Abstract
Background:
As Alzheimer’s disease (AD) biomarker testing becomes more widely available, adults may opt to learn results. Considering potential reactions to learning biomarker results can guide pre- and post-biomarker testing education and counseling programs.
Methods:
Cognitively healthy adults enrolled in observational Alzheimer’s research responded to a telephone survey about learning AD risk information (n=334; 44% Black or African American; mean age=64.9±7.0). Multiple linear regression models tested if contextual factors predicted anticipated psychological impact (distress, stigma, cognitive symptoms) or behavior change (planning, risk-reduction). Secondary analyses tested for differences in relationships by racial identity.
Results:
Internal health locus of control, concern about AD, self-identified gender, education, family dementia history, and belief in AD modifiability predicted anticipated psychological impact. Concern about AD, age, racial identity, belief in AD modifiability, research attitudes, and exposure to brain health-related social norms predicted anticipated behavior change. For Black respondents there were no gender differences in anticipated distress, whereas there were stronger relationships between health locus of control, brain health social norms, and education on outcomes compared to White respondents.
Conclusion:
Results may inform personalized and culturally tailored biomarker testing education and counseling to minimize psychological impacts and increase behavior change related to learning AD risk information.
Keywords: Alzheimer’s disease, Biomarker Disclosure, Risk Communication
INTRODUCTION
Improved characterization of preclinical Alzheimer’s disease (AD) coupled with available biomarker tests to detect early brain changes is increasing opportunities for cognitively healthy adults to learn AD biomarker results. Currently, there is ethical justification to provide AD biomarker results when participants enroll in clinical trials requiring biomarker testing for screening1. Outside of clinical trials, though, providing biomarker test results to cognitively unimpaired adults is less common. Further development of available, disease-modifying treatments coupled with advancements in less invasive, cheaper testing, such as blood-based biomarkers may increase the clinical utility of AD biomarker testing and expand accessibility for adults at-risk for dementia.
While the utility of preclinical AD biomarker testing remains uncertain in the absence of treatments, prior studies of unimpaired clinical trial enrollees demonstrate learning AD biomarker results can have personal utility., Individuals may plan for their future or reduce modifiable dementia risk factors, including making changes to exercise and diet routines2. There is also an emotional impact to learning an ‘elevated’ or ‘high-risk’ result. Prior studies report initial increases in psychological distress that resolves over the course of six month follow-up3. However, these results may not directly translate to participants in observational cohorts. While clinical trials provide a clear next step (e.g., enrollment into a clinical trial), there is not necessarily an actionable next step for those learning results in observational studies. Moreover, pre-disclosure education may differ from symptomatic populations in that there is more ambiguity and heterogeneity regarding individualized prognosis for those without symptoms. Importantly, because patterns of clinical trial enrollment in the US have been overwhelmingly White, our understanding of reactions to learning biomarker results comes from studies including mostly college-educated, White samples. Further, as few unimpaired participants in observational cohorts receive biomarker results, it is challenging to directly study these reactions. Thus, we asked about anticipated reactions to learning biomarker results of participants enrolled in our racially diverse cohorts at the Wisconsin Alzheimer’s Disease Research Center (Wisconsin ADRC) and the Wisconsin Registry for Alzheimer’s Prevention (WRAP).
In light of this knowledge gap, we administered a telephone survey4, to over 300 participants in the Wisconsin ADRC and WRAP cohorts, including 44% Black or African American participants enrolled through the African American’s Fighting Alzheimer’s in Mid-life (AA-FAIM) study. The overall objective was to identify contextual factors associated with willingness to enroll in biomarker studies that disclose results and subsequent reactions related to learning results. We previously reported that more positive attitudes towards medical research were associated with increased willingness to enroll in biomarker studies that disclose results, whereas other contextual factors did not predict willingness5. Additionally, by comparing willingness to enroll in biomarker studies that did or did not share results, we found that willingness of Black participants to enroll was potentially enhanced by opportunities to learn results5. Our prior studies also showed that Black participants willing to undergo biomarker testing and learn results were motivated by a desire to know their results and support research, and those less willing were concerned about potential negative psychological consequences of learning about increased risk, doubt about the usefulness of testing, and worry about potential physical harms of testing6. The aims of the current study were to (1) determine which contextual factors were related to anticipated psychological and behavioral reactions to learning high-risk AD biomarker results and (2) determine whether the relationships between the factors and anticipated reactions to learning high-risk AD biomarker results differed for Black and White participants. This information can inform biomarker testing and disclosure protocols for longitudinal cohorts and provide evidence for when and how to communicate results. Additionally, understanding potential reactions or varied risks and benefits of learning results from minoritized groups is needed to inform culturally tailored education materials and post-disclosure supports.
