jpad journal

AND option

OR option



N. Ali1,2, H. Tian1,3, L. Thabane4,5, J. Ma4, H. Wu6, Q. Zhong6, Y. Gao7, C. Sun6, Y. Zhu8, T. Wang8


1. First School of Clinical Medicine, Nanjing Medical University, Nanjing, Jiangsu, China; 2. Swat Institute of Rehabilitation & Medical Sciences, Swat, Khayber Pakhtoonkhwa, Pakistan; 3. School of Rehabilitation Medicine, Nanjing Medical University, Nanjing, China; 4. Department of Health Research Methods, Evidence, and Impact, McMaster University, Hamilton, Ontario, Canada; 5. Biostatistics Unit, St Joseph’s Healthcare, Hamilton, Ontario, Canada; 6. Rehabilitation Department, Nanjing Drum Tower Hospital Clinical College of Nanjing Medical University, Nanjing, Jiangsu, China; 7. Rehabilitation Department, The Affiliated Suzhou Hospital of Nanjing Medical University, Suzhou Municipal Hospital, Gusu School, Nanjing Medical University, Suzhou, Jiangsu, China; 8. Department of Rehabilitation Medicine, First Affiliated Hospital of Nanjing Medical University, Nanjing, Jiangsu, China. N. Ali and H. Tian have contributed equally.

Corresponding Author: Tong Wang, Department of Rehabilitation Medicine, The First Affiliated Hospital of Nanjing Medical University, No. 300 of Guangzhou Road, Nanjing, Jiangsu 210029, China. Tel: +86 13951680478, fax: +862583318752. E-mail:; Yi Zhu, Department of Rehabilitation Medicine, The First Affiliated Hospital of Nanjing Medical University, No. 300 of Guangzhou Road, Nanjing, Jiangsu 210029, China. Tel: +86 13705164030, fax: +862583318752. E-mail:

J Prev Alz Dis 2022;
Published online January 20, 2022,



Background and objective: Individuals with Alzheimer disease and dementia experience cognitive decline and reduction in physical capabilities. Engaging in cognitive challenges and physical exercises is effective in reducing age-related cognitive and physical decline. It is believed that physical activity in the context of cognitive challenges might enhance the process of neurogenesis in the adult brain, but how effective are such interventions? Is there enough evidence to support that dual-task training is more effective than cognitive or physical training alone? To what extent can such training improve cognitive and physical functions in patients at various stages of cognitive decline?
Methodology: This systematic review with meta-analysis summarizes the emerging evidence of dual-task training for enhancing cognitive and physical functions in older individuals with cognitive impairment, dementia or Alzheimer’s disease. A systematic search was carried out in MEDLINE, PubMed, EMBASE, and Cochrane Library with the following search terms: randomized control trials, dual-task training, SCD, MCI, dementia, and Alzheimer’s disease.
Results: A total of 21 studies with 2,221 participants were identified. The results of dual-task tanning intervention are summarized as change in global cognitive function; SMD = 0.24, (P= 0.002), memory; SMD = 0.28, (P = 0.000), executive function; SMD = 0.35, (P = 0.000), attention; SMD = −0.19, (P = 0.1), gait speed; SMD = 0.26, (P = 0.007), dual-task cost; SMD 0.56, (P = 0.000), and balance; SMD 0.36, (P = 0.004).
Conclusion: Primary analysis showed a small-to-medium positive effect of dual-task training interventions on cognitive functions and medium-to-large positive effect on gait functions and balance.

Key words: Dual-task training, cognitive functions, gait speed, balance, mild cognitive impairment, dementia, Alzheimer’s disease.




Owing to the upswing in the aging population the number of patients with Alzheimer’s disease and dementia is soaring, which poses a significant global healthcare challenge (1). With no reliable cure for these conditions, healthcare professionals are looking for an affordable and effective treatment (2). At present, there are about 50 million cases of dementia globally, and the number is expected to reach over 75 million by 2030 (3). The informal annual healthcare costs of dementia and Alzheimer’s disease account for 81.3% of the total care costs in China, and they are expected to further raise up to 114.2 billion US$ by 2030 (4). Disappointing results from the available clinical trials and laboratory findings suggest that finding a single treatment for dementia is very unlikely (5). Therefore, the focus is on the development of preventive and early treatment strategies for the cognitive decline during the early stages or preclinical stage of dementia (2, 6). One such particular stage of interest is subjective cognitive decline, which refers to the early stage of cognitive decline, manifesting as worsening of memory and diminished thinking skills, despite preserved cognitive functions based on traditional neuropsychological tests (7). These patients show a pattern of hippocampal and cortical atrophy and is considered a transitional phase between normal aging and dementia (8). In contrast, in MCI patients, other processing abilities such as attention, planning, organizing, perception, learning, reasoning, and judgment are also affected, in addition to memory loss (9). Depending on various conditions the annual conversion rate of MCI to dementia is between 5% and 20% (10). Dementia is a progressive and severe cognitive decline with motor deficits and behavioral problems, which can lead to decline in the ability to perform activities of daily life (1). Alzheimer’s disease, the most common type of dementia (60%–80% of all cases) is a progressive condition that affects memory, thinking, and behavior (11). It is believed that older people with subjective cognitive decline may be an ideal group for starting early preventive intervention programs. These patients are at a high risk of developing pathological cognitive decline in the future, and therefore, early intervention could improve their clinical outcome and reduce future burden on the healthcare system (12).
Cognitive deficit in older adults is strongly association with impaired spatiotemporal gait parameters such as slowing of gait and high stride time variability (13). It is believed that slow gait speed is the early sign of cognitive decline and an indicator of shorter life expectancy (14). In fact, it has been shown in a recent retrospective study that slow gait velocity and high gait variability occur up to 25 years before the accentuated cognitive decline (15). Higher gait variability is strongly associated with a greater degree of cognitive impairment and increased risk of fall in older adults (16).
It is possible to slowdown the progression of MCI, or even reverse the incidence of dementia through various interventions (17). Recent studies have shown that regular physical activity is very effective in reducing cognitive decline and enhancing brain functions and neurogenesis in older adults with MCI (18). Cognitive training and cognitive rehabilitation have also been reported to improve cognitive functions in older adults (19). Results from epidemiological studies have suggested that dual-task training (simultaneous or subsequent combined physical and cognitive training) may have far better results on physical and cognitive performance than either of them alone (20). Recent systematic analyses have shown that combined intervention can significantly improve cognitive functions in older people with or without dementia (21, 22). It has been suggested that dual-task training may induce combined effects of physical exercise and cognitive training (23). Physical exercise is believed to facilitate synaptic plasticity and cell proliferation, while cognitive training guides these newborn neurons into synapses with preexisting neural networks (24-26). Animal studies have shown that physical activities performed in a cognitively challenging environment are more beneficial in inducing neural and cognitive benefits than physical activities alone (27, 28). Several meta-analyses have shown that dual-task training is very effective in enhancing global cognitive function, memory, executive function, mood and activities of daily life in adults with cognitive decline (23, 29). Therefore, the hypothesis that, if dual-task training can improve cognitive and physical functions in cognitively impaired patients, the progression of subjective cognitive decline and mild cognitive impairment into Alzheimer’s and dementia could be halted. Furthermore, improvement in balance and gait will reduce the risk of fall and improve functional independence. These benefits are enormous and their clinical application will be highly appreciated.
Although several reviews and meta-analysis have highlighted the benefits of dual-task training, they have a few shortcomings. Firstly, the quality of their included studies is not very high and the majority had not included randomized control trials. Secondly, their main focus was on cognitive benefits only and had not evaluated changes in motor functions such as gait speed, dual task cost and balance. In fact, to date, there have been no such studies that have focused on the overall effects of dual-task training in individuals with subjective cognitive decline, mild cognitive impairment, dementia, and Alzheimer’s disease. The aim of this systemic review and meta-analyses was to assess the effects of dual-task training on cognitive and motor functions in older adults at various stages of cognitive decline. The secondary aim was to quantify the effects of DTT on global cognition, memory, executive functions, attention, gait speed, dual-task gait cost, and balance in the chosen population.



The review was registered in the International Prospective Register of Systematic Reviews (PROSPERO, on 5th July 2020 and is available at, The study was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines (30).

Search strategy

A systematic search was performed in MEDLINE, PUBMED, EMBASE, EBSCO, Cochran library, and Google scholar databases for papers published in English from inception to 30th September 2020. Search terms were intersections of cognitive intervention terms (subjective cognitive decline “cognitive training” OR “cognitive intervention” OR “memory training”), AND physical intervention terms (“exercise” OR “physical training” OR “aerobic training”), AND aging population terms (“aging” OR “aged” OR “older”), AND combined intervention terms (“combined” OR “combination” OR “multimodal”). Reference lists of the selected papers and related review papers were further screened and additional search of the gray literature and unpublished articles was performed.

Selection criteria

Studies were considered eligible if they met the following inclusion criteria: 1) randomized control trials (RCT); 2) older adults with SCD, MCI, dementia, or Alzheimer’s disease; 3) dual-task training; 4) having a control group for comparison (physically active or cognitive intervention, placebo, and health education). The exclusion criteria were as follows; (1) non-RCT study design; (2) case studies ;(3) review articles; (4) unpublished articles, theses, and dissertations; (5) study protocols; and (6) studies published in languages other than English. In accordance with the Cochrane Handbook of Systematic Review of Interventions, two authors (N Ali and H Tian) independently selected search results based on their titles and abstracts. Eligible articles were further screened for full-text assessment. Disagreements about the eligibility of studies were resolved by discussion with a third author (Y Zhu). The flow chart of study selection and numbers of included and excluded studies is shown in Figure 1.

Figure 1. Prisma flow diagram of the studies selected

Prisma flow chart of the systemic review and meta-analysis.


Data extraction, processing, and analysis

The main outcomes of interest were objective measures of cognitive functions and changes in motor function. Based on the popular classification used in a recent meta-analysis, cognitive outcome measures were further grouped into global cognition and cognitive domains (memory, executive function, and attention), while changes in motor functions were grouped into gait speed, dual-task gait cost, and balance. The two authors extracted data independently using spreadsheets for each study, including the information about the methodology, study population, type of intervention, delivery, frequency, duration, outcome measures, and any follow-up if present. In most of these studies, the selected outcome summary statistics were, mean difference (MD), standard deviation (SD), and number of participants before and after the intervention. If MD and SD were not available, standard mean difference from baseline, confidence interval (CI) (95%), P- and F- values, and regression coefficient were used. Open Meta [Analyst] OSX Yosemite (10.10), was used for data processing (31), while RevMan V5.4 was used for data analysis (32). In this meta-analysis we have used continuous outcome data (MD, SD, n) when the same measurements were used, however when different measures for the same outcome were used, we took standardized mean difference (SDM) to obtain their summary effect. The random-effects model was used if significant heterogeneity was found among the included studies so as to pool the effect with 95% CI; otherwise, the fixed-effect model was used.
I2 statistic was used to examine the heterogeneity of the included studies. As a general rule, large I2 of more than 75% suggests high heterogeneity, less than 60% represents moderate heterogeneity and less than 40% indicates low heterogeneity; therefore, we only pooled studies when I2 was below 60%. To determine the causes of the heterogeneity, we conducted sensitivity analysis by eliminating the included studies one by one to identify the study that is heterogeneous, and to see if the I2 statistics changed substantially. We also conducted subgroup analysis to modify the effects of the outcomes and to obtain the two-sided p-value (< 0.05 represents significant effect) (33). Furthermore, funnel plots were used to assess the publication bias and to identify outliers, studies that fall out of the normal funnel plot. Effect size was considered small (d=0.2), moderate (d=0.5), and large (d=0.8).

Risk of bias assessment

The Cochrane risk of bias assessment tool was used to measure the quality of the included studies (34). Here the risk of bias is reported in six domains, these are, selection bias, performance bias, detection bias, attrition bias, reporting bias, and other bias. These domains were rated as low, high, or unclear. Both authors (N Ali and H Tian) independently performed the risk of bias assessment and any difference in the outcome was resolved through discussion with a third author (Y Zhu). The total risk of bias judgment was obtained based on the accumulated assessment of all the above domains and is presented in Figure 2.

Figure 2. Risk of Bias assessment

Risk of bias assessment using Cochran collaboration’s risk of bias assessment tool (RoB).



A summary of the selection process is shown in Figure 1. The initial database search provided 443 relevant articles, out of which 408 were identified through the actual databases and 35 from additional sources. After the removal of the duplicates 409 remained, out of which 91 articles were selected for full-text inclusion. The reasons for exclusion the remaining studies were, irrelevant population (n=159); in-eligible design (n=85); thesis, dissertation, protocols (n=32); irrelevant interventions (n=24); review papers (n=11); and non-RCTs (n=7). Full-text scanning of the 91 articles resulted in excluding certain studies due to irrelevant outcome (n=32); no cognitive or motor outcomes (n=10); single-tasking (n=12); healthy controls (n=10); lack of control group (n=5); and published in languages other than English (n=1). Finally, two more studies (n=2) were added after cross-referencing and of the total articles, a total of 21 articles were included in the qualitative analysis and 20 papers were used for quantitative analysis.

Characteristics of the included Studies

Table 1 provides a summary of the included studies. There were 11 (n=11) studies with Mild cognitive impairment patients and the total number of the included individuals was 1176 (Intervention group 1176 and 1168 in the control group). There were four studies with dementia patients (n=4) and the total number of people in these studies was 345 (1153 in the intervention group and 142 in the control group). There were four studies of patients with subjective cognitive decline (n=4), where the total number of participants was 410 (Intervention group 207 and 203 in the control group). The remaining two studies (n=2) were of patients with Alzheimer’s disease and there were 290 participants in these studies (200 people in the intervention group and 90 in the control group). The majority of the included studies were from Japan (n=5), followed by Canada (n=4), Germany (n=3), and Italy (n=2). The remaining studies were carried out in Belgium, France, Finland, Hong Kong, Korea, Slovakia, and Taiwan.

Table 1. Characteristics of the included studies

AD, Alzheimer’s disease. MCI, mild cognitive impairment. SCD, subjective cognitive decline, MMSE, mini mental state examination. MoCa, Montreal cognitive assessment. CDT, clock drawing test. STD, Stroop word color test. RCT, randomized control trail. GCF, global cognitive functions.


Primary and secondary analysis

Effect of dual-task training on global cognitive function

Primary analysis showed that dual-task training had low-to-moderate effect on global cognitive function in older adults with SCD, MCI, and dementia or Alzheimer’s disease (SMD = 0.24 [0.11; 0.36], 95% CI, Z = 3.37, and P= 0.002). Furthermore, secondary analysis significantly reduced heterogeneity across the studies (Ch2 = 5.74, I2 = 0%, Figure 2). The overall effect for subgroup MCI, SCD, and dementia patients was low-to-moderate (SMD = 0.29 [0.16; 0.43], Z= 4.18, P= 0.000), and was negative for Alzheimer’s disease patients (SMD = −0.02 [−0.31; 0.28], Chi2 = 5.92, P=0.82), without significant heterogeneity (I2 =0%). Furthermore, funnel plot did not reveal risk of publication bias (Figure 3b).

Figure 3a. Pooled effect of dual-task training on global cognition

Forest plot efficacy of combined cognitive physical intervention compared to control group on global cognitive function. Study ID, mean difference between pre and post intervention, N= number of patients, SD standard deviation, StD standardized mean difference (95% confidence interval), random effects model.


Figure 3b. Funnel plot of dual-task training on global cognition

Funnel plot showing symmetrical distribution of studies indicating absence of publication bias.


Effects of dual-task training on domain-specific cognitive function

Effects of Dual-task training on memory

Figure 3c shows the combined effect of dual-task training on memory calculated with random effects model. Earlier there was high heterogeneity among the studies therefore, secondary analysis between various groups was performed that reduced heterogeneity significantly. SD mean with CI 95% for MCI, dementia and Alzheimer’s was 0.33 (0.2, 0.47), effect size Z = 4.81, and P = 0.000, while for subjective cognitive decline was SDM = 0.09 (-0.18, 0.37), Z = 0.68, and P =0.5. In addition there was no heterogeneity among these studies (I2 = 0%.). Furthermore the overall effect of dual task training on memory was SMD = 0.28 (0.16, 0.41), with Z = 3.75, and P = 0.000.

Figure 3c. Forest plot of the effects of dual-task training on memory

Forest plot efficacy of combined cognitive physical intervention compared to control group. Study ID, Mean difference between pre and post intervention, N= number of patients, SD standard deviation, StD standardized mean difference (95% confidence interval), weight indicates influence of individual study on the pooled results, random effects model. MCI mild cognitive impairment, AD Alzheimer’s disease, SCD subjective cognitive disorder.


Post-intervention effects of the dual-task training intervention on executive function

Figure 3d shows that dual task training had moderate effect on executive function calculated through the random-effects model. Mean SD with 95% CI was 0.35 (0.23, 0.48), with Z = 5.43 and P = 0.000. In addition, heterogeneity was low among the studies included (I2 = 8%).

Figure 3d. Forest plot of the effects of dual-task training on executive function

Forest plot efficacy of combined cognitive physical intervention compared to control group. Study ID, Mean difference between pre and post intervention, N= number of patients, SD standard deviation, StD standardized mean difference (95% confidence interval), random effects model.


Post-intervention effects of dual-task training on attention

The results of the random effects model analysis showed that dual task training has a low impact on attention; {SMD −0.19 (0.42, 0.4), Z= 1.63, and P = 0.1}. There was no heterogeneity among the included studies; (Chi2 = 1.94, P = 0.86, and I2 = 0%). One study (Maffeil et al. 2017) was excluded based on visual inspection of the funnel plot and was considered an outlier. Furthermore, attention was measured with Trail Making Test part-A by majority of the included studies (n=3), while two had (n=2), used the Strop test and one study (n=1) had used Digit Symbol Test as their measuring tool. The negative sign of the results indicate a reduction in time to complete the test representing improved attention. The results are shown in Figure 3e.

Figure 3e. Forest plot of the effects of dual-task training on attention

Forest plot efficacy of combined cognitive physical intervention compared to control group on attention. Study ID, Mean difference between pre and post intervention, N= number of patients, SD standard deviation, StD standardized mean difference (95% confidence interval), random effects model.


Post-intervention effects of dual-task training on physical functions (gait parameters)

Effects of dual-task training on gait speed

Figure 4a shows the overall effect of dual-task training on gait speed calculated through the random effects model. Mean SD with 95%CI was 0.35 (0.17, 0.54), Z = 3.72, Chi2 =6.5, and P = 0.000 for subgroups SCD, MCI and dementia. The pooled effect of gait speed for Alzheimer’s disease patients was −0.17 (−0.5, 0.16), Z=1.02, P = 0.31. Subgroup analysis highly reduced heterogeneity among the included studies, while the overall effect size was 0.26 (0.04, 0.49), Z=2.26, and P= 0.02. The two groups were significantly different from each other Z=7.3, P= 004 and I2 = 86%.

Figure 4a. Forest plot of the effects of dual-task training on gait speed

Forest plot efficacy of combined cognitive physical intervention compared to control group on gait speed. Study ID, mean difference between pre and post intervention, N= number of patients, SD standard deviation, StD standardized mean difference (95% confidence interval), random effects model.


The impact of dual-task training on dual-task gait cost (DTC)

The result showed that dual-task training had huge impact in reducing dual-task gait cost {SMD 0.65 (0.24, 0.88), Z= 3.4, and P = 0.000}. In addition there was low heterogeneity among the studies included (Chi2 = 2.88, P = 0.41, and I2 = 27%). Dual-task gait cost was measured as the percentage of the rate of dual tasking subtracted the rate of single tasking divided by rate of single tasking (35). The results are shown in Figure 4b.

Figure 4b. Pool effect size of the effects of dual-task training on Dual task gait cost

Forest plot efficacy of combined cognitive physical intervention compared to control group on dual task gait cost [(dual-task RT – simple RT) / simple RT Å~ 100]. Study ID, mean difference between pre and post intervention, N= number of patients, SD standard deviation, StD standardized mean difference (95% confidence interval), random effects model.


Effects of dual-task training on balance

Results of the random effects model analysis showed that dual task training had low-to-moderate impact on balance {SMD 0.36 (0.12, 0.61), overall effect Z= 2.87, and P = 0.004}. In addition there was no heterogeneity among the studies included (Chi2 = 4.64, and I2 = 0%). Furthermore, balance was measured with different measures by these studies, which were, Tinetti-POMA scale (Performance-oriented Mobility Assessment), Disjunctive Reaction Time (DRT-II), Falls Efficacy Scale International (FES-I), and Timed Up and Go test. The results are shown in Figure 4c.

Figure 4c. Forest plot of the impact of dual-task training on balance

Forest plot efficacy of combined cognitive physical intervention compared to control group on balance. Study ID, mean difference between pre and post intervention, N= number of patients, SD standard deviation, StD standardized mean difference (95% confidence interval), random effects model.



This review and meta analysis extend our knowledge of the beneficial effects of dual-task training on cognitive and physical functions in older individuals at various stages of cognitive impairment. In addition, here we have quantified the effects on global cognitive function, memory, executive function, attention, gait speed, dual task cost and balance. Twenty-one RCTs published between 2010 and 2020 were included in this study, representing relatively new and emerging field of research.
To our knowledge, this is the first meta-analyses about the effects of dual-task training on cognitive and motor functions in cognitively impaired older adults. By summarizing the findings of 21 RCTs of older individuals at various stages of cognitive impairment, we have analyzed the effects of dual-task training on cognitive and physical functions. A small effect size of dual-task training was found on global cognitive function and attention domain, while low-to-moderate impact on memory and executive function. We have also found low-to-moderate impact on gait speed and balance, and moderate-to-high impact on dual task cost. Surprisingly, a minimal-to-low negative impact of dual-task training was found on global cognitive function and gait speed in Alzheimer’s disease patients and minimal-to-low positive impact on memory in subjective cognitive decline patients. This might be due to the limited number of available studies in this population, stage of the cognitive impairment, and/or other age-related dysfunction (36).
Our results are in line with the findings of other reviews and meta-analysis carried out on the same population and same intervention (22, 29). These reviews have found low-to-medium effect on global cognitive function, although they had included older adults with and without cognitive problems and have also included non-RCTs in their review. Karssemeijer et al. (2017) have also reported a mild-to-moderate on executive function, attention and memory, although, attention and executive function was considered as a single domain in their study. In addition, Bruderer-Hofstetter et al. (2017), and Yang et al. (2019), have also reported similar findings in cognitive and physical performance (37, 38). Furthermore, only a few studies have focused on changes in motor functions such as gait speed and balance in individuals with cognitive impairment; thus, no reviews are available about the impact of dual-task training on dual-task cost. This might be due to the limited number of studies with gait function outcome.

Strengths, limitations and future recommendations

The main strength of our review and meta-analyses is that the evidence is much stronger than the previous reviews because we have included RCTs only. The data used to synthesize the estimated pool effect size were obtained from articles searched through a wide range of databases that reduced the incidence of bias. The aim of this meta-analysis was not only to evaluate the impact of dual-task training on older adults with cognitive problems, but also to replicate previous findings while addressing methodological and heterogeneity problems. In addition, we have analyzed the outcomes of cognitive function as well as physical outcomes, which have not been addressed before. Our findings support the clinical application of dual-task training for patients at various stages of cognitive decline. Although the intensity, duration, frequency and contents of our intervention (dual-task training) are different its cognitive and motor benefits are immense and can lead to improved mental capabilities, better coordination, fall prevention and independent life. Therefore devising a combined mind-body training program for older adults with dementia and Alzheimer’s disease could be a very good strategy for the prevention and rehabilitation of cognitive decline and fall prevention.
The findings from this study should to be considered in the context of several methodological limitations. In particular, variations of the intervention, duration, frequency, settings and the classification of cognitive impairment was not consistent among the included studies. Complexity of the dual-task and systematic differences between population groups base statistics makes the findings of this meta analysis prone to bias and differential outcome. In addition we have only included RCTs published in English and might have missed studies published in other languages. Furthermore, all the interpretations in the current meta-analyses are based on the estimated effect size and not the actual outcome; therefore, these results should be considered with caution.
Future research should focus on domain specific intervention for cognitive problems in older individuals and should aim at slowing down or preventing progression of cognitive impairment in MCI, dementia or Alzheimer’s disease. In addition, further research is needed to establish the optimum intervention intensity, duration, and frequency of dual-task training for long lasting impact. Furthermore researchers should ideally use a standardized protocol for the studies to be more conclusive and facilitate interstudy comparisons.



To conclude, this review and meta-analysis support the notion that dual-task training is an effective non-pharmacological intervention, which can improve cognitive and physical functions in older people with cognitive impairment, when compared to other therapies. We have found that 2–5 sessions (30-–120 minutes) of dual-task training per week has the potential to improve global cognition, executive function, attention and memory in cognitively impaired older adults. There is also evidence that such training programs can enhance physical functions such as gait speed, reduce dual task cost, and improve balance. These findings support that dual-task training impart diverse cognitive and physical improvements for older individuals with cognitive diseases. However, long-term studies are needed to determine the effectiveness of dual-task training in the wider community and whether or not such effects can be sustained in the long term.


Ethical approval and consent to participate: not applicable.

Consent for publication: All the authors agreed the manuscript for publication.

Availability of data and material: All the data could be made available on request.

Conflict of interest: The authors declare that this research was conducted in the absence of any commercial or financial relationships that could be considered as potential conflict of interest.

Author Contributions: Yi Zhu: Completed the funding application managed and coordinated the study and drafted the initial protocol and registration. Tong Wang: Provided the research ideas and revised the manuscript. Nawab Ali, Huifang Tian: Database search, study selection, data collection, analysis, interpretation and manuscript draft. Qian Zhong, Yaxin Gao, Han Wu, Cuiyun Sun: Revised the study design, data process and interpretation. Lehana Thabane, Jinhui Ma: Study design, guided statistical analysis and manuscript.

Funding: This work was funded by National Natural Science Foundation of China (NSFC) (Grant No. 81802244), and National Key R&D Program of China (Grant No.:2018YFC 2001600, 2018YFC 2001603) and Nanjing Municipal Science and Technology Bureau (Grant number of 2019060002).

Acknowledgments: The authors would like to thank all the authors of the included trails and their participants. We also like to thank Majid Yousafzai from Nanjing medical university for his contribution and guidance.

Trail registration: CRD42020179392.



1. Zhang XX, Tian Y, Wang ZT, et al. The epidemiology of Alzheimer’s disease modifiable risk factors and prevention. J Prev Alzheimers Dis. 2021;8(3):313-321. doi:
2. Porsteinsson AP, Isaacson RS, Knox S, et al. Diagnosis of early Alzheimer’s disease: clinical practice in 2021. J Prev Alzheimers Dis. 2021;8(3):371-386. doi:
3. Chris Lynch. World Alzheimer Report 2019: Attitudes to dementia, a global survey. Alzheimers. Dement. 2020;16(Suppl.10):e038255. doi:
4. Xu J, Wang J, Wimo A, Fratiglioni L, et al. The economic burden of dementia in China, 1990-2030: implications for health policy. Bull World Health Organ. 2017;95(1):18-26. doi:
5. Vellas B, Aisen P. New hope for Alzheimer’s disease. J Prev Alzheimers Dis. 2021;8:238–239. doi:
6. Canevelli M, Vanacore N, Blasimme A, et al. Overtreating Alzheimer’s disease. J Prev Alzheimers Dis. 2021;8:234-236. doi:
7. Perrotin A, La R, De La V, et al. Subjective cognitive decline in cognitively normal elders from the community or from a memory clinic: Differential affective and imaging correlates. Alzheimers. Dement. 2017;13(5):550-560. doi:
8. Saykin AJ, Wishart HA, Rabin LA, et al. Older adults with cognitive complaints show brain atrophy similar to that of amnestic MCI. Neurology. 2012;67:834-842. doi:
9. Albert MS, DeKosky ST, Dickson D, et al. The diagnosis is of mild cognitive impairment due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association work groups on diagnostic guidelines for Alzheimer’s disease. Alzheimers. Dement. 2011;7(3):27029. doi:
10. Hu C, Yu D, Sun X, et al. The prevalence and progression of mild cognitive impairment among clinic and community populations: A systematic review and meta-analysis. Int Psychogeriatr. 2017;29(10):1595-1608. doi:
11. Samuel BG. Diagnostic and statistical manual of mental disorders, 4th ed. (DSM-IV). The Am J Psychiatry. 1 Apr 2006. doi:
12. Livingston G, Huntley J, Sommerlad A, et al. Dementia prevention, intervention, and care, 2020 report of the lancet commission. Lancet. 2020;396(10248):413-446. doi: (20) 30367-6.
13. Montero-Odasso M, Oteng-Amoako A, Speechley M, et al. The motor signature of mild cognitive impairment: results from the gait and brain study. J Gerontol A Biol Sci Med Sci. 2014;69(11):141521. doi:
14. Studenski S, Perera S, Patel K, et al. Gait speed and survival in older adults. JAMA. 2011;305(1):50-58. doi:
15. MacDonald SWS, Hundza S, Love JA, et al. Concurrent indicators of gait velocity and variability are associated with 25-year cognitive change: a retrospective longitudinal investigation. Front Aging Neurosci.2017;9:17. doi: 3389/fnagi.2017.00017.
16. Beauchet O, Allali G, Annweiler C, et al. Gait variability among healthy adults: low and high stride-to-stride variability are both a reflection of gait stability. Gerontology. 2009;55(6):702-706. doi:
17. Hayakawa T, McGarrigle CA, Coen RF, et al. Orthostatic blood pressure behavior in people with mild cognitive impairment predicts conversion to dementia. J Am Geriatr Soc. 2015;63 (9):1868 – 1873. doi:
18. Chandra L, Thais R, Andressa B, et al. Effect of exercise on cognition, conditioning, muscle endurance, and balance in older adults with mild cognitive impairment: a randomized controlled trial. J Geriatr Phys Ther. 2019;42(2): E15-E22. doi:
19. Ball K, Berch DB, Helmers KF, et al. Effects of cognitive training interventions with older adults: a randomized controlled trial. JAMA. 2002;288(18):2271–2281. doi:
20. Combourieu L, Perrot A, Belleville S, et al. Effects of simultaneous aerobic and cognitive training on executive functions, cardiovascular fitness and functional abilities in older adults with mild cognitive impairment. Ment Health Phys. Act. 2018;15:78–87. doi:
21. Jamie L, Tait, Rachel L, et al. Influence of sequential vs. simultaneous dual-task exercise training on cognitive function in older adults. Front Aging Neurosci. 2017;9:368. doi:
22. Esther K, Justine A, Willem B, et al. Positive effects of combined cognitive and physical exercise training on cognitive function in older adults with mild cognitive impairment or dementia: A meta-analysis. Ageing Res. Rev. 2017;40:75–83. doi:
23. Law LFL, Barnett F, Yau, et al. Effects of combined cognitive and exercise interventions on cognition in older adults with and without cognitive impairment: a systematic review. Ageing Res Rev. 2014;15:6175. doi:
24. Bamidis PD, Vivas AB, Styliadis C, et al. A review of physical and cognitive interventions in aging. Neurosci Bio behav Rev. 2014;44:206-220. doi:
25. Curlik DM, Shors TJ. Training your brain: do mental and physical (MAP) training enhance cognition through the process of neurogenesis in the hippocampus? Neuropharmacology. 2013;64:506–514. doi:
26. Fissler P, Küster O, Schlee W, et al. Novelty interventions to enhance broad cognitive abilities and prevent dementia: synergistic approaches for the facilitation of positive plastic change. Prog Brain Res. 2013;207:4-0334. doi:
27. Fabel K, Kempermann G. Physical activity and the regulation of neurogenesis in the adult and aging brain. Neuromolecular Med. 2008;10(2):59–66. doi:
28. Langdon KD, Corbett D. Improved working memory following novel combinations of physical and cognitive activity. Neurorehabil Neural Repair. 2012;26(5):523–532. doi:
29. Zhu XY, Yin SF, Lang MJ, et al.The more the better? A meta-analysis on effects of combined cognitive and physical intervention on cognition in healthy older adults. Ageing Res Rev. 2016;31:67-79. doi:
30. David M, Jennifer T, Alessandro L, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ. 2009;339:b2535. doi:
31. Wallace B C, Dahabreh I J, Trikalinos T A, et al. Closing the gap between methodologists and end-users: R as a computational back-end. J Stat Softw. 2012;5(49):1-15. doi: http://hdl.handle.net10.18637/jss.v049.i05.
32. Review Manager 5.4, Software. May 2020. The Cochran Collaberation 2019, available at: 15/06/2020.
33. Julian H, James T, Jacqueline C, et al. Cochrane Handbook for Systematic Reviews of Interventions, Version 6, 2019. available at
34. Higgins JP, Green S, Cochrane Handbook for Systematic Reviews of Interventions. John Wiley & Sons,2011.available at
35. Chen YL, Pei YC. Musical dual-task training in patients with mild-to-moderate dementia: a randomized controlled trial. Neuropsychiatr Dis Treat. 2018;14:1381–1393. doi:
36. Parvin E, Mohammadian F, Amani-Shalamzari S, et al. Dual-task training affect cognitive and physical performances and brain oscillation ratio of patients with Alzheimer’s disease: a randomized controlled trial. Front Aging Neurosci. 2020;12:605317. doi:
37. Yang C, Moore A, Mpofu E, et al. Effectiveness of combined cognitive and physical interventions to enhance functioning in older adults with mild cognitive impairment: a systematic review of randomized controlled trials. Gerontologist. 2020 Nov 23;60(8):633-642. doi: 10.1093/geront/gnz149.
38. Bruderer-Hofstetter M, Rausch-Osthoff AK, Meichtry A, et al. Effective multicomponent interventionsin comparison to active control and no interventions on physical capacity, cognitive function and instrumental activities of daily living in elderly people with and without mild impaired cognition – A systematic review and network meta-analysis. Ageing Res Rev. 2018 Aug;45:1-14. doi:



X. Xu1,2,*, K.A. Chew2,*, Z.X. Wong3, A.K.S. Phua3, E.J.Y. Chong3, C.K.L. Teo3, N. Sathe3, Y.C. Chooi4, W.P.F. Chia5, C.J. Henry6,7,8, E. Chew9,10, M. Wang11, A.B. Maier12,13,14, N. Kandiah15,16, C.L.-H. Chen2,3


1. School of Public Health and the Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China; 2. Memory, Ageing and Cognition Centre (MACC), Department of Pharmacology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 3. Memory, Ageing and Cognition Centre (MACC), Department of Psychological Medicine, National University Hospital, Singapore; 4. Singapore Institute for Clinical Sciences, Agency for Science, Technology and Research (A*STAR), Singapore; 5. School of Sports, Health and Leisure, Republic Polytechnic, Singapore; 6. Clinical Nutrition Research Centre, Singapore Institute for Clinical Sciences, Agency for Science, Technology and Research (A*STAR), Singapore; 7. Department of Biochemistry, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 8. Singapore Institute of Food and Biotechnology Innovation, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 9. i Division of Neurology, Department of Medicine, National University Hospital, Singapore; 10. Department of Medicine, Yong Loo Lin School of Medicine, National University of Singapore; 11 Division of Sports Medicine & Surgery, Orthopaedic Surgery, National University Hospital, Singapore; 12. Department of Human Movement Sciences, @Age Amsterdam, Vrije Universiteit Amsterdam, Amsterdam Movement Sciences, Amsterdam, the Netherlands; 13. Healthy Longevity Translational Research Program, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 14. Centre for Healthy Longevity, @Age Singapore National University Health System, Singapore; 15. Department of Neurology, National Neuroscience Institute, Singapore; 16. Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore; *Equal contribution

Corresponding Author: Dr. Christopher Li-Hsian Chen, Department of Pharmacology, Yong Loo Lin School of Medicine, MD3, 04-01, 16 Medical Drive, National University of Singapore, Singapore 117600, Singapore. E-mail

J Prev Alz Dis 2022;
Published online January 11, 2022,



Background: The SINgapore GERiatric intervention study to reduce cognitive decline and physical frailty (SINGER) randomised controlled trial (RCT) uses a multidomain lifestyle interventions approach, shown to be effective by the Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER) trial, to delay cognitive decline.
Objective: To investigate the efficacy and safety of the SINGER multidomain lifestyle interventions in older adults at risk for dementia to delay cognitive decline.
Participants: 1200 participants between 60-77 years old, with Cardiovascular Risk Factors, Aging, and Incidence of Dementia (CAIDE) dementia risk score ≥6, fulfilling at least one of the following LIBRA index for diet, cognitive activity, physical activity and a Montreal Cognitive Assessment (MoCA) score ≥18, ≤27 points, will be recruited across Singapore.
Methods: SINGER is a 2-year multi-site RCT consisting of multidomain interventions: dietary advice, exercise, cognitive training, and vascular risk factors management. Participants will be randomised into either the Self-Guided Intervention (SGI; general lifestyle and health information and resources) or Structured Lifestyle Intervention (SLI) group. The SLI comprises diet training (6 group and 3 individual sessions over 12 months); exercise (supervised: 1-hour twice weekly for 6 months, unsupervised: 2-3/week for the rest of the study duration); cognitive sessions (15-30 minutes/session, 3/week for 6 months, together with 10 workshops in 24 months). Vascular management takes place every 3-6 months or otherwise as specified by study physicians. The primary outcome is global cognition measured using the modified Neuropsychological Battery assessing performance in various domains, such as episodic memory, executive function and processing speed. Secondary outcome measures include: domain-specific cognition and function, imaging evidence of brain and retinal changes, incidence and progression of chronic diseases, blood biomarkers, quality of life, mental health and cost-benefit analysis.
Conclusions: SINGER is part of the Worldwide-FINGERS international network, which is at the forefront of harmonizing approaches to effective non-pharmacological interventions in delaying cognitive decline in older adults at risk of dementia. By establishing the efficacy of multidomain interventions in preventing cognitive decline, SINGER aims to implement the findings into public health and clinical practices by informing policy makers, and guiding the design of community- and individual-level health promotion initiatives.

Key words: Randomized controlled trial, multi-domain life style interventions, cognitive dysfunction, aged.




The prevalence of dementia is already high, particularly in Asia (1), and is projected to increase exponentially due to a rapidly ageing population, hence, dementia will strongly impact individuals, their families and healthcare systems (2-4).
Nearly one-third of dementia cases internationally are estimated to be attributable to modifiable lifestyle, vascular or metabolic risk factors (5). Of note, reducing the prevalence of each of these risk factors by 10% or 20% per decade may reduce Alzheimer’s Disease (AD) prevalence worldwide by 8 to 15% by 2050 (6). Studies using single-domain lifestyle interventions in reducing dementia risk, such as healthy diet, cognitive training, or blood-pressure management, appear less likely to have a sustainable impact (7).
By contrast, the approach of combining multiple intervention components may have synergistic effects. However, multi-domain intervention studies have demonstrated mixed outcomes (8-12). The 3-year French Multidomain Alzheimer Preventive Trial (MAPT) compared the effect of 3 interventions on cognitive decline (9). It was reported that the combined multidomain intervention groups showed a significant decrease in cognitive decline compared with the placebo group, in participants at higher risk of dementia. Additionally, the Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER) trial produced promising results (10). This population-based 2-year randomised controlled trial (RCT) demonstrated the effectiveness of a multidomain lifestyle intervention comprising of cognitive training, physical activity, nutritional advice, social activities and vascular risk factors management on delaying cognitive decline in older adults at risk of dementia. Encouraged by these results, the Worldwide-FINGERS (WW-FINGERS) initiative was established in 2017 to adapt, test, and optimize the FINGER model in different geographical, cultural and economic settings (13, 14).
In Singapore, a 6-month pilot study demonstrated the cultural feasibility and practicality of the FINGER interventions and a set of locally adapted interventions in an Asian population (15). Hence, we propose to conduct a larger-scale trial to determine the efficacy of these interventions in an Asian population in Singapore, with the integration of novel digital intervention delivery and evaluation approaches.
In line with the WW-FINGERS international effort, this study will be investigating the efficacy and safety of the SINGER multidomain lifestyle interventions, incorporating more intensive blood pressure lowering, on cognition in older adults with increased risk of dementia. The SINGER interventions are based on standard healthcare recommendations and encompass a series of supervised and unsupervised sessions, to empower participants to continue sustaining these lifestyle modifications.



Study design

The multi-centre SINGER study is a 2-year randomized controlled trial (RCT) (Figure 1). 1200 older adults at risk of cognitive decline and dementia, determined using a Cardiovascular Risk Factors, Aging, and Incidence of Dementia (CAIDE) dementia risk score of ≥6, will be recruited from community dwelling and clinical cohorts of Singaporean older adults from August 2021.
Participants will be randomized to either a Self-Guided Lifestyle (SGL) or to a Structured Lifestyle Intervention (SLI) group, which focuses on regular aerobic and strength exercise, adherence to the FINGER diet, cognitive and social stimulation, and protocol-based vascular monitoring. Experienced neuropsychological raters evaluating cognitive outcome measures will be blinded to randomization group. Additionally, a digital administrative and patient-facing platform will be utilised. The SINGER study has been approved by the National Healthcare Group Domain-Specific Review Board and is registered under (ID: NCT05007353). Written informed consent will be obtained from all participants before enrolment into the study. The study will be conducted in accordance with Good Clinical Practice guidelines.

Figure 1. Summary of study design


Primary study objective

To determine whether randomization to a Self-Guided Intervention (SGI) versus Structured Lifestyle Intervention (SLI) group improves cognitive performance, measured by global cognitive z-scores.

Secondary study objectives

To examine (a) intervention effects on domain-specific cognition and function, (b) imaging evidence of brain and retinal changes, (c) blood biomarkers, (d) incidence and progression of chronic diseases, (e) quality of life, mental health and cost-benefit analysis.

Inclusion criteria

The SINGER study will recruit participants who are (1) aged 60-77 years, (2) at risk of dementia as determined by a CAIDE score of ≥6, (3) have modifiable lifestyle factors (fulfilling ≥1 of the LIBRA Questionnaires for diet, cognitive activity, physical activity), (4) Montreal Cognitive Assessment (MoCA) score between 18 and 27, both inclusive, (5) no plans to travel outside of Singapore for an extended period of time over the study period, (6) no physical disabilities that preclude study participation, (7) willing to complete all study-related activities for 24 months, (8) willing to be randomised to either lifestyle intervention group, (9) able to understand English or Mandarin Chinese.

Exclusion criteria

The SINGER study will exclude participants who have (1) malignant diseases, (2) dementia, (3) major depression, (4) symptomatic cardiovascular disease, (5) revascularisation within 1 year, (6) severe loss of vision, hearing or communicative ability, (7) other conditions that inhibit safe engagement in the prescribed intervention and other conditions preventing cooperation, as judged by the study physician.


Participants will be randomized into either the SGI or SLI group, generated by a random allocation sequence by non-study related staff using STATA version 14 using a 1:1 schedule, stratified by site. Treatment allocation will only be revealed to the study team after receiving information demonstrating that the patient is eligible and has consented to the trial.

Intervention programme

Participants in the SGI arm will receive general lifestyle health education information, tools, resources, and support to encourage a healthier lifestyle (16). These resources exceed those provided to older adults in Singapore. They will also receive blood laboratory testing and blood pressure monitoring during their outcome visits. General written information about the clinical significance of measurements, and advice to seek treatment if needed will be provided.
Participants in the SLI arm will be assigned the following interventions (Figure 2).

Figure 2. Study schedule for the SLI group


Diet intervention

Participants will be allocated the FINGER diet, adapted from the Mediterranean diet, which has been recommended to be an effective diet by the World Health Organization (17, 18): 10- 20% daily energy (E%) from proteins, 25-35% from fat, 45-55% from carbohydrates, 25-35g/day of dietary fibre, <5g/day salt and <5% from alcohol. They will be advised to have higher consumption of fruits and vegetables, wholegrain products, low-fat options in milk and milk products, limited added sugar intake, use of healthier oil (e.g. vegetable oil) instead of butter or saturated fat, and consumption of at least 2 portions of fish per week (10). Weight maintenance is expected.
Participants will receive 6 group-based nutrition advocacy workshops over 12 months. Each session is approximately 60-90 minutes and may be virtual if face-to-face meetings are disallowed. Discussions will include My Healthy Plate (19), food choice selection, and healthier food choices. They will also undergo 3 individual-based nutrition training sessions (30 minutes each), which are adapted to individual needs identified by the study team within the first year of the study. Participants will complete a 3-day food diary and questionnaires at each outcome visit.

Exercise intervention

Participants will be assigned to the modified FINGER exercise programme (20), which includes once to twice weekly supervised strength training sessions of up to 1 hour for 6 months. The sessions comprise of stretching, balance training, strengthening by targeting major muscle groups using free weights, TheraBands and/or weights machine, and exercise education. Additionally, a home exercise programme including aerobic exercise recommendations (30 minutes, up to 5 times/week) will be provided. Wearable activity monitors and home exercise weights will be recommended. After the initial 6-month training period, participants will be provided with a home exercise programme. An individual aerobic training program comprising activities such as walking, swimming, jogging and cycling will be planned with each participant. The progressive strength training program is based on 1 Repetition Maximum (RM) at baseline and re-measurements at month-12 and month-24 outcome visits. However, if gym access is restricted, Rating of Perceived Exertion (RPE) scale will be utilized. Participants will rate their level of difficulty for the exercises from 0 (extremely easy) to 10 (extremely hard) to assist with exercise titration, which can be done remotely. Strength training workload throughout the course of the study will be 60 to 70% 1RM/RPE 5-6, with a range of 12 to 15 repetitions for each set. 1RM/RPE will be assessed at baseline and at the end of month-6. Aerobic exercise intensity will be practised at moderate intensity levels at a recommended rating of 5-6 RPE. Participants will complete an exercise diary to allow review of the exercise prescription to achieve individualised targets: increasing from 1-2x/week for strength and 1-4x/week for aerobic in the first 6 months to 2-3x/week for strength and 3-5x/week for aerobic for the rest of the study duration.

Cognitive intervention

Ten cognitive training workshops, each lasting approximately 60-90 minutes, will take place throughout the study period. Discussions will include topics on age-related cognitive changes, memory strategies and everyday memory training. Participants will also undergo 2 periods of individualised computer-based cognitive training sessions of 6 months each, with a total of 144 sessions to be completed within 2 years. A tablet will be provided to all participants to complete the cognitive training sessions. The digital training program will take place thrice per week with each session taking 15-30 minutes (total 72 sessions in 6 months). The computerised cognitive training program will be used to conduct training; of note, a translated Mandarin version of the program will be used for Mandarin-speaking participants (10). The first 3-months will be supervised. Participants will then be advised to perform cognitive training at a recommended thrice a week unsupervised.

Vascular risk factors management

Participants will meet the study team every 3-6 months unless otherwise specified by the study physician to measure their blood pressure (BP), weight, body mass index (BMI), and hip and waist circumference. The study physician may request for additional glucose and cholesterol tests, if required. Participants will also meet the study team at screening, 12 and 24 months for detailed medical history taking and physical examination. At every visit, all participants will be provided with oral and written information and advice on healthy diets and physical, cognitive, and social activities beneficial for management of vascular risk factors and disability prevention. Blood samples will be collected at baseline and month-24. Relevant laboratory test results will be given to all participants, together with general written information about the clinical significance of measurements, and advice to seek treatment if needed. Management of metabolic and vascular risk factors will be based on Singapore Ministry of Health Clinical Practice Guidelines( 21). BP management of participants with hypertension will be based on the SPRINT study protocol to target a systolic BP of <120mmHg (22). This target has shown both cardiovascular benefits and improved cognitive outcomes (15). BP management will include meetings with the study team at the outcome measure visits, and as requested by the study physician for measurements of BP, weight and BMI, and hip and waist circumference, physical examinations, and recommendations for lifestyle management. Study physicians will initiate medication if further management is needed and arrange for further clinical follow up. Adjustment of dose will be based on a mean of three BP measurements taken after 5 minutes of quiet rest while seated (22).

Primary outcome

The primary outcome is the composite z-score of global cognitive performance at month-12 and month-24, measured using the modified Neuropsychological Battery (mNTB) (23, 24), which includes Visual Paired Associates, Logical Memory Recall of the Wechsler Memory Scale-Revised, Rey Auditory Verbal Learning, Digit Span, Word and Category Fluency test, Trail Making Test, and Letter Digit Substitution test.

Secondary outcomes

Cognitive domains

To examine intervention effects on specific cognitive domains, composite scores from the mNTB subtests (23, 24) including Episodic Memory (Visual Paired Associates tests, Logical Memory Immediate and Delayed Recall of the Wechsler Memory Scale-Revised, Rey Auditory Verbal Learning test), Executive Function (Digit Span, Word and Category Fluency test, Trail Making Test (TMT) Part B), and Processing Speed (Letter Digit Substitution test and Trail-Making Test (TMT) Part A) will be utilised. A digital cognitive battery will also be used alongside the mNTB to assess cognitive domains.

Functional abilities

To examine intervention effects on function, the Clinical Dementia Rating–Sum of Boxes (CDR-SB) (25), and functional abilities as measured on the ADCS-MCI Activities of Daily Living Inventory (ADCS-MCI-ADL) (26) will be performed.

Lifestyle factors

To examine intervention effects on lifestyle, a composite score based on self-reported physical activity, diet, and cognitive activity will be measured. Differences between intervention groups in physical function, mood, sleep quality, and subjective memory concerns will also be assessed. Scales used include the Geriatric Depression Scale (27), Pittsburgh Sleep Quality Index (28), Prospective-Retrospective Memory Questionnaire (29), Global Physical Activity Questionnaire (30), Leisure-Time Activities Questionnaire (31) and Physical Performance Test (32). See Supplementary Material.

Vascular risk factors

To examine differences between intervention groups in cardio-metabolic disease risk, changes in BP (in mmHg), lipid profile and glucose (both in mmol/L) will be measured, and incident events using serious adverse event reports assessed.


To examine intervention effects on changes in brain structural integrity, grey matter volume loss, white matter microstructure degradation, and increase of cerebrovascular disease markers (CeVD) seen on the magnetic resonance imaging (MRI) will be assessed. An evaluation of vascular polygenic risk scores (PRS) and its association with neurodegeneration, CeVD burden and cognitive decline will be performed.

Blood biomarkers

Novel and accessible blood markers to monitor AD- and CeVD-associated pathologies will be measured. Plasma concentrations of cardiac markers (High-sensitive cardiac troponin T (hs Troponin T), N-terminal pro b-type natriuretic peptide (NT-proBNP) and Growth/differentiation factor 15 (GDF 15)), and peripheral biomarkers of Aβ, tau and synaptic pathology, oxidative stress, endothelial/cardiovascular injury and degenerative protein modifications (DPMs) damaged proteins will be used.

Retinal imaging

Retinal imaging data will be used to predict cognitive decline. Retinal structural, vascular and neuronal changes will be assessed.

Cost-benefit Analysis

To develop macro (population level) and micro (individual level) systems models to simulate the causal relationship between health states, interventions/policies, and outcomes. Tests used include the Health-Related Quality of Life: 36-item Short Form Health Survey (SF-36 HRQoL) (33), Quality of Life Questionnaire (15D) (34) and 3) Resource Utilization Questionnaire: Resource Use Inventory (RUI) (35) (Supplementary Material).

Study flow

Potentially eligible participants will be pre-screened and assessed according to the inclusion and exclusion criteria. Medical history, brief clinical assessment and a brief cognitive assessment (5-min MoCA) will be conducted remotely. If participants pass these remote procedures, they will be invited on-site to undergo an electrocardiogram and further screening. Participants who pass all screening procedures will undergo full clinical, vascular, cognitive, physical and diet assessment at baseline, month-12 and month-24. Outcome assessments are divided into three sections – on-site, remote and self-administered (Table 1).

Table 1. Assessments conducted during screening and outcome visits

*First 800 participants (n=400 in each group); S = Screening visit, B = Baseline visit, M12 = Month-12 visit, M24 = Month-24 visit, SA = Self-administered assessments, O = On-site visit, R = Remote visit via telephone call. Note. DSM-IV = Diagnostic and Statistical Manual of Mental Disorders-IV, LIBRA = LIfestyle for BRAin health score, CAIDE = Cardiovascular Risk Factors, Aging, and Incidence of Dementia Risk Score, PSQI = Pittsburgh Sleep Quality Index, , MMSE = Mini-Mental State Examination, mNTB = Modified Neuropsychological Test Battery, TMT = Trail Making Test, CDR = Clinical Dementia Rating, GDS = Geriatric Depression Scale, PRMQ = Prospective and Retrospective Memory Questionnaire, HRQoL = Health-related Quality of Life: 36-item short form health survey, 15D = Quality of Life Questionnaire, ADCS-MCI-ADL = ADCS-MCI Activities of Daily Living Inventory, PPT = Physical Performance Test, GPAQ = Global Physical Activity Questionnaire


Digital Platform

In order to improve trial efficiency in a clinical trial environment altered by the COVID-19 pandemic, the SINGER trial will be using a digital platform to facilitate workflow and execution of interventions and outcome assessments. This platform will manage participants’ study activities through a dashboard to track their progress and to obtain an overview of the results and completion status of self-administered outcome measurements. Scheduling of one-to-one therapist-participant appointments, group training sessions, outcome visits and individual training sessions will also be made easier with the platform.
Participants will receive reminders to 1) attend training sessions and study visits, 2) complete their food, cognitive training, and exercise diaries, 3) log their vital signs for BP monitoring, and 4) complete their self-administered outcome measures. De-identified anthropometric measurements and laboratory results taken at the outcome visits will be available to participants. The platform will also facilitate communication between participants and the study team (e.g. discussions with other participants, video calls).
Training resources such as group workshop materials and pre-recorded exercise demonstration videos will be made available on the digital platform. Session notes will be uploaded and made visible to participants after each session. Additionally, the platform will aid in document management and data collection of clinical notes (e.g. clinical record forms), self-administered outcome questionnaires and intervention diaries.

