Science Lost DEI Programs. Now It Needs Something Better

Programs and initiatives related to diversity, equity, and inclusion (DEI) across the US have been cut and ostracized due to recent political and social backlash. This includes efforts in science, health, and the biotech and pharma industry too, which are known for both their efforts to reduce health inequities and their diverse teams. As a former DEI consultant with a background in bioethics and medical anthropology, I don’t think these teams need traditional DEI programs, which led to some improvements but were also shown to be counterproductive in certain settings. Instead, they need evidence-based strategies that actually deliver on DEI’s promise: diverse, innovative, high-performing teams.

Why We Need Diversity in STEM

Three thousand grants supported by the National Science Foundation (NSF) and National Institutes of Health (NIH) have been cut due to new anti-DEI policies.¹ These grants were flagged for having terms like ‘diverse,’ ‘race,’ ‘disparity,’ or ‘underserved.’² Delaying scientific innovation based on these terms is problematic, and so is defunding efforts to ensure diverse recruitment in research and clinical trials.³

Science, Technology, Engineering, and Mathematics (STEM) research fields have yet to achieve sustainable diversity, often being described as having a ‘leaky pipeline,’ a term describing the phenomenon of women and other marginalized people leaving academia and science due to toxic work environments. While women account for 50 percent of STEM degrees, only 15 percent are professors. Only 3.8 percent of STEM academics identify as disabled, whereas disability is the largest minority community; approximately 20 percent of people are disabled.⁴

The benefits of diverse research or healthcare teams are immense. Different backgrounds and lived experiences create environments for innovative ideas, creative problem solving, and an overall competitive advantage.⁵ However, traditional DEI efforts, including workshops, diversity statements, and employee resource groups, lack evidence of success and can often be counter-effective.^6-8 In other words, evidence shows that even without recent politicization, DEI efforts were failing to achieve their intended effects. Traditional DEI efforts did raise awareness and created some pathways for underrepresented groups. However, these gains were modest and unsustainable.⁶

Yet the outcomes that DEI aspires towards in health and science—increased innovation, healthy debate, and team alignment—are still needed. As scientists, we must follow the data that show that diversity and inclusion lead to better outcomes as well as the data supporting more effective strategies to get there.

The Data for DEI

Diversity and inclusion foster environments where different perspectives can come together to challenge each other and advance innovation.⁹ In a 2011 paper, researchers looked at the relationship between a team’s composition and publication productivity and found a positive association with interdisciplinarity, team cohesion, and fewer senior members (that is, more students and postdoctoral fellows).¹⁰ Interestingly, there was no effect on gender diversity, though this contradicts more recent research.¹¹

Chemical engineer Magda Titirici at Imperial College London noted in recent publication on DEI in science that “inclusive teams produce more impactful, relevant, and trusted research.”¹² Research shows that teams with ethnic diversity publish papers that are 10.63 percent higher in impact, mixed gender teams perform better across medical subfields, and increasing diversity in clinical trials reduces health disparities and promotes public health.^13-16

There is also evidence that a diverse research team improves decision-making, protocol preparation, and diligence, which increase not only innovative ideas and solutions but the uptake of those innovations. These benefits are evident in health science, animal behavior, psychology, physics, economics, and more.¹

Princeton University biological anthropologist Agustín Fuente pointed out the particular contradiction of ignoring the evidence of the benefits of diversity in science as “The intentional disregard for readily accessible data and well-supported analyses is not something one should expect from those who practice science.”¹ And what the evidence shows is that having a diverse team isn’t enough. Diversity plus team cohesion and trust is what makes a difference. To improve collaboration and innovation, principal investigators (PIs), academic departments, and biotech firms need to ensure that people feel safe to speak up, ask questions, and identify issues.⁵ Whether it’s neurological and cognitive, racial, gender, cultural, or educational diversity, having differences in your team is an advantage if it is cultivated to be so.

Where Traditional DEI Programs Fall Short

Cognitive-behavioral scientist and DEI practitioner at the non-profit Inequity Agents of Change, William Taylor Laimaka Cox explained in a paper published in Management Decision that most DEI approaches adopt an information deficit model approach, which “assumes recipients’ lack of key information, then try to correct that deficiency.”⁸ These approaches, which include training and historical awareness raising, rely on a type of self-introspection that can lead to feelings of guilt or blame. These feelings naturally lead to resistance.

This doesn’t mean DEI initiatives accomplished nothing. They raised awareness about implicit bias, created some pathways for underrepresented groups, and got diversity on the agenda in scientific institutions. But awareness without systemic change creates frustration. People know problems exist but lack the tools and authority to fix them.

