BLS found that 16.4% of US wage and salary workers at their sole or main job usually worked a non-day schedule in 2017-2018. That included 5.9% on evening shifts, 3.6% on night shifts, 2.4% on rotating shifts, 2.6% on irregular schedules, 0.7% on split shifts, and 1.2% on other schedules. This remains the latest comparable official BLS national breakdown by usual shift type.
BLS counted 144.295 million wage and salary workers in its 2017-2018 shift table. Applying the published 3.6% night-shift share gives about 5.2 million workers. That estimate covers people who usually worked nights at their sole or main job, not everyone who occasionally worked overnight.
In the 2025 American Job Quality Study, 27% of US employees had unpredictable schedules, 28% had unstable weekly hours, and 41% had little or no control over their schedules. Gallup based the article findings on 15,968 employees surveyed from January 13 to February 25, 2025.
Night work is associated with higher health and safety risks, but the size of the risk depends on exposure and study design. A 2025 meta-analysis of 23 cohort studies reported a 13% higher relative risk of total cardiovascular events among night-shift workers. IARC classifies night shift work as probably carcinogenic to humans, Group 2A, based on limited human evidence plus animal and mechanistic evidence.
Get practical ideas
Open a ready-made prompt to summarize the article and apply it to your team.
Shift work is not one clean number. The best national US breakdown by usual shift says 16.4% of wage and salary workers worked a non-day schedule, including 3.6% who usually worked nights. Those figures come from BLS data collected in 2017 and 2018, not from a 2026 survey.
A newer 2025 national study measures a different problem: schedule quality. It found that 27% of employees had unpredictable schedules, 28% had unstable weekly hours, and 41% had little or no control over their schedules.
Both datasets matter. They should not be blended into one headline.
Quick shift work statistics
16.4% of US wage and salary workers usually worked a non-day schedule in the 2017-2018 BLS release.
5.9% usually worked evenings, 3.6% nights, 2.4% rotating shifts, and 2.6% irregular schedules.
The BLS night-shift share represents about 5.2 million workers when applied to the release's 144.295 million worker denominator.
31.9% of workers aged 15 to 24 usually worked non-day schedules, compared with 12.3% of workers aged 35 to 44.
36.8% of leisure and hospitality workers usually worked non-day schedules. Transportation and utilities followed at 26.0%, then wholesale and retail trade at 25.4%.
In 2025, 27% of US employees had unpredictable schedules and 28% had unstable weekly hours, according to the American Job Quality Study.
The same 2025 study found that 41% had little or no control over their schedules.
The BLS 2025 absence rate was 3.2% across full-time wage and salary workers. It was 3.8% in health care and social assistance, 3.4% in accommodation and food services, and 2.9% in manufacturing.
Shift Work Data Explorer
Comparable shift prevalence, current schedule quality, 2025 absence rates, and scoped health evidence.
16.4%
usually worked a non-day schedule
BLS wage and salary workers, 2017-2018
Approximate worker counts use the same BLS denominator of 144.295 million. This is the latest comparable official US breakdown by usual shift type.
BLS (2017-2018 shifts; 2025 absence), Gallup American Job Quality Study (2025), NIOSH, Xi et al. (2025), IARC.
What counts as shift work
The numbers change because researchers ask different questions.
Measure
What it counts
What it misses
BLS usual shift
The schedule a wage or salary worker usually worked at their sole or main job
Occasional late closes, second jobs, and some weekend work
Overnight work
Any work performed during a defined overnight window
Whether nights were the worker's usual schedule
Schedule predictability
Whether an employee knew the schedule at least two weeks ahead, unless they controlled it
Whether the work happened during daytime or nighttime
Schedule stability
Whether weekly hours varied by more than 25% over a month, unless the employee chose the variation
The shift type and time of day
Schedule control
Whether the employee could influence hours, days, or time off
Advance notice and weekly-hour volatility on their own
This is why a 3.6% night-shift figure and a 27% schedule-unpredictability figure can both be true. They describe different exposures.
How many US workers usually work non-day schedules
The BLS 2017-2018 Job Flexibilities and Work Schedules release remains the latest comparable official national breakdown by usual shift type. BLS has not published a newer equivalent table, so the collection period belongs next to every figure.
Using the published shares against one denominator avoids mixing rounded category estimates and keeps the table internally consistent.
