Abstract
This study examines the perception of US individuals about their likelihood of developing dementia and their strategies to improve or maintain memory.
Disease-preventing or disease-modifying treatments do not exist for Alzheimer disease or other dementias. Adults may be unaware of strategies to reduce their risk1 and resort to marketed but ineffective options, such as ginkgo biloba or vitamin E. While these so-called treatments are relatively inexpensive, new preventive therapies may not be. Thus, individuals overestimating their risk of developing dementia could lead to inappropriate use and excessive costs.2 This analysis explores how adults aged 50 to 64 years estimate their lifetime risk of dementia and the risk-reducing strategies they pursue.
Methods
The University of Michigan National Poll on Healthy Aging (NPHA) is a nationally representative survey of adults ages 50 to 80 years, sponsored by AARP and Michigan Medicine. The NPHA uses KnowledgePanel (Ipsos Public Affairs LLC), a probability-based panel of the civilian, noninstitutionalized US population. This survey was fielded in October 2018; questions for this analysis were asked of respondents aged 50 to 64 years. The University of Michigan institutional review board reviewed this study and deemed it exempt from human subjects review because it was a study of deidentified respondents. The requirement for informed consent was therefore waived.
Along with demographic information and self-reported health status, respondents were asked, “How likely are you to develop dementia during your lifetime?” (with the possible answers being “very likely,” “somewhat likely,” and “not likely”); “Have you ever discussed ways to prevent dementia with your doctor?”; and about 4 specific strategies to “maintain or improve your memory” (with the possible answers being yes or no).
The outcome of interest was perceived likelihood of developing dementia (very/somewhat likely vs not likely). The association of respondent characteristics with perceived likelihood of developing dementia was examined with logistic regression. The final adjusted model was used to determine the expected probability of respondents’ perceived likelihood of developing dementia. Finally, memory-preserving strategies were examined overall and by perceived dementia likelihood.
Analyses used poststratification weights to draw national inferences and were performed using Stata version 15.1 (StataCorp LLC). A 2-tailed P < .05 was considered statistically significant.
Results
Among 1019 respondents aged 50 to 64 years, 48.5% (95% CI, 45.3%-51.7%) reported they were at least somewhat likely to develop dementia (those answering “somewhat”: 44.3% [95% CI, 41.1%-47.5%]; those answering “very,” 4.2% [95% CI, 3.1%-5.8%]). In adjusted analyses, non-Hispanic black respondents were significantly less likely to believe they may develop dementia (adjusted odds ratio, 0.51 [95% CI, 0.32-0.81]; P = .01; Table). Respondents who rated their mental health as fair or poor reported a higher likelihood of developing dementia (adjusted odds ratio, 2.30 [95% CI, 1.19-4.47]; P = .01) although those with similarly rated physical health did not (adjusted odds ratio, 1.46 [95% CI, 0.93-2.28]; P = .10).
Table. Respondent Characteristics and Associations With Perceived Likelihood of Developing Dementia.
