Perceptions about preclinical Alzheimer's disease biomarker collection procedure influences willingness to participate: Findings from an ethnoracially diverse study

Abstract

Background

Past research suggests that ethnoracialized groups differ in their willingness to engage in preclinical Alzheimer's disease (AD) research overall. Studies indicated that participation willingness was affected by attitudes toward research and perceived invasiveness of biomarker collection techniques. However, comparative quantitative studies are few, and minoritized groups are under-included.

Objective

In a cross-sectional online survey, we sought to explore community-based adults’ willingness to engage in preclinical AD biomarker testing, comparing their attitudes about research and different types of biomarker procedures.

Methods

We conducted an online survey with a diverse group of participants. African American (AA), American Indian/Alaska Native (AI/AN), Latinx (LTX), and Non-Hispanic White (NHW) adults aged 26–90 were asked about their research attitudes, biomarkers, and willingness to participate in specific biomarker test procedures (i.e., brain imaging via PET scanning, blood draws, and cerebrospinal fluid collection by lumbar puncture). We also assessed participants’ perceived safety, burden, and distress for each of the three biomarker collection methods. To understand the association between research willingness and ethnoracial identity, we ran linear regression models for each procedure, adjusting for age, gender, educational attainment, and attitudes toward research.

Results

The AA group expressed greater willingness to engage in biomarker testing involving blood draws than the NHW group. The AI/AN group was significantly less willing to undergo lumbar puncture than the NHW group; this difference remained after adjusting for various sociodemographic factors and research attitudes.

Conclusions

Respondents’ willingness to engage in preclinical AD biomarker research was affected by their perceptions about the testing collection procedure.

Keywords

Alzheimer's disease biomarker testing blood-based biomarkers brain scanning lumbar puncture race/ethnicity research attitudes research participation testing willingness

Introduction

Biomarkers are any human characteristic that can be used as a clinical indicator of normal function, a pathogenic process, or response to an exposure or intervention.¹ Consideration of Alzheimer's disease (AD) biomarkers in the NIA-AA² operational definition of the AD continuum has revolutionized the preclinical and diagnosis of AD. According to Mayeda et al.,³ a landmark population-based study with a diverse cohort, all-cause dementia incidence is highest for Black/African American (AA) and American Indian/Alaska Native (AI/AN) communities and considerably lower for Asian American (ASA) and Non-Hispanic White (NHW) communities. A nationwide VA-based study reported dementia incidence in Hispanic/Latinx communities was nearly double that of NHW communities.⁴ Despite the disproportionate prevalence and impact of AD dementia on ethnoracially minoritized communities, they have been under-included and underrepresented in biomarker research. Greater representation of ethnic and ancestral diversity in biomarker research is imperative and becoming more pressing in this time of “precision medicine”.^5–7 This is particularly critical, as findings suggest that biomarker cut-points defining abnormality derived from NHW samples may not be generalizable to other groups.^7,8

The underrepresentation of ethnoracially minoritized groups reflects challenges at both the scientist level⁹ and participant level.^10,11 For example, African Americans and American Indian/Alaska Natives may feel reluctance and less trusting attitudes towards biomedical and scientific research due to historical injustice. The mistreatment of the African American community in biomedical research includes but is not limited to, the Tuskegee Untreated Syphilis Study^12,13 and the Henrietta Lacks case.¹⁴ The historical mistreatment of the American Indian/Alaska Native community in biomedical research includes but is not limited to, the Barrows, Alaska Alcohol Study¹⁵ and Arizona State University's handling of the Havasupai genetic material.^16,17

Relatively few quantitative studies have compared groups’ willingness to participate in AD biomarker research. Past research suggests that overall, the ethnoracialized groups differ in their expressed willingness to participate in preclinical AD research.^18,19 In one large survey,¹⁸ minoritized groups were less willing to engage in several types of AD research. Non-Hispanic Black respondents were less willing to engage in research relative to NHWs for studies involving brain imaging (MRI and PET), blood draws, lumbar puncture, and autopsy. Compared to the NHW respondents, the Hispanic respondents and Non-Hispanic Asian respondents showed less willingness to engage in several procedures except for lumbar puncture. Research attitudes, operationalized as the summary scores from the 7-item Research Attitudes Questionnaire,²⁰ did not mediate the ethnoracial minoritized groups’ lower willingness to participate. One limitation of this investigation is its under-inclusion of non-White participants. Despite the large sample (N = 2749), the majority (87%) of the respondents were NHW, with each of the minoritized groups comprising less than 10% of the entire sample. The AA sample in particular was quite small (N = 36, or 1%).

In The Alzheimer's Biomarker Study,²¹ the investigators conducted a telephone survey of AD research participants focused on willingness to enroll in AD biomarker studies. Participants revealed clear preferences for minimally invasive procedures, with highest enrollment willingness for blood-based biomarkers and lowest enrollment willingness for the lumbar puncture procedure. Strengths of the Erickson et al. study²¹ include the relatively large proportion (44%, or 148 of 334) of Black participants. However, respondents were already enrolled in ongoing Alzheimer's disease prevention studies offering biomarker testing, i.e., a group already biased toward research participation.

The Understanding Biomarker and Genetic Research (UBIGR) Project,¹⁹ demonstrated ethnoracialized group differences in relative willingness to engage in preclinical AD biomarker research. A University of Wisconsin-Madison-based nationally derived combined group consisting of both AA and AI/AN participants expressed less willingness to undergo preclinical AD biomarker testing than NHW participants. Equally important, research attitudes, operationalized as RAQ total scores, were an important contributing factor in predicting willingness to undergo preclinical AD biomarker testing.

The present study

The present investigation was intended to extend and build upon earlier studies^18,21 by recruiting and including a more sociodemographically diverse group of participants who had no known prior experience with preclinical AD research. We specifically focused on including larger numbers of members from under-included communities and utilizing quantitative methods. While our earlier report¹⁹ focused on participants’ willingness to engage in any preclinical biomarker testing, in the present investigation, we compared participants’ attitudes about different types of biomarker tests.

We explored whether the type of biomarker sample collection method would influence respondents’ willingness to participate. We specifically inquired about different preclinical biomarker procedures, namely, PET scanning, lumbar puncture, and phlebotomy (blood draws). We compared participants’ stated willingness versus reluctance to engage in preclinical AD biomarker testing under the three different procedures, to examine whether, relative to the NHW group, the other groups would differ in their willingness to engage in preclinical AD biomarker testing, i.e., does the procedure type make a difference?

We hypothesized that members of African American and American Indian/Alaska Native groups would endorse less trusting attitudes toward biomedical and scientific research. We hypothesized that RAQ scores would be lower amongst members of Black, Indigenous People of Color (BIPOC) communities who have a documented history of egregious mistreatment by biomedical institutions (i.e., Black/African Americans and American Indian/Alaska Natives) compared to other BIPOC community groups (i.e., Hispanic/Latinx) and non-BIPOC communities, i.e., non-Hispanic White persons. We did not expect that Hispanic/Latinx participants would differ significantly from non-Hispanic White participants in terms of their attitudes toward research. We were interested in whether groups’ attitudes toward research would mediate their willingness to engage in specific types of preclinical biomarker testing.