Sample size

The targeted sample size for SINGER is 1200 participants followed over 2 years. According to projections from the SINGER-pilot study, approximately 5000 participants need to be pre-screened to reach the 1200 target. This will provide 80% power at a 5% significance level, assuming 15% drop-out over the follow-up, to detect a mean difference of 0.088 in the 2-year change in the mNTB global z-score composite between the two treatment groups, with a common standard deviation of 0.5 for the within group change. The assumed effect size for SINGER is close to the upper limit of the 95% confidence interval for the FINGER effect size (20), but lower than an earlier multi-domain lifestyle study in Singapore (36).
SINGER will be conducted in the context of several other similar trials in the Worldwide FINGER network, including the original FINGER. Thus, SINGER will not be viewed as a stand-alone assessment of a multidomain intervention, but as an important contributor to a broader assessment of efficacy. Hence the choice to target 80% power also conserves resources and provides a more efficient trial.

Statistical analysis

Primary analyses will be two-tailed and based on the intention-to-treat approach, where data from all participants will be analysed according to their original intervention assignment and full follow-up will be attempted regardless of intervention adherence. Univariate analyses will be performed to determine outliers and skewness of the data. The primary outcome for SINGER is the mNTB global composite cognitive z-score, measured by transforming the raw scores of all individual tests into standardized z-scores using the cohort-wide means and standard deviations (SD) at baseline. Additionally, individual cognitive domain z-scores will be obtained by averaging the cohort-wide composite mean and SD at baseline. Primary and secondary cognitive outcomes will be analysed using linear mixed models, with the dependent variable consisting of all composite outcomes measured from baseline through follow-up. Covariates include site (stratification factor) and clinic visit to control for potentially non-linear factors that may systematically affect both intervention groups. The fixed effects are intervention assignment and its interaction with follow-up time as a continuous variable – this interaction will be tested with one degree of freedom. Models will be fitted with restricted maximum likelihood to adjust for baseline differences among participants. Longitudinal correlations between measures collected over time within individual participants will be parameterized using an unstructured model. If this model for longitudinal covariance results is non-convergent, a first-order autocorrelation model will be used instead. The significance of the intervention will be determined based on a Wald test for the interaction between intervention assignment and time from randomization. Secondary outcomes of composite cognitive functions will be reported using 95% confidence intervals. Similarly, general linear models will evaluate if random assignment to the SGI or SLI group will affect CDR-SoB scores, functional status (ADCS-MCI-ADL), and a composite measure reflecting lifestyle practices involving diet, physical and cognitive/social activity. Furthermore, machine learning to predict future disease progression and intervention response based on multimodal biomarkers will be utilised. Logistic regression and survival analysis will be used for categorical variables.


Challenges and conclusion

Both the FINGER and SINGER interventions have been shown to be feasible in Singaporean older adults (15). However, due to the lack of sample size and short study duration, the pilot study could not determine the efficacy of the interventions. As a result, a large 2-year multi-domain lifestyle intervention RCT will be conducted to evaluate the effects of a multiple lifestyle intervention on cognition in older adults at risk of dementia in Singapore. As a large portion of the study will be conducted over a digital platform, there may potentially be a slow uptake in participants utilising the resources and interventions on the digital platform. Furthermore, some subjects may have difficulty accessing and using the digital platform. To counter this challenge, the study team will be providing each participant a tablet with the digital platform already downloaded on the device. A group workshop will also be held to teach participants how to use the digital device and platforms so that they will be able to access and participate in the interventions. Encouraging learning and socialization, and providing access to devices, may potentially be a motivating factor for participants. The WW-FINGERS international network, which SINGER is a part of, will enable a more harmonious approach to effective non-pharmacological interventions in delaying cognitive decline in older adults at risk of dementia. SINGER will extend FINGER findings in a multi-ethnic Asian population. SINGER is a part of the WW-FINGERS global interdisciplinary network ( which aims to share knowledge and experiences on trials for dementia prevention and risk reduction, harmonize data, and plan joint international initiatives for the prevention of cognitive impairment and dementia (37). SINGER will not only examine the efficacy of a large-scale lifestyle intervention, we will also explore the public health implications and translational pathway of incorporating the interventions into the Ministry of Health’s long-term vision and mission for the management of chronic disease burden nationally. The SINGER study aims to work closely with individuals, communities, public institutions and other partners to construct a framework for promoting healthy longevity.


Conflict of interest: X. Xu has nothing to disclose. K. Chew has nothing to disclose. ZX. Wong has nothing to disclose. A. Phua has nothing to disclose. E. Chong has nothing to disclose. C. Teo has nothing to disclose. N. Sathe has nothing to disclose. YC. Chooi has nothing to disclose. W. Chia has nothing to disclose. C. Henry has nothing to disclose. E. Chew has nothing to disclose. MC. Wang has nothing to disclose. AB. Maier has nothing to disclose. N. Kandiah has nothing to disclose. C. Chen reports grants from the National University of Singapore and National Medical Research Council.

Funding: This study is funded by a National Medical Research Council of Singapore Open Fund Large Collaborative Grant (OF-LCG). The sponsors had no role in the design and conduct of the study; in the collection, analysis, and interpretation of data; in the preparation of the manuscript; or in the review or approval of the manuscript.

Acknowledgements: The authors thank Dr. Miia Kivipelto, Dr. Tiia Ngandu, Dr. Alina Solomon and the FINGER team for their support and guidance throughout the planning of the project. The authors would also like to acknowledge Dr. Francesca Mangialasche, Dr. Emily Meyers, Dr. Heather Snyder, Dr. Laura Baker, Dr. Susan Landau and the US-POINTER team for their help in the development and harmonisation of the study protocol design.

Ethical standards: The SINGER study has been approved by the National Healthcare Group Domain-Specific Review Board and is registered under (ID: NCT05007353). Written informed consent will be obtained from all participants before enrolment into the study. The study will be conducted in accordance with Good Clinical Practice guidelines.





1. Alzheimer’s Disease International. Dementia in the Asia Pacific Region. 2014.
2. Chen C, Xu X, Chew E, Henry CJ, Koo EH, Singapore Intervention Study To Prevent Cognitive Impairment And Disability (Singer) Initiative. Alzheimers Dement 2017; Featured Research Session: F4–09: Global Dementia Prevention Initiative: Applicability of the Multi-Domain Interventions.
3. United Nations General Assembly. Resolution adopted by the General Assembly on 14 December 2020: United Nations Decade of Healthy Ageing (2021–2030). 2020.
4. Ministry of Health Singapore. Speech By Dr Amy Khor, Senior Minister Of State For Health, At The 13th International Congress Of The Asian Society Against Dementia, Held On Thursday 29 August 2019, At The Shangri-La Hotel Singapore. In 13th International Congress Of The Asian Society Against Dementia. 2019. Singapore.
5. Livingston G, Sommerlad A, Orgeta V, et al. Dementia prevention, intervention, and care. The Lancet Commissions 2017;390:2673–734. S0140-6736(17)31363-6
6. Norton S, Matthews FE, Barnes D, Yaffe K, Brayne C. Potential for primary prevention of Alzheimer’s disease: an analysis of population-based data. Lancet Neurology 2014;13:788-794.
7. Lisko I, Kulmala J, Annetorp M, et al. How can dementia and disability be prevented in older adults: where are we today and where are we going? Journal of Internal Medicine 2020;289(6):807-830.
8. Moll van Charante E, Richard E, Eurelings LS, et al. Effectiveness of a 6-year multidomain vascular care intervention to prevent dementia (preDIVA): a cluster-randomised controlled trial. Lancet 2016;388:797–805. S0140-6736(16)30950-3
9. Andrieu S, Guyonnet S, Coley N, et al. Effect of long-term omega 3 polyunsaturated fatty acid supplementation with or without multidomain intervention on cognitive function in elderly adults with memory complaints (MAPT): a randomised, placebo-controlled trial. Lancet Neurology 2017;16:377–89. S1474-4422(17)30040-6
10. Ngandu T, Lehtisalo J, Solomon A, et al. A 2 year multidomain intervention of diet, exercise, cognitive training, and vascular risk monitoring versus control to prevent cognitive decline in at-risk elderly people (FINGER): a randomised controlled trial. The Lancet 2015;385(9984):2255-2263.
11. de Souto Barreto P, Pothier K, et al. A Web-Based Multidomain Lifestyle Intervention for Older Adults: The eMIND Randomized Controlled Trial. The Journal of Prevention of Alzheimer’s Disease (JPAD) 2020;8:142–150.
12. Srisuwan P, Nakawiro D, Chansirikarnjana S, et al. Effects of a Group-Based 8-Week Multicomponent Cognitive Training on Cognition, Mood and Activities of Daily Living among Healthy Older Adults: A One-Year Follow-Up of a Randomized Controlled Trial. The Journal of Prevention of Alzheimer’s Disease (JPAD) 2019;7(2):112-121.
13. Kivipelto M, Mangialasche F, Snyder HM, et al. World-Wide FINGERS Network: A global approach to risk reduction and prevention of dementia. Alzheimers Dement 2020;16(7):1078-1094.
14. Rosenberg A, Mangialasche F, Ngandu T, Solomon A & Kivipelto M. Multidomain Interventions to Prevent Cognitive Impairment, Alzheimer’s Disease, and Dementia: From FINGER to World-Wide FINGERS. The Journal of Prevention of Alzheimer’s Disease (JPAD) 2019;7(1):29-36.
15. Chew KA, Xu X, Siongco P, et al. SINgapore GERiatric intervention study to reduce physical frailty and cognitive decline (SINGER)-pilot: A feasibility study. Alzheimers Dement TRCI 2021;7(1):e12141.
16. World Health Organization. A healthy lifestyle. 2021.
17. National Nutrition Council. Finnish nutrition recommendations – Diet and physical activity in balance. 2005, Edita: Oy, Finland.
18. Renzella J, Townsend N, Jewell J, et al. What national and subnational interventions and policies based on Mediterranean and Nordic diets are recommended or implemented in the WHO European Region, and is there evidence of effectiveness in reducing noncommunicable diseases? 2018. WHO Regional Office for Europe: Copenhagen.
19. Health Promotion Board. Health Promotion Board Introduces My Healthy Plate to Inculcate Healthy Eating Habits amongst Singaporeans. 2014.
20. Kivipelto M, Solomon A, Ahtiluoto S, et al. The Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER): study design and progress. Alzheimers Dement 2013;9(6):657-65.
21. Ministry of Health Singapore. Clinical Practice Guidelines (Medical). 2017.
22. The SPRINT Research Group. A Randomized Trial of Intensive versus Standard Blood-Pressure Control. N Engl J Med 2015;373:2103-2116.
23. Shen JHQ, Shen Q, Yu H, et al. Validation of an Alzheimer’s disease assessment battery in Asian participants with mild to moderate Alzheimer’s disease. Am J Neurodegener Dis 2014;3(3): p. 158.
24. Harrison J, Minassian SL, Jenkins L, et al. A neuropsychological test battery for use in Alzheimer disease clinical trials. Arch Neurol. 2007;64(9):1323–9.
25. Morris JC. The Clinical Dementia Rating (CDR): current version and scoring rules. Neurology 1993;43:2412-4.
26. Pedrosa H, De Sa A, Guerreiro M, et al. Functional evaluation distinguishes MCI patients from healthy elderly people—the ADCS/MCI/ADL scale. J Nutr Health Aging 2010;14(8):703–9.
27. Snowdon J. Validity of the geriatric depression scale. Am Geriatr Soc. 1990;38(6):722–3.
28. Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. Pittsburgh sleep quality index (PSQI). Psychiatry Res. 1989;28(2):193–213.
29. Smith G, Del Sala S, Logie RH, Maylor EA. Prospective and retrospective memory in normal ageing and dementia: A questionnaire study. Memory 2000;8(5):311–21.
30. Bull FC, Maslin TS, Armstrong T. Global Physical Activity Questionnaire (GPAQ): Nine Country Reliability and Validity Study. Journal of Physical Activity and Health 2009;6(6):790-804.
31. Karp A, Paillard-Borg S, Hui-Xin W, et al. Mental, physical and social components in leisure activities equally contribute to decrease dementia risk. Dementia and Geriatric Cognitive Disorders 2006;21(2):65-73.
32. Delbaere K, Van den Noortgate N, Bourgois J, Vanderstraeten G, Tine W, Cambier D. The Physical Performance Test as a predictor of frequent fallers: a prospective community-based cohort study. Clin Rehabil. 2006;20(1):83-90.
33. McHorney CA, Ware Jr JE, Raczek AE. The MOS 36-Item Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care. 1993;31(3): p. 247–63.
34. Sintonen H. The 15D instrument of health-related quality of life: properties and applications. Ann Med. 2001;33(5): p. 328–36.
35. Sano M, Zhu CW, Whitehouse PJ, et al. ADCS Prevention Instrument Project: pharmacoeconomics: assessing health-related resource use among healthy elderly. Alzheimer Dis Assoc Disord 2006;20:S191.
36. Ng TP, Ling LHA, Feng L, et al. Cognitive Effects of Multi-Domain Interventions Among Pre-Frail and Frail Community-Living Older Persons: Randomized Controlled Trial. Journals of Gerontology: Medical Sciences 2018;73:806–812.
37. World-Wide FINGERS. A global network of clinical trials for prevention and risk reduction of dementia. 2021.



C.W. Ritchie1,2,3, J.M.J. Waymont2,4, C. Pennington1,2,5, K. Draper2, A. Borthwick2, N. Fullerton2, M. Chantler6, M.E. Porteous1,3, S.O. Danso1, A. Green1, L. McWhirter1, G. Muniz Terrera1, S. Simpson7, G. Thompson1, D. Trépel8,9, T.J. Quinn7, A. Kilgour1,2


1. University of Edinburgh, 2. Brain Health Scotland, 3. NHS Lothian, 4. University of Aberdeen, 5. NHS Forth Valley, 6. Herriot Watt University, 7. University of Glasgow, 8. University of Dublin Trinity College, 9. Global Brain Health Institute, United Kingdom

Corresponding Author: Prof. Craig Ritchie, University of Edinburgh, United Kingdom,



In order to address the oft-cited societal, economic, and health and social care impacts of neurodegenerative diseases, such as Alzheimer’s disease, we must move decisively from reactive to proactive clinical practice and to embed evidence-based brain health education throughout society. Most disease processes can be at least partially prevented, slowed, or reversed. We have long neglected to intervene in neurodegenerative disease processes, largely due to a misconception that their predominant symptom – cognitive decline – is a normal, age-related process, but also due to a lack of multi-disciplinary collaboration. We now understand that there are modifiable risk factors for neurodegenerative diseases, that successful management of common comorbidities (such as diabetes and hypertension) can reduce the incidence of neurodegenerative disease, and that disease processes begin (and, crucially, can be detected, reduced, and delayed, prevented, or treated) decades earlier in life than had previously been appreciated. Brain Health Scotland, established by Scottish Government and working in partnership with Alzheimer Scotland, propose far-reaching public health and clinical practice approaches to reduce neurodegenerative disease incidence. Focusing here on Brain Health Scotland’s clinical offerings, we present the Scottish Model for Brain Health Services. To our knowledge, the Scottish Model for Brain Health, built on foundations of personalised risk profiling, targeted risk reduction and prevention, early disease detection, equity of access, and harnessing comprehensive data to assist in clinical decision-making, marks the first example of a nationwide approach to overhauling clinical, societal, and political approaches to the prevention, assessment, and treatment of neurodegenerative disease.

Key words: Brain health, neurodegeneration, biomarkers, Alzheimer’s disease, functional cognitive disorders.

Abbreviations: AD: Alzheimer’s Disease; APOE: Apolipoprotein E; BP: Blood pressure; CSF: Cerebrospinal fluid; FCD: Functional Cognitive Disorders; MRI: Magnetic resonance imaging; NFL: Neurofilament light; OR: Odds ratio; PAF: Population-attributable factor; Ptau: Phosphorylated tau; Ttau: Total tau.



The clinical, societal, and economic impact of neurodegenerative disease is vast and continues to grow. There are currently estimated to be over 50 million people worldwide living with Alzheimer’s disease (AD), the most common form of neurodegenerative disease giving rise to a clinical dementia syndrome, with the number of people living with AD predicted to double every twenty years (1). While current treatment options lack efficacy in reducing cognitive and functional decline associated with later-stage neurodegenerative disease (likely compounded by diagnosis often not occurring until late-stage disease), prevention and intervention are our best options to mitigate the negative impact of the anticipated rise in neurodegenerative disease prevalence. Age is the greatest risk factor for neurodegenerative disease and clinical dementia syndromes. Where life expectancy grows, so, too, does the urgency to prevent neurodegenerative disease and comorbidities, and to take decisive action to improve brain health at the earliest opportunity, from the population to the individual.
The growing consensus of the need to begin detection, prevention, and intervention earlier in the disease process arose from an accumulation of evidence from three areas: [1] modifiable risk factors for cognitive and functional decline across the life-course (2, 3), [2] evidence that pathophysiological processes underlying neurodegenerative disease begin many years before the emergence of a clinical dementia syndrome (4), and [3] from the recognised need for further development of and preparation for disease-modifying therapies and multimodal interventions for use earlier in the disease process than is currently common practice (5, 6). A recent review in the Journal of Prevention of Alzheimer’s Disease outlines thorough, practical recommendations for early detection of AD in modern clinical practice (7), providing tangible examples of how the presently proposed Scottish Brain Health Service Model may be implemented at the individual level. Further, a recent ‘A User Manual for Brain Health Services’ series (Alzheimer’s Research & Therapy, 2021, vol. 13) evidences the growing consensus of the need to move to preventative, proactive clinical practice, complementing currently existing services available to those in later-stages of neurodegenerative disease (8).
Risk profiling, reduction of modifiable risk factors, and early detection and intervention are standard clinical practice in the management of other diseases, including highly related diseases such as cardiometabolic disease. Clinicians are aware of and routinely assess for indicators of cardiometabolic risk, such as blood pressure (BP) and body mass index (BMI). Secondary prevention and intervention of concurrent conditions, such as type 2 diabetes mellitus and hypercholesterolemia, is routine practice. Risk profiling and increased focus on early disease detection have also been used to develop risk prediction algorithms, which have been successfully implemented to support clinical decision making (e.g., ‘ASSIGN’ score and ‘QRISK’ algorithms for the evaluation of future cardiovascular disease risk (9, 10)). Further, the use of risk profiling, early detection methods, and the development of risk prediction algorithms is already routine in large-scale observational cohort studies of neurodegenerative diseases, particularly of AD. Summaries of notable examples of such studies are presented in Table 1. Additionally, large-scale interdisciplinary research projects, such as the Occitanie Toulouse ‘INSPIRE’ project, continue to explore biomarkers of healthy ageing from animal and human models, with the ambition of identifying a composite biomarker to identify declines in intrinsic capacity (i.e., physiological, mental, and psychological capacity) and functional ability at the earliest possible stage (11). Each of these aforementioned research projects align with the WHO guidelines for Integrated Care for Older People (ICOPE) (12).

Table 1. Examples of large-scale observational cohort studies for neurodegenerative diseases

Note. CSF: cerebrospinal fluid; AD: Alzheimer’s disease.


The technology and infrastructure exist to integrate risk prediction and early disease detection in research settings, but have not yet been meaningfully translated into clinical practice. There is a clear and pressing need to provide effective brain health care pathways which are distinct from those in place for patients presenting with a clinical dementia syndrome or other presentations indicative of later-stage neurodegenerative disease. Existing ‘memory clinic’ and dementia services are optimised for symptomatic dementia diagnosis and post-diagnostic care. Similarly, while the general public are increasingly aware of neurodegenerative disease and dementia in later life, there remains a pervasive misconception that these diseases and their associated cognitive and functional decline are to be expected in normal ageing (13). Public health approaches to promoting brain health across the life-course are in their infancy. To address the need to move research into practice and public health, Brain Health Scotland was established by Scottish Government in 2020, in partnership with Alzheimer Scotland, to take public health and clinical approaches to reduce incident dementia in Scotland (see Figure 1).

Figure 1. Pyramid of approaches to reduce incident dementia in Scotland, from public health interventions for the Scottish population (bottom tier) to clinical Brain Health Services for the individual (top tier)


The proposed clinical approach of Brain Health Scotland’s brain health services is based on three fundamental principles, which are complementary to but designed to be distinct from the existing ‘memory clinic’ model for later-stage disease: [1] individual risk profiling, [2] early disease detection, and [3] implementation of personalised prevention plans (19). As up to 70% of all cases of a dementia syndrome will have an aetiological contribution from AD pathology (20), the care pathway for brain health services is optimised for, although not monopolised by, the detection of and risk profiling for AD.
The symbiosis between patients, clinicians, community-based clinics, data systems and research facilities forms the foundation of brain health services. Through the integration of these groups and activities, we can generate and continuously update risk prediction algorithms, providing feedback to support clinical decision making and implementation of personalised prevention plans.
Through public health interventions (Figure 1) and brain health services (Figure 2) combined, Brain Health Scotland aims to reduce the incidence of dementia in Scotland, and to ameliorate neurodegenerative disease processes much earlier in their course than is achievable with current approaches. Here, we outline the domains to be assessed and the rationale and scientific basis for them which will allow completion of the first two objectives (risk profiling and early disease detection), with implementation of personalised prevention plans (including proposals for interventions and secondary prevention approaches) to be detailed elsewhere in due course. The following is segmented into sections relating to [1] risk profiling, [2] early disease detection, and [3] general considerations around data systems, equity of access, differential diagnosis, communication, and health economics.

Figure 2. Care pathway for the Scottish model of Brain Health Services

Stage 1: generic, non-clinical support (advice, light-touch lifestyle assessment, information and signposting). Stage 2: initial clinical service (risk profiling, early disease detection, personalised prevention. Parallel referral to external services for management of comorbidities where appropriate). Stage 3: specialised clinical service (brain biomarker assessment, personalised prevention and intervention. Outwards referral to memory clinic for those with an established clinical dementia syndrome unlikely to benefit from continued care in Brain Health Services, parallel referral to external services for comorbidity management where appropriate). SBHR – Scottish Brain Health Register; CSF – cerebrospinal fluid; PDS – Post Diagnostic Support


Risk profiling

Genetic Risk

The apolipoprotein E (APOE) gene is among the most well-documented monogenic risk factors in relation to AD, with the ε4 allele being associated with increased risk of AD and the ε2 allele being a potential protective factor. A recent large neuropathologic study found individuals with ε4/ε4 genotype had an odds ratio (OR) of 11.39 (Confidence Intervals: 9.96-13.02) for developing AD and individuals with the ε2/ε2 genotype having an OR of 0.35 (CI: 0.2-0.61) for developing AD, compared to those with the more common ε3/ε3 genotype, after controlling for age, sex, and autopsy findings (with ORs greater than 1 indicating elevated odds, and ORs below 1 indicating reduced odds) (21). Further, it has been reported that allelic expression is not only related to susceptibility of developing AD, but that carriers of the ε4 allele with AD accrue greater tau accumulation and medial temporal lobe atrophy than those with AD but without the ε4 allele, among other identified heterogeneities, indicating APOE variant confers phenotypic differences in disease progression and clinical expression (22).
Genetic testing is routinely used in determining risk for hereditary disorders (e.g., Huntington’s disease) and in diseases associated with known genetic mutations (e.g., testing for mutations on the BRCA1 and BRCA2 genes associated with breast and ovarian cancers). It is important to note that, in brain health services, genetic testing for the APOE variant will be used for the determination of personal risk (i.e., there is no intention of determining APOE variant as a screening tool, rather as a method of understanding attributable genetic risk). Understanding risk attributable to APOE variant is a crucial component in determining personal risk, and allows for a more accurate determination of the proportion of potentially modifiable risk. Further considerations relating to risk disclosure are discussed later (see Section 5: Risk Disclosure).

Lifestyle risk

Lifestyle, defined by the World Health Organisation as “a way of living based on identifiable patterns of behaviour, which are determined by the interplay between an individual’s personal characteristics, social interactions, and socioeconomic and environmental living conditions”, plays a significant role in brain health and the risk for neurodegenerative disease. The 2020 Lancet Commission on dementia prevention, intervention, and care (3) lists twelve risk factors for dementia across the life-course which, if eradicated at the population level, could reduce worldwide prevalence of dementia by up to 40%. The majority of modifiable risk factors identified in the Lancet 2020 report consist of lifestyle factors or lifestyle-associated comorbidities such as diabetes (discussed further below). Each risk factor identified in the Lancet 2020 report is assigned a population attributable fraction (PAF) indicating the percentage reduction in worldwide dementia prevalence if the risk factor is eliminated entirely. In early life (<45 years), less education is the most important risk factor (PAF = 7%). In midlife (45-65 years), lifestyle risk factors include excessive alcohol consumption (>21 units/week; PAF = 1%), hypertension (PAF = 2%), and obesity (PAF = 1%). Additional factors which may be influenced by lifestyle include hearing loss (PAF = 8%) and traumatic brain injury (PAF = 3%). In later-life (>65 years), significant modifiable lifestyle factors include smoking (PAF = 5%), social isolation (PAF = 4%), physical inactivity (PAF = 2%), and exposure to air pollution across the life-course (PAF = 2%).
These modifiable risk factors, among other lifestyle factors associated with brain health (e.g., diet (23) and sleep (24)), are easily quantifiable through standard clinical measurements (e.g., BP, BMI, glycated haemoglobin) and questionnaires (e.g., physical activity levels, smoker status). Patients’ risk factor data (e.g., total modifiable risk, risk scores for individual factors, combined non-/modifiable risk) will be reassessed and updated at each visit to a Brain Health Service, enabling the clinician and patient to track longitudinal changes in individual risk factors and projected disease trajectory (for those in whom early disease is identified), and to evaluate and re-evaluate the suitability and efficacy of personalised prevention plans.

Medical comorbidities and iatrogenic effects of drugs

Multi-morbidity is increasingly prevalent, yet much of the evidence describing AD risk investigates each risk factor separately. Evidence is now emerging that multimorbidity confers a cumulative risk, with neuropsychiatric and cardiovascular disease, sensory impairment, and cancer all being highlighted as notable risk factors (25). Furthermore, multi-morbidity is associated with polypharmacy (the use of ≥ 5 daily medications), which has been associated with incident dementia (26).
Key medical comorbidities that need to be screened for and, where identified or pre-existing, added to risk prediction algorithms include diabetes (27), atrial fibrillation (28), heart failure (29, 30), hypertension (31), respiratory disease (32), and traumatic brain injury (33, 34).
Anticholinergic medication has been investigated as a potential risk factor for dementia, but the evidence is mixed (35, 36). It may be that higher anticholinergic burden and male sex increases the risk (37). There is also mixed evidence about whether use of benzodiazepines and Z-drugs increase the risk of developing dementia, or whether studies are detecting evidence of reverse causation (i.e., prodromal dementia causing sleep disturbance which led to hypnotic prescription) (36). Moreover, these drugs may drive symptoms of dementia (like confusion) as opposed to actually driving disease processes (e.g., amyloidosis).

Mental health risk

Depression and anxiety have been associated with dementia risk in numerous epidemiological studies with biological plausibility for this association focussing on dysregulation of the hypothalamic-pituitary-adrenal axis (38). Mental health conditions may mediate social isolation and socioeconomic decline and therein the secondary accumulation of other risk factors both medical and lifestyle. The association with mood and anxiety symptoms is also subject to reverse causality as, while these conditions may act as a risk factor earlier in life, they may also act as an early expression of disease. Irrespective of this risk/symptom dilemma, these conditions should be sought for and managed to improve prognosis and well-being.


Modalities for early disease detection and capturing disease expression

Disease detection


Neuroimaging methods, including computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET), and single-photon emission tomography (SPECT) have provided unparalleled insights into in vivo brain health across the life course, and have provided crucial evidence that neurodegenerative pathophysiologic processes begin long before their clinical manifestation as cognitive and functional decline.
Structural brain MRI scans (e.g., T1 MRI sequences) can provide valuable insights into baseline and longitudinal brain volumes, with hippocampal, temporal lobe, and ventricular volumes of particular relevance in early AD detection (39). Fluid attenuated inversion recovery (FLAIR), T2, and T2* (susceptibility-weighted) MR imaging protocols can be used to identify even subtle evidence of vascular disease, including cerebral small vessel disease, which is a highly prevalent brain imaging finding in adults, commonly accompanying neurodegenerative processes and acting as a leading contributor to vascular dementia (40).
PET imaging enables investigation of cortical metabolic function through the use of radiotracers. Of particular importance in the early detection of AD are measures of beta-amyloid (aβ) burden (e.g., PiB) and Tau burden (e.g., T807), both of which have been demonstrated to improve diagnostic confidence (41–43). PET imaging of cerebral glucose metabolism (e.g., Fluorodeoxyglucose-PET) is also widely used to assess hypometabolism, which is indicative of neurodegenerative disease more broadly (44).
These imaging methods are routinely included in clinical and research neuroimaging practice and there exist numerous medical image analysis pipelines for the quantification of neurodegenerative disease processes which provide clear, interpretable information on brain health, which can also be incorporated into risk prediction algorithms.

Cerebrospinal fluid and blood biomarkers

Cerebrospinal fluid (CSF) biomarkers for aiding in early AD detection have advanced considerably in recent years. There is significant evidence in favour of using CSF levels of phosphorylated tau (Ptau), total tau (Ttau), and neurofilament light (NFL) as biomarkers for preclinical AD detection (45). In isolation of other risk factor analysis, CSF biomarkers yield reasonable sensitivity but poor specificity for predicting progression from pre-clinical AD to a clinical dementia syndrome, highlighting the need to embed such biomarker assessments in multifaceted risk algorithms (46–48). The extraction of CSF through lumbar puncture (LP) is an invasive procedure and will not be suitable for all patients, but its benefits should not be overlooked.
Blood biomarkers, which are less invasive to obtain than CSF biomarkers and more accessible than CSF or neuroimaging biomarkers, are currently less widely accepted than CSF or imaging markers as being clinically effective. Blood biomarkers under particular scrutiny include plasma and serum NFL level, plasma aβ42/40 ratio, and plasma Ptau and Ttau levels. Zetterberg and colleagues (49) note that while plasma and serum NFL are good predictors of AD, they are not disease-specific. Similarly, the authors comment that aβ42/40 ratio is a strong predictor of brain amyloid burden, but this is also not specific to AD (and may be more indicative of amyloidosis, which is a prevalent feature of brain ageing). Plasma Ptau and Ttau levels are put forward as promising, AD-specific measurements, but they require further development and validation (49).

Additional and emerging biomarkers

Brain health services will provide a natural environment for the development and validation of further biomarkers and measures of early disease progress. The symbiosis of clinical and research activities will allow for efficient integration of well-established research methods and disease detection (such as electroencephalography – EEG (50)) and novel and emerging approaches (such as artificial intelligence in speech and language processing (51)) into clinical practice. This symbiosis is a key driver of the Scottish Brain Health and Dementia Research Strategy, launched in July 2021 (52).

Disease expression

We anticipate that the majority of patients accessing brain health services will not present with symptoms associated with later-stage disease (i.e., significant cognitive and functional decline). Cognitive and functional measures used in later-stage neurodegenerative disease will not necessarily be appropriate or suitably sensitive for our purposes. Here, we describe aspects of early disease expression, which are often subtle and for which further development of prodromal disease assessment methods are required.


Cognitive changes detectable in preclinical AD include poorer performance in tasks involving visuospatial processing, assessable through measures such as the Four Mountains test, which is sensitive to changes in allocentric spatial processing and indicative of change to hippocampal structure beginning in midlife (53, 54). When assessing cognitive performance at any stage of the disease-course, other factors potentially contributing to poorer performance on these tests should be considered (e.g., less education, anxiety, delirium). The transition proposed from considering neurodegenerative diseases as brain diseases with cognitive symptoms, rather than ‘cognitive disorders’, is critical to the development of brain health services. It is likely that many, if not the majority, accessing brain health services will have no cognitive symptoms observable using traditional measures such as the Mini Mental State Examination (MMSE) or Clinical Dementia Rating (CDR), which are prone to ceiling effects and are optimised for populations with overt clinical symptoms.

Behavioural and neuropsychiatric

Although sleep was not included in the 2020 Lancet Commission Report (3), there is clear and consistent evidence linking poor quality sleep with risk of cognitive decline (55). The brain in sleep is considered to be an optimised state for clearance of toxic proteins and other waste products via the glymphatic system, highlighting the potential association between poor sleep and amyloidosis (56), and is likely a symptom of neurodegenerative disease, demonstrating a bidirectional relationship similar to that of depression and cognitive decline.
Apathy has also been associated as an early symptom of neurodegenerative disease (including Parkinson’s disease (57)) and a predictor of conversion from ‘Mild Cognitive Impairment’ (MCI, early clinical disease phase) to an established dementia syndrome (later clinical disease phase) (58, 59). Collectively, neuropsychiatric and behavioural symptoms have been collected into the concept of Mild Behavioural Impairment (MBI), which is considered to be a separate and perhaps earlier indicator of neurodegeneration, worthy of inclusion in Brain Health Service assessment (60).

Gait/power and autonomic instability

Gait speed is known to reach a peak in the third decade and decline thereafter, not becoming apparent until older age (61). Slower gait speed has been found to predict incident dementia by up to seven years in older adults (62). Even in midlife, gait speed has been shown to be associated with white matter hyperintensity accumulation and cognitive decline (63). While a single measure of gait speed could be used as a predictive tool for neurodegenerative disease risk, longitudinal measurements are likely to be more useful. Similarly, weaker grip strength in midlife is associated with poorer cognitive ability, and several studies have found an association between declining grip strength in older adults and the risk of incipient dementia (64).
Autonomic dysfunction can feature as a part of any neurodegenerative disease, most commonly in Lewy body disorders, while evidence of autonomic dysfunction prevalence in AD and vascular dementia is conflicting (65). There is evidence of sympathetic dysfunction and cognitive impairment in older adults without an apparent clinical dementia syndrome (66), indicating that detection of autonomic dysfunction in older patients without dementia could represent early disease expression of neurodegenerative disorders.


Analytical approaches and data use

Risk algorithms

An individual’s risk for neurodegenerative disease can be estimated through the use of prediction algorithms and machine learning to support clinical decision making. Examples of automated or semi-automated risk prediction methods routinely used in other areas of clinical practice include the ASSIGN cardiovascular risk assessment tool (10), which is used in primary care in Scotland to estimate ten-year risk of adverse cardiovascular events in asymptomatic individuals with no clinical evidence of cardiovascular disease, and the QCancer ten-year risk tool (67) which calculates the absolute risk of an asymptomatic individual having a yet undiagnosed cancer. These online tools are only as accurate as their source data and, as such, are updated on a regular basis to reflect newly identified risk factors and changing population characteristics. Linkage of patient data between hospital and GP records and centrally held information (e.g., death certification) will allow more accurate projections than relying on information from fewer sources. The veracity of the data being collected to deliver risk prediction algorithms for the onset or progression of neurodegenerative diseases is a fundamental factor and drives the comprehensive range of data being collected, as above, on risk factors, biomarkers, and expression of disease. The data pathway will be honed over the first few years of operation, but we initially intend to include all data collected in Brain Health Services (risk factors, biomarkers, disease expression) and relevant data made available through data linkage derived from patients’ interactions with NHS services. The quantity and exact nature of data required will alter over time as the algorithm(s) and developers identify the core minimum dataset required for accurate prediction of disease progression. Output from the risk prediction and disease progression algorithm(s) will be fed back to clinicians in Brain Health Services to act as a support tool in diagnosis and disease management.

Secondary research environment

As alluded to above, brain health services will generate a vast amount of data, particularly around early- and mid-life factors associated with neurodegenerative disease. Secondary analyses of data naturally and routinely generated within brain health services, with the explicit consent of clinic patients (and anonymised to adhere with data security protocols), will provide highly valuable epidemiology data enabling further understanding of the origins of neurodegenerative diseases and enabling assessment of the efficacy of treatments and interventions in this naturally occurring Phase IV setting. Moreover, highly characterised patients, where willing, can be offered opportunities to enter clinical trials, where low screen failure rates can be anticipated – a key issue affecting trial delivery in AD. Clinically-based recruitment has been shown to be far more effective than population, register-based recruitment (68).


Differential diagnosis

While brain health services will be optimised for the early detection of neurodegenerative disease, we will inevitably detect functional and cognitive changes in patients which are attributable to other disease processes (e.g., infection, mental illness, delirium), and we will uncover incidental findings in patients who appear otherwise healthy (e.g., an incidental neuroimaging finding of a tumour). In cases where disease processes or incidental findings suggest no concurrent neurodegenerative disease and specialist services are required (e.g., oncology, neurosurgery), patients will be referred out of brain health services to the appropriate service. There may be other circumstances in which it is appropriate for brain health services to support a patient through referral into other services while maintaining their involvement in brain health service pathways (e.g., where a patient has increased risk or early indicators of neurodegenerative disease and a comorbidity requiring additional, specialist treatment).
We are working on developing a Red/Amber/Green (RAG) guideline list of conditions to aid decision making that will be subject to change as the pathway develops on the basis of experience and multi-disciplinary discussions (see Figure 3). Red conditions are those where management takes place entirely outside a Brain Health Service (e.g., brain tumour, significant mental illness including suicidal ideation and psychosis). Amber conditions are those where management may be shared with other clinical services (e.g., alcohol use disorders, Multiple Sclerosis, Motor Neurone Disease). Green conditions are neurodegenerative and cerebrovascular diseases. As services develop (both within and without the Brain Health Service), this RAG categorisation will be subject to review and change.

Figure 3. Red/Amber/Green (RAG) chart providing examples of which conditions would be treated within brain health services (green), conditions which would be managed within brain health services alongside additional specialist care (amber), and conditions which would be referred out of brain health services (red)


Functional Cognitive Disorders

Functional cognitive disorders (FCD) are conditions in which cognitive symptoms are experienced as the result of internally-inconsistent changes in attentional functions and metacognition, and not as the result of neurodegenerative disease or structural brain injury (69). They account for up to one in four memory clinic presentations, but are more frequent in attendees who have self-referred or are younger than 65 years of age (70). Sometimes unhelpfully described as the ‘worried well’, people with FCD, in contrast, have more anxiety, depression, poor sleep, and somatic symptoms than those with other causes of cognitive symptoms (70) (and McWhirter et al., 2021, under review). Clinical examination integrating consideration of linguistic and interactive features can support a positive diagnosis of FCD (71). Ongoing research aims to better understand longitudinal FCD trajectories, mechanisms, and to develop effective and scalable interventions. Low level psychological intervention alongside personal risk reductions plans may be appropriate in many cases; in others, specialist psychological or psychiatric follow-up may be required.


Risk disclosure

In developing individual risk profiles, careful consideration will need to be given to the impact of communicating the results to the recipient. A systematic review of the psychological, behavioural, and social effects of disclosing AD biomarker status found there was no increase in depression or anxiety in those to whom it was disclosed that they were carriers of the APOE ε4 variant (72). However, there was an uptake in long-term care insurance and positive health-related behaviour compared to those who tested negative for the ε4 allele. Nearly all of the subjects of studies reported in this review were first-degree relatives of someone with AD, therefore any detrimental impact of genetic testing may be mediated by the individual already perceiving themselves to be at increased risk.
In brain health services, patients will be encouraged to play an active role in the risk disclosure process. Patients will be made aware of what each test they undertake could indicate, and of the immediate and longer-term implications of test results. Ultimately, patients should be able to choose how much information they receive and how much remains known to the clinicians but is not disclosed to the patient. For example, a patient may wish to know the results of brain imaging, but may not wish to know whether they are an APOE ε4 carrier. Patients’ wishes should be accommodated, and patients should be provided with the best information available for them to decide their preferred extent of risk disclosure and testing.
Further, all patients will be provided with a personalised risk prevention and reduction plan. Again, this will be developed in partnership with the patient and be reflective of their own personal goals (73). When disclosing risk to a patient, emphasis should be placed on modifiable risk and on the positive actions that can be taken to reduce risk over time. Patients should be reassured that they will be offered follow-up appointments and will be supported in implementing their risk reduction and prevention plans. As stated previously, the rationale and implementation of personalised prevention plans and secondary prevention and intervention approaches in Brain Health Scotland’s clinical services will be elaborated upon in future publications.


Health economics

In moving towards prevention and early intervention of neurodegenerative disease, ultimately empowering individuals to live a greater proportion of life in better health, it is possible that we can alleviate some of the economic burden arising from a growing proportion of the population requiring advanced care. This is demonstrated in research indicating the increasing costs of dementia care over time (74), and in studies suggesting that efforts to prevent or intervene in specific risk factors offer the most cost-effective approach to dementia care (75). Of particular note is a recent study which found that NHS in England would save £1.863 billion per year (based on 2012/13 prices) in formal (health and social) and informal care with the successful implementation of interventions for hearing loss, hypertension, and smoking (76). Data collection to feed into ongoing health and economic analyses and reporting will form a key part of our brain health services.


Equity of access

Equity of access is a fundamental principle underpinning all of Brain Health Scotland’s activities. Many of the risk factors associated with poorer brain health accumulate and are exacerbated within underrepresented populations, and too many underserved members of the population struggle to access healthcare. In Scotland, there are also vast differences in accessibility to health services for those in rural and remote areas compared with those in better-connected towns and cities. It is hoped that through increasing accessibility of services, we can improve access to research programmes, enabling the science our clinics is based on to better represent the populations we serve. Further information on our approach to equity of access will be presented elsewhere, but includes co-design of services with marginalised communities, provision of mobile/roving brain health services, digital and telehealth services, and active and ongoing evaluation of the impact of our services within underrepresented groups.


Demonstrator sites

‘Demonstrator sites’ – Brain Health Scotland’s clinical services, cultivated to provide models of best practice and as hubs for supporting continued professional development – are currently being identified and established in partnership with Scottish Government, NHS Scotland, and Alzheimer Scotland. Brain health services will routinely undergo quality improvement assessments, with data from these exercises in initial demonstrator sites to provide insights for further development before a wider roll-out of brain health services for Scotland. This national availability is in the Scottish Programme for Government, to be completed by 2025.



It is time for research evidence to be transferred into routine clinical practice. Evidence that neurodegenerative disease processes begin many years before disease expression, particularly in diseases that give rise to a clinical dementia syndrome, is widely accepted, and methods to detect neurodegenerative disease and to reduce risk factors at the earliest opportunity are ready for clinical implementation. This is of particular importance for patients who currently fall through the gaps, such as those who may have early clinical symptoms, but their case is not severe enough to warrant referral to or ongoing engagement with memory clinics or specialist dementia care providers. Through the combined effort of brain health clinicians and academic research programmes in Scotland, we are well positioned to implement the Scottish Model of Brain Health clinical services for early disease detection, intervention, and personalised prevention, and will soon be prepared to demonstrate the impact these brain health services can have on reducing the incidence of late-stage neurodegenerative disease in the Scottish population.


Funding: Brain Health Scotland, who are overseeing the development of Brain Health Services within NHS Scotland, are funded through a grant from the Scottish Government. Brain Health Scotland is hosted legally within Alzheimer Scotland who received this grant.

Conflicts of Interest: CWR is the director of Brain Health Scotland and receives consultancy fees for this role. JMJW, AB, and NF are employed by Brain Health Scotland (Alzheimer Scotland). The following authors wish to declare additional conflicts of interest and funding received for work unrelated to the present manuscript: CWR has received consultancy fees (Biogen, Eisai, MSD, Actinogen, Roche, Eli Lilly), speaker fees (Eisai, Roche), sits on an NIHR data safety monitoring board (DSMB), and is the unpaid chair of the Brain Health Clinic Consortium and the Scottish Dementia Research Consortium. JMJW received salary support through an education grant provided to Brain Health Scotland from Biogen, and has previously received studentship funding from TauRx Pharmaceuticals. CP receives funding from the Scottish Government’s Chief Scientist Office through a Career Researcher Fellowship. LM received funding from Baillie Gifford, currently receives grant/contract funding from the Scottish Government Chief Scientist Office, is a director of a limited personal services company providing independent medical testimony in court cases, sits on the editorial board of the British Journal of Psychiatry, and on the board of directors of the British Neuropsychiatry Association. GT has received consulting fees from QMENTA. DT has received grant/contract funding from the Public Health Agency Northern Ireland, the Irish Research Council, the Health Research Board, HealthLat, the National Health and Medical Research Council, and the Global Brain Health Institute. TJQ has received grant/contract funding from NIHR, Scottish Government’s Chief Scientist Office, ESPRC, Stroke Association, Dunhill Medical, and sits on the DSMB for Novartis. The remaining authors (MEP, SOD, SS, and AK) have no funding or conflicts of interest to declare.

Ethical standard: The present manuscript is a review of the literature and proposal for a novel care pathway – this required no ethics review.

Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits use, duplication, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.



