Underrepresented voices in a Colorado Biobank: Perspectives from focus groups on motivations, return of results, and data sharing

Summary

Most participants in large cohorts, such as biobanks, are of European descent. This lack of representation has been an ongoing challenge in genomic research. Understanding the perspectives on genomics research and participation in biobanks of historically underrepresented populations could provide insight into ways to better engage with these groups. We conducted a series of virtual and in-person focus groups with individuals who self-identified as American Indian or Alaska Native (AI/AN), African American/Black (AA/B), or Hispanic/Latino (H/L) and who were enrolled in the Colorado Center for Personalized Medicine (CCPM) biobank. The focus group discussions were centered on participant experiences, including but not limited to their motivations, return of results, and data sharing. There was a total of 23 participants across the six focus groups. The majority of participants identified as AI/AN (60.9%), followed by H/L (39.1%), and AA/B (21.7%); many participants identified with multiple race/ethnicities. The motivations for participating in the biobank included the potential to advance science and health, the potential for return of results, to learn more about one’s ancestry, and a few indicated that they were interested in helping the biobank be more representative of all populations. Notably, many expressed positive feedback of the focus groups and felt that their views were valued, illustrating the importance of community-centered work. Our findings can be used to guide recruitment and engagement of biobank participants, especially from diverse backgrounds, contributing to enhanced partnerships advancing knowledge and healthcare.

Historically underrepresented biobank participants described motivations for genomics research participation, including advancing science, receiving results, and learning about ancestry. Focus groups with American Indian/Alaska Native, African American/Black, and Hispanic/Latino participants highlight the value of community-centered engagement to strengthen trust, representation, and long-term biobank partnerships.

Introduction

“This is the first time I've ever been asked about anything in terms of my healthcare and my perspective on being Native and everything.”

(Focus Group 2, Participant 086)

Elevating research participants’ voices, such as the one above, can empower communities and improve genetic research. This is critical given the advances in the rapidly growing field of precision medicine, which aims to integrate an individual’s clinical, genetic, and environmental information to advance health and care.¹ Inclusion of diverse populations is essential, both in genetic variant discovery and the clinical implementation of precision medicine tools, such as the return of pharmacogenomic results that can optimize patient outcomes.^2,3,4,5,6,7

Biobanks are valuable resources which store and collect biological samples and other pertinent phenotypic/medical data for advancing research in precision medicine and biomedical sciences.⁸ Although the upfront investments required to establish a biobank are substantial, these resources are critical for advancing scientific research. The potential to individualize care can incentivize individuals to participate in genomic research and biobanks. The participation of diverse individuals has significantly contributed to advances in genomic medicine, as shown by a few examples: H3Africa consortium,⁹ All of Us research program,¹⁰ and more recently in the Mexico Biobank.⁶ Despite these advances, most participants in large research cohorts and biobanks are of European descent and not representative of all individuals.^11,12,13 This lack of representation, which often overlaps with individuals from marginalized communities, is a persistent challenge in genomic research.^2,13,14

The ongoing exclusion and underrepresentation of diverse populations can lead to the unequitable implementation of precision medicine efforts, potentially exacerbating health disparities.^15,16,17 Studies have shown that individuals from historically underrepresented (or unrepresented) communities have specific concerns regarding genomics research. These include but are not limited to historical abuse^8,18,19 and research misconduct. Other common barriers to participation include lack of awareness, accessibility, or minimal perceived benefit to participating in genomic research.^20,21,22 This is compounded by research failing to align with community values, priorities, or research interests.^23,24 These concerns are reflected in the lack of diversity in large genomics biobanks and consortia, which, despite efforts to recruit individuals from underrepresented backgrounds, still fall short of adequately representing the communities they aim to serve.^2,8,15,23 Additionally, ethical questions remain regarding implications of genetic return of results and data sharing (e.g., types of research findings that should or should not be shared, data storage, and access). Previous research identified mixed outcomes regarding the extent that return of results and/or data sharing contribute to participation choices in biobanks^25,26,27,28 and rarely, if ever, prioritize views from underrepresented groups.

To support better research alignment and engagement with people of diverse backgrounds, we conducted virtual and in-person focus groups with individuals who self-identified as American Indian or Alaska Native (AI/AN), African American/Black (AA/B), or Hispanic/Latino (H/L) and who were enrolled in the Colorado Center for Personalized Medicine (CCPM) biobank. The CCPM biobank, developed in 2015, is an institutional collaboration between the University of Colorado Health System (UCHealth) and the University of Colorado Anschutz to serve as a unique, dual-purpose research and clinical resource accelerating precision medicine.^29,30,31

More than 240,000 UCHealth patients have enrolled in the CCPM biobank. The CCPM biobank supports scientific discovery, collaboration with partners, and actionable clinical results returns from a clinically certified setting and conducts limited genetic testing on selected genes associated with disease risk such as cancer, heart disease, and pharmacogenetics.²⁹ Biobanks, including CCPM, offer the ability to recontact, providing a unique opportunity to engage participants in a variety of research projects. There has been previous success in recontacting research participants broadly^{10,29,32,33,34} as well as with CCPM biobank participants,^35,36,37 but these efforts often engaged primarily populations of European descent.

Importantly, our focus groups provided an opportunity to recontact and engage with underrepresented individuals who have agreed to participate in the CCPM biobank (and have their genomic information generated) to further understand their perspectives as biobank participants. Our goal is to understand underrepresented individuals’ perspectives on bioethical issues related to their motivations for joining, return of results, and data sharing in the CCPM biobank. Ultimately, this research will contribute to evidence-based approaches to support participant recruitment as well as promote ethical, meaningful, and sustained engagement of participants from historically underrepresented communities.