We predicted that self-denoted ethnoracial group identity would affect willingness to engage in type of biomarker research. Earlier investigations indicate that most survey respondents, especially African Americans, express greater participation reluctance when the research procedures are perceived to be more invasive, such as brain donation²² or lumbar puncture.^18,21 Based upon this literature, we hypothesized that communities under-represented in research would be less willing to undergo biomarker tests that are generally perceived to be more invasive, namely, lumbar puncture relative to less invasive procedures such as providing a blood sample or getting a brain scan. We predicted that due to their greater familiarity with blood tests, there would be a preference for blood samples rather than brain imaging.

Methods

This investigation was part of an ongoing project, namely, the Understanding Biomarker and Genetic Research (UBIGR) Participation Study (Gooding and Gleason, co-PIs), a concerted effort to ascertain the perspectives and needs of minoritized groups to help facilitate greater inclusion in and benefit from future biomedical research.

Participants

Over 2000 adult participants (n = 2187) took an online questionnaire between July 2022 and March 2023. Participants came from two sources: community members and recruitment by a survey research firm which maintains a nationally representative panel of participants who can be invited to complete online surveys for payment. We made a special effort to recruit members of underrepresented groups through several means, including community engagement events, targeted crowdsourcing, and word-of-mouth to enhance our sample. Recruitment efforts included presentations by the first author on media outlets targeted at historically underrepresented communities, distribution of study flyers at community events, advertisements in a newspaper targeted at diverse communities, and outreach through a community advisory board.

Adults over the age of 26 who were able to give informed consent, had normal or corrected-to-normal vision, and had access to a computer or mobile device (i.e., Smartphone or tablet) with access to the internet were eligible for study inclusion. Exclusion criteria included age over 90, guardianship status, self-reported memory disorder, and/or unwillingness to indicate consent or indicate ethnoracialized group membership.

Procedure and methods

The online questionnaire was hosted by Qualtrics, a web-based survey and research software program. Participants provided informed consent electronically and completed a questionnaire that contained a standardized measure of research attitudes as well as largely structured items pertaining to sociodemographic characteristics and research willingness. The survey took approximately 30 to 45 min to complete.

Several data integrity strategies were implemented, including utilizing a “soft launch” to assess errors in survey design, only allowing respondents from one Internet Protocol address, excluding survey responses that were completed in under 10 min, and including reverse-scored items to identify straight-line responding. We also included “catch items”, i.e., infrequent items or items that would otherwise detect inattention or random response patterns. All participants were financially compensated ($25) for participating in the study. This research was approved by the University of Wisconsin-Madison Educational and Social Sciences Human Subjects Board. All procedures followed were in accordance with the ethical standards of the Helsinki Declaration of 1975, as revised in 2000, and The Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving humans.

Sociodemographic characteristics

Demographic data collected included age, self-identified ethnoracialized identity, sex at birth, gender identity, level of education, income, marital/relationship status, and first three digits of their zip code. Participants were also asked about their family history of AD dementia in first-degree relatives, and experience with genetic testing.

Attitudes towards research

We examined participants’ attitudes about biomedical research using the seven-item Research Attitudes Questionnaire (RAQ-7).²⁰ The items in the RAQ-7 pertain to trust in science, benefit of science, and feelings of altruism, three of the major reasons most individuals engage in research. Scored on a 5-point scale, ranging from 1 (strongly disagree) to 5 (strongly agree), higher summed scores indicate more favorable attitudes toward research. In the present sample, the internal reliability was good (ICC = 0.82).

Attitudes about biomarkers and health behaviors

We provided participants with definitions of biomarkers, both in the informed consent and again in the questionnaire text. After defining biomarkers and giving examples, such as blood pressure measurements to assess for heart disease risk, we presented participants with a series of statements and asked them to indicate the extent to which they agreed or disagreed with them. For example, participants’ attitudes regarding the overall utility of biomarker testing was assessed using the statement, “Biomarkers are useful even if they only point out susceptibility for a disease”.

Attitudes about willingness to participate in specific biomarker test procedures

We were primarily interested in participants’ willingness to engage in biomarker studies that involved different types of procedures, namely, those using brain imaging (i.e., positron emission tomography (PET) or lumbar puncture, and those involving a blood draw. To fully describe the procedure participants were provided with brief scenarios prior to being queried about their willingness to engage in a particular type of biomarker measurement. Verbatim text from the online survey is provided in Table 1. Participants’ willingness to engage in these biomarker collection procedures was examined by probing attitudes about the procedures’ perceived safety, invasiveness, and burden. Endorsement of the statements (e.g., “I would not want to be tested for the Alzheimer's disease biomarker using PET scanning because the test is too invasive”) was rated on a Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree). The outcome measure for each procedure was a mean composite of three items (related to perceived invasiveness, dangerousness, and burden), resulting in scores that ranged from 1 to 7.

Table 1.

Questionnaire items assessing willingness to engage in different biomarker assays.

