Night owls were 1.7 times as likely to retire early on a disability pension in a 12-year Finnish study

Some people are at their best at eleven at night and useless at seven in the morning, and most workplaces are arranged for the opposite kind of person. Night owls tend to treat that as an inconvenience: more coffee, a harder alarm, a grudge against early meetings. A long-running Finnish study suggests the mismatch may show up decades later in something more serious, the odds of being pensioned off sick before retirement age.

Tapio Räihä of the University of Oulu and colleagues tracked 5,679 people who were all born in northern Finland in 1966. At 46 each filled in a questionnaire about whether they were a morning or an evening person, and national registers then recorded who was granted a disability pension over the next 12 years. Writing in Occupational & Environmental Medicine in September, the team reports that evening types received such pensions at about 1.7 times the rate of morning types. The same group had seen a hint of this after four years, on only a handful of cases. The longer follow-up has enough to show the pattern in women as well as men.

What makes someone a night owl?

Sleep researchers call the trait chronotype: a person’s natural preference for when to sleep and when to be active. A much-cited paper on the subject describes it as largely regulated by the circadian clock, the body’s internal 24-hour timer, with people ranging from extreme early types to extreme late types with the majority falling between these extremes.

Part of it is inherited. A genetic study of almost 700,000 people found 351 places in the genome tied to being a morning person, and showed that they track real behavior: the mean sleep timing of the 5% of individuals carrying the most morningness alleles is 25 min earlier than in those carrying the fewest. An allele is one version of a gene.

In the Finnish study, chronotype came from a short questionnaire. Participants were asked how easily they woke in the morning, how tired they were during the first half hour, when they preferred to work, and whether they thought of themselves as a morning or evening person. The team then checked those answers against a genetic score for morningness in most of the cohort and found the two agreed. On the questionnaire, the researchers classified 253 men (9.7%) and 381 women (12.4%) as E-types, their shorthand for evening types. Nearly half the cohort were morning types and the rest sat in between.

How many night owls ended up on a disability pension?

A disability pension in Finland is not handed out lightly. According to the journal’s summary of the work, it is only granted in Finland after a period of severe and long term health decline, which makes it a hard marker of someone’s health having failed badly enough to end their working life.

During follow-up, 135 men (5.2%) and 249 women (8.1%) were granted their first DP, the paper’s abbreviation for disability pension. Among evening types the corresponding proportions were 8.3% and 12.1%. Compared directly with morning types, evening types of both sexes were pensioned at roughly 1.7 times the rate.

That figure comes from the plain comparison, before allowing for differences in health or sleep between the groups. The one adjustment the paper does report, for education and type of job, made no difference: adjusting for socioeconomic factors did not attenuate the results. So the gap is not simply a matter of evening types doing harder or lower-paid work.

What might link night owls to disability pensions?

The researchers’ own reading is that health is the bridge. They suggest evening types may have worse health because of the disruption of their internal body clock which can lead to sleep problems, together with less healthy habits.

The body-clock part has a name. Marc Wittmann, Till Roenneberg and colleagues in Munich coined the term social jetlag for the gap between when a person’s body wants to sleep and when work or school allows it. Their study of 501 volunteers found that late chronotypes show the largest differences in sleep timing between work and free days leading to a considerable sleep debt on work days. A night owl on a nine-to-five schedule is short of sleep for most of the week, every week, for decades.

Studies have documented the habits part better than many night owls would like. In the same Munich data, smoking was much more common among late types. In a study of 63,676 nurses aged 45 to 62 years in the United States, those who called themselves definite evening types were 54% more likely to have an unhealthy lifestyle overall, counting diet, exercise, alcohol, weight, smoking and sleep.

Night owls in other cohorts: how much is lifestyle?

Larger studies with harder outcomes help separate the clock from what travels with it.

The UK Biobank supplied the first big test. Kristen Knutson of Northwestern University and Malcolm von Schantz analyzed 433 268 adults aged 38-73 and found that definite evening types had more of almost every illness recorded, most of all psychological disorders, and a 10% higher death rate over six and a half years than definite morning types. The authors called it a small increased risk of all-cause mortality.

A Finnish team then set out to check that result over a far longer span, and reached a cooler conclusion. Christer Hublin and Jaakko Kaprio drew on a national twin cohort surveyed in 1981 and tracked deaths for 37 years. The study population included 23 854 participants. Evening types died at a 9% higher rate, but the excess shrank as the analysis took habits into account, with attenuation mainly due to smoking and alcohol. The telling detail was a subgroup: the authors reported observing no increased mortality among non-smokers who were at most light drinkers. Their verdict was that there is little or no independent contribution of chronotype to mortality.

The American nurses’ study found the same thing for diabetes. Evening types developed it far more often, and accounting for all measured lifestyle and sociodemographic factors resulted in a reduced but still positive association: most of the excess went away, and a small part did not.

The Oulu team’s own earlier work points at sleep. In its four-year analysis, the link between evening type and disability pension in men remained significant except when multiple sleep variables or working times were adjusted for. Put together, these results suggest that a late body clock matters mainly through what it leads to: lost sleep, and the smoking and drinking that are commoner among people who are awake late. That reading is ours, drawn across separate studies, and the new paper does not test it directly.

What the night owl pension figures were not adjusted for

The study watched people; it did not assign them a body clock. As the journal’s summary puts it, this is an observational study, and as such, no firm conclusions can be drawn about cause and effect. Cause could also run backward. Depression, chronic pain and poor sleep can all push bedtimes later, so some people may have been evening types at 46 because their health was already slipping.

Chronotype was measured once. The summary acknowledges that there may have been changes in health and circumstances in the years after the baseline survey which were not accounted for. The reported hazard figures are unadjusted for health and sleep, which are the likeliest explanations. And every participant was born in the same year in the same two provinces, at a latitude where winter mornings are dark until late, so the size of the effect may not carry to other places.

None of that makes the finding trivial for employers or pension systems. “People who are E-type represent a significant proportion of the population and their health and ability to continue working until retirement age pose challenges for sustainable economies,” the researchers said in the journal’s summary. The Munich group argued twenty years ago that work (and school) schedules should be adapted to chronotype whenever possible. Whether doing so would keep more evening types healthy and in work has not been tested.

On the evidence so far, the hazard for night owls lies less in the late clock than in the short sleep and the habits that tend to come with it.