MySleepMate / Field Notes

Field note

Are long naps bad for you? What those studies can and cannot show

The short answer

The headlines are real and the causal claim behind them is not established. Long daytime naps, generally over an hour, are associated in observational studies with higher rates of high blood pressure, diabetes, heart disease and mortality. One analysis linked each additional hour of daytime napping with roughly 13% higher mortality risk. But these are associations from observational data, and the researchers themselves flag the obvious alternative explanation: ill people nap more. Illness plausibly causes both the long naps and the bad outcomes, which is called reverse causation, and it is why methods like Mendelian randomisation exist. The defensible reading is not that long naps damage you. It is that a new habit of long daily naps can be a signal worth taking to a doctor.

What the studies actually found

There is a genuine and repeated pattern in the observational literature, and I do not want to wave it away.

So the split in the advice is not arbitrary. Wherever you see "keep it under 30 minutes", this literature is part of why.

The problem with all of it

Every one of those findings comes from observational data. Researchers watched large numbers of people, recorded who napped and for how long, and followed what happened. Nobody randomised anyone into a long-nap group for a decade, because you cannot.

Which leaves an alternative explanation that is at least as plausible as the headline one. People who take long daytime naps are more likely to be unwell. Undiagnosed sleep apnoea makes people nap. So do heart failure, depression, chronic pain, thyroid disease, cancer, and the ordinary frailty of advanced age. All of those also raise your risk of the outcomes being measured.

Reverse causation, in one line

The studies show that people who nap for long periods die sooner. They do not show that napping for long periods makes you die sooner. If being ill causes both the napping and the dying, you get the same graph with none of the meaning.

Researchers try to handle this by adjusting for known conditions, but adjustment only works for things you measured. Illness that has not been diagnosed yet is precisely the kind that will not be in the data, and it is precisely the kind that makes people nap. One of the systematic reviews notes that residual confounding, including factors like air pollution and rural versus urban environment, could not be accounted for because the data simply was not available.

Being fair to the researchers

Something I want to be clear about, because it would be easy to write this page as though epidemiologists had been caught out.

They have not. The researchers say this themselves. The papers explicitly discuss reverse causality, note that people taking long naps are more likely to be ill and to carry other risk factors, and describe the possibility that sicker people are more likely to experience the outcome. The caveats are in the literature, stated plainly, by the people who did the work.

What happens is that the caveats do not survive the journey. A paper says "associated with, and we note the possibility of reverse causation". A press release says "linked to". A headline says "long naps raise your risk of death". By the time it reaches you it is a causal claim, and nobody along the chain lied. Each step just dropped a qualifier.

That pattern is the same one I found with the NASA nap percentages and with most of the numbers on this site. The failure is almost never fabrication. It is attrition of nuance.

How you would settle it

You cannot randomise people into a decade of long naps, so the field uses a different tool: Mendelian randomisation.

The idea is to use genetic variants associated with a trait, in this case a tendency to daytime napping, as a stand-in for the trait itself. Because those variants are allocated at conception and are not affected by whether you later develop heart disease, they are not vulnerable to reverse causation in the way an observational measure is. If the genetic proxy still predicts the outcome, that is stronger evidence for a causal path.

Mendelian randomisation studies on napping and cardiovascular disease do exist, which is itself informative: the field considers this an open question worth the effort rather than a settled one. I have not read those papers closely enough to summarise their conclusions, and rather than paraphrase an abstract I am going to say that plainly and leave it as an open item in the research log.

The reading I would defend

That last point inverts the usual advice, and I think it is the more useful version. The popular framing treats long naps as a risk factor to eliminate. The more defensible framing treats them as a symptom to investigate. One of those tells you to set a shorter alarm. The other tells you to see a doctor, which is what the association is more likely pointing at.

I am a medical student rather than a licensed clinician, and this is squarely the territory where a page stops being the right instrument.

Questions people actually ask

Are long naps bad for you?

Observational studies associate naps over about an hour with higher rates of high blood pressure, diabetes, heart disease and mortality. But these are associations, not demonstrated causes, and researchers note that illness plausibly causes both the long napping and the poor outcomes. The evidence does not establish that long naps themselves cause harm.

Is it bad to nap every day?

Not if the naps are short. Naps of 20 to 30 minutes sit on the beneficial side of the research. The concern in the literature is about habitual long napping over years, and even there the causal direction is unresolved.

Why are long naps linked to heart disease?

The link is statistical rather than demonstrated. One leading explanation is reverse causation: conditions such as undiagnosed sleep apnoea, heart failure, depression and chronic illness all make people nap more, and all independently raise cardiovascular risk. That would produce the same association without napping causing anything.

Should I worry if I suddenly need long naps?

A recent change towards needing long daily naps, especially with daytime sleepiness despite enough time in bed, is worth raising with a doctor. That pattern is the part of this literature most likely to be meaningful, and conditions like sleep apnoea are treatable.

How long is a safe nap?

Twenty to thirty minutes is the range consistently associated with benefits and not implicated in the long-nap literature. It is also short enough to avoid waking out of deep sleep, which is a separate reason to keep naps brief.

References

Keep it in the range nobody argues about

Twenty to thirty minutes avoids both the grogginess and the literature. The calculator gives you the alarm time, counting from when you actually fall asleep.

Open the nap calculator