I. Choi1, H. La Monica1, S.L. Naismith2, A. Rahmanovic2, L. Mowszowski2, N. Glozier1
1. Central Clinical School, Brain and Mind Centre, Faculty of Medicine and Health, University of Sydney, Australia; 2. Charles Perkins Centre, Faculty of Science, School of Psychology and the Brain and Mind Centre, University of Sydney, Australia
Corresponding Author: Dr Isabella Choi, 94 Mallett Street, Camperdown, NSW 2050, Australia, firstname.lastname@example.org, +612 8627 7240.
J Prev Alz Dis 2021;
Published online June 23, 2021, http://dx.doi.org/10.14283/jpad.2021.34
Communicating personal Alzheimer’s disease risk profiles based on validated risk algorithms may improve public knowledge about risk reduction, and initiate action. This proof of concept pilot trial aimed to test whether this is feasible and potentially effective and/or harmful. Older at-risk adults (N=24) were provided with their personal Alzheimer’s disease risk profile online, which contained information on their personal risk level, scores and tailored recommendations to manage modifiable risk factors. After receiving the risk profile, participants were significantly more accurate in identifying risk and protective factors, and revised their perceived risk to be lower than their initial estimate. There was no apparent harm seen in psychological distress or dementia-related worry. This shows preliminary support for the feasibility of delivering personal dementia risk profiles to low risk, help-seeking older adults in an online format. A definitive trial examining behavioural outcomes and testing in groups with higher risk profiles is now warranted.
Key words: Risk communication, health literacy, psychological distress, prevention, Alzheimer’s disease.
Up to a third of Alzheimer’s disease cases can be prevented through improved education and reduction of modifiable risk factors (1). Growing evidence from multi-domain interventions shows that targeting modifiable risk factors can reduce risk of Alzheimer’s disease (AD) and improve cognition (2, 3). However, lack of knowledge about dementia and its risk factors among the public is a major barrier to individuals implementing behavioural and lifestyle change and, in turn, to dementia risk reduction (4).
Having an accurate understanding of one’s personal risk of future disease is considered essential for engaging in behaviours for risk reduction. Most health behaviour change models, including the Health Belief Model, identify four major constructs that surround health behaviour: health literacy, perceived susceptibility, motivation to change, and perceived barriers to change (5). However, there is poor health literacy of dementia risk, risk factors, and prevention strategies among the public. A systematic review found that almost half of the general public agreed that dementia was a normal and non-preventable part of ageing (6). Mental activity, healthy diet, physical exercise and social engagement were the most commonly nominated ways to reduce risk, but other well established risk factors such as vascular risk (including smoking, high blood pressure, high cholesterol, obesity), low education, poor mental health, brain trauma, and environmental toxins were rarely mentioned (6-8). Only around 25% of Australians were confident they could reduce risk (8). However, there is support that improving dementia health literacy has a positive impact on risk reduction. People who had a strong belief that dementia risk could be reduced, had moderate to high knowledge of risk-reduction behaviours, or had high confidence that risk reduction can be achieved were almost twice as likely to take action to reduce risk (9).
Communicating personal dementia risk level and risk factors to at-risk adults, based on validated AD risk algorithms, may be one way to improve dementia risk knowledge and engage people in risk reduction behaviours. A number of dementia risk algorithms have been validated for the general population and can identify those with high risk with acceptable predictive ability (10). For instance, the Australian National University Alzheimer’s Disease Risk Index (ANU-ADRI) includes various modifiable risk factors that have been validated for middle-age and older adults that are easily assessed via self-report (11, 12). These evidence-based algorithms, along with personalised risk factor feedback and recommendations to reduce risk, can be delivered online for wide access, allowing people to screen for their risk without having had to first consult with a physician. Older adults are able to use technology proficiently and over 90% of help-seeking older adults with some degree of cognitive impairment use the Internet at home, most commonly for emails (13). Such online dementia risk algorithms already exist and are made freely available to the public, for instance, the CAIDE Risk Score App allows users to detect their dementia risk, obtain information on modifiable risk factors, and receive suggestions on how to modify their risk (14).