METHODS
Participants
We recruited 400 participants enrolled in longitudinal observational research cohorts (i.e. Wisconsin ADRC or WRAP) to participate in the survey. Inclusion criteria for recruitment included age 45–89, self-identified ethnicity/race as either non-Hispanic Black or African American or non-Hispanic White, and no research or clinical diagnosis of mild cognitive impairment or dementia based on the National Institutes on Aging-Alzheimer’s Association (NIA-AA) diagnostic criteria7,8. We consulted with the University of Wisconsin Community Advisors on Research Design & Strategies (CARDS) to develop recruitment materials and strategies based on feedback from a community focus group with diverse racial and socioeconomic backgrounds. Eligible participants were mailed a recruitment letter and information sheet and then contacted by phone. If interested in participating, the information sheet was reviewed, and oral informed consent was obtained. The survey was administered by trained UWSC interviewers using Computer-Assisted Telephone Interview (CATI). Participants were compensated $25 for completing the survey. This study was approved by the University of Wisconsin-Madison Institutional Review Board (IRB 2019–0248).
Measures
The survey included measures assessing experience with and concern about AD9, perceived health, health locus of control (LOC)10, brain health social norms11, research attitudes12, and chronic experiences of perceived discrimination due to race, ethnicity, self-identified gender, age, or other interpersonal characteristics13. We previously reported results of exploratory and confirmatory factor analysis revealed an underlying factor structure comprising 14 items assessing anticipated psychological impact (3 factors: Distress, Stigma, Cognitive Symptoms) and 14 items assessing anticipated behavior change (2 factors: Planning, Dementia risk-reduction) associated with learning AD biomarker results and confirmed measurement invariance across White and Black or African American participants in our cohort4. These 28 items were either adapted from existing questionnaires14,15 or developed by the study team. For eight items, participants were asked to rate the importance of each item as a reason to learn their biomarker results. For the remaining 20 items, participants were asked to imagine how they might respond to learning a result that meant they were at a “higher risk of developing Alzheimer’s.” Response options were on a 5-point Likert scale for agreement (not at all, a little, somewhat, very, extremely).
Statistical Analysis
Descriptive statistics were used to evaluate demographic characteristics and response patterns on the survey items. Multiple linear regression models were used to test hypotheses that contextual factors would predict psychological or behavioral reactions to learning AD biomarker results. Outcome variables included the five factor scores of planning, dementia risk-reduction, distress, stigma, and cognitive Symptoms. Each outcome was included in a separate regression model. Predictor variables included in age (centered), self-identified gender (male, female), years of education, self-identified race (non-Hispanic White or non-Hispanic Black or African American), family history of dementia (yes or no), experience with dementia, belief in AD modifiability, health LOC, brain health social norms, attitudes towards medical research, concern about AD, healthcare access, health self-rating, memory self-rating, and perceived discrimination (see Supplemental Table 1). Additionally, we examined interactions between the predictor variables and racial identity to assess racial differences in predictor-outcome relationships using multiple linear regression models. To avoid model overfitting, we utilized stepwise backward elimination to identify the best fitting model to minimize the Akaike information criterion (AIC). We forced the main effects of age, sex, education, and family history into models as we felt these could be important covariates to consider. As a further verification of our model building, we used the glmnet R package16 to perform ridge regression as a sensitivity analysis. This penalized regression method is an alternative approach to avoiding overfitting and was used as a check on the backward selection. Using glmnet, lambda was selected as the value that minimized the cross-validation prediction error rate. For the ridge regression, predictors and outcomes were standardized, covariates were centered, and variance inflation was evaluated.
Results
Participant characteristics and descriptive statistics
A total of 334 participants (186 non-Hispanic White, 148 non-Hispanic Black or African American) completed the survey (see Table 1). Participants were about 65 years old (range=45–84), female (74%), and highly educated (58% with bachelor’s degree). Over half of participants (62%) had a family history of dementia, and most had either spent time with (26%) or cared for (69%) a person with dementia. On average, participants were “somewhat” concerned about AD (mean=3.01; range: 1–5) and “somewhat” believed modifiable factors contributed to AD (mean=3.03; range: 1–5). On average, participants rated their overall health and memory as “good” (means 3.32 and 3.27, respectively; range: 1–5) and reported high internal health LOC (mean 3.72; range: 1–5). Most participants had a healthcare provider they saw routinely (94%) and reported positive attitudes towards research (mean RAQ-7=29.9 out of 35). Most participants responded “sometimes,” when asked about their friends and family’s engagement in brain health-related activities (mean 2.72; range: 1–5). On average, Black or African American respondents had fewer years of education, lower internal health LOC, lower perceived health and memory, greater family history of dementia, and experienced a higher frequency of discrimination than White participants.
Table 1.
