163 Physician Burnout Statistics October 2026

Physician burnout statistics for 2026 show a problem that has eased since the pandemic but has not gone away. In a national survey of 1,000 physicians fielded in August 2026, 55% said they often have feelings of burnout (Physicians Foundation, Sep 2026). Doctors also burn out more often than other workers: after adjusting for age, gender, relationship status and hours worked, physicians had 1.82 times the odds of burnout of other US workers in 2023 (Mayo Clinic Proceedings, Apr 2025).

Documentation load shows up in almost every section below. In a study that checked EHR event logs against direct observation, family physicians spent 5.9 hours of an 11.4-hour workday in the electronic health record (Annals of Family Medicine, Sep 2017). Charting and inbox work, much of it done after hours, recur as drivers throughout the research.

Surveys define burnout differently, so their rates are reported side by side and never combined.

How many physicians are burned out in 2026?

Physician burnout has fallen every year since its 2021 peak, but only back to about where it stood before the pandemic, and doctors still burn out more often than other US workers. Three national measures track it with three different instruments, so every rate below names its source and its definition.

What is the physician burnout rate this year?

It depends on the instrument: the AMA's Mini-Z dataset reads lower than surveys using the full Maslach inventory or a self-report question.

• 42.9% of physicians said they felt a great deal of stress because of their job in 2025, down from 45.1% in 2024 and 50.7% in 2023 (AMA, Apr 2026).
• On the full Maslach Burnout Inventory, 45.2% of 7,643 US physicians had at least one symptom of burnout in 2023, compared with 62.8% in 2021, 38.2% in 2020, 43.9% in 2017, 54.4% in 2014 and 45.5% in 2011 (Mayo Clinic Proceedings, Apr 2025).
• 65.1% of physicians said in 2023 that they would choose to become a physician again, up from 57.1% in 2021 but still below 72.2% in 2020 (Mayo Clinic Proceedings, Apr 2025).
• The share of physicians who often feel burned out fell from 60% in 2024 to 54% in 2025 on the same self-report question (Physicians Foundation, Sep 2025).
• 58% of physicians felt debilitating levels of stress in the past year (Physicians Foundation, Sep 2026).

The rates differ because the definitions do: a review of 182 studies covering 109,628 physicians in 45 countries found burnout prevalence ranging from 0% to 80.5%, with at least 142 different definitions of burnout in use (JAMA, Sep 2018).

Which specialties have the highest and lowest burnout rates?

Emergency medicine had the highest rate in the AMA's 2025 data at 49.8%, followed by urological surgery at 49.5%, hematology/oncology at 49.3% and obstetrics and gynecology at 45.7% (AMA, Apr 2026).

• The lowest burnout rates in 2025 were in infectious diseases (23.3%), ophthalmology (25.8%), pathology (28.3%), nephrology (29.3%), dermatology (31.5%) and psychiatry (31.6%) (AMA, Apr 2026).
• In the 2023 national Maslach survey, emergency physicians had 3.01 times the adjusted odds of burnout of internal medicine subspecialists, with neurology (1.51) and general internal medicine (1.39) also elevated and general surgery subspecialties lower (0.71) (Mayo Clinic Proceedings, Apr 2025).
• At the 2021 pandemic peak, 23 of 24 specialties had a higher share of physicians with burnout symptoms than in 2020, and emotional exhaustion hit its highest level on record in 18 of 24 (Mayo Clinic Proceedings, Sep 2022).
• 44.3% of 4,905 early-career family physicians, surveyed by the American Board of Family Medicine three years after residency between 2020 and 2023, had symptoms of burnout (Annals of Family Medicine, Sep 2025).
• 64.8% of 3,741 obstetrics and gynecology residents met the criteria for burnout on a two-item Maslach measure in a January 2022 national survey taken alongside the specialty's in-training exam (Journal of Graduate Medical Education, Oct 2024).

Are women physicians and early-career doctors more burned out?

Women are, in every dataset on this page, and the gap holds after adjusting for specialty and hours. Career stage follows a curve rather than a straight line: burnout peaks in the middle years and is lowest among the most senior physicians.

• 53.3% of women physicians felt valued by their organization in 2025, against 59.6% of men (AMA, Apr 2026).
• Being a woman raised the adjusted odds of burnout by 29% in 2023 (odds ratio 1.29), after controlling for age, relationship status, specialty, hours and practice setting (Mayo Clinic Proceedings, Apr 2025).
• The gender gap widened during the pandemic: women physicians had 2.02 times the adjusted odds of burnout of men in 2021, up from 1.27 in 2020 (Mayo Clinic Proceedings, Sep 2022).
• Among early-career family physicians, 48.4% of women had burnout symptoms against 38.5% of men, an adjusted odds ratio of 1.56 (Annals of Family Medicine, Sep 2025).
• Women obstetrics and gynecology residents reported burnout more often than men, 66.9% against 56.7% (Journal of Graduate Medical Education, Oct 2024).
• Burnout peaks mid-career: in 2025 it was 48.8% for physicians 6 to 10 years out of residency or fellowship, 40.8% for those 1 to 5 years out and 35.8% for those 20 or more years out (AMA, May 2026).
• Physicians aged 65 and older had 59% lower adjusted odds of burnout than those under 35 (odds ratio 0.41) in 2023 (Mayo Clinic Proceedings, Apr 2025).
• Six in 10 residents and seven in 10 medical students reported often experiencing burnout in 2024 (Physicians Foundation, Sep 2024).
• 65% of physicians aged 45 or younger felt debilitating stress in the past year, compared with 54% of physicians aged 46 and older (Physicians Foundation, Sep 2026).

How burned out are nurses and other healthcare workers?

Nurses are the largest group in the health workforce, and they report burnout at least as often as physicians do. The nurse burnout statistics that matter most for hospitals are about intent to leave, because that is where burnout turns into unfilled shifts.

What share of nurses report burnout?

Between four and five in ten US nurses, depending on the survey. NCSBN found 45.1% of nurses felt burned out "a few times a week" or "every day" in its 2022 workforce survey of 29,472 RNs and 24,061 LPN/VNs across 45 states (NCSBN, Apr 2023).