Surveys have found that there is high interest among older adults in knowing their risk of AD (15, 16) While there may be potential benefits to disclosing risk, there are also concerns that this could cause negative psychological outcomes. For instance, about 30% of older adults who were interested in knowing their risk also actively worried about developing AD (17). Evidence from a systematic review suggests that disclosure of increased AD risk was not associated with anxiety or depression, but did lead to heightened test-related distress, long-term care insurance uptake and health-related behaviour changes (18).
This pilot trial explores the feasibility and acceptability of communicating online personal dementia risk profiles to at-risk older adults and the impact on dementia health literacy, motivation to engage in behaviour change, and potential harmful psychological effects.
This is an uncontrolled proof of concept pilot trial. The study was approved by the University of Sydney Human Research Ethics Committee (protocol number: 2019/669) and registered with the Australian New Zealand Clinical Trials Registry (ANZCTRN12619001242112p).
People attending the Healthy Brain Ageing Clinic, a specialised memory research clinic for people aged 50 years or older in Sydney, Australia, who were clinically diagnosed with mild cognitive impairment (MCI) or subjective cognitive decline (SCD) between October 2019 to May 2020 were recruited. Those who had dementia or pre-existing severe cognitive impairment due to neurological conditions, psychosis, intellectual disability, substance misuse, stroke, or acquired brain injury were excluded.
As part of standard clinic procedure, all clinic attendees completed self-report measures and were assessed by a geriatrician or neurologist, a psychologist, and a neuropsychologist for a review of medical and psychiatric history, mood, and cognitive functioning. Diagnoses were rated according to consensus, including at least two neuropsychologists and one specialist, and were used to exclude those with a dementia diagnosis and other exclusion criteria. Within two weeks of the clinic assessment, attendees received neuropsychological test result feedback over the phone from a neuropsychologist. The clinic does not provide treatment but refers attendees to suitable clinical investigations, e.g. sleep studies, if warranted, as well as clinical trials for which they are deemed eligible.
After receiving neuropsychological test feedback from a neuropsychologist, eligible attendees were invited via a telephone call to participate in the current study. Interested people had to have an email account and were emailed a survey link via REDCap (Research Electronic Data Capture), a web-based research management platform, with the Participant Information Statement and Consent Form. Participants gave consent to extract relevant data collected from their recent clinic assessment to populate their personal dementia risk profile.
Participants completed self-report baseline measures online via REDCap. Participants’ demographic and risk information were extracted from their standard clinic assessment to compile their personal dementia risk profile using the ANU-ADRI (11). Risk factors in the model included: age, gender, highest level and total number of years of education completed, body mass index, diabetes, depression, high cholesterol, traumatic brain injury, smoking status, alcohol intake, social engagement, physical activity, cognitive activity, and diet.
Within two weeks of completing the baseline self-report measures, participants were emailed a pdf document with their personal dementia risk profile. The risk profile contained standard information about dementia, an explanation of their personal dementia risk profile, and information about the ANU-ADRI risk model. Participants were presented with a visual representation of their risk level in the form of a thermometer showing their risk from 0 to 100, along with an explainer “Your risk of developing dementia is low/ moderate/ high. It is estimated that XX out of 100 people with your risk factors will develop dementia in their lifetime” (Figure 1). They were reminded that this is an estimate based on their risk factors rather than a definitive guarantee, and that there are some risk factors they cannot change but some they could potentially work on to reduce their risk. Participants also received a summary of the dementia risk factors included in the risk model and their scores on each risk factor. They were told why the risk factor was important for brain functioning and were given tailored recommendations to manage each risk factor based on their risk factor score, as well as links for more information.
One week after receiving their risk profile, participants received an email asking them to complete the online post-intervention measures in REDCap. After completing all study measures, participants were reimbursed with a $20 gift card in return for their time.