Subjects, material, and methods

Research team and reflexivity

We are a team of researchers and community members from various academic, cultural/racial/ethnic, and professional backgrounds, and we recognize that our individual perspectives shape the research process, including the framing of our questions, methodologies, and data interpretation. The research team consisted of women and non-binary individuals, many of whom are from historically underrepresented populations, including Indigenous and Hispanic/Latino. Our viewpoints are informed by our origins, education, and community connections, which guide our engagement and dedication in this research.

Recruitment

The composition of the CCPM biobank is similar to the UCHealth patient population,²⁹ which, given UCHealth systems’ large catchment area, includes diverse demographics (Table S1). This study was reviewed and approved by the Colorado Multiple Institution Review Board (COMIRB #24-0141). We employed a targeted sampling approach,³⁸ where focus group participants were randomly selected from a subset of CCPM biobank participants who identified as AI/AN, AA/B, or H/L. From April to June 2024, a total of 2,400 randomly selected CCPM biobank participants who were identified as AI/AN, AA/B, or H/L based on electronic health record (EHR) data (800 from each racial and/or ethnic group over the course of 8 weeks) were sent an initial recruitment email by the CCPM biobank participant engagement team in English (Figure S1), with a link to a Spanish translation of the materials. Attached to the email was a flyer inviting participants to complete an eligibility survey. The eligibility survey asked whether participants were at least 18 years old and whether they were participants in the CCPM biobank. Acknowledging that race and ethnicity information in EHRs can differ from individuals’ self-identified information, the eligibility survey also included a race and ethnicity question to ensure that participants self-identified as one of the three populations of interest (AI/AN, AA/B, H/L) or across multiple groups. If individuals met the eligibility criteria, they selected an available date/time to attend a focus group and received the focus group question guide, consent form, and a demographic/knowledge survey. Invited participants attended one of the six focus groups, either in person or virtually. All participants were compensated with a $60 gift card after participating in a focus group.

Although we utilized random selection within a targeted sampling approach, we acknowledge that self-selection is a known factor in studies with voluntary participation and that our participants may have volunteered based on pre-existing knowledge in these areas or a stronger desire to discuss these issues.

Study design

The study featured focus group discussions and included supplemental mixed methods using surveys and ranking activities (Figure 1). The surveys and ranking activities were used to keep track of demographics as well as to compare and contrast our findings to previous studies (e.g., return of results ranking activity) and help us identify priority areas for subsequent discussion during the focus group. The semi-structured facilitator guide was piloted by the designated facilitator with four consultants who came from similar racial and ethnic backgrounds as our target demographic and who were not part of our research team. These individuals had expertise in health research and were affiliated with the University of Colorado Anschutz. The four consultants were compensated with $60 gift cards. Revisions were made to the facilitator guide following this pilot test.

Figure 1.

Overview of the focus group activities

This figure provides an overview of the focus group discussion prompts and activities. The event began with oral consent, review of goals and formed group agreements, and then went into questions around participants’ experiences, motivations, and engagement with the CCPM biobank. Then, a return of results activity was presented, followed by questions around data sharing, and then the focus groups ended with any final thoughts. For more details, please see the sections subjects, material, and methods and supplemental information containing the focus group guide.

Written consent was obtained prior to the start of each focus group and was facilitated by two research team members. In addition, verbal consent was collected to agree to audio and visual recording. Focus groups were led by a facilitator who holds a PhD in sociology and has extensive experience with qualitative research, community-based participatory research, Indigenous knowledge systems, and research methodologies.

Before starting the focus groups, participants were asked to complete two surveys. The first survey included 12 questions related to demographics (e.g., age, race/ethnicity, education), previous research participation, and biobank preferences (e.g., receiving genetic results). The second survey was a knowledge assessment derived from a previous CCPM biobank survey with 15 true or false questions to gather information related to participant knowledge about genetics.³⁷ Previous research has shown the importance of genetics knowledge in making informed choices regarding patients’ healthcare, particularly in personalized medicine efforts.³⁹ In 2019, Schmanski et al. evaluated genetics knowledge in CCPM biobank participants; no differences were reported by gender, race, or ethnicity.³⁷ However, this previous knowledge assessment was limited to individuals identified as White and not Hispanic/Latino. Therefore, we employed the knowledge assessment to evaluate our participants’ familiarity and knowledge of genetics and to see whether there were associations we could identify between demographics as well as experiences in the biobank.

Focus groups began with participant introductions, followed by an overview of the study goals and biobank. The facilitator then led a semi-structured discussion using the piloted facilitator guide, which included 17 questions on participant perspectives, return of results, data sharing, and final thoughts. There were two notetakers/co-facilitators present at each focus group. The return of results section included questions regarding preferences and a ranking activity (adapted from Lewis et al.⁴⁰) led by co-facilitators. Participants were asked to rank genetic result characteristics (e.g., “There’s something a person can do to lower their risk of getting the condition”) into “most important,” “least important,” and “somewhat important,” as well as rank their own top three “most important” characteristics, acknowledging that all characteristics may be important for participants.

Five of the six focus groups were virtual per participants’ preferences and held over Zoom; one focus group was in-person. Audio and visual recording were used to collect data. Field notes were made during and after the focus groups. Focus groups were scheduled for 2.5 h and were conducted in English. Transcripts were generated by Rev.com and were de-identified. To promote participant validation, participants were provided with a short report synthesizing preliminary findings (August 2024), an opportunity to attend optional one-hour meetings to discuss study results (December 2025), and to provide feedback on the manuscript (March 2026).