Industries with the highest non-day schedule shares
The same BLS table shows where non-day work was concentrated. These are industry shares, not current headcounts and not absence rates.
Industry
Non-day schedule share
Evening
Night
Rotating
Irregular
Leisure and hospitality
36.8%
19.3%
2.6%
4.8%
7.1%
Transportation and utilities
26.0%
4.0%
6.3%
4.3%
6.1%
Wholesale and retail trade
25.4%
9.7%
5.1%
5.2%
2.3%
Manufacturing
16.1%
6.1%
5.7%
1.8%
0.8%
Education and health services
14.2%
4.9%
4.3%
1.4%
2.3%
The practical mix differs by sector. Leisure and hospitality carries more evening and irregular work. Transportation has a larger night and rotating component. Manufacturing splits more heavily between evenings and nights. A single "shift-worker rate" hides those operating differences.
The American Job Quality Study gives a more current view of predictability, stability, and employee control.
Gallup surveyed 18,429 US adults aged 18 to 75 who had worked for pay in the previous seven days. The schedule findings in its June 2025 article use the 15,968 respondents who worked for an employer. Surveys ran from January 13 to February 25, 2025, and the reported margin of sampling error was plus or minus 1 percentage point.
Schedule-quality measure
2025 result
Study definition
Unpredictable schedule
27%
Schedule not known at least two weeks ahead, unless the employee had substantial control
Unstable schedule
28%
Weekly hours varied by more than 25% across a month, unless the employee chose the variation
Little or no schedule control
41%
Employee lacked input into at least two of hours, workdays, or short personal time off
Low-quality schedule overall
27%
No schedule control plus a lack of predictability or stability
This is current evidence about how schedules are managed. It does not replace the BLS shift-type table because it does not separate day, evening, night, and rotating work.
2025 absence rates: useful context, not proof of cause
BLS Table 47 now reports 2025 annual averages. The table covers full-time wage and salary workers and separates occupation from industry. It does not say that shift work caused an absence.
The distinction matters. "Healthcare support" is an occupation; "health care and social assistance" is an industry. Mixing those rows produces comparisons that look precise but are not comparable.
Health and safety evidence
The strongest claims concern night work and circadian disruption. They are associations across populations, not a prediction for every individual worker.
Those estimates support a narrow conclusion: working time affects safety risk. They do not support assigning a universal dollar cost per worker or a fixed error count to every shift.
Cancer classification
The International Agency for Research on Cancer classifies night shift work as probably carcinogenic to humans, Group 2A. IARC based that classification on limited evidence in humans, sufficient evidence in experimental animals, and strong mechanistic evidence in experimental animals.
"Probably carcinogenic" is a hazard classification. It does not state the probability that one worker will develop cancer.
What managers can do with the data
The evidence does not produce one perfect rotation. It does point to a useful operating checklist:
Publish schedules early enough for employees to plan around them.
Track how much weekly hours move, not only whether every shift is filled.
Give employees a clear way to state availability and request changes.
Review night work, long shifts, overtime, and short recovery windows together.
Keep schedule changes and worked hours in the same system so the final record matches reality.
If absence is creating repeat overtime, use the overtime reduction guide. If the issue is exhaustion rather than raw coverage, the employee burnout statistics page separates workload, recovery, and retention evidence.
Methodology and source notes
Shift prevalence: BLS Table 7, collected in 2017 and 2018 and released in 2019. Percentages describe the usual shift at a worker's sole or main job.
Worker estimates: each BLS share was applied to the same published denominator of 144.295 million wage and salary workers. Results were rounded to the nearest 100,000.
Schedule quality: Gallup's American Job Quality Study fielded January 13 to February 25, 2025. Its schedule article uses 15,968 employees and a nationally weighted design.
Absence: BLS Table 47, 2025 estimates for full-time wage and salary workers. The figures exclude October, use 11 months of data, and are not strictly comparable with other annual averages. Industry and occupation rows remain separate.
Health: the page uses study-specific effect sizes and identifies whether evidence concerns night work, all shift work, or a safety comparison.
We checked for a newer official BLS table with the same usual-shift categories before retaining the 2017-2018 figures. No newer equivalent release was available. Keeping an older, clearly dated national measure is safer than presenting a newer survey with a different definition as though it were the same statistic.