Primary outcome: Dementia health literacy
Participants were asked “How likely do you think that you will get Alzheimer’s disease in your lifetime?” to assess perceived risk on a scale, where 0%=certain not to happen and 100%=certain to happen. To examine accuracy of perceived risk, the participant’s perceived risk was subtracted from their ANU-ADRI risk estimate. We recoded the difference (D) into a categorical variable, with <−10% indicating overestimation, >10% indicating underestimation, and accurate if −10% ≤ D ≤ 10%, in accordance with previous studies (19). Similarly, to examine change in perceived risk, participants’ perceived risk at post-intervention was subtracted from their perceived risk at baseline (d), and <-10% indicates increased perceived risk, >10% indicates reduced perceived risk, and −10% ≤ d ≤ 10% indicates no change.
We adapted the dementia risk and protective factors questionnaire in the MijnBreincoach survey (20) to assess for knowledge of dementia risk factors. We included five additional modifiable risk and protective factors that were identified in the ANU-ADRI (11) and the Lancet Commission for dementia prevention (21) (i.e. traumatic brain injury, social activity, sleep, education, and age), totalling 19 risk factors. Additional questions asked about barriers to improving brain health, confidence in risk reduction (8), and worry about getting dementia (7).
Secondary outcome: Motivation to Change Lifestyle and Health Behaviours for Dementia
The Motivation to Change Lifestyle and Health Behaviours for Dementia Risk Reduction (MCLHB-DRR) Scale is designed to assess beliefs and attitudes about lifestyle and health behavioural changes for dementia risk reduction among middle-aged and older adults (22). The scale includes (27) items matched onto seven subscales that reflect the seven concepts of the Health Behaviour Model. All items are rated on a 5-point Likert scale from 1 (strongly disagree) to 5 (strongly agree). The scale has moderate to high internal consistencies for the seven subscales, and moderate test-retest reliability (18). Cronbach’s alpha for each of the subscales in this study were: perceived susceptibility (.916), perceived severity (.331), perceived benefits (.715), perceived barriers (.878), cues to action (.656), general health motivation (.638), and self-efficacy (.615).
Secondary outcome: Psychological distress
The K10 is a commonly used screening scale for non-specific psychological distress validated for use in Australia (23). The K10 has also been demonstrated as having moderate sensitivity to symptom change in an Australian sample (24). Scores on the K10 range from 10 to 50, and a score of 30 or more indicates a severe level of distress. Cronbach’s alpha in this study was 0.841.
Secondary outcome: Dementia-related worry
The Dementia Worry Scale was used to assess dementia-related worry (25). It has strong internal consistency and test-retest reliability. It consists of 12 items with scores ranging from 15 to 60. Cronbach’s alpha in this study was 0.908.
We adapted a five-point scale (from 1= not at all to 5= completely) previously used to assess user experience of a dementia information website (26). Participants were asked whether the information provided was engaging and easy to understand as well as how helpful they found the risk profile and how much they felt they had learned (from 1=nothing at all to 5=a great deal). Additionally, participants were asked if they required more information about their dementia risk profile and were given the option to discuss their experience of using the risk profile with a researcher in a telephone interview.
Data was analysed using SPSS version 23.0. Descriptive statistics regarding participant and baseline characteristics were analysed. Fischer’s exact tests and paired samples t-tests were used to test for differences between outcome measures pre- and post- receiving the dementia risk profile. All p-values were two-sided with an alpha of 0.05 to test for significance.
Demographics and baseline characteristics
Overall, 24 eligible participants participated in the trial (Figure 2). Participants’ ages ranged from 53 to 87, with a mean age of 69.54 years (SD 7.69). Over half of the participants (54%) were female and majority spoke English as a first language (83%). Majority were tertiary educated (75%), 21% had completed a trade certificate, and 4% had completed high school. Majority were retired (58%), 29% were employed, and 13% were unemployed. Over half were married or in a de facto relationship (58%), 25% were widowed or divorced, and 17% had never married. The majority (75%) of participants had MCI and 25% had SCD.
