Introduction
By 2050, the general US population will be less than 50% non-Hispanic White [Reference Frey1], yet diversity within the biomedical workforce is not reflecting these rapid changes [2,3]. Recent attention has been devoted to the need for effective strategies for diversifying the biomedical workforce [Reference Rubio, Mayowski and Norman4] especially given the dual impact of systemic racism and the long-term disruption of the COVID-19 global pandemic[Reference Rubio, Mayowski and Norman4,Reference South-Paul, Campbell, Poll-Hunter and Murrell5]. The National Institutes of Health (NIH) prioritizes increasing representation in biomedical research including underrepresented racial and ethnic groups, individuals with disabilities, individuals from disadvantaged backgrounds, and women (in select biomedical research areas) [Reference Bhatti6].
Diverse biomedical workforces are important for several reasons. Research has shown that diverse research teams are more productive than more homogeneous teams and produce innovative research [Reference Valantine and Collins7]. Additionally, diversity within the workforce offers more opportunities for creative approaches to new and existing healthcare problems [Reference Oh, Galanter and Thakur8] and advocacy for change of existing power structures that cause inequities [Reference Espino and Ariza9].
Of particular importance to the efforts to diversify the biomedical workforce is the study of underrepresented researchers’ experiences within the workforce, which has implications for retention efforts. It is known that biomedical researchers from underrepresented racial and ethnic backgrounds (Black or African American, Hispanic or Latina/o, American Indian or Alaska Native, Native Hawaiian or other Pacific Islander) have higher attrition rates throughout their career trajectory compared to their White and Asian counterparts [Reference Heggeness, Evans, Pohlhaus and Mills10–Reference Shavers, Fagan and Lawrence13]. Discrimination is thought to play a large role in the high attrition rates [Reference McGee, Saran and Krulwich14,Reference Economou15], however, there may be other factors that are not being examined. Many research and healthcare agencies, including the NIH, are exploring factors and methods to learn more about these groups [Reference Valantine and Collins7].
Inclusion is known to be a key concept in determining the effectiveness of recruitment and demonstrating the impact on retention for underrepresented biomedical researchers. In fact, some argue that without inclusion, diversity efforts alone may be insufficient to impact changes in biomedical and other STEM fields [Reference Puritty, Strickland and Alia16,Reference Swartz, Palermo, Masur and Aberg17] because inclusion addresses factors such as belongingness, psychological safety, and work engagement, which create environments where diversity can thrive. While there is no gold standard definition of inclusion, most definitions include two main components. The first is having a sense of psychological safety or an individuals’ perceptions of favorable consequences related to interpersonal interactions within their work environment [Reference Newman, Donohue and Eva18–Reference Edmondson, Kramer and Cook20]. There are four components of psychological safety: (1) inclusion safety—members feel safe to belong to the team; (2) learner safety—members are able to learn through asking questions; (3) contributor safety—members feel safe to contribute their own ideas, without fear of embarrassment or ridicule; and (4) challenger safety—members can question others’ ideas or suggest significant changes to ideas, plans, or ways of working [Reference Edmondson, Kramer and Cook20,Reference Clark21]. The second component is having equitable access to opportunities and resources within the organization and being considered an “insider” that contributes to the organization’s success [Reference Randel22]. Several studies have identified an association between feelings and perceptions of inclusion and work engagement [Reference Goswami and Goswami23]; however, more research is needed to understand inclusion among underrepresented biomedical researchers. Engagement refers to an employee’s commitment to contributing to the organization’s goal [Reference Ge, He and Liu24,Reference Schullery25] and is related to work experience that is purposeful, fulfilling, and positive [Reference Schaufeli, Martinez, Pinto, Salanova and Bakker26].
The Building Up a Diverse Biomedical Research Workforce Study (Building Up) offers a unique opportunity to understand inclusion among underrepresented early-career biomedical researchers (postdoctoral fellows and early-career faculty). Building Up is a large, national trial that tested the effectiveness of an intervention to increase research productivity, belonginess and psychological safety. In this secondary study, we used qualitative interviews to explore Building Up participants’ feelings of inclusion at their institutions.
