Burnout statistics 2026: 40 verified numbers, plus 5 famous ones that fail

Between 44% and 67% of American workers are burned out right now, depending on which survey you read and where it sets the bar. The endpoints come from SHRM's 2024 employee survey and the APA's 2024 Work in America survey. Both are live and primary, which matters, because most burnout-statistics pages are copies of copies: numbers quoted from roundups that quoted press releases that no longer exist.

Every statistic below was checked at its primary source on July 3, 2026, and the exact figure was confirmed on the page. The five famous burnout statistics that fail that check get their own section near the end.

One definition anchors everything. The World Health Organization classifies burn-out in ICD-11 as "a syndrome conceptualized as resulting from chronic workplace stress that has not been successfully managed," with three dimensions: exhaustion, cynicism, and reduced professional efficacy. WHO frames it as an occupational phenomenon rather than a medical condition, and every number here follows that framing.

Burnout statistics 2026: the headline numbers

The ten most citable figures. Scope labels follow below; source URLs are in the References list.

How many workers are burned out?

The baseline is Gallup's 2018 study of about 7,500 US full-time employees: 23% felt burned out at work very often or always, another 44% sometimes, roughly two-thirds combined.

SHRM's 2024 survey found 44% of US employees feel burned out, 45% emotionally drained, and 51% used up at the end of the workday. The APA's 2024 survey found 67% experienced at least one outcome often associated with burnout in the past month. Aflac's 2024 WorkForces Report put nearly 3 in 5 American workers at moderate burnout or worse. Microsoft's 2022 Work Trend Index (20,006 knowledge workers, 11 countries) found 48% of employees and 53% of managers already burned out.

The spread is mostly instrumentation: some surveys ask about frequent burnout, others about any symptom inside a window. Our guide to the Maslach Burnout Inventory explains why validated instruments score burnout on three dimensions rather than one yes-or-no.

Burnout by age and gender

The generational gradient is steep. Per Aflac's 2024 report, 66% of millennials (ages 28-43) report moderate-to-high burnout, vs 55% of Gen X and 39% of baby boomers.

The APA's 2024 data shows the same slope:

Loneliness follows the same line: 45% of workers 18-25 feel lonely while working, vs 14% of those 65 and older.

The 2025 LeanIn.Org and McKinsey Women in the Workplace report found 6 in 10 senior-level women frequently feel burned out, vs about half of men at their level and roughly 4 in 10 employees overall. Deloitte's Women @ Work survey (5,000 women, 10 countries, fielded 2021-22) found 61% of women in middle management burned out. Among US teachers, 55% of women report frequent burnout vs 44% of men, per Gallup's 2022 data.

Burnout by profession

Educators hold the top two spots in the US. Per Gallup's 2022 panel study, 44% of K-12 workers always or very often feel burned out, the highest of any industry; among teachers it is 52%, vs 30% of all other workers. College and university workers rank second at 35%.

In the most recent national wave (7,643 physicians, surveyed October 2023 to February 2024), 45.2% of US physicians reported at least one burnout symptom, per Stanford Medicine's summary of the study. After adjusting for age, gender, relationship status, and hours worked, physicians are 82.3% more likely to be experiencing burnout than US workers in other occupations.

The same series reaches back to 2011, the best long-run burnout dataset anywhere:

What actually causes burnout

The most-cited ranking comes from Gallup's 2018 burnout research: unfair treatment at work, unmanageable workload, lack of role clarity, lack of communication and support from the manager, and unreasonable time pressure. Employees who strongly agree they are treated unfairly are 2.3× more likely to experience high burnout. Employees who feel supported by their manager are about 70% less likely to experience burnout regularly, with the same reduction for those with enough time to do their work.

The APA's 2024 survey adds the environment layer: 15% of US workers call their workplace somewhat or very toxic, splitting 30% under low psychological safety vs 3% under high.

Glassdoor's 2025 review-text analysis points the same direction. Reviews mentioning burnout are 26× more likely to also say "unsustainable," and they rate senior management 29% lower: employees blame planning, prioritization, and resourcing decisions rather than busy seasons.

WHO and the ILO tie working 55 or more hours a week to 35% higher stroke risk and 17% higher risk of dying from ischemic heart disease vs 35-40 hours, attributing 745,000 deaths in 2016 to long working hours. That is epidemiology about hours worked rather than a burnout statistic.

What burnout costs

The verified record here is thinner than the internet suggests. The only rigorous burnout-specific dollar estimate covers one profession: physician burnout costs the US about $4.6 billion a year in turnover and reduced clinical hours, roughly $7,600 per employed physician (Han et al., Annals of Internal Medicine, 2019).

Behavioral costs are better documented. Per Gallup's 2018 tracking, employees burned out very often or always are 63% more likely to take a sick day, 2.6× as likely to be actively seeking a different job, and 23% more likely to visit the ER. SHRM's 2024 data puts the retention link at about 3×: 45% of burned-out employees are actively job hunting, vs 16% of everyone else. On Glassdoor, reviews mentioning burnout rate the employer 2.68 out of 5 vs 3.61 without, and those reviewers are 59% more likely to apply for a new job shortly after posting.