Qualitative analysis

Content analysis through mixed grounded theory^41,42 was the methodological orientation used to underpin the study. Our final codebook contained 14 codes to capture relevant themes around participant engagement, return of results, and data sharing. Several codes were initially identified a priori to meet the research study aims, and emergent codes were additionally derived from field notes taken during the focus groups and, later, from transcript analysis. The codebook was tested iteratively through several phases wherein all study team members would code sections of the transcripts and compare the coded transcripts in group meetings. This enabled the team to adjust code definitions and reduce the risk of ambiguity during the coding process as well as determining which codes may not be needed and which might need to be added or collapsed with others. In addition, two independent coders identified emergent themes during the coding process, and these emergent codes were added to the codebook following discussion with the full research team. Dedoose (version 9.2.22) was used to qualitatively code the focus group transcripts. Using the final phase of the codebook, two coders independently coded the transcripts, and then a consensus meeting was held between the coders. The consensus meetings involved reviewing each code application to ensure consistency. When the coders differed in the codes they applied, each would give their rationale for their coding decisions, and the two would come to consensus on which code application was appropriate. At times, the entire research team was involved in code discussions as well. The actual application of different codes declined sharply after the first transcript was coded, and consensus mainly involved deciding the length of excerpts. Two separate individuals acted as tiebreakers when needed for consensus. Coders were additionally free to mark excerpts out of personal interest for future exploratory analysis, which allowed study team members to track areas they were interested in developing further while keeping main coding in line with study goals. Literature on focus group saturation in qualitative health care research indicates that between three and six focus groups are sufficient to capture theoretical saturation of greater than 90%.^43,44 Analysis during the coding process indicated that saturation had been reached in line with this expectation, as key concepts were discussed across multiple focus groups and no new themes were generated by the time of the last focus group.

Quantitative analysis

Analysis of survey results included demographic data, knowledge-assessment findings, and the return of results activity survey. The demographic survey included a question regarding participant preferences for delivery of genetic test results; these preferences are illustrated on a frequency bar graph (Figure S2). The knowledge assessment was analyzed to determine the number of times participants responded to a given question correctly, incorrectly, or chose “unsure” (Figure S3). An independent-samples t test was used to evaluate differences in knowledge check scores by demographic categories (Table S2). The return of results activity also included a Google JamBoard (discontinued) that tracked group consensus, which was summarized manually for each focus group. Based on the group’s consensus placement, each result characteristic type was assigned a value 1–5: 1 = least important, 3 = somewhat important, 5 = most important, with tie values allowed. All results were summarized and visualized using R version 4.4, with values less than five masked.

Results

Demographic data and recruitment

Of the 2,400 CCPM biobank participants (800 from each population of interest) contacted via email, 76 completed an eligibility survey (3.2% response rate; Figure S1). Of the 76 individuals who completed the eligibility survey, 53 were eligible to participate based on their eligibility survey responses (2.2% eligible of 2,400 contacted participants during recruitment). Notably, no participants opted to complete the eligibility survey in Spanish. Of the 53 eligible individuals, 38 were scheduled for a focus group, and 23 individuals attended one of the six focus groups.

Participant demographics across all focus groups are described in Table 1. Approximately 60% of participants identified as AI/AN or Indigenous (some individuals self-identified as Indigenous but not necessarily AI/AN), 40% identified as H/L, and 22% identified as AA/B. Almost one-quarter (22%) of participants identified with multiple racial and/or ethnic groups including at least one of the eligible populations. Over half of participants were over the age of 46 years, and 65% identified as female. About half of participants had a bachelor’s degree and higher (52%) and had previously completed other types of genetic testing (57%, e.g., commercial direct-to-consumer genetic testing). Most participants (78%) had not been involved in other biobanks. On average, participants correctly answered 72% of genetic knowledge questions (minimum = 33%, maximum = 87%; Figure S3). There were no differences in genetic knowledge by any demographic group (Table S2).

Table 1.

Demographics of the 23 focus group participants

Demographic variable	N (%)
Age (years)
18–45	7 (30.4%)
46+	16 (69.6%)
Gender
Female	15 (65.2%)
Male	8 (34.8%)
Race/ethnicity group
African American or Black	5 (21.7%)
Hispanic/Latino	9 (39.1%)
American Indian/Alaska Native or Indigenous	14 (60.9%)
Multiple racial and/or ethnic groups	5 (21.7%)
Education
Bachelor’s degree and higher	12 (52.2%)
High school, associate, or technical degree	11 (47.8%)
Other genetic testing
No	10 (43.5%)
Yes	13 (56.5%)
Other biobank
No	18 (78.3%)
Unsure/yes	5 (21.7%)

Focus group participants were asked to describe their initial encounter with the CCPM biobank to better understand when, where, and how participants joined. There was a wide range of time since self-reported recruitment, spanning from 2014 to 2024. Nine participants reported joining between 2014 and 2020, seven between 2021 and 2024, and seven participants could not remember when they joined or did not indicate when they did.

In general, participants were recruited under two major recruitment methods—either at a medical facility during an appointment (e.g., routine exams, research, emergency medicine) or joined on their own after receiving an invitation to join via email or through the UCHealth mobile application. The circumstances in which participants enrolled at a medical facility varied. Many participants described specialized medical doctor visits (diabetes, injury, cancer, etc.) while others described routine doctor check-ups or medical visits due to COVID-19 treatment or other research studies. Several enrolled through My Health Connection, a mobile or web-based application, which provides UCHealth patients access to health records and an appointment scheduler, among other useful tools,⁴⁵ and a few participants self-enrolled after being notified via email. Some participants explicitly did not remember how they were recruited or how they signed up to be part of the CCPM biobank.

Motivations for joining the CCPM biobank

We identified three major motivations for joining the biobank, none of which are mutually exclusive: altruism, return of results, and an interest in increasing biobank diversity.