• More nurses in that survey felt used up (56.4%) or emotionally drained (50.8%) a few times a week or more than said they were burned out, and 29.4% said they were at the end of their rope (NCSBN, Apr 2023).
• 62% of nurses in the same sample said their workload increased during the pandemic (NCSBN, Apr 2023).
• 48.5% of hospital nurses in New York and Illinois had high burnout after the pandemic (December 2023 to March 2024), against 48.0% before it and 51.0% during it, in a study of 50,044 nurses (Medical Care, Feb 2026).
• 61.5% of those hospital nurses said there were not enough staff after the pandemic, and medical-surgical units averaged 6.0 patients per nurse, up from 5.7 before it (Medical Care, Feb 2026).
• 47% of nurses and 32% of physicians at 60 US Magnet hospitals reported high burnout in a 2021 survey of 15,738 nurses and 5,312 physicians (JAMA Health Forum, Jul 2023).
• 87% of those hospital nurses ranked better nurse staffing as the top intervention for their own well-being, ahead of wellness and resilience programs (JAMA Health Forum, Jul 2023).
• 31.5% of US registered nurses who left a job in 2017 named burnout as a reason, and 43.4% of those thinking about leaving said burnout would contribute, in the 2018 National Sample Survey of Registered Nurses weighted to about 3.96 million RNs (JAMA Network Open, Feb 2021).
• Nurses who left because of burnout most often described a stressful work environment (68.6%) and inadequate staffing (63.0%) (JAMA Network Open, Feb 2021).
• Nurses working more than 40 hours a week had 3.28 times the odds of citing burnout as a reason for leaving, compared with those working under 20 hours (JAMA Network Open, Feb 2021).
• Only 42% of the 66 countries that answered had national or sub-national care packages for health workers' mental well-being, while 94% of responding countries reported minimum-wage laws (WHO, May 2025).

How many nurses plan to leave the profession?

Stress and burnout is the most common reason other than retirement.

• 41.5% of nurses planning to leave within five years named stress and burnout as the root cause, ahead of workload, understaffing and pay (NCSBN, Apr 2025).
• More than 138,000 nurses left the US workforce between 2022 and 2024, citing stress, burnout and retirement (NCSBN, Apr 2025).
• About 100,000 RNs left the workforce during the first two years of the pandemic, and another 610,388 said they intended to leave by 2027 (NCSBN, Apr 2023).
• 27.4% of New York and Illinois hospital nurses intended to leave their employer after the pandemic and 32.2% were dissatisfied with their job, the highest levels across three survey waves (Medical Care, Feb 2026).
• Hospital RN turnover was 17.6% in 2025, up 1.2 points, and the RN vacancy rate was 8.6%, across 527 hospitals in 40 states (NSI Nursing Solutions, Mar 2026).
• 22.7% of newly hired hospital RNs left within their first year, and first-year leavers made up 29.0% of all RN separations (NSI Nursing Solutions, Mar 2026).
• Hospitals needed an average of 78 days to recruit an experienced RN, and NSI estimates the current US RN shortage at 158,600 (NSI Nursing Solutions, Mar 2026).
• About 180,800 registered nurse openings are projected each year from 2025 to 2035, against 3,465,400 RN jobs in 2025 (BLS, Aug 2026).
• HRSA projects an 8% shortage of registered nurses in 2028, narrowing to 3% (108,960 full-time equivalents) by 2038, with nonmetro areas short by 24% in 2028 against 5% in metro areas (HRSA, Dec 2025).
• The world was 5.8 million nurses short in 2023 against a stock of 29.8 million, and WHO projects the gap will narrow to 4.1 million by 2030 (WHO, May 2025).

How does healthcare burnout compare with other jobs?

Burnout rose faster among health workers than among other US workers between 2018 and 2022, and within health care nurses report the highest rates of any role.

• 45.6% of US health workers said they felt burned out often or very often in 2022 (CDC MMWR, Nov 2023).
• Outside health care, the share burned out very often barely moved over the same period: from 12.8% to 14.1% among all other workers and from 15.1% to 15.5% among other essential workers (CDC MMWR, Nov 2023).
• Health workers who said they were very likely to look for a new job rose from 11.1% to 16.5% between 2018 and 2022, and 44.2% were at least somewhat likely in 2022 (CDC MMWR, Nov 2023).
• Reports of harassment at work among health workers more than doubled, from 6.4% in 2018 to 13.4% in 2022, and harassed health workers had 5.83 times the odds of burnout (CDC MMWR, Nov 2023).
• Health workers who trusted management had 0.40 times the odds of burnout, and those whose supervisors helped had 0.26 times the odds, compared with those without either (CDC MMWR, Nov 2023).
• Nurses had the highest burnout rate of any health care role in 2020, at 56.0%, ahead of other clinical staff (54.1%), physicians (47.3%) and non-clinical staff (45.6%), in a survey of 43,026 workers at 206 organizations (Journal of General Internal Medicine, Mar 2023).
• 41.0% of nurses in that survey intended to leave their job, against 32.6% of non-clinical staff, 32.1% of other clinical staff and 24.3% of physicians (Journal of General Internal Medicine, Mar 2023).
• Feeling overloaded at work was tied to 2.21 to 2.90 times the risk of burnout across every role studied, from physicians to non-clinical staff (Journal of General Internal Medicine, Mar 2023).

What causes burnout in medicine?

Burnout follows how the job is designed more than who is doing it. Administrative work, prior authorization and a hard-to-use EHR come up as drivers in every major survey, and the odds of burnout rise with each of them.

How much does administrative work drive physician burnout?

A great deal. Prior authorization alone takes more than a full working day of physician and staff time each week.

• Practices complete an average of 40 prior authorizations per physician each week (AMA, May 2026).
• Physicians and their staff spend 13 hours a week completing prior authorizations (AMA, May 2026).
• 40% of physicians have staff who work only on prior authorizations (AMA, May 2026).
• When physicians named the sources of their job stress in 2025, the main themes were ineffective EHR systems, leadership transparency and support, inadequate staffing and excessive administrative tasks (AMA, Apr 2026).
• Academic physicians at Massachusetts General Hospital spent 24% of their working hours on administrative duties, those with more administrative time reported more burnout, and prior authorization, clinical documentation and medication reconciliation were rated the most burdensome tasks (Academic Medicine, Feb 2017).

