Purpose: Seventy-five percent of medical students in the United States hail from families in the top 2 income quintiles as defined by the U.S. Census Bureau, while 5% of medical students represent the bottom quintile. 1 Analyses to identify potential barriers contributing to this staggering socioeconomic disparity among our future medical workforce are sparse. Characterization of the exacerbating factors and potential bottlenecks along the premedical pipeline can help inform crucial steps toward initiatives to increase representation of medical school matriculants from all socioeconomic backgrounds. This study examined U.S. medical school acceptance rates across familial income and parental education levels and identified applicant factors predictive of medical school acceptance. Approach/Methods: The American Medical College Application Service (AMCAS) collects the highest level of parental education and childhood family income for every applicant. De-identified AMCAS data were obtained for all medical school applicants from 2014 to 2019 and included the following applicant characteristics: science and nonscience GPA, Medical College Admission Test (MCAT) 2015 score, age, parental education, family income, race/ethnicity, and the number of medical school acceptances obtained. Descriptive statistical analyses assessed acceptance rates, MCAT averages, and GPA averages, stratified by applicant’s family income and parents’ education. Low income was defined as reporting a childhood family income <$75,000, and first-generation status was defined as having no parent with a bachelor’s degree. Multivariate regression analysis modeled the effects of MCAT, GPA, race/ethnicity, first-generation status, and low-income status on acceptance to at least 1 MD program. Results/Outcomes: AMCAS data from 312,898 applicants were analyzed. Of the total, 34.3% (107,396/312,898) were low-income and 19.3% (60,328/312,898) were first- generation applicants. The overall acceptance rate over the study period was 42.3%. Low-income applicants and first-generation applicants’ acceptance rates were 36.0% (38,674/107,396) and 32.7% (19,701/60,328), respectively. On univariate analysis, acceptance was negatively associated with both family income (OR: 0.602; P < .001; CI: 0.592, 0.612) and parental education (OR: 0.581; P < .001; CI: 0.570, 0.592). On multivariate analysis, among those with MCAT2015 scores (n = 142,961), medical school acceptance was most affected by average science GPA (OR: 6.63; P < .001; CI: 6.345, 6.918), underrepresented in medicine (UIM) identity (OR: 5.119; P < .001; CI: 4.923, 5.323), and MCAT score (OR: 1.186; P < .001; CI: 1.184, 1.189). Low-income status (OR: 0.968; P < .05; CI: 0.938, 0.999) was negatively associated with acceptance; however, first-generation status (OR: 0.994; P = .758) was not significantly associated with acceptance. Discussion: These results reveal that first-generation and low-income medical school applicants have a markedly lower rate of acceptance than the general applicant population. Furthermore, both first-generation and low-income status are significantly negatively associated with medical school acceptance. This effect is mediated by UIM identity and GPA and MCAT metrics. Notably, UIM identity status had a significant effect size in predicting medical school acceptance, which is potentially reflective of beneficial race-conscious admissions practices. Furthermore, the miniscule effect size of first-generation college status and low-income status serves as evidence that disadvantaged backgrounds are not widely considered among admission practices as recommended by the Liaison Committee on Medical Education (LCME) Standard on Diversity. 2 Additionally, one study’s analysis of group differences in scores on the latest version of the MCAT revealed disparities in scores and use of preparatory materials depending on an applicant attending a school with more or fewer resources. 3 Therefore, first-generation college and low-income applicants’ relatively poorer acceptance rates may be mitigated by early academic support and access to cost-prohibitive MCAT preparatory materials. Significance: First-generation college and low-income applicants experience lower medical school acceptance rates, but high grades and test scores mediate the effect, in combination with UIM status. These applicant populations require increased support and mentorship to achieve proportional representation among medical students.
Read full abstract