Methods

Data

Publications, citations, and field normalization come from OpenAlex, an open (CC0) catalog of scholarly works. Data refreshes daily. This site is not affiliated with or endorsed by ACGME or any program. The program list comes from the CMS teaching hospital list (public domain) and each program's own website.

Coverage

How much of the program universe has been collected so far. Updated after every pipeline run.

Research environment (every program)

Each program is matched to the OpenAlex institution of its sponsor. Some programs map to more than one institution, such as a university and its medical center. Works are then counted across those institutions and everything affiliated with them, such as hospitals and the VA, in the specialty's subfields (listed below). The window runs from 2022. Impact measures use only works published before 2025, because recent papers haven't had time to be cited.

Works / yr

Average number of articles and reviews per complete calendar year whose primary topic falls in the specialty's subfields, with at least one author at the program's institution or an affiliated one. This reflects the size of the institution, not just quality.

2-yr growth

Works in the last two complete years compared with the two years before.

% top 10% of field

Share of works in the top 10% of citations for their field and publication year (OpenAlex citation-normalized percentile). Independent of size.

% above world average

Share of works with field-weighted citation impact (FWCI) above 1, meaning cited more than the world average for comparable work. Independent of size.

Internal Medicine:

Internal Medicine, Infectious Diseases, Nephrology, Endocrinology, Diabetes and Metabolism, Cardiology and Cardiovascular Medicine, Pulmonary and Respiratory Medicine, Gastroenterology, Hepatology, Hematology, Oncology, Rheumatology, Geriatrics and Gerontology, Critical Care and Intensive Care Medicine, Immunology and Allergy

Family Medicine:

Family Practice, General Health Professions

Pediatrics:

Pediatrics, Perinatology and Child Health

Psychiatry:

Psychiatry and Mental Health

Emergency Medicine:

Emergency Medicine

General Surgery:

Surgery

Obstetrics and Gynecology:

Obstetrics and Gynecology, Reproductive Medicine

Neurology:

Neurology

Anesthesiology:

Anesthesiology and Pain Medicine, Critical Care and Intensive Care Medicine

Diagnostic Radiology:

Radiology, Nuclear Medicine and Imaging

Pathology:

Pathology and Forensic Medicine

Dermatology:

Dermatology

Orthopaedic Surgery:

Orthopedics and Sports Medicine

Otolaryngology:

Otorhinolaryngology

Ophthalmology:

Ophthalmology

Urology:

Urology

Physical Medicine and Rehabilitation:

Rehabilitation, Physical Therapy, Sports Therapy and Rehabilitation

Neurological Surgery:

Neurology, Surgery

Radiation Oncology:

Oncology, Radiology, Nuclear Medicine and Imaging

This measures the research happening where a program trains, not the program itself. Large academic systems produce more works partly because they're large, which is why the two percentage measures matter more for comparison. A multi-campus system (Mayo Clinic, for example) is counted as one institution.

Program leadership (where available)

Some programs also have a verified roster of their program director, associate program directors, and core faculty. It's compiled from the program's own website and matched to OpenAlex authors: by ORCID when available, otherwise by name among people affiliated with the institution, scored on name match and topic overlap. Only clear matches are accepted automatically. When OpenAlex splits one person into a large profile and small fragments, the profile with at least five times the works of every close rival is accepted, and that choice is recorded for audit. The rest is reviewed by hand.

Faculty h-index

Lifetime h-index for each listed faculty member; the median and interquartile range across the program.

Faculty works

Unique works with at least one faculty member as an author, within the window.

Median FWCI

Field-weighted citation impact of those works.

Trainee first-author works

Works where a resident, chief resident, or fellow is the first author.

Privacy

• Trainees (residents, chief residents, fellows) are counted in totals but never listed by name.
• Anyone listed can ask not to be. Request removal. They're removed from the page, and their work still counts toward program totals.

Limitations

• These numbers describe research output only. They say nothing about the quality of clinical training.
• Matching a program to an institution is a judgment call; mappings are reviewed and can be corrected.
• Author disambiguation in OpenAlex has errors. Review reduces them but doesn't eliminate them.
• Programs come from the CMS list of teaching hospitals (Medicare GME recipients). Each hospital is checked against its own program web pages to record which programs it sponsors, and the list grows daily.
• A specialty is defined by the subfields of its papers' primary topics. That fits internal medicine well but undercounts specialties whose research OpenAlex files elsewhere: clinical neurology often lands in neuroscience, and surgical subspecialties share “Surgery.” Comparisons within a specialty stay fair, because every program is measured the same way, but absolute counts for those specialties run low. Separate children's hospitals aren't yet counted toward pediatrics.
• Percentages are hidden when a program has fewer than 20 citation-mature works, because they'd be noise.