The 16 most populous specialties self-reported by medical graduates in 2016 are displayed. The following specialties were excluded: allergy, colorectal surgery, medical genetics, neurosurgery, physical medicine and rehabilitation, preventive medicine, radiation oncology, thoracic surgery, nonmedical, undecided, other unclassified, or those with missing survey responses to intended specialty. Combined medicine and pediatrics was excluded because data were not available for 2010. OB/GYN indicates obstetrics and gynecology.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Grischkan J, George BP, Chaiyachati K, Friedman AB, Dorsey ER, Asch DA. Distribution of Medical Education Debt by Specialty, 2010-2016. JAMA Intern Med. 2017;177(10):1532–1535. doi:10.1001/jamainternmed.2017.4023
Among medical students graduating with educational loans, the mean debt was $32 000 in 1986, which is approximately $70 000 in 2017 dollars.1 Rising tuitions and a growing reliance on loans increased mean medical education debt to $190 000 by 2016.2,3 Alongside these well-known trends is a quieter increase in the proportion of graduates without debt. The combination of these observations suggests a concentration of debt among fewer individuals—a finding that is obscured by population means.1 We sought to analyze the trends in the distribution of medical education debt by focusing on the increase in graduates without such debt.
We used deidentified data from the 2010-2016 Association of American Medical Colleges Graduation Questionnaire (response rates, 78%-83%). We limited analyses to graduates who responded to questions about self-reported medical education debt. We adjusted figures to 2016 dollars4 and stratified medical education debt into the following 4 categories: no debt, $1 to $100 000, $100 001 to $200 000, and greater than $200 000. We performed a sensitivity analysis on the distribution of debt using total educational debt (ie, including debt incurred before medical school). We additionally examined changes in scholarship funds over time. This work was deemed not to be human subjects research.
The number of medical school graduates reporting on debt was 12 786 of 13 904 (92.0%) in 2010 and 13 610 of 15 232 (89.4%) in 2016. Among those with debt, mean amount of debt was $161 739 in 2010 and $179 068 in 2016 (Table). Comparing 2010 and 2016, the proportion of graduates reporting no debt increased from 16.1% (2056 of 12 786) in 2010 to 26.9% (3655 of 13 610) in 2016, while other debt categories remained relatively stable or declined (Figure, A). The exception was those with greater than $300 000 in debt, where the proportion of graduates doubled from 2.1% (263 of 12 786) in 2010 to 4.2% (565 of 13 610) in 2016.
By specialty (Figure, B), the cohort with no debt experienced the largest absolute increase between 2010 and 2016. Six specialties experiencing the largest absolute increase in no debt were radiology (17.1% in 2010 and 30.4% in 2016), dermatology (23.1% in 2010 and 36.1% in 2016), neurology (18.1% in 2010 and 30.7% in 2016), obstetrics and gynecology (11.5% in 2010 and 24.7% in 2016), ophthalmology (26.3% in 2010 and 39.9% in 2016), and pathology (20.8% in 2010 and 34.8% in 2016). When examining total educational debt, the mean amount of debt increased to $177 362 in 2010 and $197 426 in 2016, but distribution by specialty remained unchanged.
The mean scholarship funding among recipients did not substantially increase to account for growth in medical graduates without debt, from $53 065 in 2010 to $58 136 in 2016. Among those without debt, the mean amount of scholarship funding declined from $135 186 in 2010 to $52 718 in 2016.
Although the real increase in medical student debt is well known, this study offers 3 new observations. First, the proportion of students graduating with no debt is also increasing. Although this finding seems positive, when paired with a decline in scholarship funding within this debt-free cohort, the finding suggests a concentration of medical students with wealthy backgrounds. Second, when paired with an increase in aggregate per capita debt, this finding suggests that medical education debt is concentrated in fewer individuals. Such concentration disguises individual debt burdens behind aggregate debt estimates that, although high, are still lower than what an increasing number of students face. Third, these changing distributions vary considerably by specialty choice. There is no clear association between specialty-specific proportions of graduates without debt and the income typical of those specialties, but primary care–oriented fields seem to have less of an increase in graduates without debt.
The causal associations among debt, specialty choice, and income are challenging to disentangle.5 Conceptually, debt is likely to be less of a determinant of specialty choice than is future income. There is also no consensus on the balance across the medical specialties to best meet the United States’ workforce needs. But to the extent that specialty choice is important and that indebtedness may be associated with it, we need to begin to examine second-order debt effects and, in particular, its distribution across specialties.
Corresponding Author: David A. Asch, MD, Center for Health Care Innovation, Perelman Center for Advanced Medicine, University of Pennsylvania, 3400 Civic Center Blvd, 14-171, Philadelphia, PA 19104 (email@example.com).
Accepted for Publication: June 26, 2017.
Correction: This article was corrected on October 2, 2017, to fix an error in an Author Affiliation.
Published Online: September 5, 2017. doi:10.1001/jamainternmed.2017.4023
Author Contributions: Dr George had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Mr Grischkan and Dr George contributed equally to this study.
Study concept and design: Grischkan, George, Chaiyachati, Friedman.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Grischkan, George, Chaiyachati, Asch.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Grischkan, George, Friedman.
Study supervision: Chaiyachati, Friedman, Asch.
Conflict of Interest Disclosures: None reported.