1. Prince M, Wimo A, Guerchet M, et al. World Alzheimer Report 2015 The Global Impact of Dementia An AnalysIs of Prevalence, Incidence, Cost and Trends.; 2015. Accessed January 25, 2021.
2. Livingston G, Sommerlad A, Orgeta V, et al. Dementia prevention, intervention, and care. Lancet. 2017;390(10113):2673-2734. doi:10.1016/S0140-6736(17)31363-6
3. Livingston G, Huntley J, Sommerlad A, et al. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet. 2020;396(10248):413-446. doi:10.1016/S0140-6736(20)30367-6
4. Sperling RA, Aisen PS, Beckett LA, et al. Toward defining the preclinical stages of Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimer’s Dement. 2011;7(3):280-292. doi:10.1016/j.jalz.2011.03.003
5. Ritchie CW, Russ TC, Banerjee S, et al. The Edinburgh Consensus: Preparing for the advent of disease-modifying therapies for Alzheimer’s disease. Alzheimer’s Res Ther. 2017;9(1):1-7. doi:10.1186/s13195-017-0312-4
6. Yu JT, Xu W, Tan CC, et al. Evidence-based prevention of Alzheimer’s disease: Systematic review and meta-analysis of 243 observational prospective studies and 153 randomised controlled trials. J Neurol Neurosurg Psychiatry. 2020;91(11):1201-1209. doi:10.1136/jnnp-2019-321913
7. Porsteinsson A, Isaacson R, Knox S, Sabbagh M, Rubino I. Diagnosis of Early Alzheimer’s Disease: Clinical Practice in 2021. J Prev Alzheimer’s Dis. 2021;8(3). doi:10.14283/JPAD.2021.23
8. Altomare D, Molinuevo JL, Ritchie C, et al. Brain Health Services: organization, structure, and challenges for implementation. A user manual for Brain Health Services—part 1 of 6. Alzheimer’s Res Ther 2021 131. 2021;13(1):1-11. doi:10.1186/S13195-021-00827-2
9. Hippisley-Cox J, Coupland C, Vinogradova Y, Robson J, May M, Brindle P. Derivation and validation of QRISK, a new cardiovascular disease risk score for the United Kingdom: Prospective open cohort study. Br Med J. 2007;335(7611):136-141. doi:10.1136/bmj.39261.471806.55
10. Tunstall-Pedoe H. Cardiovascular risk and risk scores: ASSIGN, Framingham, QRISK and others: How to choose. Heart. 2011;97(6):442-444. doi:10.1136/hrt.2010.214858
11. Ader I, Pénicaud L, Andrieu S, et al. Healthy Aging Biomarkers: The INSPIRE’s Contribution. J frailty aging. 2021;10(4):313-319. doi:10.14283/JFA.2021.15
12. WHO. Integrated care for older people: guidelines on community-level interventions to manage declines in intrinsic capacity. Accessed November 4, 2021.
13. Cations M, Radisic G, Crotty M, Laver KE. What does the general public understand about prevention and treatment of dementia? A systematic review of population-based surveys. Werner P, ed. PLoS One. 2018;13(4):e0196085. doi:10.1371/journal.pone.0196085
14. Bateman RJ, Xiong C, Benzinger TLS, et al. Clinical and Biomarker Changes in Dominantly Inherited Alzheimer’s Disease. N Engl J Med. 2012;367(9):795-804. doi:10.1056/NEJMoa1202753
15. Bateman RJ, Benzinger TL, Berry S, et al. The DIAN-TU Next Generation Alzheimer’s prevention trial: Adaptive design and disease progression model. Alzheimer’s Dement. 2017;13(1):8-19. doi:10.1016/j.jalz.2016.07.005
16. Ellis KA, Bush AI, Darby D, et al. The Australian Imaging, Biomarkers and Lifestyle (AIBL) study of aging: Methodology and baseline characteristics of 1112 individuals recruited for a longitudinal study of Alzheimer’s disease. Int Psychogeriatrics. 2009;21(4):672-687. doi:10.1017/S1041610209009405
17. Ritchie CW, Ritchie K. The PREVENT study: A prospective cohort study to identify mid-life biomarkers of late-onset Alzheimer’s disease. BMJ Open. 2012;2(6):1893. doi:10.1136/bmjopen-2012-001893
18. Ritchie CW, Molinuevo JL, Truyen L, Satlin A, Van der Geyten S, Lovestone S. Development of interventions for the secondary prevention of Alzheimer’s dementia: The European Prevention of Alzheimer’s Dementia (EPAD) project. The Lancet Psychiatry. 2016;3(2):179-186. doi:10.1016/S2215-0366(15)00454-X
19. Frisoni GB, Molinuevo JL, Altomare D, et al. Precision prevention of Alzheimer’s and other dementias: Anticipating future needs in the control of risk factors and implementation of disease-modifying therapies. Alzheimer’s Dement. 2020;16(10):1457-1468. doi:10.1002/alz.12132
20. Sosa-Ortiz AL, Acosta-Castillo I, Prince MJ. Epidemiology of Dementias and Alzheimer’s Disease. Arch Med Res. 2012;43(8):600-608. doi:10.1016/j.arcmed.2012.11.003
21. Reiman EM, Arboleda-Velasquez JF, Quiroz YT, et al. Exceptionally low likelihood of Alzheimer’s dementia in APOE2 homozygotes from a 5,000-person neuropathological study. Nat Commun. 2020;11(1):1-11. doi:10.1038/s41467-019-14279-8
22. Emrani S, Arain HA, DeMarshall C, Nuriel T. APOE4 is associated with cognitive and pathological heterogeneity in patients with Alzheimer’s disease: a systematic review. Alzheimer’s Res Ther 2020 121. 2020;12(1):1-19. doi:10.1186/S13195-020-00712-4
23. Scarmeas N, Stern Y, Tang M-X, Mayeux R, Luchsinger JA. Mediterranean diet and risk for Alzheimer’s disease. Ann Neurol. 2006;59(6):912-921. doi:10.1002/ana.20854
24. Minakawa E, Wada K, Nagai Y. Sleep Disturbance as a Potential Modifiable Risk Factor for Alzheimer’s Disease. Int J Mol Sci. 2019;20(4):803. doi:10.3390/ijms20040803
25. Grande G, Marengoni A, Vetrano DL, et al. Multimorbidity burden and dementia risk in older adults: The role of inflammation and genetics. Alzheimer’s Dement. January 2021:alz.12237. doi:10.1002/alz.12237
26. Park H-Y, Park J-W, Song HJ, Sohn HS, Kwon J-W. The Association between Polypharmacy and Dementia: A Nested Case-Control Study Based on a 12-Year Longitudinal Cohort Database in South Korea. Laks J, ed. PLoS One. 2017;12(1):e0169463. doi:10.1371/journal.pone.0169463
27. Biessels GJ, Staekenborg S, Brunner E, Brayne C, Scheltens P. Risk of dementia in diabetes mellitus: A systematic review. Lancet Neurol. 2006;5(1):64-74. doi:10.1016/S1474-4422(05)70284-2
28. Kwok CS, Loke YK, Hale R, Potter JF, Myint PK. Atrial fibrillation and incidence of dementia: A systematic review and meta-analysis. Neurology. 2011;76(10):914-922. doi:10.1212/WNL.0b013e31820f2e38
29. Eggermont LHP, De Boer K, Muller M, Jaschke AC, Kamp O, Scherder EJA. Cardiac disease and cognitive impairment: A systematic review. Heart. 2012;98(18):1334-1340. doi:10.1136/heartjnl-2012-301682
30. Wolters FJ, Segufa RA, Darweesh SKL, et al. Coronary heart disease, heart failure, and the risk of dementia: A systematic review and meta-analysis. Alzheimer’s Dement. 2018;14(11):1493-1504. doi:10.1016/j.jalz.2018.01.007
31. Sharp SI, Aarsland D, Day S, Sønnesyn H, Ballard C. Hypertension is a potential risk factor for vascular dementia: systematic review. Int J Geriatr Psychiatry. 2011;26(7):661-669. doi:10.1002/gps.2572
32. Russ TC, Kivimäki M, Batty GD. Respiratory Disease and Lower Pulmonary Function as Risk Factors for Dementia: A Systematic Review With Meta-analysis. In: Chest. Vol 157. Elsevier Inc; 2020:1538-1558. doi:10.1016/j.chest.2019.12.012
33. Gardner RC, Burke JF, Nettiksimmons J, Kaup A, Barnes DE, Yaffe K. Dementia risk after traumatic brain injury vs nonbrain trauma: The role of age and severity. JAMA Neurol. 2014;71(12):1490-1497. doi:10.1001/jamaneurol.2014.2668
34. Godbolt AK, Cancelliere C, Hincapié CA, et al. Systematic review of the risk of dementia and chronic cognitive impairment after mild traumatic brain injury: Results of the international collaboration on mild traumatic brain injury prognosis. Arch Phys Med Rehabil. 2014;95(3 SUPPL):S245-S256. doi:10.1016/j.apmr.2013.06.036
35. Grossi CM, Richardson K, Fox C, et al. Anticholinergic and benzodiazepine medication use and risk of incident dementia: A UK cohort study. BMC Geriatr. 2019;19(1):276. doi:10.1186/s12877-019-1280-2
36. Hafdi M, Hoevenaar-Blom MP, Beishuizen CRL, Moll van Charante EP, Richard E, van Gool WA. Association of Benzodiazepine and Anticholinergic Drug Usage With Incident Dementia: A Prospective Cohort Study of Community-Dwelling Older Adults. J Am Med Dir Assoc. 2020;21(2):188-193.e3. doi:10.1016/j.jamda.2019.05.010
37. Liu YP, Chien WC, Chung CH, Chang HA, Kao YC, Tzeng NS. Are Anticholinergic Medications Associated With Increased Risk of Dementia and Behavioral and Psychological Symptoms of Dementia? A Nationwide 15-Year Follow-Up Cohort Study in Taiwan. Front Pharmacol. 2020;11:1. doi:10.3389/fphar.2020.00030
38. Smith SM, Vale WW. The role of the hypothalamic-pituitary-adrenal axis in neuroendocrine responses to stress. Dialogues Clin Neurosci. 2006;8(4):383. doi:10.31887/DCNS.2006.8.4/SSMITH
39. Scheltens P, Fox N, Barkhof F, De Carli C. Structural magnetic resonance imaging in the practical assessment of dementia: Beyond exclusion. Lancet Neurol. 2002;1(1):13-21. doi:10.1016/S1474-4422(02)00002-9
40. Wardlaw JM, Smith EE, Biessels GJ, et al. Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol. 2013;12(8):822-838. doi:10.1016/S1474-4422(13)70124-8
41. Altomare D, Caprioglio C, Assal F, et al. Diagnostic value of amyloid-PET and tau-PET: a head-to-head comparison. Eur J Nucl Med Mol Imaging. February 2021:1-12. doi:10.1007/s00259-021-05246-x
42. Villemagne VL, Doré V, Bourgeat P, et al. Aβ-amyloid and Tau Imaging in Dementia. Semin Nucl Med. 2017;47(1):75-88. doi:10.1053/j.semnuclmed.2016.09.006
43. Villemagne VL, Doré V, Burnham SC, Masters CL, Rowe CC. Imaging tau and amyloid-? proteinopathies in Alzheimer disease and other conditions. Nat Rev Neurol. 2018;14(4):225-236. doi:10.1038/nrneurol.2018.9
44. Dave A, Hansen N, Downey R, Johnson C. FDG-PET Imaging of Dementia and Neurodegenerative Disease. Semin Ultrasound, CT MRI. 2020;41(6):562-571. doi:10.1053/j.sult.2020.08.010
45. Olsson B, Lautner R, Andreasson U, et al. CSF and blood biomarkers for the diagnosis of Alzheimer’s disease: a systematic review and meta-analysis. Lancet Neurol. 2016;15(7):673-684. doi:10.1016/S1474-4422(16)00070-3
46. Kokkinou M, Beishon LC, Smailagic N, et al. Plasma and cerebrospinal fluid ABeta42 for the differential diagnosis of Alzheimer’s disease dementia in participants diagnosed with any dementia subtype in a specialist care setting. Cochrane Database Syst Rev. 2021;2. doi:10.1002/14651858.cd010945.pub2
47. Noel-Storr AH, Flicker L, Ritchie CW, et al. Systematic review of the body of evidence for the use of biomarkers in the diagnosis of dementia. Alzheimer’s Dement. 2013;9(3):e96-e105. doi:10.1016/j.jalz.2012.01.014
48. Ritchie C, Smailagic N, Noel-Storr AH, et al. Plasma and cerebrospinal fluid amyloid beta for the diagnosis of Alzheimer’s disease dementia and other dementias in people with mild cognitive impairment (MCI). Cochrane Database Syst Rev. 2014;2014(6):CD008782-CD008782. doi:10.1002/14651858.CD008782.pub4
49. Zetterberg H. Blood-based biomarkers for Alzheimer’s disease—An update. J Neurosci Methods. 2019;319:2-6. doi:10.1016/j.jneumeth.2018.10.025
50. Rossini PM, Di Iorio R, Vecchio F, et al. Early diagnosis of Alzheimer’s disease: the role of biomarkers including advanced EEG signal analysis. Report from the IFCN-sponsored panel of experts. Clin Neurophysiol. 2020;131(6):1287-1310. doi:10.1016/j.clinph.2020.03.003
51. De La Fuente Garcia S, Ritchie CW, Luz S. Artificial Intelligence, Speech, and Language Processing Approaches to Monitoring Alzheimer’s Disease: A Systematic Review. J Alzheimer’s Dis. 2020;78(4):1547-1574. doi:10.3233/JAD-200888
52. Scottish Dementia Research Consortium; Alzheimer Scotland; Brain Health Scotland. Scottish Brain Health and Dementia Research Strategy. Published July 2021. Accessed September 6, 2021.
53. Dounavi ME, Mak E, Wells K, et al. Volumetric alterations in the hippocampal subfields of subjects at increased risk of dementia. Neurobiol Aging. 2020;91:36-44. doi:10.1016/J.NEUROBIOLAGING.2020.03.006
54. Ritchie K, Carrière I, Howett D, et al. Allocentric and Egocentric Spatial Processing in Middle-Aged Adults at High Risk of Late-Onset Alzheimer’s Disease: The PREVENT Dementia Study. J Alzheimer’s Dis. 2018;65(3):885-896. doi:10.3233/JAD-180432
55. Xu W, Tan chen-chen, Zou J-J, cao X-P, Tan lan, department neurology. Sleep problems and risk of all-cause cognitive decline or dementia: an updated systematic review and meta-analysis. J Neurol Neurosurg Psychiatry. 2020;91:236-244. doi:10.1136/jnnp-2019-321896
56. Rasmussen M, Mestre H, Nedergaard M. The glymphatic pathway in neurological disorders. Lancet Neurol. 2018;17(11):1016-1024. doi:10.1016/S1474-4422(18)30318-1
57. Dujardin K, Sockeel P, Delliaux M, Destée A, Defebvre L. Apathy may herald cognitive decline and dementia in Parkinson’s disease. Mov Disord. 2009;24(16):2391-2397. doi:10.1002/MDS.22843
58. Robert PH, Berr C, Volteau M, et al. Apathy in patients with mild cognitive impairment and the risk of developing dementia of Alzheimer’s disease: A one-year follow-up study. Clin Neurol Neurosurg. 2006;108(8):733-736. doi:10.1016/J.CLINEURO.2006.02.003
59. Richard E, Schmand B, Eikelenboom P, et al. Symptoms of Apathy Are Associated with Progression from Mild Cognitive Impairment to Alzheimer’s Disease in Non-Depressed Subjects. Dement Geriatr Cogn Disord. 2012;33(2-3):204-209. doi:10.1159/000338239
60. Ismail Z, Agüera-Ortiz L, Brodaty H, et al. The Mild Behavioral Impairment Checklist (MBI-C): A rating scale for neuropsychiatric symptoms in pre-dementia populations. J Alzheimers Dis. 2017;56(3):929. doi:10.3233/JAD-160979
61. Cruz-Jentoft AJ, Bahat G, Bauer J, et al. Sarcopenia: Revised European consensus on definition and diagnosis. Age Ageing. 2019;48(1):16-31. doi:10.1093/ageing/afy169
62. Dumurgier J, Artaud F, Touraine C, et al. Gait Speed and Decline in Gait Speed as Predictors of Incident Dementia. Journals Gerontol Ser A Biol Sci Med Sci. 2016;72(5):glw110. doi:10.1093/gerona/glw110
63. Rasmussen LJH, Caspi A, Ambler A, et al. Association of Neurocognitive and Physical Function With Gait Speed in Midlife. JAMA Netw open. 2019;2(10):e1913123. doi:10.1001/jamanetworkopen.2019.13123
64. Adamo DE, Anderson T, Koochaki M, Fritz NE. Declines in grip strength may indicate early changes in cognition in healthy middle-aged adults. Wylie GR, ed. PLoS One. 2020;15(4):e0232021. doi:10.1371/journal.pone.0232021
65. Allan LM. Diagnosis and Management of Autonomic Dysfunction in Dementia Syndromes. Curr Treat Options Neurol. 2019;21(8):1-11. doi:10.1007/s11940-019-0581-2
66. Dalise AM, Prestano R, Fasano R, Gambardella A, Barbieri M, Rizzo MR. Autonomic Nervous System and Cognitive Impairment in Older Patients: Evidence From Long-Term Heart Rate Variability in Real-Life Setting. Front Aging Neurosci. 2020;12. doi:10.3389/fnagi.2020.00040
67. Usher-Smith J, Emery J, Hamilton W, Griffin SJ, Walter FM. Risk prediction tools for cancer in primary care. Br J Cancer 2015 11312. 2015;113(12):1645-1650. doi:10.1038/bjc.2015.409
68. Vermunt L, Muniz-Terrera G, Ter Meulen L, et al. Prescreening for European Prevention of Alzheimer Dementia (EPAD) trial-ready cohort: impact of AD risk factors and recruitment settings. Alzheimers Res Ther. 2020;12(1). doi:10.1186/S13195-019-0576-Y
69. Ball H, McWhirter L, Ballard C, et al. Functional cognitive disorder: dementia’s blind spot. Brain. 2020;143(10):2895-2903. doi:10.1093/BRAIN/AWAA224
70. McWhirter L, Ritchie C, Stone J, Carson A. Functional cognitive disorders: a systematic review. The lancet Psychiatry. 2020;7(2):191-207. doi:10.1016/S2215-0366(19)30405-5
71. Alexander M, Blackburn D, Reuber M. Patients’ accounts of memory lapses in interactions between neurologists and patients with functional memory disorders. Sociol Health Illn. 2019;41(2):249-265. doi:10.1111/1467-9566.12819
72. Bemelmans SASA, Tromp K, Bunnik EM, et al. Psychological, behavioral and social effects of disclosing Alzheimer’s disease biomarkers to research participants: a systematic review. Alzheimer’s Res Ther. 2016;8(1):1-17. doi:10.1186/s13195-016-0212-z
73. Saunders S, Sheehan S, Evans A, Luz S, Ritchie CW, Terrera GM. The electronic person specific outcome measure (ePSOM) programme: Overview and survey results. Alzheimer’s Dement. 2020;16(S6):e040097. doi:10.1002/ALZ.040097
74. Wimo A, Guerchet M, Ali G-C, et al. The worldwide costs of dementia 2015 and comparisons with 2010. Alzheimer’s Dement. 2017;13(1):1-7. doi:10.1016/j.jalz.2016.07.150
75. Breeze P, Thomas C, Thokala P, Lafortune L, Brayne C, Brennan A. The Impact of Including Costs and Outcomes of Dementia in a Health Economic Model to Evaluate Lifestyle Interventions to Prevent Diabetes and Cardiovascular Disease. Med Decis Mak. 2020;40(7):912-923. doi:10.1177/0272989X20946758
76. Mukadam N, Anderson R, Knapp M, et al. Effective interventions for potentially modifiable risk factors for late-onset dementia: a costs and cost-effectiveness modelling study. Lancet Heal Longev. 2020;1(1):e13-e20. doi:10.1016/s2666-7568(20)30004-0



Q.Y. Chen1, Y. Yin2, L. Li1, Y.J. Zhang1, W. He1, Y. Shi3


1. Department of Neurology, The Third Affiliated Hospital of Qiqihar Medical University, Qiqihar 161000, P.R. China; 2. Department of Teaching & Research, The Third Affiliated Hospital of Qiqihar Medical University, Qiqihar 161000, P.R. China; 3. College of Medical Technology, Qiqihar Medical University, Qiqihar 161006, P.R. China

Corresponding Author: Yan Shi College of Medical Technology, Qiqihar Medical University, No. 333, Bukui Street, Jianhua District, Qiqihar 161006, Heilongjiang Province, P.R. China E-mail:, ORCID:0000-0001-6254-3201, Tel: +86-0452-2663346

J Prev Alz Dis 2021;
Published online November 17, 2021,



BACKGROUND: Alzheimer’s disease (AD) is a major cause of dementia, which is a growing global health problem and has a huge impact on individuals and society. As the modifying role of geniposidic acid (GPA) has been suggested in AD, this study sets out to determine if and how GPA treatment affects AD progression in mice.
METHODS: Potential downstream target genes of GPA during AD were identified by bioinformatics analysis, revealing GAP43 as a primary candidate protein. Then, mPrP-APPswe/PS1De9 AD transgenic mice were treated with GPA via intragastric administration. This allowed for gain- and loss-of-function assays of candidate proteins being carried out with or without GPA treatment, after which behavioral tests could be conducted for mice. Cortical neuron apoptosis was measured by TUNEL staining, Amyloid β-protein (Aβ) expression in cerebral cortex by Thioflavin-s staining, and Aβ, IL-1β, IL-6, IL-4 and TNF-α levels in cerebral cortex by ELISA. GAP43 expression in cerebral cortex of mice was detected by immunohistochemistry. Primary cortical neurons of embryonic mice were isolated and induced by Aβ1-42 to construct AD cell model. Cell viability was assessed by CCK-8, and axon growth by immunofluorescence.
RESULTS: GPA administration significantly improved the cognitive impairment, reducing Aβ accumulation and neuronal apoptosis in AD mice, and alleviated inflammation and axonal injury of Aβ1-42-induced neurons. GAP43 was shown experimentally to be the target of GPA in AD. Silencing of GAP43 repressed the neuroprotective effect of GPA treatment on AD mice. GPA elevated GAP43 expression via PI3K/AKT pathway activation and ultimately improved nerve injury in AD mice.
CONCLUSION: GPA activates a PI3K/AKT/GAP43 regulatory axis to alleviate AD progression in mice.

Key words: Alzheimer’s disease, geniposidic acid, GAP43; PI3K/AKT pathway, neuroinflammation, neuron, axon, nerve injury.



A s the principal cause of dementia, Alzheimer’s disease (AD) is a growing global health problem which has a huge impact at an individual and societal level (1, 2). This disease is caused by amyloid plaques and neurofibrillary tangles that accumulate in the brain and result in gradual cognitive decline (3). As a neurodegenerative and prominent protein conformational disease, AD presents itself as memory loss and progressive neurocognitive dysfunction (4). Despite several treatments have been developed to alleviate mild symptoms, no drugs are currently in use which improve cognition or prevent AD progression (5). Therefore, furthering our understanding of molecular mechanism of AD is of great clinical significance and could reveal novel therapeutic targets.
Geniposidic acid (GPA), one of the main active components of Gardenia jasminoides J. Ellis (Rubiaceae), has several beneficial physiological effects such as blood pressure regulation, cancer prevention, repairing soft tissue injury, osteoporosis treatment, anti-inflammation, anti-thrombosis, and anti-platelet aggregation (6). Moreover, it has been shown that GPA can relieve the spatial learning, memory deficits, and neuroinflammation in AD-model mice (7). Interestingly, a prior study has reported that geniposide can upregulate growth-associated protein 43 (GAP43) which leads to a reduction in fluoxetine-suppressed neurite outgrowth in Neuro2a neuroblastoma cells (8). However, a direct relationship between GPA and GAP43 has not been fully explored. As an axonal membrane protein which is abundantly expressed in neuronal growth cones, GAP43 plays a critical role in the stabilization of synapse structure, axonal regeneration, and neural growth (9). Importantly, upregulation of GAP43 has been identified to be involved in the attenuation of behavioral deficits, modification of synaptic structure, and acceleration of neurite outgrowth in AD (10). Intriguingly, geniposide has also been shown to activate PI3K/AKT signaling, reducing neonatal mouse brain injury after hypoxic-ischemia treatment (11). In addition, activation of the PI3K/AKT pathway is thought to lead to the upregulation of GAP43 during hair cell protection in the neonatal murine cochlea (12). Furthermore, activation of the PI3K/AKT pathway might be capable of subduing cognitive deficits in AD (13).
Given these findings we hypothesized that the suppressive role GPA treatment plays in AD is via a PI3K/AKT/GAP43 regulatory axis. Therefore, we set out to determine whether GPA treatment affects Aβ accumulation, neuronal apoptosis, neuroinflammation and axonal injury via the PI3K/AKT/GAP43 regulatory axis.


Materials and methods

Bioinformatics analysis

CTD database ( and SymMap database ( were used to predict the targets of GPA. Gene expression dataset GSE28146 related to AD in mice species was obtained from Gene Expression Omnibus database ( GPL570 was the platform file of the gene expression dataset. The samples of the gene expression dataset were grouped, including 8 control samples and 7 AD samples. Differential analysis was conducted using R language «limma» package ( to screen the differentially expressed genes (DEGs) with |logFC| > 1 and p < 0. 05 as the screening criteria. The top 50 DEGs with the minimum p value were selected to draw a heat map using «pheatmap» package ( of R language. The drug-target network was visualized by Cytoscape 3.5.1 software.

Drug treatment

GPA (purity: HPLC ≥ 98%, wkq16052101) was purchased from Sichuan Victory Biological Technology Co., Ltd. (Chengdu, China) and dissolved into distilled water at a final concentration of 2.5 mg/mL, 5 mg/mL and 7.5 mg/mL. The GPA administration time of mice was 10:00-12:00 a.m.

Animal treatment and grouping

Experimental male mPrP-APPswe/PS1De9 AD transgenic mice (AD mice, male, aged 6 months) were purchased from Beijing HFK Biotechnology Co., Ltd. (Beijing, China). The animals were kept at ambient temperature of 23 ± 1℃ with a 12 h light/dark cycle and relative humidity of 55 ± 5%, and free access to food and water. Before the experiment, mice were left to adapt to this environment for 2 to 3 days. AD mice were randomly arranged into treatment group and control group. AD mice were treated with GPA (25 mg/kg, 50 mg/kg, and 75 mg/kg) or normal saline (NS) via intragastric administration for three months. At the same time, 50 μg/kg LY294002 (S1105, Selleck, Houston, TX, USA) was injected 30 min after GPA administration or short hairpin RNA (sh)-negative control (NC)/sh-GAP43 adenovirus (1 × 1011 plaque-forming units, 10 μL, Shanghai Geneland Biotech Co., Ltd., Shanghai, China) was injected every two weeks into the lateral ventricle (The injection site relative to bregma position: 1 mm in the positive position, 1.5 mm in the right position, and 3.5 mm in the depth). The drug was injected into the ipsilateral ventricle with a Hamilton syringe (Hamilton Company, Renault, NV, USA). The needle was left for 5 min after injection and then slowly withdrawn for more than 5 min. After the needle was removed, the burr hole was sealed with bone wax. The 90 AD mice were randomly assigned into 9 groups with 10 mice in each group: AD + NS group (AD mice were fed with NS), AD + GPA 25 mg/kg group (AD mice were given with GPA at 25 mg/kg by gavage), AD + GPA 50 mg/kg group (AD mice were given with GPA at 50 mg/kg by gavage), AD + GPA 75 mg/kg group (AD mice were given with GPA at 75 mg/kg by gavage), AD + GPA + sh-NC group (AD mice were given with GPA at 75 mg/kg by gavage, and sh-NC adenovirus was injected into lateral ventricle), AD + GPA + sh-GAP43 group (AD mice were given with GPA 75 mg/kg by gavage, and sh-GAP43 adenovirus was injected into lateral ventricle), AD + GPA + overexpression (oe)-NC (GPA at 75 mg/kg was given to AD mice and oe-NC adenovirus was injected into lateral ventricle), AD + GPA + LY294002 + oe-NC (GPA at 75 mg/kg was given to AD mice, and oe-NC adenovirus and PI3K inhibitor LY294002 (14) were injected into the lateral ventricle at the same time), AD + GPA + LY294002 + oe-GAP43 (GPA at 75 mg/kg was given to AD mice, and oe-GAP43 adenovirus and PI3K inhibitor were injected into lateral ventricle at the same time). Next, 10 C57BL/6 mice (WT mice) of the same age were fed with NS as control: WT + NS (WT mice were fed with NS) group. Morris water maze (MWM) test and new object recognition test were carried out as described below. After behavioral experiment, mice were euthanized and brain tissue was prepared into paraffin section. The experiment was approved by the animal ethics committee of Qiqihar Medical University.

Animal behavioral experiment

MWM test was performed 90 to 95 days after administration. The MWM instrument (Chinese Academy of Medical Sciences, Beijing, China, DMS-2) consisted of a circular container (100 cm in diameter and 40 cm in height), a recording and analysis system, and a digital camera (TOTA Group Limited, Japan). The instrument was located in a low light test room. The water was made opaque by adding food grade white colorant and kept at 20 ± 2℃. During the training, the cylindrical escape platform (10 cm in diameter) was located in the center of the north and South quadrant of the pool, with a depth of about 0.5 cm. The space training test was conducted for 5 consecutive days, and the mice were put into the pool at the planned starting position every day. The mice were trained to find the platform within 60 s. If the mouse did not find the platform within 60s, the mouse was guided to the platform, and the mouse was allowed to stand on the platform for 20 s after reaching the platform. This was repeated four times a day. The swimming track of each mouse was recorded by a camera. The exploration experiment was carried out within 24 h after the 5-day positioning navigation test, and the platform was removed. The mice swam in the swimming pool for 60 s. The swimming trajectory of the mice within 60 s was recorded and analyzed. Within 24 h after the 5-day WMW experiment, the spatial exploration test of memory recovery ability was carried out, and the platform of shadow was removed. The mice were placed in a water tank with their faces facing the wall in a randomly selected quadrant. The crossing times of the platform location within 60 s were recorded. After that, in order to eliminate the potential impact of motor ability and visual obstacles on the experimental results, the visual platform test was conducted after the spatial exploration test. All the mice were placed in the same quadrant, and the mice were able to see the platform. The time required for mice to find the visible platform was recorded.
On the first day of the new object recognition test, the mice were gently placed in the test box for 5 min. On the next day, mice were gently placed in the test box with 2 identical objects. On the third day, the mice were gently placed in the test box, but one of the objects had been replaced by a new one. When the mice directly touched the object with their mouth, forehead or nose, the exploration time was recorded. The discovery index was that the time spent on a new object was divided by the cumulative time spent on two objects.

Thioflavin-s (Th-S) staining

Thioflavin was a fluorescent dye which was usually adopted to stain senile plaques. Firstly, the brain tissues of mice were fixed with paraformaldehyde, dehydrated, and embedded in paraffin. The paraffin-embedded tissues were cut with a slicer and fixed. The slices were dehydrated, hydrated in distilled water, and stained with Mayer’s hematoxylin for 5 min. The slides were exposed to Th-S solution (1% in distilled water) for 5 min. The slices were immersed in 70% alcohol for 5 min and sealed by glycerin gelatin. The slices were observed with an optical microscope (Olympus BX 41 microscope, 40 × magnification), and the number of senile plaques was calculated by Image Pro Plus software.

Enzyme-linked immunosorbent assay (ELISA)

The levels of interleukin (IL)-1β (Beyotime, Shanghai, China, PI301), IL-6 (Beyotime, PI326), IL-4 (Beyotime, PI612), TNF-α (Beyotime, PT512), and Aβ 1-42 (Mosake Biological Technology Co., Ltd., Wuhan, China, 69-21411) in mouse cortical homogenate were detected according to the instructions of ELISA kit. The antibody was diluted to a protein content of 1-10 μg/mL through 0.05 M carbonate coated buffer solution (PH = 9). The 0.1 mL diluent was added into the reaction well of each polystyrene plate which was kept at 4℃ overnight. The next day the plate was washed with washing buffer for 3 times. The 0.1 mL diluted sample to be tested was added into the coated reaction well and incubated at 37℃ for 1 h. The 0.1 mL fresh diluted enzyme-labeled antibody was added to each reaction well. The 0.1 mL tetramethylbenzidine substrate solution was added into each reaction well for 10-30 min of incubation at 37℃, followed by addition of 0.05 mL of 2 M sulfuric acid into each reaction well. At 450 nm, the optical density (OD) value of each well was measured using a microplate reader and corrected to blank control wells. Each experiment was repeated three times.

TdT-mediated dUTP-biotin nick end-labeling (TUNEL) staining

A TUNEL apoptosis detection kit (Millipore Corp., Billerica, MA, USA) was used to detect apoptosis in brain tissues according to the instructions provided. Paraffin mouse brain tissue sections were dewaxed in xylene for 5-10 min, with fresh xylene for 5-10 min, with anhydrous ethanol for 5-10 min, with 90% ethanol for 2 min, and with 70% ethanol for 2 min, and washed with distilled water for 2 min. The sections were treated with 20 μg/mL protease K (ST532, Beyotime) at 20-37℃ for 15-30 min. The sections were incubated in 3% hydrogen peroxide solution prepared by phosphate buffer saline (PBS) for 20 min at room temperature, and then incubated with biotin-labeled solution at 37℃ for 60 min. The sections were incubated with labeling reaction termination solution at room temperature for 10 min. The sections were incubated in 50 μL Streptavidin-horseradish peroxidase (HRP) solution at room temperature for 30 min. The 0.2-0.5 mL diaminobenzidine (DAB) chromogenic solution was added for incubation at room temperature for 5-30 min, followed by sealing. The sections were observed and photographed under an inverted microscope. Ten visual fields in each group were randomly selected and the number of positive cells and total cells in each were counted. The cells with brownish yellow nuclei were noted as apoptotic positive cells, while those with blue nuclei were classed as normal healthy cells. The apoptosis rate was expressed as the percentage of brown-yellow cells.

Isolation and culture of primary cortical neurons

Primary cortical neurons were prepared from the cortex of 19-day embryo mice. In short, the embryonic mouse cortex was dissected in cold PBS. The tissues were collected and washed in PBS, then treated with 0.25% trypsin at 37℃ for 15 min. Finally, fetal bovine serum (FBS, Gibco, Carlsbad, California, USA) was added at the final concentration of 10% to stop the trypsinization. After, the neurons were precipitated by 10-min centrifugation at 800 rpm and resuspended with the neural basal medium (Gibco) containing 10% FBS in a cell incubator at 37℃ and 5% CO2. The next day, the basic medium was discarded and renewed with fresh medium in the presence of serum. On the third day of culture, the proliferation of glial cells was inhibited by cytarabine at final concentration of 10 μmoL/L, and glial cells were aspirated after 24 h. The cells cultured in vitro for 7-21 days were used for the experiment.

Identification of primary cortical neurons

On the third and seventh day of seeding, the morphological changes of the cells were observed using an optical microscope. The purity of neurons was identified by Neuron Specific Enolase (NSE) immunofluorescence staining. The cells were fixed with 4% paraformaldehyde at room temperature for 10 min. After the paraformaldehyde was removed, the cells were permeabilized for 15 min by treating with PBS supplemented with 0.12% Triton X100, and blocked with 5% bovine serum albumin (BSA) at room temperature for 1 h. Then cells were further cultured with specific rabbit anti-NSE (ab180943, 1:100, Abcam, Cambridge, UK) in a wet box at 4℃ overnight, and incubated with fluoresceinisothiocyanat (FITC)-labeled rabbit anti secondary antibody (ab6717, 1:100, Abcam) in the dark at room temperature for 1 h, followed by the addition of 4’,6-Diamidino-2-Phenylindole at room temperature in the dark for 5 min. Cells were sealed with anti-quench sealing tablet. NSE positive cells were observed under a fluorescence microscope. The percentage of NSE positive cells in the total number of cells in the visual field was the purity of neurons (Figure S1).

Cell transfection

The core plasmid (pLKO.1) and auxiliary plasmid (psPAX2, pMD2.G) of target gene silencing sequence were applied to package the silencing lentivirus. The core plasmid (pHAGE-CMV-MCS-IzsGreen) and auxiliary plasmid (psPAX2, pMD2.G) of target gene cDNA sequence were adopted to package the overexpression lentivirus. The lentivirus was purchased from Shanghai Sangon Biotechnology Co. Ltd. (Shanghai, China). The primer sequence and plasmid were constructed by Shanghai Sangon Biotechnology Co., Ltd. The packaging virus and the target vector were co-transfected into 293T cells by Lipo2000 (11668-019, Invitrogen, Carlsbad, CA, USA). The supernatant was collected after 48 h of cell culture. The supernatant after filtration and centrifugation contained virus particles. Virus titer was detected. Viruses in exponential phase of growth were harvested and were arranged into sh-NC group (silencing lentivirus control group), sh-GAP43-1 group (silencing GAP43 lentivirus 1 group), and sh-GAP43-2 group (silencing GAP43 lentivirus 2 group) according to the different transfectants. The medium was renewed 8-h post transfection. After 96 h of transfection, the transfection efficiency was observed under an inverted fluorescence microscope. Silencing lentiviral transfection and silencing sequences are shown in Table S1.

Establishment of AD cell model in vitro

For AD cell model construction, 1 mg of Aβ1-42 (Anaspec, San Jose, CA, USA) lyophilized powder was dissolved in hexafluoroisopropanol (HFIP) (Sigma-Aldrich, St Louis, MO, USA) at the final concentration of 1 mg/mL, followed by water bath and ultrasonic wave treatment for 10 min. Aβ1-42 was stored at room temperature in dark for 5-24 h. The 0.1 mg powder was obtained by nitrogen drying and stored at -80℃. Then, the powder was added with dimethyl sulfoxide (DMSO) to the final concentration of 1 mg/mL, and diluted with PBS to the working concentration of 100 μM/L, and cultured in a constant temperature incubator at 37℃ for more than 3 h, allowing a Aβ1-42 oligomer to form. The activated Aβ1-42 (20 μM/L) was added to cortical neuron culture medium for 24 h to construct AD cell model.

Cell counting kit (CCK)-8 assay

A CCK-8 (K1018, APExBIO, Boston, MA, USA) kit was used to evaluate the cell viability of cell model which was treated with GPA or Aβ1-42. The cells were placed in 96 well plates (100 μL/well; 1 × 104 cells/well). Firstly, cells were treated with 20 μM/L GPA for 2 h, and then treated with activated Aβ1-42 respectively for 3 h, 6 h, 12 h, and 24 h. Six duplicated wells were set up. At the same time, the control group was added with the same volume of DMSO as Aβ1-42. After treatment, 10 μL CCK8 solution was added to each well and incubated at 37℃ for 2 h.
Next, the absorbance at 450 nm was measured using a microplate reader. The cell viability (%) was calculated by the formula [(As – Ab)/(Ac – Ab)] × 100%. As represents the absorbance value of supernatant from exposed or false exposed dishes, Ac represents the absorbance of the well containing the supernatant in the normal control, and Ab represents the absorbance value of the culture well containing 10% CCK-8 solution.

Immunofluorescence staining

The culture plate was washed twice with PBS at 37℃ and the cells were fixed in 4% paraformaldehyde (pH 7.4) for 15 min. After blocking in 3% normal Donkey Serum (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) and permeating in 0.1% Triton X-100, cells were cultured with mouse anti-class III β-tubulin monoclonal antibody (1: 200, Beyotime) overnight in the dark room at 4℃. Afterwards, Cy3 donkey anti-mouse Immunoglobulin G (IgG) (1:400, Jackson ImmunoResearch Laboratories) and Dylight488 donkey anti-rabbit IgG (1:400, Jackson ImmunoResearch Laboratories) were incubated with cells. Then, the cells were stained with Hoechst 33342 (Sigma-Aldrich). The samples were observed under a confocal microscope (LSM 710; Carl Zeiss, Oberkochen, Germany), and the length of axon growth was measured by ZEN2009 software (Carl Zeiss). For each group and experiment, three fields of vision were observed in each cover glass. Each experiment was repeated three times.

Western blot analysis

Radio-Immunoprecipitation assay cell lysis buffer containing phenylmethylsulfonyl fluoride (P0013B, Beyotime) was added to lyse cells. The supernatant was harvested and the total protein concentration of each sample was measured with a bicinchoninic acid kit as per the manufacturers protocol (P0011, Beyotime). The adjusted protein concentration was 1 μg/μL. The volume of sample in each tube was set at 100 μL. The sample was boiled at 100℃ for 10 min to denature the protein which was stored at – 80℃ for use. According to the size of the target protein, 8%-12% sodium dodecyl sulfate gel was prepared, and the proteins were added to the lanes in equal amounts for electrophoresis. The protein on the gel was transferred to a polyvinylidene fluoride membrane (1620177, Bio-Rad Laboratories, Hercules, CA, USA). The membrane was sealed with 5% skimmed milk or 5% BSA at room temperature for 1 h. The membrane was probed with primary rabbit antibodies (Abcam) to PI3K (ab154598, 1:1000,) and GAP43 (ab75810, 1:1000) and primary rabbit antibodies [Cell Signaling Technologies (CST), Beverly, MA, USA] to AKT (#9272, 1:1000), phosphorylation (p)-AKT (Serine473, #9271, 1:1000), and β-actin (#4970, 1:5000) overnight at 4℃. The membrane was re-probed with horseradish peroxidase (HRP)-labeled goat anti-rabbit IgG (ab6721, 1:5000, Abcam) secondary antibody at room temperature for 1 h. The membrane was immersed in electrogenerated chemiluminescence reaction solution (1705062, Bio-Rad Laboratories) at room temperature for 1 min. The liquid was removed, and the membrane was covered with plastic wrap. Strip exposure imaging was performed on the Image Quant LAS 4000C gel imager (Amersham Biosciences/GE Healthcare, Piscataway, NJ, USA). β-actin was used as the internal reference. The ratio of gray value of target band to internal reference band was used as the relative expression level of protein. Each experiment was repeated three times.

Reverse transcription quantitative polymerase chain reaction (RT-qPCR)

Total RNA was extracted from cells and mouse cerebral cortex using Trizol (16096020, Thermo Fisher Scientific Inc., Waltham, Massachusetts, USA). Then the total RNA was reversely transcribed into complementary (cDNA) by using PrimeScript RT Kit (Takara Biotechnology Ltd., Dalian, China). RT-qPCR experiment was carried out with a RT-qPCR kit (Q511-02, NanJing Vazyme Biotech Co., Ltd, Nanjing, China) according to the instructions. PCR amplification was performed on a Bio-rad quantitative real-time PCR system CFX96. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was utilized as the internal reference. The primer sequence was designed and provided by Shanghai Sangon Biotechnology Co. Ltd. The 2-ΔΔCt method was the multiple ratio relationship between the experimental group and the control group. The primer sequences are described in Table S2. Each experiment was repeated three times.

Immunohistochemistry (IHC) staining

The brain tissue sections of mice were baked at 60℃ for 20 min, placed in xylene solution successively, and then soaked for 15 min after xylene solution replacement. Mouse brain tissue sections were dehydrated with 100%, 95%, 90%, 85% and 80% ethanol. After that, 3% H2O2 was added to each section, which was soaked at room temperature for 10 min to block the endogenous peroxidase activity. The sections were added to citric acid buffer, heated in a microwave oven for 3 min, and cultured with antigen repair solution at room temperature for 10 min. The sections were incubated with normal goat serum blocking solution (Shanghai Sangon Biotechnology Co., Ltd.) at room temperature for 20 min. Then, the sections were incubated with primary rabbit anti-GAP43 antibody (ab75810, 1:500, Abcam) for a night in the dark room. The next day, the sections were incubated with goat anti-rabbit IgG secondary antibody (ab6721, 1: 1000, Abcam) for 30 min, and then added with streptavidin-biotin complex (vector company, USA) in a 37℃ incubator for 30 min. Diaminobenzidine chromogenic Kit (P0203, Beyotime) was utilized to add one drop of chromogenic reagent to the specimen for coloring for 6 min. The sections were stained in hematoxylin for 30 s. The sections were put into 70%, 80%, 90%, 95% ethanol and anhydrous ethanol for 2 min each time. Finally, the sections were cleared twice in xylene for 5 min and sealed with neutral resin. The sections were observed under an upright microscope (BX63, Olympus Optical Co., Ltd, Tokyo, Japan). Each experiment was repeated three times.

Statistical analysis

SPSS 21.0 (IBM Corp. Armonk, NY, USA) was adopted for statistical analysis. The measurement data is presented as the mean ± standard deviation throughout this study. Unpaired t-tests were implemented for comparison between the two groups. One-way analysis of variance (ANOVA) was applied for comparison among multiple groups, while repeated measurement ANOVA was utilized to compare the data at different time points, followed by Tukey’s post-hoc test. p < 0. 05 was considered to be statistically significant difference.



GPA alleviates cognitive impairment in AD mice

In order to explore the role of GPA in the cognitive dysfunction of AD mice, the behavior of AD mice was initially assessed. MWM test results showed that after 5 days of positioning navigation test, the time for mice to find the hidden platform reduced gradually. Compared with the WT group, the escape latency of the AD group was significantly longer, while the escape latency of the AD + GPA group was significantly reduced (Figure 1A). On the last day of the MWM test, the hidden platform was removed for exploratory testing. The results showed that compared with the WT group, the times crossing the platform was significantly reduced in the AD group. In contrast to the AD group, mice in the AD + GPA group crossed the platform more regularly (Figure 1B).

Figure 1. GPA alleviates cognitive dysfunction in AD mice

A, The escape latency of mice on different days detected by the positioning navigation test. B, The times that mice crossed the platform in each group. C, Swimming time of mice in each group in the target quadrant, D, Exploration index of mice on day 1 in new object recognition test. E, Exploration index of mice on day 2 in new object recognition test. * p < 0.05 vs. the WT group. # p < 0.05 vs. the AD group, n=10.


Visual platform test results showed that, compared with the WT group, the time it took for mice to find the visual platform was significantly increased in the AD group, while this time was dramatically decreased in the GPA treated AD + GPA group (Figure 1C). The results of the new object recognition test demonstrated that there was no significant difference in the exploration index of all groups when the mice were placed in a box with two identical objects (Figure 1D). When one of the objects was replaced by a new one on the next day, the exploration index in the AD group was strikingly lower than that of the WT group. Compared with the AD group, the exploratory index of the AD + GPA group was clearly elevated (Figure 1E). In conclusion, GPA treatment significantly improved the cognitive impairment of AD mice in a dose dependent manner.

GPA alleviates Aβ accumulation, neuronal apoptosis, neuroinflammation and axonal injury in AD mice

We then turned our attention to examine the mechanisms underpinning GPA-mediated alleviation of cognitive impairment, more specifically effects on Aβ deposition, neuroinflammation, and axonal injury in AD mice. ELISA analysis revealed that the level of Aβ in the AD group was notably higher than that in the WT group. Compared with the AD group, the level of Aβ in the AD + GPA group was reduced, with a clear inverse correlation between GPA concentration and Aβ levels (Figure 2A). In addition, the use of Th-S staining further demonstrated that GPA triggered reduction of the accumulation of Aβ in cerebral cortex (Figure 2B). TUNEL analysis showed, that compared with the WT group, the apoptosis rate was higher in the AD group, an effect that was reduced when mice were given GPA, where again a dose dependent response was observed. (Figure 2C). As reflected by ELISA analysis, the expression of IL-6, TNF-α and IL-1β was strikingly enhanced and the expression of IL-4 was reduced in the AD group, while an opposite trend was observed in the AD + GPA group (Figure 2D). Next, primary cortical neurons of mice were induced by Aβ1-42 to provide an in vitro cell model of AD, followed by identification of primary cortical neurons, shown in Supplementary result (Figure S1A-C). Then, as CCK-8 assay demonstrated, when compared with the DMSO group, cell viability of the Aβ group was markedly decreased. In contrast to the Aβ group, cell viability of the GPA + Aβ group was noticeably increased (Figure 2E). Immunofluorescence microscopy revealed that axon length was observably shorter in the Aβ group than in the DMSO group, which was longer in the GPA + Aβ group (Figure 2F).

IGPA reduces Aβ level, neuronal apoptosis, inflammation, and axonal injury in AD mice

A, ELISA to detect the level of Aβ in the brain of mice in each group. B, Th-S staining and quantitative map of the brain. C, TUNEL staining to measure the apoptosis rate in each group. D, The expression of inflammatory factors in cerebral cortex determined by ELISA. E, CCK-8 to assess cell viability in each group. F, The growth of axons observed by immunofluorescence staining. * p < 0.05 vs. the WT/DMSO group. # p < 0.05 vs. the AD/Aβ group. Each cell experiment was repeated three times


In conclusion, GPA, in an apparently dose-related manner, reduces Aβ accumulation and neuronal apoptosis in AD mouse brain, while also alleviating inflammation and axonal injury. It is highly likely that these processes underpin the improved cognitive function following GPA treatment, thus indicating that GPA may play a neuroprotective role in AD.

GAP43 is a downstream target of GPA

In order to further explore the target of GPA, the differential analysis of AD cohort GSE28146 was analyzed. In total, 770 DEGs were screened in GSE28146, among which 379 significantly upregulated and 391 significantly downregulated genes downregulated (Figure 3A). The top 50 genes with significant difference were identified and used to plot a heat map (Figure 3B). The 19 GPA targets that were predicted by CTD database and symmap database were intersected with AD-related genes that were obtained by GeneCards database and DEGs, which obtained candidate genes GAP43 and IL-1β (Figure 3C). The drug-target regulatory network was obtained by Cytoscape 3.5.1 software, as shown in Figure 3D, in which genes labeled with yellow were the common target. GAP43 was significantly downregulated in AD microarray dataset GSE28146 (Figure 3E). Overall, this bioinformatics analysis indicates that GAP43 is a probable downstream target of GPA in AD.

Figure 3. GAP43 is a downstream target of GPA as predicted by bioinformatics analysis

A, Volcano map of DEGs in AD in GSE28146. Red dots indicated upregulated genes, and green dots indicated downregulated genes. B, Heat map of expression of candidate genes. The color scale from blue to red indicates level of gene expression from low to high. C, Venn diagram of intersection of DEGs, AD-related genes and drug targets, D, The drug-target regulatory network. E, Expression box diagram of GAP43 in AD in GSE28146.


GAP43 silencing inhibits GPA-mediated improvements in cognitive function

In order to further verify whether GPA alleviated cognitive impairment and nerve injury in AD mice through GAP43, the mRNA level of GAP43 in the cerebral cortex of each group was measured by RT-qPCR. RT-qPCR results found that GAP43 expression in the AD group was substantially lower than that of the WT group. Compared with the AD group, GAP43 expression and positive rate of GAP43 in the AD + GPA group were significantly enhanced as GPA concentration (Figure 4A, B). In order to explore whether GAP43 is a target of GPA, sh-GAP43 adenovirus were injected into the lateral ventricle of AD mice to achieve a local silencing of GAP43 expression during administration period of GPA. Western blot results indicated that GAP43 expression in the AD + GPA + sh-NC group was substantially higher than that in the AD + sh-NC group. In contrast to the AD + GPA + sh-NC group, GAP43 expression was obviously lower in the AD + GPA + sh-GAP43 group, confirming a GAP43 knockdown (Figure 4C). The result of MWM tests demonstrated that the escape latency of mice in the AD + GPA + sh-NC group was significantly shorter than that of the AD + sh-NC group. Additionally, compared with the AD + GPA + sh-NC group, the escape latency of mice in the AD + GPA + sh-GAP43 group was noticeably decreased (Figure 4D). Also, on the last day of the MWM test, the hidden platform was removed for exploratory test. These results showed that, in comparison with the AD + sh-NC group, the times of crossing platform in the AD + GPA + sh-NC group was increased, while the AD + GPA + sh-GAP43 group displayed the opposite results (Figure 4E). The swimming time of each group in the target quadrant is shown in Figure 4F. Compared with the AD + sh-NC group, the swimming time of mice in the AD + GPA + sh-NC group was strikingly increased. Compared with the AD + GPA + sh-NC group, the swimming time of mice was remarkably reduced in the AD + GPA + sh-GAP43 group. The result of new object recognition test showed that the exploratory index of the AD + GPA + sh-NC group was prominently higher than that of AD + sh-NC group when one of the objects was replaced by a new one. However, in comparison with the AD + GPA + sh-NC group, the exploration index of the AD + GPA + sh-GAP43 group was observably decreased (Figure 4G-H). In summary, silencing of GAP43 reduced GPA-mediated improvements in cognitive function in AD mice.

Figure 4. GPA-mediated alleviation of cognitive dysfunction is interrupted when GAP43 expression is silenced in AD mice

A, mRNA expression of GAP43 in cerebral cortex of mice detected by RT-qPCR. B, GAP43 expression in cerebral cortex of mice determined by IHC staining. * p < 0.05 vs. the WT group. # p < 0.05 vs. the AD group. C, The silence efficiency of GAP43 measured by Western blot. D, Escape latency of mice in different days in the positioning and navigation test. E, The times of mice crossing the platform in each group. F, Swimming time of mice in target quadrant. G, Exploration index of mice on day 1 in new object recognition test. H, Exploration index of mice on day 2 of new object recognition test. * p < 0.05 vs. the AD + sh-NC group. # p < 0.05 vs. the AD + GPA + sh-NC group.


GAP43 silencing reduces GPA-mediated changes in Aβ deposition, neuronal apoptosis, neuroinflammation, and axonal injury

We subsequently investigated whether silencing of GAP43 could affect the function of GPA in Aβ deposition, neuronal apoptosis, neuroinflammation and axonal injury in AD mice. Based on the result of ELISAs, compared with the AD + sh-NC group, Aβ level in the AD + GPA + sh-NC group was dramatically decreased. However, the AD + GPA + sh-GAP43 group had opposite result (Figure 5A). The results of Th-S staining were identical to that of the ELISAs (Figure 5B). TUNEL staining results displayed that the apoptosis rate was markedly reduced in the AD + GPA + sh-NC group in comparison with the AD + sh-NC group, which was reverse in the AD + GPA + sh-GAP43 group relative to the AD + GPA + sh-NC group (Figure 5C). ELISA results showed that lower IL-6, TNF-α and IL-1β expression and higher IL-4 expression in the AD + GPA + sh-NC group than in the AD + sh-NC group but higher IL-6, TNF-α and IL-1β expression and lower IL-4 expression in the AD + GPA + sh-GAP43 group than in the AD + GPA + sh-NC group (Figure 5D). GAP43 was silenced in Aβ1-42-induced cells treated with GPA. The silence efficiency is shown in Figure 5E. sh-GAP43-1 (sh-GAP43 group) with better silence efficiency was selected for the next experiment. As shown in CCK-8 results, higher cell viability was noted in the GPA + Aβ + sh-NC group than in the Aβ + sh-NC group, which was reverse in the GPA + Aβ + sh-GAP43 group when compared to the GPA + Aβ + sh-NC group (Figure 5F). Immunofluorescence staining highlighted that the axon length of Aβ + GPA + sh-NC was longer than that of the Aβ + sh-NC group. Compared with the Aβ + GPA + sh-NC group, the shorter axon length was observed in the Aβ + GPA + sh-GAP43 group (Figure 5G). These results suggested that GAP43 knockdown inhibits the alleviating effects GPA treatment has on Aβ deposition, neuroinflammation, neuronal apoptosis and axonal injury in AD mice.

Figure 5. Silencing of GAP43 leads to an increase in Aβ deposition, neuroinflammation, neuronal apoptosis, and axonal injury

A, The Aβ level in the brain of each group was detected by ELISA. B, Th-s staining in brain tissue of mice in each group. C, TUNEL staining to evaluate the apoptosis rate of neurons in each group. D, The expression of inflammatory factors in cerebral cortex was detected by ELISA. * p < 0.05 vs. the AD + sh-NC group. # p < 0.05 vs. the AD + GPA + sh-NC group. E, The silencing efficiency of GAP43 (* p < 0.05 vs. the sh-NC group). F, CCK-8 assay for cell viability. G, The growth of axons observed by immunofluorescence staining. * p < 0.05 vs. the Aβ + sh-NC group. # p < 0.05 vs. the Aβ + GPA + sh-NC group. All cell experiments were repeated three times.


GPA alleviates cognitive impairment in AD mice by upregulating GAP43 through PI3K/AKT signaling

It has been reported in the literature that GPA can activate the PI3K/AKT pathway, and PI3K/AKT pathway can regulate the expression of GAP43 (15, 16). However, whether GPA can alleviate AD by activating PI3K/AKT/GAP43 regulatory axis remains unclear. Therefore, we speculated that GPA upregulates GAP43 expression through PI3K/AKT pathway to improve cognitive impairment and alleviate nerve injury in AD mice.
In order to verify this hypothesis, PI3K, AKT and p-AKT expression was examined by Western blot. This experiment revealed that, when compared with the WT group, PI3K and p-AKT expression was significantly decreased in the AD group, whereas the AD + GPA group displayed the opposite result (Figure 6A). PI3K inhibitor LY294002 was used to combine treatment with GAP43. Based on the result of Western blot, PI3K, GAP43 and p-AKT expression in the AD + GPA + LY294002 + oe-NC group was obviously reduced in contrast to the AD + GPA + oe-NC group. There was no significant difference in PI3K and p-AKT expression, but GAP43 expression was significantly increased in the AD + GPA + LY294002 + oe-GAP43 group (Figure 6B). These results indicate that activation of PI3K/AKT pathway upregulates GAP43 expression, and that GAP43 expression was markedly diminished following treatment with a PI3K inhibitor. MWM test result showed that the escape latency of mice was substantially shorter in the AD + GPA + oe-NC group than in the AD + GPA + LY294002 + oe-NC group, as well as shorter in the AD + GPA + LY294002 + oe-GAP43 group (Figure 6C). Exploratory test results described that the times of mice crossing platform were noticeably reduced in the AD + GPA + LY294002 + oe-NC group in comparison with the AD + GPA + oe-NC group, whilst the AD + GPA + LY294002 + oe-GAP43 group had reverse result relative to the AD + GPA + LY294002 + oe-NC group (Figure 6D). The swimming time in the target quadrant is exhibited in Figure 6E. Compared with the AD + GPA + oe-NC group, the swimming time of mice in the AD + GPA + LY294002 + oe-NC group was observably reduced. Compared with the AD + GPA + LY294002 + oe-NC group, the swimming time of mice in the AD + GPA + LY294002 + oe-GAP43 group was significantly enhanced. Based on the result of new object recognition test, the exploratory index of the AD + GPA + LY294002 + oe-NC group was significantly lower than that of the AD + GPA + oe-NC group when one of the objects was replaced by a new one. However, compared with the AD + GPA + LY294002 + oe-NC group, the exploration index of the AD + GPA + LY294002 + oe-GAP43 group was observably diminished (Figure 6F-G).
Collectively, these results indicate that GPA activates PI3K/AKT signaling which in turn elevates GAP43 expression, alleviating cognitive dysfunction in AD mice.

Figure 6. GPA induces GAP43 upregulation to attenuate cognitive dysfunction in AD mice through activation of the PI3K/AKT pathway

A, The PI3K, AKT and p-AKT expression in the brain of each group detected by Western blot. * p < 0.05 vs. the WT group. # p < 0.05 vs. the AD group. B, The PI3K, AKT, p-AKT and GAP43 expression in the brain of each group determined by Western blot. C, Escape latency of mice on different days in the positioning and navigation test. D, The Times of crossing the platform in each group. E, Swimming time of mice in target quadrant. F, Exploratory index of mice on day 1 of new object recognition test. G, Exploratory index of mice on day 2 of new object recognition test. * p < 0.05 vs. the AD + GPA + oe-NC group. # p < 0.05 vs. the AD + GPA + LY294002 + oe-NC group


GPA reduces Aβ levels, neuroinflammation, and neuronal damage in AD mice through activation of PI3K/AKT/GAP43 regulatory axis

The aim of this section of the study was to further verify the possible effects of GPA on via PI3K/AKT/GAP43 axis in the setting of Aβ deposition, neuroinflammation, neuronal apoptosis and axonal injury in AD mice. We adopted an ELISA approach, which showed that in comparison with the AD + GPA + oe-NC group, the Aβ level in the AD + GPA + LY294002 + oe-NC group was markedly augmented. However, compared with the AD + GPA + LY294002 + oe-NC group, the Aβ level in the AD + GPA + LY294002 + oe-GAP43 group was strikingly reduced (Figure 7A). The result of Th-S staining revealed a similar finding (Figure 7B). TUNEL staining indicated that apoptosis rate in the AD + GPA + LY294002 + oe-NC group was higher than that of the AD + GPA + oe-NC group, while the AD + GPA + LY294002 + oe-GAP43 group had opposite result relative to the AD + GPA + LY294002 + oe-NC group (Figure 7C). ELISA results documented lower IL-4 expression and higher IL-6, TNF-α, and IL-1β expression in the AD + GPA + LY294002 + oe-NC group than in the AD + GPA + oe-NC group, which was reverse in the AD + GPA + LY294002 + oe-GAP43 group relative to the AD + GPA + LY294002 + oe-NC group (Figure 7D). In a CCK-8 assay, when compared with the Aβ + GPA + LY294002 + oe-NC group, diminished cell viability was observed in the Aβ + GPA + LY294002 + oe-GAP43 group (Figure 7E). Western blot analysis demonstrated that in comparison with the DMSO group, p-AKT and GAP43 expression in the Aβ group was appreciably reduced; compared to the Aβ group, p-AKT and GAP43 expression was augmented in the Aβ + GPA + oe-NC group; p-AKT and GAP43 expression in the Aβ + GPA + LY294002 + oe-NC group was substantially lower than that of the AD + GPA + oe-NC group; in contrast to the AD + GPA + LY294002 + oe-NC group, p-AKT expression in the AD + GPA + LY294002 + oe-GAP43 group was not significantly different, but GAP43 expression was appreciably increased (Figure 7F). Immunofluorescence staining documented that relative to the Aβ + GPA + LY294002 + oe-NC group, the axon length of the Aβ + GPA + LY294002 + oe-GAP43 group was strikingly augmented (Figure 7G).
In summary, GPA upregulates GAP43 expression, leading to a decline in Aβ levels, inflammation, and neuronal damage in AD mice in a process that requires PI3K/AKT signaling pathway.