Almost all participants (96%; 23/24) were considered to have Low Risk of developing AD (ANU-ADRI score of less than 17), and one participant (4%) was considered as having High Risk (ANU-ADRI score of greater than 27). Participants’ perceived risk of developing AD ranged from 10-99 (M=51.63, SD=23.85), with the majority of participants overestimating their personal risk (87.5%; 21/24).
Pre-post change on dementia health literacy
All participants completed the post-intervention questionnaires. After receiving their personal dementia risk profile, participants’ perceived risk of developing AD ranged from 3-81 (M=38.83, SD=25.02). There was a significant decrease in perceived risk among the group (p = .010). A total of 18 participants (75%) still overestimated their level of risk whereas the remaining 6 correctly identified their level of risk (25%). Eleven participants (45.8%) reported a reduction in their perceived risk, eleven (45.8%) reported no change, and two participants’ (8.3%) perceived risk had increased.
The average number of correctly identified risk and protective factors increased from a mean of 11.42 items (SD= 4.50) at baseline to 13.96 items (SD= 3.98) (t1,23= -3.839, p=.001) at follow-up.
Pre-post change on motivation to engage in behaviour change and psychological effects
There was a significant reduction on the perceived susceptibility subscale of the MCLHB (t1,23=4.416, p<.001) from baseline (M=12.86; SD=3.30) to 1-week follow up (M=10.29; SD=4.21), but no change on the other subscales. There was no change on the K10 or Dementia Worry Scale.
Participants’ self-reported worry about getting dementia was significantly reduced at follow-up, from 2.75 (SD=.85) to 2.29 (SD=.91) (t1,23= 3.412, p= .002). The most common barriers to reducing risk factors at baseline were lack of knowledge (45.8%; 11/24), followed by health problems (25%; 6/24). At 1-week follow up, the most common barriers were lack of motivation (29.2%; 7/24), health problems (29.2%; 7/24), and difficulty with organisation (25%; 6/24). There was no change in confidence to take action to change risk.
Overall, 70.9% (17/24) of participants agreed that the information in the personal risk profile was engaging, 79.2% (19/24) agreed the information was easy to understand, 79.2% (19/24) agreed it was helpful, and 79.2% (19/24) reported they learned a good deal from their personal risk profile.
Two participants participated in the optional telephone interview with a researcher. Both expressed surprise at their lower than expected AD risk feedback, and identified difficulty addressing some of their risk factors (e.g. getting motivated to exercise). One participant agreed that seeking guidance from a health professional may support them to work on their risk factors.
This pilot study aimed to explore the feasibility, acceptability and potential impact of providing an online personal dementia risk profile to help-seeking older adults at risk of developing AD on risk knowledge, motivation to change health-related behaviours, and psychological effects. To our knowledge, this is the first study focusing on the effects of communicating personal risk profile using risk algorithms based on epidemiological risk factors. Communication of the personal dementia risk profile led to more accurate knowledge of AD risk factors and improved understanding of perceived susceptibility among patients with MCI and SCD. Importantly, there was no negative effect of communicating the personal risk profile online on psychological distress or dementia-related worry among our participants. Participants mostly had a low risk of developing AD, but still reported reduced worry about getting dementia after receiving their risk profile. These findings support the feasibility and acceptability of using dementia risk algorithms to deliver personal risk profiles to low risk older adults in an online format, and indicate that providing this information can improve AD health literacy without a negative impact on psychological wellbeing.