Sample and descriptive characteristics
| Total (n=334) |
Black or African American (n=148) | White (n=186) | p-value | |
|---|---|---|---|---|
| Demographics/personal factors | ||||
| Age (mean, standard deviation) | 64.9 (7.0) | 65.3 (8.4) | 65.1 (7.64) | 0.67 |
| Gender (n, % women) | 248 (74%) | 107 (72.3%) | 141 (75.8%) | 0.55 |
| Education (mean, standard deviation) | 15.5 (2.4) | 14.9 (2.4) | 16.0 (2.3) | <0.001 |
| Perceived health (1=poor, 5=excellent) |
3.32 (0.9) | 2.93 (0.9) | 3.63 (0.9) | <0.001 |
| Perceived memory (1=poor, 5=excellent) |
3.27 (0.8) | 3.09 (0.9) | 3.41 (0.7) | <0.001 |
| Health locus of control (1=not at all, 5=a tremendous amount) |
3.72 (0.5) | 3.66 (0.5) | 3.77 (0.5) | 0.04 |
| Experiential factors | ||||
| Healthcare access (n, % yes) | 313 (94%) | 140 (95%) | 173 (93%) | 0.71 |
| Attitudes towards medical research1 (Scale range = 7–35) |
29.9 (3.5) | 29.6 (3.7) | 30.2 (3.3) | 0.16 |
| Brain health social norms (1=none, 5=very often) |
2.72 (0.8) | 2.78 (0.8) | 2.67 (0.7) | 0.20 |
| Perceived discrimination2 (Scale range = 9–45) |
16.6 (6.4) | 21.1 (6.3) | 13.0 (3.6) | <0.001 |
| Alzheimer’s-related factors | ||||
| Family history of dementia (n, % positive) | 208 (62.3%) | 75 (50.7%) | 133 (71.5%) | <0.001 |
| Experience with dementia None (n,%) Spent time with (n,%) Cared for (n,%) |
18 (5.4%) 86 (25.7%) 229 (68.6%) |
10 (6.8%) 35 (23.6%) 102 (68.9%) |
8 (4.3%) 51 (27.4%) 127 (68.3%) |
0.50 |
| Concern about Alzheimer’s disease (1=Not at all, 5=Extremely) |
3.01 (1.2) | 2.86 (1.3) | 3.12 (1.1) | 0.06 |
| Belief in AD modifiability (1=Not at all, 5=Extremely) |
3.03 (0.8) | 3.10 (0.9) | 2.98 (0.7) | 0.19 |
| Anticipated reactions to learning biomarker results | ||||
| Distress factor score | −0.000000628 (0.954) | −0.0235 (1.04) | 0.0187 (0.885) | 0.69 |
| Stigma factor score | 0.000000466 (0.884) | 0.143 (0.957) | −0.113 (0.807) | 0.01 |
| Cognitive symptoms factor score | 0.0000000623 (0.940) | 0.176 (1.00) | −0.140 (0.864) | <0.001 |
| Planning factor score | 0.000000239 (0.965) | 0.229 (0.880) | −0.182 (0.992) | <0.001 |
| Risk-Reduction factor | −0.000000266 (0.933) | 0.177 (0.827) | −0.141 (0.990) | <0.05 |
Predictors of anticipated psychological impact – Distress, Stigma, and Cognitive Symptoms
Female gender (β=0.32, p<.01), fewer years of education (β=−0.05, p<.05), lower internal health LOC (β=−0.39, p=.001), and higher concern about AD (β=0.23, p<.001) significantly predicted anticipated distress. No family history of dementia (β=−0.24, p<.05), higher belief in modifiable causes of AD (β=0.13, p<.05), lower internal health LOC (β=−0.45, p<.001), and higher concern about AD (β=0.19, p<.001) significantly predicted anticipated stigma. Lower internal health LOC (β=−0.33, p<.01) and higher concern about AD (β=0.16, p<.001) predicted anticipated cognitive symptoms (Table 2). Age, racial identity, social norms, research attitudes, experience with dementia, healthcare access, health self-rating, memory self-rating, or perceived discrimination were not associated with anticipated distress, stigma, or cognitive symptoms. Overall, lower internal health LOC and higher concern about AD were the most consistent predictors of anticipated psychological impact after learning high-risk biomarker results.
Table 2.
Primary analyses - Statistical parameters from multiple linear regression models
| Distress Factor | Stigma Factor | Cognitive symptoms Factor | Planning Factor | Risk-Reduction Factor | |
|---|---|---|---|---|---|
| Predictors |
Estimates
(CI) |
Estimates
(CI) |
Estimates
(CI) |
Estimates
(CI) |
Estimates
(CI) |
| (Intercept) | 0.43 (−1.04 – 1.90) |
0.47 (−0.90 – 1.83) |
0.40 (−1.05 – 1.86) |