Two big figures need correct labels every time they appear. Gallup's estimate that low engagement cost the world economy about $10 trillion in 2024 (9% of global GDP) is an engagement figure rather than a burnout figure. The Goh, Pfeffer, and Zenios study (Management Science, 2016) ties ten workplace stressors to about 120,000 excess US deaths a year and 5-8% of national healthcare spending ($125-190 billion); that is a workplace-stressors figure rather than a burnout figure. Both reappear below.

How long burnout recovery takes

The honest headline: no validated average recovery timeline exists. A 2023 systematic review of return-to-work interventions for burnout (8 studies) concluded that heterogeneity plus moderate-to-high risk of bias "precludes us from drawing firm conclusions." Most "recovery takes X months" claims online have no primary source behind them.

What the data does support: help aimed at the workplace beats help aimed only at the person. In a Swedish controlled study of workers on sick leave for burnout, 89% of the group that received a structured workplace dialogue meeting were back at work at 18 months, vs 73% of controls. The 2023 review found workplace-directed intervention was the only kind that significantly improved return to work; among workers 45 and under, the effect held at 30 months, 88.6% vs 69.7%. A 2017 review in Occupational Medicine adds the predictors: sick leave beyond 6 months predicts worse odds of coming back, while unimpaired sleep and part-time sick leave predict better ones.

Our main guide to burnout recovery covers the interventions, and how long burnout recovery takes walks through the timeline question.

Is burnout rising? 2022 to 2026

Most indicators point up. Glassdoor reviews mentioning burnout rose 32% year over year as of Q1 2025, the highest since that dataset began in 2016 and 50% above pre-pandemic Q4 2019. Aflac found US workers reporting high stress rose to 38% in 2024 from 33% in 2023.

The APA's 2025 survey added a job-insecurity layer: 54% of US workers say job insecurity significantly affects their stress, and 44% worry a downturn could cost them their job within 12 months, up from 36% in 2024.

Globally, Gallup's 2025 wave found 20% of employees engaged and 40% who felt a lot of stress the previous day, with the squeeze at the manager level: manager engagement (22%) barely exceeds individual contributors (19%).

Physicians are the exception: down from the 62.8% pandemic peak in 2021 to 45.2% in 2023, though still above most pre-pandemic waves.

5 burnout statistics you should stop citing

Each of these circulates in burnout listicles right now, fails primary-source verification or carries the wrong label, and was checked on July 3, 2026.

1. "Workplace stress costs US employers $300 billion a year"

Usually credited to the American Institute of Stress. AIS's own workplace-stress page no longer contains the figure; no methodology was ever published, and the number traces to 1990s-era extrapolations. Cite instead: Goh, Pfeffer, and Zenios (2016), workplace stressors linked to 5-8% of US healthcare spending and about 120,000 deaths a year, labeled as stressors rather than burnout.

2. "Burnout costs the global economy $1 trillion a year"

Usually credited to WHO. WHO's actual wording, on its mental-health-at-work fact sheet, is that 12 billion working days are lost every year to depression and anxiety at a cost of US$1 trillion in lost productivity. That is a depression-and-anxiety figure rather than a burnout figure; the misattribution is so pervasive that stress.org's own workplace page repeats it. Cite instead: WHO's exact wording, or a correctly labeled cost figure from the section above.

3. "77% of professionals have experienced burnout at their current job"

Usually presented as current Deloitte data. The survey is real but was fielded in 2015, on 1,000 US professionals, and the original source page now redirects to Deloitte's generic About page. Cite instead: SHRM 2024 (44%) or APA 2024 (67%), both fielded in 2024 and both still live.

4. "Burnout costs the US $125-190 billion in healthcare spending"

Usually credited to "Harvard/Stanford research on burnout." The paper measured ten workplace stressors, including unemployment, lack of insurance, and shift work, rather than burnout. Cite instead: the same paper with its true label, or Han et al.'s $4.6 billion for burnout-specific cost among physicians.

5. "95% of HR leaders say burnout is sabotaging workforce retention"

Usually credited to Kronos and Future Workplace (2017). The original press release no longer exists; kronos.com now redirects to UKG's homepage after the 2020 merger, and the survey instrument was never independently documented. Cite instead: SHRM 2024's verifiable retention link, 45% vs 16% actively job hunting.

Two more to verify before citing

Gallup's "76% experience burnout at least sometimes" (2020) is likely real, but Gallup's live page no longer renders the figure; until re-verified, the 2018 split of 23% and 44% is the safe citation. McKinsey's "about 1 in 4 employees report burnout symptoms" (2022) comes from a real 15-country survey, but McKinsey's primary page was unreachable throughout this research, so it stays off the verified list.

Methodology and how to cite this page

Every statistic above was fetched at its primary source on July 3, 2026: the source organization's own domain, or the lead institution's release for paywalled journals. Verification meant the page loads and contains the exact figure. Two well-known statistics that could not be verified that day were excluded rather than quoted from copies. Figures with a scope wider than burnout carry their label inline every time.

To cite this page: Smartonic, "Burnout Statistics 2026," verified at primary sources July 3, 2026, smartonic.co/blog/burnout-statistics-2026. Before citing any burnout statistic, click through to its primary source, and check whether it sits in the section above. All figures get re-verified annually; the verification date changes only when the check has been re-run.