Additionally, impacts resulting from these interventions are often short-lived and do not change ingroup behavior.^6,16 A 2021 meta-analysis found that in the studies that reported achieving some goals of the DEI programs, their effect was relatively small, limited in scope, and temporary.¹⁷ More concerning, some research has demonstrated counter-effective impacts of information deficit-based interventions, such as a group of researchers finding that diversity awareness training led to poorer treatment of non-White employees.¹⁶

The Solution: Productive and Safe Debate, Innovative Collaboration, and Supported Flexibility

Scientific teams need ways to cultivate collaboration alongside debate, diverse perspectives along with effective communication strategies, and the flexibility to modify and pivot as data is produced. Cox suggested an empowerment-based approach, which assumes people already want equality and inclusion and just need the right resources to create it. He argued that thinking of bias and discrimination as habits to break communicates that bias is something that requires ongoing effort but that it is something in a person’s control.⁸

Setting up teams to be agents of change by ensuring increased systemic support has the highest likelihood of success. For example, efforts to address wealth and social capital inequalities by intentionally creating more mentoring opportunities and expanding searches for talent are more likely to succeed. This is why there is scientific evidence supporting affirmative action.⁶ These efforts require a shift in mindset away from individual behavior and toward designing an organizational culture and systems to support multiple perspectives, working styles, and communication preferences. Labs need well-functioning people just as much as they need well-functioning machines, and the best way to ensure both is ongoing service and support.⁵

When barriers to participation are predicted, addressed proactively, and embedded in team culture and policy, inclusion, diversity, and all the benefits that follow become a natural, almost invisible part of the team. Below are three examples of common culture issues and potential solutions that emphasize culture design rather than traditional DEI.

The PI that alienates their team: Brilliant scientists become brilliant PIs, but they rarely receive people management training. Science stands out as a rare field where a person’s merit gets them a team but not the skills to lead that team effectively.

The Solution: Embedding people management and leadership training into the hiring and promotion of PIs. Training should include science industry specific strategies to cultivate supportive teams and help team members achieve their professional aspirations. New PIs should also learn to think about team composition issues that can clog team progress, understanding that each team has its own dynamics.⁵ Some PIs may need refreshers or additional support, but arming them with these strategies will improve cohesion, engagement, and innovation.

The multicultural team with different ideas on what team membership means: The US has been relying on immigrants to build the best teams in science and technology for decades. But once those team members are recruited to a US-based team, interpersonal conflict and a lack of team cohesion may follow.¹⁸ One frequent concern is effective communication.

The Solution: Thinking creatively about how to create communication protocols that attend to English as a second language and different culture norms regarding when and how to speak up to others is an exciting opportunity. One study showed that using technology to help multicultural teams communicate asynchronously improved communication and scientific collaborations.¹⁹ While this may not be relevant or feasible for all teams, this kind of out-of-the-box thinking can generate sustainable solutions.

Conflict between corporate leadership and scientists: When science and capitalism merge, discord can arise. The best-known example of this situation is the tobacco industry; scientists were sounding alarms about health impacts of tobacco that corporate leaders downplayed for profit. Science can be slow, identifies more issues than it solves, and adheres to professional and ethical guidelines that can hinder industry transparency and complicate health-protective policy and practice.²⁰ Differences in the language used to talk about progress and innovation lead to communication misfires that affect the relationship between scientific and corporate teams.

The Solution: Teams that focus on collaboration from the very beginning, with shared workflows and learning sessions to align language and priorities, can ensure expectations are realistic. Including liaisons who understand both sides of the business can help bridge gaps in understanding. Incentivizing upholding scientific integrity and transparent policies will increase trust both within and outside of biotech and pharma companies.

In a 2022 publication, Vanderbilt University cell biologist Antentor Hinton Jr and health inequities researcher W. Marcus Lambert at SUNY Downstate Health Sciences University explained, “Simply put, culture is the essence of an organization that determines how individuals behave and the organization runs.”²¹ However, they noted, DEI efforts rarely get the space they need to improve team culture in science. Taking a more embedded, proactive approach by integrating support and inclusion throughout organizational policies and practices will generate a sense of belonging and respect. As a result, brilliant science can proceed unencumbered by interpersonal or systemic barriers.

The science is clear: Diversity and inclusion are good for science. Although traditional DEI efforts have not been as effective as we hoped, it doesn’t mean the goals of the field are no longer important. It just means we have to rethink how to reach those goals.