−1.82 (−3.29 – −0.36) |
−2.80 (−4.20 – −1.39) |
| Age (centered) | −0.01 (−0.02 – 0.01) |
−0.00 (−0.02 – 0.01) |
0.01 (−0.00 – 0.02) |
−0.01
*
(−0.03 – −0.00) |
−0.02
***
(−0.04 – −0.01) |
| Gender (Female) |
0.32
**
(0.09 – 0.56) |
0.18 (−0.03 – 0.40) |
0.02 (−0.21 – 0.25) |
−0.04 (−0.27 – 0.20) |
0.05 (−0.17 – 0.28) |
| Years of Education |
−0.05
*
(−0.10 – −0.01) |
−0.03 (−0.07 – 0.01) |
−0.00 (−0.05 – 0.04) |
−0.03 (−0.07 – 0.02) |
−0.02 (−0.06 – 0.03) |
| Race (Black) | −0.15 (−0.43 – 0.12) |
0.12 (−0.13 – 0.37) |
0.12 (−0.15 – 0.39) |
0.37
**
(0.10 – 0.64) |
0.29
*
(0.03 – 0.55) |
| Health Self-rating | 0.00 (−0.13 – 0.14) |
−0.01 (−0.13 – 0.12) |
−0.05 (−0.18 – 0.08) |
0.01 (−0.13 – 0.14) |
−0.10 (−0.23 – 0.03) |
| Memory Self-rating | 0.01 (−0.14 – 0.15) |
0.05 (−0.08 – 0.19) |
−0.12 (−0.27 – 0.02) |
−0.04 (−0.18 – 0.11) |
0.07 (−0.07 – 0.21) |
| Health Locus of Control |
−0.39
***
(−0.60 – −0.17) |
−0.45
***
(−0.65 – −0.25) |
−0.33
**
(−0.55 – −0.12) |
−0.12 (−0.33 – 0.10) |
0.13 (−0.08 – 0.34) |
| Healthcare Access | 0.18 (−0.26 – 0.62) |
0.17 (−0.23 – 0.58) |
−0.02 (−0.45 – 0.41) |
−0.09 (−0.53 – 0.34) |
−0.10 (−0.52 – 0.32) |
| Research Attitudes | 0.02 (−0.00 – 0.05) |
0.01 (−0.02 – 0.04) |
0.02 (−0.01 – 0.04) |
0.06*** (0.03 – 0.09) |
0.04* (0.01 – 0.06) |
| Social Norms | −0.10 (−0.24 – 0.03) |
−0.10 (−0.23 – 0.02) |
−0.02 (−0.15 – 0.11) |
−0.03 (−0.17 – 0.10) |
0.14* (0.01 – 0.27) |
| Discrimination | 0.01 (−0.01 – 0.03) |
0.01 (−0.01 – 0.03) |
0.02 (−0.00 – 0.04) |
0.01 (−0.01 – 0.03) |
−0.00 (−0.02 – 0.02) |
| Family History | −0.14 (−0.39 – 0.10) |
−0.24
*
(−0.46 – −0.01) |
−0.05 (−0.29 – 0.19) |
−0.07 (−0.31 – 0.17) |
0.01 (−0.22 – 0.25) |
| Experience with Dementia | |||||
| Spent time with | 0.22 (−0.28 – 0.72) |
0.27 (−0.19 – 0.74) |
−0.01 (−0.50 – 0.49) |
−0.14 (−0.63 – 0.36) |
−0.11 (−0.59 – 0.36) |
| Cared for | 0.07 (−0.42 – 0.57) |
0.26 (−0.20 – 0.71) |
−0.05 (−0.54 – 0.44) |
−0.20 (−0.69 – 0.29) |
−0.15 (−0.62 – 0.32) |
| Concern About AD | 0.23*** (0.14 – 0.32) |
0.19
***
(0.10 – 0.27) |
0.16
***
(0.07 – 0.26) |
0.15*** (0.06 – 0.25) |
0.14** (0.05 – 0.23) |
| Belief in AD modifiability | 0.08 (−0.05 – 0.20) |
0.13
*
(0.01 – 0.25) |
0.09 (−0.03 – 0.22) |
0.22
***
(0.09 – 0.34) |
0.27
***
(0.15 – 0.39) |
| Overall Model R2 (R2 Adjusted) | 0.19 (0.14) | 0.19 (0.15) | 0.17 (0.13) | 0.21 (0.17) | 0.24 (0.19) |
p<.05,
p<.01,
p≤.001
Predictors of anticipated behavior change – Planning and Risk-Reduction
Younger age (β=−0.01, p<.05), Black or African American racial identity (β=0.37, p<.01), higher belief in AD modifiability (β=0.22, p<.001), more positive attitudes towards medical research (β=0.06, p<.001), and greater concern about AD (β=0.15, p<.001) were associated with anticipated planning behaviors (Table 2). Similarly, younger age (β=−0.02, p <.001), Black or African American racial identity (β=0.29, p<.05), higher belief in AD modifiability (β=0.27, p<.001), more positive attitudes towards medical research (β = 0.04, p <.05), greater concern about AD (β=0.14, p<.01), and more brain health-related social norms (β=0.14, p<.05) were associated with anticipated dementia risk-reduction behaviors. Self-identified gender, years of education, family history of dementia, experience with dementia, health LOC, healthcare access, health self-rating, memory self-rating, or perceived discrimination were not associated with anticipated planning or risk-reduction behavior changes. Overall, most consistent predictors of anticipated behavioral changes included younger age, Black or African American racial identity, higher belief in AD modifiability, more positive research attitudes, and greater concern about AD.
