Cold showers and sick days: what a 3,018-person trial found, and what it could not show

**Research review** ## What is known, what is not, and the risks Cold water can put real strain on the heart and breathing, so the risks come first. - **What is known:** in a trial of 3,018 adults, a routine cold finish to an ordinary shower, for 30 days and then optional for 60 more days, cut self-reported sick-leave absence by about 29% compared with controls. The number of illness days did not change significantly. No serious adverse events were judged related to the intervention: 20 serious events were reported, all considered unrelated [S-001]. - **What is not known:** whether cold showers change how often people actually get ill. The trial measured time off work for illness, not illness itself [S-001]. - **Risks:** when healthy people are fully immersed in cold water, heart rate can rise sharply and the heart rhythm can be disturbed. A review argues that these mechanisms may be linked to rare deaths once put down to drowning or hypothermia. That is a hypothesis drawn from the physiology, not a measurement in each case [S-002]. If you have heart, circulatory or blood-pressure problems, talk to a doctor first. ## The hook A video of cold water sent the topic viral. Some people say cold "boots the immune system", and others call it a trend. What did researchers actually measure, and where do the data stop? ## What was measured - **The group.** 3,018 adults aged 18 to 65, with no severe comorbidity and no routine experience of cold showers. They were randomised to four groups: a hot shower that ended with 30, 60 or 90 seconds of cold, or a control group [S-001]. - **Protocol.** 30 consecutive days of a cold finish, then 60 days of cold showers at the participant's own discretion. 79% of participants in the intervention groups completed the 30-day protocol [S-001]. - **Main outcomes.** Illness days, and sickness absence from work for illness [S-001]. ## What was found - Self-reported sickness absence was lower in the cold-shower groups: incidence rate ratio 0.71, about 29% lower [S-001]. - No significant group effect was found for illness days [S-001]. - No serious adverse events were judged related to the intervention; 20 serious events were all considered unrelated [S-001]. ## What this does not show - **Immunity.** The trial did not measure immune markers, so it cannot show that cold "strengthens" the immune system. No immunity conclusion follows from these data. - **Illness days.** The primary outcome of illness days did not change. The reduction is in self-reported time off work, which is not the same as fewer illnesses. - **Cold water in general.** The trial tested a cold finish to an ordinary shower, not ice baths or immersion. Ice-cold immersion carries a different risk (see above), and applying these results to it has not been checked. - **Dopamine, lasting effects, "armour against stress".** These do not appear in the sources we opened. Those claims from the original post are not used. ## What to do next If you want to try it, talk to a doctor first, especially if you have heart, circulatory or blood-pressure problems. This article gives no dosing: the trial protocol is described as study data, not as an instruction. ## Question for readers Have you noticed a difference in how often you get ill, or how many days you miss work? What exactly were you measuring? ## What was verified and what was meant | Claim | Status | Source | |---|---|---| | Randomised trial by Buijze et al., PLoS ONE, 2016 | verified against the full text (PMC5025014, 8 Nov 2026) | S-001 | | 3,018 participants aged 18 to 65, no severe comorbidity, randomised to four groups | verified against the full text | S-001 | | 30 days of a cold finish, then 60 days optional; 79% completed the 30-day protocol | verified against the full text | S-001 | | Self-reported sickness absence 29% lower (rate ratio 0.71) | verified against the full text | S-001 | | No significant effect on illness days | verified against the full text | S-001 | | No serious adverse events judged related to the intervention; 20 serious events, all considered unrelated | verified against the full text | S-001 | | Sharp heart-rate rise and rhythm disturbance in healthy people immersed in cold water | verified against the review abstract; described as a plausible link, not as a measurement in each case | S-002 | | A correction to the trial was published in 2018 | verified against the full text of the correction (PMC6072101) | S-003 | | What the 2018 correction changes: the data underlying Table 2 had been left out of Supporting Information, and were added as file S1. It reports no changed figures for 29%, 0.71 or the 30/60/90-second arms | verified against the full text of the correction | S-003 | | "Dopamine up 200% for 4–6 hours" | not supported; dropped | — | | "Strengthens immunity" and "armour against stress" | not supported; dropped | — | | "15 seconds, then 2–3 minutes" and other durations as advice | not from a source; dropped. The 30/60/90-second arms appear only as trial protocol | S-001 | | Advice on time of day, fasting, rubbing down, a glass of hot water | not supported; dropped | — | | Contraindications "recent flu" and "adrenal problems" | not supported by the trial; dropped. The general advice is to ask a doctor if you have heart or circulatory problems | — | | The historical section (Porfiry Ivanov, Wim Hof) | not checked in this article; Wim Hof is covered in #381 | — | | The author's personal experience | not provided; no block added | — | Author: Serge Luboff. Reviewer: not assigned. Medical review: not done. This is not medical advice and does not replace a consultation with a doctor.