However, there was little evidence in this study that providing personal risk profiles as a standalone intervention was sufficient to motivate change in behaviours to address AD risk factors. Although providing the dementia risk profile addressed one main barrier for risk reduction at baseline (i.e. lack of knowledge of dementia risk factors), participants reported that lack of motivation, health problems, and difficulty with organisation became the main barriers after receiving their risk profile. This suggests that older adults need extra support to effect behavioural change. The personal dementia risk profile could potentially be used as part of a collaborative, shared decision-making approach to address these barriers by guiding and engaging users, carers and clinicians to choose several high impact or easy-to-change risk factors to focus on, and by providing feedback on the change in risk level if risk factors are modified. Trials are underway to test the impact of a tailor-made online lifestyle programme targeting modifiable risk factors on risk score and health behaviours and compliance to health advice (27). There may also be a role for clinicians to follow up with specific guidance on addressing health problems and to assist the older adult to develop a personalised risk reduction plan. A recent rapid review on approaches to healthy ageing interventions for older adults demonstrated that optimal interventions are those that incorporate collaborative approaches with shared decision-making and behavioural change techniques (28). In this regard, the personal dementia risk profile represents a useful tool that clinicians can draw on to present evidence-based, tailored, health and risk information, which in turn can stimulate a collaborative decision-making process around which health/lifestyle factors to target and how best to achieve long-lasting behaviour change.
An interesting finding was that most participants overestimated their risk of developing AD, even after receiving their personal risk profile. This is possibly reflective of our cohort which was composed of help-seeking older adults seeking an evaluation at a memory clinic and were concerned with developing dementia. The continuing high levels of perceived risk at follow up is unsurprising given that previous research has found that even among individuals who accurately recalled their communicated AD risk, over 50% did not fully adjust their perceived risk to match the communicated risk, and that high baseline AD risk perception was the strongest predictor of overestimation of risk (29). It is also possible that participants may not have readily accepted the communicated risk after receiving lower than expected risk feedback, as seen by interview participants expressing surprise at their communicated risk.
Nonetheless, this has implications for supporting clinicians to communicate AD risk information to people with MCI. A survey found that 90% of neurologists said they counselled MCI patients on their risk of AD in general terms but only 60% communicated AD risk in numeric terms (30). Our findings provide preliminary support that patients with MCI or SCD understood numeric AD risk information and risk factors even when it is communicated online without support, and that the multifaceted approach with a clear visual representation and accompanying explanatory text may have facilitated understanding. Clinicians are encouraged to discuss numeric risk estimates with patients with visual aids, explain how these are estimated from risk models, and explore reasons for discord to improve risk appraisal.
There are several important considerations in interpreting the findings. The study sample was a well-educated inner-city cohort who have expressed concern about their memory and were highly motivated to seek help. Participants already knew that they did not have dementia. Their relatively positive reactions to their personal risk profile may reflect their personal interest in managing their brain health or because they were reassured of having low risk. It is unclear whether a general population or primary care sample, who had subjective memory concerns and not assessed for AD, would have similar reactions. It should be noted that a number of older adults approached to take part in the trial declined due to not wanting to know their risk or because they felt overwhelmed. Further research is needed to explore these concerns about knowing one’s personal risk.
The majority of our sample had low risk of developing AD, and it is unknown how moderate or high risk adults would respond to their personal dementia risk profile. There is some indication that high risk individuals, such as those who screen positive to genetic biomarkers, have heightened test-related distress (18). In order for dementia risk profiles to be widely and safely distributed to older adults in public health programs, particularly if they are to be delivered online in the absence of immediate clinical support, it is important to understand how moderate or high risk older adults react to their personal dementia risk profile and to monitor any potential adverse reactions. Finally, this was a pilot trial with a small sample size and short-term follow up. Longer-term follow up and randomised controlled trials to examine effects of communication of personal risk of developing AD are required.
The application of dementia risk algorithms to identify those at risk and to promote and encourage risk reduction behaviour is still in its early stages. This study provides preliminary support for the utility of using risk models that incorporate accessible and potentially modifiable risk factors to communicate personal dementia risk profiles to at-risk older adults.
Acknowledgements: The authors would like to acknowledge Professor Kaarin J. Anstey and Dr Sarang Kim for permission to use the ANU Alzheimer’s Disease Risk Index and their advice on adapting it to the Healthy Brain Ageing clinic measures. We thank the participants who have helped make this research possible.
Funding: This study was supported by a Dementia Australia Research Foundation project grant award. 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.