Altruism was the most frequent motivation identified in our focus groups. This motivation is characterized by participants speaking about a desire to donate a sample to help advance science or contribute to “greater good.” When reflecting on their feelings about joining the biobank, one participant stated, “If it helps someone, I want it to be available to them. And I don't care if it's a family member or someone a million miles away. If it helps people, I hope it's available to them” (FG3, PART013).

Notably, some participants voiced that they joined the biobank to gain more information about conditions they had been diagnosed with or for which they were seeking diagnosis. These participants were motivated by a desire to have their genetic data as a tool to guide clinical care or personal research into their conditions. One participant described using personal results in conversations with doctors:

Or the benefit of like, my, my reading about what they sent back to me and being able to inform my doctor and say, you might not wanna do this later on if you decide to do this <laugh>. You know, so itself, I think it helps encourage dialogue that way (FG2, PART094).

Following this, the same participant went on to state that, “as we know, if we don’t help ourselves then nobody will help us anyway” (FG2, PART094). This quote points to a perception within this focus group that this type of research and self-advocacy is necessary and beneficial to receive proper care.

There was also interest in receiving genetic test results back to learn more about oneself and one’s family. When reflecting on one’s motivation to join the biobank, one participant explained that, “I might as well provide some information so I can have some information back. You know, you kind of turn it around to make it positive for yourself” (FG2, PART094). There was also a recognition of how the results could be used for one’s care; one participant discussed involvement with a university as an expert patient partner, a background that gave them insight into the benefits of the research process. The participant stated, “I’m familiar with research being done in a lot of areas, and so that made me more willing to be participating in research because I’ve seen how research really affects the outcome of patient care” (FG3, PART119), highlighting how previous exposure and research familiarity can impact perspectives, trust, and willingness to participate in research.

Uniquely, several participants spoke specifically to the fact that they were motivated by a desire to increase diversity within the biobank. These participants correctly identified the fact that biobanks tend to consist of primarily European-descent individuals and chose to enroll to mitigate perceived inequities. As one participant explained, “My assumption is there's not a lot of people that look like me that have information in the bank. And so it might be helpful for research in terms of people that look like me” (FG5, PART964). An Indigenous participant in a different focus group said that “Well, I know us as Native Americans, there's different things about us. You know, our blood and some certain things we carry, certain problems we share like with sugar problems, diabetes and all this stuff. So I do know that we're different from other races of people” (FG1, PART971). While rooted in altruism, this motivation is distinct in its focus to improve research outcomes for underrepresented groups.

Return of results

Participants were asked to rank eight characteristics of the types of results one could hypothetically receive (Figure 2A). “The condition is life threatening” was chosen as the most important from both individual and group rankings (Figure 2B). “Affects similar age or older” was chosen as the least important from the individual and group rankings. Importantly, a “least important” ranking does not imply total lack of importance. In individual ranking, participants were allowed to indicate all characteristics as “most important,” and many participants indicated challenges with both individual and group ranking. To illustrate, one participant indicated,

I'm not sure if I understood the questions, but to me, I ranked all of those most important. They're all very important to me in, in their own ways. If this program could help with one of those questions and give you that information, that's, that's terribly important … (FG6, PART062).

Figure 2.

Summary of rankings of hypothetical characteristics of results

(A and B) Participants were asked to rank eight characteristics of results. Individual rankings are shown in (A), where light blue = most important, green = somewhat important, and yellow = least important. Only each individual’s top three most important characteristics of results are included in dark blue in (B), which also shows group ranking of each focus group’s top three most important characteristics in pink, represented as a percentage of the possible choices (total participants/total focus groups). Note that focus group one’s responses to “prevalent in community” are not included here as the question was changed after that group.

Thus, rather than presenting the participants’ discussion around the return of results activity as a hierarchy of importance (i.e., most to least importance), we focus on the factors and reasoning participants considered when doing the activity.

Participants most frequently considered the actionability of results and the potential relevance of the results at the individual, familial, and communal levels. One participant shared that having information would be helpful to prevent harmful experiences:

For me, I have … very primary experiences. You know … I have experiences with not being with my mental health, not being able to think clearly. So, if there's other stuff that I could be warned about that I would want to know that, 'cause the experiences I've had with this is not a good one and I want avoid it (FG1, PART061).

Another participant was interested in actionable results that could improve one’s life expectancy and quality of life, explaining that “If I can correct what's going on, if I can make myself a lot more healthier, if there's some things that the doctors know that, hey, you do this, we can, we can see, see you making it to 85 or, you know, 90” (FG6, PART457).

Many participants also highlighted the importance of results for their family, as one participant indicated, “You find out you've got something, but there's, you know, my sons have the ability to be able to do something in their lifestyle early on to prevent them from having that condition, then I would say that's pretty damn important” (FG4, PART378). While participants recognized the relevance of results for themselves and family, there was also interest in results for their community. One participant stated, “I would wanna know that too, so I can educate people in my community that we're predisposed to certain things or we need to be aware of certain things” (FG5, PART966). Hence, actionability and relevance of results were key factors in return of results to participants.

Yet there was also recognition that there are challenges and barriers to access, such as being able to afford medical care. One participant jokingly stated that “we really don’t have the funds” for mental health care (FG4, PART121), which highlights a challenge in funding after obtaining return of results. There was also consensus across focus groups that as one gets older, it is expected that health burden would also increase. Moreover, while there are preventive actions people can take, participants noted that people in general may not take these steps, explaining that, “Well, I think a lot of it ties into the mental health too. There’s a lot of things people can do to quit drinking, but do they do that?” (FG2, PART086).