Variation in anticipated reactions by racial group
On average, Black or African American participants exhibited higher anticipated stigma and cognitive symptoms, as well as higher planning and risk-reduction factor scores, compared with White participants; there were no racial differences in anticipated distress (see Table 1). A significant interaction of race x gender (β=−0.15, p=.01) demonstrated no differences between Black men and women on anticipated distress, whereas White women anticipated experiencing more distress than White men. Additionally, a significant race x health LOC interaction (β=0.26, p=.01) indicated that higher internal health LOC was associated with less anticipated distress for Black respondents, while there was no relationship between anticipated distress and health LOC for White respondents. For the outcome of anticipated cognitive symptoms, a significant interaction between race and social norms (β=0.25, p<.001) showed that greater exposure to friends/family engaging in brain health activities was associated with more anticipated cognitive symptoms in White respondents, but fewer anticipated cognitive symptoms in Black respondents. For the outcome of planning, significant interactions of race x education (β=−0.06, p<.01) demonstrated that Black respondents with higher education anticipated more planning behavior, whereas White respondents with higher education anticipated less planning behavior. Additionally, a race by perceived discrimination interaction (β=0.03, p<.05) indicated that White respondents who experienced greater discrimination anticipated more planning behavior, whereas there was no difference by discrimination experienced for Black respondents. No significant interaction effects were observed for outcomes of risk-reduction and stigma. Secondary models are displayed in Table 3 and significant interaction results are displayed in Figure 1. Results from sensitivity analyses using ridge regression models demonstrated similar patterns, supporting that the backward selection models were appropriate (see Supplemental Table 2).
Table 3.
Secondary analyses - Statistical parameters from backwards regression models
| Distress Factor | Stigma Factor | Cognitive symptoms Factor | Planning Factor | Risk-Reduction Factor | |
|---|---|---|---|---|---|
| Predictors |
Estimates
(CI) |
Estimates
(CI) |
Estimates
(CI) |
Estimates
(CI) |
Estimates
(CI) |
| (Intercept) | 1.60 (0.59 – 2.61) |
0.96 (−0.05 – 1.97) |
0.22 (−1.17 – 1.61) |
−1.96 (−3.34 – −0.58) |
−2.63 (−3.91 – −1.35) |
| Age (Centered) | −0.01 (−0.02 – 0.01) |
−0.00 (−0.02 – 0.01) |
0.01 (−0.00 – 0.02) |
−0.01
*
(−0.03 – −0.00) |
−0.02
***
(−0.03 – −0.01) |
| Gender (Female) |
−0.12
*
(−0.24 – −0.01) |
−0.09 (−0.19 – 0.02) |
0.00 (0.11 – 0.12) | 0.04 (−0.07 – 0.16) | −0.02 (−0.13 – 0.09) |
| Years of Education | −0.04 (−0.08 – 0.00) |
−0.02 (−0.06 – 0.02) |
−0.00 (−0.05 – 0.04) |
−0.02 (−0.06 – 0.02) |
−0.01 (−0.05 – 0.04) |
| Race (Black) |
−0.98
*
(−1.73 – −0.22) |
−0.47
*
(−0.85 – −0.10) |
−0.75
***
(−1.12 – −0.37) |
−0.27 (−1.39 – 0.86) |
−0.48 (−1.41 – 0.44) |
| Health Self-rating | -- | -- | -- | -- | −0.09 (−0.20 – 0.03) |
| Memory Self-rating | -- | -- | −0.12 (−0.25 – 0.00) |
-- | -- |
| Health Locus of Control (LOC) |
−0.40
***
(−0.61 – −0.19) |
−0.45
***
(−0.64 – −0.25) |
−0.35
***
(−0.55 – −0.15) |
−0.18 (−0.39 – 0.02) |
0.07 (−0.13 – 0.28) |
| Healthcare Access | -- | -- | -- | -- | -- |
| Research Attitudes | -- | -- | 0.02 (−0.01 – 0.05) |
0.06
***
(0.03 – 0.09) |
0.04
**
(0.01 – 0.07) |
| Social Norms | −0.09 (−0.22 – 0.04) |
−0.09 (−0.22 – 0.03) |
0.03 (−0.10 – 0.16) |
-- | 0.12 (−0.01 – 0.25) |
| Discrimination | 0.02 (−0.00 – 0.04) |
0.02 (−0.00 – 0.04) |
0.01 (−0.01 – 0.03) |
-- | |
| Family History | 0.04 (−0.08 – 0.16) |
0.10 (−0.01 – 0.21) |
0.03 (−0.08 – 0.14) |
0.07 (−0.04 – 0.18) |
0.00 (−0.10 – 0.11) |
| Experience with Dementia | |||||
| Spent time with | −0.07 (−0.39 – 0.24) |
−0.19 (−0.49 – 0.10) |
-- | -- | -- |
| Cared for | 0.11 (−0.09 – 0.31) |
0.12 (−0.07 – 0.31) |
-- | -- | -- |
| Concern About AD |
0.22
***
(0.14 – 0.31) |
0.17
***
(0.09 – 0.25) |
0.17
***
(0.08 – 0.26) |
0.15
***
(0.06 – 0.24) |
0.12
**
(0.04 – 0.21) |
| Belief in AD modifiability | -- |
0.13
*
(0.02 – 0.25) |
-- |
0.23
***
(0.11 – 0.35) |
0.27
***
(0.15 – 0.39) |
| Race x Age | -- | -- | 0.01 (−0.00 – 0.02) |
-- | |
| Race x Gender |
−0.15
*
(−0.26 – −0.03) |
−0.08 (−0.18 – 0.03) |
-- | -- | -- |
| Race x Education | -- | -- | -- |
−0.06
**
(−0.11 – −0.02) |
−0.04 (−0.08 – 0.00) |
| Race x Health Self-rating | -- | -- | -- | -- | -- |
| Race x Memory Self-rating | -- | -- | -- | -- | -- |
| Race x Health LOC |
0.26
*
(0.06 – 0.46) |
-- | -- | -- | 0.17 (−0.03 – 0.37) |
| Race x Healthcare Access | -- | -- | -- | -- | -- |
| Race x Research Attitudes | 0.02 (−0.01 – 0.05) |
-- | |||
| Race x Social Norms | -- | -- |
0.25
***
(0.12 – 0.38) |
-- | -- |
| Race x Discrimination | -- | 0.02 (−0.00 – 0.04) |