Figure 7. GPA activates the PI3K/AKT/GAP43 regulatory axis to reduce Aβ deposition, neuroinflammation, neuronal apoptosis, and axonal injury in AD mice

A, The Aβ level in the brain of each group detected by ELISA. B, Th-S staining in brain tissue of mice in each group. C, TUNEL staining to measure the apoptosis rate of neurons in each group. D, The inflammatory factors in cerebral cortex determined with ELISA. * p < 0.05 vs. the AD + GPA + oe-NC group. # p < 0.05 vs. the AD + GPA + LY294002 + oe-NC group. E, Cell viability assessed using CCK-8 (* p < 0.05 vs. the Aβ + GPA + LY294002 + oe-NC group). F, Western blot to detect the related protein expression (*p < 0.05 vs. the DMSO group. # p < 0.05 vs. the Aβ group, & p < 0.05 vs. the Aβ + GPA + oe-NC group, ^ p < 0.05 vs. the Aβ + GPA + LY294002 + oe-NC group). G, The growth of axons observed by immunofluorescence staining (* p < 0.05 vs. the Aβ + GPA + LY294002 + oe-NC group). All cell experiments were repeated three times



Pathologically, AD characterized by amyloid plaques which lead to dementia in increasing sections of the elderly population (17). GPA has been documented to have therapeutic effects on AD (7). However, at the time of publication, the mechanism underlying these beneficial effects remains unclear and required further investigation. Therefore, we adopted in vitro and in vivo AD model systems to examine the mechanism by which GPA acts in AD, revealing that GPA treatment activated the PI3K/AKT pathway, leading to an increase in GAP43 expression. This in turn thus improves cognitive function by reducing Aβ accumulation, neuronal apoptosis, neuroinflammation, and axonal injury, thus exerting a neuroprotective effect on AD mice.
We initially found that GPA improved cognitive impairment, and alleviated Aβ accumulation, neuronal apoptosis, neuroinflammation, and axonal injury in AD mice in a dose dependent manner. Concurring with our results, research conducted by Zhou et al. uncovered that memory deficits, cognitive impairment, Aβ deposition, and neuroinflammation were repressed after GPA treatment in AD mice (7). In addition, GPA treatment reduces cellular apoptosis in mice with D-galactosamine and lipopolysaccharide-induced hepatic failure (18). Moreover, neuronal geniposide activity, is widely recognized as a derivative of GPA (6), has been reported to be associated with increased gene expression related to cell growth and repair, and inhibited apoptosis and inflammation of neurons (19). Geniposide has also been demonstrated to repress amyloid deposition and behavioral and cognitive impairments of AD mice (20). Therefore, it is highly likely that GPA has neuroprotective effects on AD.
Further mechanistic analysis revealed that GAP43 is a target of GPA in AD. A prior study identified that after geniposide treatment, Neuro2a neuroblastoma cells had elevation of GAP43 expression, supporting our finding (8). Importantly, our data clarified that silencing of GAP43 negated the alleviating role of GAP in cognitive impairment, Aβ accumulation, neuronal apoptosis, neuroinflammation, and axonal injury in AD mice. As a presynaptic protein, GAP43 is overexpressed during neuronal development and synaptogenesis where it plays a crucial role in the orchestration of learning and memory functions, axonal outgrowth, and synaptic plasticity (21). Interestingly, the downregulation of GAP43 has been detected in the brain tissues of AD mice (22). Additionally, GAP43 is also involved in the decline of neuronal apoptosis in rats with spinal cord injury (23). A previous work uncovered the finding that GAP43 upregulation is associated with relief and suppression of inflammation in mouse diabetic retinopathy (24). Activation of GAP43 has also been shown to lead to attenuation of cognitive impairment in rats with subthreshold convulsant discharge (25). Research conducted by Liu et al. demonstrated that GAP43 overexpression causes the promotion of axonal regeneration in the spinal cord of rats with spinal cord injury (26). Based on our results and the existing literature, GAP43 overexpression might participate in the neuroprotective effects of GPA on AD.
Another central finding of this project was that GPA activates the PI3K/AKT pathway to upregulate GAP43, thus alleviating cognitive impairment by reducing Aβ accumulation, neuronal apoptosis, neuroinflammation, and axonal injury in AD mice. Concordant with our finding, GPA has been shown to activate the PI3K/AKT pathway in human melanocytes (15). It has also been elucidated in another work that activation of the PI3K/AKT pathway was capable of elevating GAP43 expression during hair cell protection in the neonatal murine cochlea (12). Intriguingly, activation of the PI3K/AKT pathway could contribute to the repression of cognitive impairment in AD rat (13). In line with our results, activation of the PI3K/AKT pathway ameliorates learning and memory dysfunction, the histology structure of damaged neurons in hippocampal area, and neuronal apoptosis in AD mice (27).
Our data demonstrates that GPA relieves cognitive impairment in a process that leads to a reduction in Aβ accumulation, neuronal apoptosis, neuroinflammation, and axonal injury in AD mice. Mechanistically, we have revealed that GPA exerts a neuroprotective effect on AD in mice via activation of the PI3K/AKT/GAP43 regulatory axis (summarized in Figure 8). This study uncovers a mechanism by which GPA suppresses AD progression, imparting an improved understanding of the pathogenesis of AD and providing novel potential therapeutic targets for AD treatment. However, considering the limitations of our study, more extensive research is required to investigate the specific mechanism of GPA in AD, with eventual translation to clinical trials. Absorption of GPA is affected by different processing methods in rats following crude Gardeniae Fructus administration based on a comparative pharmacokinetics of GPA, Geniposide, Genipin, and Crocetin (28). Moreover, the toxicity of GPA has not been discovered yet (29). Future studies should be conducted to foresee the future investigations of the therapeutic interest of this compound.

Schematic summarizing the molecular mechanism by which GPA alleviates nerve injury in AD mice through a PI3K/AKT/GAP43 regulatory axis


Funding: This work is supported by Clinical Research Special Fund of Qiqihar Academy of Medical Sciences (QMSI2020L-11).

Acknowledgements: We would like to give our sincere appreciation to the reviewers for their helpful comments on this article.

Declaration of conflicting interests: The authors declare no conflict of interest.

Ethical Statement: The experiment was approved by the animal ethics committee of Qiqihar Medical University.







1. Scheltens P, Blennow K, Breteler MM, et al. Alzheimer’s disease. Lancet 2016;388:505-517, doi: 10.1016/S0140-6736(15)01124-1.
2. Lane CA, Hardy J, Schott JM. Alzheimer’s disease. Eur J Neurol 2018;25:59-70, doi: 10.1111/ene.13439.
3. Robinson N, Grabowski P, Rehman I. Alzheimer’s disease pathogenesis: Is there a role for folate? Mech Ageing Dev 2018;174:86-94, doi: 10.1016/j.mad.2017.10.001.
4. Tiwari S, Atluri V, Kaushik A, Yndart A, Nair M. Alzheimer’s disease: pathogenesis, diagnostics, and therapeutics. Int J Nanomedicine 2019;14:5541-5554, doi: 10.2147/IJN.S200490.
5. Joe E, Ringman JM. Cognitive symptoms of Alzheimer’s disease: clinical management and prevention. BMJ 2019;367:l6217, doi: 10.1136/bmj.l6217.
6. Gao Y, Chen ZY, Liang X, Xie C, Chen YF. Anti-atherosclerotic effect of geniposidic acid in a rabbit model and related cellular mechanisms. Pharm Biol 2015;53:280-285, doi: 10.3109/13880209.2014.916310.
7. Zhou Z, Hou J, Mo Y, et al. Geniposidic acid ameliorates spatial learning and memory deficits and alleviates neuroinflammation via inhibiting HMGB-1 and downregulating TLR4/2 signaling pathway in APP/PS1 mice. Eur J Pharmacol 2020;869:172857, doi: 10.1016/j.ejphar.2019.172857.
8. Chen MK, Peng CC, Maner RS, Zulkefli ND, Huang SM, Hsieh CL. Geniposide ameliorated fluoxetine-suppressed neurite outgrowth in Neuro2a neuroblastoma cells. Life Sci 2019;226:1-11, doi: 10.1016/j.lfs.2019.04.003.
9. Zhang F, Ying L, Jin J, et al. GAP43, a novel metastasis promoter in non-small cell lung cancer. J Transl Med 2018;16:310, doi: 10.1186/s12967-018-1682-5.
10. Long ZM, Zhao L, Jiang R, et al. Valproic Acid Modifies Synaptic Structure and Accelerates Neurite Outgrowth Via the Glycogen Synthase Kinase-3beta Signaling Pathway in an Alzheimer’s Disease Model. CNS Neurosci Ther 2015;21:887-897, doi: 10.1111/cns.12445.
11. Liu F, Wang Y, Yao W, Xue Y, Zhou J, Liu Z. Geniposide attenuates neonatal mouse brain injury after hypoxic-ischemia involving the activation of PI3K/Akt signaling pathway. J Chem Neuroanat 2019;102:101687, doi: 10.1016/j.jchemneu.2019.101687.
12. Hayashi Y, Yamamoto N, Nakagawa T, Ito J. Insulin-like growth factor 1 induces the transcription of Gap43 and Ntn1 during hair cell protection in the neonatal murine cochlea. Neurosci Lett 2014;560:7-11, doi: 10.1016/j.neulet.2013.11.062.
13. Kamel AS, Abdelkader NF, Abd El-Rahman SS, Emara M, Zaki HF, Khattab MM. Stimulation of ACE2/ANG(1-7)/Mas Axis by Diminazene Ameliorates Alzheimer’s Disease in the D-Galactose-Ovariectomized Rat Model: Role of PI3K/Akt Pathway. Mol Neurobiol 2018;55:8188-8202, doi: 10.1007/s12035-018-0966-3.
14. Biskind MS, Schreier H. On the significance of nutritional deficiency in diabetes. Exp Med Surg 1945;3:299-316.
15. Lu W, Zhao Y, Kong Y, et al. Geniposide prevents H2 O2 -induced oxidative damage in melanocytes by activating the PI3K-Akt signalling pathway. Clin Exp Dermatol 2018;43:667-674, doi: 10.1111/ced.13409.
16. Zhang L, Yue Y, Ouyang M, Liu H, Li Z. The Effects of IGF-1 on TNF-alpha-Treated DRG Neurons by Modulating ATF3 and GAP-43 Expression via PI3K/Akt/S6K Signaling Pathway. Neurochem Res 2017;42:1403-1421, doi: 10.1007/s11064-017-2192-1.
17. Gao LB, Yu XF, Chen Q, Zhou D. Alzheimer’s Disease therapeutics: current and future therapies. Minerva Med 2016;107:108-113.
18. Kim SJ, Kim KM, Park J, Kwak JH, Kim YS, Lee SM. Geniposidic acid protects against D-galactosamine and lipopolysaccharide-induced hepatic failure in mice. J Ethnopharmacol 2013;146:271-277, doi: 10.1016/j.jep.2012.12.042.
19. Liu W, Li G, Holscher C, Li L. Neuroprotective effects of geniposide on Alzheimer’s disease pathology. Rev Neurosci 2015;26:371-383, doi: 10.1515/revneuro-2015-0005.
20. Zhang Z, Wang X, Zhang D, Liu Y, Li L. Geniposide-mediated protection against amyloid deposition and behavioral impairment correlates with downregulation of mTOR signaling and enhanced autophagy in a mouse model of Alzheimer’s disease. Aging (Albany NY) 2019;11:536-548, doi: 10.18632/aging.101759.
21. Sandelius A, Portelius E, Kallen A, et al. Elevated CSF GAP-43 is Alzheimer’s disease specific and associated with tau and amyloid pathology. Alzheimers Dement 2019;15:55-64, doi: 10.1016/j.jalz.2018.08.006
22. Reutzel M, Grewal R, Silaidos C, et al. Effects of Long-Term Treatment with a Blend of Highly Purified Olive Secoiridoids on Cognition and Brain ATP Levels in Aged NMRI Mice. Oxid Med Cell Longev 2018;2018:4070935, doi: 10.1155/2018/4070935.
23. Ni Y, Gu J, Wu J, Xu L, Rui Y. MGMT-Mediated neuron Apoptosis in Injured Rat Spinal Cord. Tissue Cell 2020;62:101311, doi: 10.1016/j.tice.2019.101311.
24. Mohammad HMF, Sami MM, Makary S, Toraih EA, Mohamed AO, El-Ghaiesh SH. Neuroprotective effect of levetiracetam in mouse diabetic retinopathy: Effect on glucose transporter-1 and GAP43 expression. Life Sci 2019;232:116588, doi: 10.1016/j.lfs.2019.116588.
25. Wang MJ, Jiang L, Chen HS, Cheng L. Levetiracetam Protects Against Cognitive Impairment of Subthreshold Convulsant Discharge Model Rats by Activating Protein Kinase C (PKC)-Growth-Associated Protein 43 (GAP-43)-Calmodulin-Dependent Protein Kinase (CaMK) Signal Transduction Pathway. Med Sci Monit 2019;25:4627-4638, doi: 10.12659/MSM.913542.
26. Liu H, Xiong D, Pang R, et al. Effects of repetitive magnetic stimulation on motor function and GAP43 and 5-HT expression in rats with spinal cord injury. J Int Med Res 2020;48:300060520970765, doi: 10.1177/0300060520970765.
27. Yao Y, Wang Y, Kong L, Chen Y, Yang J. Osthole decreases tau protein phosphorylation via PI3K/AKT/GSK-3beta signaling pathway in Alzheimer’s disease. Life Sci 2019;217:16-24, doi: 10.1016/j.lfs.2018.11.038.
28. Yang X, Li J, Yang X, He J, Chang YX. Comparative Pharmacokinetics of Geniposidic Acid, Genipin-1-beta-Gentiobioside, Geniposide, Genipin, and Crocetin in Rats after Oral Administration of Crude Gardeniae Fructus and Its Three Processed Products Using LC-MS/MS. Evid Based Complement Alternat Med 2020;2020:1642761, doi: 10.1155/2020/1642761.
29. Khanal T, Kim HG, Choi JH, et al. Biotransformation of geniposide by human intestinal microflora on cytotoxicity against HepG2 cells. Toxicol Lett 2012;209:246-254, doi: 10.1016/j.toxlet.2011.12.017.




J. Cummings1, P. Aisen2, L.G. Apostolova3, A. Atri4, S. Salloway5, M. Weiner6


1. Chambers-Grundy Center for Transformative Neuroscience, Department of Brain Health, School of Integrated Health Sciences, University of Nevada Las Vegas (UNLV), Las Vegas, NV, USA; 2. Alzheimer’s Treatment Research Institute, University of Southern California, San Diego, CA, USA; 3. Departments of Neurology, Radiology, Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana, USA; 4. Banner Sun Health Research Institute, Banner Health, Sun City, AZ; Center for Brain/Mind Medicine, Harvard Medical School, Boston, MA, USA; 5. Butler Hospital and Warren Alpert Medical School of Brown University, Providence RI, USA; 6. Departments of Radiology and Biomedical Imaging, Medicine, Psychiatry and Neurology, University of California San Francisco, San Francisco, CA, USA

Corresponding Author: Jeffrey Cummings, MD, ScD, 1380 Opal Valley Street, Henderson, NV 89052,, T: 702-902-3939

J Prev Alz Dis 2021;4(8):398-410
Published online July 20, 2021,



Aducanumab has been approved by the US Food and Drug Administration for treatment of Alzheimer’s disease (AD). Clinicians require guidance on the appropriate use of this new therapy. An Expert Panel was assembled to construct Appropriate Use Recommendations based on the participant populations, conduct of the pivotal trials of aducanumab, updated Prescribing Information, and expert consensus. Aducanumab is an amyloid-targeting monoclonal antibody delivered by monthly intravenous infusions. The pivotal trials included patients with early AD (mild cognitive impairment due to AD and mild AD dementia) who had confirmed brain amyloid using amyloid positron tomography. The Expert Panel recommends that use of aducanumab be restricted to this population in which efficacy and safety have been studied. Aducanumab is titrated to a dose of 10 mg/kg over a 6-month period. The Expert Panel recommends that the aducanumab be titrated to the highest dose to maximize the opportunity for efficacy. Aducanumab can substantially increase the incidence of amyloid-related imaging abnormalities (ARIA) with brain effusion or hemorrhage. Dose interruption or treatment discontinuation is recommended for symptomatic ARIA and for moderate-severe ARIA. The Expert Panel recommends MRIs prior to initiating therapy, during the titration of the drug, and at any time the patient has symptoms suggestive of ARIA. Recommendations are made for measures less cumbersome than those used in trials for the assessment of effectiveness in the practice setting. The Expert Panel emphasized the critical importance of engaging in a process of patient-centered informed decision-making that includes comprehensive discussions and clear communication with the patient and care partner regarding the requirements for therapy, the expected outcome of therapy, potential risks and side effects, and the required safety monitoring, as well as uncertainties regarding individual responses and benefits.

Key words: Alzheimer’s disease, aducanumab, Aduhelm™, appropriate use, titration, ARIA, amyloid imaging, MRI.


Aducanumab (Aduhelm™) has been approved by the US Food and Drug Administration (FDA) for the treatment of Alzheimer’s disease (AD). The Prescribing Information for aducanumab (1) provides key facts on aducanumab such as dose, titration, pharmacokinetics, and side effects. The Clinical Studies section describes the clinical trials that led to the approval of aducanumab. Many details of the clinical use of this new agent are not detailed in the Prescribing Information (1) and there is a need for specific recommendations regarding how to use aducanumab appropriately. Experts with experience in AD research, AD clinical trials and drug development, AD clinical care, and use of aducanumab were assembled to develop consensus recommendations for the appropriate use of aducanumab in clinical practice.
The Prescribing Information (1) provides the “on label” prescribing instructions. The Expert Panel recommends that the appropriate use of aducanumab in real-world clinical practice should pragmatically mimic the use of aducanumab in the EMERGE and ENGAGE clinical trials that led the FDA to approve aducanumab. After the initial Prescribing Information was published, the FDA adjusted the indication section from “indicated for the treatment of Alzheimer’s disease” to “indicated for the treatment of Alzheimer’s disease…should be initiated in patients with mild cognitive impairment or mild dementia stage of the disease, the population in which treatment was initiated in clinical trials” (1, 2). Some of the Expert Panel recommendations are more specific or more restrictive than the information provided in the Prescribing Information (1). The recommendations are within the scope of use articulated in the Prescribing Information (1) The Expert Panel describes the appropriate use of aducanumab for the practicing clinician; we do not address trial outcomes, approval strategies, cost, insurance coverage, or reimbursement issues. The Expert Panel recommendations apply to practices in the Unites States where aducanumab is currently approved. Recommendations may change as more data on the use of aducanumab and more data from the trials become available. These recommendations are meant to assist practitioners in using aducanumab safely; they do not replace clinician judgement in the delivery of care to individual patients.



Aducanumab is a monoclonal antibody directed to the N-terminus of the amyloid beta peptide (Aß). It was derived through a process of reverse translation in which blood lymphocytes from healthy elderly individuals who were cognitively normal or had unusually slow cognitive decline served as a source of antibody genes for the generation of recombinant human antibodies (3).
The Expert Panel recommends that patients treated with aducanumab closely resemble those included in the pivotal clinical trials (4, 5). Pragmatic adjustments will be required for use of aducanumab outside of the trial setting, and the translation of clinical trial protocol requirements to clinical practice is summarized in Table 1. Efficacy and safety have been assessed in the early AD population of patients with mild cognitive impairment (MCI) due to AD and mild dementia due to AD confirmed by amyloid positron emission tomography (PET) and are unknown for individuals with preclinical AD, those with more severe AD dementia, or those with cognitive impairment that is not confirmed to be AD by Aß studies.

Table 1. Clinical trial enrollment criteria and appropriate use criteria for aducanumab in clinical practice

Aß – amyloid beta protein; AD – Alzheimer’s disease; APOE – apolipoprotein E; CDR – Clinical Dementia Rating; cm – centimeter; CSF – cerebrospinal fluid; HIV – human immunodeficiency virus; MMSE – Mini Mental State Examination; MoCA – Montreal Cognitive Assessment; MRI – magnetic resonance imaging; PET – positron emission tomography; RBANS – Repeatable Battery for the Assessment of Neuropsychological Status; TIA – transient ischemic attack


Appropriate Patient


The Expert Panel recommends that patients appropriate for treatment with aducanumab have a diagnosis of early AD established by a diagnostic evaluation that includes: 1) detailed history that is sufficient to establish the nature and time course of cognitive symptoms, functional changes, and behavioral status; 2) objective corroboration of cognitive decline using standardized testing; 3) detailed neurological and physical examination; 4) review of all current medications and supplements; 5) laboratory testing sufficient to exclude other concomitant disorders that can cause cognitive decline including a complete blood count, electrolyte panel, thyroid stimulating hormone, lipids and triglycerides, liver function tests, and serum vitamin B12 level; and 6) magnetic resonance imaging (MRI) of the brain to rule out other conditions that could present with cognitive decline (e.g., normal pressure hydrocephalus, vascular dementia, slow going neoplasm, subdural hematoma) and to assess possible exclusions for use of aducanumab (discussed below) (6-8). This assessment will determine if the patient has clinical findings consistent with early AD.
Patients with early AD meet the clinical criteria of stage 3 and 4 of the FDA staging approach (9). Stage 3 consists of individuals with subtle or more apparent detectable abnormalities on sensitive neuropsychological measures and mild but detectable functional impairment. The functional impairment in this stage is not severe enough to warrant a diagnosis of overt dementia. Stage 4 includes individuals with cognitive impairment and mild but definite functional decline.
To quantify the cognitive and functional changes, early AD patients in the aducanumab trials had scores on the Clinical Dementia Rating (CDR) (10) global rating of 0.5. This instrument assesses cognitive (memory, orientation, judgment, and problem solving) and functional (community affairs, home and hobbies, and personal care) domains. In addition, trial participants had Mini Mental State Examination (MMSE) (11) scores of 24-30. The MMSE is commonly used in clinical practice and is a useful tool for identifying appropriate patients. The standard error of measurement on the MMSE is 1 point, and the minimum detectable difference is 3 points (12, 13). The test-retest reliability of MMSE is 2-4 points (14). These studies indicate that scores of 21 and higher would not be detectably different from the range of MMSE scores of patients included in the pivotal trials (MMSE range of 24-30). The Phase 1B study of aducanumab had encouraging results in patients with MMSE scores of 20-30 (15). The Expert Panel recommends that patients with MMSE scores of 21 or higher or who have a similar level of performance on an alternate reliable and valid assessment are appropriate for treatment with aducanumab. An alternative assessment that provides reliable information similar to that of the MMSE is the Montreal Cognitive Assessment (MoCA) (16). The MoCA is a more challenging test than the MMSE resulting in lower scores when compared to the MMSE. Scores of 17 and higher on the MoCA are equivalent to MMSE scores of 21-30 in early symptomatic AD (17). In settings where neuropsychological testing is available, a diagnosis of early AD can be based on more extensive cognitive, functional, and behavioral assessments (18).

Use of cognitive enhancing agents in aducanumab candidates

A newly diagnosed patient with MCI due to AD may be started on aducanumab since cholinesterase inhibitors and memantine are not approved for this stage of AD. Patients with early AD may be on a cholinesterase inhibitor or memantine when referred for possible treatment with aducanumab; these patients can remain on their standard of care while being treated with aducanumab. Patients diagnosed with mild AD dementia can have treatment with aducanumab before or following initiation of treatment with a cholinesterase inhibitor. If patients with MCI progress to mild AD dementia, treatment with a cholinesterase inhibitor (donepezil, rivastigmine, galantamine) can be considered. Memantine is not approved for mild AD dementia. If patients progress to moderate or severe AD, memantine treatment can be considered as monotherapy or in conjunction with a cholinesterase inhibitor (19).

Amyloid status

All patients included in the pivotal trials had positive amyloid positron emission tomography (PET). Demonstration of amyloid burden is critical to establishing the presence of the target for amyloid lowering therapies. The clinical diagnosis of AD is often not confirmed by amyloid studies and up to 40% of patients diagnosed with early AD do not have the amyloid pathology when studied with amyloid imaging (20). Appropriate Use Criteria of amyloid imaging suggest that the imaging is appropriate when: a) there is a cognitive complaint and cognitive impairment has been objectively confirmed impairment; b) AD is a possible diagnosis, but the diagnosis is uncertain after a comprehensive evaluation by a dementia expert; and c) knowledge of the presence or absence of amyloid-beta pathology is expected to increase diagnostic certainty and alter management (21). These criteria are fulfilled in the situation where a patient is being considered for treatment with aducanumab: they have the symptoms of early AD, additional diagnostic certainty is needed, and management will be based on the outcome.
Three amyloid PET tracers are approved by the FDA: florbetapir, florbetaben, and flutametamol (22-24). Table 2 provides the criteria for a positive scan for each tracer. Scan interpretation is best done by radiologists or nuclear medicine specialists; training programs for amyloid PET interpretation are available for each ligand. The Expert Panel recommends that programs offering aducanumab treatment and using amyloid PET to confirm the diagnosis of AD should ensure the availability of individuals properly trained in amyloid PET interpretation.


Table 2. Criteria for a positive amyloid PET for the three approved amyloid PET tracers (from drugs@FDA: FDA-Approved Drugs)


Lumbar puncture and assessment of cerebrospinal fluid (CSF) biomarkers (Aβ42, Aβ40 total tau, phosphorylated tau [p-tau]) provide an alternative to amyloid PET and are more widely available (25). Several CSF measures can be indicative of the presence of AD including low Aβ42, low Aβ42/Aβ40 ratio, abnormal Aβ42/tau ratios, and abnormal Aβ42/p-tau ratios (26-28). Practitioners should use Clinical Laboratory Improvement Amendments (CLIA)-certified facilities and follow the laboratory’s guidelines for optimal AD-related assays. If CSF results are ambiguous, amyloid imaging is recommended. Amyloid PET and CSF AD signature studies provide equally valid information (29); CSF Aβ42 levels correlate inversely with brain amyloid on PET with CSF levels declining as Aβ is deposited in the cortex (30). Changes in CSF Aβ42 levels precede changes in amyloid PET (31); individuals with abnormal CSF and normal amyloid PET imaging are usually without symptoms and they lack evidence of amyloid plaques which are the target of aducanumab. The Expert Panel recommends that these patients not be treated with aducanumab. Re-imaging with amyloid PET in 1-3 years may be warranted in this group of individuals.
Lumbar puncture can be performed by physicians, nurse practitioners, or physicians’ assistants/associates with low patient morbidity and high safety (32). Lumbar puncture may not be possible in those with pathological or surgical changes of the lumbar spine; fluoroscopic guidance may be useful in such cases. Lumbar puncture is contraindicated in those with clotting disorders or who are on anticoagulants. Prothrombin time (PT) and partial thromboplastin time (PTT) can be obtained to ensure normal clotting parameters before proceeding with lumbar puncture.
Amyloid imaging or CSF biomarker analyses in persons with the clinical features of early AD will reveal that some of these cognitively impaired individuals do not have AD, exhibit evidence of neurodegeneration, and fulfill criteria for suspected non-Alzheimer pathology (SNAP) (33). Discovery of the non-amyloid status of these individuals assists clinicians in making management decisions (34). The Expert Panel recommends that individuals with SNAP not be treated with aducanumab.
Lumbar puncture with findings consistent with AD or PET with elevated brain amyloid confirm the diagnosis of AD in patients with the clinical syndrome of early AD. Failure to confirm the diagnosis of AD with amyloid biomarkers could result in administering aducanumab to patients who do not have AD and who lack the target pathology of the agent. The Expert Panel recommends that all patients considered for treatment with aducanumab have the diagnosis of AD confirmed by clinically validated amyloid studies such as amyloid PET or CSF analysis.

Genetic testing

Genetic testing to determine the apolipoprotein E (APOE) genotype of the participants was required in the pivotal trials. ARIA of the effusion (ARIA-E) or hemorrhagic (ARIA-H) type are more common in APOE ε 4 (APOE-4) gene carriers and understanding this effect in trials is important (35). ARIA may be more common in APOE-4 homozygotes and can be severe (36). The Prescribing Information (1) instructions for use of aducanumab do not require APOE genotyping and the dosing and monitoring of individuals with and without an APOE-4 allele are identical. The Expert Panel recommends that patients and care partners be engaged in a patient-centered discussion of the risk that an APOE-4 genotype confers for the risk of ARIA. This discussion will determine if genotype information would influence their decision to be treated with aducanumab and if they wish to pursue APOE genotyping.
If patients, care partners, or referring clinicians request APOE genotyping prior to the decision to use aducanumab or if the individual has determined their genotype through a commercial service, the Expert Panel recommends that the clinician be prepared to discuss the increased risk for ARIA in the presence of an APOE-4 allele as well as the consequences, monitoring, and management of ARIA if it occurs (discussed below). Genotyping provides transgenerational information on risk of AD for first degree relatives. Parents, siblings, and children of APOE-4 heterozygotes have a 50% chance of being an APOE-4 carrier with an increased risk of AD, and first-degree relatives of APOE-4 homozygotes have a 100% chance of being APOE-4 carriers and have an increased risk of AD. Clinicians may request genetic counseling to assist patients and caregivers in understanding the implications of their genotype (37, 38).

Neurological, medical, and psychiatric illness

The Expert Panel recommends that patients with neurological disorders that could account for or contribute to the clinical syndrome of the patients not be treated with aducanumab. This would include patients with parkinsonism, evidence of stroke or widespread white matter ischemic changes, or rapidly progressive dementia. Similarly, recent major psychiatric illness may compromise the ability to adhere to therapy and treatment should be deferred until behavioral stability is established. Poorly controlled or serious medical illnesses (e.g., cancer, heart failure) were exclusions for trial participation and if such illnesses are present in an individual being considered for treatment with aducanumab, the medical condition should be managed and stable prior to initiating treatment. Exclusionary factors are often less rigorous in routine care than in clinical trials but should not be so different as to threaten the generalizability of the trial results to the patient or increase the risk of treatment (39).
Aducanumab has not been studied for its reproductive or teratogenic effects and aducanumab should be administered to younger sexually active AD patients only if they are using contraceptive methods.

Clotting status

Aducanumab is associated with ARIA. Patients with evidence of microhemorrhage on MRI (discussed below) or with clotting abnormalities or who were on anticoagulants were excluded from the pivotal trials. It is not known if these exclusions affected the rate of microhemorrhage associated with aducanumab therapy. The risk of severe ARIA in a person receiving anticoagulants or with a clotting disorder is sufficient to exclude them from treatment with aducanumab. Platelet anti-aggregation agents are allowable as concomitant therapy. Lumbar puncture for confirmation of amyloid status should not be performed on patients being treated with anticoagulants; the occurrence of perispinal hemorrhage and spinal cord compression are low but can occur and the risk should be avoided (40).

Concomitant Medications

There are no adverse drug-drug interactions noted in the Prescribing Information (1). Drugs used in routine care of patients with AD were allowed to be used by participants in the pivotal trials. The Expert Panel agreed that aducanumab may be co-administered with other drugs used in the treatment of AD including cholinesterase inhibitors (donepezil, rivastigmine, galantamine), memantine, and psychotropic agents (antidepressants, antipsychotics, hypnotics).

MRI prior to initiating treatment

Concern for the occurrence of ARIA motivated avoiding administration of aducanumab to patients who had evidence of substantial cerebrovascular disease at baseline in the pivotal trials. The protocol excluded patients who had acute or subacute hemorrhage, macrohemorrhage, greater than 4 microhemorrhages, cortical infarction (>1.5 cm), 1 lacunar infarction (>1.5 cm), diffuse white matter disease, or any areas of superficial siderosis (41). The Expert Panel recommends that these exclusions be observed in clinical practice when choosing appropriate patients for treatment with aducanumab. An MRI including T1, T2 or fluid attenuated inversion recovery (FLAIR), T2* gradient recalled echo (GRE) sequences or susceptibility weighted imaging (SWI), and diffusion weighted imaging should be obtained within 1 year of initiating treatment with aducanumab (and more recently if there is any evidence of stroke since the last MRI). A 3-Tesla magnet MRI will reveal more microhemorrhages than a 1.5 Tesla magnet device, and SWI sequences will reveal more ARIA than GRE images (42). Changes from a baseline scan is the basis for ARIA-related decision making, and the Expert Panel recommends that practitioners use the same MRI device with the same imaging protocol for a given patient whenever possible to assist in comparing the images. Computerized tomography (CT) does not provide sufficient information to determine risk at baseline or to monitor ARIA; individuals who cannot have an MRI (e.g., have a pacemaker incompatible with MRI, metallic brain vessel aneurysm clip, or metallic object in an eye) should not be treated with aducanumab.

Knowledgeable engagement

In the clinical trials of aducanumab, informed consent from the patient and care partner were required for participation. In clinical care, formal informed consent is not required but a similar approach should be used to ensure that the patient and care partner/family member/companion understand the requirements for treatment and the expected outcome of therapy. Patients with early AD have the cognitive capacity to comprehend the possible benefit or harms of aducanumab treatment. Key aspects of informed therapy include discussion of requirements for monthly infusions and periodic MRI and the risk of adverse events including ARIA. The anticipated duration of therapy is indefinite and longer treatment with disease-modifying agents is expected to have greater effects on the disease course (43); the optimal duration of therapy is unknown and it may be possible to reduce the frequency of infusions when amyloid levels have been substantially reduced but this has not yet been determined. Those considering aducanumab therapy should understand that the expected benefit is slowing of cognitive and functional decline; improvement of the current clinical state is not anticipated. Patients should have disease state education regarding the course of AD and the availability of cognitive enhancing agents. Educational programs can improve mood, reduce anxiety, and ameliorate caregiver burden (44). The Expert Panel recommends that appropriate use of aducanumab includes providing information on the requirements for treatment and the expected outcomes, potential risks and side effects, and burdens related to administration and monitoring.
Special efforts are required to engage minority patients and to communicate the need for care and the opportunities for treatment. Minority patients report being “unheard” in medical conversations (45). Historically, use of AD therapies such as cholinesterase inhibitors has been less in African American, Latino, and Asian populations than among White AD patients (46). Addressing concerns about the deleterious effects and stigma of diagnosis and raising awareness of potential benefits of disease identification and treatment may influence the willingness of minority patients to discuss cognitive symptoms with clinicians (47). Minority patients often prefer clinicians who share their language and culture (48). The Expert Panel recommends that clinicians strive to engage diverse patients in diagnosis and treatment discussions with the goal of achieving equity among diverse groups in the use of aducanumab.

Appropriate Treatment

Aducanumab initiation and Titration

Aducanumab infusions are done monthly and require approximately one hour to complete. Infusions should be at least 21 days apart. The first and 2nd infusion dose is 1 mg/kg; the 3rd and 4th infusions are with doses of 3 mg/kg; the 5th and 6th infusions are dosed at 6 mg/kg; the 7th infusion and beyond involve monthly infusions of 10 mg/kg (Figure 1). Aducanumab is supplied in vials of 170 mg/1.7 mL or 300 mg/3 mL and is added to an infusion bag of 100 mL of 0.9% sodium chloride. The data from the pivotal trials and the Phase 1B trial of aducanumab suggest that 10 mg/kg is the target dose (15). Lower doses may not produce benefit and may cause ARIA. The Expert Panel recommends that patients be titrated to 10 mg/kg. If that is not possible, the clinician should engage in a patient-centered discussion as to whether to continue treatment with lower doses of aducanumab.

Figure 1. Aducanumab dosing and MRI monitoring schedule (Prescribing Information (1) and Expert Panel recommendation; © J Cummings; illustrator M de la Flor, PhD)


Management of missed doses has not been studied. The Expert Panel recommends that if a patient misses a dose, the next infusion should be administered as soon as possible at the dose administered in the previous infusion. If a patient misses three or more doses and requires continued treatment, titration should be re-initiated beginning at a dose level one step below that previously administered (e.g., if the patient was at 6 mg/kg previously, they would resume at a dose level of 3 mg/kg) with the dose increased every other month as described for treatment initiation.
Infusions may be done in a clinician’s office; in general infusion centers providing intravenous (IV) therapies to patients with cancer, arthritis, or other disorders; in specialized aducanumab infusion facilities; or at home. Home infusions are administered by a visiting nurse. General infusion center personnel may not be familiar with interacting with cognitively impaired patients and may require specialized training to ensure that the patient has a positive experience fostering a sense of well-being and conducive to treatment adherence. Clinicians should ask patients about any recent symptoms suggestive of ARIA before each infusion. Evidence of coagulopathy, symptoms suggestive of stroke, or poorly controlled blood pressure may be reasons to defer therapy and reevaluate the patient.

ARIA monitoring and management

The most common adverse event produced by aducanumab is ARIA. Aducanumab is associated with a substantially increased rate of ARIA compared to rates observed in natural history studies or trial placebo groups. ARIA (ARIA-E and ARIA-H) occurred in 35.2% of patients on high dose aducanumab compared to an occurrence rate of 2.7% in the placebo group (Table 3) (5). Among those receiving aducanumab, ARIA-E was most commonly observed in participants who were APOE-4 gene carriers (43%) and least often in those without the APOE-4 gene (20.3%). Both symptomatic and asymptomatic ARIA are more common in APOE-4 gene carriers. Thirty percent of ARIA-E were mild (< 5cm on FLAIR imaging with hyperintensity confined to one location); 58% were moderate (5-10 cm involving more than one location); and 13% were severe (> 10 cm) (2). Most ARIA occurs in the first 8 months of treatment during the titration period but can occur any time in the treatment course. ARIA was successfully managed in most patients participating in the pivotal trials without discontinuing treatment; ARIA led to discontinuation from the trials in 6.2% of patients on aducanumab and 0.6% of patients on placebo.

Table 3. Occurrence of ARIA in the entire population and in participants with and without the APOE-4 allele in the two pivotal trials combined (10 mg/kg dose) (5)


Most ARIA events (74%) detected by MRI have no accompanying symptoms. Among those with symptomatic ARIA, symptoms were mild in 67.7%, moderate in 28.3%, and severe in 4% (4). The most common symptoms reported were confusion or altered mental status (5%), dizziness (4%), visual disturbances (2%), and nausea (2%) (2). ARIA episodes typically resolved in 4-16 weeks.

MRIs should be obtained at least 1 year prior to the initiation of treatment and more recently (preferably within 6 months) if there is any suggestion of an intervening central nervous system event (e.g., sudden worsening, transient ischemic attacks). After treatment initiation, MRIs should be obtained before the 5th infusion (before initiating the 6 mg/kg dose); prior to the 7th infusion (before infusion of the first dose of 10 mg/kg); and before the 12th infusion (e.g., before the 6th dose of 10 mg/kg). Given the rate of ARIA-E with the 10 mg/kg dose in the phase 3 studies, especially among APOE-4 carriers, some clinicians may decide to obtain an MRI before the 10th dose, after 3 doses of 10 mg/kg have been administered to avoid failure to detect ARIA that may require active management. MRI studies for ARIA should include FLAIR, T2* GRE and quick DWI. An optional 4th sequence would be either 3D T1 or 3D T2 SPACE (depending on the type of MRI available). In addition to these scheduled MRIs, patients should have an MRI whenever they have symptoms suggestive of ARIA such as headache, vomiting and/or nausea, confusion, dizziness, visual disturbance, gait difficulties, loss of coordination, tremor, transient ischemic attack, new onset seizures, or significant and unexpected acute cognitive decline.
If patients with ARIA (ARIA-E or ARIA-H) have symptoms, treatment should be suspended, and a clinical assessment and neurological examination performed (Figure 2). MRI should be repeated in 1 month; if the ARIA-E has resolved or the ARIA-H is stabilized, treatment can be resumed. If ARIA-E has not resolved and ARIA-H is worsening, treatment is withheld, and monthly MRIs obtained until treatment can be re-initiated or a decision is made to terminate treatment. If three or more doses are missed before restarting aducanumab, the dose should be re-titrated as described above. Aducanumab should not be re-initiated in patients with severe symptomatic ARIA (e.g., seizure, stroke-like syndromes).

Figure 2. Management strategy for ARIA. Patients with severe symptomatic ARIA are not re-titrated and are not candidates for further treatment with aducanumab (Expert Panel recommendation; © J Cummings; illustrator M de la Flor, PhD)


If patients are asymptomatic and the MRI reveals severe or moderate ARIA-E or severe or moderate ARIA-H (Table 4), treatment is suspended, and management follows the procedures described for patients with symptoms (Figure 2). If asymptomatic patients have mild ARIA-E or mild ARIA-H, treatment is continued, and MRIs are obtained at monthly intervals until ARIA-E is resolved or ARIA-H is stable. There is limited information on best practices for management of moderate ARIA-E or moderate ARIA-H and recommendations may evolve.

Table 4. MRI severity levels of ARIA-E and ARIA-H as described in the aducanumab Prescribing Information (1)


Clinicians providing aducanumab need access to MRI facilities and to radiologists familiar with detection and reporting of ARIA-E and ARIA-H. Inexperienced readers may fail to detect signs of ARIA when interpreting scans (35, 49). CT is not sufficient for ARIA monitoring.

Non-ARIA side effect monitoring

Overall adverse events were experienced by 86.9% of patients on placebo and 91.6% of patients on high dose aducanumab in the pivotal trials (5). Adverse events reported more often in patients receiving aducanumab included headache (20.5% vs 15.2% in placebo), falls (15% vs 11.8% in placebo), and diarrhea (8.9% vs 6.8% in placebo). Serious adverse events occurred in 13.9% of patients on placebo and 13.6% of patients receiving aducanumab. There were 5 fatalities among patients on placebo and 8 among those on aducanumab. The Expert Panel recommends vigilance for all potential side effects in patients treated with aducanumab with special attention to headache, falls, and diarrhea.

Effectiveness monitoring

Efficacy was assessed in the pivotal trials using the Clinical Dementia Rating – Sum of Boxes (CDR-sb) (10), Alzheimer’s Disease Assessment Scale – Cognitive Subscale (ADAS-cog) (50), Alzheimer’s Disease Cooperative Study Activities of Daily Living MCI version (ADCS-ADL-MCI) scale (51), MMSE (11), and the Neuropsychiatric Inventory (NPI) (52). These tools were used to assess patients directly (ADAS-Cog; MMSE; portions of the CDR-sb) or through interviews with care partners (ADCS-ADL-MCI; NPI; parts of the CDR-sb). The time of administration of this panel is approximately 2 hours and some of the instruments take substantial training and experience to be administered reliably (e.g., CDR-sb) (53). Use of such a battery is impractical in many medical or neurological practice settings. Objective assessments requiring less time and training may provide insight in the patient’s course; and the clinician should employ tools commonly used in practice. No improvement in cognition or function is anticipated with disease modifying therapy (DMT); slowing of decline and prolongation of the optimal clinical state is the goal of treatment (43). The heterogeneity of decline in early AD makes it difficult to conclude that a slowly progressive disorder is being slowed more by aducanumab (54).
Several means of monitoring treatment effects in the open label practice environment can be considered. The mean change on the MMSE over twelve months in the placebo group in PRIME was (-2.5), in ENGAGE (-3.5), and in EMERGE (-3.3). This provides a range of scores against which the decline in the patient on aducanumab might be compared. The drug-placebo differences observed in EMERGE may guide clinician expectations for the impact of aducanumab on disease progression: this included 18%-27% differences on cognitive decline, 40% difference on functional decline, and 87% difference in behavioral changes. The decline in the late period of MCI due to AD is predictable based on observations in the early MCI period (55). The clinician and care partner may observe differences in the rate of change when aducanumab is introduced and titrated to the 10 mg/kg dose.
The MMSE (11) is commonly used in clinical settings and may be used to monitor patients treated with aducanumab. The MoCA is an alternative to the MMSE (16). The AD8 is a brief informant interview assessing orientation, judgement, memory, and function (56, 57). The AD8 has been shown to have concurrent validity with the CDR used in the pivotal trials and distinguishes patients with MCI (CDR 0.5) from normal elderly with sensitivity of 74% and specificity of 86%. The NPI-Questionnaire is a brief version of the NPI that can be completed by the informant and reviewed by the clinician (58). These three tools are related to or derived from instruments used in the aducanumab pivotal trials. The Functional Activities Questionnaire (FAQ) is a functional rating scale relevant to early AD and is sufficiently brief to be used to assess functional abilities in patients treated with aducanumab (59). The FAQ has good discriminant validity in distinguishing MCI from dementia and performed similarly to the ADCS-ADL-MCI scale in comparative studies (60). These tools are sufficiently brief to be used in practice settings and could be considered for use in evaluating patients receiving aducanumab. Clinicians familiar with CDR administration may consider annual administration of this instrument to assess patient cognitive and functional abilities. The Expert Panel recommends that objective, validated tests to be used longitudinally to assess patients treated with aducanumab.

Stopping therapy

The appropriate timing and strategies for stopping aducanumab therapy have not been studied. Stopping treatment might be informed by patient preferences, care partner decisions, or clinician recommendations based on a perceived lack of effect, ARIA-related concerns, or inability of the patient to adhere to the treatment regimen. Aducanumab should be stopped in all patients manifesting severe symptoms (e.g., seizures, stroke-like manifestations) in the presence of ARIA. Stopping treatment in other ARIA-related circumstances depends on whether ARIA-E resolves after suspending therapy, whether ARIA-H stabilizes when treatment is withheld, the patient’s clinical status, and clinician-patient alignment on the benefit/harm ratio of resuming treatment.
Aducanumab has not been tested in patients with moderate or severe AD and progression into the more advanced phases of AD will prompt reassessment of treatment continuation. Progression into moderate dementia is signaled by progression to CDR global score of 2.0, decline of MMSE scores below 20, and loss of autonomy on key ADLs. The Expert Panel recommends that clinicians carefully review the evidence of benefit and the potential risk in patients who progress to moderate dementia after appropriate use of aducanumab in early AD.

Primary Care Clinicians collaboration

The availability of aducanumab may create a demand for detection, diagnosis and treatment of early AD that can overwhelm an unprepared healthcare system (61). Providing treatment with aducanumab requires high proficiency and sufficient resources including close collaborations with comprehensive multi-disciplinary teams. With too few specialists currently available to respond to the possible number of patients who are candidates for treatment, there are opportunities to forge new models of hub-and-spoke dementia specialist-primary care collaborations and peer-to-peer counseling to partially fill these needs and respond to workforce gaps. The Expert Panel recommends including community organizations, Alzheimer Association chapters, primary care clinicians, memory-care enabled nurses and nurse practitioners, and other creative collaborations and solutions to meet the needs of patients seeking care and encountering the difficulty of being assessed because of shortages of memory care specialists in the current health care system (62-64).

Appropriate Patient Discussions

Aducanumab is an unprecedented therapy; it is the first drug approved for treatment of AD based on plaque lowering and addressing the underlying pathophysiology of AD. Clinicians, patients, care partners, and stakeholders of the healthcare system must learn and adjust to the new therapeutic circumstances. Discussions with patients and care partners are particularly important. They require information regarding the possible benefits of aducanumab, the side effects including ARIA, and the likely need for long term adherence to treatment. Dementia medication discontinuation rates have been shown to be higher in African American and Hispanic patients than White patients; racially and ethnically appropriate strategies may be required to optimize adherence (65). Referral to the Alzheimer’s Association ( and other trusted community sources can assist the clinician in providing reliable information.

Aducanumab Treatment in Non-AD Amyloid-Bearing Conditions and Atypical AD

Autosomal dominant AD is produced by mutations of presenilin 1, presenilin 2, or the amyloid precursor protein gene. Patients typically develop amyloid plaques as evidenced by amyloid PET in their mid to late 30’s and progress to MCI due to AD and mild AD dementia at age 45 to 55 (66). The individuals have the canonical features of AD at autopsy (67). Few if any of these patients were included in the aducanumab clinical trials. The Expert Panel Recommends that if patients with autosomal dominant AD meet all other criteria for aducanumab treatment described in Table 1, they could be considered candidates for aducanumab and the option can be discussed with families. They should be informed of the scarcity of data in patients with the inherited form of AD.
Individuals with Down syndrome essentially uniformly develop brain amyloid plaques and often have symptoms of dementia in midlife (68, 69). The presence of amyloid plaques in Down syndrome suggests that treatment with aducanumab may be beneficial. There are many differences between Down syndrome and late onset AD, and the Expert Panel recommends against treating Down syndrome patients with aducanumab until more data are available.
Patients with AD may present with atypical syndromes such as logopenic aphasia, posterior cortical atrophy, or frontal AD (70). These patients have metabolic scans that reflect the regional dysfunction corresponding to their clinical presentation; other biomarkers are characteristic of AD (71). Few patients with atypical features were included in the aducanumab trials. The Expert Panel recommends that if patients with atypical AD meet all the criteria for the appropriate use of aducanumab, they can be considered as candidates for aducanumab treatment while cautioning patients and families that little information regarding use of aducanumab is available on patients with these clinical profiles.
Patients with dementia with Lewy bodies (DLB) have MCI that progresses to dementia. They have characteristic clinical features including parkinsonism, visual hallucinations, fluctuating cognition, and rapid eye movement sleep behavior disorder (72). Patients with DLB may have pure Lewy body pathology or may have concomitant Lewy body changes and Aβ plaques. Those with Aβ plaques will have positive amyloid PET imaging (73). The Expert Panel Recommends that patients with DLB not be treated with aducanumab; the effect of treatment in patients with mixed amyloid and Lewy body pathology is unknown.
The ability to image cognitively normal individuals or conduct lumbar puncture and CSF analyses allows the detection of persons in the preclinical phases of the AD continuum. These individuals have amyloid plaques in the brain but are cognitively normal. All participants in the aducanumab clinical trials were symptomatic and met criteria for MCI due to AD or mild AD dementia. There are no data on the utility of treating individuals in the preclinical disease state with aducanumab. The Expert Panel recommends against treating patients in the preclinical phase of AD with aducanumab until additional data are available.
Care partners seek means of improving quality of life for their loved one regardless of the degree of the patient’s dementia-related disability. Patients with moderate to severe AD and their caregivers will seek information about aducanumab and may wish to be treated. There are no data available on the use of aducanumab in moderate and severe AD. The Expert Panel recommends against beginning aducanumab therapy in patients with moderate to severe AD (e.g, those with cognitive deficits beyond mild severity and requiring substantial assistance with activities of daily living). These patients require comprehensive compassionate care, and their support must continue regardless of DMT therapy status. Multidisciplinary interventions at this stage can significantly improve quality of life (64, 74).
The amyloid, tau, neurodegeneration (AT(N)) framework is influential in the biomarker classification of AD (75). Using this approach, A+T-N-, A+T+N-, and A+T+N+ patients would be considered candidates for treatment with aducanumab if they have early AD and meet all treatment criteria (Table 1). A+T-N+ patients may have some disorder such as vascular pathology in addition to amyloidosis that may impact aducanumab therapy. Further evaluation of these patients is required before proceeding with therapy.
Patients with cerebral amyloid angiopathy may have positive amyloid PET (76). Use of aducanumab in these patients may promote ARIA (77). The Expert Panel recommends that aducanumab not be used in patients with cerebral amyloid angiopathy.

Potential Future Changes in Appropriate Use of Aducanumab

AD science is evolving rapidly in both diagnostic and therapeutic technologies. Blood tests that assist in the diagnosis of AD could have a transformative influence on the care of AD patients and the appropriate use of aducanumab. Blood assays that determine the Aß42/40 ratio have good correspondence with amyloid PET status (receiver operator curse area under the curve [AUC] 0.88) and this improves when combined with patient age and APOE-4 genotype (AUC 0.94) (78). Plasma hyperphosphorylated tau (p-tau) 181 and p-tau 217 are abnormal in early AD and correlates significantly with amyloid burden on PET (79-81). One of these plasma-based markers or a panel of markers possibly including APOE genotype could eventually provide a diagnosis of brain amyloidosis in patients with symptoms of early AD or could function as a case-finding tool to identify patients likely to have an abnormal amyloid PET.
Blood tests may not be the only means of identifying amyloidosis in patients with the clinical syndrome of early AD. Amyloid is deposited in the retina in AD, and retinal imaging might be another means of detecting central nervous system amyloidosis (82, 83). Digital biomarkers could play a role in case finding or diagnostic confirmation. Voice and language analyses, for example, are promising means of identifying early AD (84, 85).
Currently, aducanumab treatment is administered with the plan of continuing at least until the patient reaches the moderate stage of AD dementia. However, once significant amyloid lowering has been achieved it may be possible to reduce the frequency of infusions. The durability of amyloid lowering was explored in a trial with another amyloid-targeting monoclonal antibody (86) with encouraging results.
Prevention of AD is an important goal of AD research. Trials of aducanumab during the preclinical phases of AD when the brain has high levels of amyloid but cognition remains largely normal may expand the range of individuals appropriate for treatment (87)
Patients with Down syndrome that meet all the other criteria for treatment with aducanumab may become treatment-eligible when additional studies have been conducted and additional data are available (88).