In the next series of questions, we ask you how you feel about different methods of collecting biomarker information.
Method One: PET Scan
Consider the following scenario. There's a brain marker for ALZHEIMER's DISEASE that can be measured using a type of brain imaging called a PET scan. With the PET scan, a “tracer” containing a small amount of radioactive dye is injected into a vein in your arm. After one hour, you are placed in a scanner for approximately an hour. The biomarker would be measured using your brain scan image. How would you feel about that? Would you want to be tested in that manner?
I WOULD NOT want to be tested for the ALZHEIMER'S DISEASE biomarker using PET scanning because the test is too invasive.
1	2	3	4	5	6	7
Strongly Disagree	Disagree	Mildly Disagree	Neither Disagree nor Agree	Mildly	Agree Agree	Strongly Agree
I WOULD NOT want to be tested for the ALZHEIMER'S DISEASE biomarker using PET scanning because the test is too dangerous.
1	2	3	4	5	6	7
Strongly Disagree	Disagree	Mildly Disagree	Neither Disagree nor Agree	Mildly	Agree Agree	Strongly Agree
I WOULD NOT want to be tested for the ALZHEIMER'S DISEASE biomarker using PET scanning because the test is too burdensome.
1	2	3	4	5	6	7
Strongly Disagree	Disagree	Mildly Disagree	Neither Disagree nor Agree	Mildly	Agree Agree	Strongly Agree
Method Two: Lumbar Puncture
Now consider this alternative method. Rather than a brain scan, the biomarker for ALZHEIMER's DISEASE would be measured using a lumbar puncture. The lumbar procedure is also known as a spinal tap. You’d be given a local anesthetic and a very thin needle would be inserted into the space between the bones of your spine. The biomarker would be measured using your cerebrospinal fluid results. How would you feel about that? Would you want to be tested in that manner?
I WOULD NOT want to be tested for the ALZHEIMER'S DISEASE biomarker because the Lumbar Puncture test is too invasive.
1	2	3	4	5	6	7
Strongly Disagree	Disagree	Mildly Disagree	Neither Disagree nor Agree	Mildly	Agree Agree	Strongly Agree
I WOULD NOT want to be tested for the ALZHEIMER'S DISEASE biomarker because the Lumbar Puncture test is too dangerous.
1	2	3	4	5	6	7
Strongly Disagree	Disagree	Mildly Disagree	Neither Disagree nor Agree	Mildly	Agree Agree	Strongly Agree
I WOULD NOT want to be tested for the ALZHEIMER'S DISEASE biomarker because the Lumbar Puncture test is too burdensome.
1	2	3	4	5	6	7
Strongly Disagree	Disagree	Mildly Disagree	Neither Disagree nor Agree	Mildly	Agree Agree	Strongly Agree
Method Three: Blood Draw
Here's one more method to consider: the biomarker for ALZHEIMER'S DISEASE would be measured by drawing some blood. The biomarker would be measured using your blood test results. How would you feel about that? Would you want to be tested in that manner?
I WOULD NOT want to be tested for the ALZHEIMER'S DISEASE biomarker using blood draw because the test is too invasive.
1	2	3	4	5	6	7
Strongly Disagree	Disagree	Mildly Disagree	Neither Disagree nor Agree	Mildly	Agree Agree	Strongly Agree
I WOULD be willing to be tested for the ALZHEIMER'S DISEASE biomarker using blood draw because the test is NOT dangerous (REVERSE CODED)
1	2	3	4	5	6	7
Strongly Disagree	Disagree	Mildly Disagree	Neither Disagree nor Agree	Mildly	Agree Agree	Strongly Agree
I WOULD NOT be willing to be tested for the ALZHEIMER'S DISEASE biomarker using blood draw because the test is too burdensome.
1	2	3	4	5	6	7
Strongly Disagree	Disagree	Mildly Disagree	Neither Disagree nor Agree	Mildly	Agree Agree	Strongly Agree

Statistical analysis

Unless otherwise noted, analyses were computed using SPSS version 27 (IBM SPSS, Armonk, NY, USA). We used descriptive statistics to provide information on sample characteristics. We calculated means, standard deviations (SDs), and percentages to evaluate response patterns on survey items. Participants were excluded due to their displays of random responding, inattentive response patterns, and/or excessive missing data. Participants who missed more than 10% of the items were removed in order to reduce bias. Other participants (n = 60) were excluded from analyses due to their refusal to disclose their ethnicity, and 14 participants were excluded due to their nonbinary gender. Although the inclusion of the nonbinary participants did not change the results in any meaningful way, subgroups for which there were fewer than 5 in each cell were omitted from further statistical analysis. Some of the cases satisfied more than one reason for exclusion. The total number of participants excluded from analysis was 362 (16.55%). This resulted in a final sample of 1825 (83.44%) participants.

After deleting these cases, we used multivariate imputation by chained equations (MICE) in R software²³ to impute missing data. One hundred and fifty-one (8.27%) of the sample had at least one missing response; none of the missing data pertained to the perceptions regarding biomarker testing procedure or self-reported ethnoracial identity.

To examine the intraclass consistency, a component of construct reliability, we calculated Cronbach's alpha. We conducted chi-square analyses to compare Likert response categories across groups. Post hoc analysis of chi-squared tests used Fisher's exact approach, with Bonferroni corrected significant levels (i.e., 0.05/(nresponses*4groups) = adjusted significance level).²⁴ To investigate the association between ethnoracial identity and willingness to engage in biomarker research, we ran separate linear regression models for each of the three biomarker procedures (R package stats): 1) we assessed the association between research willingness and ethnoracial identity (unadjusted); 2) next, we adjusted for age, gender, and educational attainment; and 3) we adjusted the regression models by adjusting for attitudes towards research, as measured by the total RAQ score. For Models 2 and 3, all the predictor variables were entered simultaneously. To permit comparisons with the extant literature, we compared each of the underrepresented ethnoracial groups to the NHW group. In all models, ethnoracial identity was entered as a nominal predictor. To permit comparisons with the extant literature, we designated the NHW group as the reference group; selecting the NHW group as the reference group does not imply a “gold standard”.

Results

Participant characteristics

Participants were invited to self-identify their primary ethnoracial group. We classified participants into four ethnoracialized categories of Black/African American (AA), American Indian/Alaska Native (AI/AN), Latinx (LTX), and Non-Hispanic White (NHW). Individuals from African ethnic origins, as well as those from sub-Saharan Africa, North Africa, and the Caribbean, were classified as “Black/African American”. Individuals recognized by one of the Tribal Nations or indigenous Alaskan groups (e.g., Aleuts, Inupiat, Yuit, Athabascans, Tlingit, and Haida) were classified as “American Indian/Alaska Native”. Individuals from Latin America, Mexico, South America, Cuba, Spain, Portugal, or Puerto Rico were classified in the “Hispanic/Latinx” group. Individuals from North America or European origins other than Spain and Portugal were classified as “Non-Hispanic White”. The participants were drawn from nearly every state in the U.S., along with the District of Columbia. Table 2 provides the self-reported sociodemographic characteristics of the sample.

Table 2.

Participant demographic characteristics.

Variable	Total Sample (N = 1825)	AA (N = 599)	AI/AN (N = 352)	LTX (N = 149)	NHW (N = 725)	p
Age						<0.001
Mean (SD)	45 (± 12.75)	50.88 (± 13.4)	47.78 (±13.9)	42.66 (± 10.49)	39.28 (± 8.9)
range	26–89	26–88	26–86	27–73	26–89
Gender						<0.001
Male	1046 (57.3)	267 (44.5)	187 (53.1)	81 (54.4)	511 (70.5)
Female	779 (42.7)	332 (55.4)	165 (46.9)	68 (45.6)	214 (29.5)
Education						<0.001
HS or less	314 (17.2)	167 (27.9)	74 (21.0)	32 (21.5)	41 (5.7)
Some college (13–15)	645 (35.3)	244 (40.7)	154 (43.8)	62 (41.6)	185 (25.5)
College or higher	866 (47.4)	188 (31.3)	124 (35.2)	55 (36.9)	499 (68.8)
Marital status						<0.001
Single, never married	376 (20.6)	207 (34.6)	91 (25.9)	26 (17.4)	52 (7.2)
Separated	191 (10.5)	54 (9.0)	37 (10.5)	23 (15.4)	77 (10.6)
Married	1020 (55.9)	225 (37.6)	151 (42.9)	81 (54.4)	563 (77.7)
Widowed	61 (3.3)	27 (4.5)	23 (6.5)	3 (2.0)	8 (1.1)
Divorced	175 (9.6)	86 (14.4)	50 (14.2)	16 (10.7)	23 (3.1)
Income						<0.001
<20,000	240 (13.2)	125 (20.9)	62 (17.6)	27 (18.1)	26 (3.6)
20,000–40,000	422 (23.1)	194 (32.4)	84 (23.9)	26 (17.4)	118 (26.3)
40,000–60,000	251 (13.8)	100 (16.7)	57 (16.2)	29 (19.5)	65 (9.0)
60,000–80,000	388 (21.3)	76 (12.7)	74 (21.0)	43 (28.9)	195(26.9)
80,000–100,000	320 (17.5)	28 (4.7)	40 (11.4)	9 (6.0)	243 (33.5)
>100,000	164 (9.0)	56 (9.3)	26 (7.3)	12 (8.1)	70 (28.8)
Did not disclose	40	20	9	3	8

Means and Standard Deviations (SD) provided for total sample, as well as each of the four ethnoracialized groups: AA: Black/African American; AI/AN: American Indian/Alaska Native; LTX: Latinx; NHW: Non-Hispanic White.

