Data show that representation matters for underrepresented populations in science, technology, engineering, and mathematics (STEM). Representation increases a sense of belonging [1,2,3,4], retention [1], and success [4,5,6]. It shows trainees what is possible and what is welcomed within their chosen profession. However, sexual and gender minority [e.g., lesbian, gay, bisexual, transgender, queer, and the many other affirmative ways people choose to self-identify (LGBTQ+)] visibility relies on self-disclosure. Unfortunately, ~30–50% of LGBTQ+ people in STEM [7] or in non-STEM workplaces [8] report that they are not out at work, meaning they are not open at work about their identity, with 35% reporting that they feel compelled to lie at work to conceal their identity. The high rates of nondisclosure are likely connected to LGBTQ+ people in STEM reporting more experiences of harassment, social isolation, career devaluation, depression, insomnia, and minor health issues compared to cisgender (i.e., their gender identity aligns with birth sex), heterosexual peers [9, 10].

Professional academic societies have taken initial steps to capture LGBTQ+ representation among their attendees. The American College of Neuropsychopharmacology (ACNP) is one such professional society, made up of neuropsychopharmacologists, psychiatrists, psychologists, and neuroscientists. The ACNP annual conference features leading research from preclinical and clinical investigators to promote a high-quality translational experience for members and nonmembers. Membership within ACNP is divided into different tiers that generally correspond to career seniority and achievements, including associate members, members, fellows, and emeritus. Membership is competitive with an acceptance rate of about 50% in 2020 as there are only a few open membership slots per year. Diversity within ACNP membership has been discussed, especially pertaining to male/female gender and racial diversity. In 2020, ACNP added a voluntary LGBTQ+ affiliation question to its demographic profile (i.e., Do you identify as LGBTQ+?), which is available to all members and nonmembers. People that chose to update their profile and respond could answer Yes, No, or Prefer not to answer. All other demographic questions were mandatory. These data were de-identified prior to being shared.

As of May 2021 (members have updated their profiles since the 2020 ACNP conference), there were 1282 members that had ACNP profiles (see Table 1). Only 456 members responded to the LGBTQ+ question (35.6% response rate). Within the respondents, 21 members identified as LGBTQ+ (4.6%; see Fig. 1A), 414 members did not identify as LGBTQ+ (90.7%), and 21 members preferred not to answer (4.6%). When looking at representation within each membership tier of those that responded, 4% of associate members, 4.6% of members and member emeritus, and 5.3% of fellows and fellow emeritus identify as LGBTQ+. Despite the low response rate, at first glance, the percentages do not seem too alarming. A recent large survey of STEM professional societies found only about 3% of respondents identified as LGBTQ+ [9] and a 2020 Gallup poll estimated that 5.6% of the United States population identifies as LGBTQ+. When split by age, the Gallup poll estimated 8.2% of Millennials and 3% of Generation X and Baby boomers identify as LGBTQ+. However, the voluntary nature of ACNP’s LGBTQ+ question raises questions about the accuracy of these data. People are more likely to answer voluntary questions that directly pertain to them; therefore, the ~5% is likely inflated from the true LGBTQ+ percentage among ACNP members, which could be 1.6% at the lowest (21/1282). If the 35.6% that responded are indeed representative of ACNP members, then the majority of LGBTQ+ members of ACNP (i.e., the LGBTQ+ members in the 64.4% of nonrespondents) did not to disclose their identity. This could be for a few possible reasons: they do not feel that disclosing their identity matters, they are not out to their colleagues, or they fear harassment and did not trust that the data would remain de-identified and confidential. Importantly, this was the only voluntary question on the demographic profile, despite having a nondisclosure option. Therefore, it is also possible that members that already have profiles did not see the question if most members decided not to update their profiles.

Table 1 ACNP member response breakdowns to the demographic profile question: do you identify as LGBTQ+?.
Full size table
Fig. 1: LGBTQ+ affiliation within ACNP members and nonmembers.
figure 1

ACNP member (A) and nonmember (B) response breakdowns to the demographic profile question: Do you identify as LGBTQ+?.

Full size image

As of May 2021, there were 11,666 nonmembers in the ACNP system. Of the 635 nonmembers that answered the question on their demographic profile, 81 identified as LGBTQ+ (12.8%; see Fig. 1B) and 515 did not identify as LGBTQ+ (81.1%), while 39 selected they prefer not to answer (6.1%). There were 1,149 nonmembers that updated their profile in 2020 but did not answer the question. A response rate was not calculated for the nonmembers because it would not accurately represent the response rate since the nonespondent nonmembers could be one-time attendees from years ago with no incentive to update their profile. Likewise, the nonmembers that updated their profile in 2020 but did not answer the question does not capture nonmembers that attended the 2020 conference and already had a profile made and did not update their profile in 2020. While the same self-report qualifications apply, the increase percentage raises additional points. Nonmembers are more likely to be trainees based on the nature of requirements for membership so it follows national trends that LGBTQ+ representation would be higher among nonmembers than members. However, it will be important to track member and nonmember representation going forward to assess whether LGBTQ+ people are underrepresented at the membership level, relative to population percentages.