Methods
Participants were recruited for this secondary, qualitative study from Building Up [Reference White, Proulx, Morone, Murrell and Rubio27], which is a two-arm cluster randomized trial at 25 academic medical institutions (23 primarily white institutions and 2 minority serving institutions) that tests the effectiveness of two intervention arms with varying intensity of 4 intervention components (monthly sessions, networking, coursework, and mentoring) to increase research productivity. For this study, a subsample of 33 participants were randomly selected from the Building Up participant pool and were emailed an invitation to participate in an interview. We sought to oversample from the two minority serving institutions (MSI), in the case that participant’s experiences were qualitatively different and to have a sample with approximately equal numbers from each intervention arm. Interviews were conducted from February 23 to June 18, 2022. If a potential participant declined or did not respond, another was randomly selected. A single Institutional Review Board at the University of Pittsburgh approved the protocol. Participants provided electronic informed consent and were informed that their responses were confidential.
We employed a qualitative description approach, in which researchers seek to describe participants’ thoughts and feelings on study topics without abstracting to the theoretical models, as is often the goal with qualitative research in the social sciences. This type of qualitative approach is common in qualitative studies in medical and health sciences contexts [Reference Sandelowski28,Reference Kim, Sefcik and Bradway29] and tends to produce actionable insights.
The primary investigator developed an interview guide covering domains of interest, including how the participant came into their current position, how they described their workday (what they liked/disliked about a typical day and what they wanted to change), and how they described their working unit (commitments to diversity, equity, and inclusion [DEI]; experiences of microaggressions, racism, and discrimination; and workplace culture). The guide was reviewed and piloted by members of the Building Up study.
Interviewing duties were split between the primary investigator and an interviewer from the Qualitative, Evaluation, and Stakeholder Engagement Research Services (Qual EASE) at the University of Pittsburgh. Interviews were conducted via Zoom or telephone, per the participants’ preference, and were audio recorded. The mean completion time for all interviews was 50 minutes (range 20 minutes – 1 hour 24 minutes) and mean completion times were similar for each interviewer (49 minutes and 52 minutes). Interview style was slightly different between interviewers, but coding was similar.
Verbatim transcripts of the audio recordings were produced by research assistants at Qual EASE. Following transcription, the primary investigator inductively developed a codebook based on the content of the interviews; the codebook was reviewed and approved by the qualitative methodologist and interviewer. The primary investigator then co-coded all transcripts with an experienced coder from Qual EASE and adjudicated all coding disagreements to full agreement. The completed coding served as a basis for a thematic analysis [Reference Braun and Clarke30,Reference Braun and Clarke31] conducted by the primary investigator. The resulting themes were shared and discussed with the rest of the study team as a form of investigator triangulation and to allow team members to provide feedback to refine the themes. Participant characteristics were presented using frequencies and percentages for categorical variables and mean and standard deviation for continuous variables.
Results
Of the 33 people we contacted, 25 (16 faculty and 9 post-doctoral fellows) responded and were interviewed. The eight participants that did not participate either did not respond to the email (N = 6) or no-showed (N = 2) for their appointments. Of those that did respond, participants’ racial/ethnic identities were as follows: Black (36%), more than one race (28%), Asian (12%), Latino/a (12%), White (8%), and Middle Eastern (4%). Two participants (8%) reported a disability and 76% of participants were from the control arm. Compared to the main study, participants in this analysis were slightly older (36 vs. 39), had slightly more women (80% vs. 84%), were less likely to report a disability (6% vs. 8%), and more likely to hold faculty positions (64% vs. 53%) [Reference White, Proulx, Morone, Murrell and Rubio27]. Most participants were in academic positions, although a few participants had transitioned to industry positions. The participants were working at 16 of the 25 institutions included in the larger study including one of the MSI (participants from the other MSI did not respond). For most, regardless of type of position, their primary work activities included meetings, research, teaching, administrative duties, mentoring, and supervisory roles.
Our research team identified four distinct themes: (1) Participants appreciated the flexibility, versatility, and sense of fulfillment of their positions which enhanced feelings of inclusion; (2) Greater psychological safety led to a greater sense of belonging to a research community; (3) Participants had varied experiences of inclusion in the presence of microaggressions, racism, and discrimination; and (4) Access to opportunities and resources increased feelings of value within the workplace.