Overall, the participants were middle-aged, with a mean age of 45 (± 12.75). There was a significant group difference in mean age, F(3, 1821) = 117.31, p < 0.001. The AA group was significantly older than the AI/AN group (p = 0.001), and the LTX (p < 0.001). The AI/AN participants were significantly older than the LTX participants, p < 0.001. The NHW group was significantly younger than the AA, AI/AN, and LTX groups, all ps < 0.001. There were more male respondents than females, though the four groups differed in terms of proportion of females, χ²(3) = 94.18, p < .0001. Significantly more of the AA participants were female whereas more of the NHW participants were male, p < 0.001.

The groups varied according to average level of educational attainment. More detailed description can be found in Supplemental Table 1; most (52.5%) of our participants did not have a college degree. The NHW participants were significantly more likely to have earned a college degree, p < 0.001 than AA and AI/AN participants. Most (55.9%) participants were married. The NHW participants were significantly more likely to be married compared to either the AA or AI/AN participants, p < 0.001. The AI/AN participants were significantly more likely to be widowed than the NHW participants, p < 0.001.

Finally, participants were asked to estimate their total annual income. Forty (2.2%) of the total sample opted not to disclose their income. Overall, the self-reported income of the sample displayed considerable range, with a nearly normal distribution from an annual salary of under $20K (13%) to over $100K (nearly 9%). We found significant group differences in terms of proportion of ethnoracialized participants in each income range, χ²(18) = 373.53, p < 0.001. Over half (53.3%; 319/599) of the AA participants reported an annual sample of $40,000 or less, whereas nearly 58% (203/352) AI/AN participants and 55% (82/149) of LTX participants reported an annual sample of $60,000 or less. In contrast, in the NHW group, 60.4% (438/725) of the participants reported an annual income of at least $60,000.

Attitudes towards research

Total scores on the RAQ ranged from 7 to 35, with a mean of 27 (SD = 4.44). Figure 1 depicts the overall sample's distribution of RAQ total scores. The four groups’ median total RAQ scores differed significantly, F(3, 1821) = 13.82, p < 0.001. Follow-up tests revealed that AA participants did not differ from the AI/AN participants (n.s.), though they differed from the LTX and NHW groups (p < 0.01 and 0.001, respectively). Similarly, the AI/AN participants differed significantly from the LTX and NHW participants (p < 0.001).

Figure 1.

Research attitude questionnaire scores for overall sample (N = 1825).

Table 3 compares the distribution of responses for each of the RAQ items across ethnoracialized groups. The groups differed significantly in terms of the proportion of their endorsements of the RAQ items. Chi-square analysis revealed a group difference in response to RAQ item 1, “I have a positive view about medical research in general,” χ² (6) = 31.49, p < 0.001.

Table 3.

Distribution of research attitudes responses by Ethnoracial Group^a.

		Level	of	Agreement	N	(%)
RAQ Item^b		Overall Sample	AA	AI/AN	LTX	NHW	p value
1. I have a positive view about medical research in general.	Disagree	105 (5.8)	27 (1.9)	24 (6.8)	6 (4.0)	48 (6.6)
	Neutral	317 (17.4)	112 (18.7)	89 (25.3)	23 (15.4)	93 (12.8)
	Agree	1403 (76.9)	460 (76.8)	239 (67.9)	120 (80.5)	584 (80.6)	***
3. Medical researchers can be trusted to protect the interests of people who take part in their research studies.	Disagree	130 (7.1)	40 (6.7)	36 (10.2)	8 (5.4)	46 (6.3)
	Neutral	415 (22.7)	176 (29.4)	93 (26.4)	26 (17.4)	120 (16.6)
	Agree	1280 (70.1)	383 (63.9)	223 (63.4)	115 (77.2)	559 (77.1)	***
4. We all have some responsibility to help others by volunteering for medical research.	Disagree	178 (9.8)	75 (12.5)	44 (12.5)	10 (6.7)	49 (6.8)	***
	Neutral	479 (26.2)	192 (32.1)	108 (30.7)	43 (28.9)	136 (18.8)
	Agree	1168 (64.0)	332 (55.4)	200 (56.8)	96 (64.4)	540 (74.5)
6. Society needs to devote more resources to medical research.	Disagree	85 (4.7)	23 (3.8)	17 (4.8)	4 (2.7)	41 (5.7)
	Neutral	346 (19.0)	102 (17.0)	84 (23.9)	33 (22.1)	127 (17.5)
	Agree	1394 (76.4)	474(79.1))	251 (71.3)	112 (75.2)	557 (76.8)
8. Participating in medical research is generally safe.	Disagree	133 (7.3)	46 (7.7)	35 (9.9)	7 (0.05)	45 (6.2)
	Neutral	557 (30.5)	225 (37.6)	126 (35.8)	52 (39.6)	154 (21.2)
	Agree	1135 (62.2)	328 (54.8)	191 (54.3)	90 (60.4)	526 (72.6)	***
9. If I volunteer for medical research, I know my personal information will be kept private and confidential.	Disagree	153 (8.4)	64 (10.7)	43 (12.2)	5 (3.4)	41 (5.7)	***
	Neutral	426 (23.3)	176 (29.4)	93 (26.4)	34 (22.8)	123 (17.0)
	Agree	1246 (68.3)	359 (59.9)	216 (61.4)	110 (73.8)	561 (77.4)
11. Medical research will find cures for many major diseases during my lifetime.	Disagree	90 (4.9)	28 (4.7)	23 (6.5)	4 (2.7)	35 (4.8)
	Neutral	372 (20.4)	162 (45.7)	94 (26.7)	29 (19.5)	87 (12.0)
	Agree	1363 (74.7)	409 (68.3)	235 (66.8)	116 (77.9)	603 (83.2)	***

Note. ^aAttitudes towards research measured by the Research Attitudes Questionnaire (RAQ; Rubright et al., 2011). Abbreviations for the ethnoracialized groups are as follows: AA= Black/African American; AI/AN= American Indian/Alaska Native; LTX= Latinx; NHW= NonHispanic White.