Does the EHR itself raise the odds of burnout?

Yes. Too little time for documentation, EHR work at home and a poorly usable system each raise the odds of burnout, for nurses as well as physicians.

• About 70% of physicians who use an EHR reported at least one form of health IT-related stress, in a statewide survey of 1,792 Rhode Island physicians (Journal of the American Medical Informatics Association, Dec 2018).
• Physicians who spent too much time on the EHR at home had 1.9 times the odds of burnout, and those who said the EHR adds to the frustration of their day had 2.4 times the odds (Journal of the American Medical Informatics Association, Dec 2018).
• Orthopedic surgeons (86.5%), general internists (86.0%) and family physicians (83.2%) reported the most health IT-related stress, and hospitalists the least (38.9%) (Journal of the American Medical Informatics Association, Dec 2018).
• Each 1-point improvement in a physician's EHR usability score was linked to 3% lower odds of burnout, after adjusting for specialty, hours worked and nights on call (Mayo Clinic Proceedings, Nov 2019).
• Physicians who used computerized physician order entry had 29% higher odds of burnout, in a national survey of 6,375 US physicians (Mayo Clinic Proceedings, Jun 2016).
• Among 10,315 US family physicians, an appropriate amount of EHR time at home was associated with 0.58 times the odds of burnout (JAMA Network Open, Nov 2024).
• Nurses rated their EHRs 57.6 on the System Usability Scale, in the bottom 24% across industries, and each 1-point gain in usability was linked to 2% lower odds of burnout (Journal of the American Medical Informatics Association, Apr 2021).
• Nurses in hospitals with poorer EHR usability had 41% higher odds of burnout than nurses in hospitals with better usability, across 343 hospitals and 12,004 nurses (Medical Care, Jul 2021).

How do work hours, staffing and inbox messages add up?

Every extra hour worked, every extra patient per nurse and every extra patient message raises the odds of burnout, and patient messages rose by more than half after 2020.

• The median US physician worked 50 hours a week in 2021, and each additional weekly hour raised the odds of burnout by 2% (Mayo Clinic Proceedings, Sep 2022).
• 61.3% of US physicians reported adequate control over their workload, and poor control over workload, patient load, team, schedule and areas of accountability were each independently associated with burnout (Annals of Internal Medicine, Nov 2024).
• Clinicians with the highest volume of patient call messages had almost 4 times the odds of burnout of those with the fewest, and no other EHR workload measure was significantly linked to burnout (Journal of the American Medical Informatics Association, Jul 2020).
• Messages from patients rose to 157% of their pre-pandemic level in 2020, and each additional patient message added 2.32 minutes of EHR time per day, across 366 US health systems (Journal of the American Medical Informatics Association, Dec 2021).
• At UCSF Health, the average physician's weekly messages from patients rose from 16.76 to 30.33 between the year before the pandemic and the year after it began (JAMA Internal Medicine, Oct 2023).
• Primary care physicians received 55.5% more patient medical advice requests per 8 hours of scheduled clinic time in 2022-2023 than before the pandemic (Annals of Family Medicine, Jan 2024).
• Nurse burnout was 9 percentage points higher in for-profit hospitals than in nonprofits, and two thirds of that gap (67%) ran through lower nurse staffing, in 2024 data on 941 hospitals and 17,368 bedside nurses (Medical Care, Sep 2026).
• Health workers with childcare stress during the pandemic had 80% higher odds of burnout, in a survey of 58,408 workers at 208 US organizations (JAMA Network Open, Jul 2022).
• Family physicians who described their care team as highly efficient had 0.61 times the odds of burnout, and working with a registered nurse raised the odds of a highly efficient team by 35% (JAMA Network Open, Nov 2024).

How much time do doctors spend on EHR documentation?

Documentation is the part of the job that keeps growing, and more of it now happens after the clinic closes. Most figures below come from EHR audit logs and some from time-and-motion observers, so each bullet keeps its own unit and none are added together.

How many hours a day do physicians spend in the EHR?

In log-data studies, close to half of a primary care physician's working day goes to the EHR, and national Epic data shows the total kept rising after 2020.

• Clerical and administrative tasks such as documentation, order entry, billing and coding took 157 minutes a day (44.2% of family physicians' EHR time), and inbox management took another 85 minutes (23.7%) (Annals of Family Medicine, Sep 2017).
• In a time-and-motion study of 57 physicians in four specialties, 27.0% of the office day went to direct face time with patients and 49.2% to EHR and desk work; even inside the exam room, 37.0% of the time went to the EHR and desk work (Annals of Internal Medicine, Sep 2016).
• Across about 100 million outpatient encounters with 155,000 US physicians at 417 health systems, physicians averaged 16 minutes and 14 seconds of EHR time per encounter, with chart review taking 33%, documentation 24% and ordering 17% (Annals of Internal Medicine, Jan 2020).
• The clinic a physician works in matters: median EHR time per visit ranged from 23.5 to 47.9 minutes across 31 clinics, and having a clinic pharmacy technician was associated with 7.87 fewer EHR minutes per visit (JAMA Network Open, Nov 2023).
• Specialty drives the total: primary care physicians averaged 115.0 minutes of active EHR time a day, against 85.7 minutes for medical specialists and 45.6 minutes for surgeons, across 351 US health systems in 2019 (JAMA Internal Medicine, Mar 2021).
• In national Epic data on 280,712 ambulatory physicians, primary care physicians' active EHR time rose 6.5% after the pandemic began, from a mean of 10.6 to 11.3 hours a week (June 2019 to March 2022); those with more than 40 visits a week went from 14.1 to 15.2 hours (JAMA Internal Medicine, Feb 2025).
• Among 141 academic primary care physicians, EHR time per 8 hours of scheduled clinic rose 28.4 minutes (7.8%) between 2019-2020 and 2022-2023, driven by time in orders (+58.9%) and the inbox (+24.4%) (Annals of Family Medicine, Jan 2024).
• Observers watching 211 primary care visits at three VA sites found that 35% of in-person visit time, 46% of phone visit time and 39% of video visit time went to EHR work (Journal of General Internal Medicine, Jan 2025).
• Across 215,207 US ambulatory physicians on Epic (September 2020 to May 2021), the mean was 27.2 minutes of EHR time per visit, 9.5 minutes of it on notes and 4.4 minutes after hours (Journal of General Internal Medicine, Nov 2022).
• After Medicare's 2021 rules cut required visit documentation, 84% of 87 primary care physicians at UCSF said time spent on the review of systems fell, but audit logs showed median documentation time per visit dropped only 1.16 minutes, from 31.55 to 30.39 minutes (Journal of General Internal Medicine, Feb 2025).