Theme 1: Participants appreciated the flexibility, versatility, and sense of fulfillment of their positions which enhanced feelings of inclusion
Many participants described working in environments that were conducive to inclusion. Most participants enjoyed the flexibility (ability to organize schedule including working remotely and “freedom” to make decisions about how projects are chosen and conducted) and versatility (use of multiple skillsets), of their positions, which they said enhanced their sense of creativity and innovation. Specifically, participants enjoyed how they could set their own schedules to align with when they felt they functioned best. They also discussed how the research they did was rewarding because it led to new discoveries and methods of addressing existing problems. Having a sense of autonomy while developing new research or running lab experiments led participants to feel productive. Being able to conduct and make progress on their projects brought a sense of fulfillment and inclusion.
Additionally, when participants’ work was in line with their overall career goals, they reported feeling fulfilment with a significant sense of purpose. This helped participants feel part of the institution’s overall mission, which is a component of inclusion. This sense of fulfillment was amplified when participants felt supported via assistance to complete their work or recognition of their efforts by their mentor and/or supervisor. Many participants also commented on how the sense of fulfilment was enhanced because they were making a difference as an underrepresented biomedical researcher.
At times, participants felt there were barriers to having the flexibility that they needed. For example, some participants discussed how having duties, such as clinical and/or teaching responsibilities, sometimes took precedence over the research that they wanted to focus on. Other times, they reported feeling undervalued when they were asked to do service work such as DEI service that was meaningful but did not contribute to their research goals. When leadership’s priorities were not in line with the participant’s, the participant did not have as much autonomy around how they spent their time. While this did not change their sense of fulfillment about their research, it often elicited feelings of conflict, frustration, and stress as they tried to accomplish all the tasks that they needed to do, which challenged their sense of inclusion. At times participants expressed uncertainty about how to navigate these situations with their superiors. In some instances, participants described their experiences as a “tax” of being underrepresented compared to their majority counterparts.
See Table 1 for representative quotes.
Quotes representing theme 1 – participants appreciated the flexibility, versatility, and sense of fulfillment of their positions which enhanced feelings of inclusion
Theme 2: Greater psychological safety led to a greater sense of belonging to a research community
Participants discussed how elements of inclusion safety or feeling safe to belong to a team, a component of psychological safety, affected their feelings of being in a research community. (Participants did not specifically use the term “inclusion safety” or “psychological safety” in their descriptions.) One example included feeling welcomed by their colleagues when joining their working unit via welcoming emails or other invitations to be part of the research team. Participants also described how collaborative and supportive their environment was. These environments were described as not being competitive but, rather, were contexts in which participants were able to discuss and receive support from their coworkers, mentors, leadership, and collaborative networks outside of their immediate workplace environments. Participants also reported that greater psychological safety allowed them to advocate for additional support they might need to conduct their research.
Participants mentioned that “strong” leadership increased their feelings of psychological safety. They described strong leaders in the following ways: leading by example, being open to listening and addressing the participant’s or others concerns, advocating for and supporting participants when needed, being receptive to the needs of all, and being approachable. Strong leadership played a role in feeling supported in their research community and often set the tone for other interactions within the work environment.
When participants reported lower trust of their colleagues, mentors, and/or supervisors, and thus lower inclusion safety, they did not feel as connected to their research community. For example, participants offered scenarios in which they were ignored during introductions to coworkers or were not spoken to by senior research faculty when in the elevator together which made them feel as though they were not accepted within the research community. At times, participants attributed being ignored to their underrepresented status and perceived that their majority counterparts were not having the same experience. Other regular interactions also made them feel like they were not part of the team. These missed interactions were perceived as important because they were seen as opportunities to network, collaborate, and generally be viewed as a colleague. Others with an international background sometimes felt that they were not taken seriously because they were not from the U.S or because English was not their first language. Participants also cited lack of diversity, generational differences, differences in how different degrees were valued, and lack of support from leadership as playing a role in strained interactions that led to lower inclusion safety. Additionally, some participants discussed how they did not know how to interact with certain individuals (particularly those with majority status) or felt uncomfortable doing so, which often led to self-induced isolation. These experiences resulted in feelings of stress and of being undervalued and sometimes contributed to individuals leaving a department or an institution.
See Table 2 for representative quotes.