^bRAQ item numbers pertain to the item numbers on the original measure. The RAQ is scored on a 5-point Likert-type scale; strongly disagree to strongly agree. “Strongly disagree” and “disagree” were collapsed, and “Strongly agree” and “agree” were collapsed.

*p < .05 **p < .10 ***p < .001

Compared to the NHW group, the AI/AN group was more likely to respond in a neutral direction (i.e., neither agree nor disagree). The NHW participants were more likely to agree with this RAQ item. No other between-group differences were observed for RAQ item 1.

The groups differed significantly in terms of their endorsement of RAQ item 3, which stated “Medical researchers can be trusted to protect the interests of people who take part in their research studies”, χ² (6) = 46.20, p < 0.001. The AI/AN participants were more likely to disagree with this sentiment compared to the other three groups, p = 0.01. The AA participants were more likely to respond neutrally to RAQ item 3. Compared to the NHW group, a significantly smaller proportion of both the AA and AI/AN groups agreed that medical researchers could be trusted to protect the interests of the participants in their studies, p < 0.001. There were no differences between the LTX and NHW groups on the RAQ item regarding medical researchers’ trustworthiness.

RAQ item 4, “We all have some responsibility to help others by volunteering for medical research” also elicited differential responses from the four ethnoracialized groups, χ² (6) = 63.89, p < 0.001. The NHW group was more likely to agree with that attitude, compared to the AA group, p < 0.001 and the AI/AN group, p < 0.01. The AA group was more likely to endorse a neutral stance (i.e., neither agree nor disagree) in response to RAQ item 4, relative to the NHW group, p < 0.001. There were no differences between the LTX and NHW groups on the RAQ item regarding a sense of obligation to volunteer for research.

Overall, most (76.4%) of the respondents agreed with the sentiment expressed on item 6 of the RAQ, namely, “Society needs to devote more resources to medical research.” We observed no significant group differences in terms of endorsements of RAQ item 6, χ²(6) = 12.74, p = 0.05.

The groups responded differentially to RAQ item 8, “Participating in medical research is generally safe”, χ²(6) = 61.86, p < 0.001. Follow-up tests revealed that the AA participants were significantly less likely than the NHW participants to agree about the safety of medical research. In terms of RAQ item 9, the AA participants were less likely to agree that if they volunteered for medical research, their personal information would be kept private and confidential, p < 0.001. Similarly, AI/AN participants were less likely to agree than NHW participants, p < 0.01. The LTX and NHW groups did not differ significantly in their percent agreement, n.s. Compared to NHW participants, the AA participants were more likely to respond neutrally (i.e., neither agree nor disagree), p < 0.001; the AI/AN participants were also more likely to be more neutral than the NHW participants, p = 0.001. The LTX and NHW groups responded similarly to RAQ item 9.

RAQ item 11 probed respondents’ optimism regarding the outcome of medical research. Overall, the groups differed in responding to “Medical research will find cures for many major diseases during my lifetime”, χ²(6) = 61.90, p < 0.001. The AA participants were more likely to respond neutrally than the NHW participants, p < 0.001. Similarly, compared to the NHW respondents, the AI/AN participants were significantly more likely to respond neutrally, p = 0.001, than to endorse agreement, p < 0.001. We observed no statistically significant differences between the LTX and NHW groups on RAQ item 11. In summary, then, these RAQ data suggested that AA and AI/AN participant groups endorsed less trusting attitudes toward research compared to the NHW and LTX groups.

Attitudes about biomarkers in general

Participants were also asked about their general attitudes toward health screenings, genetic testing, and other health behaviors to place the questions about biomarkers in context. Over half (55.1%) of the respondents were either willing or had already engaged in genetic testing for medical disorders, such as prenatal screening. The majority (62%) of the respondents stated that they had been or intended to be screened for cancer, compared to 31% who responded negatively and 7.5% who were undecided. In contrast, only 39% of the respondents had any experience with genealogy testing in which their DNA was collected.

Most (71%,) of the participants agreed with the statement that “biomarkers are useful even if they only point out susceptibility for a disease”; 71.2% endorsed the overall utility of biomarkers. Chi-square analysis revealed significant group differences in the ethnoracialized group endorsements, [χ²(6) = 58.68, p ≤ 0.001]. AAs were significantly more likely to respond neutrally than NHWs, p < 0.001], whereas the NHWs were significantly more likely to disagree with the statement than the other three groups, p ≤ 0.001.

Attitudes towards specific types of biomarker test procedures

Overall, participants appeared more willing to engage in biomarker testing involving PET scanning and blood collection and less willing to engage in lumbar puncture tests (see Figure 2).

Figure 2.

Comparison of relative willingness by biomarker type.

To test our a priori hypotheses, we sought to compare ethnoracialized groups’ relative willingness to engage in different types of preclinical biomarker tests. Table 4 shows the results of the linear regression analyses predicting each of the ethnoracial groups’ willingness to undergo specific types of preclinical biomarker tests in comparison to the NHW group. In keeping with the literature's tradition, the NHW group was used as the reference group, and the other three participant groups, namely, the AA, AI/AN, and LTX groups, were compared to them. This practice does not imply that the NHW group is “the norm”. We followed this practice to permit comparisons with the extant literature.

Table 4.

Association of ethnoracialized group identity and willingness to undergo biomarker testing^a.

Black/African American Group (N = 599)
Procedure Type						95.0% CI
		B	S.E.	t-value	p	Lower	Upper
PET scan	Model 1	−0 . 34	0.09	3.89	0.0001	−0.518	−0.171
	Model 2	−0.06	0.10	0.60	0.547	−0.259	0.137
	Model 3	−0.08	0.10	0.76	0.450	−0.274	0.122
Lumbar puncture	Model 1	0.13	0.09	1.41	0.159	−0.051	0.309
	Model 2	0.11	0.11	1.06	0.290	−0.096	0.32
	Model 3	0.10	0.11	0.91	0.363	−0.111	0.304
Blood draw	Model 1	−0.43	0.07	6.12	<0.0001	−0.575	−0.296
	Model 2	−0.16	0.08	1.99	0.045	−0.322	−0.002
	Model 3	−0.16	0.08	2.00	0.045	−0.323	−0.003

American Indian/ Alaska Native Group (N = 352)
Procedure Type						95.0% CI
		B	S.E.	t-value	p	Lower	Upper
PET scan	Model 1	−0 . 42	0.10	4.06	<0.0001	−0.626	−0.218
	Model 2	−0.16	0.011	1.50	0.135	−0.377	0.051
	Model 3	−0.20	0.11	1.80	0.071	−0.411	0.017
Lumbar puncture	Model 1	0.48	0.11	4.42	<0.0001	0.265	0.687
	Model 2	0.50	0.11	4.34	<0.0001	0.272	0.721
	Model 3	0.46	0.11	4.03	<0.0001	0.237	0.686
Blood draw	Model 1	−0.28	0.08	3.40	<0.0001	−0.449	−0.12
	Model 2	−0.06	0.09	0.71	0.476	−0.235	0.11
	Model 3	−0.07	0.09	0.74	0.458	−0.239	0.108