What is pajama time and how much do doctors log?

Pajama time is EHR work done outside scheduled hours: evenings, early mornings and weekends. In an American Board of Family Medicine survey of 9,653 family medicine residents in October 2024, 32.3% reported 3 hours or more of pajama time a night, which was linked to 1.61 times the odds of burnout (Academic Medicine, Mar 2026).

• Measured as EHR time between 5:30 pm and 7:00 am and on weekends, pajama time varied about eightfold by clinic, from a median 1.7 to 13.1 minutes per visit, and each extra year since residency added 0.15 minutes per visit (JAMA Network Open, Nov 2023).
• Family physicians logged 86 minutes (1.4 hours) of EHR work after clinic hours per weekday, a figure that spreads their 51 minutes of weekend EHR use across the working week (Annals of Family Medicine, Sep 2017).
• The 21 physicians who kept after-hours diaries reported 1 to 2 hours of extra work each night, mostly EHR tasks (Annals of Internal Medicine, Sep 2016).
• From 2019-2020 to 2022-2023, primary care physicians' EHR time outside scheduled hours on clinic days rose 8.2%, and time on days with no scheduled patients rose 19.9% (Annals of Family Medicine, Jan 2024).
• Primary care physicians averaged 29.8 minutes of after-hours EHR time a day in 2019, against 26.2 minutes for medical specialists and 16.0 minutes for surgeons (JAMA Internal Medicine, Mar 2021).
• Before switching on an ambient AI scribe, outpatient clinicians at a Philadelphia academic health system averaged 50.6 minutes of after-hours EHR work per workday in a 2024 pilot (JAMA Network Open, Feb 2025).
• In Alberta, Canada, 71 hospital specialists logged roughly 24 to 40 minutes of pajama time a day after a system-wide Epic rollout, and after-hours EHR time kept rising from 2019 to 2022 (Journal of Medical Internet Research, Apr 2026).
• At one US institution, female surgeons spent 35% more time in the EHR than male surgeons and more time on unscheduled days and outside scheduled hours, across 593 attending surgeons in 2022-2023 (American Journal of Surgery, Feb 2026).
• Nurse after-shift charting is barely measured: a scoping review of documentation-burden studies found only 1 study of after-hours EHR work that included nurses (Journal of the American Medical Informatics Association, Jan 2021).

Do nurses and US clinicians document more than peers elsewhere?

US clinicians do: on the same vendor's software they spend more time in the record and receive far more messages than clinicians abroad. Hospital nurses at one US academic health system spent an average 31.11% of each 12-hour shift in EHR flowsheets in February 2023, while nurses in focus groups estimated documentation took 10% to 50% of their shift (JMIR Nursing, Apr 2025).

• US clinicians spent more time on notes (40.7 vs 30.7 minutes a day) and on in-basket messages (12.5 vs 4.80 minutes), and received 33.8 messages a day against 12.8 outside the US (JAMA Internal Medicine, Dec 2020).
• US clinicians logged 26.5 minutes a day in the EHR after hours against 19.5 minutes elsewhere, and 77.5% of their note text came from templates or copy-paste, against 60.8% outside the US (JAMA Internal Medicine, Dec 2020).
• Across 840 nurses at Canada's largest academic mental health hospital, active EHR time per patient per shift averaged 19.3 minutes on inpatient units, 14.8 in emergency and 6.3 in ambulatory care (Applied Clinical Informatics, Dec 2025).
• In a national Epic analysis, the 7.2% of physicians who wrote notes mainly by dictation or voice recognition averaged 4.8 minutes on notes per visit, against 13.1 minutes for those who relied mostly on copy-paste (Journal of General Internal Medicine, Nov 2022).

That last finding is an association, not proof that dictation saves time: physicians who choose medical dictation software or a transcription service may write different notes from those who type. It does show how rarely voice was used for clinical notes in 2020 and 2021: about 7% of physicians.

What does burnout cost health systems and patients?

Burnout costs money through turnover, lost clinical hours and safety events, and peer-reviewed studies have now priced the first two in dollars.

How much does burnout-related turnover cost?

Billions a year nationally and thousands per physician, and every nurse who leaves carries a measurable replacement cost on top.

• At the level of a single organization, the annual cost of burnout-related turnover and reduced hours is about $7,600 per employed physician (Annals of Internal Medicine, May 2019).
• Primary care physician turnover adds about $979 million a year in excess health care spending for public and private payers, and $260 million of that is attributable to burnout-related turnover (Mayo Clinic Proceedings, Feb 2022).
• At two Stanford-affiliated hospitals, physicians who reported burnout had 2.68 times the adjusted odds of leaving within two years, and the recruitment cost of burnout-attributable departures was estimated at $15.5 million to $55.5 million over that period (BMC Health Services Research, Nov 2018).
• Each bedside RN who leaves costs a US hospital an average $60,090 (NSI Nursing Solutions, Mar 2026).
• NSI estimates the average hospital loses $5.19 million a year to RN turnover, and every one-point change in RN turnover costs or saves it $295,000 a year (NSI Nursing Solutions, Mar 2026).

The NSI figures cover all RN turnover, whatever the reason, so they price a departure rather than burnout itself.

How many doctors and nurses plan to leave or cut their hours?

Fewer than at the pandemic peak, but still a large minority, and the most common things physicians say would keep them are less documentation and less EHR work at home.