Quotes representing theme 2 –greater psychological safety led to a greater sense of belonging to a research community
Theme 3: Participants had varied experiences of inclusion in the presence of microaggressions, racism, and discrimination
Participants described aspects of inclusion safety as they discussed different ways that they interpreted and experienced microaggressions, racism, and discrimination directly or indirectly which could have occurred in their work unit or within the broader institution. Often during the interview, participants would describe how they suppressed experiences of microaggressions, racism, and discrimination as a coping strategy so that they could focus on their research. For this reason, participants may not have recalled all the negative experiences they had. Additionally, because ignoring the occurrence was a way of dealing with microaggressions, racism, and discrimination, they too might have chosen not to share certain occurrences. Many participants reflected on how they were not often asked for input about these experiences within their work culture.
Participants who recalled direct experiences of microaggressions, racism, or discrimination described times when they personally experienced an interaction that called into question their ability or character in a negative stereotypical manner. Although these experiences were often reported as being stressful, participants had a variety of ways of dealing with them. Some participants reported the occurrence to their supervisor or another reporting body, some bonded with other coworkers to deal with the occurrence (this was most common when discriminatory patient behavior was involved), and others ignored the experience in favor of keeping the peace. Participants who reported having organizational support stated that their complaints were taken seriously and were more likely to report feelings of inclusion and belonging compared to those that did not report feeling supported.
Participants had different reactions to indirect experiences of microaggression, racism, and discrimination (i.e., observed or shared experiences that did not directly happen to them). Some participants were able to offer support to an individual that experienced a microaggression or instance of racism or discrimination, while other participants were not sure how to show support. Often, participants experienced stress due to indirect experiences, especially if they did not feel the situation was addressed well. Such indirect experiences sometimes caused participants to question the values and standards within the workplace, which affected their feelings of inclusion. In rare cases, participants reported leaving or wanting to leave their institution due to indirect experiences.
See Table 3 for representative quotes.
Quotes representing theme 3 – participants had varied experiences of inclusion in the presence of microaggressions, racism, and discrimination
Theme 4: Access to opportunities and resources increased feelings of value within the workplace
Access to opportunities and resources were important for inclusion. Participants described how access to opportunities, including opportunities to communicate their needs, increased their feelings of being valued, which is related to contributor safety or feeling safe to share ideas. Participants described that sharing needs as an underrepresented researcher felt vulnerable because it could be interpreted as not being qualified for their job. Participants appreciated opportunities to discuss their research plans with other coworkers, or to have a wide range of individuals on their mentoring and collaborative teams, especially those from underrepresented groups. In some situations, participants commented on how not being able to work with mentors from underrepresented groups was not as important to them because they felt supported by their existing mentors. Participants greatly valued moments when they asked for things that they needed and their needs were taken seriously—particularly if they felt vulnerable when asking.
Regarding career advancement, participants were very appreciative when they were presented with opportunities that would contribute to their career growth. This largely included being able to share their research progress with the scientific community, being able to collaborate with other researchers, or having opportunities to develop skills that would enhance their work. While some participants did mention opportunities to do service work, such as creating DEI curriculum, they often mentioned that these did not count toward their advancement and took a considerable amount of time. While many discussed enjoying these types of activities, they also wanted to be considered for opportunities that would enhance advancement at their institutions (i.e. publications and grants).
Regarding resources, participants valued having access to resources (e.g., competitive pay, access to labs and lab equipment, startup packages, trainings, and personnel) that guided them as they were developing their research careers. These resources either lessened the amount of work the participant had and/or allowed them to be more effective and efficient. Others who did not have access to these resources mentioned that having them would make a significant difference. Generally, having access to these resources increased the participant’s feeling of value in the workplace because they felt that leadership was investing in them, and they had the tools to be more effective at their jobs.
Participants specifically described the importance of improving the effectiveness of DEI policies and practices, as resources, leading to climate and culture change, ensuring that the participant’s unique needs are met. Participants mentioned that effective practices and policies reflect and in fact drive actual—as opposed to nominal—culture changes within the workplace. Participants expressed how these culture shifts facilitate communication of unique challenges faced by early career researchers, especially those with underrepresented status. When participants did not feel genuine intentions behind their institution’s DEI policies, they felt less inclusion and did not feel as valued in their workplace.
See Table 4 for representative quotes.
Quotes representing theme 4 –access to opportunities and resources increased feelings of value within the workplace
PI: Principal Investigator.
