Researchers analyzed three socioeconomic markers within a common stage-decomposition framework. The primary comparison grouped records classified as EO1 or EO2 into a lower Education-Occupation group, reflecting lower parental education and/or nonprofessional occupational categories. The higher group combined EO3, EO4 and EO5, representing degree-holding parents in administrative, managerial or professional occupations. Separate analyses examined first-generation college students and users of AMCAS fee assistance. Available data covered 2018–2019 through 2025–2026 for Education-Occupation status, 2020–2021 through 2025–2026 for first-generation status and 2022–2023 through 2025–2026 for fee-assistance use.

All three focal groups had lower acceptance and applicant-to-matriculant pipeline rates than their comparison groups in 2025–2026. First-generation applicants’ acceptance rate was 11.2 percentage points below the comparison rate, while the acceptance rate for AMCAS fee-assistance users was 7.3 points lower. The corresponding pipeline-rate gaps were 11.2 points for first-generation applicants and 22 points for fee-assistance users. By contrast, post-acceptance matriculation-yield gaps were comparatively small, ranging from −1.0 to −0.4 percentage points across the three comparisons. In the aggregate data, the largest downstream difference therefore appeared in the full pipeline for fee-assistance users, not in enrollment among those who had already been accepted.

The study quantified the stage contributions with two complementary descriptive calculations. A direct representation decomposition tracked changes in each group’s share from applicants to acceptees and from acceptees to matriculants. It attributed 96.6% to 97.9% of the decline from applicant share to matriculant share to the application-to-acceptance stage. The second calculation, a stage equalization, replaced each focal group’s acceptance rate with the rate for its reference group while leaving the other stages unchanged. Doing so reduced the pipeline-rate gaps by 95.7% to 97.6%. The close agreement between these descriptive results places nearly all of the measurable difference at the application-to-acceptance transition in the available national aggregates.

Because the analysis was based on aggregate AAMC counts and descriptive trend and decomposition methods, its findings describe where visible national patterns were concentrated rather than identifying causes of individual admissions decisions. The authors identified the application-to-acceptance stage as the priority for further individual- and institution-level research into the mechanisms behind these patterns and their implications for admissions policy and downstream outcomes. The results provide a stage-specific account of the admissions pipeline: acceptance rates differed substantially across the three socioeconomic markers, whereas subsequent matriculation yields among accepted applicants differed much less. The study also used different available periods for the three markers, so the comparisons describe the national aggregate records available for each measure during 2018–2019 through 2025–2026.

Subject of Research: National socioeconomic differences across the U.S. medical school admissions pipeline

Article Title: Where do socioeconomic differences emerge in the U.S. medical school admissions pipeline a repeated cross-sectional analysis using AAMC FACTS data from 2018 to 2026

Article References: Qian, M., Yan, C., & Liu, Y. (2026). Where do socioeconomic differences emerge in the U.S. medical school admissions pipeline a repeated cross-sectional analysis using AAMC FACTS data from 2018 to 2026. BMC Medical Education. https://doi.org/10.1186/s12909-026-10284-w

Image Credits: AI Generated

DOI: 10.1186/s12909-026-10284-w

Keywords: medical school admissions, socioeconomic disparities, AAMC FACTS, acceptance rates, matriculation yield, first-generation students, fee assistance, education-occupation status, admissions pipeline, socioeconomic, differences, emerge