High-Income Households Overrepresent US Medical School Student Body

Arman A. Shahriar, BS

The US medical school student body is overrepresented with individuals from high-income households relative to the national population, according to findings from a new research letter.

The new data reviewing socioeconomic makeup of the 2017-2019 US medical school student body highlighted the need to better represent persons of lower economic status and diverse racial and ethnic groups in the future class of clinicians.

Led by Arman A. Shahriar, BS, of the University of Minnesota Medical School, the team of US-based investigators sought to compare recent medical school student composition with that of the national population, based on household incomes, race, ethnicity, and sex. Workforce diversity has been previously linked to improved health care quality, they wrote—yet postsecondary education students, including those in medical school, have disproportionately represented high-income households.

“Significant attention has been rightfully directed toward racial-, ethnic-, and gender-based diversity in medical schools, with little mention of socioeconomic diversity, which is a less visible form of diversity,” they explained.

Shariar and colleagues assessed data from the Association of American Medical Colleges Matriculating Student Questionnaire (AAMC-MSQ) from 2017-2019, compared to the Census Bureau’s Current Population Survey Annual Social and Economic Supplement (CPS-ASEC) from 2016-2018, to interpret matriculating allopathic medical students’ parental income versus the national average, respectively.

Their AAMC-MSQ annual data included a 65% to 71% response rate for each observed year. The team categorized questionnaire and census respondents into household income groups based on national quintile-based income limits; they additionally compared medical student household incomes with the general population for each race and ethnicity.

A representation index (RI) was used to define the ratio of proportions for medical student subgroups overrepresenting (>1.0) or underrepresenting (<1.0) the general US population. Statistical significance was set at P values <.05.

Investigators observed 44,903 AAMC-MSQ respondents, of whom 30,373 (67.6%) reported parental income. Of that group, approximately half (50.5%) belonged in the highest quintile of household incomes; another quarter (24.0%) were in the nation’s top 5% of income. More than half (52.4%) were women.

The sample included 54.0% non-Hispanic White students, 21.3% non-Hispanic Asian students, 10.9% Hispanic students of any race, and 6.4% non-Hispanic Black students. Investigators stated the assessment captured 46.4% of all matriculating allopathic medical students in the US.

Shariar and colleagues observed a consistent overrepresentation of the top 5% household incomes in the medical school subgroups versus the general population, including:

• Non-Hispanic White (28.6% vs 5.9%; RI, 4.8; P <.001)
• Non-Hispanic Asian (21.7% vs 9.5%; RI, 2.3; P <.001)
• Hispanic of any race (14.7% vs 2.2%; RI, 6.6; P <.001)
• Non-Hispanic Black (9.1% vs 1.7%; RI, 5.3; P <.001)

The highest quintile for parental income was overrepresented for each observed race and ethnicity among medical school students, while the lower 3 quintiles were each underrepresented. Investigators did not that variability by medical student sex and parental income was not as pronounced.

As the team noted, the ideal of demographic representation among physicians is “widely accepted,” but progress is lacking in efforts to reach such a goal.

“Our findings suggest that underlying the lack of progress may be the inaccessibility of the profession to low-income students, who, owing to powerful historical and contemporary forces, like structural racism, are disproportionately students who identify as Black or Hispanic,” they wrote. “A low socioeconomic status significantly decreases the likelihood that a student who is interested in medicine will apply or gain acceptance into medical school.”

They advocated for improved assessment of prospective students’ socioeconomic disadvantage during the admissions process; already present requirements including written essays and validation questions on applications could help indicate qualified students with a disadvantaged household income.

“Likewise, common metrics, such as grade point average and Medical College Admission Test scores, can be adjusted for socioeconomic disadvantage,” Shariar and colleagues noted. “Long-term solutions will require upstream engagement, including community partnerships and targeted investments in pipeline programs.”

Despite limitations including the validity of self-reported parental incomes and lacking AAMC-MSQ responses, the team concluded that further exploration into the makeup of medical school student bodies should consider the factors resulting in differing income, racial and ethnic representation.

The study, “Socioeconomic Diversity of the Matriculating US Medical Student Body by Race, Ethnicity, and Sex, 2017-2019,” was published online in JAMA Network Open.