Discussion
We identified four main themes that improved inclusion at the institutional level among Building Up participants. The themes illustrate how inclusion is enhanced when there is a sense of flexibility and versatility around the work that is being done. Inclusion is further enhanced when the participants felt psychological safety within their research community; felt supported in the presence of microaggressions, racism, and discrimination; and felt that the general environment was supportive in terms of opportunities and resources. However, our findings suggest nuanced or subtle differences in the ways that participants perceived inclusion within their work environments. For example, if a participant felt supported within their work environment, experiences of racism, microaggressions, or discrimination may not have been as detrimental to their sense of inclusion as those participants who did not feel supported. Additionally, participants expressed differences in how they navigated issues at work. For example, some participants focused on finding supportive groups in the face of situations that seemed hostile due to discrimination while others were bothered by hostility.
These findings highlight that underrepresented biomedical researchers, especially those who are early-career, perceive differences between what they need, compared to their majority counterparts, to reach the same level of psychological safety in their workplace. To elaborate, underrepresented participants expressed a desire to have their unique challenges recognized by others. Many indicated their underrepresented status meant that they needed additional resources to function equivalently to their well-represented counterparts. For example, some participants isolated because they were not sure how to interact with their well-represented counterparts or who to seek help from. While we did not include well-represented early-career biomedical researchers in this research, several studies have indicated that they do not report the same level of adversity as their underrepresented counterparts [Reference Chicca and Shellenbarger32]. In addition, our findings suggest that psychological safety can be inhibited by missed opportunities to interact with colleagues and senior faculty. It is important to understand how and under what circumstances these missed opportunities impact inclusion and what counteractions may be employed to mitigate these impacts.
While these findings are consistent with other published literature, this study provides more context around how unconscious bias can affect inclusion. Much of the literature focuses on unconscious bias and the presence of discrimination as independent barriers to inclusion [Reference Swartz, Palermo, Masur and Aberg17,Reference Allen and Garg33–Reference Whelan35]. That is, an individual from an underrepresented background will not be able to feel included where unconscious bias or discrimination is present. While it is not preferable to have unconscious bias and discrimination, ridding our institutions of unconscious bias that leads to discrimination is a gargantuan task that will take massive amounts of resources and time [Reference Braxton36,Reference Thomas37]. Our data suggest that there are ways to improve inclusion among underrepresented biomedical researchers even when discrimination is present. Developing more robust interventions that specifically increase psychological safety, irrespective to when discrimination is present, and enhance the availability of opportunities and resources that strengthen feelings of value and belongingness is indicated.
Our study also provides more context around the relationship between psychological safety and the policies and practices of a workplace. Policies and practices are important components of psychological safety. However, our findings indicate that the presence of policies and practices alone did not increase psychological safety or feelings of inclusion. Instead, participants indicated that it was important that policies and practices were in line with the daily operations of the workplace and that they reflected the values of the members of the workplace. Thus, it is important for leaders and administrators to assess the climate of their workplace and to develop policies and practices with realistic plans for implementation and with genuine expectations and timelines [Reference Burnett and Aguinis38]. For example, creating policies that endorse diversifying the workforce when 80% of the employees do not believe there is a problem with diversity is ill advised as that does not match the climate of the workplace. These policies will likely feel inauthentic to people from underrepresented backgrounds, may cultivate feelings of mistrust, and will not be executable in a reasonable time frame.
Anti-DEI legislation will likely have a significant impact on the opportunities and protections that are established and supported within institutions from recruitment to retention [Reference Dhanani, Arena and Bogart39–Reference Cox and Nguyen41]. The erosion of DEI policies may have negative effects on hiring personnel that establish and reinforce DEI policies, establishing programs that focus on improving outcomes for marginalized populations (such as Building Up), and providing protections for minoritized groups [Reference Feder40]. It is possible that these changes will cause fear and uncertainty about the best ways to continue DEI-focused work [Reference Feder40]. Additionally, without specific policies and programs awareness of the unique needs of underrepresented researchers may be greatly diminished leading to more barriers and less feelings of inclusion [Reference Dhanani, Arena and Bogart39]. However, as participants alluded to, there are many things that can be done in the workplace to support underrepresented researchers beyond policies. For example, leadership can be intentional about modeling equitable practices within the workplace and holding others accountable for doing the same. In line with this example, Cox and Nguyen suggest movement towards accountability practices in the face of threats to DEI polices [Reference Cox and Nguyen41]. While DEI policies and programs focus on the means to achieve diversity goal process (e.g., anti-bias training), accountability practices focus on accountability to the goal itself (e.g., equitable career advancement). Institutions and workplaces will likely still be able to implement accountability practices that monitor and assess goal achievement and promote belonging and inclusion without violating laws [Reference Feder40,Reference Cox and Nguyen41]. Accountability structures and practices have been associated with inclusion [Reference Corley42] and career advancement of underrepresented individuals [Reference Cox and Nguyen41].