• Among 37,112 physicians at 160 US organizations, intent to reduce clinical hours in the next year fell from 25.6% in 2022 to 22.5% in 2024, and intent to leave the organization within two years fell from 19.9% to 15.1% (JAMA Network Open, May 2026).
• Of physicians planning to cut their hours, 52.6% said less EHR work outside office hours would keep them at their current workload, and 52.5% named less documentation and less work outside of work (JAMA Network Open, May 2026).
• At the end of 2021, 40.3% of US physicians said they were likely or definitely going to reduce clinical hours in the next 12 months, up from 16.1% in 2011 and 19.8% in 2014 (Mayo Clinic Proceedings, Nov 2023).
• In the same national survey, 25.6% of physicians said they were likely or definitely going to leave their current practice within 24 months in 2021, compared with 18.4% in 2011 and 26.6% in 2014 (Mayo Clinic Proceedings, Nov 2023).
• At Mayo Clinic, each 1-point rise on a 7-point emotional exhaustion scale raised the odds that a physician actually cut their full-time equivalent over the next 24 months by 43%, measured from payroll records (Mayo Clinic Proceedings, Apr 2016).
• In a 2020 survey of 20,665 US health care workers at 124 institutions, burnout more than doubled the odds of intending to leave (odds ratio 2.57) and of intending to cut hours (2.15) (Mayo Clinic Proceedings: Innovations, Quality & Outcomes, Dec 2021).
• Across 25 studies of 32,271 physicians, burnout was linked to more than three times the odds of turnover intention (odds ratio 3.10) (BMJ, Sep 2022).
• In two San Francisco primary care systems, burnout predicted actual clinician turnover (adjusted odds ratio 1.57), and 30% of clinicians were no longer working in the same system two to three years later (Annals of Family Medicine, Jan 2019).
• Attrition from clinical practice for any reason, among 712,395 physicians who treated Medicare patients, rose from 3.5% in 2013 to 4.9% in 2019 (Annals of Internal Medicine, Oct 2025).
• Women physicians who left clinical practice did so at a median age of 49, compared with 64 for men (Journal of General Internal Medicine, Apr 2026).
• Federal projections show a shortage of 141,160 full-time-equivalent physicians in 2038, including 70,610 in primary care, with nonmetro areas facing a 58% shortage against 5% in metro areas (HRSA, Dec 2025).

Does burnout affect patient safety?

The evidence points that way for both physicians and nurses, though most of it is cross-sectional and self-reported, so it shows association rather than proof of cause.

• Physicians with burnout had 2.22 times the odds of reporting a major medical error in the prior three months, in a national survey of 6,695 US physicians in which 10.5% reported such an error (Mayo Clinic Proceedings, Jul 2018).
• Among physicians who reported a major error, 77.6% had burnout symptoms, compared with 51.5% of those who did not (Mayo Clinic Proceedings, Jul 2018).
• In a separate measure from the same BMJ meta-analysis, burnout was linked to more than twice the odds of low professionalism (odds ratio 2.33) and of patient dissatisfaction (2.22) (BMJ, Sep 2022).
• A meta-analysis of 85 studies covering 288,581 nurses in 32 countries tied nurse burnout to weaker safety climate, lower safety grades, more hospital-acquired infections, more patient falls, more medication errors and more missed care, though not to mortality (JAMA Network Open, Nov 2024).
• In Pennsylvania hospitals, nurse burnout was the only factor that stayed significantly linked to urinary tract and surgical site infections, and the authors modeled that cutting burnout by 30% would have meant 6,239 fewer infections and savings of up to $68 million a year (American Journal of Infection Control, Aug 2012).
• A meta-analysis of 82 studies with 210,669 health care providers found burnout consistently linked to lower quality (r = -0.26) and lower safety (r = -0.23) of care, effects the authors describe as small to medium (Journal of General Internal Medicine, Oct 2016).

Do AI scribes and voice documentation reduce burnout?

The first randomized trials say ambient documentation helps, by minutes a day rather than hours. Time savings run to minutes a day, the drop in exhaustion is clearer, and the drafts still contain errors a clinician has to catch. For adoption of AI across hospitals beyond documentation, see our AI in healthcare statistics.

How much documentation time do ambient AI scribes save?

In randomized trials the savings range from no measurable change for one product to a modest daily cut for another, and most studies find no drop in after-hours EHR work.

• Even when given a scribe, physicians in the UCLA trial used it in about a third of visits: DAX in 33.5% of 24,696 visits and Nabla in 29.5% of 23,653 (NEJM AI, Nov 2025).
• Neither scribe in the UCLA trial changed the time physicians spent in the EHR outside scheduled hours or on days with no scheduled patients (NEJM AI, Nov 2025).
• In a 24-week stepped-wedge randomized trial of 66 clinicians at an academic health system in Wisconsin and Illinois, 38% of 71,487 notes were drafted with ambient AI, and time spent on notes fell by 0.36 hours (about 22 minutes) a day (NEJM AI, Nov 2025).
• Across 8,581 clinicians at 5 US academic health systems, adopting an AI scribe was associated with 16.0 fewer minutes of documentation and 13.4 fewer minutes of total EHR time per 8 scheduled patient hours (JAMA, Apr 2026).
• In that 5-site study, scribe adopters delivered 0.49 more visits a week, EHR time outside work hours did not change, and the largest gains went to primary care clinicians, advanced practice clinicians, women and those using the scribe in at least half their visits (JAMA, Apr 2026).
• In a Duke randomized crossover trial of 160 outpatient clinicians comparing two ambient scribes, the better product saved 3.19 more minutes a day in notes than the other, and neither made a meaningful difference to pajama time (Journal of the American Medical Informatics Association, Feb 2026).
• In a pre-post study of 46 clinicians from 17 specialties at a Philadelphia academic health system, ambient scribing was associated with 20.4% less time in notes per appointment and 30.0% less after-hours work per workday (JAMA Network Open, Feb 2025).
• In a Stanford emergency department, ambient AI was used in only 11.2% of 8,740 eligible encounters and by 35 of 92 attendings, but on-shift documentation time was 28% shorter when it was used (2:45 vs 3:50 minutes) (Annals of Emergency Medicine, Feb 2026).
• At UCSF Health, physicians who adopted an AI scribe saw 0.80 more patients a week and billed 1.81 more RVUs a week than non-adopters, worth about $3,044 a year per physician at 2025 Medicare rates, with no change in claim denials (JAMA Network Open, Jan 2026).
• Health systems interviewed by the Peterson Health Technology Institute reported that when an ambient scribe is widely available, typically 20% to 50% of clinicians adopt it (Peterson Health Technology Institute, Mar 2025).
• 28% of physicians in the AMA's 2026 survey used AI to document billing codes, medical charts or visit notes, and 7 in 10 saw it as a way to automate tasks that contribute to burnout (AMA, Mar 2026).

