Night Shift Work: What the Evidence Says, Stated Carefully

How many people actually work nights

The best US measurement comes from the Bureau of Labor Statistics' Leave and Job Flexibilities Module of the American Time Use Survey, covering 2017–18 and released in September 2019.

It found that 84% of wage and salary workers worked a regular daytime schedule. The remaining 16% worked a non-daytime schedule: 6% worked evenings and 4% worked nights, with the balance on rotating, split or irregular schedules. Part-time workers were about twice as likely as full-time workers to work a non-daytime schedule — 27% against 14%.

By industry, non-daytime work concentrates where you would expect: leisure and hospitality (37%), transportation and utilities (26%), and wholesale and retail trade (25%).

Internationally, IARC's own summary of the exposure puts it at about 20% of workers in North America, Europe and elsewhere employed outside the standard daytime shift — a broader definition than the BLS "nights" category, and worth not conflating with it.

One thing this data does not tell you is how many of those people are alone while they do it. No agency publishes that cut, and we are not going to estimate one.

What IARC concluded in 2019 — and what it explicitly did not

This is the single most misreported finding in the field, so it is worth quoting the agency rather than the coverage.

In June 2019 an IARC Monographs Working Group of 27 experts from 16 countries classified night shift work as probably carcinogenic to humans, Group 2A. IARC defines night shift work for this purpose as work, including transmeridian air travel, during the regular sleeping hours of the general population.

The evidence basis is specific: limited evidence in humans, with positive associations observed for cancers of the breast, prostate, colon and rectum; sufficient evidence in experimental animals for the carcinogenicity of alteration in the light–dark schedule; and strong mechanistic evidence in experimental systems.

Now the part that gets dropped. IARC states that its classification indicates the strength of the evidence that something causes cancer — it identifies a hazard, not a level of risk. In its own words, it "does not indicate the level of cancer risk associated with exposure at different levels," and agents in the same group can carry very different risks. "Limited evidence" is a defined term meaning a positive association was observed but chance, bias or confounding could not be ruled out. The Working Group did not attempt to estimate what the level of cancer risk might be for different levels of night shift work exposure.

So: night shift work is classified as a probable carcinogen. It is not established that a given person's night shifts will raise their cancer risk by any particular amount, and IARC deliberately declined to say so.

The cardiovascular evidence, including the part that cuts the other way

The largest and most cited synthesis is Vyas and colleagues' systematic review and meta-analysis in the BMJ in 2012, covering 34 observational studies and 2,011,935 people, including 6,598 myocardial infarctions, 17,359 coronary events and 1,854 ischaemic strokes.

Shift work was associated with myocardial infarction (risk ratio 1.23, 95% confidence interval 1.15 to 1.31), with coronary events (1.24, 1.10 to 1.39) and with ischaemic stroke (1.05, 1.01 to 1.09).

And then the finding that rarely makes it into a marketing page: shift work was not associated with increased rates of mortality, whether vascular cause-specific or overall.

The authors also translated the relative risks into population terms, using a Canadian shift-work prevalence of 32.8%: population attributable risks of 7.0% for myocardial infarction, 7.3% for all coronary events and 1.6% for ischaemic stroke. Those are population-level attributions, not personal risk figures, and should not be quoted as "your risk goes up 7%".

Separately, NIOSH's training materials for nurses summarise the mechanism plainly: shift work and long work hours raise health and safety risks by disturbing sleep and circadian rhythms, and can make chronic conditions with a circadian component harder to control.

The drive home is the sharpest documented hazard

The most striking experimental finding in this literature is not about cancer or heart disease. It is about the twenty minutes after the shift ends.

Lee and colleagues, in the Proceedings of the National Academy of Sciences in 2016, put 16 night-shift workers through two two-hour daytime driving sessions on a closed track — one after a normal night's sleep, one after a night shift. After the night shift there were 11 near-crashes, occurring in 6 of the 16 drives, and 7 of the 16 drives were terminated early for safety reasons. After normal sleep, across 16 baseline drives, there were no near-crashes and no early terminations at all. Lane excursions ran at 3.09 per minute post-shift against 1.49 at baseline.

