Physician Burnout by Specialty Statistics 2025

Emergency medicine physicians had the highest burnout percentage among six specialties highlighted in the American Medical Association's 2024 national physician comparison: 52.2% reported at least one symptom of burnout. Family medicine followed at 46.4%, then obstetrics and gynecology at 45.8%.

These numbers are useful benchmarks, but they are not a ranking of every medical specialty in the United States. The comparison reflects nearly 18,000 physician responses from more than 100 participating health systems and organizations across 43 states. The organizations and specialties represented in a voluntary survey affect the result.

The AMA released these 2024 findings in 2025. A page labeled 2025 statistics should describe the newest reported measurement period, not imply that responses were collected in 2025.

Burnout rates by specialty

The AMA reported these 2024 burnout percentages for six specialties. Burnout meant reporting at least one symptom on the measure used in its Organizational Biopsy program.

Source: American Medical Association 2024 national physician comparison. The earlier figures are the comparison values published with the 2024 estimates. Subtraction gives the percentage-point change.

All six published specialty estimates decreased. Internal medicine had the largest reduction in this group, falling 9.4 percentage points. That change does not establish a cause. Survey participation, organizational mix, working conditions, and the period being measured can all influence a repeated cross-sectional comparison.

Across physicians from all participating organizations, 43.2% reported at least one burnout symptom in 2024. The comparable AMA figure was 48.2% in 2023 and 53% in 2022. This is a substantial improvement from the recent peak, but more than four in ten respondents still met the survey threshold.

How the specialties compare

Emergency medicine was 9.0 percentage points above the 43.2% participating-organization average. Family medicine was 3.2 points above it, and obstetrics and gynecology was 2.6 points above it. Pediatrics, internal medicine, and hospital medicine were below that benchmark.

Those gaps are descriptive, not adjusted comparisons. They do not account for age, career stage, work hours, night call, staffing, ownership, local demand, or organizational culture. They do not prove that specialty alone produced the difference.

Frontline specialties share pressures, but the work is not interchangeable. Emergency physicians face unpredictable volume, acute clinical risk, shift work, and boarding constraints. Primary care physicians often manage large inboxes, longitudinal coordination, documentation, and payer requirements. Obstetrician-gynecologists combine outpatient work, procedures, call, and time-sensitive deliveries.

The AMA published job-stress results for the same specialties, but stress and burnout were separate indicators. Emergency medicine had 51.1% reporting job stress; obstetrics and gynecology, 50.7%; family medicine, 48.8%; pediatrics, 44.1%; internal medicine, 43.3%; and hospital medicine, 40.6%. A physician can report high stress without crossing a burnout threshold, so the measures should not be merged.

Job satisfaction is not simply the inverse of burnout. In the same comparison, satisfaction ranged from 73.1% in emergency medicine to 80% in obstetrics and gynecology. A respondent may find clinical work meaningful while also experiencing emotional exhaustion.

Why estimates differ

Burnout prevalence depends on the population, instrument, threshold, and timing. The Agency for Healthcare Research and Quality describes burnout as a long-term stress reaction involving emotional exhaustion, depersonalization, and a reduced sense of personal accomplishment. Studies may use the Maslach Burnout Inventory, a single-item measure, or another validated instrument. Their cutoffs are not automatically equivalent.

A national survey published in 2022 found that 62.8% of physicians had at least one manifestation of burnout in 2021, compared with 38.2% in 2020. That study sampled physicians nationally during the COVID-19 pandemic and used its own design. It should not be spliced into the AMA specialty table as though it were another year from the same series.

A separate survey of California physicians conducted from June 2021 through March 2022 reported 45.4% burnout. Researchers applied weighting to address specialty representation and response bias. Its geography, collection period, and sampling method differ from the 2024 AMA comparison.

Training status changes the population as well. A large study of second-year resident physicians found 45.2% reported burnout symptoms, with significant specialty differences. Resident findings should not be presented as rates for all practicing physicians because schedules, supervision, autonomy, and career stage differ.

Even within one specialty, an average can conceal workplace differences. Research on physicians and advanced practice professionals has linked stronger teamwork and safety climates with lower odds of burnout. A specialty label cannot capture schedule control, staffing adequacy, leadership behavior, documentation design, or psychological safety.

What practices can measure

A solo or independent practice cannot infer its own risk from a national specialty percentage. It can use the benchmark to start a structured review, then measure the conditions it can change.

Track work hours and where they occur. Separate scheduled clinical time, same-day documentation, after-hours EHR work, inbox management, prior authorization, billing review, and staff supervision. A weekly total alone will not reveal which work is creating spillover.

Track operational signals alongside a confidential well-being measure. Examples include unsigned charts, inbox age, prior-authorization turnaround time, unused appointment capacity, staff vacancies, and time spent on tasks that do not require physician judgment. These measures diagnose workflow; they do not diagnose a mental health condition.

Use a validated survey consistently. Changing the question or threshold between periods makes trend interpretation unreliable. In a very small practice, protect anonymity and avoid granular reporting that could identify an employee.

Administrative redesign should focus on systems. AHRQ summarizes workplace factors associated with burnout, including time pressure, chaotic environments, low control over pace, and unfavorable organizational culture. Suggested approaches include removing unnecessary work, providing documentation time, improving information-technology support, and building teams around workflow improvement.

Delegation can be part of that work when tasks are appropriate, documented, and supervised. Scheduling, records follow-up, referral coordination, and defined billing support may be candidates. Clinical decisions, unapproved medical advice, and work restricted by law or payer rules are not. The specialty statistics do not prove that hiring a virtual assistant will reduce burnout, so outcomes should be measured rather than promised.

Our overall physician burnout rate statistics page covers national trend context. The physician work-week statistics brief separates hours from burnout, and the after-hours EHR statistics page focuses on documentation outside scheduled time.

Frequently asked questions

Which physician specialty had the highest burnout rate in 2024?

Emergency medicine had the highest percentage among the six specialties highlighted in the AMA comparison, at 52.2%. The report was not a complete ranking of every specialty.

What was the family medicine burnout rate?

The 2024 AMA comparison reported 46.4% for family medicine, down from the 51.0% comparison figure published by the AMA.

Did physician burnout improve in 2024?

Yes, within the AMA's participating organizations. Overall burnout fell from 48.2% in 2023 to 43.2% in 2024, and each of the six highlighted specialty percentages declined.

Can burnout rates from different surveys be compared directly?

Not without checking the instrument, threshold, population, sampling, and collection period. Two surveys can produce different prevalence estimates even when both are methodologically sound.

Do high specialty rates mean every physician in that field is burned out?

No. A specialty percentage estimates the surveyed share meeting a defined threshold. It does not predict an individual's condition or explain why a respondent met that threshold.

Methodology and sources

We treated the AMA's 2024 national physician comparison as the primary dataset because it reports the same measure across six specialties in one period. We calculated percentage-point changes by subtracting the published comparison figure from the 2024 estimate. National, government, and peer-reviewed research explains why estimates from different surveys should remain separate.

This article reports group-level workplace research and is not medical advice or a diagnostic tool. Physicians experiencing distress can seek confidential support through their employer, state physician health program, or an appropriate licensed clinician. In an immediate crisis in the United States, call or text 988.