Does poor sleep make you catch colds?
Every autumn the advice returns: sleep well or you will get ill. The obvious objection is that being ill makes you sleep badly, so the association could run entirely backwards. One study closed that loophole in the only way it can be closed — measure people's sleep first, then give them the virus — and the answer came back clearly.
On this page
Why most sleep-and-immunity studies cannot settle this
The usual design asks people how they slept and how often they were ill, and finds that poor sleepers get sick more. The trouble is that the arrow could point either way, and probably points both. An infection you have not noticed yet raises your breathing rate, your heart rate and your temperature, and fragments your night — so a week of bad sleep can be the first symptom rather than the cause. Self-reported sleep makes it worse still, because feeling ill changes how you remember the week.
The study that fixed the direction
A 2015 study took a different route. Participants wore actigraphy for a week so their sleep was measured rather than remembered. Then they were quarantined and given a rhinovirus — the common cold — through nasal drops, and monitored to see who actually developed an infection. Because the exposure came after the sleep measurement and was identical for everyone, the direction of the arrow is not in question: shorter sleep beforehand predicted more colds afterwards.
That design is rare and expensive, which is why this one study carries so much of the argument. It is also why the finding survives the usual objections: nobody was ill when their sleep was recorded, and nobody chose their own dose of virus.
It is duration doing the work, not effort
The variable that predicted infection was how long people slept, not how virtuous their routine was. That fits what the sleep-restriction literature shows elsewhere: cutting sleep to six hours for a fortnight produces deficits comparable to two nights of total deprivation, and — the part that matters here — people in that state consistently underestimate how impaired they are. You do not feel your way to knowing you are short. You count the hours.
What your ring adds
Two things, in different directions. Looking backwards, it gives you the duration number honestly — the one you would otherwise estimate, badly, in exactly the state where estimates fail. Looking forwards, it often sees an infection arriving before you do: respiratory rate is unusually stable night to night, so a clear multi-night rise above your own normal stands out, and it tends to move alongside a raised resting heart rate and a temperature deviation a day or two before symptoms.
That combination is the useful one. A short week followed by a breathing rate creeping up is a much better reason to move a hard session than either signal alone.
What to actually do with this
Nothing exotic. The lever with the evidence behind it is total sleep time in the days before an exposure you can anticipate — travel, a conference, the first week of term, a small child returning to nursery. Protecting seven hours across that week is a more defensible intervention than any supplement marketed for immunity, and it is the one thing in this article that a trial actually tested.
In Vitra, mark those weeks as a period and compare total sleep, resting heart rate and respiratory rate against your own baseline — computed on your own machine, with your health data staying there. This is general information, not medical advice: if you are unwell, or your resting numbers stay elevated for more than a few days, talk to a clinician rather than to a chart.
Frequently asked questions
- Prather et al. (2015), Sleep: participants had their sleep measured by actigraphy, were then deliberately exposed to a cold virus under quarantine, and the shorter sleepers caught more colds
- Van Dongen et al. (2003), Sleep: six hours a night for two weeks produced cognitive deficits equivalent to up to two nights of total sleep deprivation — and participants underestimated their own impairment
- Oura: respiratory rate is estimated at rest from minute-by-minute changes in nighttime heart rate, and averages 12–20 breaths per minute in healthy adults
Pedro Thomaz builds Vitra, a desktop app that reads Oura data against your own baseline instead of a population average. He has worn a ring daily for years and reads these same numbers every morning — which is where most of what is written here comes from. Vitra is not a medical device and nothing on this blog is medical advice.
Local AI on your Mac or PC. One-time purchase, 7-day trial, no Vitra subscription.
Download Vitra →