Aducanumab is a new treatment for AD. It provides opportunities and challenges for its introduction into the management of AD patients. Aducanumab requires substantial infrastructure for appropriate administration: expert clinicians skilled in recognition of early AD; amyloid PET or lumbar puncture capability; experts in amyloid PET interpretation or CSF analysis: infusion center availability; and access to MRI and experts in recognition and management of ARIA (Table 5). Genetic counseling may be required in some circumstances, and all patients and care partners require education and support. Building this infrastructure for the appropriate use of aducanumab will require time, resources, and creative planning. Appropriate use of aducanumab requires a commitment to patient-centered care and best-practices for the safe delivery of this new treatment.

Table 5. Resources needed for the appropriate use of aducanumab (Expert Panel Recommendations)


Disclosure and Conflicts of Interest: JC has provided consultation to Acadia, Alkahest, AriBio, Avanir, Axsome, Behren Therapeutics, Biogen, Cassava, Cerecin, Cerevel, Cortexyme, EIP Pharma, Eisai, GemVax, Genentech, Green Valley, Grifols, Janssen, Jazz, Karuna, LSP, Merck, Novo Nordisk, Otsuka, ReMYND, Resverlogix, Roche, Signant Health, Sunovion, Suven, United Neuroscience, and Unlearn AI pharmaceutical and assessment companies. Dr. Cummings owns the copyright of the Neuropsychiatric Inventory. Dr Cummings has the following research support: NIGMS P20GM109025; NINDS U01NS093334; NIA R01AG053798; NIA P20AG068053; NIA R35AG71476. LGA has provided consultation to Eli Lilly, Biogen, and Two Labs. LGA receives the following research support: NIA U01 AG057195, NIA R01 AG057739, NIA P30 AG010133, Alzheimer Association LEADS GENETICS 19-639372, Roche Diagnostics RD005665, AVID Pharmaceuticals, Life Molecular Imaging. LGA has received honoraria for participating in independent data safety monitoring boards and providing educational CME lectures and programs. LGA has stock in Cassava Sciences. AA has received honoraria for consulting; participating in independent data safety monitoring boards; providing educational lectures, programs, and materials; or serving on advisory boards for AbbVie, Acadia, Allergan, the Alzheimer’s Association, Axovant, AZ Therapies, Biogen, Grifols, Harvard Medical School Graduate Continuing Education, JOMDD, Lundbeck, Merck, Roche/Genentech, Novo Nordisk, Sunovion, and Suven. PA has received research funding from NIA, FNIH, the Alzheimer’s Association, Janssen, Lilly and Eisai, and personal fees from Biogen, Merck, Roche, Abbvie, ImmunoBrain Checkpoint, Rainbow Medical and Shionogi. SS was a site PI and co-chair of the investigator steering committee for the ENGAGE trial and he receives research support and consultancy fees from Lilly, Biogen, Avid, Eisai, Genentech, and Roche. MW has served on Advisory Boards for Eli Lilly, Cerecin/Accera, Roche, Alzheon, Inc., Merck Sharp & Dohme Corp., Nestle/Nestec, PCORI/PPRN, Dolby Family Ventures, National Institute on Aging (NIA), Brain Health Registry and ADNI. He serves on Editorial Boards for Alzheimer’s & Dementia, TMRI and MRI. He has provided consulting and/or acted as a speaker to Cerecin/Accera, Inc., BioClinica, Nestle/Nestec, Roche, Genentech, NIH, The Buck Institute for Research on Aging, FUJIFILM-Toyama Chemical (Japan), Garfield Weston, Baird Equity Capital, University of Southern California (USC), Cytox, and Japanese Organization for Medical Device Development, Inc. (JOMDD) and T3D Therapeutics. He holds stock options with Alzheon, Inc., Alzeca, and Anven.

Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits use, duplication, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.



1. U.S. Food & Drug Administration. Drugs@FDA: FDA-Approved Drugs. Aducanumab. Reference ID 4822820 2021; Available from:
2. U.S. Food & Drug Administration. Drugs@FDA: FDA-Approved Drugs – Aducanumab. Reference ID 4807032. 2021
3. Arndt JW, Qian F, Smith BA, et al. Structural and kinetic basis for the selectivity of aducanumab for aggregated forms of amyloid-beta. Sci Rep. 2018;8:6412; DOI: 10.1038/s41598-018-24501-0.
4. Haeberlein SB, Von Hehn C, Tian Y, et al., editors. EMERGE and ENGAGE topline results: two phase 3 studies to evaluate aducanumab in patients with early Alzheimer’s disease. Clinical Trials on Alzheimer’s Disease; 2019 December 5, 2019; San Diego, CA.
5. Sandrock A. Aducanumab for the treatment of Alzheimer’s disease. U.S. Food & Drug Administration. Peripheral and Central Nervous System Drugs Advisory Committee. November 6, 2020.
6. Atri A. The Alzheimer’s disease clinical spectrum: diagnosis and management. Med Clin North Am. 2019;103:263-293; DOI: 10.1016/j.mcna.2018.10.009.
7. McKhann GM, Knopman DS, Chertkow H, et al. The diagnosis of dementia due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7:263-269; DOI: 10.1016/j.jalz.2011.03.005.
8. Albert MS, DeKosky ST, Dickson D, et al. The diagnosis of mild cognitive impairment due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7:270-279; DOI: 10.1016/j.jalz.2011.03.008.
9. Food and Drug Administration. Early Alzheimer’s Disease: Developing Drugs for Treatment Guidance for Industry. U. S. Department of Health and Human Services Food and Drug Administration Center for Drug Evaluation and Research (CDER) Center for Biologics Evaluation and Research (CBER). February, 2018.
10. Morris JC. The Clinical Dementia Rating (CDR): current version and scoring rules. Neurology. 1993;43:2412-2414; DOI: 10.1212/wnl.43.11.2412-a.
11. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189-198; DOI: 10.1016/0022-3956(75)90026-6.
12. Feeney J, Savva GM, O’Regan C, King-Kallimanis B, Cronin H, Kenny RA. Measurement error, reliability, and minimum detectable change in the Mini-Mental State Examination, Montreal Cognitive Assessment, and Color Trails Test among community living middle-aged and older adults. J Alzheimers Dis. 2016;53:1107-1114; DOI: 10.3233/JAD-160248.
13. Hensel A, Angermeyer MC, Riedel-Heller SG. Measuring cognitive change in older adults: reliable change indices for the Mini-Mental State Examination. J Neurol Neurosurg Psychiatry. 2007;78:1298-1303; DOI: 10.1136/jnnp.2006.109074.
14. Tombaugh TN. Test-retest reliable coefficients and 5-year change scores for the MMSE and 3MS. Arch Clin Neuropsychol. 2005;20:485-503; DOI: 10.1016/j.acn.2004.11.004.
15. Sevigny J, Chiao P, Bussiere T, et al. The antibody aducanumab reduces Abeta plaques in Alzheimer’s disease. Nature. 2016;537:50-56; DOI: 10.1038/nature19323.
16. Nasreddine ZS, Phillips NA, Bedirian V, et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc. 2005;53:695-699; DOI: 10.1111/j.1532-5415.2005.53221.x.
17. Trzepacz PT, Hochstetler H, Wang S, Walker B, Saykin AJ, Alzheimer’s Disease Neuroimaging I. Relationship between the Montreal Cognitive Assessment and Mini-mental State Examination for assessment of mild cognitive impairment in older adults. BMC Geriatr. 2015;15:107; DOI:10.1186/s12877-015-0103-3.
18. Jak AJ, Preis SR, Beiser AS, et al. Neuropsychological criteria for mild cognitive impairment and dementia risk in the Framingham Heart Study. J Int Neuropsychol Soc. 2016;22:937-943; DOI:10.1017/S1355617716000199.
19. Deardorff WJ, Feen E, Grossberg GT. The use of cholinesterase inhibitors across all stages of Alzheimer’s disease. Drugs Aging. 2015;32:537-547; DOI: 10.1007/s40266-015-0273-x.
20. Sevigny J, Suhy J, Chiao P, et al. Amyloid PET screening for enrichment of early-stage Alzheimer disease clinical trials: experience in a phase 1b clinical trial. Alzheimer Dis Assoc Disord. 2016;30:1-7; DOI: 10.1097/WAD.0000000000000144.
21. Johnson KA, Minoshima S, Bohnen NI, et al. Appropriate use criteria for amyloid PET: a report of the Amyloid Imaging Task Force, the Society of Nuclear Medicine and Molecular Imaging, and the Alzheimer’s Association. Alzheimers Dement. 2013;9:e-1-16; DOI: 10.1016/j.jalz.2013.01.002.
22. Clark CM, Schneider JA, Bedell BJ, et al. Use of florbetapir-PET for imaging beta-amyloid pathology. JAMA. 2011;305:275-283; DOI: 10.1001/jama.2010.2008.
23. Seibyl J, Catafau AM, Barthel H, et al. Impact of training method on the robustness of the visual assessment of 18F-florbetaben PET scans: results from a phase-3 study. J Nucl Med. 2016;57:900-906; DOI: 10.2967/jnumed.115.161927.
24. Buckley CJ, Sherwin PF, Smith AP, Wolber J, Weick SM, Brooks DJ. Validation of an electronic image reader training programme for interpretation of [18F]flutemetamol beta-amyloid PET brain images. Nucl Med Commun. 2017;38:234-241; DOI: 10.1097/MNM.0000000000000633.
25. Shaw LM, Arias J, Blennow K, et al. Appropriate use criteria for lumbar puncture and cerebrospinal fluid testing in the diagnosis of Alzheimer’s disease. Alzheimers Dement. 2018;14:1505-1521; DOI: 10.1016/j.jalz.2018.07.220.
26. Hansson O, Lehmann S, Otto M, Zetterberg H, Lewczuk P. Advantages and disadvantages of the use of the CSF amyloid beta (Abeta) 42/40 ratio in the diagnosis of Alzheimer’s disease. Alzheimers Res Ther. 2019;11:34; DOI: 10.1186/s13195-019-0485-0.
27. Skillback T, Farahmand BY, Rosen C, et al. Cerebrospinal fluid tau and amyloid-beta1-42 in patients with dementia. Brain. 2015;138:2716-273; DOI: 110.1093/brain/awv181.
28. van Harten AC, Wiste HJ, Weigand SD, et al. Detection of Alzheimer’s disease using ELECSYS amyloid-beta 1-42, p-tau and t-tau assays. Alzheimer Dement. 2021;In Press
29. Palmqvist S, Zetterberg H, Mattsson N, et al. Detailed comparison of amyloid PET and CSF biomarkers for identifying early Alzheimer disease. Neurology. 2015;85:1240-1249; DOI: 10.1212/WNL.0000000000001991.
30. Fagan AM, Mintun MA, Mach RH, et al. Inverse relation between in vivo amyloid imaging load and cerebrospinal fluid Abeta42 in humans. Ann Neurol. 2006;59:512-519; DOI: 10.1002/ana.20730.
31. Palmqvist S, Mattsson N, Hansson O, Alzheimer’s Disease Neuroimaging I. Cerebrospinal fluid analysis detects cerebral amyloid-beta accumulation earlier than positron emission tomography. Brain. 2016;139:1226-1236; DOI: 10.1093/brain/aww015.
32. Hampel H, Shaw LM, Aisen P, et al. State-of-the-art of lumbar puncture and its place in the journey of patients with Alzheimer’s disease. Alzheimers Dement. 2021;Online ahead of print:DOI: 10.1002/alz.12372.
33. Jack CR, Jr., Knopman DS, Chetelat G, et al. Suspected non-Alzheimer disease pathophysiology–concept and controversy. Nat Rev Neurol. 2016;12:117-124; DOI: 10.1038/nrneurol.2015.251.
34. Rabinovici GD, Gatsonis C, Apgar C, et al. Association of amyloid positron emission tomography with subsequent change in clinical management among medicare beneficiaries with mild cognitive impairment or dementia. JAMA. 2019;321:1286-1294; DOI: 10.1001/jama.2019.2000.
35. Sperling RA, Jack CR, Jr., Black SE, et al. Amyloid-related imaging abnormalities in amyloid-modifying therapeutic trials: recommendations from the Alzheimer’s Association Research Roundtable Workgroup. Alzheimers Dement. 2011;7:367-385; DOI: 10.1016/j.jalz.2011.05.2351.
36. VandeVrede L, Gibbs DM, Koestler M, et al. Symptomatic amyloid-related imaging abnormalities in an APOE epsilon4/epsilon4 patient treated with aducanumab. Alzheimers Dement (Amst). 2020;12:e12101; DOI: 10.1002/dad2.12101.
37. Green RC, Roberts JS, Cupples LA, et al. Disclosure of APOE genotype for risk of Alzheimer’s disease. N Engl J Med. 2009;361:245-254; DOI: 10.1056/NEJMoa0809578.
38. Langlois CM, Bradbury A, Wood EM, et al. Alzheimer’s Prevention Initiative Generation Program: Development of an APOE genetic counseling and disclosure process in the context of clinical trials. Alzheimers Dement (N Y). 2019;5:705-716; DOI: 10.1016/j.trci.2019.09.013.
39. Flather M, Delahunty N, Collinson J. Generalizing results of randomized trials to clinical practice: reliability and cautions. Clin Trials. 2006;3:508-512; DOI: 10.1177/1740774506073464.
40. Bodilsen J, Mariager T, Vestergaard HH, et al. Association of lumbar puncture with spinal hematoma in patients with and without coagulopathy. JAMA. 2020;324:1419-1428; DOI: 10.1001/jama.2020.14895.
41. Mayo Clinic. Clinical Trials. 221AD301 Phase 3 study of aducanumab (BIIB037) in early Alzheimer’s disease [Available from:
42. Goos JD, van der Flier WM, Knol DL, et al. Clinical relevance of improved microbleed detection by susceptibility-weighted magnetic resonance imaging. Stroke. 2011;42:1894-1900; DOI: 10.1161/STROKEAHA.110.599837.
43. Cummings J, Fox N. Defining disease modifying therapy for Alzheimer’s disease. J Prev Alzheimers Dis. 2017;4:109-115; DOI: 10.14283/jpad.2017.12.
44. Cartz-Piver L, Polin C, Diebolt C, et al. Therapeutic patient education in Alzheimer’s disease and related disorders: Clinical experience and perspectives. Rev Neurol (Paris). 2020;176:200-207; DOI: 10.1016/j.neurol.2019.07.028.
45. Cuevas AG, O’Brien K, Saha S. African American experiences in healthcare: “I always feel like I’m getting skipped over”. Health Psychol. 2016;35:987-995; DOI: 10.1037/hea0000368.
46. Mehta KM, Yin M, Resendez C, Yaffe K. Ethnic differences in acetylcholinesterase inhibitor use for Alzheimer disease. Neurology. 2005;65:159-162; DOI: 10.1212/01.wnl.0000167545.38161.48.
47. Gleason CE, Dowling NM, Benton SF, Kaseroff A, Gunn W, Edwards DF. Common sense model factors affecting African Americans’ willingness to consult a healthcare provider regarding symptoms of mild cognitive impairment. Am J Geriatr Psychiatry. 2016;24:537-546; DOI: 10.1016/j.jagp.2015.08.005.
48. Cuevas AG, O’Brien K, Saha S. What is the key to culturally competent care: Reducing bias or cultural tailoring? Psychol Health. 2017;32:493-507; DOI: 10.1080/08870446.2017.1284221.
49. Bechten A, Wattjes MP, Purcell DD, et al. Validation of an MRI rating scale for amyloid-related imaging abnormalities. J Neuroimaging. 2017;27:318-325; DOI: 10.1111/jon.12422.
50. Rosen WG, Mohs RC, Davis KL. A new rating scale for Alzheimer’s disease. Am J Psychiatry. 1984;141:1356-1364; DOI: 10.1176/ajp.141.11.1356.
51. Galasko D, Bennett D, Sano M, et al. An inventory to assess activities of daily living for clinical trials in Alzheimer’s disease. The Alzheimer’s Disease Cooperative Study. Alzheimer Dis Assoc Disord. 1997;11 Suppl 2:S33-39
52. Cummings JL, Mega M, Gray K, Rosenberg-Thompson S, Carusi DA, Gornbein J. The Neuropsychiatric Inventory: comprehensive assessment of psychopathology in dementia. Neurology. 1994;44:2308-2314; DOI: 10.1212/wnl.44.12.2308.
53. Tractenberg RE, Yumoto F, Jin S, Morris JC. Sample size requirements for training to a kappa agreement criterion on clinical dementia ratings. Alzheimer Dis Assoc Disord. 2010;24:264-268; DOI: 10.1097/WAD.0b013e3181d489c6.
54. Sun N, Mormino EC, Chen J, Sabuncu MR, Yeo BTT, Alzheimer’s Disease Neuroimaging I. Multi-modal latent factor exploration of atrophy, cognitive and tau heterogeneity in Alzheimer’s disease. Neuroimage. 2019;201:116043; DOI: 10.1016/j.neuroimage.2019.116043.
55. Giorgio J, Landau SM, Jagust WJ, Tino P, Kourtzi Z, Alzheimer’s Disease Neuroimaging I. Modelling prognostic trajectories of cognitive decline due to Alzheimer’s disease. Neuroimage Clin. 2020;26:102199; DOI: 10.1016/j.nicl.2020.102199.
56. Galvin JE, Roe CM, Xiong C, Morris JC. Validity and reliability of the AD8 informant interview in dementia. Neurology. 2006;67:1942-1948; DOI: 10.1212/01.wnl.0000247042.15547.eb.
57. Galvin JE, Roe CM, Powlishta KK, et al. The AD8: a brief informant interview to detect dementia. Neurology. 2005;65:559-564; DOI: 10.1212/01.wnl.0000172958.95282.2a.
58. Kaufer DI, Cummings JL, Ketchel P, et al. Validation of the NPI-Q, a brief clinical form of the Neuropsychiatric Inventory. J Neuropsychiatry Clin Neurosci. 2000;12:233-239; DOI: 10.1176/jnp.12.2.233.
59. Pfeffer RI, Kurosaki TT, Harrah CH, Jr., Chance JM, Filos S. Measurement of functional activities in older adults in the community. J Gerontol. 1982;37:323-329; DOI: 10.1093/geronj/37.3.323.
60. Kaur N, Belchior P, Gelinas I, Bier N. Critical appraisal of questionnaires to assess functional impairment in individuals with mild cognitive impairment. Int Psychogeriatr. 2016;28:1425-1439; DOI: 10.1017/S104161021600017X.
61. Liu J, Hlayka J, Hillestad R, Mattke S. Assessing the preparedness of the US health care infrastructure for an Alzeimer’s treatment. RAND Corporation, Los Angeles, 2017 (
62. Reuben DB, Tan ZS, Romero T, Wenger NS, Keeler E, Jennings LA. Patient and caregiver benefit from a comprehensive dementia care program: 1-year results from the UCLA Alzheimer’s and Dementia Care Program. J Am Geriatr Soc. 2019;67:2267-2273; DOI: 10.1111/jgs.16085.
63. Reuben DB, Gill TM, Stevens A, et al. D-CARE: The Dementia Care Study: design of a pragmatic trial of the effectiveness and cost effectiveness of health system-based versus community-based dementia care versus usual dementia care. J Am Geriatr Soc. 2020;68:2492-2499; DOI: 10.1111/jgs.16862.
64. Callahan CM, Boustani MA, Weiner M, et al. Implementing dementia care models in primary care settings: The Aging Brain Care Medical Home. Aging Ment Health. 2011;15:5-12; DOI: 10.1080/13607861003801052.
65. Thorpe CT, Fowler NR, Harrigan K, et al. Racial and ethnic differences in initiation and discontinuation of antidementia drugs by medicare beneficiaries. J Am Geriatr Soc. 2016;64:1806-1814; DOI: 10.1111/jgs.14403.
66. Bateman RJ, Xiong C, Benzinger TL, et al. Clinical and biomarker changes in dominantly inherited Alzheimer’s disease. N Engl J Med. 2012;367:795-804; DOI: 10.1056/NEJMoa1202753.
67. Shepherd C, McCann H, Halliday GM. Variations in the neuropathology of familial Alzheimer’s disease. Acta Neuropathol. 2009;118:37-52; DOI: 10.1007/s00401-009-0521-4.
68. Lott IT, Head E. Dementia in Down syndrome: unique insights for Alzheimer disease research. Nat Rev Neurol. 2019;15:135-147; DOI: 10.1038/s41582-018-0132-6.
69. Ballard C, Mobley W, Hardy J, Williams G, Corbett A. Dementia in Down’s syndrome. Lancet Neurol. 2016;15:622-636; DOI: 10.1016/S1474-4422(16)00063-6.
70. Paterson RW, Toombs J, Slattery CF, et al. Dissecting IWG-2 typical and atypical Alzheimer’s disease: insights from cerebrospinal fluid analysis. J Neurol. 2015;262:2722-2730; DOI: 10.1007/s00415-015-7904-3.
71. Graff-Radford J, Yong KXX, Apostolova LG, et al. New insights into atypical Alzheimer’s disease in the era of biomarkers. Lancet Neurol. 2021;20:222-234; DOI: 10.1016/S1474-4422(20)30440-3.
72. McKeith IG, Boeve BF, Dickson DW, et al. Diagnosis and management of dementia with Lewy bodies: Fourth consensus report of the DLB Consortium. Neurology. 2017;89:88-100; DOI: 10.1212/WNL.0000000000004058.
73. Orad RI, Shiner T. Differentiating dementia with Lewy bodies from Alzheimer’s disease and Parkinson’s disease dementia: an update on imaging modalities. J Neurol. 2021; DOI: 10.1007/s00415-021-10402-2.
74. Callahan CM, Boustani MA, Unverzagt FW, et al. Effectiveness of collaborative care for older adults with Alzheimer disease in primary care: a randomized controlled trial. JAMA. 2006;295:2148-2157; DOI: 10.1001/jama.295.18.2148.
75. Jack CR, Jr., Bennett DA, Blennow K, et al. NIA-AA research framework: toward a biological definition of Alzheimer’s disease. Alzheimers Dement. 2018;14:535-562; DOI: 10.1016/j.jalz.2018.02.018.
76. Raposo N, Planton M, Peran P, et al. Florbetapir imaging in cerebral amyloid angiopathy-related hemorrhages. Neurology. 2017;89:697-704; DOI: 10.1212/WNL.0000000000004228.
77. Greenberg SM, Cordonnier C, Schneider JA, et al. Off-label use of aducanumab for derebral amyloid angiopathy. Lancet Neurol. 2021;Epub ahead of print;
78. Schindler SE, Bollinger JG, Ovod V, et al. High-precision plasma beta-amyloid 42/40 predicts current and future brain amyloidosis. Neurology. 2019;93:e1647-e1659; DOI: 10.1212/WNL.0000000000008081.
79. Moscoso A, Grothe MJ, Ashton NJ, et al. Time course of phosphorylated-tau181 in blood across the Alzheimer’s disease spectrum. Brain. 2021;144:325-339; DOI: 10.1093/brain/awaa399.
80. Karikari TK, Benedet AL, Ashton NJ, et al. Diagnostic performance and prediction of clinical progression of plasma phospho-tau181 in the Alzheimer’s Disease Neuroimaging Initiative. Mol Psychiatry. 2021;26:429-442; DOI: 10.1038/s41380-020-00923-z.
81. Mattsson-Carlgren N, Janelidze S, Palmqvist S, et al. Longitudinal plasma p-tau217 is increased in early stages of Alzheimer’s disease. Brain. 2020;143:3234-3241; DOI: 10.1093/brain/awaa286.
82. Koronyo Y, Biggs D, Barron E, et al. Retinal amyloid pathology and proof-of-concept imaging trial in Alzheimer’s disease. JCI Insight. 2017;2; DOI: 10.1172/jci.insight.93621.
83. Koronyo-Hamaoui M, Koronyo Y, Ljubimov AV, et al. Identification of amyloid plaques in retinas from Alzheimer’s patients and noninvasive in vivo optical imaging of retinal plaques in a mouse model. Neuroimage. 2011;54 Suppl 1:S204-217; DOI: 10.1016/j.neuroimage.2010.06.020.
84. Thomas JA, Burkhardt HA, Chaudhry S, et al. Assessing the utility of language and voice biomarkers to predict cognitive impairment in the Framingham Heart Study Cognitive Aging Cohort Data. J Alzheimers Dis. 2020;76:905-922; DOI: 10.3233/JAD-190783.
85. Konig A, Satt A, Sorin A, et al. Automatic speech analysis for the assessment of patients with predementia and Alzheimer’s disease. Alzheimers Dement (Amst). 2015;1:112-124; DOI: 10.1016/j.dadm.2014.11.012.
86. Mintun MA, Lo AC, Duggan Evans C, et al. Donanemab in early Alzheimer’s disease. N Engl J Med. 2021;384:1691-1704; DOI: 10.1056/NEJMoa2100708.
87. Sperling R, Mormino E, Johnson K. The evolution of preclinical Alzheimer’s disease: implications for prevention trials. Neuron. 2014;84:608-622; DOI: 10.1016/j.neuron.2014.10.038.
88. Hartley D, Blumenthal T, Carrillo M, et al. Down syndrome and Alzheimer’s disease: Common pathways, common goals. Alzheimers Dement. 2015;11:700-709; DOI: 10.1016/j.jalz.2014.10.007.



Z. Jiayuan1, J. Xiang-Zi2,*, M. Li-Na1, Y. Jin-Wei1, Y. Xue3


1. Psychology Nursing, Harbin Medical University, Daqing, Heilongjiang) China; 2. Business management department, Suzhou Industrial Park Institute of Service Outsourcing, Suzhou, Jiangsu, China; 3. Neurology/Daqing Longnan Hospital, Daqing, Heilongjiang, China. * co-first author

Corresponding Author: Meng Li-Na, No.39 Xinyang Street, Harbin Medical University, Daqing, Heilongjiang Province, China, Tel: 86-18604586122, Email:

J Prev Alz Dis 2021;
Published online July 5, 2021,



Background: The Objective: To assess the effectiveness of a mindfulness-based Tai Chi Chuan on physical performance and cognitive function among cognitive frailty older adults.
Design: A single-blind,three-arm randomized controlled trial.
Setting: Three communities in Daqing, China.
Participants: The study sample comprised 93 men and women aged 65 years or older who were able to walk more than 10 m without helping tools, scored 0.5 on Clinical Dementia Rating (CDR) and absence of concurrent dementia, identified pre-frailty (scored 1-2 on Fried Frailty Criteria) and frailty older adults (scored 3-5 on Fried Frailty Criteria).
Intervention: Subjects were randomly allocated to three groups: Group1, which received mindfulness intervention (formal and informal mindfulness practices); Group 2, which received Tai-Chi Chuan intervention; Group 3, which received MTCC intervention.
Measurements: The primary outcomes was cognitive frailty rate(measured by Fried Frailty Criteria and Clinical Dementia Rating-CDR) , the secondary outcome were cognitive function (measured by Min-Mental State Examination-MMES) and physical level (measured by Short physical performance battery- SPPB, Timed up and Go test-TUG and the 30-second Chair test). They were all assessed at Time 1-baseline, Time 2-after the end of 6-month intervention and the follow up (Time 3-half year after the end of 6-month intervention).
Results: The baseline characteristics did not differ among the groups.Improvements in the cognitive function (MMES), physical performance (SPPB, TUG, 30-second Chair test) were significantly difference between time-group interaction (p<.05). The rate of CF was significantly different among groups at 6-month follow-up period (χ2=6.37, p<.05). A lower prevalence of frailty and better cognitive function and physical performance were found in the Group 3 compared with other two groups at the follow-up period (p<.05).
Conclusions: MTCC seems to be effectively reverse CF, improving the cognitive and physical function among older adults, suggesting that MTCC is a preferably intervention option in community older adults with cognitive frailty.

Key words: Cognitive frailty, mindfulness, Tai Chi, physical, cognitive.




Physical and cognitive impairment frequently overlap in older adults. Recent studies have showed that an interrelationship between physical frailty and cognitive impairment existed (1). Therefore a new conceptual construct—cognitive frailty (CF)—characterized by the simultaneous existence of both cognitive impairment and frailty, was proposed in 2013 by an international consensus group (2). A recent meta-analysis which focused on CF older adults in community showed that the CF was found to be a significant predictor of the short-term and the long-term mortality and dementia, disability, and other adverse health outcomes (3). Older adults with CF are more likely to develop dementia than the two individual components, and could provide a new direction for healthy aging (4). Early diagnosis, detection and intervention of CF are of great significance for delaying the occurrence and development of dementia (5). A series of evidence suggested that physical exercise was associated with improvements in health related outcomes in older adults with frailty (6-9). However, currently no optimal interventions can be recommended for cognitive function and physical health promotion in older adults with CF as the evidence base is small and of limited quality.
In the last decade, Tai Chi Chuan (TCC) has been widely adopted in physical exercise aiming to improve physical performance in community older adults in China, which being proposed as a high efficiency exercise by the Centers for Disease Control and Prevention (CDC) (10). The pulling movement in TCC is beneficial to improve the flexibility and coordination of the body and reduce the probability of injury. Several studies proved that TCC could effectively promote physical fitness and muscle protein anabolism in frail older adults and was more effective than conventional exercise approaches for reducing the incidence of falls (11-14). The training spirit of TCC is to complete harmony of body and mind which is similar to the mindfulness skills. Mindfulness is broadly defined as present-focused, non-judgmental awareness (15). Evidence shows that mindfulness-based interventions has yielded positive effects on enhancing the ability of cognitive reserve and slowing down the aging-associated cognitive decline, promoting active aging among mild cognitive impairment (MCI) older adults in the community (9). Regarding to the consistency of core mechanism between mindfulness and TCC, some researchers have mixed systematic mindfulness training with TCC to investigate the health benefits both in physical and psychology (16-17). Mindfulness-based Tai Chi Chuan (MTCC) may be more suitable for CF older adults, however, whether MTCC can reduce cognitive frailty is currently unknown.
With the continuing growth of the CF population and the community health care policy in China, implementing a more feasible and effective intervention would widely promote healthy aging. Consequently, in this study we aim to use a rigorous randomized control trial design to investigate the effects of an MTCC intervention on cognitive function and physical performance in the CF older adults. We hypothesized that compared with mindfulness intervention and TCC exercise, MTCC would be greatly more effective in both cognitive and physical aspects among CF populations.



Study Design

This study was a single-blind, three-arm randomized controlled trial conducted in three communities in Daqing, China. All participants who met the inclusion criteria and agree to participant finished the written informed consent, and then they were allocated to group 1 (Mindfulness group) or group 2 (TCC group) or group 3 (MTCC group) by lottery method. The study was approved by the the committee on ethics in research of Harbin Medical University. Our reporting in the manuscript adheres to the CONSORT 2010 guidelines. The study was registered under Chinese Clinical Trials Registry (ChiCTR2100042851).


Participants were recruited by putting up a poster through collaboration with three communities. The inclusion criteria included: 1- aged 65 years or older; 2- had no serious mental or physical disease; 3- could walk without helping tools; 4- a score of 0.5 in CDR ; 5- pre-frail and frail older adults. These with dementia or undergoing similar or other physical and cognitive intervention were excluded. Finally 93 participants were recruited (Figure 1).

Figure 1. Study flow diagram


Randomization and Masking

Participants were randomly assigned to either a mindfulness intervention or a TCC group or a MTCC group using sequentially numbered, opaque, sealed envelopes with NCR (no carbon required) paper, conducted by a trained research assistant independent of the study design.The implement of interventions and data collection were conducted by independent teams that blinded to to group allocation.


The programs in this study were established and carried out by professional team which consisted of two psychologists (qualified in mindfulness intervention for eleven years), two physical education and sports specialists (qualified in TCC program for more than ten years) and one rehabilitation medicine specialist (major in geriatric rehabilitation). All the participants received a six-month intervention which consisted two stage: the first stage was a 1-hour group intervention (varied in size) twice weekly for three months in community facilities, such as senior or community centers and the second stage was a 1-hour individual practice twice weekly for three months. To avoid the group contamination, the interventions were separated in time and sites. The details of each group intervention were as follows,
Group 1 (mindfulness intervention): All participants received a booklet about mindfulness skills. It consisted of four basic forms of meditation practices (body scan, walking meditation, gentle yoga, sitting meditation). During group intervention, each session began with a 10-minute short-review aiming to solving the existing problems, and then a 45-minute exercises and a 5-minute summary. After participants had a good grasp of basic mindfulness practice, they were taught to integrate mindfulness into daily life such as eating, hearing, smelling, observing.During individual session, participants were required to continue mindfulness training under the supervision and guidance for 3 months.
Group 2 (TCC intervention): All participants received a picture booklet about induction of TCC. 24-Simplified TCC was conducted. During group intervention, each session began with a 10-minute warm-up (including muscle stretching and joint movement) aiming to avoid injure, and then a 45-minute exercises and a 5-minute cool-down activity (deep breathing and relaxing). Participants were taught how to carry out different TTC forms such as “Starting Posture”, “Hold the Lute”, “Cloud Hands” , “Turn and Kick with Left Heel”, etc. During individual session, participants were required to continue TCC training under the supervision and guidance for 3 months.
Group 3 (MTCC intervention): All participants received a booklet about MTCC program.The MTCC involved practice of a core of modified exercise forms mixed together mindfulness and TCC postures aimed at stimulating and integrating body, sensory, and cognitive systems, the practice focused on present, non-judgment, peaceful state involving TCC movements.
During group intervention, participants were taught how to be mindfulness and find connections between mindfulness and TCC practice. Besides, the modified MTCC forms were introduced to participants. After they grasped the basic practice skills, they began to carry out different MTTC forms including the seated formats, standing formats and stepping formats. Each session began with a 10-minute warm-up (including muscle stretching and joint movement) aiming to avoid injure, and then a 45-minute exercises and a 5-minute cool-down activity (deep breathing and relaxing). During individual session, participants were required to continue MTCC training under the supervision and guidance for 3 months.
The theme of each group were shown in Table 1. If all participants have any problems during intervention period, they could contact their tutor at any time.

Table 1. The theme of each group during group intervention


Variables and Outcomes

Demographic data was collected by self-reported questionnaires: age, sex and BMI. Assessment were conducted at baseline (Time 1-one week before commencing the program), after the intervention (Time 2-six month after commencing the program) and follow-up (Time 3-one year after commencing the program).
The primary outcome was the rate of CF at follow up period among different groups. We defined no cognitive frailty as non-frail (scored 0 on Fried Frailty Criteria) and without MCI (CDR=0). 1- Fried Frailty Criteria, which is based on the five Cardiovascular Health Study criteria defined as: slowness/unintentional weight loss/weakness/exhaustion/low physical activity (18); 2- Clinical Dementia Rating (CDR) was used to assess the cognitive function from six aspects :memory/orientation/judgment and problem solving/community affairs/home and hobbies/and personal care.The levels ranged from 0-3 (none to severe). The higher scores, the poorer cognitive performance (19).
The secondary outcomes were change of cognitive function and physical performance of participants. The assessment included: 1- Cognitive function-The Mini-Mental State Examination (MMSE) consists of 11 items was used to assess the cognitive function from five aspects :orientation/memory/attention/language ability /comprehensive/judgment. The score ranged from 0-30 that higher score indicate better cognitive function (20); 2- Physical function-Short physical performance battery (SPPB) was used to assess gait speed, chair stand, and balance tests. Total scores ranged 0-12 that higher points indicate better physical function (21); The Timed Up and Go test (TUG) was used to assess mobility and requires both static and dynamic balance, the time it took form the beginning movement until returned to the seated position was recorded in seconds (22); The 30-second Chair test was administered to assess the core strength which calculated by times (23).

Data analysis

The sample size was calculated by using the G-power 3.1 program with a power (1- β) of 0.95, a significance level of 0.05, From related data (6), we set an effect size of 0.21, finial a sample size was 75 participants, and assuming a 20% estimate loss of follow-up. Finally the sample size was 93. Statistical analyses were performed using SPSS 22.0 (IBM Corp., Armonk, NY, USA). All analyses followed the intent-to-treat principle. Categorical variables were expressed as percentages and continuous variables with mean and SD. Demographic characteristics and basic CF level were compared using a t-test and Chi-square tests. Between-group differences for the effects of intervention were assessed using the Chi-square test for CF rate.Three-way analysis of variance (ANOVA) compared outcome variables between the three groups at three assessment points (three time points × three groups). The post hoc Bonferroni test was used to assess changes within groups. Set p <0.05 as statistically significant.



There were two participants lose to follow up, one in group 1 (because of refuse to continue the study) and one in group 3 (because of be hospitalized) respectively and finally a total of 91 participants completed all the assessment.

Demographic and baseline characteristic

Baseline characteristics were not significantly different among the three groups. The mean age was 71.4±4.6 years (range = 65-81 y), and 56% (n=51) were female. The details of physical performance and cognitive function variables across groups at the baseline were illustrated in (Table 2.Characteristics of Participants). Fried’s Frailty Criteria, MMSE, SPPB, TUG, and 30-second Chair test showed no significantly different and has the comparability among groups (P > .05).

Table 2. Characteristics of Participants (N = 91)

Note. BMI- Body Mass Index;MMSE- Mini Mental State Examination ;SPPB-Short Physical Performance Battery; TUG,Time Up and Go;SD-Standard Deviation


Effects of interventions on CF rate after one year

The Group 1 (mindfulness group) showed there were 2 participants reversed to no CF state and the reversed rate was 6.7%; Group 2 (TCC group) showed there were 4 participants reversed to no CF state and the reversed rate was 12.9%; Group 3 (MTCC group) showed there were 9 participants reversed to no CF state and the reversed rate was 30%.The distribution of CF significantly differed among groups in the follow-up period (χ2=6.37, P=0.041).

Effects of interventions on physical performance and cognitive function

Table 3 revealed the changes in frailty, physical performance and cognitive function from baseline to follow-up (Time 1-Time 3) in the three groups. Group 3 showed the changes better outcomes than the Group 1 and Group 2 at the post hoc significance level in the scores for frailty(p=.039), SPPB (p=.004),TUG ( p<.000), MMSE (p=.018) in the follow up period. However, difference in score for 30-second chair test (p=.112) did not reach statistical significance difference (p>.05). The variables of physical performance (SPPB, TUG, 30-second chair test) and cognitive function (MMSE) showed significant Group×Time interaction (p<.05). The details were shown in Table 3 and Figure 2.

Table 3. Changes in all variables across time among the groups (N = 91)

Note. SPPB-Short Physical Performance Battery; TUG,Time Up and Go;MMSE- Mini Mental State Examination ;SD = standard deviation


Figure 2. Changes in all variables across time among the groups

A-C: Physical Performance. D:cognitive function. SPPB- Short Physical Performance Battery; TUG, Time Up and Go; MMSE- Mini Mental State Examination



To our best knowledge, this is the first three-arm RCT study to compare the effects of three intervention programs on cognitive function and physical performance in community older adults with CF. The main finding in this study revealed that the 6-month MTCC intervention was optimal in reversing cognitive frailty, significantly improving the cognitive function and physical performance. According to the definition of CF by the International Academy on Nutrition and Aging (IANA) (24), in our study participants were recruited based on the cognitive state (a point of 0.5 in CDR, without concurrent dementia) with pre-frailty (scored 1-2 on Fried Frailty Criteria) or frailty level (scored 3-5 on Fried Frailty Criteria). The participants in MTCC group showed a highest reverse rate of CF (30%) at 6-month follow-up. Moreover, the MTCC group also showed significant improvement from baseline in SPPB, TUG and MMSE at the end of intervention and follow-up period.
Regarding the improvement of physical performance, participants in the TCC group and MTCC group gained more benefits than mindfulness group. Both of the TCC and MTCC group received the Chinese traditional exercise program- Tai Chi Chuan. In our study, we applied the 24-simplified TCC to participants which was more suitable for older adults. After 6-month intervention, the physical function of gait speed, static and dynamic balance, range of motion, reflex control and core strength were increased. Previous studies have reported that TCC training leads to positive changes in neuromuscular functions in older adults (13, 25). The findings from our study are aligned with other RCT trails on the effect of TCC exercise on promoting physical benefits in older adults (26-27). As one of the highest-tier evidence-based health-promoting and disease prevention exercise, TCC has been widely adopted by communities in China. In our study, TCC exercise training program was adjusted to be easily implementable according to the condition and demand of older adults. Compared with TCC group, MTCC received modified Tai Chi Chuan program that was a mindfulness-based TCC consisted of training of mind-body coordination, the beneficial effects on physical function was more significant. The aim of practice was to achieve the state of «motion in quietness» and focus on the moment. In MTCC program, participants were taught to carry out gentle TCC postures under the continuous attention of consciousness to avoid brute force so as to improve the effect of exercise.
In terms of cognitive function, significant change in MMSE was found between group and time interaction in the present study. What’s more, the MMSE score in MTCC group yielded the highest among the three groups. The participants in MTCC group received the mindfulness skills which combined with TCC exercise, both of them were beneficial to cognitive recovery. The systemic review reported that mindfulness training has a positive effect on cognitive functions in a wide group of populations, including the aging adults with early cognitive degeneration (28). A recent study found that the mindfulness practice could enhance functional brain connectivity in the default mode network in older adults with cognitive impairment and reduce hippocampal volume atrophy in MCI with positive impacts on brain regions most related to dementia (29). Regarding to the receptivity of the older adults, we integrated the systemic mindfulness skills into daily life and exercise which was easily for older adults to grasp and practice everyday. After the 6-month intervention , most of CF older adults in MTCC group could practice by themselves and continue to practice in future. Training with mindfulness involved TCC, participants cultivated self-regulation of attention, allowing attention to be maintained on the immediate exercise experience allowing for increased recognition in the present that helped them to achieve a “harmony of body and mind” state to promote health.
One major strength of our study was that we adopted a multidomain program – MTCC-involving not only cognitive training (mindfulness skills) but also combined with physical activity (TCC) to facilitate training-driven brain plasticity and better performance in physical and cognitive functions. To our best knowledge, this was the first RCT design to investigate the effectiveness of MTCC program on the cognitive function and physical performance in CF older adults. The MTCC training yielded a notable result in reversing the CF. After 6-month follow up, thirty percentage of participants in MTCC group had no frailty and MCI which was very impressive. For Chinese community older adults, the MTCC program were easy to grasp and apply to their daily life. The lose of follow-up rate was very low (2.75%) which represented participants adhere with the study. Moreover, most of participants reflected the program was helpful and useful and they would like to practice after the end of intervention. Previous studies have proved that cognitive intervention or physical exercise had beneficial effects on older adults (30-32). It was recommended that the comprehensive program which led to both promotion in physical performance and cognitive function should be targeted to the older adults (33-34), however fewer feasible interventions could systematically and effectively improve the conditions of CF older adults. In this study we designed a MTCC program tailored for older adults which was easily implementable and efficient. In accordance with hypothesis, a six-month MTCC training program resulted in a long-term effect on a significant reduction in the CF rate and promotion in the cognitive and physical functions among older adults with CF. During the MTCC training, participants were required to focus on the present moment, involving self awareness and understanding of every movement process which involved posture shift and center of gravity shift. The weight of the body changed all the time to strengthen the ability of control the balance of the body (35-36). The outstanding benefits of MTCC may due to capture the similarities of mindfulness and TCC and combine them to create the overlapping effect through practice.
Although we use a rigorous and well-controlled randomized trials, the study has several limitations. Firstly, due to the limitations of funding, the validated biochemical measures for CF older adults were not used in this study, further study investigations using subjective and objective indicators are recommended to investigate the mechanism of the interventions.
Secondly, although this trial was conducted in multicenter, these communities were in one province. To generalize the findings could be enhanced by involving more states.



In conclusion, our findings proved that mindfulness practice and TCC exercise were beneficial in the domains of cognitive function and physical performance among CF older adults. Moreover, the current randomized controlled trail showed evidence that the tailored MTCC program combined mindfulness practice and TCC exercise was most effective in reversing CF state. MTCC was a efficacious and innovative program that could serve as a model to improve cognitive function and physical performance in community older adults with CF in China.


Funding: This study was supported by grants from the Ministry of education, humanities and social sciences research projects [Fund No.17YJCZH129].

Acknowledgments: All authors were grateful for all the participants and community staffs in this study for their corporation.

Conflict of interest: All authors have declared no conflicts of interest for this article.

Ethical standard: The study design was approved by the Institutional Review Board of Harbin Medical University. All participants signed written informed consent.



1. Sargent, L.,R. Brown, Assessing the Current State of Cognitive Frailty: Measurement Properties. J Nutr Health Aging, 2017;21(2): p.152-160. DOI:10.1007/s12603-016-0735-9
2. Robertson D.A.,Savva G.M.,Kenny R.A.Frailty and cognitive impairment:A review of the evidence and causal mechanisms.Ageing Res Rev,2013,12:840-851. DOI:10.1016/j.arr.2013.06.004
3. Bu, Z, Huang, A, Xue, M,Li, Q,Bai, Y,Xu, G.Cognitive frailty as a predictor of adverse outcomes among older adults: A systematic review and meta-analysis.Brain Behav.2020,06:e01926. DOI:10.1002/brb3.1926
4. Shimada H, Doi T,Lee S,Makizako, H,Chen, LK,Arai, H. Cognitive frailty predicts incident dementia among community-dwelling older people.J Clin Med,2018,7:250. DOI:10.3390/jcm7090250
5. Tarazona-Santabalbina FJ,Gómez-Cabrera MC,Pérez-Ros P,et al.A multicomponent exercise intervention that reverses frailty and improves cognition,emotion,and social networking in the community- dwelling frail elderly:a randomized clinical trial. J Am Med Dir Assoc,2016;17(5):426-33. DOI:10.1016/j.jamda.2016.01.019
6. Yoon, DH; Lee, JY; Song, W.Effects of Resistance Exercise Training on Cognitive Function and Physical Performance in Cognitive Frailty: A Randomized Controlled Trial.J Nutr Health Aging.2018,22:944-951. DOI:10.1007/s12603-018-1090-9
7. Bherer L,Erickson K I,Liu-Ambrose T.A review of the effects of physical activity and exercise on cognitive and brain functions in older adults.J Aging Res,2013;6:57058. DOI:10.1155/2013/657508
8. Apóstolo, JL,Cardoso, DFRosa, AI,Paúl, C.The effect of cognitive stimulation on nursing home elders:A randomized controlled trial. J Nurs Scholarsh, 2014, 46:157-166. DOI:10.1111/jnu.12072
9. Klainin-Yobas, P; Kowitlawakul, Y; Lopez, V; et al.The effects of mindfulness and health education programs on the emotional state and cognitive function of elderly individuals with mild cognitive impairment: A randomized controlled trial.J Clin Neurosci.2019,68:211-217. DOI:10.1016/j.jocn.2019.05.031
10. Lee PS, Jung M, Abraham A, Lei-Rivera L, Kim AH, Tippett WJ. Efficacy of Tai Chi as a technique for vestibular rehabilitation – a pre-liminary quasi-experimental study. J Phys Ther. 2012; 5 :6-13. 805044&partnerID=40&md5=dc32ad7dfe0ef13edc6aa3c46be3ef28.
11. Sattin RW,Easley KA,Wolf SL,Kutner, MH.Reduction in fear of falling through intense tai chi exercise training in older,transitionally frail adults. J Am Geriatr Soc,2005;53:1168-78. DOI:10.1111/j.1532-5415.2005.53375.x
12. Villareal DT,Smith GI,Sinacore DR,Shah, K,Mittendorfer, B.Regular multicomponent exercise increases physical fitness and muscle protein anabolism in frail,obesity. Obesity,2011;19:312-8. DOI:10.1038/oby.2010.110
13. Hu, Y, Kattan, C, Kontos, D,Zhu, W,Hernandez, ME.Benefits of tai ji quan practice on neuromuscular functions in older adults: A Systematic Review and meta-analysis.Complement Ther Clin Pract.2020,42:101295. DOI:10.1016/j.ctcp.2020.101295
14. Li, F,Harmer, P, Eckstrom, E,Fitzgerald, K,Chou, LS,Liu, Y.Effectiveness of Tai Ji Quan vs Multimodal and Stretching Exercise Interventions for Reducing Injurious Falls in Older Adults at High Risk of Falling: Follow-up Analysis of a Randomized Clinical Trial.JAMA Intern Med,2019, 2: e188280. DOI:10.1001/jamanetworkopen.2018.8280
15. Kabat-Zinn J. Mindfulness-Based Interventions in Context: Past, Present, and Future. Clin Psychol Sci Pract. 2003;10(2):144-156. DOI:10.1093/clipsy.bpg016
16. Zhang, JY, Qin, SD, Zhou, YQ,Zhao S,Su H.A randomized controlled trial of mindfulness-based Tai Chi Chuan for subthreshold depression adolescents.Neuropsychiatr Dis Treat.2018,14:2313-2321. DOI:10.2147/NDT.S173255
17. Burschka, JM, Keune, PM,Oy UH, Oschmann, P,Kuhn, P.Mindfulness-based interventions in multiple sclerosis: beneficial effects of Tai Chi on balance, coordination, fatigue and depression.BMC Neurol.2014,14:165. DOI:10.1186/s12883-014-0165-4
18. Fried, L.P.,Tangen, CM,Walston, J,et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci, 2001,56: p. M146-56. DOI:10.1093/gerona/56.3.m146
19. Park, S., Park, SE,Kim, MJ,et al. Development and validation of the Pictorial Cognitive Screening Inventory for illiterate people with dementia. Neuropsychiatr Dis Treat, 2014;10: p. 1837-45. DOI:10.2147/NDT.S64151
20. Li, X, Dai, J, Zhao, S, Liu, W,Li, H.Comparison of the value of Mini-Cog and MMSE screening in the rapid identification of Chinese outpatients with mild cognitive impairment..Medicine (Baltimore).2018,97:e10966 DOI:10.1097/MD.0000000000010966
21. Guralnik, J.M., Ferrucci, L,Pieper, CF,et al. Lower extremity function and subsequent disability: consistency across studies, predictive models, and value of gait speed alone compared with the short physical performance battery. J Gerontol A Biol Sci Med Sci, 2000,55: p.M221-31. DOI:10.1093/gerona/55.4.m221
22. Sasaki, K.,Senda, M,Nishida, K, Ota, H. Preoperative time required for the timed «up and go» test in women with hip osteoarthritis could predict a deep venous thrombosis complication after total hip arthroplasty. Acta Med Okayama, 2010,64: p. 197-201. DOI:10.18926/AMO/40012
23. Rikli RE,Jones CJ.Development and validation of criterion-referenced clinically relevant fitness standards for maintaining physical independence in later years. Gerontologist,2013; 53: 255-67. DOI:10.1093/geront/gns071
24. Abellan van Kan, G.,Rolland, Y., Bergman, H,Morley, JE,Kritchevsky, SB,Vellas, B.The I.A.N.A Task Force on frailty assessment of older people in clinical practice. J Nutr Health Aging, 2008. 12: p. 29-37. DOI:10.1007/bf02982161
25. Fink D, Houston K. Implementing an evidence-based tai ji quan program in a multicultural setting: a pilot dissemination project. J Sport Health Sci. 2014;3:27-31. DOI:org/10.1016/j.jshs.2013.10.003
26. Li, F,Harmer, P, Chou, LS. Dual-Task Walking Capacity Mediates Tai Ji Quan Impact on Physical and Cognitive Function.Med Sci Sports Exerc.2019,51:2318-2324. DOI:10.1249/MSS.0000000000002051
27. Li, F, Harmer, P, Liu, Y,Chou, LS.Tai Ji Quan and global cognitive function in older adults with cognitive impairment:a pilot study.Arch Gerontol Geriatr.2014,58:434-9. DOI:10.1016/j.archger.2013.12.003
28. Ng, TKS; Fam, J; Feng, L; et al.Mindfulness improves inflammatory biomarker levels in older adults with mild cognitive impairment: a randomized controlled trial.Transl Psychiatry.2020,10:21. DOI:10.1038/s41398-020-0696-y
29. Chiesa A, Calati R, Serretti A. Does mindfulness training improve cognitive abilities? A systematic review of neuropsychological findings. Clin Psychol Rev 2011;31:449–64. DOI:10.1016/j.cpr.2010.11.003
30. Gine-Garriga, M.,Roqué-Fíguls, M,Coll-Planas, LSitjà-Rabert,Salvà, A. M Physical exercise interventions for improving performance-based measures of physical function in community-dwelling, frail older adults: a systematic review and meta-analysis. Arch Phys Med Rehabil, 2014. 95: p.753-769 e3. DOI:10.1016/j.apmr.2013.11.007
31. Cameron, I.D.,Fairhall, N,Langron, C,et al. A multifactorial interdisciplinary intervention reduces frailty in older people: randomized trial. BMC Med, 2013;11: p. 65. DOI:10.1186/1741-7015-11-65
32. Arts, M.H., Collard, RM,Comijs, HC.Physical Frailty and Cognitive Functioning in Depressed Older Adults: Findings From the NESDO Study. J Am Med Dir Assoc, 2016;17: p.3. DOI:10.1016/j.jamda.2015.07.016
33. Srisuwan P, Nakawiro D, Chansirikarnjana S, Kuha O, Chaikongthong P.Suwannagoot T.Effects of a Group-Based 8-Week Multicomponent Cognitive Training on Cognition, Mood and Activities of Daily Living among Healthy Older Adults: A One-Year Follow-Up of a Randomized Controlled Trial.The Journal of Prevention of Alzheimer’s Disease 2020;2:112-120. DOI:10.14283/jpad.2019.42
34. Vellas B., Scrase D., Rosenberg G.A., Andrieu S., Araujo de Carvalho I., Middleton L.T. WHO Guidelines on Community-Level Interventions to Manage Declines in Intrinsic Capacity: The Road to Prevention Cognitive Decline in Older Age? The Journal of Prevention of Alzheimer’s Disease 2018;3:165-167. DOI:10.14283/jpad.2018.26
35. Hallisy KM.Tai Chi beyond Balance and fall prevention:health benefits and its potential role in combatting social isolation in the aging population. Curr Geriatr Rep,2018;7:1-12. DOI:org/10.1007/s13670-018-0233-5
36. Chen EW,Fu AS,Chan KM,Tsang, WW.The effects of Tai Chi on the balance control of elderly persons with visual impairment:a randomized clinical trial.Age Ageing,2012;41:254-9. DOI:10.1093/ageing/afr146


A.P. Porsteinsson1, R.S. Isaacson2, S. Knox3, M.N. Sabbagh4, I. Rubino5
1. University of Rochester School of Medicine and Dentistry, Rochester, NY, USA; 2. Weill Cornell Medical Center and New York-Presbyterian, New York, NY, USA;
3. Biogen International GmbH, Baar, Switzerland; 4. Cleveland Clinic Lou Ruvo Center for Brain Health, Las Vegas, NV, USA; 5. Biogen Inc, Cambridge, MA, USA

Corresponding Author: Sean Knox, MBChB. Biogen International GmBH, Neuhofstrasse 30, 6340 Baar, Switzerland. Phone: +41413921976; Email:

J Prev Alz Dis 2021;3(8):371-386
Published online June 9, 2021,


Alzheimer’s disease is a progressive, irreversible neurodegenerative disease impacting cognition, function, and behavior. Alzheimer’s disease progresses along a continuum from preclinical disease, to mild cognitive and/or behavioral impairment and then Alzheimer’s disease dementia. Recently, clinicians have been encouraged to diagnose Alzheimer’s earlier, before patients have progressed to Alzheimer’s disease dementia. The early and accurate detection of Alzheimer’s disease-associated symptoms and underlying disease pathology by clinicians is fundamental for the screening, diagnosis, and subsequent management of Alzheimer’s disease patients. It also enables patients and their caregivers to plan for the future and make appropriate lifestyle changes that could help maintain their quality of life for longer. Unfortunately, detecting early-stage Alzheimer’s disease in clinical practice can be challenging and is hindered by several barriers including constraints on clinicians’ time, difficulty accurately diagnosing Alzheimer’s pathology, and that patients and healthcare providers often dismiss symptoms as part of the normal aging process. As the prevalence of this disease continues to grow, the current model for Alzheimer’s disease diagnosis and patient management will need to evolve to integrate care across clinical disciplines and the disease continuum, beginning with primary care. This review summarizes the importance of establishing an early diagnosis of Alzheimer’s disease, related practical ‘how-to’ guidance and considerations, and tools that can be used by healthcare providers throughout the diagnostic journey.