-- |
0.03
*
(0.00 – 0.05) |
-- |
| Race x Family History | -- | -- | -- | -- | -- |
| Race x Experience w Dementia | |||||
| Any | 0.19 (−0.12 – 0.51) |
−0.11 (−0.40 – 0.18) |
-- | -- | -- |
| Caregiver | −0.20 (−0.40 – 0.00) |
−0.06 (−0.25 – 0.13) |
-- | -- | -- |
| Race x Concern about AD | -- | -- | -- | -- | -- |
| Race x Belief in AD modifiability | -- | -- | -- | -- | 0.10 (–0.02 – 0.21) |
| Overall Model R2 (R2 Adjusted) | 0.22 (0.18) | 0.22 (0.17) | 0.20 (0.17) | 0.25 (0.21) | 0.25 (0.22) |
p<.05,
p<.01,
p≤.001.
Used backwards selection with AIC criteria, forced age, gender, education, race, and family history into all models as potential covariates, only removed primary variables once their interaction with race were removed.
Figure 1.
Significant interaction effects between race and health locus of control, brain health social norms, years of education, and perceived discrimination on anticipated reactions to learning biomarker results.
DISCUSSION
We investigated contextual factors associated with anticipated reactions to learning AD biomarker results in an observational cohort of cognitively healthy middle-aged and older adults. Results demonstrated that lower internal health LOC (e.g., feeling less directly responsible for one’s health) and greater concern about dementia were the most consistent predictors of anticipated psychological impact, including distress (e.g., anger, sadness, anxiety), perceived stigma (e.g., difficulty sharing results with loved ones), or cognitive symptoms (e.g., feeling less able to remember or concentrate) when imagining learning high-risk AD biomarker results. In contrast, the most consistent predictors of anticipated behavior changes included younger age, Black or African American racial identity, higher belief in AD modifiability, more positive attitudes about medical research, and greater concern about dementia. Concern about dementia was the only factor that predicted both anticipated psychological impact and behavior change.
Lower health LOC may be associated with greater anticipated psychological impact related to learning results because individuals feel they have less control over the result or their ability to reduce their risk. Prior studies suggest higher internal health LOC is positively associated with psychological function and health behaviors17. This result suggests the importance of targeting strategies to increase internal LOC as a part of pre- or post-disclosure counseling to reduce the likelihood of psychological distress related to learning results. Secondary analyses suggested that higher internal health LOC was associated with less anticipated distress for Black respondents, while there was no difference in anticipated distress for White respondents with high or low internal health LOC. Thus, increasing internal health LOC may be particularly relevant for Black participants in this cohort. Additionally, participants who expressed greater concern about developing Alzheimer’s dementia endorsed anticipated psychological impact. While this is expected, it suggests that pre-existing concerns may lead to or exacerbate post-result reactions. These pre-existing concerns may be due to worries about stigma and discrimination , as has been shown in prior survey studies of the general population18,19. This could be addressed by providing focused pre- and post-testing education about protecting privacy or reducing risks for discrimination. In addition to increasing risk for psychological impacts, greater concern about dementia was also associated with anticipated behavior change after learning results, suggesting concern may also be a motivator for taking action on risk-reduction or planning activities after learning results.
In contrast, age, racial identity, social norms, research attitudes, experience with dementia, healthcare access, health self-rating, memory self-rating, or perceived discrimination were not associated with anticipated distress, stigma, or cognitive symptoms. These null findings are interesting in that they suggest that some experiential factors (e.g., access to healthcare, social norms related to brain health, racial identity, perceived discrimination, attitudes about medical research) and some individual factors (e.g., age, ratings of personal health and memory abilities) do not influence anticipated psychological symptoms post-disclosure. Although these factors were non-significant across the total sample, in secondary analyses we observed some variation across racial groups for social norms whereby greater exposure to friends/family engaging in brain health activities was associated with more anticipated cognitive symptoms in White respondents, but fewer anticipated cognitive symptoms in Black respondents. Additionally, although there were no gender differences between Black men and women on anticipated distress, White women anticipated experiencing more distress than White men.
Results demonstrated the most consistent predictors of anticipated behavior changes after learning biomarker results included younger age, Black or African American racial identity, higher belief in AD modifiability, more positive attitudes about medical research, and greater concern about AD. Prior studies in similar aged samples have also reported that age, family history of dementia, concern/fear of developing dementia, and desire to improve knowledge were associated with willingness to engage in AD risk reduction behaviors, such as seeing a doctor or making a lifestyle change11,20. Although most studies of reactions to learning biomarker results have been conducted with cognitively unimpaired adults aged 65 and older, these results suggest that younger adults may be more likely to engage in behavior change after learning their results than older adults. This may suggest an increased benefit for sharing results with those under 65. Currently, sharing AD biomarker results with cognitively unimpaired adults is less common given potential risks of discrimination, undue burden/worry, stigma, and lack of clinical utility. Although research on the impact of behavior change and dementia risk-reduction is still ongoing, available literature suggests risk factors present at midlife (e.g., obesity, diabetes) are associated with increased dementia risk21. It is not surprising that those who report a higher belief in AD modifiability, (i.e., AD is caused by multiple factors including modifiable factors such as physical or mental inactivity, diet, or stress) are more likely to engage in behavior change after learning a high-risk AD biomarker result, as these individuals believe they have more control over their overall AD dementia risk. This result may support more pre-disclosure education about the multiple modifiable and non-modifiable risk factors for AD dementia outcomes, which may result in greater likelihood for individuals to engage in greater post-disclosure brain healthy behaviors. Additionally, the finding that Black or African American participants are more likely to engage in behavior change after learning high-risk results compared to non-Hispanic White participants suggests a need to ensure availability of culturally-informed resources to guide actions to reduce overall dementia risk, including facilitating infrastructure to support people in making behavior changes, such as increasing access to healthy food options in food-scarce areas or other disparate barriers facing those from under-served areas.
Furthermore, although this particular sample has highly positive attitudes towards research, those within this group who endorse more positive research attitudes also report greater willingness to engage in behavior change post-disclosure. Although this correlation could result from an unmeasured third factor (e.g., personality characteristics like agreeableness), it may also suggest that building long-standing relationships and trust with participants is associated with the likelihood of benefit after biomarker testing. Although years of education and perceived discrimination were not associated with anticipated planning or risk-reduction behaviors across the total sample, results of secondary analyses demonstrated that Black respondents with higher education anticipated more planning behavior and no differences by discrimination for Black respondents. White respondents with higher education anticipated less planning behavior, while White respondents who experienced greater discrimination anticipated more planning behavior. In contrast, no other predictors were associated with anticipated planning or changes risk-reduction behavior.
Strengths of this study include the novelty of the survey to assess willingness to learn biomarker results, anticipated psychological and behavioral reactions, and contextual factors, as well as representation of nearly half of respondents from the Black or African American community. A limitation of this study is that respondents were a convenience sample of people already enrolled in a longitudinal observational AD study. Although these participants reflect a relevant group of people who are undergoing biomarker testing and may be eligible for clinical trials, results are unlikely to be generalized to the general public. Additionally, this study included hypothetical vignettes rather than reactions to learning actual biomarker results. Future directions of this work include assessing actual short- and long-term psychological and behavioral reactions to learning biomarker results, describing more specific components of recommended pre-disclosure educational sessions and evaluating the impact on post-disclosure reactions, and developing personalized post-disclosure supports and identifying who will benefit most from them.
In conclusion, this survey demonstrated that lower internal health LOC and greater concern about AD most consistently predicted anticipated psychological impact, while younger age, Black or African American racial identity, higher belief in AD modifiability, positive research attitudes, and greater concern about AD most consistently predicted anticipated behavior change. These findings may inform the development of culturally-tailored practices to maximize planning and minimize the psychological impact of learning AD risk information. For example, providing more education on chronic health conditions, raising awareness of modifiable risk factors for dementia, ensuring giving back and fostering trust in research in the community, and addressing specific concerns about AD may help increase the benefits of learning results. Moreover, given the interest in learning results and anticipated greater behavior change, prioritizing opportunities for Black participants to learn their biomarker results as well as personalizing post-disclosure action plans for dementia risk-reduction and planning behaviors is also recommended.
Supplementary Material
Acknowledgments
We extend our deepest thanks to the WRAP and WADRC participants and staff for their invaluable contributions to the study. We gratefully acknowledge the assistance of Shana Stites, Jason Karlawish, and their team for providing feedback on the Alzheimer’s Biomarker Survey during development. We would also like to acknowledge the University of Wisconsin Survey Center for their assistance with survey development and data collection. This study was supported by funding from the National Institute on Aging (R03 AG062975 (Clark), R01 AG054059 (Gleason), R01 AG027161, P30 AG062715). This material is the result of work supported by resources and the use of facilities at the William S. Middleton Memorial VA Hospital in Madison, WI.
Conflict of interest and source of funding:
The authors declare that they have no affiliations with or involvement in any organization or entity with any financial interest in the subject matter or materials discussed in this manuscript. This publication was supported by funding from the National Institute on Aging (R03 AG062975 (Clark), R01 AG054059 (Gleason), R01 AG027161, P30 AG062715).
Data availability statement.
The datasets generated during and/or analyzed during the current study are not publicly available, but are available from the corresponding author on reasonable request.