| Characteristic | Respondents, No. | Respondents, No. (%) | Odds Ratio (95% CI) | P Value | Adjusted Odds Ratio (95% CI)a | P Value | Probability (95% CI)b | |
|---|---|---|---|---|---|---|---|---|
| Very or Somewhat Likely (n = 444) | Unlikely (n = 576) | |||||||
| Age range, y | ||||||||
| 50-54 | 303 | 136 (44.9) | 167 (55.1) | 1 [Reference] | NA | 1 [Reference] | NA | 44.9 (39.1-50.7) |
| 55-59 | 388 | 199 (51.0) | 189 (49.0) | 1.27 (0.93-1.75) | .13 | 1.23 (0.89-1.70) | .21 | 49.9 (44.9-54.9) |
| 60-64 | 328 | 162 (49.5) | 166 (50.5) | 1.20 (0.87-1.66) | .28 | 1.24 (0.88-1.74) | .22 | 50.0 (44.3-55.6) |
| Sex | ||||||||
| Male | 499 | 228 (45.7) | 271 (54.3) | 1 [Reference] | NA | 1 [Reference] | NA | 45.4 (41.0-49.8) |
| Female | 520 | 269 (51.1) | 251 (48.9) | 1.24 (0.96-1.61) | .0 | 1.26 (0.96-1.64) | .09 | 50.9 (46.3-55.4) |
| Race/ethnicity | ||||||||
| Non-Hispanic | ||||||||
| White | 753 | 385 (51.1) | 368 (48.9) | 1 [Reference] | NA | 1 [Reference] | NA | 51.0 (47.4-54.7) |
| Black | 93 | 36 (37.1) | 57 (62.9) | 0.56 (0.36-0.89) | .01 | 0.51 (0.32-0.81) | .004 | 35.3 (25.8-44.8) |
| Hispanic | 101 | 46 (47.5) | 55 (52.5) | 0.86 (0.56-1.32) | .50 | 0.85 (0.54-1.33) | .47 | 47.1 (37.0-57.1) |
| Other, non-Hispanic | 72 | 30 (43.6) | 42 (56.4) | 0.74 (0.42-1.32) | .30 | 0.78 (0.44-1.39) | .40 | 45.2 (32.0-58.4) |
| Education | ||||||||
| High school or less | 337 | 177 (51.8) | 160 (48.2) | 1 [Reference] | NA | 1 [Reference] | NA | 49.9 (44.2-55.6) |
| Some college | 339 | 164 (47.7) | 175 (52.3) | 0.85 (0.62-1.16) | .30 | 0.87 (0.62-1.21) | .39 | 46.4 (41.1-51.8) |
| Bachelor’s degree or higher | 343 | 156 (45.0) | 187 (55.0) | 0.76 (0.58-1.21) | .09 | 0.91 (0.64-1.30) | .62 | 47.7 (42.0-53.4) |
| Total annual household income, $ | ||||||||
| <30 000 | 150 | 79 (51.4) | 71 (48.6) | 1 [Reference] | NA | 1 [Reference] | NA | 45.9 (37.3-54.6) |
| 30 000-59 999 | 191 | 98 (50.9) | 93 (49.1) | 0.98 (0.62-1.53) | .93 | 1.10 (0.69-1.77) | .68 | 48.3 (41.0-55.6) |
| ≥60 000 | 678 | 320 (46.9) | 358 (53.1) | 0.84 (0.58-1.21) | .34 | 1.13 (0.74-1.73) | .57 | 48.9 (44.7-53.0) |
| Physical health status | ||||||||
| Excellent or very good | 445 | 184 (40.9) | 261 (59.1) | 1 [Reference] | NA | 1 [Reference] | NA | 42.9 (37.8-47.9) |
| Good | 413 | 223 (52.9) | 190 (47.1) | 1.63 (1.23-2.16) | .001 | 1.49 (1.10-2.01) | .01 | 52.5 (47.2-57.7) |
| Fair or poor | 157 | 87 (56.6) | 70 (43.4) | 1.89 (1.29-2.78) | .001 | 1.46 (0.93-2.28) | .10 | 52.0 (43.2-60.8) |
| Mental health status | ||||||||
| Excellent or very good | 721 | 315 (43.8) | 406 (56.2) | 1 [Reference] | NA | 1 [Reference] | NA | 44.7 (40.8-48.7) |
| Good | 234 | 140 (57.2) | 94 (42.8) | 1.71 (1.25-2.35) | .001 | 1.48 (1.05-2.08) | .03 | 54.3 (47.3-61.3) |
| Fair or poor | 60 | 39 (66.3) | 21 (33.7) | 2.53 (1.41-4.52) | .002 | 2.30 (1.19-4.47) | .01 | 64.6 (50.5-78.7) |
Abbreviation: NA, not applicable.