Key words: Alzheimer’s disease, early diagnosis, diagnostic work-up.


Dementia is among the greatest global health crises of the 21st century. Currently, more than 50 million people are living with dementia worldwide (1), with this number estimated to triple to 152 million by 2050 as the world’s population grows older (2). Alzheimer’s disease (AD) is the most common cause of dementia and is thought to account for 60–80% of dementia cases (3). Currently, the total annual cost for AD and other dementias in the USA is $305 billion and is predicted to increase to more than $1.1 trillion by 2050 (3). This substantial economic burden includes not only healthcare and hospice support for patients with AD (3) but also lost productivity from patients and caregivers (4).
AD is a progressive, neurodegenerative disease associated with cognitive, functional, and behavioral impairments, and characterized by two underlying pathological hallmarks: the progressive accumulation of extracellular amyloid beta (Aβ) plaques and intracellular neurofibrillary tangles (NFTs) (3). In AD, aggregated Aβ plaques are deposited within the brain as a result of either reduced Aβ clearance or excessive production (5); plaque deposition typically occurs ~20 years before the onset of cognitive impairment (6, 7). NFTs are formed by the abnormal accumulation of hyperphosphorylated-tau protein (5); these can be detected 10–15 years before the onset of symptoms (6, 7).
AD follows a progressive disease continuum that extends from an asymptomatic phase with biomarker evidence of AD (preclinical AD), through minor cognitive (mild cognitive impairment [MCI]) and/or neurobehavioral (mild behavioral impairment [MBI]) changes to, ultimately, AD dementia. A number of staging systems have been developed to categorize AD across this continuum (7–9). While these systems vary in terms of how each stage is defined, all encompass the presence/absence of pathologic Aβ and NFTs, as well as deficits in cognition, function, and behavior (7–9). As a result, subtle but important differences exist in the nomenclature for each stage of AD depending on the selected clinical and research classifications (Figure 1).

Figure 1. Stages within the Alzheimer’s disease continuum

The AD continuum can be classified into different stages from preclinical AD to severe AD dementia; the nomenclature associated with each stage varies between the different clinical and research classifications. This figure provides a summary of the different naming conventions that are used within the AD community and the symptoms associated with each stage of the continuum; *Mild behavioral impairment is a construct that describes the emergence of sustained and impactful neuropsychiatric symptoms that may occur in patients ≥50 years old prior to cognitive decline and dementia (112); Abbreviations: Aβ, amyloid beta. AD, Alzheimer’s disease. FDA, Food and Drug Administration. IWG, International Working Group. MCI, mild cognitive impairment. NIA-AA, National Institute on Aging—Alzheimer’s Association

Preclinical AD, as the earliest stage in the AD continuum, comprises a long asymptomatic phase, in which individuals have evidence of AD pathology but no evidence of cognitive or functional decline, and their daily life is unaffected (8) (Figure 1). The duration of preclinical AD can vary between individuals, but typically lasts 6–10 years depending on the age of onset (10, 11). The risk of progression from preclinical AD to MCI due to AD (with/without MBI) depends on a number of factors, including age, sex, and apolipoprotein E (ApoE) status (11, 12); however, not all individuals who have underlying AD pathology will go on to develop MCI or AD dementia (13, 14). A recent meta-analysis of six longitudinal cohorts followed up for an average of 3.8 years found that 20% of patients with preclinical AD progressed to MCI due to AD (11). A further study by Cho et al., with an average follow-up rate of 4 years, found that 29.1% of patients with preclinical AD progressed to MCI due to AD (12).
For patients who do progress to MCI due to AD (with/without MBI), initial clinical symptoms typically include short-term memory impairment, followed by subsequent decline in additional cognitive domains (15) (Figure 1). On a day-to-day basis, an individual with MCI due to AD may struggle to find the right word (language), forget recent conversations (episodic memory), struggle with completing familiar tasks (executive function), or get lost in familiar surroundings (visuospatial function) (15, 16). As individuals have varying coping mechanisms and levels of cognitive reserve, patients’ experiences and symptomology vary widely; however, patients tend to remain relatively independent at this stage, despite potential marginal deficits in function. The prognosis for patients with MCI due to AD can be uncertain; one study that followed up patients with MCI due to AD for an average of 4 years found that 43.4% progressed to AD dementia (12). Other studies reported 32.7% and 70.0% of individuals with MCI due to AD progress to AD dementia within 3.2 and 3.6 years of follow-up, respectively (17, 18). Patients who do progress to AD dementia will develop severe cognitive deficits that interfere with social functioning and will require assistance with activities of daily living (7) (Figure 1). As the disease progresses further, increasingly severe behavioral symptoms will develop that significantly burden patients and their caregivers, and the disease ultimately results in severe loss of independence and the need for round-the-clock care (3).
An early diagnosis of AD can provide patients the opportunity to collaborate in the development of advanced care plans with their family, caregivers, clinicians, and other members of the wider support team. Importantly, it also enables patients to seek early intervention with symptomatic treatment, lifestyle changes to maintain quality of life, and risk-reduction strategies that can provide clinically meaningful reductions in cognitive, functional, and behavioral decline (19–22). It can also help reduce healthcare system costs and constraints: a study by the Alzheimer’s Association found that diagnosing AD in the early stages could save approximately $7 trillion. These savings were due to lower medical and long-term care costs for patients with managed MCI than for those with unmanaged MCI and dementia (3). Furthermore, an early diagnosis will be vital for patients when a therapy addressing the underlying pathology of AD becomes available; currently 19 biologic compounds are under Phase 2 or 3 investigation (23). Physicians will need to be prepared for the approval of these treatments, to optimize the potential benefit and prolong preservation of patients’ cognitive function and independence beyond that associated with current standard of care (19).
As the prevalence of AD continues to grow, the advancement of AD patient diagnosis will require an orchestrated effort, starting in the primary care setting and subsequently involving multiple healthcare provider (HCP) specialties (e.g., nurse practitioner [NP] or physician assistant [PA]) throughout the disease continuum. Galvin et al. recently highlighted the need for HCPs to work as an integrated, patient-centered care team to accommodate the growing and diverse population of patients with AD, beginning with diagnosis (24). For patients to receive a timely diagnosis, it is vital to implement an approach that minimizes the burden placed on the patient, clinician, and healthcare system (25). Here, we summarize the importance of establishing an early diagnosis of AD, related practical ‘how-to’ guidance and considerations, and tools that can be used by healthcare providers throughout the diagnostic journey.

The importance of an early diagnosis

Historically, a diagnosis of AD has been one of exclusion, and one only made in the latter stages of disease (26); however, the disease process can take years to play out, exacting a significant toll on the patient, caregiver, and healthcare system along the way (27).
To mitigate this burden, the early and accurate detection of AD-associated symptoms in clinical practice represents a critically needed but challenging advancement in AD care (19, 28–30). Usually, a patient with early signs/symptoms of AD will initially present in a primary care setting (30). For some patients, minor changes in cognition and/or behavior may be detected during a routine wellness visit or an appointment to discuss other comorbidities (24). As the PCP is often the first to observe a patient’s initial symptomatology, it is vital they recognize the early signs and symptoms, and understand how to use the most appropriate assessment tools designed to detect these early clinical effects of the disease.
Because the neuropathologic hallmarks of AD (Aβ plaques and NFTs) can be detected decades prior to the onset of symptoms (6, 7), biomarkers reflecting this underlying pathology represent an important opportunity for early identification of patients at greatest risk of developing MCI due to AD. Biomarkers support the diagnosis of AD (especially important early on when symptoms can be subtle), and the U.S. Food and Drug Administration (FDA) has recently published guidelines that endorse their use in this population (9). The National Institute on Aging—Alzheimer’s Association (NIA-AA) has recently created a research framework that acknowledges the use of biomarkers for diagnosing AD in vivo and monitoring disease progression (7).
Important biomarker information can be gathered from imaging modalities such as magnetic resonance imaging (MRI) and positive emission tomography (PET) that visualize early structural and molecular changes in the brain, respectively (25, 30). Fluid biomarker testing, such as cerebrospinal fluid (CSF) can also be used; CSF biomarkers can directly reflect the presence of Aβ and aggregated tau within the brain (7, 31). As will be discussed in more depth later in this article, a large number of clinical studies have shown that Aβ and tau biomarkers can contribute diagnostically important information in the early stages of disease (32). There is ongoing research to expand the current range of tests that can be used by clinicians as part of the multistage diagnostic process (25). For instance, once approved, blood-based biomarkers could be used to identify patients at risk of developing AD and for monitoring disease progression (33, 34), which would also reduce the current capacity constraints associated with PET imaging (25).

Practical guide for an early diagnosis of Alzheimer’s disease in clinical practice

As already raised, recent recommendations for evolving AD care to a more patient-centric, transdisciplinary model include guidance on realizing an efficient diagnostic process—one in which HCPs, payers, and specialists are encouraged to combine their efforts to ensure the early warning signs of AD are not overlooked (24). The recommendations include dividing the diagnosis of AD into the following steps: detect, assess/differentiate, diagnose, and treat (Figure 2). We present here a practical guide for the early diagnosis of AD, based on this outlined approach, including a case study to highlight each of these key steps.

Figure 2. A stepwise infographic to highlight key stages within the diagnostic process, along with the recommended tests to support each step

The diagnostic process for AD can be divided into the following steps: detect, assess/differentiate, diagnose, and treat. It is important for clinicians to utilize appropriate tests when investigating a patient suspected of having AD in the early stages. Here, we highlight the most valuable tests for each step and which ones should be used in a primary care or specialist setting; *FDG-PET is usually considered after a diagnostic work-up; Abbreviations: A-IADL-Q, Amsterdam Instrumental Activities of Daily Living Questionnaire. Aβ, amyloid beta. Ach, acetylcholine. BG, blood glucose. CSF, cerebrospinal fluid. FAQ, Functional Activities Questionnaire. FAST, Functional Analysis Screening Tool. FDG-PET, fluorodeoxyglucose-PET. GDS, Geriatric Depression Scale. IQCODE, Informant Questionnaire on Cognitive Decline in the Elderly. Mini-Cog, Mini Cognitive Assessment Instrument. MMSE, Mini-Mental State Examination. MoCA, Montreal Cognitive Assessment. MRI, magnetic resonance imaging. NMDA, N-Methyl-D-aspartic acid. NPI-Q, Neuropsychiatric Inventory Questionnaire. PCP, primary care physician. PET, positive emission tomography. p-tau, phosphorylated tau. QDRS, Quick Dementia Rating System. TSH, thyroid-stimulating hormone. t-tau, total tau

Table 1. Patient case study

Abbreviations: Aβ, amyloid beta. ApoE, apolipoprotein E. HgbA1c, hemoglobin A1c. MoCA, Montreal Cognitive Assessment. MRI, magnetic resonance imaging. PCP, primary care physician. p-tau, phosphorylated tau. t-tau, total tau

Step 1: Detect

The role of primary care in the early detection of AD

The insidious and variable emergence of symptoms associated with AD and other dementias can make recognition extremely challenging, particularly in a primary care setting (30, 35). Clinicians often have limited time with patients, so it is vital that they are able to quickly and accurately recognize the early signs and symptoms associated with AD (Table 2) (3, 30, 36), and training for nurses, NPs, and PAs who may have more time to observe patients should provide substantial benefits. Although extremely variable, initial symptoms may include short-term memory loss or psychological concerns, including depressive symptoms and a loss of purpose (36).

able 2. Symptoms associated with suspected early stage Alzheimer’s disease

Patients, family members, and even HCPs themselves may present barriers to the diagnosis of early-stage AD. Patients may hide their symptoms or even avoid making an appointment until their symptoms significantly affect their day-to-day life due to fear of the stigma associated with a diagnosis of AD (19). Additionally, patients, family members, and PCPs/HCPs may dismiss or misinterpret symptoms as simply part of the normal aging process (30). Retrieving information from a trusted family member or informant/caregiver is essential when trying to assess a patient for suspected AD, as this perspective can provide a more objective understanding of the daily routine, mood, and behavior of the patient, and how this may have changed over time (30). For patients presenting with even subtle symptoms associated with AD, it is important that the PCP/HCP conducts an initial assessment to confirm the presence of symptoms using a validated assessment for early-stage AD detection (Figure 2; Step 2: Assess/Differentiate).

Case study: Presentation

A 63-year-old Caucasian male (J.K.) presented to his PCP with short-term memory loss over the last 2 years (Table 1A). Accompanied by his wife, he acknowledged his job had been affected by issues with his short-term memory; however, he considered his memory similar to that of his peers. His wife reported that people at work had started to notice him struggling to keep up, and also that family had to remind him of his upcoming appointments. He admitted to having intermittent depressive symptoms and anxiety, as well as irritability. Based on the patient’s symptoms, the PCP felt his presentation warranted further clinical assessment.

Step 2: Assess and differentiate

Primary care: Initial assessment when a patient presents

When a patient initially presents with symptoms consistent with early stages of AD, a clinician must first conduct a comprehensive clinical assessment to rule out other potential non-AD causes of cognitive impairment (Figure 2). PCPs are well placed to conduct these initial assessments, as they may not require specialist input or hospital tests. During the initial assessment, the primary objective of the clinician should be to exclude possible reversible causes of cognitive impairment, such as depression, or vitamin, hormone, and electrolyte deficiencies (37). The initial assessment should include a thorough history to identify potential risk factors associated with AD, including a family history of AD or related dementias in first-degree relatives (31, 38). Other known risk factors for AD that should be identified include age, female sex, ApoE ε4 status, physical inactivity, low education, diabetes, and obesity (3). It is also important to review for pre-existing medical conditions or prescribed medications that could be a cause of the patient’s cognitive impairment (36). Additionally, when conducting a thorough history, open-ended, probing questions should be directed to both the patient and the informant to ascertain how the patient’s cognition has changed over time and how the cognitive deficits affect their everyday activities; example questions for the initial assessment are detailed in Table 3 (30). Engaging with informants/caregivers is key to capturing additional information to help support all assessments. A routine differential diagnosis of AD begins with a detailed history, physical and neurologic examinations, and bloodwork analyses, followed by cognitive assessments and functional evaluation (Figure 2).

Table 3. Example questions for a clinician conducting an initial assessment with a patient and caregiver (30)

Primary care: Physical examination and blood analyses

A physical examination and blood tests can identify comorbid contributory medical conditions and reversible causes of cognitive impairment. A physical examination, including a mental status and neurological assessment, should be conducted to detect conditions such as depression and, for example, to look for signs such as issues with speaking or hearing as well as signs that could indicate a stroke (37). As part of the physical exam, a physician may ask the patient about diet and nutrition, review all medications (to see if these are the cause of any cognitive impairment, e.g. anti-cholinergics, analgesics, or sleep aids and anxiolytics), check blood pressure, temperature and pulse, and listen to the heart and lungs (36, 39).
Blood tests can rule out potentially treatable illnesses as a cause of cognitive impairment, such as vitamin B12 deficiency or thyroid disease (37). Suggested blood analyses include: 1) complete blood cell count; 2) blood glucose; 3) thyroid-stimulating hormone; 4) serum B12 and folate; 5) serum electrolytes; 6) liver function; and 7) renal function tests (30). Although not routinely used in clinical practice, clinicians may request ApoE genotyping, as this can help assess the genetic risk of developing AD. ApoE is the dominant cholesterol carrier within the brain that supports lipid transport and injury repair (40, 41), and the APOE gene exists as three polymorphic alleles: APOE ε2, ε3, and ε4. The ε4 allele of ApoE is associated with increased AD risk, whereas the ε2 allele is protective (40, 42). The number of ApoE ε4 alleles a person carries increases their risk of developing AD and the age of disease onset (43). Homozygous ε4 carriers (those with two copies of the ε4 allele) have the greatest risk of developing AD and the lowest average age of onset (43). In some practice settings, ApoE genotyping can only be conducted by a genetic counselor; a referral for more comprehensive genetic testing may be considered by the HCP if there is a family history of early-onset AD or dementia. Consumer tests are also becoming more readily available for patients wanting to determine their risk of developing diseases such as AD based on genetic risk factors (44).

Primary care: Cognitive, functional, and behavioral assessments

Cognitive assessments

If a patient is suspected of having AD following an initial assessment in primary care, and they are <65 years old, or if the case is complex, a referral to a dementia specialist such as a neurologist, geriatrician, or geriatric psychiatrist may be required for further evaluation. The specialist would then use an appropriate battery of cognitive, functional, and behavioral tests to assess the different aspects of disease, and ultimately to confirm diagnosis. However, not all patients with suspected cognitive deficits are immediately referred to a dementia specialist at this stage, which is only partly due to limited numbers of specialists (25) (Figure 2). In clinical practice, a two-stage process is often employed. This involves an initial ‘triage’ step conducted by non-specialists to clinically assess and select those patients who require further evaluation by a dementia specialist (45). During this ‘triage’ step, there are several clinical assessments available to non-specialists for assessing the presence of cognitive and functional impairments and behavioral symptoms (Table 4) (28, 35, 46–55).

Table 4. Cognitive, functional, and behavioral assessments to support the diagnosis of Alzheimer’s disease in a primary care and specialist setting

*Personal communication; Abbreviations: AD, Alzheimer’s disease. A-IADL-Q, Amsterdam Instrumental Activities of Daily Living Questionnaire. FAQ, Functional Activities Questionnaire. FAST, Functional Assessment Screening Tool. GDS, Geriatric Depression Scale. IQCODE, Informant Questionnaire on Cognitive Decline in the Elderly. MCI, mild cognitive impairment. Mini-Cog, Mini Cognitive Assessment Instrument. MMSE, Mini-Mental State Examination. MoCA, Montreal Cognitive Assessment. NPI-Q, Neuropsychiatric Inventory Questionnaire. QDRS, Quick Dementia Rating System

Previous research has shown that clinicians have a tendency to choose one assessment over another due to their familiarity with the assessment, time constraints, or specific resources available to them within their clinic (30), but clinicians need to be aware of, and prepared to use, the most patient-appropriate assessments: the cultural, educational, and linguistic needs of the patient are important considerations (30, 36, 56–58). Some assessments have been translated into different languages or shortened, or have education-adjusted scoring classifications, where required (56–58).
Cognitive assessments that can be conducted quickly (<10 minutes), such as the Mini-Mental State Examination (MMSE) or Montreal Cognitive Assessment (MoCA), can be used by non-specialists to identify the presence and severity of cognitive impairment in patients before referring to a dementia specialist (Table 4) (36). Both the MMSE and MoCA are used globally in clinical practice, particularly in primary care, but vary in terms of their sensitivity to identify AD in the early stages (28, 59). The MMSE is sensitive and reliable for identifying memory and language deficits in general but has limitations in identifying impairments in executive functioning (59). MoCA was originally developed to improve the detection of MCI (28) and is more sensitive than the MMSE in its assessment of memory, visuospatial, executive, and language function, and orientation to time and place (59). Both tests are relatively easy to administer and take around 10 minutes to complete. Neither assessment requires extensive training by the clinician, although MoCA users do need to undergo a 1-hour certification as mandated by the MoCA Clinic and Institute (28, 60).
For time-constrained clinicians, the Mini Cognitive Assessment Instrument (Mini-Cog) may be an appropriate tool to assess cognitive deficits that focus on memory, and components of visuospatial and executive function (Table 4). The assessment includes the individual learning three items from a list, drawing a clock, and then recalling the three-item list. The Mini-Cog can be useful for clinicians in primary care, as it requires no training and the results are easy to interpret. As an alternative to these tests, PCPs might also consider using an informant-based structured questionnaire such as the AD8 or Informant Questionnaire on Cognitive Decline in the Elderly to help guide discussions with the patient and caregiver (Table 4) (28).

Functional assessments

Functional assessments are valuable in identifying changes in a patient’s day-to-day functioning through the evaluation of their instrumental activities of daily living (IADLs). IADLs are complex activities that are necessary for the individual to function independently (e.g., cooking, shopping, and managing finances) and can be impaired during the early stages of cognitive impairment. While it is possible that functional decline may occur as a part of normal aging, a decline in a person’s IADL performance is strongly associated with neurodegenerative diseases such as AD (61). In the early stages of AD, patients may be functionally independent, and any impairment in IADLs may be subtle, such as difficulties paying bills or driving to new places. A patient’s functional independence is essential for their well-being and mental health (62), particularly in the early stages of the disease when the individual may still be working and socializing relatively independently (3). Consequently, functional independence is one of the most important clinical features for patients with AD. As the disease progresses, and patients have increasing functional impairment, this significantly impacts on their independence, and subsequently their and their family/caregiver’s quality of life.
Functional assessment is, therefore, an integral part of the diagnostic process for AD. The Functional Activities Questionnaire (FAQ) is an informant questionnaire that assesses the patient’s performance over a 4-week period and may take only a few minutes to complete (Table 4). The questionnaire is scored from ‘normal’ to ‘dependent’, using numerical values assigned to categories, with higher scores indicative of increasing impairment (47). Previous research has shown that the FAQ has high sensitivity and reliability for detecting mild functional impairment in patients with MCI (47).
Determining an individual’s functional independence can be challenging and the clinician may require additional input from an informant to determine a patient’s functional decline and their ongoing ability to conduct activities of daily living (37). The clinician can gain greater insight through the informant into the patient’s day-to-day life and any issues the patient is having at home. This type of information is vital to the clinician, and when combined with other assessment tools, can help to narrow the differential diagnosis.

Behavioral assessments

Patients with suspected AD may experience several behavioral symptoms such as anxiety, disinhibition, apathy, and depression (Table 2). In the early stages of disease, such symptoms are generally associated with poor long-term outcomes and caregiver burden, and are particularly distressing to both patients and their families (63). It is important for clinicians to use appropriate assessments to identify behavioral and psychiatric symptoms that are caused by neurodegenerative diseases, such as AD, rather than by alternative causes, such as a mood disorder.
The Geriatric Depression Scale (GDS) and Neuropsychiatric Inventory Questionnaire (NPI-Q) can be used by clinicians to assess neuropsychiatric symptoms in patients for whom early-stage AD is suspected (Table 4). The GDS is a 15-item (or longer 30-item) questionnaire that assesses mood, has good reliability in older populations for detecting depression, and can be completed by the patient within 5–10 minutes (63). The NPI-Q can be used in conjunction with or as an alternative to the GDS. The NPI-Q is completed by a knowledgeable informant or caregiver who can report on the patient’s neuropsychiatric symptoms. The NPI-Q can be conducted in around 5 minutes to determine both the presence and severity of symptoms across several neuropsychiatric domains including depression, apathy, irritability, and disinhibition (49). Consequently, as it assesses depression, it can be used as an alternative to GDS if time constraints do not allow for both to be completed.
Behavioral symptoms can be non-specific, so it is important for clinicians to consider and rule out other potentially treatable causes of impairment when assessing this domain. For example, depression is associated with concentration and memory issues (64); apathy can occur in non-depressed elderly individuals and can impact cognitive function (65). Signs/symptoms such as social withdrawal, feelings of helplessness, or loss of purpose should be investigated closely, as these could be indicative of depression alone. It is important for clinicians to recognize that if changes over time in cognitive symptoms and mood symptoms match, then depression is most likely to be the root cause of subtle cognitive decline, rather than AD (28).

Primary care clinician checklist

If AD is still suspected following clinical assessment, referral to a specialist for further diagnostic testing, including imaging and fluid biomarkers, may be required. It is important the clinician confirms the following checks/assessments before the patient undergoes further evaluation:

Primary care clinician checklist

• Confirm medical and family history
• Review the patient’s medications for any that could cause cognitive impairment
• Perform blood tests to eliminate potential reversible causes of cognitive impairment
• Conduct a quick clinical assessment to confirm the presence of cognitive impairment

Specialist role in assessment

Following the initial assessment in primary care, further cognitive, behavioral, functional, and imaging assessments can be carried out in a specialist setting. With their additional AD experience, access to other specialties, and possibly fewer time constraints than the PCP, the specialist is able to conduct a more comprehensive testing battery, using additional clinical assessments and biomarkers to determine causes of impairment and confirm diagnosis (Figure 2).

Cognitive assessments

Because the cognitive impacts of early-stage AD may vary from patient to patient, it is important to consider which cognitive domains are affected in these early stages when considering which assessments to use. Specialists are able to conduct a full neuropsychological test battery that covers the major cognitive domains (executive function, social cognition/emotions, language, attention/concentration, visuospatial and motor function, learning and memory); preferably, a battery should contain more than one test per domain to ensure adequate sensitivity in capturing cognitive impairment (66). This step can help with obtaining an in-depth understanding of the subtle changes in cognition seen in the early stages of AD and enables the clinician to monitor subsequent changes over time.
Typically, episodic memory, executive function, visuospatial function, and language are the most affected cognitive domains in the early stages of AD (29, 67, 68). Currently, most cognitive assessment tools focus on a subset of the overall dimensions of cognition; it is therefore vital the clinician chooses the correct test to assess impairment in these specific cognitive domains that could be indicative of AD in the early stages. As cognitive impairment in the early stages of AD can be subtle and vary significantly between individuals (29), clinicians must choose appropriate, sensitive tests that can detect these changes and account for a patient’s level of activity and cognitive reserve (29). If there is large disparity in results across cognitive assessments, it is important for the clinician to shape their assessments based on the patient’s history. If the patient’s history is positive for neurodegenerative disease, but one assessment does not reflect this, it is important to conduct further tests to ascertain the cause of the cognitive impairment.
The Quick Dementia Rating System (QDRS) can be used by specialists to assess cognitive impairment (Table 4). This short questionnaire (<5 minutes) is completed by a caregiver/informant and requires no training. The QDRS assesses several cognitive domains known to be affected by AD, including memory, language and communication abilities, and attention. The questionnaire can reliably discriminate between individuals with and without cognitive impairment and provides accurate staging for disease severity (28).

Functional assessments

The Amsterdam IADL Questionnaire (A-IADL-Q) and Functional Assessment Screening Tool (FAST) can both be used to assess a patient’s functional ability (Table 4) (53). The A-IADL-Q is a reliable computerized questionnaire that monitors a patient’s cognition, memory, and executive functioning over time. This questionnaire is completed by an informant of the patient and takes 10 minutes to complete (53). For patients with suspected early stage AD, the A-IADL-Q is a useful tool to monitor subtle changes in IADL independence over time and is less influenced by education, gender, and age than other functional assessments (53). The FAST is a useful assessment for clinicians to identify the occurrence of functional and behavioral problems in patients with suspected AD. The questionnaire is completed by informants who interact with the patient regularly; informants are required to answer Yes/No to a number of questions focusing on social and non-social scenarios (55).

Structural imaging

Structural imaging, such as MRI, provides clinically useful information when investigating causes of cognitive impairment (69) (Figure 2). MRI is routinely conducted to exclude alternative causes of cognitive impairment, rather than support a diagnosis of AD (37, 70). It is well known that medial temporal lobe atrophy is the best MRI marker for identifying patients in the earliest stages of AD (70, 71); however, specific patterns of atrophy may also be indicative of other neurodegenerative diseases. Atrophy alone is rarely sufficient to make a diagnosis. MRI findings can help to narrow the differential diagnosis, and the results should be considered in the context of the patient’s age and clinical examination (69–71).
Clinicians are advised to take a stepwise approach when reviewing structural imaging reports of a patient with suspected AD. These steps include: 1) excluding brain pathology that may be amenable to surgical intervention (e.g., the scan will show regions of hyper- or hypointensity rather than a uniform signal); 2) assessing for brain microbleeds (e.g., looking at signal changes within different areas of the brain can identify vascular comorbidities); and 3) assessing atrophy (e.g., medial temporal lobe atrophy is characteristic of AD) (69). Radiologists can conduct a quick and easy visual rating of any medial temporal lobe atrophy; these results can then be utilized by the specialist, in conjunction with a clinical assessment, to determine the likely cause of cognitive impairment. If the clinician is unable to determine a differential diagnosis, additional confirmatory tests can be requested.
Fluorodeoxyglucose-PET (FDG-PET) is a useful structural imaging biomarker that can support an early and differential diagnosis (72); however, specialists usually prefer to use this after their initial diagnostic work-up. As the brain relies almost exclusively on glucose as its source of energy, FDG (a glucose analog) can be combined with PET to identify regional patterns of reduced brain metabolism and neurodegeneration (70,72). FDG-PET is not recommended for diagnosing patients with preclinical AD, as there is no way to ascertain whether the hypometabolism is directly related to AD pathology (73); however, clinicians may refer patients with more established symptomatology for an FDG-PET scan to identify regions of glucose hypometabolism and neurodegeneration that could be indicative of AD (70).

Case study: Assess/differentiate

The initial assessment by the primary care clinician revealed that J.K.’s medical history was significant for hypertension, dyslipidemia, mild obesity, and glucose intolerance (Table 1B). There was no history of cerebrovascular events, significant head injuries, or focal findings on the neurologic exam. Besides the vascular risk factors, no medical conditions or current medications were found to be likely contributors to the cognitive deficit. The patient had a positive family history of dementia, where the onset typically occurred in the late 60s. Genotyping showed the patient to be a homozygous carrier of two ApoE ε4 alleles. Blood tests revealed elevated serum glucose and C-reactive protein but were otherwise normal. The patient had an unremarkable mental status examination, and his MoCA score was 21/30, with points lost on orientation, recall, and naming (Table 1C).
The patient was referred to a memory clinic for further assessment. The dementia specialist referred the patient for an MRI that predominantly showed mild small vessel disease and mild generalized atrophy with a significant reduction in hippocampal volume and ratio. Based on his medical and family history, cognitive assessments, and structural imaging results, the specialist deemed the severity of cognitive impairment to be in the mild range; consequently, the specialist referred the patient for biomarker assessment to determine the underlying cause.

Step 3: Diagnose

Historically, AD was only diagnosed postmortem until we developed the ability to ascertain the underlying pathology associated with the disease in new ways, namely imaging and fluid biomarkers. However, despite supportive results from single- and multicenter trials, the use and reimbursement of imaging and fluid biomarkers to support the diagnosis of AD still vary considerably between countries (70).

Imaging biomarkers

Recent advances have allowed physicians to visualize the proteins associated with AD, namely Aβ and tau, via PET scanning. Amyloid PET is currently the only imaging approach recommended by the Alzheimer’s Association and the Amyloid Imaging Task Force to support the diagnosis of AD (70). Amyloid PET utilizes tracers (florbetapir, flutemetamol, and florbetaben) that specifically bind to Aβ within amyloid plaques; a positive amyloid PET scan will show increased cortical retention of the tracer in regions of Aβ deposition within the brain (74), thus confirming the presence of Aβ plaques in the brain (74, 75) and directly quantify brain amyloid pathology (76), thus making it a useful tool to supplement a clinical battery to diagnose AD (3, 74). However, a positive amyloid PET scan alone does not definitively diagnose clinical AD, and these results must be combined with other clinical assessments, such as cognitive assessment, for an accurate diagnosis (74). It is also important to note that amyloid PET is expensive and not readily reimbursed by health insurance providers (70); if it is not possible to access amyloid PET, biomarker confirmation can be assessed using CSF.

Fluid biomarkers

An additional or alternative tool to amyloid PET is the collection and analysis of CSF for the presence of biomarkers associated with AD pathology. Patients who have symptoms suggestive of AD can be referred for a lumbar puncture to analyze their CSF for specific AD-associated biomarkers (3). CSF biomarkers are measures of the concentrations of proteins in CSF from the lumbar sac that reflect the rates of both protein production and clearance at a given timepoint (7). Lumbar punctures can be conducted safely and routinely in an outpatient setting or memory clinic (77). However, many patients still worry about the pain and possible side effects associated with the procedure and may require additional information and support from the clinician to undertake the procedure (77). Appropriate use criteria are available for HCPs to help identify suitable patients for lumbar puncture and CSF testing (78). For example, individuals presenting with persistent, progressing, and unexplained MCI, or those with symptoms suggestive of possible AD, should be referred for lumbar puncture and CSF testing (78). However, lumbar puncture and CSF testing are not recommended for determining disease severity in patients who have already received a diagnosis of AD or in lieu of genotyping for suspected autosomal dominant mutation carriers (78).
Because there is strong concordance between CSF biomarkers and amyloid PET, either can be used to confirm Aβ burden (79). As such, CSF biomarkers are widely accepted within the AD community to support a diagnosis (80). AD biomarkers from the brain can be detected in CSF well before the onset of overt clinical symptoms in early-stage AD (6, 7). Core AD CSF biomarkers, such as Aβ42 (one of two main isoforms of Aβ and a major constituent of Aβ plaques) and phosphorylated tau (p-tau) and total tau (t-tau), can be measured to determine the presence of disease (80).
When interpreting CSF analyses for a patient with suspected AD, it is important to remember that AD is associated with decreased CSF Aβ42 and increased tau isoforms (32). Decreased CSF Aβ42 levels are a reflection of increased Aβ aggregation and deposition within the brain (32), and the concentration of CSF Aβ42 directly relates to the patient’s amyloid status (e.g., the presence or absence of significant amyloid pathology) and the total amount of Aβ peptides (e.g., Aβ42 and Aβ40) (32). Specialists’ use of ratios of these CSF biomarkers (e.g., Aβ42/40) rather than single CSF biomarkers alone has been shown to adjust for potential differences in Aβ production and provide a better index of the patient’s underlying amyloid-related pathology (81). The increase in CSF p-tau and t-tau associated with AD may directly reflect the aggregation of tau within the brain and neurodegeneration, respectively (32). P-tau in CSF provides a direct measure of the amount of hyperphosphorylated tau in the brain, which is strongly suggestive of the presence of NFTs, whereas CSF t-tau can predict the level of neurodegeneration in a patient with suspected AD; however, t-tau is also increased in other neurologic conditions (32).
Ultimately, the clinical decision to use amyloid PET or CSF to confirm amyloid and tau pathology can be affected by several practical factors (Table 5) (70, 77, 80, 82–85).

able 5. Comparison of key CSF and amyloid PET considerations for amyloid confirmation

Abbreviations: CSF, cerebrospinal fluid. PET, positron emission tomography

Emerging diagnostic tools

Access constraints for amyloid PET have driven the need for alternative sensitive and specific CSF and blood-based biomarkers that can detect AD-associated pathology in the early stages (86). Significant efforts have been undertaken over the last decade to identify blood-based biomarkers to: 1) detect AD pathology; 2) identify those at risk of developing AD in the future; and 3) monitor disease progression (33, 34, 87). At present, only a limited number of approved blood-based assays are available to clinicians to detect AD pathology (88); however, several novel assays are currently under investigation, including those measuring various phosphorylated forms of tau, including p-tau181 and p-tau217 (89). Investigational use of plasma p-tau181 (an isoform of tau) has been shown to differentiate AD from other neurodegenerative diseases and predict cognitive decline in patients with AD (33). CSF p-tau217 (a different isoform of tau) is a promising biomarker under investigation for detecting preclinical and advanced AD (86, 90). Given that blood testing is already a well-established part of clinical routines globally and can easily be performed in a variety of clinical settings, blood-based biomarkers could in future serve as the potential first step of a multistage diagnostic process. This would be a benefit to clinicians, particularly those in primary care, by helping to identify individuals requiring a referral to a specialist for diagnostic testing (87).

Case study: Diagnose

J.K. underwent a lumbar puncture for CSF analysis, which showed decreased Aβ42 and increased p-tau and t-tau protein (Table 1D). Based on the results from the genotyping, cognitive assessments, MRI, and CSF biomarkers, the clinician confirmed that the likely cause of the patient’s cognitive deficits was early-stage AD, especially in view of a positive family history of dementia with similar age of onset.

Step 4: Treat

The role of the clinician following a diagnosis of early-stage AD is to discuss the available management and treatment options while providing emotional and practical support to the patient, caregiver, and family where appropriate (37). Clinicians can also refer the patient and their caregiver(s) to social services for further support, as well as help connect them with reliable sources of information and even local research opportunities and clinical trials.
One important role for a clinician treating a patient diagnosed with early-stage AD is to closely monitor the patient’s disease progression through regular follow-up appointments (e.g., every 6–12 months); clinicians should encourage patients (and the caregiver) to make additional follow-up appointments, especially should symptoms worsen. Routine cognitive and functional assessments (Table 4) should be used to monitor disease progression; these tools can be used to identify unexpected trends, such as rapid decline, which could prompt the need for additional medical evaluation such as blood tests, imaging, or biomarker analyses. Results from such tests could help guide management and/or treatment decisions over the course of the patient’s disease.
Non-pharmacologic therapies (e.g., diet and exercise) may be employed for patients with early AD, with the goal to maintain or even improve cognitive function and retain their ability to perform activities of daily living. For patients in the early stages of disease, dietary changes (e.g., following a healthy diet high in green, leafy vegetables, fish, nuts, and berries), physical exercise, and cognitive training have demonstrated small but significant improvements in cognition (36, 91). Non-pharmacologic therapies can have a positive impact on quality of life and are generally safe and inexpensive (36); however, compliance with these non-pharmacologic therapies should be monitored by the clinician. Research suggests that multimodal therapies, such as cognitive stimulation therapy, may also be more effective when used in combination with pharmacologic treatments (91).
Several pharmacologic treatments have received regulatory approval to treat the symptoms of mild to severe AD dementia. Acetylcholinesterase inhibitors (donepezil, rivastigmine, and galantamine) and N-methyl-D-aspartate receptor antagonists (memantine) can be prescribed to patients to temporarily ameliorate the symptoms of AD dementia such as cognitive and functional decline (92–96). Meta-analyses of donepezil, rivastigmine, and galantamine have shown that patients with mild-to-moderate AD dementia experience some benefits in cognitive function, activities of daily living, and clinician-rated global clinical state (93, 94, 97). Furthermore, treatment with acetylcholinesterase inhibitors and/or memantine has also been shown to modestly improve measures of global function and temporarily stabilize measures of activities of daily living (96). However, it is important to note that these drugs provide only temporary, symptomatic benefit and that not all patients respond to treatment (36, 98). Critically, none of the current drugs available address the underlying pathophysiology or alter the ultimate disease course.
Following AD diagnosis, a comprehensive approach toward clinical care can be individualized based on the patient’s specific AD risk factors (20, 21). Clinicians should consider managing uncontrolled vascular risk factors (e.g., hypertension, hyperlipidemia, diabetes) with antithrombotics, antihypertensives, lipid-lowering, and/or antidiabetic agents, respectively, to reduce the risk of cerebrovascular ischemia and stroke, and subsequent cognitive decline (36, 99). They should also consider the management of the patient’s behavioral symptoms. For most patients in the early stages of disease, behavioral symptoms will be relatively mild, and no pharmacologic management is required; however, pharmacologic treatment, such as a low-dose selective serotonin reuptake inhibitor, can be prescribed for patients with AD-associated depression and anxiety (100, 101).

Specialist clinician checklist

The specialist’s role is critical to further evaluating the initial checks/assessments, providing the diagnosis, and developing the individualized patient management plan:
• Identify deficits to specific cognitive domains using appropriate tests
• Confirm functional performance, using patient and caregiver assessments
• Perform structural imaging to complete assessment of the patient
• Confirm diagnosis with imaging or fluid biomarkers
• Develop a personalized management and follow-up plan
• Direct the patient to additional support resources such as the Alzheimer’s Association

Case study: Treat

Following diagnosis, J.K. was advised on the available management options and research opportunities (Table 1E). The specialist emphasized the need to control his vascular risk factors and suggested lifestyle modifications to optimize the management of his other medical problems. The patient’s neuropsychiatric symptoms were considered mild and did not require pharmacologic intervention. The patient was also provided with details for a local social worker and directed toward further disease-specific information from the Alzheimer’s Association related to his disease. The patient was encouraged to return for additional follow-up visits so that his disease and associated symptoms could be appropriately monitored and managed.

Future perspectives

An early diagnosis of AD will become increasingly important as treatments that alter the underlying disease pathology become available—particularly given the expectation that such treatments will be more effective in preserving cognitive function, and thus prolonging independence, when given early in the course of the disease (19). The approval of such treatments will likely lead to an increased awareness of cognitive impairment and other AD-associated symptoms among both the public and non-specialists, such as those in primary care settings. This may encourage more patients/family members to seek help at an earlier stage of disease than is currently seen in community practice. Increased use of sensitive screening measures to proactively assess for the presence of AD symptoms will help identify patients suspected of having early AD. Assessment of cognitive impairment during a Medicare Annual Wellness Visit is inconsistent; the U.S. Preventative Services Task Force, whilst recognizing the importance of MCI, has maintained its decision that there is insufficient evidence to support the mandate of cognitive screening. However, sensitive screening procedures, along with the availability of disease-modifying treatments, are likely to change their recommendations. There is also a need for a mandated, standardized screening approach internationally. Together, this will result in an increase in patients requiring diagnosis, increasing the demand for specialists to evaluate and diagnose, the need for amyloid confirmation, and wait times for patients, which will collectively put further pressure on an already-stretched healthcare infrastructure (25).
Nevertheless, efforts continue within the AD field to streamline the diagnostic process. Planning for and implementing change will not only improve patient management now but also help prepare healthcare systems for an approved disease-modifying treatment for AD. A flexible, multidisciplinary team approach is recommended to integrate the care needed to detect, assess, differentiate, diagnose, treat, and monitor a diverse AD population (24). The development of tests that could be carried out routinely in a primary care setting, such as blood-based AD biomarkers, would help PCPs and non-specialists identify which patients may need further evaluation or referral to a specialist (25). Interest also remains high in advancing imaging techniques, such as amyloid and tau PET, to support a diagnosis of AD. Although amyloid and tau PET are not currently readily available, they may be useful for specialists in the future to determine disease staging or track progression, or as a surrogate marker of cognitive status (74). The introduction of new screening and diagnostic tools could ultimately help lower the burden on specialists and ensure patients are diagnosed in a timely manner.


Consensus within the AD community has recently shifted to encourage the diagnosis of AD as early as possible. This shift will enable patients to plan their future and consider symptomatic therapies and lifestyle changes that could reduce cognitive deficits and ultimately help preserve their quality of life. Promisingly, new, potentially disease-modifying therapeutic candidates are on the horizon that could be effective in early AD by targeting and ameliorating the underlying biological mechanisms (92, 102). This paper has outlined a menu of practical tools for clinicians to use in the real world to support an early diagnosis of AD and how they may best be incorporated into current clinical practice. Ultimately, a coordinated, multidisciplinary approach that encompasses primary care and specialist expertise is required to ensure timely detection, assessment and differentiation, diagnosis, and management of patients with AD.

Authors’ contributions: All authors participated in the review of the literature and in the drafting and reviewing of the manuscript. All authors read and approved the final version of the manuscript for submission.

Funding: The authors developed this manuscript concept during an assessment of Alzheimer’s disease educational needs. The development of this manuscript was funded by Biogen. Editorial support was provided by Jodie Penney, MSc, PhD, Helios Medical Communications, Cheshire, UK, which was funded by Biogen.

Acknowledgements: The authors would like to acknowledge and thank Dr. Giovanni Frisoni, Geneva University Neurocenter, for his contribution towards the development of this manuscript.

Conflict of Interest: AP reports personal fees from Acadia Pharmaceuticals, Alzheon, Avanir, Biogen, Cadent Therapeutics, Eisai, Functional Neuromodulation, MapLight Therapeutics, Premier Healthcare Solutions, Sunovion, and Syneos; grants from Alector, Athira, Avanir, Biogen, Biohaven, Eisai, Eli Lilly, Genentech/Roche, and Vaccinex. RI has nothing to disclose. MS reports personal fees from Alzheon, Athira, Biogen, Cortexyme, Danone, Neurotrope, Regeneron, Roche-Genentech, and Stage 2 Innovations; stock options from Brain Health Inc, NeuroReserve, NeuroTau, Neurotrope, Optimal Cognitive Health Company, uMethod Health, and Versanum Inc. Additionally, he has intellectual property rights with Harper Collins. SK and IR report employment with Biogen.

Electronic supplementary material: Practical Guidance extender video.

Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits use, duplication, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.