Among the time studies above, only the UCLA, UW Health and Duke trials were randomized. The others are pre-post or cohort designs, and their after-hours findings disagree, so they should not be averaged.

Do AI scribes lower burnout scores?

In the short term and on self-reported measures, yes. Follow-up runs weeks to months, response rates are often low, and professional fulfillment barely moves.

• In the 30-day study across 6 health systems, severe burnout fell from 18.4% to 12.2%, but the study had no control group, 60.3% of enrolled clinicians completed both surveys, the surveys were not anonymous, and the scribe vendor's staff were co-authors (JAMA Network Open, Oct 2025).
• In the 66-clinician randomized trial, the share with high work exhaustion fell from 56.1% at baseline to 35.4% by the end of the study, while professional fulfillment rose by a non-significant 0.14 points on a five-point scale (NEJM AI, Nov 2025).
• In the UCLA trial, both scribes raised total Mini-Z well-being scores by about 2.7 to 2.8 points on a 10 to 50 scale; DAX lowered physician task load by 39.9 points on a 0 to 400 scale, while Nabla's 31.7-point drop was not statistically significant. The authors say these secondary results need confirming in larger trials (NEJM AI, Nov 2025).
• In Duke's 160-clinician crossover trial, both ambient scribes reduced personal and work burnout on the Copenhagen Burnout Inventory, and workflow satisfaction rose by 2.51 and 1.91 points on a 7-point scale for the two products (Journal of the American Medical Informatics Association, Feb 2026).
• In an emergency department crossover study where 18 of 20 enrolled faculty used DAX and Abridge for 3 weeks each, both tools scored a median 73.5 on the System Usability Scale, and the apparent preference for one tool mostly reflected which tool doctors tried first (Applied Clinical Informatics, Mar 2026).
• Across Spain's Quirónsalud hospital network, use of an ambient AI scribe grew from 2.7% to about 31% of outpatient visits in 16 months, passing 2.33 million assisted encounters and 7,147 clinicians who used it at least once (Frontiers in Digital Health, Jul 2026).
• Clinician experience improved in 5 of 7 domains in the Quirónsalud surveys, but consultation length differed by less than half a minute (15.01 vs 14.65 minutes) and survey response rates were 14.7% and 7.8% (Frontiers in Digital Health, Jul 2026).
• On a 48-bed medical-surgical unit at Cedars-Sinai, a six-month nurse-led pilot of an AI voice documentation app with 90 RNs and 33 nursing assistants cut incidental overtime from an average of 97 to 48.5 hours a month, in a single unit with no control group (American Journal of Nursing, Aug 2026).
• In 22 interviews with physicians from a California ambient scribe pilot, all 16 comments on cognitive demand and 10 of 11 on work-life integration were positive, while views on note accuracy and style were largely negative (JAMA Network Open, Mar 2025).

How accurate are speech recognition and AI scribe notes?

Raw speech recognition is far from clean before a person edits it, and ambient scribes still make occasional clinically significant errors.

• Speech recognition drafts of 217 dictated clinical notes contained 7.4 errors per 100 words in 2016 data, falling to 0.4% after medical transcriptionist editing and 0.3% in the physician-signed note (JAMA Network Open, Jul 2018).
• Even after review, 42.4% of physician-signed notes in that sample still contained at least one error, and 6.4% of the errors left in signed notes were clinically significant (JAMA Network Open, Jul 2018).
• A systematic review of 122 studies of speech recognition in clinical documentation from 1990 to 2018 found reported word error rates from 7.4% to 38.7% and error-containing documents from 4.8% to 71% (Journal of the American Medical Informatics Association, Feb 2019).
• In a controlled study of 10 physicians at Brigham and Women's Hospital, dictated notes were longer than typed ones (320.6 vs 180.8 words), carried fewer uncorrected errors (1.5 vs 2.9 per note, most of the typed ones minor misspellings) and scored higher on quality (7.7 vs 6.6), while taking about the same time (International Journal of Medical Informatics, May 2020).
• Physicians in the UCLA trial said both scribes produced at least one clinically significant inaccuracy "occasionally" (2.7 and 2.8 on a five-point scale), most often omissions, and one mild patient-safety event was reported (NEJM AI, Nov 2025).
• Asked whether the scribe "generated notes at least as good as my own", UCLA physicians answered neutral on average (3.4 for DAX, 2.9 for Nabla on a five-point scale), the lowest score of any usability question (NEJM AI, Nov 2025).
• In the 66-clinician randomized trial, reviewers rated ambient AI notes between 3.97 and 4.99 out of 5 across the PDSQI-9 quality domains, and diagnostic billing codes improved with ambient AI use (NEJM AI, Nov 2025).

Methodology and Sources

Every statistic on this page was traced to the organization that produced the data: federal agencies (CDC, HRSA, BLS), professional and regulatory bodies (AMA, NCSBN, WHO), peer-reviewed journals, and organizations that ran their own surveys (Physicians Foundation, NSI Nursing Solutions, Peterson Health Technology Institute). We do not cite statistics roundups, vendor marketing pages or news coverage, and any number whose trail ended at one was dropped. Each link was opened on 2 October 2026 to confirm the page loads, the number appears on it and the date is right; where a publisher blocks automated reads, the number was checked on an archived copy of the same URL or on the paper's NCBI record. Two widely repeated papers are excluded because they were retracted: a 2018 JAMA study of resident burnout by specialty and a 2018 JAMA Internal Medicine meta-analysis of burnout and patient safety. Burnout rates from different instruments are reported separately and never averaged, and observational studies are described as associations. Nothing here is medical advice. The page is refreshed monthly, and fast-moving figures such as AI scribe trials and AI adoption are replaced first.