Notably, ACNP, through the Diversity and Inclusion Task Force, has taken steps to be inclusive of LGBTQ+ people. At the 2019 conference they handed out pronoun buttons and at the 2020 conference they successfully enabled LGBTQ+ people to be invited to attend under the diversity invitation bank, included pronoun options on conference profiles, hosted a “Sex differences, gender bias, and trans-inclusive research practices” discussion, and hosted a LGBTQ+ networking session. Anecdotally, the well-attended networking session was mostly comprised of nonmember trainees with only a few ACNP members, most of whom were allies, not LGBTQ+ community members. Efforts like this should continue within ACNP and begin or continue within other scientific societies. Efforts should also ensure access to gender neutral bathrooms at conference sites, consider LGBTQ+ identity as a historically excluded minority for travel awards, ensure the conference is held within a country that affirms LGBTQ+ identities, and support LGBTQ+ scientist nominations for membership and leadership positions.

With regards to data collection going forward, ACNP should consider making LGBTQ+ affiliation questions mandatory (while still including a nondisclosure option) like their required demographic questions pertaining to race, ethnicity, gender, and disability status, or specifically encourage members to update their profiles so the collected data is representative. They should also consider tracking representation at the leadership level. However, extreme caution should be used here as small numbers of people in leadership could lead to LGBTQ+ people being identifiable. In addition, ACNP and other societies looking to follow in ACNP’s footsteps should be aware that by merging all LGBTQ+ people together, they may achieve LGBTQ+representation similar to the U.S. or international population within membership tiers, while specific groups under the LGBTQ+ umbrella remain underrepresented, like transgender researchers, whom face additional barriers. Other scientific societies including, but not limited to, Society for Neuroscience, Society for Biological Psychiatry, the International Behavioral Neuroscience Society, and the American Society for Pharmacology and Experimental Therapeutics, should strongly consider adding a LGBTQ+ identity question to their attendee/membership profiles in order to track representation of LGBTQ+ researchers within the society. Importantly, societies should also examine how people with intersecting identities are represented within the society and create equitable policies for groups with multiple underrepresented identities.

ACNP and many scientific societies should also modify their gender question. For the 2020 ACNP conference, the only options were male and female and conference attendees were required to pick one or the other. The response options force those with nonbinary, genderqueer, Two-Spirit, or another nonconforming gender identity to select a gender they are not, and could signal that the society is not welcoming or accepting of people that do not conform to male or female genders. In addition, the current ACNP prefix options do not include a non-gendered prefix, like Mx or None, for trainees that have yet to earn their doctorate. Many scientific societies, although not all, only include male/female options or include an “other” response in their profiles, which is exclusionary language. Instead of just a binary choice, the gender question and prefix selection should at least include a nonbinary/gender nonconforming option or a fill-in response, as well as a nondisclosure option.

Data are critical to identifying where disparities lie and are crucial for implementing policy changes, like identifying which groups qualify as underrepresented in STEM. Within the U.S., the National Science Foundation (NSF) tracks the STEM participation rates of women, racial and ethnic groups, disabled people, and those from disadvantaged backgrounds, but does not collect sexual orientation or gender identity (SOGI) information. Other federal agencies, like NIH, use NSF’s data to determine who is eligible for diversity-specific programs. By not collecting SOGI data, NSF and NIH are unable to systematically investigate underrepresentation or disparities experienced by LGBTQ+ researchers. While it is commendable that ACNP is collecting some of these data, changes need to be made to their system to include inclusive gender identity questions and improve response rates. It is imperative that ACNP and other societies collect these data, work together to advocate on behalf of LGBTQ+ scientists, and use their substantial influence to persuade NIH, NSF, and other federal organizations to collect these data. From continued pressure of LGBTQ+ advocates including Dr. Jon Freeman, the American Association for the Advancement of Science and the American Educational Research Association, NSF began piloting SOGI questions for its U.S. STEM workforce surveys on ~8,000 participants in April 2021. This is one way ACNP and other societies can partner with existing efforts to encourage NSF to collect these data. While challenging, societies can coordinate a collective response, to maximize the impact, to persuade NSF to add SOGI questions to all surveys and to persuade NIH to add LGBTQ+ identity or SOGI questions to their eRA Commons profiles for representation and potential funding disparities to be systematically examined within STEM fields. Not only will collection of these data affect policy, but including these questions shows that LGBTQ+ researchers are welcome and expected to be in STEM. LGBTQ+ people exist within STEM, our existence should be embraced, and our experiences matter. LGBTQ+ people deserve to be recognized and celebrated for who we are within STEM and cease being erased by exclusionary data collection practices.

Funding and disclosure

The author declares no competing interests.