aamc.org does not support this web browser.

Report on Residents

Table C5. Physician Retention in State of Residency Training, by Last Completed GME Specialty and Gender

Residents Who Completed Training, 2012-21

The table below displays the number and percentage of individuals who completed residency training from 2012 through 2021, are not currently active in any GME program, and are practicing physicians in or out of the state of residency training, by specialty and gender. Data are for graduates of MD-granting and DO-granting U.S. and Canadian schools and of international medical schools. For example, 48.5% of men (205) and 60.8% of women (522) who completed a residency program in Allergy and Immunology from 2012 through 2021 went on to practice in the same state where they completed residency training. The data are displayed by the most recently completed ACGME-accredited specialty or subspecialty. Counts and percentages shown for specialties do not include counts and percentages for the related subspecialties. Please contact residentreport@aamc.org with any comments or questions.

Selected Finding: Overall, 58.5% of women who completed residency from 2012 through 2021 are practicing in the state where they did their residency training, compared with 52.3% of men. The pattern of retention varies across specialties. 

Download Excel version

ACGME-Accredited Specialties and Subspecialties Completed Men Women Total
In State Out of State In State Out of State
Number Percent Number Percent Number Percent Number Percent

Note: Individuals whose gender was unavailable were excluded. Data for physicians not represented in the AMA Physician Masterfile may be missing. About 10% of individuals who completed their residencies and are included in GME Track® data are not represented as active physicians in the AMA Physician Masterfile. As a result, practicing physicians may be underrepresented. Specialties with no reported residents were excluded.

Source: GME Track® as of GMEdate, YYYY1, and AMA Physician Masterfile as of AMAdate, YYYY2.