Glossary

• EHR: electronic health record, the software clinicians use to chart, order tests and message patients.
• Pajama time: EHR work done outside scheduled hours, usually evenings, early mornings and weekends.
• Burnout: a work-related syndrome of emotional exhaustion, depersonalization and a low sense of accomplishment. Studies measure it with different instruments, which is why rates differ.
• MBI: Maslach Burnout Inventory, the most widely used research instrument for burnout.
• Mini-Z: a short burnout and well-being survey used in the AMA's Organizational Biopsy and in several AI scribe studies.
• Copenhagen Burnout Inventory: a burnout questionnaire with personal, work-related and patient-related scales.
• PCP: primary care physician.
• RN: registered nurse. LPN/VN: licensed practical or vocational nurse.
• FTE: full-time equivalent.
• Prior authorization: an insurer's requirement that a clinician get approval before a test, drug or procedure is covered.
• CPOE: computerized physician order entry, the part of the EHR used to place orders.
• SUS: System Usability Scale, a 0 to 100 usability score used across industries.
• In-basket: the EHR inbox for patient messages, results and staff requests.
• Ambient AI scribe: software that listens to a clinical visit and drafts the note for the clinician to review.
• RCT: randomized controlled trial. Pre-post study: a study that compares the same group before and after a change, with no control group.
• RVU: relative value unit, the measure Medicare uses to price physician work.
• PDSQI-9: a nine-item instrument for rating the quality of AI-generated clinical summaries.
• NCSBN: National Council of State Boards of Nursing. HRSA: Health Resources and Services Administration. CDC MMWR: the Centers for Disease Control and Prevention's Morbidity and Mortality Weekly Report. BLS: U.S. Bureau of Labor Statistics. NSI: NSI Nursing Solutions, which runs an annual hospital retention survey. VA: US Department of Veterans Affairs.