REFERENCES
- 1.Kim SYH, Karlawish J, Berkman BE. Ethics of genetic and biomarker test disclosures in neurodegenerative disease prevention trials. Neurology. 2015;84(14):1488–1494. doi: 10.1212/WNL.0000000000001451 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Largent EA, Harkins K, Dyck CH van, et al. Cognitively unimpaired adults’ reactions to disclosure of amyloid PET scan results. PLOS ONE. 2020;15(2):e0229137. doi: 10.1371/journal.pone.0229137 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Grill JD, Raman R, Ernstrom K, et al. Short-term Psychological Outcomes of Disclosing Amyloid Imaging Results to Research Participants Who Do Not Have Cognitive Impairment. JAMA Neurol. 2020;77(12):1504–1513. doi: 10.1001/jamaneurol.2020.2734 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Clark LR, Erickson CM, Jonaitis EM, et al. Anticipated reactions to learning Alzheimer’s disease biomarker results. Alzheimers Res Ther. 2022;14(1):85. doi: 10.1186/s13195-022-01027-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Erickson CM, Chin NA, Ketchum FB, et al. Predictors of Willingness to Enroll in Hypothetical Alzheimer Disease Biomarker Studies that Disclose Personal Results. Alzheimer Dis Assoc Disord. Published online March 16, 2022. doi: 10.1097/WAD.0000000000000490 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ketchum FB, Erickson CM, Chin NA, et al. What Influences the Willingness of Blacks and African Americans to Enroll in Preclinical Alzheimer’s Disease Biomarker Research? A Qualitative Vignette Analysis. J Alzheimers Dis JAD. 2022;87(3):1167–1179. doi: 10.3233/JAD-215521 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 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 J Alzheimers Assoc. 4;7(3):263–269. doi: 10.1016/j.jalz.2011.03.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 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 J Alzheimers Assoc. 2011;7(3):270–279. doi: 10.1016/j.jalz.2011.03.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Roberts JS, Connell CM. Illness Representations Among First-Degree Relatives of People With Alzheimer Disease. Alzheimer Dis Assoc Disord. 2000;14(3):129–136. [DOI] [PubMed] [Google Scholar]
- 10.Wallston KA, Wallston BS, DeVellis R. Development of the Multidimensional Health Locus of Control (MHLC) Scales. Health Educ Monogr. 1978;6(2):160–170. doi: 10.1177/109019817800600107 [DOI] [PubMed] [Google Scholar]
- 11.Seifan A, Ganzer CA, Vermeylen F, et al. Development and validation of the Alzheimer’s prevention beliefs measure in a multi-ethnic cohort-a behavioral theory approach. J Public Health Oxf Engl. 2017;39(4):863–873. doi: 10.1093/pubmed/fdw145 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Rubright JD, Cary MS, Karlawish JH, et al. Measuring how people view biomedical research: Reliability and validity analysis of the Research Attitudes Questionnaire. J Empir Res Hum Res Ethics. 2011;6(1):63–68. doi: 10.1525/jer.2011.6.1.63 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Williams DR, Yu Y, Jackson JS, et al. Racial Differences in Physical and Mental Health: Socio-economic Status, Stress and Discrimination. J Health Psychol. 1997;2(3):335–351. doi: 10.1177/135910539700200305 [DOI] [PubMed] [Google Scholar]
- 14.Ryan MM, Gillen DL, Grill JD. Reasons for undergoing amyloid imaging among cognitively unimpaired older adults. Ann Clin Transl Neurol. 2021;8(8):1646–1655. doi: 10.1002/acn3.51414 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Chung WW, Chen CA, Cupples LA, et al. A New Scale Measuring Psychological Impact of Genetic Susceptibility Testing for Alzheimer’s Disease. Alzheimer Dis Assoc Disord. 2009;23(1):50–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Friedman JH, Hastie T, Tibshirani R. Regularization Paths for Generalized Linear Models via Coordinate Descent. J Stat Softw. 2010;33(1):1–22. doi: 10.18637/jss.v033.i01 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Dogonchi M, Mohammadzadeh F, Moshki M. Investigating the Relationship between Health Locus of Control and Health Behaviors: A Systematic Review. Open Public Health J. 2022;15(1). doi: 10.2174/18749445-v15-e2208010 [DOI] [Google Scholar]
- 18.Stites SD, Gill J, Largent EA, et al. The relative contributions of biomarkers, disease modifying treatment, and dementia severity to Alzheimer’s stigma: A vignette-based experiment. Soc Sci Med 1982. 2022;292:114620. doi: 10.1016/j.socscimed.2021.114620 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Stites SD, Rubright JD, Karlawish J. What features of stigma do the public most commonly attribute to Alzheimer’s disease dementia? Results of a survey of the U.S. general public. Alzheimers Dement J Alzheimers Assoc. 2018;14(7):925–932. doi: 10.1016/j.jalz.2018.01.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.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]
- 21.Livingston G, Huntley J, Sommerlad A, et al. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. The Lancet. 2020;396(10248):413–446. doi: 10.1016/S0140-6736(20)30367-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated during and/or analyzed during the current study are not publicly available, but are available from the corresponding author on reasonable request.