Ethical standards: The study was approved by the University of Sydney Human Research Ethics Committee (protocol number: 2019/669) and registered with the Australian New Zealand Clinical Trials Registry (ANZCTRN12619001242112p).
Conflict of interest: The authors confirm that there are no known conflicts of interest associated with this publication.
1. Norton S, Matthews FE, Barnes DE, Yaffe K, Brayne C. Potential for primary prevention of Alzheimer’s disease: an analysis of population-based data. The Lancet Neurology. 2014;13(8):788-794 http://dx.doi.org/10.1016/S1474-4422(14)70136-X.
2. Isaacson RS, Hristov H, Saif N, et al. Individualized clinical management of patients at risk for Alzheimer’s dementia. Alzheimer’s & dementia : the journal of the Alzheimer’s Association. 2019;15(12):1588-1602 https://doi.org/10.1016/j.jalz.2019.08.198.
3. 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 http://dx.doi.org/10.1016/ S0140-6736(15)60461-5.
4. Kim S, Sargent-Cox KA, Anstey KJ. A qualitative study of older and middle-aged adults’ perception and attitudes towards dementia and dementia risk reduction. J Adv Nurs. 2015;71(7):1694-1703 https://doi.org/10.1111/jan.12641.
5. Janz NK, Becker MH. The Health Belief Model: A Decade Later. Health Educ Q. 1984;11:1-47 https://doi.org/10.1177/109019818401100101.
6. 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. PLoS One. 2018;13(4):e0196085 https://doi.org/10.1371/journal.pone.0196085.
7. Low L-F, Anstey KJ. Dementia literacy: Recognition and beliefs on dementia of the Australian public. Alzheimer’s & Dementia. 2009;5(1):43-49 https://doi.org/10.1016/j.jalz.2008.03.011.
8. Smith BJ, Ali S, Quach H. Public knowledge and beliefs about dementia risk reduction: a national survey of Australians. BMC Public Health. 2014;14(1):661 https://doi.org/10.1186/1471-2458-14-661.
9. Smith BJ, Ali S, Quach H. The motivation and actions of Australians concerning brain health and dementia risk reduction. Health Promot J Austr. 2015;26(2):115-121 http://dx.doi.org/10.1071/HE14111.
10. Hou X-H, Feng L, Zhang C, Cao X-P, Tan L, Yu J-T. Models for predicting risk of dementia: a systematic review. J Neurol Neurosurg Psychiatry. 2018:jnnp-2018-318212 http://dx.doi.org/10.1136/jnnp-2018-318212.
11. Anstey KJ, Cherbuin N, Herath PM. Development of a New Method for Assessing Global Risk of Alzheimer’s Disease for Use in Population Health Approaches to Prevention. Prevention Science. 2013;14(4):411-421 http://dx.doi.org/10.1007/s11121-012-0313-2.
12. 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;9(1):e86141 https://doi.org/10.1371/journal.pone.0086141.
13. LaMonica HM, English A, Hickie IB, et al. Examining Internet and eHealth Practices and Preferences: Survey Study of Australian Older Adults With Subjective Memory Complaints, Mild Cognitive Impairment, or Dementia. J Med Internet Res. 2017;19(10):e358 https://doi.org/10.2196/jmir.7981.
14. Sindi S, Calov E, Fokkens J, et al. The CAIDE Dementia Risk Score App: The development of an evidence-based mobile application to predict the risk of dementia. Alzheimer’s & Dementia : Diagnosis, Assessment & Disease Monitoring. 2015;1(3):328-333 http://dx.doi.org/10.1016/j.dadm.2015.06.005.
15. Sheffrin M, Stijacic Cenzer I, Steinman MA. Desire for predictive testing for Alzheimer’s disease and impact on advance care planning: a cross-sectional study. Alzheimers Res Ther. 2016;8(1):55. https://doi.org/10.1186/s13195-016-0223-9
16. Wikler EM, Blendon RJ, Benson JM. Would you want to know? Public attitudes on early diagnostic testing for Alzheimer’s disease. Alzheimers Res Ther. 2013;5(5):43 http://dx.doi.org/10.1186/alzrt206.