Latinx Group (N = 149)
Procedure Type						95.0% CI
		B	S.E.	t-value	p	Lower	Upper
PET scan	Model 1	−0 . 47	0.14	3.26	0.001	−0.753	−0.187
	Model 2	−0.34	0.14	2.36	0.018	−0.622	−0.058
	Model 3	−0.32	0.14	2.26	0.024	−0.605	−0.042
Lumbar puncture	Model 1	−0.25	0.15	1.70	0.088	−0.545	0.039
	Model 2	−0.21	0.15	1.44	0.150	−0.513	0.079
	Model 3	−0.20	0.15	1.33	0.183	−0.496	0.095
Blood draw	Model 1	−0.51	0.12	4.47	<0.0001	−0.744	−0.29
	Model 2	−0.41	0.12	3.54	0.0004	−0.638	−0.183
	Model 3	−0.41	0.12	3.52	0.0004	−0.637	−0.181

To be consistent with the extant literature, our reference group is a sample of non-Hispanic White individuals (N = 725).

Model 1: unadjusted.

Model 2 adjusted for age, gender, and educational attainment.

Model 3 further adjusted for RAQ scores.The bolded numbers depict significant difference.

The AA group expressed more willingness to undergo biomarker testing using PET scanning compared to the NHW group [p = 0.0001, CI: −0.52 to 0.17], though this difference was no longer statistically significant after adjusting age, gender, and education [p = 0.547, CI: −0.26 to 0.14]), and attitudes towards research [p = 0.45, CI: −0.27 to 0.12]. Similarly, the AI/AN group expressed more willingness to be tested using PET scans [p < 0.0001, CI: −0.63 to −0.22], but the difference did not reach statistical significance after adjusting for age, gender, and education [p = 0.14, CI:−0.38 to 0.05], and attitudes towards research [p = 0.07, CI: −0.41 to 0.02]. Compared to the NHW participants, the LTX respondents expressed significantly more willingness to engage in preclinical biomarker testing using PET scanning [p = 0.001, CI:−0.75 to −0.19]. This difference remained significant after adjustment for age, gender, and education [p = 0.018, CI: −0.62 to −0.06] and adjustment for attitudes towards research [p = 0.024, CI: −0.61 to −0.04].

The AA participants did not significantly differ from the NHW participants in terms of their stated unwillingness to engage in biomarker testing with lumbar puncture (ps ranged from 0.16 to 0.36). The AI/AN group was significantly less willing to engage in biomarker testing with lumbar puncture than the NHW group [p < 0.0001, CI: −0.58 to −0.30]. The difference in willingness remained statistically significant even after adjusting for age, gender, and education [p < 0.0001, CI: 0.27 to 0.72] and total RAQ scores [p < 0.0001, CI: 0.24 to 0.69]. The LTX group did not differ significantly from the NHWs in terms of their willingness to engage in biomarker testing with lumbar puncture in the unadjusted analysis [p = 0.09, CI: −0.55 to 0.04], the analysis adjusting for sociodemographic variables [p = 0.15, CI: −0.51 to 0.08], or the analysis further adjusting for RAQ scores [p = 0.18, CI:0.50 to 0.10].

AA respondents expressed significantly more willingness to engage in blood draws than NHW respondents [p < 0.0001 CI:−0.58 to −0.30]. The AA group's lower reluctance to engage remained significant even after controlling for sociodemographic variables [p = 0.045, CI: −0.32 to −0.002] and further controlling for total RAQ scores [p = 0.045, CI: −0.32 to = 0.003]. The LTX group displayed a similar pattern of willingness as the AA group, expressing significantly more willingness than the NHW group to do blood draws [p < 0.0001, CI: −0.74 to −0.29], a difference which remained significant after controlling for sociodemographic variables [p < 0.001 CI: −0.64 to −0.118] and for the RAQ [p < 0.001, CI: −0.64 to −0.18]. The AI/AN participants appeared more reluctant to engage in blood draws than the NHW participants [p < 0.0001 CI: −0.45 to −0.12], though there were no significant differences after adjusting for sociodemographic variables [p = 0.48 CI: −0.24 to 0.11] and RAQ scores [p = 0.46, CI:−0.24 to −0.11].

Discussion

This report is an update and extension of preliminary work,¹⁹ where we modeled the factors that affected the willingness of different ethnoracialized groups to engage in genetic and biomarker research. We were able to increase the number of AI/AN participants as well as include members from the LTX community. We also took a closer look at participants’ perceptions regarding three different biomarker testing procedures, to explore whether the procedure type would influence respondents’ willingness to participate in research. We specifically inquired about brain (PET) imaging, CSF collection (lumbar puncture), and phlebotomy (blood draws). Across the entire sample (n = 1825), respondents’ perceptions regarding potential burden, distress, and safety concerns influenced their relative willingness to engage in preclinical AD research. Consistent with Erikson et al.,²¹ the overall sample endorsed greater enrollment willingness for blood-based biomarker testing and brain imaging and considerably less willingness for the CSF biomarker testing (i.e., lumbar puncture).

However, we also observed significant ethnoracial group differences in willingness to engage in preclinical AD research. In contrast to prior research,²¹ the AA respondents did not differ from the NHWs in terms of their willingness to undergo CSF biomarker collection (i.e., lumbar puncture). After adjusting for sociodemographic factors and research attitudes, as measured by the RAQ, the AA and NHW groups did not differ in their willingness to undergo PET scanning. Contrary to Erickson et al.,²¹ we observed that the AAs were significantly more willing than the NHWs to undergo a blood-draw for preclinical AD research. Although the magnitude of the group difference decreased after accounting for age, gender, education, and attitudes toward research, the difference remained. The divergent findings between Erickson et al. and our study could be accounted for by differences in our sampling method. The participants in the Erickson et al.²¹ investigation were recruited from existing AD research cohorts, and therefore were already engaged in research activities. The participants in our study are a community-based sample. Our sample is more likely to represent community members that investigators will try to recruit for preclinical AD biomarker testing.

Our findings regarding the acceptability of blood draws are consistent with past investigations of other AA samples in which AA participants expressed greater willingness to provide blood samples for research,²⁵ and greater willingness to participate in studies requiring blood samples than brain imaging (MRI).²⁶ Similarly, our AA participants’ relative unwillingness to engage in preclinical AD biomarker studies using lumbar puncture is consistent with the findings of Howell et al.,²⁷ in which proportionally more AAs than NHWs raised concerns regarding lumbar puncture as a reason for nonparticipation.