A logistic regression model (0 = unlikely; 1 = somewhat/very likely) was adjusted for all respondent characteristics presented. For example, compared with non-Hispanic white respondents, non-Hispanic black respondents have 0.51 odds of believing they are somewhat or very likely to develop dementia.
Reflects margins from the adjusted logistic regression model. For example, if the response pattern of non-Hispanic black respondents were applied to all poll respondents, respondents overall would predict a 35.3% likelihood of developing dementia.
Only 5.2% (95% CI, 4.0%-6.8%) of respondents had discussed dementia prevention with their physician (Figure). In contrast, 31.6% (95% CI, 28.7-34.6) endorsed using fish oil or ω-3 fatty acids, and 39.2% (95% CI, 36.1%-42.4%) used other vitamins or supplements. Discussion with a physician was the only strategy that varied by perceived likelihood of developing dementia, being more common among respondents with a higher perceived likelihood (7.1% [95% CI 5.1%-9.8%]) vs those with a lower perceived likelihood (3.6% [95% CI 2.2%-5.7%]; P = .02).
Figure. Actions Taken to Prevent Memory Loss by Perceived Likelihood of Developing Dementia.
The percentage of poll respondents who endorsed specific strategies in response to the following question: “Do you take or do any of the following to maintain or improve your memory?” Responses are grouped by perceived likelihood of developing dementia (somewhat/very likely vs not likely). A χ2 test was used to compare particular strategies endorsed by perceived likelihood of developing dementia. All comparisons were nonsignificant with the exception of discussion with a physician, which was endorsed more frequently by those who believed they were at least somewhat likely to develop dementia (7.1% [95% CI, 5.1%-9.8%] vs 3.6% [95% CI, 2.2%-5.7%]; P = .02).
Discussion
Among US adults aged 50 to 64 years in this poll, nearly 50% believe they are at least somewhat likely to develop dementia. Non-Hispanic black American individuals have a higher prevalence of dementia than other racial or ethnic groups,3 but in this survey, they perceived their risk as lower relative to other groups. Those with fair to poor physical health did not accurately perceive that their likelihood of developing dementia was potentially higher than respondents with very good or excellent physical health. In contrast, fair to poor mental health had the largest association with perceived likelihood of dementia, even though less evidence suggests that poor mental health is causally linked with dementia.4
Poll respondents report engaging in a variety of strategies to maintain or improve memory that are not evidence based. While managing chronic medical conditions, such as diabetes or cardiovascular disease, could reduce dementia risk,4 few respondents appear to have discussed this with their physician.
Given repeated failures of disease-preventing or disease-modifying treatments for dementia, interest in treatment and prevention have shifted earlier in the disease process. Adults in middle age may not accurately estimate their risk of developing dementia, which could lead to both overuse and underuse if preclinical dementia treatments become available. Policy and physicians should emphasize current evidence-based strategies of managing lifestyle and chronic medical conditions to reduce the risk of dementia.5
References
- 1.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. doi: 10.1111/jan.12641 [DOI] [PubMed] [Google Scholar]
- 2.Langa KM, Burke JF. Preclinical Alzheimer disease—early diagnosis or overdiagnosis? JAMA Intern Med. 2019. doi: 10.1001/jamainternmed.2019.2629 [DOI] [PubMed] [Google Scholar]
- 3.Langa KM, Larson EB, Crimmins EM, et al. A comparison of the prevalence of dementia in the United States in 2000 and 2012. JAMA Intern Med. 2017;177(1):51-58. doi: 10.1001/jamainternmed.2016.6807 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.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 [DOI] [PubMed] [Google Scholar]
- 5.Lourida I, Hannon E, Littlejohns TJ, et al. Association of lifestyle and genetic risk with incidence of dementia. JAMA. 2019;322(5):430-437. doi: 10.1001/jama.2019.9879 [DOI] [PMC free article] [PubMed] [Google Scholar]