Sources

• A Pragmatic Randomized Controlled Trial of Ambient Artificial Intelligence to Improve Health Practitioner Well-Being, NEJM AI
• After-Hours Use of the Electronic Health Record Among Medical and Surgical Specialists in Alberta, Canada, Journal of Medical Internet Research
• AI-Powered Voice Documentation: A Nurse-Led Pilot Program, American Journal of Nursing
• AMA: 2025 AMA prior authorization physician survey
• AMA: Mid-career physicians report higher burnout rates
• AMA: More than 80% of physicians use AI professionally
• AMA: Physician burnout rate continues to decline, falling to nearly 42%
• AMA: These 9 physician specialties report the highest burnout rates
• AMA: What's behind the persistent gender gap in physician burnout
• Ambient Artificial Intelligence Scribes and Physician Financial Productivity, JAMA Network Open
• Arndt et al.: More Tethered to the EHR: EHR Workload Trends Among Academic Primary Care Physicians, 2019-2023, Annals of Family Medicine
• Arndt et al.: Tethered to the EHR, Annals of Family Medicine
• Blackley et al.: Speech recognition for clinical documentation from 1990 to 2018, a systematic review, Journal of the American Medical Informatics Association
• Burnout and Health Care Workforce Turnover, Annals of Family Medicine
• Burnout and Well-Being in Trainees: Findings From a National Survey of US Obstetrics and Gynecology Residents, Journal of Graduate Medical Education
• CDC MMWR: Vital Signs, Health Worker-Perceived Working Conditions and Symptoms of Poor Mental Health, Quality of Worklife Survey, United States, 2018-2022
• Changes in Clinician Time Expenditure and Visit Quantity With Adoption of AI-Powered Scribes: A Multisite Study, JAMA
• Cimiotti et al.: Nurse staffing, burnout, and health care-associated infection, American Journal of Infection Control
• Comparing ambient scribes: a randomized crossover clinical trial, Journal of the American Medical Informatics Association
• COVID-Related Stress and Work Intentions in a Sample of US Health Care Workers, Mayo Clinic Proceedings: Innovations, Quality & Outcomes
• Crossover Evaluation of Two Ambient AI Scribe Tools in the Emergency Department, Applied Clinical Informatics
• Differences in Total and After-hours Electronic Health Record Time Across Ambulatory Specialties, JAMA Internal Medicine
• Duggan et al.: Clinician Experiences With Ambient Scribe Technology to Assist With Documentation Burden and Efficiency, JAMA Network Open
• Electronic Health Record Time Allocation Among Primary Care Clinicians at the Veterans Health Administration, Journal of General Internal Medicine
• Evaluating Nurses' Perceptions of Documentation in the Electronic Health Record, JMIR Nursing
• Gardner et al.: Physician stress and burnout: the impact of health information technology, Journal of the American Medical Informatics Association
• Gender Differences in Surgeon Electronic Health Record Utilization, American Journal of Surgery
• Hamidi et al.: Estimating institutional physician turnover attributable to self-reported burnout and associated financial burden, BMC Health Services Research
• Han et al.: Estimating the Attributable Cost of Physician Burnout in the United States, Annals of Internal Medicine
• Hilliard et al.: Are specific elements of electronic health record use associated with clinician burnout more than others?, Journal of the American Medical Informatics Association
• Hodkinson et al.: Associations of physician burnout with career engagement and quality of patient care, BMJ
• Holmgren et al.: Assessing the impact of the COVID-19 pandemic on clinician ambulatory electronic health record use, Journal of the American Medical Informatics Association
• Holmgren et al.: Assessment of Electronic Health Record Use Between US and Non-US Health Systems, JAMA Internal Medicine
• Holmgren et al.: Changes in Physician Electronic Health Record Use With the Expansion of Telemedicine, JAMA Internal Medicine
• HRSA: Health Workforce Projections
• HRSA: Nurse Workforce Projections, 2023-2038
• Kutney-Lee et al.: Electronic Health Record Usability: Associations With Nurse and Patient Outcomes in Hospitals, Medical Care
• Lasater, McHugh and Aiken: Hospital Nursing Has Not Returned to Pre-COVID Conditions, Medical Care
• Li et al.: Nurse Burnout and Patient Safety, Satisfaction, and Quality of Care, JAMA Network Open
• Lukac et al.: Ambient AI Scribes in Clinical Practice: A Randomized Trial, NEJM AI
• McHugh et al.: Patient Outcomes and Nurse Understaffing in For-Profit Hospitals, Medical Care
• Melnick et al.: The association between perceived electronic health record usability and professional burnout among US nurses, Journal of the American Medical Informatics Association
• Melnick et al.: The Association Between Perceived Electronic Health Record Usability and Professional Burnout Among US Physicians, Mayo Clinic Proceedings
• Moy et al.: Measurement of clinical documentation burden among physicians and nurses using electronic health records, a scoping review, Journal of the American Medical Informatics Association
• NCSBN: NCSBN Research Highlights Small Steps Toward Nursing Workforce Recovery; Burnout and Staffing Challenges Persist
• NCSBN: NCSBN Research Projects Significant Nursing Workforce Shortages and Crisis
• NSI Nursing Solutions: 2026 NSI National Health Care Retention & RN Staffing Report
• Olson et al.: Use of Ambient AI Scribes to Reduce Administrative Burden and Professional Burnout, JAMA Network Open
• Overhage and McCallie: Physician Time Spent Using the Electronic Health Record During Outpatient Encounters, Annals of Internal Medicine
• Pajama time and burnout: the burden of after-hours electronic health record use on family medicine residents, Academic Medicine
• Peterson Health Technology Institute: Adoption of AI in Healthcare Delivery Systems, Early Applications and Impacts
• Physician and Nurse Well-Being and Preferred Interventions to Address Burnout in Hospital Practice, JAMA Health Forum
• Physician Intent to Reduce Hours and Intent to Leave Organizations, JAMA Network Open
• Physician Note Composition Patterns and Time on the EHR Across Specialty Types, Journal of General Internal Medicine
• Physician Perspectives on Ambient AI Scribes, JAMA Network Open
• Physician use of speech recognition versus typing in clinical documentation, International Journal of Medical Informatics
• Physicians Foundation: Examining Physician, Resident and Student Wellbeing and Impact of the Current Healthcare Landscape
• Physicians Foundation: How Practice Conditions Shape Physician Wellbeing
• Physicians Foundation: The State of America's Physicians, 2025 Wellbeing Survey
• Preiksaitis et al.: Ambient Artificial Intelligence Scribe Adoption and Documentation Time in the Emergency Department, Annals of Emergency Medicine
• Primary Care Physician Perceptions of the Impact of CMS E/M Coding Changes and Associations with Changes in EHR Time, Journal of General Internal Medicine
• Rao et al.: The Impact of Administrative Burden on Academic Physicians, Academic Medicine
• Real-world evaluation of an ambient AI scribe in Spanish outpatient care after 2.3 million uses, Frontiers in Digital Health
• Relationships of Educational Debt With Hours Worked and Burnout Symptoms Among Early-Career Family Physicians, Annals of Family Medicine
• Rotenstein et al.: Prevalence of Burnout Among Physicians: A Systematic Review, JAMA
• Rotenstein et al.: Sex Differences in Physician Attrition from Clinical Practice, Journal of General Internal Medicine
• Rotenstein et al.: System-Level Factors and Time Spent on Electronic Health Records by Primary Care Physicians, JAMA Network Open
• Rotenstein et al.: Team and Electronic Health Record Features and Burnout Among Family Physicians, JAMA Network Open
• Rotenstein et al.: Trends in and Predictors of Physician Attrition From Clinical Practice, Annals of Internal Medicine
• Salyers et al.: The Relationship Between Professional Burnout and Quality and Safety in Healthcare: A Meta-Analysis, Journal of General Internal Medicine
• Shah et al.: Prevalence of and Factors Associated With Nurse Burnout in the US, JAMA Network Open
• Shanafelt et al.: Career Plans of US Physicians After the First 2 Years of the COVID-19 Pandemic, Mayo Clinic Proceedings
• Shanafelt et al.: Changes in Burnout and Satisfaction With Work-Life Integration in Physicians and the General US Working Population Between 2011 and 2023, Mayo Clinic Proceedings
• Shanafelt et al.: Changes in Burnout and Satisfaction With Work-Life Integration in Physicians During the First 2 Years of the COVID-19 Pandemic, Mayo Clinic Proceedings
• Shanafelt et al.: Longitudinal Study Evaluating the Association Between Physician Burnout and Changes in Professional Work Effort, Mayo Clinic Proceedings
• Shanafelt et al.: Relationship Between Clerical Burden and Characteristics of the Electronic Environment With Physician Burnout and Professional Satisfaction, Mayo Clinic Proceedings
• Sinsky et al.: Allocation of Physician Time in Ambulatory Practice: A Time and Motion Study in 4 Specialties, Annals of Internal Medicine
• Sinsky et al.: Association of Work Control With Burnout and Career Intentions Among U.S. Physicians, Annals of Internal Medicine
• Harry et al.: Childcare Stress, Burnout, and Intent to Reduce Hours or Leave the Job During the COVID-19 Pandemic Among US Health Care Workers, JAMA Network Open
• Sinsky et al.: Health Care Expenditures Attributable to Primary Care Physician Overall and Burnout-Related Turnover, Mayo Clinic Proceedings
• Tawfik et al.: Physician Burnout, Well-being, and Work Unit Safety Grades in Relationship to Reported Medical Errors, Mayo Clinic Proceedings
• The Association of Work Overload with Burnout and Intent to Leave the Job Across the Healthcare Workforce During COVID-19, Journal of General Internal Medicine
• Trends in Physician Electronic Health Record Time and Message Volume, JAMA Internal Medicine
• U.S. Bureau of Labor Statistics: Occupational Outlook Handbook, Registered Nurses
• Variations in Nursing Documentation Time in a Mental Health Setting, Applied Clinical Informatics
• WHO: State of the World's Nursing 2025
• Zhou et al.: Analysis of Errors in Dictated Clinical Documents Assisted by Speech Recognition Software and Professional Transcriptionists, JAMA Network Open