That is a small study on a closed track, and we will not inflate it: 16 participants is 16 participants. But the contrast — 11 near-crashes against zero — is not a subtle effect.

At population scale, NHTSA estimated 633 traffic fatalities involving drowsy drivers in 2023, 1.5% of all traffic deaths, down 9.6% from 700 in 2022. NHTSA itself treats this as an undercount: it describes under-reporting as a predominant issue for drowsy-driving estimates, notes that police-reported crash data likely underestimate drowsiness as a factor, and observes that drivers may be reluctant to admit they dozed off and that hyperarousal after a crash tends to erase the evidence an officer would need.

Folkard and Tucker's 2003 review in Occupational Medicine reached the operational version of the same conclusion: safety declines over successive night shifts, with increasing hours on duty, and with time since the last rest break. Their paper states the direction; the specific percentages often attributed to it are not in the abstract, so we do not quote any.

Working nights and working alone are two problems that usually arrive together

Most people who work nights are not alone, and most people who work alone are not on nights. Where the two overlap — the lone security officer, the community nurse on a night visit, the courier finishing at 2am, the petrol-station attendant — the regulators treat it as a distinct risk, not a sum of two ordinary ones.

HSE's guidance for employers of lone workers names late evening or early morning work, when there are fewer workers around, as one of the key workplace violence risks, alongside carrying money or valuable equipment and holding a position of authority over customers. It also says the thing that no shift pattern can design away: lone working does not automatically imply a higher risk of violence, but it does make workers more vulnerable, because the lack of nearby support means they may be less able to prevent an incident from occurring.

HSE's recommended controls are ordinary and specific. Pre-agreed intervals of regular contact between the lone worker and the employer. Devices that can raise the alarm manually or automatically. And — the one that maps most directly onto a check-in — implementing a robust system to ensure a lone worker has returned to their base or home once their work is completed.

In Australia, the model WHS Regulations require a person conducting a business or undertaking to manage the risks of remote or isolated work, including by providing effective communication with the worker. Remote or isolated work is defined by isolation from the assistance of others because of location, time or the nature of the work — which is to say, working nights can make work "isolated" even in a city.

What a check-in covers, and what it doesn't

Nothing on this page is an argument that an app fixes night work. It does not shorten a shift, restore a circadian rhythm, or keep anyone awake on the drive home. If you are too tired to drive, the answer is not to check in — it is not to drive.

The narrow thing a daily check-in does is close the notice gap. If you finish at 4am, live alone, and something goes wrong between the car park and your front door, the honest question is how many hours pass before anyone wonders. A check-in makes that a matter of hours rather than of whether anyone happens to call.

If you work nights, the practical setup is to put your check-in window after you would normally be home and asleep — not at a fixed morning time that you would miss every rest day. Timezone-aware reminders and a check-in note ("nights this week, home by 5") do more for a night worker than any feature marketed at them.

I'm Alive is a daily check-in app. You confirm you're okay once a day; if you don't, the people you chose are told, with your location. The daily check-in is free (Try It, $0) and works on iPhone and Android anywhere in the world. Automatic alerts to your contacts start on Protect Me ($29.99/year), with location on the move on Protect Me On The Move ($39.99/year); there is a one-time Stay Connected lifetime upgrade at $4.99. Check-first phone fall detection is live on iPhone only (app v2.3.1) on Protect Me and above, stays off until you switch it on, and only runs while the app is open — it is not background detection. It alerts your own trusted contacts; it never calls emergency services, and it is best-effort, not a medical device. There is no Android fall detection and no Apple Watch fall detection. The app interface is currently in English only. It is a consumer app: it is not a certified lone-worker safety system, there is no 24/7 monitoring centre behind it, and using it does not discharge an employer's legal duties.