Participants also had variable levels of interest in ancestry-related results, and the discussions around ancestry brought up some unique points. Ancestry was initially perceived by many participants to relate only to lineage and ethnic background, while later the addition of the word “genetic” to ancestry results opened the discussion to how genetic ancestry is perceived based on this subtle distinction. For many participants, ancestry and genetic ancestry are not the same; the former is tied to race or ethnicity (which participants expressed knowledge and understandings of), while the latter is often associated with health outcomes. Therefore, genetic ancestry may be regarded with more importance because it could impact health research. Below we outline some of the differences in perspectives when talking about ancestry vs. genetic ancestry.

Individuals interested in ancestry information were curious to know more about their biological family, especially those participants that were adopted. Others were interested in ancestry to learn more about themselves, for example,

I didn't express this at the beginning but I'm half Cherokee, my mom's full blood and my dad is full Euro mix of whatever. And I kind of just wanted to know where I ended up falling in that spectrum of that mash of genetics, <laugh>. And haven't really done it. It's just something that's always like in the back of my head (FG1, PART061).

However, for some participants, ancestry results may seem unnecessary as some expressed that they are confident in or well-aware of their ethnic background. One participant illustrated this point well, stating that, “Well, for me, I was looking at the ancestry thing as far as ethnicity and that kind of thing. And so it wasn't that important to me because I know my ethnicity …” (FG3, PART119). Ancestry return of results can also be perceived as a challenge to an individual’s identity, as one participant stated, “I think, I always hear my husband when I talk, when people say stuff like that, he is like, ‘I don't need you to tell me who I am.’ You know what, he gets mad, you know?” (FG2, PART086). Lastly, there is a level of distrust, a belief that ancestry results are not reliable or can be incorrect: “Yeah. To be fair, I also have some distrust of some scientist telling them, ‘Well, this is what your ancestry tells you.’ Really? I don't think so” (FG2, PART094).

When the research team specifically asked about genetic ancestry results, many participants took a more positive view of the potential results, including those that had expressed disinterest in ancestry results earlier. The inclusion of “genetic” was referred to as a keyword by one Indigenous participant in separating data-driven science from the subjective nature of the broader ancestry concept. Later in that conversation, a different participant clarified similar ideas:

No, I think that's what I, that's what I meant by genetics. I mean, like, if the data that shows this is where this is coming from, or this is how this is expressed, these kind of traits are expressed because of this. Not this, you know, this makes you this kind of Indian. I'm not, I'm not that. I don't need to know that (FG2, PART086).

Overall, there were mixed views into the importance of ancestry-related results, where many participants were interested in ancestry for health-related implications and less so to learn more about one’s background.

Data sharing

Participant opinions on data sharing are crucial to understand, especially for underrepresented communities who have been harmed by unethical research practices in the past. We note that participants had different understandings of what types of data might be available to researchers or otherwise potentially leaked from the biobank. Many made a strict divide between data that only included genetic information versus sharing data that was linked to their EHRs or other personally identifiable information.

Many were supportive of data sharing to contribute to health research, a point of view that aligned with the altruistic motivations frequently seen in our participant group. As mentioned above, these positive views on data sharing applied only for de-identified data, which are what the biobank shares with approved researchers. The idea of data being shared for approved research purposes was frequently contrasted against the idea of data being shared for profit. One participant stated that personal comfort levels with data sharing depended entirely on who it was being shared with:

If it’s being shared with the scientific community for science, sure. You know, if it is just being given out to the highest bidder, forget it, you know? That’s where that fear of that compensation comes in. Okay. So why do you want my data, you know? (FG2, PART094).

A few participants also mentioned that time and education influenced their perspectives on participation and data sharing. One stated that, “I think when I was a little bit younger and I very much was just such a private person, I wouldn't have been so open to having my blood drawn or giving samples. But once I started to understand how beneficial it is to my health and my family's health, my neighbor’s health, it really just opened up my mind.” (FG3, PART182) The desire to help others is a frequent driver for participants and has an impact on many aspects of the biobank experience.

Participants were aware of potential risks but also discussed their willingness to contribute their data despite this. Many described weighing potential risks to their data against the potential benefits that might arise from participation, including benefits to self and others. The participants’ individual willingness to participate and share their data was frequently juxtaposed with their opinions that their friends and families might not participate in research like this for various reasons. One participant stated that, “I think it's lack of trust in the research community. You know, like with the Tuskegee study and things like that, I'm just kind of scanning like all my friends in my mind and none of them would do it” (FG5, PART513). The participant then ascribed their own willingness to participate in research to their experience working at a university, explaining that, “I have maybe a little more trust in or, you know, I'm willing to like, kind of trust the process a little bit more” (FG5, PART513). Another participant mentioned being comfortable with participating in the biobank but understood why other community members would be wary of participation. This participant had to explain why they joined the biobank to their community due to stigmas from data misuse in historically marginalized communities:

I know that the people in my community feel very protective of themselves … the first thing that comes out is how do you know it's not gonna be used against you? And the next thing is that somehow it'll be used against my community in general, like saying Latina people are this way or that way in a derogatory sense … all of these concerns that we'd have to crawl over in order to be able to get people to participate (FG3, PART119).

Notably, this participant acknowledged these concerns yet suggested they could be overcome to encourage participation. Despite discussing an important risk to their community, the participant is still supportive of research participation. This characteristic was shared across our participant population, who largely focused on the research benefits of data sharing as opposed to the risks. Even though individuals engaged in risk analysis before participating, they decided that participation was still the best decision due to potential benefits.

Additional concerns about data access and use outside of the research context were shared by participants. While participants were comfortable with sharing de-identified data for research use, there were intricacies that arose when discussing results linked to their EHR. Several participants voiced concerns about how their genetic data might negatively impact health insurance or life insurance coverage. One participant shared that,

The only thing that freaked me out a little bit is because I was applying for a thing, life insurance … And they said like, we're gonna request your genetic information. And I'm like, are they gonna start like, not giving me certain benefits or like rejecting me from getting something because I have something? You know, like if I wouldn't know, they wouldn't know. But then since I signed myself into like the biobank and other people can get my genetic information (FG5, PART966).