This study has both strengths and limitations. As for its strengths, we assessed perceptions of inclusion among a diverse group of early-career biomedical researchers from backgrounds underrepresented in science. Additionally, the study examined experiences of underrepresented individuals using the NIH definition of underrepresented. Even though we were not able to report on differences between specific groups (e.g., by race, gender identity, sexuality), we believe that the data is robust enough to provide transferable insight into underrepresented individuals’ experiences. Future research should include within-group differences, which may reveal differences in experiences of inclusion. For example, gender differences in personality traits might influence how individuals from different genders perceive inclusion [Reference Weisberg, DeYoung and Hirsh43]. Understanding the circumstances in which underrepresented early-career biomedical researchers feel included can lead to more effective policy and practice changes that may influence retention. Another strength is that we sampled from a large research study of early-career biomedical researchers and that we interviewed participants at various institutions throughout the United States. While there were differences in the positions and resources of participants at each institution, some clear commonalities appeared regarding experiences of inclusion.
Our study included some limitations. All interviews used participants’ accounts of their experiences in the workplace with the majority of participants coming from the control study arm. Participants may have made errors in the order in which they remembered their experiences, or they may have not recalled specific aspects of their experiences and the participant’s study arm may have influenced their participation or responses. However, participants were asked to base their feelings of inclusion on their experiences within their work unit and institution and were not asked to discuss their experiences in the study. Because psychological safety and inclusion are largely based on perception, we do not believe these errors affected our study findings. Participants may have also under-reported negative experiences due to concern over self-perception or the possible impact it might have had to their institution if they were identified. This study also only evaluated experiences and perceptions of inclusion among academic centers. Biomedical research done in other settings, such as industry, may have other defining factors. Lastly, while we sought to be inclusive of all 25 institutions included in the larger Building Up study, our sample included participants who were available and self-selected to be interviewed. This may have introduced bias in the sampling process.
Even given these limitations, our findings have several implications that indicate various future research directions. Our findings suggest several areas of intervention that may improve how individuals experience inclusion, including thoughtfully implementing policies and practices, providing additional focused resources for people from underrepresented backgrounds, enhancing opportunities for communication and collaboration, and tailoring ways to engage leadership to provide and model effective ways to interact within the workplace. These findings also offer new lines of inquiry. For example, different gender specific personality characteristics (i.e., women are often found to be more agreeable than men) may have impacts on their experiences of inclusion. By better understanding how nuance may affect implementation strategies, there may be better ways to improve feelings of inclusion within academic medical centers which can improve belongingness, psychological safety, and work engagement which are related to retention efforts.
Data availability statement
The datasets [GENERATED/ANALYZED] for this study can be found in the [NAME OF REPOSITORY] [LINK]. Please see the “Availability of data” section of Materials and data policies in the Author guidelines for more details.
Acknowledgements
We would like to thank all of the Building Up and Qualitative, Evaluation, and Stakeholder Engagement Research Services (Qual EASE) staff and team members for their support with this project. Special thanks goes to Flor Cameron and Balchandre Kenkre for their assistance with conducting and coding the interviews. We would also like to thank Marie Norman for her assistance with developing the discussion section of the manuscript.
Author contributions
CMM developed the concept and design, collected the data, and structured the manuscript. MH advised CMM and assisted with the development of the concept, design, and manuscript structure. DMR, TGW, and GS advised on the development of the content and validated the content. GEW, NEM, and AJM validated the content. The authors read and approved the final manuscript.
Funding statement
This study was funding by the National Institute of General Medical Sciences; Grant number: U01-GM132133-03S1.
Competing interests
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential competing interest.
Open access