To our knowledge, the UBIGR study is unique in its examination of AI/AN participants’ views regarding specific preclinical biomarker testing techniques. Relative to the NHW group, the AI/AN group was significantly less willing to undergo lumbar puncture; this unwillingness remained consistent after adjusting for various sociodemographic factors and research attitudes. There are different cultural values associated with different aspects of one's body, such as the sacredness of one's hair. This should be explored to better understand and respect the wishes of the AI/AN community. In contrast, while the AI/AN group initially appeared more willing to undergo brain imaging compared to the NHWs, the differences became nonsignificant after adjusting for socioeconomic factors and research attitudes. The comparison between AI/AN and NHW participants in terms of blood draws displayed a similar pattern, i.e., no significant difference after adjustments for age, gender, education, and research attitudes were made. Future work in this area would also benefit from examination of within-group differences in the AI/AN group, which is characterized by rich cultural heterogeneity.

The LTX group did not differ from the NHWs in terms of their willingness to engage in preclinical biomarker testing using lumbar puncture. Compared to the NHWs, the LTX group was more willing to undergo brain scanning, even after statistical adjustments. Similarly, the LTX group endorsed significantly greater willingness to undergo blood-based biomarker (BBBM) testing than the NHW group, a difference which survived adjusting for age, gender, education, and research attitudes. Ultimately, our data are not sufficiently granular to address why the LTX group may have expressed greater willingness to participate in biomarker research, but it is likely multifactorial. For example, LTX individuals are less likely to have access to insurance-covered health care²⁸ and research participation may be a means to gain access to expensive features of clinical care such as biomarker testing. Further research is needed in order to discern the underlying motivations. Replication is needed to determine the robustness of this particular finding.

In summary, we observed significant ethnoracial group differences in willingness to participate in preclinical AD biomarker testing. Whereas prior research from our group demonstrated that members from historically under-included groups were willing to engage in preclinical AD research, the present research advances this area of study by further exploring, with a larger and more diverse sample, relative preferences regarding biomarker collection procedures. Our investigation differs from prior investigations in several meaningful ways. First, the sample characteristics differentiate it from earlier work. While the overall sample size is large, it is notable that 60% of the sample is comprised of historically underrepresented participants. Equally important, our sample was community-derived, i.e., less than 10% of the participants were recruited from existing ADRD research cohorts, and our sample was generally middle-aged. Our sample was more diverse in terms of educational background and income level than most of those reported upon in the literature. We believe that this is a strength that allows us to generalize our findings to the wider population.

Another important difference we noted was that our sample displayed a greater range of responses on the RAQ. Although several published investigations^18,29 reported sample mean scores of 28–29, our sample mean was 27, indicating a less positive view of research overall. We observed that attitudes towards research, as measured by total RAQ scores, had a significant effect on willingness to engage in different types of biomarker procedures, over and above self-identified ethnoracial group. Perhaps the greater variability in RAQ scores permitted us to examine the mediating effect of attitudes toward research on willingness to engage in research. In contrast to Salazar and colleagues,¹⁸ we found that adjusting for total RAQ scores attenuated the differences between the AA and NHW groups in terms of likelihood to undergo PET scans. Similarly, adjusting for RAQ scores attenuated the differences between the AI/AN and NHW respondents in terms of willingness to undergo PET scanning and blood draws. Notably, we did not observe the same pattern of mediation among the LTX group. We believe that this may reflect the differential history of medical mistrust of the groups.

Advantages of the present study include a large sample, including, to our knowledge, one of the largest cohorts of AA and AI/AN participants to date. Comparing the present investigation to a larger sample of AA research study participants,²⁵ we noted some similarities. We, too, found that AA participants are generally willing to participate in health -related research such as biomarker studies. Although the Lang et al.²⁵ sample of AA participants was larger than our sample (733 versus 599), their investigation was drawn solely from the Triad region of North Carolina. In contrast, our sample was drawn from across the United States and may be more generalizable. Another advantage is our inclusion of a relatively younger sample of middle-aged individuals; most investigations of willingness to enroll in preclinical AD studies,¹⁸ engage an older sample, with a mean age of 60 or above. Future investigations of Black/AA participants should examine within-group differences, given the group's heterogeneity.

Limitations

The present study's cross-sectional design, while efficient for examining associations between attitudes and testing willingness, limits the ability to infer a causal relationship. Our use of an online survey enabled broad geographic representation, though online surveys may also introduce selection bias. Individuals without internet access or those less comfortable with digital technology may be underrepresented in our sample. Nonetheless, our findings were generally consistent with prior investigations that relied upon paper questionnaires³⁰ and telephone surveys,²¹ namely, as the perceived risk and burden of the biomarker collection method increased, participation willingness decreased. Our participants were asked to rate reasons for relative unwillingness to engage in biomarker testing. Consistent with the extant literature, they cited perceived invasiveness of the lumbar puncture procedure and safety concerns. However, one study limitation is that we did not specifically probe for concerns regarding radiation exposure.

Another limitation of our study is the relatively small number of LTX participants who were surveyed. The group subsumed under the term “Latino” or “Latinx” is considerably broad, including Mexican American/Chicano, Puerto Rican, and groups from Latin America and/or the Caribbean. The participants included in the present study may be a very heterogeneous group, differing in terms of country of origin and language preferences. Future work in this area would be enhanced by recruiting a larger sample that would allow examination of within-group differences in terms of subclassifications (i.e., Caribbean Hispanic, South American Hispanic, European Hispanic) and years of acculturation. Admittedly, we regard these findings as a small albeit important beginning step in terms of including a heretofore under-included community.

Conclusions

In the present study, survey respondents rated their relative willingness to engage with preclinical AD biomarker testing using different procedures. Considering all four ethnoracial groups, there was relatively greater endorsement for the blood-based biomarkers (BBBM) than the two more conventional procedure types, i.e., brain imaging and lumbar puncture.

Reliance upon plasma biomarkers may help improve the diverse representation of samples in AD.^6–8 Potential advantages of BBBM tests include less invasiveness, lower cost, and less dependence upon biomedical research infrastructure such as specialized facilities for diagnostic imaging.^7,31 Just as the integration of biomarkers into the clinical diagnosis of AD dementia was paradigm-shifting, we assert that BBBM in AD disease may be a “game-changer”. The findings from the current investigation suggest that BBBM may be better received by a more diverse group of community members. This is auspicious given the urgent need to validate AD biomarkers for under-included and underrepresented populations, to establish cut-points across groups who differ in terms of ethnoracial and socioeconomic backgrounds and concomitant social determinants of health.

Supplemental Material

sj-docx-1-alz-10.1177_13872877241307255 - Supplemental material for Perceptions about preclinical Alzheimer's disease biomarker collection procedure influences willingness to participate: Findings from an ethnoracially diverse study

Supplemental material, sj-docx-1-alz-10.1177_13872877241307255 for Perceptions about preclinical Alzheimer's disease biomarker collection procedure influences willingness to participate: Findings from an ethnoracially diverse study by Diane Carol Gooding, Carol A Van Hulle, Megan Zuelsdorff, Jordan P Lewis, Fabu P Carter, Hector Salazar, Shenikqua Bouges, Taryn T James, Alexander Gee and Carey E Gleason in Journal of Alzheimer's Disease

Footnotes

Acknowledgments

The authors appreciate the study participants and community liaisons who made this study possible. Special thanks to the community groups for their assistance with publicizing the study. We thank the anonymous reviewers whose incisive comments enhanced the quality of our work.