This sentiment echoed across several focus groups. Some participants were vaguely aware that protections exist for genetic data in these cases, but the lack of clarity and knowledge on the topic contributed to confusion and unease. An additional concern was raised about the use of genetic data by law enforcement, and participants worried that they might get falsely accused of a crime due to forensic analysis.

Some participants reasoned that their data may already be available online. One participant reflected on lived experiences and prior knowledge of data breaches in research settings and how it influenced their perspectives, stating, “It’s probably out there anyway, so I don’t care” (FG5, PART513). A participant with a military background shared that the government took many blood samples from enlisted personnel, which influenced the participant’s views on data sharing. The participant stated, “You’re wondering what they’re doing with it … I’m kinda wary, but I really can’t do nothing ‘cause they got everything they need, I guess, by now …” (FG1, PART982). While participants acknowledged that some of their information or data may be available online, they were still overall open to data sharing for research purposes.

Key takeaways from focus groups

We summarized the possible benefits (Table 2) and risks (Table 3) participants raised regarding biobank participation. One key benefit noted was that donating samples could advance knowledge and positively impact future generations (Table 2). However, some participants also raised concerns about a lack of transparency from the biobank and how this could contribute to a lack of trust in research (Table 3).

Table 2.

Potential benefits and opportunities of participating in biobank research with supporting quotations

Category	Benefits and opportunities	Quotation
Research impacts	•Donating samples could have meaningful and rewarding impacts on research for generations to come	•“So maybe something in a hundred, 150 years from now they find in my blood that they won’t find anywhere else.” (FG2, PART086)
	•Donating samples could provide tangible benefits for underrepresented populations	•“The benefit I would say for my community would be for there to be more research on whether it is something that’s more likely in the Latinx community … in Mexico there’s way less research, way less funding.” (FG4, PART247)
Return of results	•Return of personal results might inform future medical care	•“I thought it was important to know if something runs through my family line. Because maybe I can start looking for it earlier and start treating it” (FG5, PART513)
	•Return of personal results might inspire positive lifestyle changes	•“… everybody knows they’re supposed to exercise, but your test said, hey, if you do this will happen. Give a positive outcome or maybe be more motivating.” (FG2, PART096)

Table 3.

Potential challenges and risks of participating in biobank research with supporting quotations

Category	Challenges and risks	Quotation
Research impacts	•Potential harm to oneself or one’s community	•“You don’t want your data weaponized against you … And your genetics weaponized against you” (FG2, PART094) “And your family and your people” (FG2, PART086)
Return of results	•Potential for unwanted concerning results (i.e., “ignorance is bliss”)	•“The risk is always finding something that you don’t wanna know about” (FG2, PART094)
	•Violation of privacy (e.g., biopiracy)	•“Bio-privacy is a huge thing right now. And I think it’s like there’s not a lot of legislation surrounding it and our privacy and protections …” (FG4, PART081)
Data sharing	•Challenges with life insurance and/or health insurance	•“Just being denied life insurance coverage, or, like very high-cost insurance” (FG4, PART247)
	•Concern about who has access to their data	•“Will we suddenly get arrested walking down the street because they identified our DNA mistakenly for crime? There’s just a lack of trust out there, and safeguarding the information” (FG5, PART513)

Participants also provided recommendations; for instance, one participant proposed the idea of making information more accessible so they could share it with others, stating, “I could share to some people in my family who might be afraid or worried or want to know, you know, be able to say, ‘Hey, look, this is what happens in my genotype. You could do this too if you go to the website and sign up.’ That might lead to disseminating more information” (FG6, PART062). Moreover, having a timeline or status of where they are in the biobank process (e.g., consented, sample collected, results returned) would limit potential confusion. One participant suggested that the biobank utilize success stories from diverse individuals enrolled in the biobank: “What would be helpful is to have people that are of the same cultural background who have done the biobank thing, have given the results and had positive results, to have those people promoting it” (FG5, PART513). In addition, to support participant autonomy, participants in our study suggested having a platform that allowed them to opt in or out of certain communications would prevent email fatigue and allow them to engage at the level they felt comfortable with. Participants also provided their preferences on how they would prefer to be contacted regarding their results, with the majority opting for email communications (Figure S2). When asked about culturally tailoring the return of results experience, Indigenous participants suggested including a traditional medicine person when discussing genetic results, especially if the results might be unfavorable or troubling. Finally, many participants expressed their enjoyment of participating in these focus groups, and one participant indicated that they would be more willing to participate in future studies (Table 4). This approach highlights the potential of personalized engagement and similar approaches to engage with underrepresented groups in research.

Table 4.

Feedback on the focus group experience with example quotations

Feedback topic or participant feedback	Example quotation
Participants wanted face to face interaction with researchers to feel engaged and learn about research goals	“ … I have to say, seeing you guys, seeing your faces and having the ability to talk to you and see what your interests are would definitely make me less shy in the future to participate in more … if a research study or something were to come out of the same establishment, then I would definitely be interested in participating more” (FG3, PART554)
Participants valued the opportunity to be a voice for a minority community	“ … that is definitely an interest to me and that is being of a minority community and just being a voice for that population if I can be” (FG3, PART554)
Participants wanted to be informed of research	“Yeah, I’d like to stay up to date, you know, through, you know, messages or, you know, even more of these kind of deals ‘cause you know, this is my first one. I didn’t even have a clue really what this was going to be. I thought I had to go donate blood somewhere and talk about it” (FG1, PART982)
Participants valued feeling included in research as members of minority communities	“Having been given an opportunity to have that voice, I think it just is a good thing. And being able to provide that insight, because I don’t think it’s something that people actively seek. I mean, how many times do we have, oh, you know, Natives still exist. Oh, you guys are still around, you know. And it’s not an exaggeration. There are people that really think that … That we’re all, we’re relegated to the museum, so having to be able to say no we’re actually living, breathing, we’re actually contributing to the genetic pool” (FG2, PART094)

Discussion

Here, we recontacted historically underrepresented participants enrolled in the CCPM biobank and centered their voices on motivations for biobank participation, return of results, and data sharing. We gained critical insights that can inform future strategies to strengthen engagement with these communities.