17. Roberts JS, McLaughlin SJ, Connell CM. Public beliefs and knowledge about risk and protective factors for Alzheimer’s disease. Alzheimer’s & Dementia. 2014;10(5):S381-S389 http://dx.doi.org/10.1016/j.jalz.2013.07.001.
18. 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. Alzheimers Res Ther. 2016;8(1):46 http://dx.doi.org/10.1186/s13195-016-0212-z.
19. Rimer BK, Halabi S, Sugg Skinner C, et al. Effects of a mammography decision-making intervention at 12 and 24 months. Am J Prev Med. 2002;22:247-257 https://doi.org/10.1016/S0749-3797(02)00417-8.
20. Heger I, Deckers K, van Boxtel M, et al. Dementia awareness and risk perception in middle-aged and older individuals: baseline results of the MijnBreincoach survey on the association between lifestyle and brain health. BMC Public Health. 2019;19(1):678 https://doi.org/10.1186/s12889-019-7010-z.
21. Livingston G, Sommerlad A, Orgeta V, et al. Dementia prevention, intervention, and care. The Lancet. 2017;390(10113):2673-2734 https://doi.org/10.1016/S0140-6736(17)31363-6.
22. Kim S, Sargent-Cox K, Cherbuin N, Anstey KJ. Development of the motivation to change lifestyle and health Behaviours for Dementia Risk Reduction Scale. Dement Geriatr Cogn Dis Extra. 2014;4(2):172-183 https://doi.org/10.1159/000362228.
23. Kessler R, Andrews G, Colpe L, et al. Short screening scales to monitor population prevalences and trends in non-specific psychological distress. Psychol Med. 2002;32(06):959-976 https://doi.org/10.1017/S0033291702006074.
24. Perini SJ, Slade T, Andrews G. Generic effectiveness measures: Sensitivity to symptom change in anxiety disorders. J Affect Disord. 2006;90:123-130 https://doi.org/10.1016/j.jad.2005.10.011.
25. Kinzer A, Suhr JA. Dementia worry and its relationship to dementia exposure, psychological factors, and subjective memory concerns. Applied Neuropsychology: Adult. 2016;23(3):196-204 https://doi.org/10.1080/23279095.2015.1030669.
26. Farrow M. User perceptions of a dementia risk reduction website and its promotion of behavior change. JMIR research protocols. 2013;2(1):e15-e15 https://doi.org/10.2196/resprot.2372.
27. Vrijsen J, Abu-Hanna A, Maeckelberghe ELM, et al. Uptake and effectiveness of a tailor-made online lifestyle programme targeting modifiable risk factors for dementia among middle-aged descendants of people with recently diagnosed dementia: study protocol of a cluster randomised controlled trial (Demin study). BMJ Open. 2020;10(10):e039439 http://dx.doi.org/10.1136/bmjopen-2020-039439.
28. Owusu-Addo E, Ofori-Asenso R, Batchelor F, Mahtani K, Brijnath B. Effective implementation approaches for healthy ageing interventions for older people: A rapid review. Arch Gerontol Geriatr. 2021;92:104263 https://doi.org/10.1016/j.archger.2020.104263.
29. Linnenbringer E, Roberts JS, Hiraki S, Cupples LA, Green RC. “I know what you told me, but this is what I think:” Perceived risk of Alzheimer disease among individuals who accurately recall their genetics-based risk estimate. Genet Med. 2010;12(4):219-227 https://doi.org/10.1097/GIM.0b013e3181cef9e1.
30. Roberts JS, Karlawish JH, Uhlmann WR, Petersen RC, Green RC. Mild cognitive impairment in clinical care: A survey of American Academy of Neurology members. Neurology. 2010;75(5):425-431 https://doi.org/10.1212/WNL.0b013e3181eb5872.