1. Hughes J. This is one of the biggest global health crises of the 21st century. World Econ Forum 2017.’s%20Day%2C%20the,crises%20of%20the%2021st%20century (accessed May 18, 2020).
2. Alzheimer’s Disease International. World Alzheimer Report 2019. Attitudes to dementia 2019. (accessed February 13, 2020).
3. Alzheimer’s Association. 2020 Alzheimer’s disease facts and figures. Alzheimers Dement 2020;16:391–460.
4. Deb A, Thornton JD, Sambamoorthi U, Innes K. Direct and indirect cost of managing alzheimer’s disease and related dementias in the United States. Expert Rev Pharmacoecon Outcomes Res 2017;17:189–202.
5. Serrano-Pozo A, Frosch MP, Masliah E, Hyman BT. Neuropathological alterations in Alzheimer disease. Cold Spring Harb Perspect Med 2011;1:a006189.
6. Bateman RJ, Xiong C, Benzinger TL, et al. Clinical and biomarker changes in dominantly inherited Alzheimer’s disease. N Engl J Med 2012;367:795–804.
7. Jack CR, Bennett DA, Blennow K, et al. NIA-AA Research Framework: toward a biological definition of Alzheimer’s disease. Alzheimers Dement 2018;14:535–62.
8. Dubois B, Feldman HH, Jacova C, et al. Revising the definition of Alzheimer’s disease: a new lexicon. Lancet Neurol 2010;9:1118–27.
9. U.S. Food and Drug Administration (FDA). Early Alzheimer’s disease: developing drugs for treatment guidance for industry 2018. (accessed August 11, 2020).
10. Insel PS, Weiner M, Mackin RS, et al. Determining clinically meaningful decline in preclinical Alzheimer disease. Neurology 2019;93:e322–33.
11. Vermunt L, Sikkes SA, van den Hout A, et al. Duration of preclinical, prodromal, and dementia stages of Alzheimer’s disease in relation to age, sex, and APOE genotype. Alzheimers Dement 2019;15:888–98.
12. Cho SH, Woo S, Kim C, et al. Disease progression modelling from preclinical Alzheimer’s disease (AD) to AD dementia. Sci Rep 2021;11:4168.
13. Knopman DS, Parisi JE, Salviati A, et al. Neuropathology of cognitively normal elderly. J Neuropathol Exp Neurol 2003;62:1087–95.
14. Bennett DA, Schneider JA, Arvanitakis Z, et al. Neuropathology of older persons without cognitive impairment from two community-based studies. Neurology 2006;66:1837–44.
15. Kazim SF, Iqbal K. Neurotrophic factor small-molecule mimetics mediated neuroregeneration and synaptic repair: emerging therapeutic modality for Alzheimer’s disease. Mol Neurodegener 2016;11:50.
16. Tolbert S, Liu Y, Hellegers C, et al. Financial management skills in aging, MCI and dementia: cross sectional relationship to 18F-florbetapir PET cortical β-amyloid deposition. J Prev Alzheimers Dis 2019;6:274–82.
17. Ye BS, Kim HJ, Kim YJ, et al. Longitudinal outcomes of amyloid positive versus negative amnestic mild cognitive impairments: a three-year longitudinal study. Sci Rep 2018;8:5557.
18. Roberts RO, Aakre JA, Kremers WK, et al. Prevalence and outcomes of amyloid positivity among persons without dementia in a longitudinal, population-based setting. JAMA Neurol 2018;75:970–9.
19. Dubois B, Padovani A, Scheltens P, Rossi A, Dell’Agnello G. Timely diagnosis for Alzheimer’s disease: a literature review on benefits and challenges. J Alzheimers Dis 2016;49:617–31.
20. Isaacson RS, Ganzer CA, Hristov H, et al. The clinical practice of risk reduction for Alzheimer’s disease: a precision medicine approach. Alzheimers Dement 2018;14:1663–73.
21. Isaacson RS, Hristov H, Saif N, et al. Individualized clinical management of patients at risk for Alzheimer’s dementia. Alzheimers Dement 2019;15:1588–602.
22. Gauthier SG. Alzheimer’s disease: the benefits of early treatment. Eur J Neurol 2005;12:11–6.
23. Cummings J, Lee G, Ritter A, Sabbagh M, Zhong K. Alzheimer’s disease drug development pipeline: 2020. Alzheimers Dement NY 2020;6:e12050.
24. Galvin JE, Aisen P, Langbaum JB, et al. Early stages of Alzheimer’s disease: evolving the care team for optimal patient management. Front Neurol 2021;11:592302.
25. Liu JL, Hlavka JP, Hillestad R, Mattke S. Assessing the preparedness of the U.S. Health Care System infrastructure for an Alzheimer’s treatment 2017. (accessed May 5, 2018).
26. Sabbagh MN, Lue L-F, Fayard D, Shi J. Increasing precision of clinical diagnosis of Alzheimer’s disease using a combined algorithm incorporating clinical and novel biomarker data. Neurol Ther 2017;6:83–95.
27. Balasa M, Gelpi E, Antonell A, et al. Clinical features and APOE genotype of pathologically proven early-onset Alzheimer disease. Neurology 2011;76:1720–5.
28. Galvin JE. Using informant and performance screening methods to detect mild cognitive impairment and dementia. Curr Geriatr Rep 2018;7:19–25.
29. Sabbagh MN, Boada M, Borson S, et al. Early detection of mild cognitive impairment (MCI) in primary care. J Prev Alzheimers Dis 2020;7:165–70.
30. Galvin JE, Sadowsky CH, NINCDS-ADRDA. Practical guidelines for the recognition and diagnosis of dementia. J Am Board Fam Med 2012;25:367–82.
31. Aisen PS, Cummings J, Jack CR, et al. On the path to 2025: understanding the Alzheimer’s disease continuum. Alzheimers Res Ther 2017;9:60.
32. Blennow K, Zetterberg H. Biomarkers for Alzheimer disease – current status and prospects for the future. J Intern Med 2018;284:643–63.
33. Karikari TK, Pascoal TA, Ashton NJ, et al. Blood phosphorylated tau 181 as a biomarker for Alzheimer’s disease: a diagnostic performance and prediction modelling study using data from four prospective cohorts. Lancet Neurol 2020;19:422–33.
34. Janelidze S, Mattsson N, Palmqvist S, et al. Plasma P-tau181 in Alzheimer’s disease: relationship to other biomarkers, differential diagnosis, neuropathology and longitudinal progression to Alzheimer’s dementia. Nat Med 2020;26:379–86.
35. Iliffe S, Robinson L, Brayne C, et al. Primary care and dementia: 1. diagnosis, screening and disclosure. Int J Geriatr Psychiatry 2009;24:895–901.
36. Arvanitakis Z, Shah RC, Bennett DA. Diagnosis and management of dementia: review. JAMA 2019;322:1589–99.
37. Robinson L, Tang E, Taylor J-P. Dementia: timely diagnosis and early intervention. BMJ 2015;350:h3029.
38. Zucchella C, Bartolo M, Pasotti C, Chiapella L, Sinforiani E. Caregiver burden and coping in early-stage Alzheimer disease. Alzheimer Dis Assoc Disord 2012;26:55–60.
39. Pfistermeister B, Tümena T, Gaßmann K-G, Maas R, Fromm MF. Anticholinergic burden and cognitive function in a large German cohort of hospitalized geriatric patients. PLoS One 2017;12:e0171353.
40. Liu C-C, Liu C-C, Kanekiyo T, Xu H, Bu G. Apolipoprotein E and Alzheimer disease: risk, mechanisms, and therapy. Nat Rev Neurol 2013;9:106–18.
41. Sadigh-Eteghad S, Sabermarouf B, Majdi A, Talebi M, Farhoudi M, Mahmoudi J. Amyloid-beta: a crucial factor in Alzheimer’s disease. Med Princ Pract 2015;24:1–10.
42. Karch CM, Goate AM. Alzheimer’s disease risk genes and mechanisms of disease pathogenesis. Biol Psychiatry 2015;77:43–51.
43. Ungar L, Altmann A, Greicius MD. Apolipoprotein E, gender, and Alzheimer’s disease: an overlooked, but potent and promising interaction. Brain Imaging Behav 2014;8:262–73.
44. 23andMe. What to know about health report test result – 23andMe. 23&Me® 2020. (accessed October 15, 2020).
45. Hendry K, Green C, McShane R, et al. AD-8 for detection of dementia across a variety of healthcare settings. Cochrane Database Syst Rev 2019;3:CD011121.
46. Patnode CD, Perdue LA, Rossom RC, et al. Screening for cognitive impairment in older adults: an evidence update for the U.S. Preventive Services Task Force. Rockville (MD): Agency for Healthcare Research and Quality (US); Report No.: 19-05257-EF-1, 2020.
47. Teng E, Becker BW, Woo E, Cummings JL, Lu PH. Subtle deficits in instrumental activities of daily living in subtypes of mild cognitive impairment. Dement Geriatr Cogn Disord 2010;30:189–97.
48. Marshall GA, Amariglio RE, Sperling RA, Rentz DM. Activities of daily living: where do they fit in the diagnosis of Alzheimer’s disease? Neurodegener Dis Manag 2012;2:483–91.
49. Rosenberg PB, Mielke MM, Appleby BS, Oh ES, Geda YE, Lyketsos CG. The association of neuropsychiatric symptoms in MCI with incident dementia and Alzheimer disease. Am J Geriatr Psychiatry 2013;21:685–95.
50. Ismail Z, Emeremni CA, Houck PR, et al. A comparison of the E-BEHAVE-AD, NBRS and NPI in quantifying clinical improvement in the treatment of agitation and psychosis associated with dementia. Am J Geriatr Psychiatry 2013;21:78–87.
51. Bowden VM, Bowden CL. The Journal of Neuropsychiatry and Clinical Neurosciences. JAMA 1992;268:1473–4.
52. Galvin JE. The quick dementia rating system (QDRS): a rapid dementia staging tool. Alzheimers Dement Amst 2015;1:249–59.
53. Koster N, Knol DL, Uitdehaag BM, Scheltens P, Sikkes SA. The sensitivity to change over time of the Amsterdam IADL Questionnaire(©). Alzheimers Dement 2015;11:1231–40.
54. Sikkes SA, Pijnenburg YA, Knol DL, de Lange-de Klerk ES, Scheltens P, Uitdehaag BM. Assessment of instrumental activities of daily living in dementia: diagnostic value of the Amsterdam Instrumental Activities of Daily Living Questionnaire. J Geriatr Psychiatry Neurol 2013;26:244–50.
55. LaRue RH. Functional Assessment Screening Tool (FAST). In: Volkmar FR, editor. Encycl. Autism Spectr. Disord., New York, NY: Springer New York, 2018:1–2.
56. Lindgren N, Rinne JO, Palviainen T, Kaprio J, Vuoksimaa E. Prevalence and correlates of dementia and mild cognitive impairment classified with different versions of the modified Telephone Interview for Cognitive Status (TICS-m). Int J Geriatr Psychiatry 2019;34:1883–91.
57. Wang H, Fan Z, Shi C, et al. Consensus statement on the neurocognitive outcomes for early detection of mild cognitive impairment and Alzheimer dementia from the Chinese Neuropsychological Normative (CN-NORM) Project. J Glob Health 2019;9:020320.
58. Franzen S, van den Berg E, Goudsmit M, et al. A systematic review of neuropsychological tests for the assessment of dementia in non-western, low-educated or illiterate populations. J Int Neuropsychol Soc 2020;26:331–51.
59. Costa A, Bak T, Caffarra P, et al. The need for harmonisation and innovation of neuropsychological assessment in neurodegenerative dementias in Europe: consensus document of the Joint Program for Neurodegenerative Diseases Working Group. Alzheimers Res Ther 2017;9:27.
60. MoCA Test Inc. Upcoming mandatory training for MoCA testing. MoCA Montr – Cogn Assess 2021. (accessed February 3, 2021).
61. Tabira T, Hotta M, Murata M, et al. Age-related changes in instrumental and basic activities of daily living impairment in older adults with very mild Alzheimer’s disease. Dement Geriatr Cogn Disord Extra 2020;10:27–37.
62. Martyr A, Nelis SM, Quinn C, et al. The relationship between perceived functional difficulties and the ability to live well with mild-to-moderate dementia: findings from the IDEAL programme. Int J Geriatr Psychiatry 2019;34:1251–61.
63. Ismail Z, Smith EE, Geda Y, et al. Neuropsychiatric symptoms as early manifestations of emergent dementia: provisional diagnostic criteria for mild behavioral impairment. Alzheimers Dement 2016;12:195–202.
64. McAllister-Williams RH, Bones K, Goodwin GM, et al. Analysing UK clinicians’ understanding of cognitive symptoms in major depression: a survey of primary care physicians and psychiatrists. J Affect Disord 2017;207:346–52.
65. Richard E, Schmand B, Eikelenboom P, Yang SC, Ligthart SA. Symptoms of apathy are associated with progression from mild cognitive impairment to Alzheimers disease in non-depressed subjects. Dement Geriatr Cogn Disord 2012;33:204–9.
66. Weintraub S, Besser L, Dodge HH, et al. Version 3 of the Alzheimer Disease Centers’ Neuropsychological Test Battery in the Uniform Data Set (UDS). Alzheimer Dis Assoc Disord 2018;32:10–7.
67. Harrison JE, Hendrix S. Chapter 21 – The assessment of cognition in translational medicine: a contrast between the approaches used in Alzheimer’s disease and major depressive disorder. In: Nomikos GG, Feltner DE, editors. Handb. Behav. Neurosci., vol. 29, Elsevier, 2019:297–308.
68. Roberts R, Knopman DS. Classification and epidemiology of MCI. Clin Geriatr Med 2013;29:753–72.
69. Harper L, Barkhof F, Scheltens P, Schott JM, Fox NC. An algorithmic approach to structural imaging in dementia. J Neurol Neurosurg Psychiatry 2014;85:692–8.
70. Frisoni GB, Boccardi M, Barkhof F, et al. Strategic roadmap for an early diagnosis of Alzheimer’s disease based on biomarkers. Lancet Neurol 2017;16:661–76.
71. Dubois B, Feldman HH, Jacova C, et al. Advancing research diagnostic criteria for Alzheimer’s disease: the IWG-2 criteria. Lancet Neurol 2014;13:614–29.
72. Johnson KA, Fox NC, Sperling RA, Klunk WE. Brain imaging in Alzheimer disease. Cold Spring Harb Perspect Med 2012;2:a006213.
73. Dubois B, Hampel H, Feldman HH, et al. Preclinical Alzheimer’s disease: definition, natural history, and diagnostic criteria. Alzheimers Dement 2016;12:292–323.
74. Villemagne VL, Doré V, Burnham SC, Masters CL, Rowe CC. Imaging tau and amyloid-β proteinopathies in Alzheimer disease and other conditions. Nat Rev Neurol 2018;14:225–36.
75. Clark CM, Pontecorvo MJ, Beach TG, et al. Cerebral PET with florbetapir compared with neuropathology at autopsy for detection of neuritic amyloid-β plaques: a prospective cohort study. Lancet Neurol 2012;11:669–78.
76. Wong DF, Rosenberg PB, Zhou Y, et al. In vivo imaging of amyloid deposition in Alzheimer disease using the radioligand 18F-AV-45 (florbetapir F 18). J Nucl Med 2010;51:913–20.
77. Duits FH, Martinez-Lage P, Paquet C, et al. Performance and complications of lumbar puncture in memory clinics: results of the multicenter lumbar puncture feasibility study. Alzheimers Dement 2016;12:154–63.
78. Shaw LM, Arias J, Blennow K, et al. Appropriate use criteria for lumbar puncture and cerebrospinal fluid testing in the diagnosis of Alzheimer’s disease. Alzheimers Dement 2018;14:1505–21.
79. Hansson O, Seibyl J, Stomrud E, et al. CSF biomarkers of Alzheimer’s disease concord with amyloid-β PET and predict clinical progression: a study of fully automated immunoassays in BioFINDER and ADNI cohorts. Alzheimers Dement 2018;14:1470–81.
80. Blennow K, Dubois B, Fagan AM, Lewczuk P, de Leon MJ, Hampel H. Clinical utility of cerebrospinal fluid biomarkers in the diagnosis of early Alzheimer’s disease. Alzheimers Dement 2015;11:58–69.
81. Hansson O, Lehmann S, Otto M, Zetterberg H, Lewczuk P. Advantages and disadvantages of the use of the CSF Amyloid β (Aβ) 42/40 ratio in the diagnosis of Alzheimer’s Disease. Alzheimers Res Ther 2019;11:34.
82. Lee SA, Sposato LA, Hachinski V, Cipriano LE. Cost-effectiveness of cerebrospinal biomarkers for the diagnosis of Alzheimer’s disease. Alzheimers Res Ther 2017;9:18.
83. Dolgin E. Alzheimer’s disease is getting easier to spot. Nature 2018;559:S10–2.
84. Morris E, Chalkidou A, Hammers A, Peacock J, Summers J, Keevil S. Diagnostic accuracy of 18F amyloid PET tracers for the diagnosis of Alzheimer’s disease: a systematic review and meta-analysis. Eur J Nucl Med Mol Imaging 2016;43:374–85.
85. Engelborghs S, Niemantsverdriet E, Struyfs H, et al. Consensus guidelines for lumbar puncture in patients with neurological diseases. Alzheimers Dement Amst 2017;8:111–26.
86. Barthélemy NR, Bateman RJ, Hirtz C, et al. Cerebrospinal fluid phospho-tau T217 outperforms T181 as a biomarker for the differential diagnosis of Alzheimer’s disease and PET amyloid-positive patient identification. Alzheimers Res Ther 2020;12:26.
87. O’Bryant SE, Mielke MM, Rissman RA, et al. Blood-based biomarkers in Alzheimer disease: current state of the science and a novel collaborative paradigm for advancing from discovery to clinic. Alzheimers Dement 2017;13:45–58.
88. C2N Diagnostics. Press release. Alzheimer’s breakthrough: C₂N first to offer a widely accessible blood test. C2N Diagn 2021. (accessed January 25, 2021).
89. Barthélemy NR, Horie K, Sato C, Bateman RJ. Blood plasma phosphorylated-tau isoforms track CNS change in Alzheimer’s disease. J Exp Med 2020;217:e20200861.
90. Janelidze S, Stomrud E, Smith R, et al. Cerebrospinal fluid p-tau217 performs better than p-tau181 as a biomarker of Alzheimer’s disease. Nat Commun 2020;11:1683.
91. Chen J, Duan Y, Li H, Lu L, Liu J, Tang C. Different durations of cognitive stimulation therapy for Alzheimer’s disease: a systematic review and meta-analysis. Clin Interv Aging 2019;14:1243–54.
92. Cummings J, Fox N. Defining disease modifying therapy for Alzheimer’s disease. J Prev Alzheimers Dis 2017;4:109–15.
93. Birks JS, Harvey RJ. Donepezil for dementia due to Alzheimer’s disease. Cochrane Database Syst Rev 2018;6:CD001190.
94. Birks JS, Chong LY, Grimley Evans J. Rivastigmine for Alzheimer’s disease. Cochrane Database Syst Rev 2015;9:CD001191.
95. Tariot PN, Farlow MR, Grossberg GT, et al. Memantine treatment in patients with moderate to severe Alzheimer disease already receiving donepezil: a randomized controlled trial. JAMA 2004;291:317–24.
96. Cummings J. New approaches to symptomatic treatments for Alzheimer’s disease. Mol Neurodegener 2021;16:2.
97. Loy C, Schneider L. Galantamine for Alzheimer’s disease and mild cognitive impairment. Cochrane Database Syst Rev 2006;1:CD001747.
98. Atri A. The Alzheimer’s Disease Clinical Spectrum. Med Clin N Am 2019;103:263–93.
99. Berlowitz DR, Foy CG, Kazis LE, et al. Effect of intensive blood-pressure treatment on patient-reported outcomes. N Engl J Med 2017;377:733–44.
100. Porsteinsson AP, Drye LT, Pollock BG, et al. Effect of citalopram on agitation in Alzheimer’s disease – the citAD randomized controlled trial. JAMA 2014;311:682–91.
101. Sheline YI, Snider BJ, Beer JC, et al. Effect of escitalopram dose and treatment duration on CSF Aβ levels in healthy older adults: a controlled clinical trial. Neurology 2020;95:e2658–65.
102. Sheehan B. Assessment scales in dementia. Ther Adv Neurol Disord 2012;5:349–58.
103. Haubois G, Annweiler C, Launay C, et al. Development of a short form of Mini-Mental State Examination for the screening of dementia in older adults with a memory complaint: a case control study. BMC Geriatr 2011;11:59.
104. Horton DK, Hynan LS, Lacritz LH, Rossetti HC, Weiner MF, Cullum CM. An Abbreviated Montreal Cognitive Assessment (MoCA) for dementia screening. Clin Neuropsychol 2015;29:413–25.
105. Mini-Cog©. Mini-Cog© In other languages. Mini-Cog© Lang 2021. (accessed April 21, 2021).
106. Carnero Pardo C, de la Vega Cotarelo R, López Alcalde S, et al. Assessing the diagnostic accuracy (DA) of the Spanish version of the informant-based AD8 questionnaire. Neurologia 2013;28:88–94.
107. Harrison JK, Fearon P, Noel-Storr AH, McShane R, Stott DJ, Quinn TJ. Informant questionnaire on cognitive decline in the elderly (IQCODE) for the diagnosis of dementia within a secondary care setting. Cochrane Database Syst Rev 2015;3:CD010772.
108. Sanchez MA, Correa PC, Lourenço RA. Cross-cultural adaptation of the “Functional Activities Questionnaire – FAQ” for use in Brazil. Dement Neuropsychol 2011;5:322–7.
109. Kim G, DeCoster J, Huang C-H, Bryant AN. A meta-analysis of the factor structure of the Geriatric Depression Scale (GDS): the effects of language. Int Psychogeriatr 2013;25:71–81.
110. Mapi Research Trust. NPI – Officially distributed by Mapi Research Trust. Neuropsychiatr Inventory Quest NPI-Q 2021. (accessed April 21, 2021).
111. Mapi Research Trust. A-IADL-Q-SV – Amsterdam Instrumental Activity of Daily Living Questionnaire – Short version. Amst Instrum Act Dly Living Quest – Short Version -IADL-Q-SV n.d. (accessed April 21, 2021).
112. Ismail Z, Agüera-Ortiz L, Brodaty H, et al. The Mild Behavioral Impairment Checklist (MBI-C): a rating scale for neuropsychiatric symptoms in pre-dementia populations. J Alzheimers Dis 2017;56:929–38.



T.P. Ng1,2, T.S. Lee3, W.S. Lim4, M.S. Chong2, P. Yap5, C.Y. Cheong5, K.B. Yap6, I. Rawtaer7, T.M. Liew8, Q. Gao1, X. Gwee1, M.P.E. Ng2, S.O. Nicholas2, S.L. Wee2,9


1. Gerontological Research Programme, Department of Psychological Medicine, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 2. Geriatric Education and Research Institute, Singapore; 3. Duke-NUS Medical School, Singapore; 4. Department of Geriatric Medicine, Tan Tock Seng Hospital, Singapore; 5. Department of Geriatric Medicine, Khoo Teck Puat Hospital, Singapore; 6. Department of Geriatric Medicine, Ng Teng Fong General Hospital, Singapore; 7. Department of Psychiatry, Sengkang General Hospital, Singapore; 8. Department of Psychiatry, Singapore General Hospital, Singapore; 9. Health and Social Sciences Cluster, Singapore Institute of Technology, Singapore

Corresponding Author: A/P Tze Pin Ng, Department of Psychological Medicine, Yong Loo Lin School of Medicine, National University of Singapore, Level 9, NUHS Tower Block, 1E Kent Ridge Road, Singapore 119228, Fax: 65-67772191, Email:

J Prev Alz Dis 2021;3(8):335-344
Published online April 26, 2021,



Background: Mild cognitive impairment (MCI) is a critical pre-dementia target for preventive interventions. There are few brief screening tools based on self-reported personal lifestyle and health-related information for predicting MCI that have been validated for their generalizability and utility in primary care and community settings.
Objective: To develop and validate a MCI risk prediction index, and evaluate its field application in a pilot community intervention trial project.
Design: Two independent population-based cohorts in the Singapore Longitudinal Ageing Study (SLAS). We used SLAS1 as a development cohort to construct the risk assessment instrument, and SLA2 as a validation cohort to verify its generalizability.
Setting: community-based screening and lifestyle intervention
Participants: (1) SLAS1 cognitively normal (CN) aged ≥55 years with average 3 years (N=1601); (2) SLAS2 cohort (N=3051) with average 4 years of follow up. (3) 437 participants in a pilot community intervention project.
Measurements: The risk index indicators included age, female sex, years of schooling, hearing loss, depression, life satisfaction, number of cardio-metabolic risk factors (wide waist circumference, pre-diabetes or diabetes, hypertension, dyslipidemia). Weighted summed scores predicted probabilities of MCI or dementia. A self-administered questionnaire field version of the risk index was deployed in the pilot community project and evaluated using pre-intervention baseline cognitive function of participants.
Results: Risk scores were associated with increasing probabilities of progression to MCI-or-dementia in the development cohort (AUC=0.73) and with increased prevalence and incidence of MCI-or-dementia in the validation cohort (AUC=0.74). The field questionnaire risk index identified high risk individuals with strong correlation with RBANS cognitive scores in the community program (p<0.001).
Conclusions: The SLAS risk index is accurate and replicable in predicting MCI, and is applicable in community interventions for dementia prevention.

Key words: Mild cognitive impairment, dementia, metabolic syndrome, diabetes, cardiovascular risk factors.



The number of people worldwide who are living with dementia is currently estimated at about 50 million, and doubles every 20 years (1). Currently, there is a lack of effective disease-modifying treatments for Alzheimer’s disease (AD) and dementia. Based on research evidence over the past decade of risk and protective factors for dementia, a population prevention approach currently offers a promising prospect for reducing the worldwide burden of dementia (2). A viable population-based strategy is the identification of individuals whose psychosocial, lifestyle and health characteristics put them at significant risks of developing dementia in its precursor and reversible stage of mild cognitive impairment (MCI). Such individuals represent population target for multi-domain lifestyle behavioral and health interventions. Accurate screening for risks of MCI and early dementia should also be validated for use at low cost in community and primary care settings. This is especially important in low- and middle-income countries where the majority of persons with dementia live.
There are a number of earlier reports of validated risk indexes that predict the onset of Alzheimer’s disease and dementia. However, few risk indexes have been evaluated for assessing the risk of the MCI precursor of dementia, using easily available variables in the clinical setting (3-9). The Mayo Clinic Study of Aging (MSCA) basic risk score comprises 12 general demographic and clinical variables (such as age, education, marital status, diabetes, hypertension, and body mass index [BMI 7 The Australian National University Alzheimer’s Disease Risk Index (ANU-ADRI), originally developed and validated for predicting AD and dementia, comprises up to 15 similar variables (including age, sex, education, social engagement, physical activity, cognitively stimulating activities, smoking, alcohol consumption, depression, traumatic brain injury BMI, diabetes, cholesterol, fish intake and pesticide exposure) (8). Although scores in both risk indexes were strongly associated with the progression from CN to MCI, their predictive ability was limited (with C-statistic of 0.60 for both). Not all the predictive variables in the ANU-ADRI are easily available even in the study cohort of cognitive normal persons. Both risk indexes have not been reported to be externally validated in an independent population sample. It is possible that the identified risk scores may perform differently outside of the original study population. Both risk indexes were developed and evaluated in Caucasian populations. It cannot be presumed that the risk indexes are generalizable to ethnically different populations. Furthermore, to date, there are no translational research reports of the deployment and evaluation of field versions of MCI risk prediction tools in pragmatic intervention trials in the community setting.
In this study, we used risk prediction model data from the Singapore Longitudinal Ageing Study (SLAS-1) population-based cohort to derive a risk index that predicted incident MCI or dementia from 3-5 (mean 4.5) years of follow up, using risk scores for age, sex, education, depression, life satisfaction, hearing loss, and the number of cardio-metabolic risk factors (waist circumference, pre-diabetes or diabetes, hypertension, triglycerides, and high-density lipoprotein levels). We evaluated the external validity of the SLAS risk index by assessing its predictive accuracy among participants in an independent validation cohort (SLAS-2). Finally, we designed a brief self-administered questionnaire field version of the risk index and evaluated its application outside of the SLAS study cohorts in a pilot community-based interventional trial. This was a National Innovation Challenge (NIC) community project which aimed to screen and enroll high-risk individuals for targeted multi-domain lifestyle interventions for dementia prevention.



Study design and participants

The Singapore Longitudinal Ageing Study (SLAS) is an ongoing population-based observational prospective cohort study of ageing and health transition among older adults, aged 55 and above in Singapore. The study was approved by the National University of Singapore Institutional Review Board. Written informed consent was obtained from all participants.
Two distinct population cohorts were recruited in separate waves of recruitment of community-dwelling older adults in geographically different locations. SLAS-1 cohort participants who were aged 55 years and above and residents in the South East region of Singapore (N=2804) were recruited in 2003-2004 and were followed up at two re-assessment visits approximately 3 years apart in 2005-2007 and 2007-2009 respectively. The participants in the SLAS-2 cohort (N=3246) were aged 55 and above and residents in the South West region who were recruited in 2009-2013, with follow up assessments 3 to 5 (mean 4.5) years apart conducted in 2013-2018. Trained nurses visited the participants at home to perform face-to face questionnaire interviews, and clinical and neurocognitive assessments and blood draws were performed in a local study site center. The background and methodology of the SLAS research are described in detail elsewhere (10, 11).

SLAS-1 development cohort study

We used data from the SLAS-1 development cohort to derive a prediction model and risk scores for incident MCI or dementia. The development cohort comprised cognitive normal participants (N=2611), after excluding small numbers of Malay and Indian participants, those with stroke, Parkinson’s disease, other brain disorders and injury (N=101), prevalent MCI (N=478) and dementia (N=42), and undetermined neurocognitive status (N=4) at baseline. During the follow-up period, 227 participants died and 296 were lost to follow up. The MCI risk prediction model was based on longitudinal data analysis of 1,610 individuals, who were followed up to incident MCI (N=149) or dementia (N=14).

SLAS-2 validation cohort study

We evaluated the external validity of the SLAS-1 risk model by applying the prediction algorithm on individual participants in the SLAS-2 validation cohort. We used the calculated risk scores to assess its accuracy in predicting prevalent and incident cases of MCI or dementia. Both prevalent and incident cognitive outcomes were evaluated because the practical application of the risk prediction tool should include the identification of prevalent cases of MCI and interventions to prevent MCI progression to dementia. At baseline, after excluding 219 participants with missing data on neurocognitive status, there were a total of 3051 participants with known neurocognitive status: cognitive normal (CN)=2700, MCI =265, dementia=86) and had complete data on risk factors. A total of 1323 cognitive normal participants followed up for an average of 4.5 years were re-assessed on their neurocognitive status (CN=1157, MCI=69, dementia=6), after excluding 45 participants with missing data.

Mild Cognitive Impairment (MCI) and Dementia

As described in detail in a previous publication (12), neurocognitive disorder (MCI and dementia) among study participants was ascertained from initial screening using MMSE and self or informant reports of subjective cognitive decline, using the IQCODE (13) followed by subsequent assessment using Clinical Dementia Rating Scale (14) and neuropsychological evaluation for cognitive domains: memory (RAVLT immediate recall, RAVLT delayed recall, Visual Reproduction immediate recall, Visual Reproduction delayed); executive function (Symbol Digit Modality Test; Design Fluency; Trail Making Test B, language (Categorical Verbal Fluency); visuospatial skills (Block Design); attention (Digit Span Forwards, Digit Span Backwards, Spatial Span Forwards and Backwards), before final clinical assessments including MRI and additional blood tests and consensus diagnosis by a panel of geriatricians and psychiatrists.
MCI was defined according to published criteria (15, 16): cognitive concern expressed by the participant or informant; cognitive impairment in one or more domains (executive function, memory, language, or visuospatial); normal functional activities; and absence of dementia, and operationalized as follows:
1) Subjective cognitive complaints from a single question asking whether the subject had more problems with memory than most, or an informant report of memory decline, “Do you think your family member’s memory or other mental abilities have declined?” or IQCODE ≥3.0.
2) Cognitive deficits defined as Modified Chinese MMSE score of 24-27, or decline of 2 more points from baseline, a neurocognitive domain score that was 1 to 2 SD below the age-and-education adjusted means, or decline from baseline of 0.5 SD from serial measurements;
3) No functional dependency in performing instrumental daily living activities (Lawton et al);
4) Clinical Dementia Rating Scale score of 0 or 0.5;
5) No dementia.

Dementia was diagnosed based on the Diagnostic and Statistical Manual of Mental Disorders, DSM-4R (American Psychiatric Association, 1994) (17), with evidence of cognitive deficit from objective assessment (MMSE ≤23 or neuropsychological domain scores below 2 SD of age-education-adjusted mean), and evidence of social or occupational function impairment (dependency in one or more Instrumental ADL or Clinical Dementia Rating score ≥1).
Participants who did not meet these criteria for MCI or dementia were classified as cognitive normal.

Predictor variables

We developed the risk prediction model from stepwise selection of a parsimonious set of significant predictor variables from among 20 known variables that represent known psychosocial, lifestyle and health risk factors for the onset and development of MCI and dementia.
1. Age, gender, education (none, 6 or less years, and more than 6 years), marital status (single, widowed, divorced), APOE- ε4 genotype carrier status;
2. Physical, social or productive activities based on the number and frequencies on a 3-point Likert scale (0=never or less than once a month; 1=sometimes (once a month or more but less than once a week); 2=often (at least once a week) of usual participation in 18 different categories of physical, social and productive activities; level of physical, social or productive activities was scored as the sum of frequencies across all activities, with higher score denoting higher level of participation; hearing loss (self-reported or whisper test);
3. Psychosocial variables: living alone, loneliness, life satisfaction as reported in a previous publication, depression (Geriatric Depression Scale ≥5) (18);
4. Smoking (current or past smoking versus never smoking); alcohol consumption (one or more drinks daily);
5. Cardio-metabolic and vascular risk factors: body mass index (BMI)≥27.5kg/m2, waist circumference ≥90cm for male and ≥80cm for female; pre-diabetes and diabetes (raised fasting plasma glucose (FPG) (FPG ≥ 100 mg/dL (5.6 mmol/L), or on treatment for previously diagnosed type 2 diabetes); raised triglyceride (TG) level ( ≥ 150 mg/dL (1.7 mmol/L) or on anti-lipid treatment); reduced high density lipoprotein (HDL) cholesterol (< 40 mg/dL (1.03 mmol/L) in males and < 50mg/dL (1.29 mmol/L) in females or on anti-lipids treatment; hypertension (raised blood pressure(BP) (systolic BP ≥ 130 or diastolic BP ≥ 85 mm Hg, or on treatment of previously diagnosed hypertension).

Field version questionnaire and validation

We created a field version of the risk prediction tool by designing a simple self-administered checklist questionnaire of the 11 risk factor items that respondents endorse, and derive a summary risk score for each individual. (Table 1 and Figure 1) We deployed the risk screening questionnaire in a community-based National Innovation Challenge trial project in which individuals with high risk scores were identified and enrolled for multi-domain lifestyle interventions for dementia prevention. A total of 437 community-living older persons aged ≥55 years without ADL disability, visual impairment, or known neurodegenerative disorders (dementia, Parkinson’s and other diseases) were administered the risk screening questionnaire at neighourhood senior activity centres. A total of 194 participants who scored ≥6 points on the risk index were enrolled into the 6-month multi-domain intervention or control programmes. At the same time, they were assessed at baseline (pre-intervention) with the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) (19, 20). Raw scores were standardized to T scores with a mean of 100 and standard deviation of 15 in the global index score, with higher scores indicating better performance.

Table 1. SLAS Index of Risk Prediction for Pre-dementia and Dementia

Figure 1. Predicted risk of MCI/Dementia and risk prediction score


Statistical Analyses

In the SLAS-1 developmental cohort, we examined a total of 20 available demographic, psychosocial, lifestyle behavior and health risk variables known from the literature review or demonstrated in our previous published reports to be risk or protective factors for the risk of MCI or dementia. We modeled the cumulative incidence (rather than time-to-event) of new cases of MCI or dementia that develop within the 4.5-year period in logistic regression as it was considered to be more informative and relevant for the purpose of predicting the risk or probability of MCI, and it would be more interpretable for lay users of the information. We used stepwise procedures to select a parsimonious set of 7 surrogate variables in the final logistic model that predict incident MCI or dementia. The outcome variable combined MCI and dementia since the follow up of these cognitive normal participants identified incident cases of MCI as well as a small number of dementia cases (N=14, diagnosed without known progression from MCI). Results are presented using complete case analysis as they were similar to results using imputation methods for missing data. Estimated regression coefficients in the logistic model were used as weights to derive simple point scores (from 0 to 3) associated with each risk indicator, and a summed risk score. In the SLAS-2 validation cohort, we used the same prediction algorithm to calculate the summary risk score for each participant and estimated the risks of prevalent and incident MCI-dementia associated with each risk score category. The predictive accuracy of the risk index was assessed by the area under the curve (AUC) using receiving operating curve techniques. In the NIC community –based interventional trial, the risk scores (ranging from 6 to 11) of 194 participants were evaluated on the trend of association with mean RBANS T-score of global cognitive performance using ANOVA with significance tests for linear trend.



SLAS1 development cohort

The baseline characteristics and risk factors of cognitively normal participants who presented with incident MCI or dementia (N=163) at follow up and their counterparts (N=1438) are shown in Table 2. Older age, female sex, low education, hearing loss, depression and low life satisfaction were significant predictors of incident MCI-dementia. Diabetes was the only cardio-metabolic risk factor among the metabolic syndrome components that was significantly associated with MCI-dementia, but the number of metabolic syndrome components was more strongly predictive of MCI-dementia.

Table 2. Baseline characteristics and risk factors of SLAS-1 development cohort of cognitive normal participants by MCI-dementia follow-up status


Table 3 shows the estimates of the strength of association of individual risk factors with incident MCI-dementia in the final risk prediction model, and associated weighted risk scores. The summed risk scores for individual participants ranged from 0 to 11, and was associated respectively with cumulative incidence of MCI-dementia ranging from 0% to 50%, (Figure 2) or 45% increased odds of incident MCI-dementia per point score increase. The AUC was 0.73, 95% confidence interval (95%CI): 0.70-0.75. Using an optimum risk score of ≥6, the sensitivity was 0.748 (95%CI: 0.677-080), specificity was 0.672 (0.647-0.695) and positive predictive value was 0.169 (95%CI: 0.144-0.199), based on the observed cumulative incidence of MCI-dementia of 10.2%
SLAS-2 validation cohort

Figure 2. Predicted cumulative probabilities of incident MCI-dementia by risk index score in SLAS-1 developmental cohort

Table 3. Estimates of association of SLAS risk factors with and risk score predictors of MCI-dementia and ROC measure of discriminant accuracy


The baseline characteristic and risk profile of participants overall and among cognitive normal participants who were followed up are shown in Table 4. In cross-sectional cohort, the risk index ranged from 0 to 12 (mean of 5.7 and SD of 2.4) and were associated with increasing prevalence of MCI-dementia ranging from 0% to 45% (Figure 3A). The AUC was 0.74 (95%CI: 0.72-0.77), sensitivity was 0.755 (95%CI: 0.707-0.797), specificity was 0.675 (95%CI: 0.576-0.613) and PPV was 0.209, 95%CI: 0.175-0.217) based on the overall prevalence of 11.5%.
In the follow up cohort of cognitive normal participants, risk scores from 0 to 12 were associated with increasing cumulative incidence of MCI-dementia from 0% to 36%. (Figure 3B) The AUC was 0.74 (95%CI: 0.69-0.79); sensitivity=0.720 (95%CI: 0.609-0.809), specificity=0.672 (95%CI: 0.644-0.698), PPV=0.124 (95%CI: 0.097-0.159), based on the observed overall cumulative incidence of 6.1%.

Figure 3a. Predicted probabilities of prevalent MCI-dementia by risk index score in SLAS-2 validation cohort


Figure 3b. Predicted cumulative probabilities of incident MCI-dementia by risk index score in SLAS-2 validation cohort

Table 4. SLAS-2 validation cohort baseline characteristics


Field version questionnaire risk score validation

The risk scores range in values from 6 to 11 (mean of 8.0 and SD of 1.1), and were significantly associated with decreasing trend of mean RBANS T-score from 105 to 95 (p for linear trend <0.001) (Figure 4).

Figure 4. RBANS T-scores by risk score among persons with risk scores of six and above who are selected for targeted intervention



We describe in this report the development of a risk index that predicts average 4.5-years risk of MCI or dementia in an Asian population. It comprises established personal, lifestyle and health factors that are easily measured and commonly used in primary care. The risk index showed good predictive accuracy both in the development cohort (AUC=0.73) and the external validation cohort (AUC=0.74). Equally important is that the scores predict a wide range of cumulative risk estimates of MCI-dementia with the highest risk score predicting a probability of 50% in the development cohort. In the validation cohort, the highest risk score predicted 45% probability of prevalent MCI-dementia and 36% probability of incident MCI-dementia. A lower prediction performance is usually demonstrated in external validation samples compared to original development samples (21, 22).
Several other risk prediction tools for MCI have been previously described. They include the Mayo Clinic Study of Aging (MCSA) basic risk score comprising age, education, marital status, diabetes, hypertension, body mass index (BMI), which has a AUC of 0.60 (7). The Australian National University-Alzheimer’s Disease Risk Index (ANU-ADRI) (8) is computed from 11 to 15 predictive variables, including self-reports of age, education, BMI, diabetes, depressive symp¬toms, high cholesterol, head trauma, smoking, alcohol consumption, social engagement (marital status, size of social network, quality of social network, level of social activities), physical activity (number of hours performing mild, moderate and vigorous activities), cognitively stimulating activities (number of cognitive activities undertaken), BMI, and (as available) cholesterol, fish intake and pesticide exposure. It also showed a AUC of 0.61 for incident MCI/dementia (5). In comparison, the SLAS risk index has higher predictive accuracy. While these studies have evaluated the predictive accuracy of their risk scoring tools within the development cohort, there are no reports of calibrating the performance of these tools in external validation cohorts. In our study, the external validity and generalizability of the SLAS risk prediction tool was demonstrated in an independent population sample (SLAS-2). This shows that the risk index is transportable to another population that differed in geographical location and other characteristics.

Furthermore, we re-designed and created a brief questionnaire version of the risk prediction index for field application in a community interventional trial. The questionnaire was self-administered or administered by a healthcare or community service provider within 2 minutes. In Singapore, basic health screening is easily accessible in primary care clinics at no or minimal costs for older residents, and include blood tests for fasting glucose and lipids, blood pressure and body size measurements. Given the advantage of ease of collection of self-reported data, there was no material loss of accuracy of information ascertainment, as shown by the RBANS cognition data. We demonstrated that the risk index questionnaire was able to discriminate individuals with varying levels of cognitive function. Using a pre-determined cutoff of 6 and above (associated with less than 10% predicted risk of MCI, or equivalent to the combined scores for non-modifiable risk factors of age, sex and education), the RBANS T-score stands at about 105 for the threshold risk scores of 6 and 7 and are close to local population norms (17), but dropped to 97 and 95 for higher score of 8 and above. The 10-point difference is more than the minimum clinically important difference of 8 for RBANS total score (18). Although the data empirically suggests a critical threshold risk score of 8 and above, practically a lower screening threshold of 6 or 7 is optimal. As yet, we found no other studies that have validated the field performance of a risk index for MCI in a pragmatic intervention trial in the community setting. Only a Portuguese questionnaire version of the ANU-ADRI was assessed in Brazil in a primary care study to be reliable and valid for assessing risk for AD (not MCI), by showing a moderate negative linear relation between the ANU-ADRI and MMSE scores (23). Our data thus show the feasibility of using the risk index to screen and identify individuals with prevalent MCI or at risk of MCI who are likely to benefit from early targeted multi-domain lifestyle intervention in a randomized controlled community trial which is ongoing.

Strengths and limitations

Our risk index for predicting MCI was based on risk factors from a single study. Other approaches of modelling with data from meta-analyses of risk factors from multiple studies may produce more predictive accuracy. A Shanghai risk model of 10 risk predictors was derived from a meta-analysis of 38 Chinese population studies and reported higher AUC from multivariate logistic regression model of 0.77. Notably, the predictive performance of the risk index was verified on a cross-sectional study predicting prevalent cases of MCI (6). Interestingly, the Shanghai study showed that better AUC of 0.86 was obtained using the same data using the Rothman-Keller model with additional model parameterizations. Further work should explore this new approach in developing risk prediction models of MCI and dementia.
Our risk index was derived from available data on 20 candidate risk and protective factors in a single population study. However, they are not much less wide and diverse in coverage as those reported in published meta-analyses. In general, risk prediction is improved with more independent predictors that contribute additionally to the model, but at greater costs of data collection. This may be done by adding information from cognitive, MRI or other testing, which may be desirable in clinical settings but does not suit the purpose of a risk index for population-based prevention interventions in the community. Our risk index was intentionally designed to be brief, inexpensive and non-invasive. It can be self-administered or administered by non-highly trained staff, hence based on a minimalist set of risk factors predicting MCI or dementia. It turns out to compare quite favorably or outperform relatively lengthy risk prediction instruments. It is possible that the weights derived for the SLAS risk index are study specific. Studies in other heterogeneous populations using the same risk prediction model may have optimal weights that differ for specific populations. More studies in ethnically similar or different populations elsewhere are needed. This risk prediction model appears to be successful in predicting risks of MCI-dementia in a Chinese population in Singapore. Whether it will be equally successful in ascertaining risks in Chinese populations in other places such as Taiwan or China, or to non-Chinese populations elsewhere needs to be verified independently with further validation studies.
The relatively short length of follow up of this cohort is notable, but is not a limitation. Risk prediction requires a specification of the time span, and in this case, a 3 to 5-year time span for predicting the risk of MCI or early dementia appears to be appropriate. The relatively young age of the study cohorts is also notable. A diversity of risk factors has their own age-developmental trajectories in predicting future risk of dementia, hence risk prediction instruments are of necessity applicable at different ages. Hypertension, obesity and cholesterol are known to be risk factors for late-life dementia when they are measured at middle age. The SLAS risk index is therefore optimal in identifying individuals at increased risk of dementia at a younger age and suits the primary purpose of early population-level interventions to prevent late-life dementia.



We developed and validated the SLAS risk index for predicting the risk of MCI or dementia over 4.5 years. We observed a high level of predictive accuracy which was replicated in an external validation population, and demonstrated the feasibility of applying a field version self-reported questionnaire of the risk index in a pilot community intervention project for dementia prevention.


Declarations: The study was approved by the National University of Singapore Institutional Review Board. Written informed consent was obtained from all participants.

Potential Conflict of Interest: The authors declare that they have no competing interests.

Availability of data: The datasets generated and analysed during the current study are not publicly available due to local data regulations and institutional policies but may be available from the corresponding author on reasonable request with permission from relevant authorities.

Funding: The study was supported by the Agency for Science Technology and Research (A*STAR) Biomedical Research Council (BMRC) [grant number 08/1/21/19/567] and the National Medical Research Council [grant number: NMRC/1108/2007].

Authors’ contributions: TPN reviewed the literature, designed the study, analyzed the data, drafted and revised the manuscript. SON provided additional data analysis. TSL, WSL, MSC, KBY, PY, CYC, IR, TML, QG, XYG, and NPEM contributed to the study design and data collection. All authors reviewed the results and drafts, and approved the final manuscript.

Acknowledgments: We thank the following voluntary welfare organizations for their support: Geylang East Home for the Aged, Presbyterian Community Services, St Luke’s Eldercare Services, Thye Hua Kwan Moral Society (Moral Neighbourhood Links), Yuhua Neighbourhood Link, Henderson Senior Citizens’ Home, NTUC Eldercare Co-op Ltd, Thong Kheng Seniors Activity Centre (Queenstown Centre) and Redhill Moral Seniors Activity Centre.

Ethical standards: The National University of Singapore (NUS) Institutional Review Board (IRB) approved the study, and all participants gave informed consent before participating.



1. Alzheimer’s Disease International. World Alzheimer Report 2010: The Global Economic Impact of Dementia. London: Alzheimer’s Disease International, 2010.
2. Norton S, Matthews FE, Barnes D, et al. Potential for primary prevention of Alzheimer’s disease: an analysis of population-based data. Lancet Neurol 2014;13:788–794.
3. Kivipelto M, Ngand T, Laatikainen T, et al. Risk score for the prediction of dementia risk in 20 years among middle aged people: a longitudinal, population-based study. Lancet Neurol 2006;5:735-741.
4. Reitz C, Tang MX, Schupf N, et al. A summary risk score for the prediction of Alzheimer disease in elderly persons. Arch Neurol 2010;67:835–841.
5. Anstey KJ, Cherbuin N, Herath PM, et al. A self-report risk index to predict occurrence of dementia in three independent cohorts of older adults: the ANU-ADRI. PLoS One 2014;DOI: 10.1371/journal.pone.0086141.
6. Vos SJB, van Boxtel MPJ, Schiepers OJG, et al. Modifiable Risk Factors for Prevention of Dementia in Midlife, Late Life and the Oldest-Old: Validation of the LIBRA Index. J Alzheimers Dis 2017;58:537-547.
7. Pankratz VS, Roberts RO, Mielke MM, et al. Predicting the risk of mild cognitive impairment in the Mayo Clinic Study of Aging. Neurology 2015;84:1433-1442.
8. ndrews SJ, Eramudugolla R, Velez JI, et al. Validating the role of the Australian National University Alzheimer’s Disease Risk Index (ANU-ADRI) and a genetic risk score in progression to cognitive impairment in a population-based cohort of older adults followed for 12 years. Alzheimers Res Ther 2017;9:16.
9. Wang B, Shen T, Mao L, et al. Establishment of a Risk Prediction Model for Mild Cognitive Impairment among Elderly Chinese. J Nutr Health Aging 2020;24:255-261.
10. Niti M, Yap KB, Kua EH, et al. Physical, social and productive leisure activities, cognitive decline and interaction with APOE-epsilon4 genotype in Chinese older adults. Int Psychogeriatr 2008;20:237-251.
11. Wei K, Nyunt MSZ, Gao Q, et al. Frailty and Malnutrition: Related and Distinct Syndrome Prevalence and Association among Community-Dwelling Older Adults: Singapore Longitudinal Ageing Studies. J Am Med Dir Assoc 2017;18:1019-1028.
12. Ng TP, Feng L, Nyunt MS, et al. Metabolic Syndrome and the Risk of Mild Cognitive Impairment and Progression to Dementia: Follow-up of the Singapore Longitudinal Ageing Study Cohort. JAMA Neurol 2016;73:456-463.
13. Jorm AF. A short form of the Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE): development and cross-validation. Psychol Med. 1994 Feb 24(1):145-53
14. Morris JC. The Clinical Dementia Rating (CDR): current version and scoring rules. Neurology. 1993 Nov;43(11):2412-4. doi: 10.1212/wnl.43.11.2412-a. PMID: 8232972.
15. Petersen RC. Mild cognitive impairment as a diagnostic entity. J Intern Med 2004; 256:183–194.
16. Winblad B, Palmer K, Kivipelto M, et al. Mild cognitive impairment: beyond controversies, towards a consensus: report of the International Working Group on Mild Cognitive Impairment. J Intern Med 2004;256:240–246.
17. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-IV. Washington, DC: American Psychiatric Association, 1994.
18. Nyunt MSZ, Jin AZ, Fones CSL, et al. Criterion-based validity and reliability of the Geriatric Depression Screening Scale (GDS-15) in a large validation sample of community-living Asian older adults. Aging Ment Health 2009;13:376-382.
19. Collinson SL, Fang SH, Lim ML, et al. Normative Data for the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) in Elderly Chinese. Arch Clin Neuropsychol 2014;29:442-455.
20. Phillips R, Qi G, Collinson SL, et al. The Minimum Clinically Important Difference in the Repeatable Battery for the Assessment of Neuropsychological Status. Clin Neuropsychol 2015;29:905-923.
21. Laupacis A, Sekar N, Stiell IG. Clinical prediction rules: A review and suggested modifications of methodological standards. JAMA 1997;277:488–494.
22. Schonberg MA, Davis RB, McCarthy EP, et al. External validation of an index to predict up to 9-year mortality of community-dwelling adults aged 65 and older. J Am Geriatr Soc 2011;59:1444-1451.
23. Borges MK, Jacinto AF, Citero VA. Validity and reliability of the Brazilian Portuguese version of the Australian National University – Alzheimer’s Disease Risk Index (ANU-ADRI). Dement Neuropsychol, 2018;12:235-243.




K. Pun1, C.W. Zhu1,2, M.T. Kinsella1, M. Sewell1, H. Grossman1,2, J. Neugroschl1, C. Li1, A. Ardolino1, N. Velasco1, M. Sano1,2


1. Icahn School of Medicine at Mount Sinai, New York, NY, USA; 2. James J. Peters VA Medical Center, Bronx, NY, USA

Corresponding Authors: Carolyn W. Zhu, PhD, Department of Geriatrics & Palliative Medicine, Icahn School of Medicine at Mount Sinai and JJP VA Medical Center, 130 West Kingsbridge Road, Bronx NY 10468, USA. Email:, Telephone: 718-584-9000 ext. 6146, Fax: 718-741-4211.

J Prev Alz Dis 2021;3(8):292-298
Published online April 24, 2021,



Objectives: This report describes the efficacy and utility of recruiting older individuals by mail to participate in research on cognitive health and aging using Electronic Health Records (EHR).
Methods: Individuals age 65 or older identified by EHR in the Mount Sinai Health System as likely to have Mild Cognitive Impairment (MCI) were sent a general recruitment letter (N=12,951). A comparison group of individuals with comparable age and matched for gender also received the letter (N=3,001).
Results: Of the 15,952 individuals who received the mailing, 953 (6.0%) responded. 215 (1.3%) declined further contact. Overall rate of expression of interest was 4.6%. Of the 738 individuals who responded positively to further contact, 321 indicated preference for further contact by telephone. Follow-up of these individuals yielded 30 enrollments (0.2% of 15,952). No differences in response rate were noted between MCI and comparison groups, but the comparison group yielded higher enrollment. 6 individuals who were not the intended recipients of mailing but nevertheless contacted our study were also enrolled.
Conclusions: Mailings to individuals identified through a trusted source, such as a medical center from which they have received clinical care, may be a viable means of reaching individuals within this age group as this effort yielded a low rejection rate. However, EHR information did not enhance study enrollment. Implications for improving recruitment are discussed.

Key words: Recruitment methods, electronic health records, cognitive health, mild cognitive impairment.



It is increasingly recognized that the pathophysiological process of Alzheimer’s disease (AD) begins years and maybe decades prior to the onset of clinical symptoms (1-7). Over the past several decades, both pharmacological and non-pharmacological lifestyle interventions have been studied for the prevention of cognitive decline or dementia in older adults with or without risk factors for AD. While important innovations in ongoing trials include identification of novel targets, development of multidomain interventions, identification and validation of biomarker or genetic targets, and improving outcome measures, the biggest challenge remains the recruitment of participants, espcially for long studies of non or minimally affected individuals (8-11).
One of the critical components for the success of these studies is identifying and recruiting individuals at high risk of developing dementia, both for observational studies investigating the natural history of prodromal and early disease stages and for interventional studies aimed at disease prevention or modification. Individuals with Mild Cognitive Impairment (MCI) have an increased risk of developing dementia compared with their cognitively normal peers (12). However, outreach to older adults for studies in these areas is often difficult because disease may be undetected in its mildest forms and awareness of future problems may be low. Studies have used diagnostic codes in administrative data or medical records to identify cases, however accuracy of diagnostic codes for cognitive impairment is limited (13-16). Identification of non-demented individuals with MCI from electronic health records (EHR) can be challenging, often depending on unstructured text for detection but several algorithms have been reported for such case ascertainment (17-19). Machine learning algorithms that have been developed to increase precision of identification have yet to be used in outreach efforts (20).
Campaigns to improve outreach and recruitment have often used mass mailings, defined as letters without a specific addressee, as they permit an inexpensive way to reach large numbers of potential participants and do not require technologies that may be less available to older cohorts. In general, rates of recruitment by this method can be low but the large numbers reached can allow studies to achieve needed sample sizes. For example, the Lifestyle Interventions and Independence for Elders (LIFE) Study reported that directly mailing a study brochure to households with age-eligible residents obtained from commercial databases and voter registration lists yielded 59.5% of randomized cases (21). However, the number of contacts needed to achieve this recruitment is quite high.
We made the assumption that trusted sources such as a medical center from which individuals have received clinical care may increase their interest in research participation and improve response to outreach efforts. The Alzheimer’s Disease Research Center (ADRC) at the Icahn School of Medicine at Mount Sinai has been continuously funded by the National Institutes of Health and has more than 30 years of experience that is highly recognized in the community. The ADRC offers ongoing opportunities to participate in a variety of clinical studies ranging from observational studies with and without imaging to pharmacological and non-pharmacological intervention studies. Building on that reputation, we undertook a focused mailing outreach effort using EHR, to engage individuals in research on cognitive health and aging. This report describes our experience and evaluates the efficacy of implementing a mass mailing within a trusted healthcare system to a cohort likely to have MCI based on their EHR.



Using EHR to Identify MCI and Comparison Groups

The EHR was used to create a group of individuals likely to have MCI. The MCI group included those age 55 to 90 who received care from the Mount Sinai Health System between spring 2013 and 2018. Inclusion criteria were: presence of an International Classification of Diseases, Ninth Revision (ICD-9) diagnosis code consistent with memory loss (MCI, memory loss, or dementia), or a similar Tenth Revision (ICD-10) diagnosis code of MCI or other amnesia. Exclusion criteria included presence of Alzheimer’s disease, Parkinson’s disease, schizophrenia, Huntington’s Chorea, epilepsy, multiple sclerosis, substance use disorders, tobacco use disorders, bipolar and major depressive disorders, and use of anti-dementia medications. Full inclusion/exclusion criteria are available upon request.
To evaluate whether the MCI group provided a recruitment benefit over that from an unselected comparison group, a comparison group was selected from individuals age 65 years or older, seen in the same health care system over the same time interval. They were further matched by sex to the MCI group. Those with a dementia diagnosis were excluded. The sample size of the comparison group was approximately 20% of the MCI group.
The ADRC staff were blinded to group status when contacting interested individuals, but group status was available for analysis at the end of recruitment efforts.

Mailing Process

A letter printed on institutional letterhead without specific salutation was sent to all individuals in the MCI and comparison groups. The letter 1) acknowledged the individual’s connection to the health system; 2) invited the individual to learn more about research in cognitive health and aging; 3) provided options to contact the ADRC via telephone, mail, or email; 4) informed the individual that they could “opt out” of further contact; and 5) informed the individual that the ADRC may reach out to them in the future if interest in further contact was expressed.
The letter included a returnable mailing slip on which individuals could confirm or deny interest in further contact by the ADRC and specify their preferred method of contact. A prepaid envelope to return the mailing slip was also included. Mailing materials are shown in Supplemental Materials Figure 1.

Response to Mailing and Expressions of Interest

Individuals who received the mailing were able to actively express interest in learning more about research programs at the ADRC either by 1) contacting the ADRC directly by phone; 2) contacting the ADRC directly by email; or 3) returning the included mailing slip, denoting interest for further contact, and specifying interest in further contact by telephone, email, or mail. A small number of individuals contacted us directly by email, all of whom also contacted us by telephone and expressed interest in further contact by telephone. Because the majority of responses across all contact methods preferred telephone, the current analysis describes our outreach efforts to those individuals regardless of initial response method. These included individuals who called the ADRC directly as well as those who emailed us directly or returned the mailing slip and expressed interest in further contact by telephone.

Recruitment Efforts

ADRC staff proceeded to call all individuals who expressed interest in further contact by telephone. Interested individuals were offered the opportunity to participate in observational and interventional studies available at the ADRC at the time of contact. ADRC staff explained study details and then offered to schedule initial screening appointments for all individuals who remained interested.

Estimating Staff Time and Effort

ADRC staff attempted to contact interested individuals by telephone using the following protocol: When individuals answered calls or responded to voice messages left by ADRC staff, calls continued until a decision regarding research participation was reached. If there was no response to a voice message within two weeks or it was not possible to leave a message, staff made up to two additional follow-up calls. A total of three call attempts were made to minimally or non-responsive individuals, with attempts made to vary the day and time of the call. Staff time and effort were estimated as follows: scheduling visits: 20 minutes, handling requests for more information or time: 15 minutes, determining ineligibility: 15 minutes, determining not interested: 10 minutes, and leaving a voice message: 5 minutes.



Response to Mailing and Expression of Interest

Of the 15,952 individuals who were sent the mailing, 114 contacted the ADRC by telephone directly and expressed interest, 839 sent returnable mailing slips, and the remaining 14,999 did not respond (Figure 1). Of the 839 who returned mailing slips, 624 (3.9% of all individuals who were sent the mailing) expressed interest in further contact, while 215 (1.3%) declined further contact. Among the 624 mailing slip respondents who expressed interest in further contact, 207 (1.3% of all individuals who were sent the mailing) indicated preference for further contact by telephone, 240 (1.5%) by email, and 177 (1.1%) by mail. Taken together, overall rate of expression of interest from the mailing was 4.6%.

Figure 1. Response to Mailing and Expression of Interest


This analysis focused on the 321 individuals who expressed interest in further contact by telephone (i.e., 2% of the entire mailing). Therefore, the 417 individuals who expressed interest in further contact through mail and email are not described in this report, and follow up on this cohort was left for future efforts.

Recruitment Outcomes

Of these 321 individuals who expressed interest for further contact from the ADRC by telephone, 30 (9.3%) enrolled in a study, 57 (17.8%) were not eligible due to medical comorbidities or study contraindications, 82 (25.6%) were not interested in research participation after speaking with ADRC staff, 137 (42.7%) did not definitively respond after at least three contact attempts, and 15 (4.7%) were not eligible for currently enrolling studies but remained interested in future participation (Table 1).

Table 1. Recruitment Outcomes for Individuals who Expressed Interest in Further Contact by Telephone (N=321)

Study Participation by MCI and Comparison Group

Among the 321 individuals who expressed interest in further contact by telephone, 227 (70.7%) individuals were from the MCI group (Table 2). This represents 1.8% of the 12,951 individuals from the MCI group. 53 of the 321 individuals (16.5%) were from the comparison group, representing 1.8% of the 3,001 individuals in the comparison group. An additional 41 (12.8%) individuals were incidental contacts who were not the intended recipients of the mailing but nevertheless contacted the ADRC. Group status for these individuals is by definition unknown.

Table 2. MCI and Comparison Group Status of Individuals who Expressed Interest and Individuals who Enrolled


Of the individuals who enrolled in studies, 15 of 30 were from the MCI group, representing a 0.1% enrollment rate from the total MCI group, while 9 of 30 were from the comparison group, representing a 0.3% enrollment rate from the total comparison group. A two-sample test of proportions shows that the difference in enrollment rate between MCI and comparison groups is statistically significant (p=0.019). An additional 6 of the 30 enrolled individuals were incidental contacts whose group status was unknown.

Enrollment by Study Type

Of the 30 individuals who enrolled, 7 (23.3%) enrolled in observational studies which did not include imaging, 21 (70.0%) enrolled in observational studies which included imaging, and 2 (6.7%) enrolled in an intervention trial (Table 3). Of the 7 individuals enrolled in observational studies without imaging, 5 (71.4%) were from the MCI group and 2 (28.6%) were from the comparison group. Among the 21 individuals who enrolled in observational studies with imaging, 9 (42.9%) were from the MCI group, 6 (28.6%) were from the comparison group, and 6 (28.6%) were from the incidental contact group. Finally, of the 2 individuals who enrolled in interventional trials, 1 (50.0%) was from the MCI group and 1 (50.0%) was from the comparison group. Detailed descriptions of these studies are included in Supplemental Materials Table 1.

Table 3. Enrollment by Study Type

Staff Time and Effort

ADRC staff logged 463 calls and spent an estimated total of 87 hours communicating with the 321 individuals who expressed interest in further contact with the ADRC by telephone (Table 4). Distribution of call logs and estimated time spent by recruitment outcome are also reported. On average, three hours (177 minutes) of staff time were required to enroll one participant. These estimates are limited to communication with individuals by telephone and offer a conservative summary of the total time and effort dedicated to this recruitment effort.

Table 4. Staff Time and Effort Required to Contact Individuals who Expressed Interest for Further Contact by Telephone



In this report we described our experience with a mailing outreach effort to engage individuals in research on cognitive health and aging. Individuals who had recent contact with the medical center were identified through the Mount Sinai Health System’s Electronic Health Record. Despite recent contact with the medical center and proximity to the site (over 95% of mailing addresses were within New York, New Jersey, and Connecticut), overall rate of expression of interest was approximately 5%, which is somewhat higher than comparable efforts of targeted mailing in similar age groups. For example, the Medical, Epidemiologic and Social Aspects of Aging (MESA) Study used a commercial mailing list to recruit women age 55-80 for a trial of behavioral techniques for the prevention of urinary incontinence that reached over 48,000 individuals and reported a 3.3% initial positive response rate (22). Efforts to recruit for dementia prevention trials using Medicare beneficiary lists and voter registration polls reported response rates between 0.4 to 2.0%.23 In the AD Anti-inflammatory Prevention Trial (ADAPT), over 3.5 million mailings were sent to Medicare beneficiaries 70 years or older in specified zip codes. The trial enrolled 2,518 volunteers at 6 sites over 44 months. Across trial sites, enrollment ranged from 0.4 to 1.9 participants per 1,000 mailings (24, 25). Unlike our study, this report included calls made to all individuals who received the mailing and did not opt out. The Ginkgo Evaluation of Memory (GEM) study to evaluate ginkgo biloba to prevent dementia mailed brochures to approximately 243,000 individuals age 75 and over who were assumed to be dementia free from purchased lists, voter registration lists, and university lists. Using telephone follow up to those who received the mailing, the study team attempted to reach 14,603 mail recipients, reached 12,186 and enrolled 1.3% of those who received the initial mailing (26).

Higher response rates have been reported in the literature when mailings came directly from primary care providers and when research activity was home-based and required no travel. For example, the Screening, Recruitment, and Baseline Characteristics for the Strategies to Reduce Injuries and Develop Confidence in Elders (STRIDE) Study for fall prevention found that approximately 12% of mailings resulted in expressions of interest (27). Higher response rates have been noted outside the US. Andersen et al mailed a questionnaire examining self-reported cognitive function to more than 11,807 community residents in Norway. 3,767 (31%) responded to the mailing. Of these, 438 met criteria for cognitive impairment and 292 were willing to undergo clinical evaluation (28). Notably, the cohort was invited to join a study for symptomatic individuals which may be of greater interest than studies of disease prevention.
In our study, about one third of the individuals who responded to the mailing contacted us by telephone. Of note, while 1.5% (240) used the returnable mailing slip and indicated preference for future contact by email, those who contacted the ADRC directly by email also contacted us by telephone, which may indicate less confidence in initiating email connections in this age group. Those who used the returnable mailing slip to provide telephone contact were nearly twice as likely to be lost to follow up relative to individuals who expressed interest by telephone directly. Future work might attempt to determine if requiring telephone follow-up would identify a more specifically interested group. Optimizing approaches to identifying sufficient numbers of interested individuals is critical to efficient and cost effective outreach.
Contrary to our expectations, we found greater interest and higher enrollment among those in the comparison group than in the MCI group. Of note, among those who enrolled, the MCI group was more likely to participate in observational studies and less likely to enroll in imaging or interventional studies. There are several possible reasons for this difference. Cognitive impairment and dementia in health records may be associated with other serious health problems (29). These health problems may be more prominent than cognitive impairment for these individuals and may reduce their interest in research on this topic. Meanwhile, the comparison group may be less likely to suffer from comorbid conditions and may be motivated by interest in prevention or protection against cognitive decline. Only 2 participants enrolled in clinical trials, highlighting challenges in recruitment to prodromal AD drug trials. Given these low frequencies, results from this study should be replicated. Understanding these themes will be important topics for future research.
While this mailing did not specifically invite non-recipients to join, the recruitment effort was prepared to accept them. We identified several individuals who described the mailing but were unsure of the source from which it came. This incidental interest is obviously a positive outcome for improving recruitment. Some mailing efforts encourage anyone to reply and this approach may be worthy of consideration. In particular, these contacts may be a source of “high interest” individuals, and efforts to identify features to improve outreach to them could provide an important contribution to recruitment science.
Several studies report the use of an opt-out approach to recruitment. In general those who opt-out prior to contact are low but often not reported. In our study, very few mailing recipients refused further contact (<2% of total number of mailings). Others have reported similarly low rates (23). The opt-out option can be executed in different ways including the “pre-mailing” requirement or a delay in outreach before initiating contact to those who received mail (23). These options require significant expense and time delay. In our study we reached out only to those who contacted us. Without an opt-out option, when unsolicited telephone calls followed the initial mailing, rate of no-interest is high, with reports of over 50% in several studies (24-26). The discordance between the low opt-out rate and the high no-interest rate reported suggest that opt-out resources may be effective in allowing studies to call more people. However, the resources and expenses needed to call those who do not demonstrate any initial interest are high. Future work may focus on identifying criteria for more efficient recruitment of individuals who do not opt out.
There are several limitations to fully appreciate this report in the context of recruitment to studies of cognitive health and aging. While there is growing use of algorithms to identify undetected dementia using EHR, little work has been done on identifying MCI (30, 31). Accuracy of diagnostic codes for dementia and cognitive impairment in medical records is limited (13-15). The study aimed to target a wide group of potential participants, however, it is possible that suitable participants were excluded if they were prescribed cholinesterase inhibitors for their MCI but did not have a formal diagnosis of MCI recorded in their medical records. Furthermore, our algorithm was based on EHR data collected across a five year period, and key patient information may not be the most up to date. We also did not have important variables such as time since diagnosis to examine their usefulness as potential predictors of response. We have little data on the accuracy of our mailing as “return to sender” information was not available to us. Recruitment efforts served multiple studies which may not have been open throughout the entire outreach period. However, the center had a variety of observational and interventional studies open at any time, and the option to be contacted for future studies was always provided. Ongoing efforts include study-specific mailing outreach to individuals who did not decline further contact, including those who expressed their preference for future contact by mail or email. Finally, while response to this mailing effort was slightly higher than expected, our experience is limited to a single site.



Focused mailing outreach efforts continue to represent a valuable means of engaging older adults in research on cognitive health and aging. However, using EHR to identify individuals who likely have cognitive impairment did not appear to increase response or participation rates. As demonstrated among the incidental contacts, a written letter has the potential to spark interest beyond its intended recipients. Mailings from known healthcare systems build upon an established relationship and may foster the development of a preliminarily defined, trial-ready cohort.


Acknowledgments and funding: Funding for this initiative was provided by the Alzheimer’s Disease Research Center (ADRC) at the Icahn School of Medicine at Mount Sinai (P50AG005138), the Alzheimer’s Therapeutic Research Institute (R01AG047992), and the National Center for Advancing Translation Sciences (UL1TR001433). Drs. Sano, Grossman, and Zhu also are supported by the Department of Veterans Affairs, Veterans Health Administration. The views expressed in this article are those of the authors and do not necessarily represent the views of the Department of Veterans Affairs. The sponsors had no role in the design and conduct of the study; in the collection, analysis, and interpretation of data; in the preparation of the manuscript; or in the review or approval of the manuscript.

Conflict of Interest: No conflict of interest has been declared by the authors.

Ethics standards: The study was approved by the Icahn School of Medicine Institutional Review Board (IRB). All participants gave informed consent before participating.





1. Schneider LS. Recruitment methods for United States Alzheimer disease prevention trials. J Nutr Health Aging. 2012;16(4):331-335.
2. Sperling R, Mormino E, Johnson K. The evolution of preclinical Alzheimer’s disease: implications for prevention trials. Neuron. 2014;84(3):608-622.
3. Vellas B. Update on Prevention Trials in Alzheimer’s Disease. J Prev Alzheimers Dis. 2014;1(3):168-175.
4. Weninger S, Carrillo MC, Dunn B, et al. Collaboration for Alzheimer’s Prevention: Principles to guide data and sample sharing in preclinical Alzheimer’s disease trials. Alzheimers Dement. 2016;12(5):631-632.
5. Alber J, Lee AKW, Menard W, Monast D, Salloway SP. Recruitment of At-Risk Participants for Clinical Trials: A Major Paradigm Shift for Alzheimer’s Disease Prevention. J Prev Alzheimers Dis. 2017;4(4):213-214.
6. Brookmeyer R, Abdalla N. Design and sample size considerations for Alzheimer’s disease prevention trials using multistate models. Clin Trials. 2019;16(2):111-119.
7. Rafii MS, Aisen PS. Alzheimer’s Disease Clinical Trials: Moving Toward Successful Prevention. CNS Drugs. 2019;33(2):99-106.
8. Galvin JE. Prevention of Alzheimer’s Disease: Lessons Learned and Applied. J Am Geriatr Soc. 2017;65(10):2128-2133.
9. Andrieu S, Coley N, Lovestone S, Aisen PS, Vellas B. Prevention of sporadic Alzheimer’s disease: lessons learned from clinical trials and future directions. Lancet Neurol. 2015;14(9):926-944.
10. Wang J, Tan L, Yu JT. Prevention Trials in Alzheimer’s Disease: Current Status and Future Perspectives. J Alzheimers Dis. 2016;50(4):927-945.
11. Vellas B, Bain LJ, Touchon J, Aisen PS. Advancing Alzheimer’s Disease Treatment: Lessons from CTAD 2018. J Prev Alzheimers Dis. 2019;6(3):198-203.
12. Petersen RC, Morris JC. Mild cognitive impairment as a clinical entity and treatment target. Arch Neurol. 2005;62(7):1160-1163; discussion 1167.
13. Albrecht JS, Hanna M, Kim D, Perfetto EM. Predicting Diagnosis of Alzheimer’s Disease and Related Dementias Using Administrative Claims. J Manag Care Spec Pharm. 2018;24(11):1138-1145.
14. Gianattasio KZ, Prather C, Glymour MM, Ciarleglio A, Power MC. Racial disparities and temporal trends in dementia misdiagnosis risk in the United States. Alzheimers Dement (N Y). 2019;5:891-898.
15. Zhu CW, Ornstein KA, Cosentino S, Gu Y, Andrews H, Stern Y. Misidentification of Dementia in Medicare Claims and Related Costs. J Am Geriatr Soc. 2019;67(2):269-276.
16. Dell’Agnello G, Desai U, Kirson NY, et al. Reliability of coded data to identify earliest indications of cognitive decline, cognitive evaluation and Alzheimer’s disease diagnosis: a pilot study in England. BMJ Open. 2018;8(3):e019684.
17. Gilmore-Bykovskyi AL, Block LM, Walljasper L, Hill N, Gleason C, Shah MN. Unstructured clinical documentation reflecting cognitive and behavioral dysfunction: toward an EHR-based phenotype for cognitive impairment. J Am Med Inform Assoc. 2018;25(9):1206-1212.
18. Graham SA, Lee EE, Jeste DV, et al. Artificial intelligence approaches to predicting and detecting cognitive decline in older adults: A conceptual review. Psychiatry Res. 2020;284:112732.
19. Amra S, O’Horo JC, Singh TD, et al. Derivation and validation of the automated search algorithms to identify cognitive impairment and dementia in electronic health records. J Crit Care. 2017;37:202-205.
20. Uspenskaya-Cadoz O, Alamuri C, Wang L, et al. Machine Learning Algorithm Helps Identify Non-Diagnosed Prodromal Alzheimer’s Disease Patients in the General Population. J Prev Alzheimers Dis. 2019;6(3):185-191.
21. Marsh AP, Lovato LC, Glynn NW, et al. Lifestyle interventions and independence for elders study: recruitment and baseline characteristics. J Gerontol A Biol Sci Med Sci. 2013;68(12):1549-1558.
22. Messer KL, Herzog AR, Seng JS, et al. Evaluation of a mass mailing recruitment strategy to obtain a community sample of women for a clinical trial of an incontinence prevention intervention. Int Urol Nephrol. 2006;38(2):255-261.
23. Sano M, Jacobs D, Andrews H, et al. A multi-center, randomized, double blind placebo-controlled trial of estrogens to prevent Alzheimer’s disease and loss of memory in women: design and baseline characteristics. Clin Trials. 2008;5(5):523-533.
24. Adapt Research Group, Meinert CL, McCaffrey LD, Breitner JC. Alzheimer’s Disease Anti-inflammatory Prevention Trial: design, methods, and baseline results. Alzheimers Dement. 2009;5(2):93-104.
25. Meinert CL, Breitner JC. Chronic disease long-term drug prevention trials: lessons from the Alzheimer’s Disease Anti-inflammatory Prevention Trial (ADAPT). Alzheimers Dement. 2008;4(1 Suppl 1):S7-S14.
26. Fitzpatrick AL, Fried LP, Williamson J, et al. Recruitment of the elderly into a pharmacologic prevention trial: the Ginkgo Evaluation of Memory Study experience. Contemp Clin Trials. 2006;27(6):541-553.
27. Gill TM, McGloin JM, Latham NK, et al. Screening, Recruitment, and Baseline Characteristics for the Strategies to Reduce Injuries and Develop Confidence in Elders (STRIDE) Study. J Gerontol A Biol Sci Med Sci. 2018;73(11):1495-1501.
28. Andersen F, Engstad TA, Straume B, et al. Recruitment methods in Alzheimer’s disease research: general practice versus population based screening by mail. BMC Medical Research Methodology. 2010;10(1):35.
29. Wilkinson T, Ly A, Schnier C, et al. Identifying dementia cases with routinely collected health data: A systematic review. Alzheimers Dement. 2018.
30. Barnes DE, Zhou J, Walker RL, et al. Development and Validation of eRADAR: A Tool Using EHR Data to Detect Unrecognized Dementia. J Am Geriatr Soc. 2020;68(1):103-111.
31. Reuben DB, Hackbarth AS, Wenger NS, Tan ZS, Jennings LA. An Automated Approach to Identifying Patients with Dementia Using Electronic Medical Records. J Am Geriatr Soc. 2017;65(3):658-659.



M. Kosaner Kließ1, R. Martins1, M.P. Connolly1,2

1. Health Economics, Global Market Access Solutions Sarl, St-Prex Switzerland; 2. Unit of Pharmacoepidemiology & Pharmacoeconomics, Department of Pharmacy, University of Groningen, Groningen, The Netherlands

Corresponding Author: Mark Connolly PhD, Unit of Pharmacoepidemiology & Pharmacoeconomics, Department of Pharmacy, University of Groningen, Antonius Deusinglaan 1, 9713 AV Groningen, The Netherlands,,

J Prev Alz Dis 2021;3(8):362-370
Published online April 24, 2021,



Background: Alzheimer’s Disease is the most common cause of dementia, affecting memory, thinking and behavior. Symptoms eventually grow severe enough to interfere with daily tasks. AD is predicted to increase healthcare spending and costs associated with formal and informal caregiving. The aim of this study was to identify and quantify the contribution of the different cost components associated with AD.
Methods: A structured literature review was conducted to identify studies reporting the economic burden of Alzheimer`s Disease beyond the healthcare setting. The search was conducted in Medline, Embase and EconLit and limited to studies published in the last 10 years. For each identified cost component, frequency weighted mean costs were calculated across countries to estimate the percentage contribution of each component by care setting and disease severity. Results obtained by each costing approach were also compared.
Results: For community-dwelling adults, the percentage of healthcare, social care and indirect costs to total costs were 13.9%, 17.4% and 68.7%, respectively. The percentage of costs varied by disease severity with 26.0% and 10.4% of costs spent on healthcare for mild and severe disease, respectively. The proportion of total spending on indirect costs changed from 60.7% to 72.5% as disease progressed. For those in residential care, the contribution of each cost component was similar between moderate and severe disease. Social care accounted on average for 85.9% of total costs.
Conclusion: The contribution of healthcare costs to the overall burden was not negligible; but was generally exceeded by social and informal care costs.

Key words: Indirect costs, healthcare costs, Alzheimer’s Disease, societal perspective, economic evaluation.



Many chronic diseases pose significant economic and humanistic burden for patients, families, and society as a whole. For example, it has been estimated that the indirect costs of lost economic productivity of people with chronic diseases are almost 300% greater than the direct costs of healthcare (1). The economic consequences of health-related employment inactivity of people with chronic conditions can also extend to the government due to increased spending on support programs and lost tax revenues (2, 3). Fewer people working, earning income and paying taxes generates lost tax revenue for the government and increasing dependency on public benefits support (4). The externalities of poor health can further extend to family members or friends who may reduce or discontinue their work in order to provide informal care (5-8). Furthermore, informal caregiving can impact the well-being of those providing care, which is shown to be proportional to the amount of care provided (9, 10) suggesting that as the Alzheimer’s disease (AD) population grows, the externalities of the condition also expand.
Researchers have increasingly studied the cost related to informal caregiving due to its significant impact on families as well as the overall contribution to the total economic burden of many chronic conditions (11). Studies have also examined how including the cost of informal care can influence findings of cost-effectiveness studies, where inclusion of the cost of informal care can determine the likelihood that interventions are considered cost-effective or not (12). Many determinants can influence the amount of informal care provided, including age, gender, geographic region, caregiver relationship, the level of dependence of the person requiring care and the amount of social services being provided (1, 13).
The importance of informal caregiving is exemplified by AD, which is a progressive chronic condition with increasing global prevalence (14). AD is a continuum with the first clinically recognizable stage being Mild Cognitive Impairment (MCI) (15). MCI refers to individuals who function similarly to their peers and suffer some cognitive impairment, but it is not sufficiently severe for it to be considered dementia (16). As the disease progresses, symptoms gradually worsen and in the later stages patients typically lose their independence and become dependent on formal or informal care. As a result, AD is predicted to increase healthcare spending and costs associated with formal and informal caregiving compared to an average aging population.
This is particularly important as AD progresses, and more intensive care is required (17-19). Increasing demands are placed on informal care at a time when the proportion of working aged adults is decreasing in many advanced economies, which could influence economies and labor markets (20). There is growing evidence of the significant economic burden that AD poses on the healthcare system as well as on patients and their families. To further understand the contribution of healthcare costs to overall costs attributed to AD, we have reviewed the literature to identify studies that provide comprehensive estimates of financial burden including productivity losses, informal care costs, institutionalization costs and other economic domains. We believe that dissecting the cost components can give a more complete picture of the overall burden of AD, emphasize the major cost drivers associated with AD, and in the end serve as a foundation for future policy frameworks.

Study aims

The aim of this literature review was to provide an overview of the different cost components associated with AD and estimate the proportion of overall costs of AD that are attributable to healthcare in comparison with all other attributable costs incurred by individuals, households and society.



Search strategy

A comprehensive search strategy was constructed using controlled vocabulary and free-text terms relating to the population, outcomes and study designs of interest. Population terms included those related to AD and mixed dementia, as well as neurocognitive disorders other than AD, and those defined by the Diagnostic and Statistical Manual of Mental Disorders (DSM-V) and recognized patient societies, in order to reduce irrelevant studies. Outcome terms were clustered around five concepts: labor force participation and income, disposable income, social security, disability allowances and indirect costs. These measures are typically not included in randomized trials; or are reported as secondary outcomes for which studies are not powered to analyze. Additionally, when these data are collected alongside randomized trials they are intervention-specific, restricted to shorter follow-up periods and of limited generalizability due to strict trial inclusion criteria (21).Therefore, a search filter for observational studies formed the last search concept. The search was limited to humans and to studies published in the last 10 years. No language limitations were predefined. The full strategy provided in Supplement 1 was used for searching MEDLINE (PubMed) and adapted for searching EMBASE (OVID) and EconLIT. Backwards snowballing was conducted on eligible studies to identify further relevant research.

Study eligibility


Individuals identified with MCI likely due to AD or AD with or without another form of dementia were included along with their caregivers. Populations limited to a single gender or AD in combination with non-dementia health conditions were excluded.


Comparisons of AD to a cognitively normal population or between different stages of AD were of interest.


For the patient and caregiver, the outcomes of interest included direct and indirect healthcare costs; these including but not limited to income, labor force participation, economic (in)activity, work adaptation; disposable income; social insurance allowance or benefit; disability allowance and caregiver’s allowance. Studies assessing total societal costs which included health costs and the cost of each component as well as the total were included. However, studies reporting only on a single component of economic impact, e.g., only informal costs or health costs only, were excluded.

Study design

Non-interventional, observational studies providing an overview of AD were included. Interventional studies were kept in if they reported relevant outcomes; however, they were of less priority. Randomized or quasi-randomized clinical trials, traditional and systematic literature reviews, qualitative studies, methodological papers or study protocols, economic modeling studies, comments, editorials and letters were excluded. Studies with less than 10 subjects per arm were also omitted.

Study selection

References were downloaded into ENDNOTE version 9.3. Study titles and abstracts were screened against the eligibility criteria described above by a single reviewer. The full texts of relevant studies were subsequently obtained and screened by two independent reviewers. Posters of conference abstracts were sought if the material had not been published in a journal manuscript. Uncertainties between reviewers were resolved by discussion with a third reviewer.

Data extraction and synthesis

Data were extracted from each study by a single reviewer on study design and duration, country, care setting, sample size and age, disease diagnosis and disease severity; measurement and costing of resources (costing approach, costing year and currency), and the absolute mean and variance of each cost component and of the total costs. The resource items comprising each cost component were also recorded.
The percentage of total costs covered by each component was calculated for the overall AD population in each study and by disease severity. Outcomes from cross-sectional studies and at baseline from longitudinal studies were narratively synthesized. For each cost component, frequency weighted mean costs were calculated to summarize results across countries by disease severity, and per country when multiple studies were available. For this purpose, all costs were inflated to 2019 using country specific consumer price index values (22) and then converted to Euros. Primary analysis was based on studies that used the human capital approach for valuing indirect costs (23) and repeated for each care setting. When studies reported multiple analyses, results obtained with supervision time from a caregiver or family member were included. A separate assessment was conducted on studies that valued informal care using the labor replacement approach, i.e., by using the cost for hiring a professional caregiver. Results obtained with the two costing approaches for the community setting were compared. Economic elements not included in the estimation of the total societal costs, i.e., income, were narratively summarized.



The search yielded 2250 results. After removing duplicates, the titles and abstracts of 1740 records were screened of which 143 were considered relevant for full-text screening. Of these, 3 were conference abstracts for which journal publications were identified; 1 was a repeated publication; 10 provided an insufficient description of methods or results and 108 met at least one exclusion criteria. 21 publications were included in a narrative synthesis. Five publications were further included in synthesis after backward snowballing. Study selection is depicted in Figure 1.

Figure 1. Flow of study selection


Characteristics of individual studies

Ten publications reported results from the GERAS I (18, 24-28), GERAS II (29, 30), and extensions of the GERAS to Japan (31) and the USA (32). The remaining 20 publications included the ECO, EVOCOST and Codep-AD studies from Spain (33-36); the ECAD from Ireland (17, 37); one multinational study (38); a cluster-randomized observational study from China (39, 40); and others from France (41), Germany (42), Sweden (43), and the USA (44-48). Together there were 17 studies with unique methodologies.
One retrospective case-control study from the USA used a claims database to assess patient and caregiver medical costs in comparison to a cognitively healthy spouse-patient dyad (47-50). Based on population survey data also from the USA, Ton (46) assessed the relationship between cognitive decline (MCI and AD) and household income in addition to patient medical costs.
The total socioeconomic burden was estimated in 15 studies. The characteristics of these are summarized in Table 1. Two studies used random sampling to identify study sites (33, 39, 40). In the remaining studies, participants were conveniently sampled from their healthcare settings by their local healthcare providers. Longitudinal studies (9 studies) limited their sample to community-dwelling adults, with exception of the ECO study that also included individuals from a residential setting. Three studies further restricted their sample by disease severity: the EVOCOST study focused on adults with moderate disease severity (34); the GERAS-US study (32) compared mild AD against MCI; and Zhu (45) compared adults with MCI against cognitively healthy adults. Cross-sectional studies (6 studies) included a broad sample from the community and residential setting, except for Gervès et al (2014) who studied community-dwelling adults; and most did not specify an age-limit for inclusion (35, 36, 38, 42, 43). Disease severity was defined by the Mini-mental State Exam (MMSE) scores in 14 studies; and by the Clinical Dementia Rating (CDR) in the ECO and Codep-AD studies (33, 35, 36). Discrepancy was observed between studies in the diagnostic criteria for AD and disease staging based on MMSE scores. Two studies staged disease severity by dependency level (36, 44).

Table 1. Characteristics of studies assessing total socioeconomic burden of AD

1. National Institute of Neurological and Communicative Disorders, and Stroke and Alzheimer’s Disease and Related Disorders Association criteria; 2. National Institute on Aging and Alzheimer’s Association Alzheimer’s criteria; 3 .International Working Group on Mild Cognitive Impairment (J Intern Med 2014; 256(3):240-246) ; AD: Alzheimer`s Disease. CN: China. DE: Germany. ESP: Spain. FR: France. IE: Ireland. IT: Italy. JPN: Japan. MCI: Mild cognitive impairment. NR: Not reported. NA: Not available. SWE: Sweden. UK: United Kingdom. USA: United States of America. MMSE: Mini-Mental State examination. FAQ: Functional Activities Questionnaire. CDR: Clinical Dementia Rating. GDS: Global Deterioration Scale.


Overall, adults with MCI likely to be due to AD were included in 3 studies (17, 32, 37, 42); their outcomes were reported separately from adults with AD in the GERAS-US (32).
All 16 studies included patient health care, social care and informal care in their estimation of total socioeconomic burden. There were minimal differences across studies in the resource items assessed as most studies used the Resource Utilization in Dementia (RUD) (52) or RUD-Lite (53)instruments for measuring resource utilization. The case-control study by Zhu (45) differed from the others by using the Resource Use Inventory (54) to capture resource utilization and not valuing the use of informal care in MCI. It is also noteworthy that Reese (42) conducted their economic evaluation from the perspective of the German statutory health insurance; formal and informal care were assessed together as a component of social care. This evaluation also estimated productivity losses of the patient and caregiver. Productivity loss of the caregiver was evaluated independently from informal care in one other study where informal care was accounted as lost leisure time (35). Informal care was accounted as productivity loss in one study each from the USA (44) and China (39, 40). The Chinese study further considered intangible costs which accounted for 4.2% of total costs. Additionally, healthcare costs of the caregiver were evaluated by GERAS I, GERAS II-Spain and GERAS-US.
The contribution of patient health and social care and indirect costs to total societal costs, without caregiver health care and intangible costs, were calculated across all studies. Indirect costs related to informal care and productivity loss when evaluated separately.

Cost components by setting

The cost components attributed to the MCI population were obtained from a single study where the largest component of overall costs was patient health care costs (50.9%) followed by informal care costs (40.1%) when using the human capital approach. The case-control study by Zhu (45) found hospitalization to be the largest component of medical costs and that adults with MCI required significantly more informal care than cognitively healthy adults.
In community-dwelling adults with AD, the weighted mean contribution of health care costs was 26.0%, 15.7% and 10.4% for mild, moderate and severe forms of AD, respectively; and averaged 13.9% across all severity levels. Results summarized in Table 2 show that the weighted mean contribution of indirect cost to the overall cost burden was substantially high and increased as disease progressed representing 60.7%, 67.1% and 72.5% for mild, moderate and severe AD, respectively. Country-level data presented in Supplement 2 show that patient health care costs formed a greater component of total costs in the USA compared to European countries at all disease severity levels; and the least in Italy where informal care costs exceeded 80% of total costs. Further, social care costs composed a larger amount of the total costs in Japan and Sweden, and even exceeded the contribution of informal care in Sweden.

Table 2. Weighted mean (min-max) contribution of each cost component to total costs across countries

† only one MCI study identified.


For adults living in residential care, the weighted mean contribution of cost components was similar between moderate and severe AD, as shown in Table 2. Across severity levels, patient social care formed 85.9% of total costs and patient health care was slightly larger than that of informal care (8.6% vs. 5.5%). Further, the percentage contribution of each cost component was similar between countries. The difference in minimum and maximum values between Germany, Spain, Sweden, UK and USA were 3.1%, 8.5% and 5.8% for patient health care, social care and indirect costs, respectively between Germany, Spain, Sweden, UK and USA. Country-level data are tabulated in Supplement 2.
In studies that assessed both community and residential care settings, the percentage contribution of cost components varied between countries in terms of social care (15.6%-83.9%) and informal care (9.4%-67.8%). Looking at country-level data (Supplement 2), this outcome was heavily influenced by high social care costs and little informal care in Sweden. Additionally, social care constituted a smaller component of total costs than patient healthcare in China (15.6% vs. 32.5%) than in European countries.

Comparison of costing approaches

The choice of method for costing informal caregiving time had a substantial impact on the distribution of cost components in the early stages of cognitive decline. Using the labor replacement approach increased the weighted mean contribution of patient healthcare to total costs for MCI (79.9% vs. 50.9%) and mild AD (39% vs. 26%), as shown in Figure 2. Country-level results provided in Supplement 2 show that this was especially true for the USA where the contribution of patient healthcare almost doubled (36.2% to 65.4%). Smaller, but observable changes also occurred in Spain, Germany and Italy. Data for these countries came from analyses that excluded supervision time from informal care. Additional analysis was carried out using results from the GERAS studies to explore how the inclusion of supervision time influences results. Across France, Germany, UK, Spain and Italy, the weighted mean contribution of patient health and social care were equally elevated by 5% to 6% with the exclusion of supervision time from informal care calculations. Results are presented in Supplement 3.

Figure 2. Comparison of labor replacement method and human capital approach for valuing indirect costs and influence on percentage contribution of each cost component


Contribution of caregiver healthcare to overall costs

Across the GERAS-I countries, Spain and the USA, caregiver healthcare costs accounted for 6.9% of total costs in adults with MCI likely due to AD (32) and 3.7% of total costs in those with AD. As shown in Table 3, the percentage contribution of this component to the total cost decreased substantially from mild (11.5%) to moderate AD (4.4%) and reached 2.3% for severe AD. Across AD severity levels, the contribution of caregiver healthcare costs showed little variation between countries (3% – 4.2%).

Table 3. Weighted mean (min-max) contribution of caregiver healthcare costs

AD: Alzheimer`s Disease. MCI Mild Cognitive Impairment. HC: Healthcare. SC: Social care.


Impact of AD on other socioeconomic aspects

Ton et a (2017) (46) demonstrated that in the USA not only adults with AD but also those with MCI had greater medical expenditure and less household income than cognitively healthy adults (<0.001). This result remained highly significant after adjusting for age, sex, race, education, marital status, residential region and comorbidities (<0.015). Another study demonstrated that, compared to MCI, significantly more individuals with mild AD were pushed to an income below the federal poverty level. Patients’ employment rates were found to significantly drop from 21.4% to 9.4%; and the number of employed adults who reduced their work significantly rose from 3.2% to 13.8% (32). In the broader AD population, a significant relationship between dependency and household income has not been found (44).
When examining the impact of AD on household expenditure (47, 48), an US study indicated that annual health care costs were double the amount of costs of a cognitively healthy household ($6,028 vs. $2,951). Patient health care costs were significantly higher than age, sex and comorbidity-matched adults ($4408 vs. $1473, p<0.001). Spousal caregivers accumulated significantly higher costs for AD-related and mental health prescription; but on average were not significantly different from spouses of cognitively health adults.



The rising costs of treating AD and the impact on households and caregivers has been a topic of concern for researchers, policy-makers and planners for many years (55). The work described here helps to put expenditure into perspective to understand major cost drivers in the delivery of care to people with AD. This review has illustrated that in community-dwelling adults with AD, patient healthcare costs constitute the smallest component of the total cost burden representing, on average, 13.9% across all AD severity levels. Furthermore, the contribution of healthcare costs to the overall cost burden decreases as disease progresses and as informal care needs increase. As described here, the costs of informal care represent approximately 60% of total costs, and reach 72.5% of the total cost burden in severe AD. The difference between the contribution of patient healthcare and indirect costs was substantially reduced in early stages of AD when using the replacement labor approach to valuing informal care. This may be due to higher employment rates of the caregiver of adults with MCI and mild AD compared to the later stages; and that this is disregarded with the use of a uniform cost to value caregiving time. Robinson (32) reported employment rates of 48.3% and 43.4% respectively for patients with MCI and mild AD; with later stages of AD this tends to drop below 30% (18, 30).
Variation in the distribution of the cost components in the community and residential professional care settings emphasize the importance of studying each setting separately. When costs were pooled across settings, results were heavily influenced by residency care and showed high variability between countries. It is important to put the informal care costs into perspective as these represent lost earnings for individuals with significant economic consequences (56). Therefore, interventions that delay progression can offer economic benefits due to reduced need for informal and formal care.
We observed that the distribution of cost components was relatively similar between European countries. In Italy, however, there was a heavy reliance on informal care and little utilization of medical care which became even more apparent with increasing disease severity. The provision of long-term care by the family may be due to differences in the formalization of and access to healthcare compared to other European countries (29). The greater contribution of community care in Japan, compared to European countries, may be due to the caregiver being an adult-child of the person with AD (31), and in Sweden due to the availability of different social care structures (38). Such factors have been considered in other comparisons of country-level data (24, 26).
This review identified few studies evaluating the broader economic burden of MCI likely due to AD, probably because of the recent introduction of this term and the difficulty to establish this diagnosis (57). These studies demonstrated that individuals with MCI likely due to AD require social care and informal care more than their age-matched peers; and that this is further increased in those with mild AD dementia (32, 45). A similar trend is seen with caregiver health care costs when they are included in the estimation of total costs. These results highlight the importance of reporting disaggregated outcomes across early stages of cognitive decline. As more sensitive diagnostic methods become available to detect changes in cognition and more therapies become available to slow down progression early in the AD continuum, the need to explore the wide socioeconomic impact of cognitive decline will become more pertinent.
The results of this review should be interpreted with caution as a small number of studies were included. A larger number of studies might have been identified by removing the search limit on publication dates. The intention of this search limit was to identify studies reflecting current treatment practices. As part of a rapid review, study screening and data extraction were carried out mostly by a single reviewer, and the quality of the included studies were not assessed due to limited time and resources. The exclusion of quality appraisal is justifiable as a meta-analysis of study results was not possible. The analysis was nonetheless quantitative in nature and would not have benefited from the inclusion of qualitative evidence. Calculation of a frequency-weighted mean cost across countries was seen as a descriptive method for summarizing estimated costs per person. Differences in criteria for disease diagnosis and staging were not considered in data synthesis. Only the extensions of the GERAS study applied the more recent diagnostic criteria from the National Institute on Aging and Alzheimer’s Association Alzheimer’s (NIA-AAA) (58). Study-level results differed more between diagnostic criteria than between disease staging based on MMSE scores. Differences in AD severity categorization are likely to generate cost data somewhat different in absolute terms. There is a clear trend in the data showing that a reduction in the proportional contribution of healthcare costs is accompanied by an increase in the contribution of indirect costs, as severity progresses (Figure 2). The authors believe that this overall trend is unlikely to be substantially altered were AD categories more homogeneous.
NIA-AAA criteria distinguish AD dementia from earlier stages of cognitive decline, not limited to memory loss alone, and from other dementing conditions. They also recognize the additional use of imaging methods or biomarker analysis in increasing certainty in diagnosis, particularly for the differential diagnosis of MCI likely due to AD. However, at time of publication ancillary testing was described as optional clinical tools, advocating more investigational research on their use and standardization (57, 58). The Alzheimer`s Disease Neuroimaging Initiative has played an important role in the quest to find sensitive biomarkers and diagnostic tests; and have developed standardized methods for clinical tests, magnetic resonance imaging, positron emission tomography and cerebrospinal fluid biomarkers (59). Multi-modal use of neuroimaging and biological markers has been recommended as the way forward for detecting changes in cognition throughout the AD pathophysiology (60), and for predicting future decline (59). Blood biomarkers have also been developed as a non-invasive, low-cost alternative to cerebrospinal fluid biomarkers; and have shown to be effective in differentiating AD, MCI and cognitively normal controls (59, 61). These recent advances will likely impact the incidence of MCI due to AD and AD dementia and their associated health care costs. Study-level results from this review suggest the contribution of patient health care costs to be lower and that of social care costs to be higher with NIA-AAA criteria compared to older diagnostic criteria. Future observational studies reflecting the use of modern methods are needed to explore this hypothesis.



Healthcare costs can cover up to 30% of the overall burden of AD; but is generally exceeded by the costs associated with social care and informal care in the community setting the contribution of indirect costs to overall costs increases and that of patient healthcare decreases as disease progresses. As people transition from community care to residential care, the proportion of spending on social care increases and that of indirect costs substantially decreases. Such a transition allows some caregivers to regain independency and rejoin the labor force. The reliance on informal care in the community setting is likely due to the differing availability and organization of social care between countries particularly in the earlier, less dependent stages of AD.


Funding: This work was funded by Biogen.

Disclosures: During the peer review process, Biogen had the opportunity to review and comment on the manuscript. The authors had full editorial control of the manuscript and provided their final approval on all content to be published. The authors hold no financial interests in the sponsoring company.

Conflict of interest: The authors report no further conflicts of interest in relation to the work described here.

Ethical standards: The analysis reported here is based on previously reported literature. No individual patient data has been collected for this study and no ethics approval was required.

Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits use, duplication, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.





1. Courbage C, Montoliu-Montes G, Wagner J. The effect of long-term care public benefits and insurance on informal care from outside the household: Empirical evidence from Italy and Spain. Eur J Health Econ 2020;21:1131-47.
2. Connolly MP, Kotsopoulos N, Postma MJ, Bhatt A. The fiscal consequences attributed to changes in morbidity and mortality linked to investments in health care: A government perspective analytic framework. Value Health 2017;20(2):273-7.
3. Black D. Working for a healthier tomorrow. Department for Work and Pensions, 2008. Accessed 15 November 2020.
4. Schofield D, Shrestha RN, Zeppel MJ, et al. Economic costs of informal care for people with chronic diseases in the community: Lost income, extra welfare payments, and reduced taxes in Australia in 2015–2030. Health Soc Care Community 2019;27(2):493-501.
5. Farré M, Kostov B, Haro JM, et al. Costs and burden associated with loss of labor productivity in informal caregivers of people with Dementia. J Occup Environ Med 2018;60(5):449-56.
6. Schofield D, Zeppel MJ, Tanton R, et al. Informal caring for back pain: Overlooked costs of back pain and projections to 2030. Pain 2020;161(5):1012-8.
7. Nguyen HT, Connelly LB. The effect of unpaid caregiving intensity on labour force participation: Results from a multinomial endogenous treatment model. Soc Sci Med 2014;100:115-22.
8. Eikemo TA. Hidden health care in European health systems. Scand J Public Health 2018;46(4):433-5.
9. Chen L, Fan H, Chu L. The hidden cost of informal care: An empirical study on female caregivers’ subjective well-being. Soc Sci Med 2019;224:85-93.
10. Conde-Sala JL, Garre-Olmo J, Turró-Garriga O, Vilalta-Franch J, López-Pousa S. Quality of life of patients with Alzheimer’s disease: Differential perceptions between spouse and adult child caregivers. Dement Geriatr Cogn Disord 2010;29(2):97-108.
11. Oliva-Moreno J, Trapero-Bertran M, Peña-Longobardo LM, del Pozo-Rubio R. The valuation of informal care in cost-of-illness studies: A systematic review. Pharmacoeconomics 2017;35(3):331-45.
12. Krol M, Papenburg J, van Exel J. Does including informal care in economic evaluations matter? A systematic review of inclusion and impact of informal care in cost-effectiveness studies. Pharmacoeconomics 2015;33(2):123-35.
13. Oliva-Moreno J, Peña-Longobardo LM, García-Mochón L, del Río Lozano M, Mosquera Metcalfe I, García-Calvente MdM. The economic value of time of informal care and its determinants (The CUIDARSE Study). PLoS One 2019;14(5):e0217016.
14. Nichols E, Szoeke CE, Vollset SE, et al. Global, regional, and national burden of Alzheimer’s disease and other dementias, 1990–2016: A systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol 2019;18(1):88-106.
15. 2018 Alzheimer’s disease facts and figures. Alzheimers Dement 2018;14(3):367-429.
16. Petersen RC. Mild cognitive impairment as a diagnostic entity. J Intern Med 2004;256(3):183-94.
17. Gillespie P, O’Shea E, Cullinan J, et al. Longitudinal costs of caring for people with Alzheimer’s disease. Int Psychogeriatr 2015;27(5):847-56.
18. Wimo A, Reed CC, Dodel R, et al. The GERAS study: A prospective observational study of costs and resource use in community dwellers with Alzheimer’s disease in three European countries – Study design and baseline findings. J Alzheimers Dis 2013;36(2):385-99.
19. Reed C, Belger M, Scott Andrews J, et al. Factors associated with long-term impact on informal caregivers during Alzheimer’s disease dementia progression: 36-month results from GERAS. Int Psychogeriatr 2020;32(2):267-77.
20. Gerrard L. Women’s unpaid dementia care and the impact on employment Alzheimer’s Society, 2020. Accessed 15 November 2020.
21. Drummond MF, Sculpher MJ, Claxton K, Stoddart GL, Torrance GW. Using clinical studies as vehicles for economic evaluations. Methods for the economic evaluation of health care programmes. Oxford: Oxford University Press; 2015. p. 267-310.
22. Inflation (CPI) (indicator) [Internet]. 2020 [cited 5 November 2020oecd].
23. Brazier J, Ratcliffe J, Saloman J, Tsuchiya A. Measuring and valuing health benefits for economic evaluation. 2017. Oxford university press, Oxford.
24. Dodel R, Belger M, Reed C, et al. Determinants of societal costs in Alzheimer’s disease: GERAS study baseline results. Alzheimers Dement 2015;11(8):933-45.
25. Rapp T, Andrieu S, Chartier F, et al. Resource use and cost of Alzheimer’s disease in France: 18-month results from the GERAS observational study. Value Health 2018;21(3):295-303.
26. Reed C, Happich M, Argimon JM, et al. What drives country differences in cost of Alzheimer’s disease? An explanation from resource use in the GERAS study. J Alzheimers Dis 2017;57(3):797-812.
27. Lenox-Smith A, Reed C, Lebrec J, Belger M, Jones RW. Resource utilisation, costs and clinical outcomes in non-institutionalised patients with Alzheimer’s disease: 18-month UK results from the GERAS observational study. BMC Geriatr 2016;16(1):195.
28. Boess FG, Lieb M, Schneider E, Zimmermann TM, Dodel R, Belger M. Costs due to Alzheimer’s disease in Germany – Recent results of the observational GERAS study. Gesundheitsokonomie und Qualitatsmanagement 2016;21(5):232-41.
29. Bruno G, Mancini M, Bruti G, Dell’Agnello G, Reed C. Costs and resource use associated with Alzheimer’s disease in Italy: Results from an observational study. J Prev Alzheimers Dis 2018;5(1):55-64.
30. Olazarán J, Agüera-Ortiz L, Argimón JM, et al. Costs and quality of life in community-dwelling patients with Alzheimer’s disease in Spain: Results from the GERAS II observational study. Int Psychogeriatr 2017;29(12):2081-93.
31. Nakanishi M, Igarashi A, Ueda K, et al. Costs and resource use associated with community-dwelling patients with Alzheimer’s disease in Japan: Baseline results from the prospective observational GERAS-J Study. J Alzheimers Dis 2020;74(1):127-38.
32. Robinson RL, Rentz DM, Andrews JS, et al. Costs of early stage Alzheimer’s disease in the United States: Cross-sectional analysis of a prospective cohort study (GERAS-US). J Alzheimers Dis 2020;75(2):437-50.
33. Coduras A, Rabasa I, Frank A, et al. Prospective one-year cost-of-illness study in a cohort of patients with dementia of Alzheimer’s disease type in Spain: The ECO study. J Alzheimers Dis 2010;19(2):601-15.
34. Salva A, Frank-Garcia A, Lereun C, Gimeno V, Milea D, Bineau S. 12-month cost of illness study of moderate Alzheimer’s disease patients in Spain: The evocost study. Value Health 2012;15(7):A549.
35. Darba J, Kaskens L, Lacey L. Relationship between global severity of patients with Alzheimer’s disease and costs of care in Spain; Results from the Co-Dependence study in Spain. Eur J Health Econ 2015;16(8):895-905.
36. Darbà J, Kaskens L. Relationship between patient dependence and direct medical-, social-, indirect-, and informal-care costs in Spain. Clinicoecon Outcomes Res 2015;7:387-95.
37. Gillespie P, O’Shea E, Cullinan J, et al. The effects of dependence and function on costs of care for Alzheimer’s disease and Mild Cognitive Impairment in Ireland. Int J Geriatr Psychiatry 2013;28(3):256-64.
38. Gustavsson A, Brinck P, Bergvall N, et al. Predictors of costs of care in Alzheimer’s disease: A multinational sample of 1222 patients. Alzheimers Dement 2011;7(3):318-27.
39. Jia J, Wei C, Chen S, et al. The cost of Alzheimer’s disease in China and re-estimation of costs worldwide. Alzheimers Dement 2018;14(4):483-91.
40. Yan X, Li F, Chen S, Jia J. Associated factors of total costs of Alzheimer’s disease: A cluster-randomized observational study in China. J Alzheimers Dis 2019;69(3):795-806.
41. Gervès C, Chauvin P, Bellanger MM. Evaluation of full costs of care for patients with Alzheimer’s disease in France: The predominant role of informal care. Health Policy 2014;116(1):114-22.
42. Reese JP, Heßmann P, Seeberg G, et al. Cost and care of patients with Alzheimer’s disease: Clinical predictors in german health care settings. J Alzheimers Dis 2011;27(4):723-36.
43. Mesterton J, Wimo A, By A, Langworth S, Winblad B, Jönsson L. Cross sectional observational study on the societal costs of alzheimer’s disease. Curr Alzheimer Res 2010;7(4):358-67.
44. Michaud TL, High R, Charlton ME, Murman DL. Dependence stage and pharmacoeconomic outcomes in patients With Alzheimer disease. Alzheimer Dis Assoc Disord 2017;31(3):209-17.
45. Zhu CW, Sano M, Ferris SH, Whitehouse PJ, Patterson MB, Aisen PS. Health-related resource use and costs in elderly adults with and without Mild Cognitive Impairment. J Am Geriatr Soc 2013;61(3):396-402.
46. Ton T, DeLeire, T., May, S. G., Hou, N., Tebeka, M. G., Chen, E., Chodosh, J. The financial burden and health care utilization patterns associated with amnestic Mild Cognitive Impairment. Alzheimers Dement 2017;13(3):217-24.
47. Brixner DI, McAdam-Marx C, Van Den Bos J, Perlman D. Cost one year post diagnosis for Alzheimer patients and their spouses vs. matched controls. J Am Geriatr Soc 2010;58:S53.
48. Brixner D, McAdam-Marx C, Perlman D, Bos JVD. Approach to matching Alzheimer`s disease patients and their spouses to assess caregiver burden in an administrative claims database. Value Health 2010;13.
49. Suehs BT, Shah SN, Davis CD, et al. Household members of persons with Alzheimer’s disease: Health conditions, healthcare resource use, and healthcare costs. J Am Geriatr Soc 2014;62(3):435-41.
50. Gilden DM, Kubisiak JM, Kahle-Wrobleski K, Ball DE, Bowman L. Using U.S. Medicare records to evaluate the indirect health effects on spouses: A case study in Alzheimer’s disease patients. BMC Health Serv Res 2014;14:291.
51. Lacey LA, Niecko T, Leibman C, Liu E, Grundman M. Association between illness progression measures and total cost in Alzheimer’s disease. J Nutr Health Aging 2013;17(9):745-50.
52. Wimo A, Jonsson L, Zbrozek A. The Resource Utilization in Dementia (RUD) instrument is valid for assessing informal care time in community-living patients with dementia. J Nutr Health Aging 2010;14(8):685-90.
53. Wimo A, Gustavsson A, Jönsson L, Winblad B, Hsu MA, Gannon B. Application of Resource Utilization in Dementia (RUD) instrument in a global setting. Alzheimers Dement 2013;9(4):429-35.e17.
54. Sano M, Zhu CW, Whitehouse PJ, et al. ADCS Prevention Instrument Project: pharmacoeconomics: assessing health-related resource use among healthy elderly. Alzheimer Dis Assoc Disord 2006;20(4 Suppl 3):S191-202.
55. Deb A, Sambamoorthi U, Thornton JD, Schreurs B, Innes K. Direct medical expenditures associated with Alzheimer’s and Related Dementias (ADRD) in a nationally representative sample of older adults – An excess cost approach. Aging Ment Health 2018;22(5):619-24.
56. 2011 Alzheimer’s disease facts and figures. W V Med J 2011;107(3):82-3.
57. Albert MS, DeKosky ST, Dickson D, et al. The diagnosis of Mild Cognitive Impairment due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 2011;7(3):270-9.
58. McKhann GM, Knopman DS, Chertkow H, et al. The diagnosis of dementia due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimer’s & dementia : the journal of the Alzheimer’s Association 2011;7(3):263-9.
59. Weiner MW, Veitch DP, Aisen PS, et al. 2014 Update of the Alzheimer’s Disease Neuroimaging Initiative: A review of papers published since its inception. Alzheimer’s & dementia : the journal of the Alzheimer’s Association 2015;11(6):e1-e120.
60. Cavedo E, Lista S, Khachaturian Z, et al. The Road Ahead to Cure Alzheimer’s Disease: Development of Biological Markers and Neuroimaging Methods for Prevention Trials Across all Stages and Target Populations. The journal of prevention of Alzheimer’s disease 2014;1(3):181-202.
61. Sabbagh MN, Boada M, Borson S, et al. Early Detection of Mild Cognitive Impairment (MCI) in Primary Care. J Prev Alzheimers Dis 2020;7(3):165-70.