Distinctively, this study focuses on the perspectives of individuals from underrepresented backgrounds who have enrolled in a large-scale biobank, contextualizing their lived experiences and motivations. For example, understanding participant motivations for joining the biobank can guide recruitment efforts, particularly for historically underrepresented communities—a key step toward increasing diversity in research cohorts, which remains a goal for many health institutions and biobanks.^2,4,10,46 Unique to this study, some focus group participants were aware of the lack of representation in biomedical research, and this in part drove their motivations for joining the biobank and countering health inequities. Many expressed confusion regarding the research process and/or how their samples had contributed to research, as well as a limited ability to understand the results that were returned to them. This lack of transparency about how the biobank operates impacts participants’ ability to fulfill the reasons that initially motivated them to join the biobank. When a participant opts to join the biobank for a particular reason and expects something in return, whether that be the return of personal results or the sense of satisfaction at contributing to research, the nature of the relationship turns to one of reciprocity and mutual obligation. It has been noted⁴⁷ that a stronger focus on reciprocity in large-scale genomics research has potential to aid in engagement efforts with underrepresented communities, given that these communities have historical mistrust stemming from research that took from them without providing any semblance of benefit in return. In order to effectively maintain an ongoing relationship built on common interests and reciprocity, biobanks that offer return of results or research reports would benefit from clear and detailed communication of expectations and follow-up communications.

Clear and detailed communication from the biobank serves multiple purposes: it increases recruitment through participant networks, strengthens representation, and enhances the biobank’s reputation. Our study indicates that a positive experience with the biobank, in which participants feel that they have contributed value and are valued in return, may prompt them to recommend the biobank to others in their lives. This type of word-of-mouth recruitment, also known as snowball recruitment, has been found to be effective in historically underrepresented populations.^48,49 Therefore, biobanks should share findings in an accessible manner and provide regular updates, so participants can more easily identify ways they are helping to advance scientific knowledge and medical care.

We also found a strong relationship between motivations to join and participants’ feelings toward data sharing. As data-sharing practices become increasingly open,^50,51 it is important to consider how communities perceive and experience these approaches.^52,53 For example, the creation of the collective benefit, authority to control, responsibility, ethics (CARE) principles to supplement the findable, accessible, interoperable, reusable (FAIR) principles is an important step in ensuring Indigenous data sovereignty for Indigenous communities worldwide and highlights the importance of Indigenous peoples having the right to own, collect, and govern how their data are used.^54,55 Although there were positive perceptions of data sharing for public benefit and knowledge contributions, we must also recognize the potential concerns and limitations of how data sharing may be used to exploit historically underrepresented communities.⁵⁶

The potential for genetic data to be used outside of a research or clinical context also presented its own set of concerns for participants. Some reported concerns about their genetic information being used by employers for insurance decisions and fears that their data may be provided to law enforcement and used to falsely prosecute them. In the United States, the Genetic Information Nondiscrimination Act (GINA)⁵⁷ of 2008 governs the ways in which genetic information may be accessed or used by employers and protects against many forms of discrimination based on genetic information but does not prohibit the use of genetic information to influence eligibility for life insurance or disability insurance.⁵⁸ Our participants expressed mixed understandings of the protections that are available to them, indicating that further work ought to be done when communicating the potential benefits and risks of participating in genetic research.

The choice to participate in a biobank is a thoughtful balance between perceived risks and potential benefits. Consistent with previous studies, the choice to participate in research by historically underrepresented groups can be influenced by systemic barriers and inequities, such as past historical abuses, privacy violations,⁵⁹ inappropriate or stigmatizing research topics,⁶⁰ and helicopter/vampire science.⁶¹ One of the most significant perceived benefits of joining the biobank for participants was receiving genetic information either to guide healthcare treatment or to learn more about themselves. Although focus group participants acknowledged the risks of sharing their sample, the opportunity to receive personal health information through return of results provided a beneficial, reciprocal relationship with the biobank. Moreover, biobanks should carefully evaluate how genetic results are being returned, establish clear priorities for return of results, and recognize that these priorities may vary across populations. Ultimately, some participants expressed concerns, but many felt that their involvement in the CCPM biobank was worthwhile.

Similar to Lewis et al., the return of results activity found that a life-threatening condition was ranked as the most important and ancestry-related results were ranked among lower importance.⁴⁰ Previous work has found that biobank participants and general patients are interested in clinically actionable results.⁶² Currently, the CCPM biobank returns CLIA-certified genetic test results around disease risk (e.g., hereditary cancer, heart conditions, rare genetic conditions),²⁹ pharmacogenetics, and other results.⁶³ Biobanks often return health-related results, including life threatening, with fewer offering ancestry results.^{29,32,33,34,35,36,37} However, we found distinct discussions around ancestry that might differ from the general population. We identified an interest in ancestry for health-related implications and, for some, to learn more about one’s background, yet mindful framing of ancestry results is necessary. There is growing interest in providing research participants with the option to choose what kind of results they would be interested in, including the ability to decline information with no cures (as a way of supporting autonomy).⁶⁴

Our findings provide further insights into ongoing debates regarding return of results, specifically around actionability thresholds and participant choice(s). Participants favored returning results with clinical actionability, while acknowledging the difficulties and barriers to access, particularly within mental health services; some focus groups even discussed how mental health ties into physical health. Relevance is also important not only for themselves or their families but to their broader communities as well, highlighting the importance of each individual’s identity in shaping how they perceive relevance and personal utility and contributing to discussions around community consent.⁶⁵ The ranking activity suggested support for participant choice in return of results, particularly for ancestry-related results, reinforcing advocacy and desires for preference-based return of results approaches with appropriate guidance, as opposed to return of results policies that provide the same uniform information to all enrolled participants.

We had the following limitations. We had a low response (3.2%) and eligibility rate (2.2%). While we hesitate to draw conclusions, we note that this illustrates ongoing challenges to reach historically underrepresented populations. Further, we encountered logistical challenges, such as working with a separate entity to reach our participants, and email being the primary method of contact. Despite these limitations, we engaged AI/AN at a higher rate compared to other target groups, as well as compared to other studies where AI/AN often represent less than 1%.^13,66,67 This representation of AI/AN participants could be related to our project having multiple Indigenous research team members and principal investigator, highlighting the importance of representation, subject matter coherence, and having a diverse team with whom participants can connect. While our sample size was relatively small, we were able to reach saturation in the topics of interest. Our findings were generally consistent across our focus groups; however, there may be limited transferability to other populations of similar backgrounds, given the cultural and experiential diversity within these groups, yet these findings also align with previous work with people of color and their health motivations. Because our study is focused on participant engagement, it is important to highlight that these findings are derived from individuals already willing to participate in research and that additional research is needed to engage those biobank participants less likely to self-select for research studies. Five of our six focus groups were conducted over Zoom because of participant preference; virtual focus groups have their own associated limitations, as well as their own unique benefits. Previous studies on virtual focus groups^68,69,70 have highlighted negative impacts on rapport, a lack of ability to read nonverbal cues, frustration from technical difficulties, and a loss of contextual information. Literature on virtual focus groups⁶⁹ indicates that the identified limitations can be mitigated by employing a robust study design that accounts for the unique challenges associated with the format and by having a research team that is prepared to address technological challenges. Our focus groups were attended by multiple notetakers that were trained in the technology being used, which enabled them to address any issues that arose. The virtual format also allowed the team to accommodate participant schedules and overcome potential participation barriers such as transportation access. Another potential limitation is that participants who opted into this study may already be more inclined to participate in research and may view research more positively.

Here, we examined the views of historically underrepresented populations (AI/AN, AA/B, H/L) enrolled in the CCPM biobank regarding topics of interest. The focus groups allowed many participants to feel heard, sharing their desire for more updates from the CCPM biobank regarding their results and the impact of their participation. Recontacting participants provided valuable insights into building trust and equity in research. We share key recommendations for researchers interested in engaging with individuals from historically underrepresented backgrounds in biobanks or genetics research broadly. These recommendations were developed from our project experiences (Table 5). We recommend that researchers and biobanks/cohorts consider the following: (1) ensure that your team is composed of diverse individuals, (2) incorporate recontact of participants as part of the research study design, (3) provide alternatives and flexibility in the venue and format for recontact, and (4) use multiple types of recruitment strategy. These recommendations and practices can improve recruitment and engagement of historically underrepresented populations. Ultimately, this study contributes to evidence-based approaches that can support recruitment and contribute to lasting engagement and retention of participants from historically underrepresented communities within biobanks and genomic research.

Table 5.

Recommendations for re-/engaging participants in biobanks and large cohort studies

Broad recommendations	Why it matters
Ensure that your team is composed of diverse individuals	Having team members from similar backgrounds can build trust, comfort, and helps bring diverse perspectives to the study design and data interpretation
Incorporate recontacting participants as a part of the research study design from the outset	Participants value learning how their contributions advance research. Recontact also supports “member-checking,” ensuring findings are accurately represented
Provide alternatives and flexibility in venue and format when recontacting participants	Virtual formats increase accessibility and reduce barriers such as transportation, while in-person sessions remain valuable
Use multiple types of recruitment strategies when recontacting participants	Recruiting in multiple formats is beneficial to reach more people across different age groups, racial/ethnic groups, and sexes

Data and code availability

There are restrictions to the availability of the dataset and code supporting the current study. Data and code will not be deposited in a public repository because of the increased risk to participant confidentiality inherent in qualitative data.

Acknowledgments

We would like to acknowledge and thank our focus group participants who shared their time, knowledge, experiences, and stories with us. Their efforts will help shape future engagement efforts. We would also like to thank F. Leah Nez (F.L.N.) and Samantha Lopez (S.L.) for their instrumental roles in study design and data analysis, respectively. This work was supported by the NHGRI R35HG011319 Genomic Innovator Award Bioethics Supplement.

Author contributions

The focus group material was designed by K.G.C., R.K.J., D.E.M., F.L.N. (not a co-author), S.D.S., C.A.S., and B.L.M. The focus groups were facilitated by D.E.M. and a combination of two team members who acted as notetakers and observers: C.A.S., B.L.M., F.L.N. (not a co-author), and S.D.S. The study’s principal investigators were K.G.C. and R.K.J. Qualitative analysis was led by L.O.R. and S.L. (not a co-author). Quantitative analysis was led by S.D.S., which involved analyzing survey data and creating graphs. Manuscript writing was led by C.A.S., B.L.M., and L.O.R. and reviewed by all team members.

Declaration of interests

R.K.J. and K.G.C. are faculty members of the CCPM biobank.

Web resources

• University of Colorado Anschutz Biobank, https://medschool.cuanschutz.edu/cobiobank

• CCPM Biobank, https://www.uchealth.org/services/biobank/