The mission of the National Academy of Medicine (NAM) is to improve health for all by advancing science and accelerating health equity through independent, authoritative, and trusted advice nationally and globally. NAM members are elected in recognition of professional achievement and commitment to service in activities of the National Academies of Sciences, Engineering, and Medicine. Since 2008, the NAM has cultivated membership diversity through an adapted choice architecture, producing remarkable change. This process may be of interest to other mission-based organizations considering strategies to improve membership diversity.
The NAM has nearly 2400 members, and each year members nominate 900 or more candidates, from whom 100 will be elected. The core criteria for membership are:
In 2007, the NAM Council began to evaluate diversity due to concerns about potential biases in the discovery, nomination, and election process, which limited capacity to provide broad expertise and advice to the nation. The Council prioritized demographic diversity, defined as age (50 years or younger), gender, race and ethnicity (Alaska Native or American Indian, Asian or Asian American, Black, Hispanic, Middle Eastern or North African, and Native Hawaiian), and geography (recently changed to states with less than 10 members). To address important issues requiring subject matter expertise, each year the Council also defines a subject emphasis. Recent topics have included climate change and prevention of firearm injury. The Council set a general goal of electing a more diverse, representative membership and, in 2008, adopted choice architecture after considering other options. Choice architecture, as described by Thaler et al, "can help nudge people to make better choices (as judged by themselves) without forcing certain outcomes." Choice architecture is a product of social science research that offers simplifying strategies when there are numerous and complex choices. It provides structure to choice-making and can affect outcomes, like increasing diversity. The NAM adapted choice architecture maintains core membership criteria and honors member prioritization of candidates but uses a diversity ballot development architecture (complex choice structuring) that feeds back on the candidate discovery and nomination process (incentive to nominate more diverse candidates).
At the beginning of each nomination period, the NAM Membership Committee hosts virtual seminars to help members understand the election process and facilitate candidate discovery. The seminars describe the full election process, notably how diversity categories factor into consideration during final ballot development. This creates a feedback loop increasing the likelihood of more diverse nominations. Nominations feature the nominee's demographic characteristics in addition to accomplishments, expertise, potential contributions to the NAM, and a short bibliography.
Nominations are made within NAM subsections, and most subsections start with a confidential poll that prioritizes their nominees. NAM sections then hold a priority ballot vote of all nominees from their section, with the ability to see subsections candidate prioritization such that the first round of voting informs the next. In these 2 rounds, every nominee is considered and ranked: first by their peers (subsection) and then by the section. Diversity categories do not impose any preference architecture in these voting rounds.
The NAM Membership Committee, comprised of the chair and vice chair of each section, uses section priority ballot scores to build a final ballot for the full membership to vote a final time. The candidates receiving the top scores in each section populate nearly 40% of the final ballot. Section leaders next use section priority scores to propose nominees in demographic, at large, and age categories. This step creates a choice architecture that uses a diversity structure to organize nominees according to their sectional rankings in building the rest of the final ballot. Transparency about this ballot construction step creates a subtle but influential feedback loop on candidate nominations.
Subject emphasis candidates broaden experience and expertise diversity and are vetted by an expert panel of NAM members who prepare an advisory ranking that considers candidate impact in the emphasis area. The Membership Committee uses the panel ranking as well as diversity and section scores when selecting up to 5 subject emphasis candidates for the ballot.
Since 2008, the NAM has improved the diversity of elected members. We attribute this to an adapted choice architecture that influences candidate discovery and nominations as well as the creation of the final membership ballot. In 2008, women were 34% of newly elected members but this rose to 50% in 2022 and was 40% in 2023 (Figure A). Individuals from underrepresented race and ethnicity groups comprised approximately half of newly elected members in 2022 (49%) and 2023 (51%) (Figure B). Geographic diversity success has been more limited as the lack of NAM members in many states reduces awareness of worthy candidates. In 2022, 8 members were elected from an underrepresented state, but in 2023, only 1 such member was elected. Annual variation is a hallmark of choice architecture as opposed to quotas.
Increasing membership diversity is a long-term NAM effort. We hope that the health issues that are prominent in the US and other countries, including mental health, pandemic resilience, and health equity, will benefit from the NAM's membership diversity. For example, the Academy's Grand Challenge on Climate Change, Human Health, and Equity has been accelerated by the election of members who have related expertise. Choice architecture increases diversity without constraining members' ability to make their own choices, but it takes time for it to influence nomination diversity. One unintended consequence may have been that some sections were slower to recognize or use nomination strategies favored by the adapted choice architecture and reduced the likelihood of their candidates appearing on the ballot, but this has largely been self-correcting. We hope this informs other mission-based organizations seeking to increase membership diversity.
Corresponding Author: Robert L. Phillips Jr, MD, MSPH, National Academy of Medicine, The Center for Professionalism & Value in Health Care, American Board of Family Medicine, 1016 NW 16th St, Ste 800, Washington, DC 20036 ([email protected]).
Conflict of Interest Disclosures: Dr Phillips was chair of the Membership Committee of the National Academy of Medicine when this manuscript was written. No other disclosures were reported.
Additional Contributions: We are grateful to Donna Duncan, Sara Hitzig, BSFS, Jamal Samuel, Chiquita Boston, and Mi So Park, MPH, who make the National Academy of Medicine (NAM) Membership and Governance process so successful. We are also grateful to NAM President, Victor J. Dzau, MD, for his very helpful reviews of this manuscript; compensation was not received. We thank Jamie Durana, MA, and Laura DeStefano, BA, National Academy of Science, for their contributions to first drafts of the manuscript and for producing the figure; they did not receive compensation. Finally, we are grateful to members and staff whose efforts more than a decade ago laid the foundations of the choice architecture that has helped the NAM serve the country better.