ORCID iDs

Diane Carol Gooding

Shenikqua Bouges

Carey E Gleason

Author contributions

Diane Carol Gooding (Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Writing – original draft); Carol A Van Hulle (Data curation; Formal analysis; Writing – original draft); Megan Zuelsdorff (Writing – review & editing); Jordan P Lewis (Writing – review & editing); Fabu P Carter (Project administration; Resources; Supervision); Hector Salazar (Resources); Shenikqua Bouges (Writing – review & editing); Taryn T James (Writing – review & editing); Alexander Gee (Writing – review & editing); Carey E Gleason (Conceptualization; Funding acquisition; Methodology; Resources; Supervision; Writing – review & editing).

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Department of Medicine Intramural Faculty Competition Grant (CEG, DCG) and the Leon Epstein Faculty Research Award (DCG). Additional funding was provided by NIH-NIA R01AG054059 (CEG, PI). NIH-NIAP30AG062715 (ADRC Center Grant) and NIH-NIAR01AG027161.

Declaration of conflicting interests

Megan Zuelsdorff is an Editorial Board Member of this journal but was not involved in the peer-review process or had access to any information regarding its peer-review.

The remaining authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability

The data supporting the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Supplemental material

Supplemental material for this article is available online.

References

Califf

. Biomarker definitions and their applications. Exp Biol Med (Maywood) 2018; 243: 213–221.

Jack

Bennett

Blennow

, et al. NIA-AA Research framework: toward a biological definition of Alzheimer’s disease. Alzheimers Dement 2018; 14: 535–562.

Mayeda

Glymour

Quesenberry

, et al. Inequalities in dementia incidence between six racial and ethnic groups over 14 years. Alzheimers Dement 2016; 12: 216–224.

Kornblith

Bahorik

Boscardin

, et al. Association of race and ethnicity with incidence of dementia among older adults. JAMA 2022; 327: 488–1495.

Canevelli

Bruno

Grande

, et al. Race reporting and disparities in clinical trials on Alzheimer’s disease: a systematic review. Neurosci Biobehav Rev 2019; 101: 122–128.

Gleason

Zuelsdorff

Gooding

, et al. Alzheimer’s disease biomarkers in Black and non-hispanic White cohorts: a contextualized review of the evidence. Alzheimers Dement 2022; 18: 1545–1564.

Karikari

. Blood tests for Alzheimer’s disease: increasing efforts to expand and diversify research participation is critical for widespread validation and acceptance. J Alzheimers Dis 2022; 90: 967–974.

Kang

Korecka

Lee

, et al. Alzheimer disease biomarkers: moving from CSF to plasma for reliable detection of amyloid and tau pathology. Clin Chem 2023; 69: 1247–1259.

Passmore

Kisicki

Gilmore-Bykovskyi

, et al. There's not much we can do…” researcher-level barriers to the inclusion of underrepresented participants in translational research. J Clin Transl Sci 2021; 6: e4.

10.

Lincoln

Chow

Gaines

, et al. Fundamental causes of barriers to participation in Alzheimer’s clinical research among African Americans. Ethn Health 2021; 26: 585–599.

11.

Boyd

Railey

Kirkpatrick

, et al. Communication about Alzheimer’s disease and research among American Indians and Alaska natives. Alzheimers Dement 2022; 8: e12302.

12.

Heintzelman

. The Tuskegee Syphilis Study and its implications for the 21^st century. New Soc Worker 2003; 10: 1–2.

13.

Alsan

Wanamaker

. Tuskegee and the health of black men. Q J Econ 2018; 133: 407–455.

14.

Skloot

. The immortal life of Henrietta lacks. New York: Crown Publishing Co, 2010.

15.

Foulks

. Misalliances in the barrow alcohol study. Am Indian Native Alaska Mental Health Res 1989; 2: 7–17.

16.

Santos

. Genetic research in native communities. Prog Comm Health Partnersh 2008; 2: 321–327.

17.

Garrison

. Genomic justice for native Americans: impact of the Havasupai case on genetic research. Sci Technol Human Values 2013; 38: 201–223.

18.

Salazar

Hoang

Gillen

, et al. Racial and ethnic differences in older adults’ willingness to be contacted about Alzheimer’s disease research participation. Alzheimers Dement 2020; 6e: 12023.

19.

Gooding

Carter

Umucu

, et al. Factors affecting the willingness of African-American and American Indian/Alaska native communities to engage in genetic and biomarker research: the UBIGR study. Biomark Neuropsychiatry 2024; 10: 100090.

20.

Rubright

Cary

Karlawish

, 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: 63–68.

21.

Erickson

Chin

Ketchum

, et al. Predictors of willingness to enroll in hypothetical Alzheimer disease biomarker studies that disclose personal results. Alzheimer Dis Assoc Disord 2022; 36: 125–132.

22.

Caban-Holt

Cuccaro

Lloyd

, et al. Attitudes and perceptions about brain donation among African Americans: implications for recruitment into Alzheimer’s disease research. J Alzheimers Dis 2024; 97: 1621–1627.

23.

Van Buuren

Groothuis-Oudshoorn

. Mice: multivariate imputation by chained equations (MICE) in R. J Stat Softw 2011; 45: 1–67.

24.

Shan

Gerstenberger

. Fisher’s exact approach for post hoc analysis of a chi-squared test. PloS One 2017; 12: e0188709.

25.

Lang

Kelkar

Byrd

. African American participation in health-related research studies: indicators for effective recruitment. J Public Health Manag Pract 2013; 19: 110–118.

26.

Byrd

Edwards

Kelkar

, et al. Recruiting intergenerational African American males for biomedical research studies: a major research challenge. J Natl Med Assoc 2011; 103: 480–487.

27.

Howell

Parker

Watts

, et al. Research lumbar punctures among African Americans and Caucasians: perception predicts experience. Front Aging Neurosci 2016; 8: 296.

28.

Terriquez

Joseph

. Ethnoracial inequality and insurance coverage among latino young adults. Soc Sci Med 2016; 168: 150–158.

29.

Zhou

Elashoff

Kremen

, et al. African Americans are less likely to enroll in preclinical Alzheimer’s disease clinical trials. Alzheimers Dement 2017; 3: 57–64.

30.

Nuño

Gillen

Dosanjh

, et al. Attitudes toward clinical trials across the Alzheimer’s disease spectrum. Alzheimers Res Ther 2017; 9: 81.

31.

Suridjan

van der Flier

Monsch

, et al. Blood-based biomarkers in Alzheimer's disease: future directions for implementation. Alzheimers Dement (Amst) 2023; 15: e12508.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB