Shift work and health: what studies show for the heart, diabetes and sleep, and what remains unclear

Hospitals, warehouses, bakeries, power stations and police forces all run through the night, which means millions of people are at work while their bodies expect to be asleep. Most of them know what a run of nights does to the following week. Fewer are told what the research says about years of it.

Two professional bodies, the American Academy of Sleep Medicine and the Sleep Research Society, describe the core problem as misalignment between work schedules and the biological regulation of waking alertness and sleep. Large studies now put numbers on the consequences. A review led by Luciana Torquati at the University of Queensland pooled 21 studies and found shift workers had modestly more heart attacks and strokes than day workers, with the gap opening after about five years on the job. The link to diabetes points the same way. The link to dying earlier is much weaker, and part of the excess can be traced to things that are not fixed features of the work.

Shift work covers more than night shifts

Researchers use the term for any schedule outside a regular daytime pattern: permanent nights, early starts, evenings, or schedules that rotate through all of them. The definitions differ from study to study, which is one reason results are hard to compare. A long-running American study of nurses, for example, counted nurses as rotating night shift workers if they did at least three night shifts a month on top of days and evenings.

Not all shifts strain the body equally. A review in The BMJ by Göran Kecklund and John Axelsson of Stockholm University found that the effect on sleep mainly concerns acute sleep loss in connection with night shifts and early morning shifts.

Heart disease in shift workers: a modest excess that builds over years

The Queensland team gathered studies that followed workers over time and counted heart attacks, strokes and deaths from cardiovascular disease (CVD). The review included 21 studies with a total of 173 010 unique participants. Pooled together, the risk of any CVD event was 17% higher among shift workers than day workers.

Time on shifts was what mattered most. Short stints showed nothing. The authors report that the association seems to appear only after the first five years of exposure, and that from then on there was a 7.1% increase in risk of CVD events for every additional five years of exposure. A worker with 20 years of shifts would, on that estimate, carry roughly a fifth more risk than one who had just passed the five-year mark.

The largest single study gives a sense of scale. Céline Vetter and colleagues at Brigham and Women’s Hospital in Boston analyzed 189 158 initially healthy women followed up over 24 years in two cohorts of American nurses. Among the older group, women with ten or more years of rotating nights developed coronary heart disease (CHD), the kind caused by narrowed arteries around the heart, at a rate of about 6 per 1,000 women each year. Among those who had never worked nights the rate was a little over 4. After adjustment for other risk factors, the long-serving night workers had about 18% higher risk.

That study also offers the most encouraging result in the field. The link was stronger in the first half of the follow-up than in the second, a pattern the authors called suggestive of waning risk after cessation of shift work. In the younger cohort, where work histories were updated every few years, longer time since quitting shift work was associated with decreased CHD risk. The excess appears to recede once the nights stop.

Whether any of this shortens life is less clear. A meta-analysis that pooled deaths from all causes, in which a total of 958,674 cohort participants were included, found no meaningful difference for shift work in general and a 6% higher death rate for night work specifically. Its authors concluded cautiously that night work may be associated with higher risk of all-cause, cardiovascular and cancer mortality.

Shift work, diabetes and weight

The evidence on blood sugar runs parallel. Yong Gan and colleagues at Huazhong University of Science and Technology combined twelve studies with 28 independent reports involving 226 652 participants and found that people who had ever worked shifts were 9% more likely to have diabetes. The figure was higher among men and the rotating shift group, at about 37% for men.

The Stockholm review pulled the whole field together. It is based on 38 meta-analyses and 24 systematic reviews, and it lists links between shift work and accidents, type 2 diabetes, weight gain, coronary heart disease, stroke and cancer, while stressing a caveat that applies to all of them: the original studies showed mixed results.

Two explanations for the health effects of shift work

There are two broad explanations, and they are not rivals.

The first is the body clock. Almost every organ keeps time, expecting food and activity by day and rest by night. Frank Scheer’s laboratory, at the same Boston hospital, tested what happens when behavior and clock are pulled apart. Ten adults (5 female) underwent a 10-day laboratory protocol in which each “day” was stretched to 28 hours, so that within a few days they were eating and sleeping at times when their bodies expected the opposite. When meals and sleep were about 12 hours out of step, blood sugar rose even though the body released more insulin to control it, blood pressure went up, and sleep became much less efficient. In three of the eight volunteers with complete data, blood sugar after meals moved into the range typical of a prediabetic state. These were healthy people, and the change took days.

The second explanation is what shift work does to daily habits. A University of Glasgow team studied 238 661 UK Biobank participants who were in paid employment or self-employed and followed them for about 11 years. Shift workers had 11% more cardiovascular disease and 25% more cardiovascular deaths. The researchers then asked how much of that ran through measurable differences between shift and day workers: smoking, short sleep, poor sleep, body fat and blood sugar. Those factors are known as mediators, meaning the steps through which an effect travels. Mediators collectively explained 52.3% of the associations. About half the excess, in other words, came with things that can in principle be changed, and about half was left unexplained.

Kecklund and Axelsson noticed that the diseases linked to shift work are the same ones linked to chronic short sleep, and proposed that the two share mechanisms. They also said this had not been shown: additional research is needed to determine whether insufficient sleep is a causal pathway for the health effects of shift work.

What sleep medicine guidance says about shift work

The guidance that exists is aimed at employers and regulators more than individuals. In 2021 the two sleep societies published principles for setting shift lengths. Their starting point was that the traditional tools, fixed maximum durations of work shifts and minimum durations of time off, do not do enough. By themselves, such arrangements are not sufficient to curb risks to performance, safety, and health, because a rule about hours takes no account of the time of day at which they fall. The statement sets out a process for weighing those risks and does not prescribe a single schedule.

The Queensland reviewers made a similar point about scale. Shift work is not going away, and they called for policies and initiatives to reduce the cardiovascular risk of the people who do it.

Whether a particular shift pattern suits someone with heart disease, diabetes or a sleep disorder depends on their condition and treatment, which their doctor or occupational health service assesses.

Why shift work studies are hard to interpret

Every study cited here observed workers. None assigned them to schedules. Shift workers differ from day workers in income, type of job and often in smoking and diet, and adjusting for those differences is imperfect.

A second problem cuts the other way. People who cope badly with nights may move to day work when they can, so those who remain on shifts for decades may be an unusually resilient group. That would make long-term shift work look safer than it is.

The studies also disagree with each other more than chance would explain. The Queensland analysis found high variation between its 21 studies, though it also judged that the majority of the studies were ranked low-to-moderate risk of bias. Vetter’s team noted that earlier results on heart disease have been inconsistent and limited by short follow-up. And “shift work” covers a hospital roster, a bakery’s 4 a.m. start and a permanent night watch, which may not carry the same risks.

Nobody has yet shown that a particular schedule design, nap policy or meal pattern lowers the long-term risk. Those are the questions the next studies would need to test.

The risks of shift work are modest, build over years and, in the one large study that checked, eased after the night shifts stopped.