Cold Showers and Ice Baths: A Dutch Trial Found Fewer Sick Days, a New Review Found Inflammation. Both Are Right

Cold water has become the most visible wellness ritual of the decade. Ice baths on balconies, plunge tubs in gyms, cold showers prescribed by every podcast host who owns a wetsuit. The claims range from the modest, that it wakes you up, to the sweeping: immunity, fat loss, depression, longevity. The evidence is smaller than the claims and more interesting than the sceptics allow.

The Dutch shower trial

The largest experiment is a randomised trial from the Netherlands published in PLOS One in 2016. Geert Buijze and colleagues recruited 3,018 adults with no habit of cold showering and randomised them to finish their daily shower with 30, 60 or 90 seconds of cold water for thirty consecutive days, or to carry on as usual. The primary outcome was days off work sick.

The cold shower groups reported a 29 per cent reduction in sickness absence compared with the control group. The length of the cold exposure made no difference; thirty seconds did as well as ninety. There was, however, no difference in the number of days participants felt ill. They were ill just as often; they took fewer days off. The authors suggested the cold shower might reduce the severity of symptoms, or might simply change how people respond to feeling unwell. Both readings fit the data. Four in five participants in the cold groups completed the full thirty days, which is a high figure for any health intervention and worth noting in itself.

The 2025 review

In January 2025 a team at the University of South Australia published a systematic review and meta-analysis in PLOS One of randomised trials of cold water immersion in healthy adults: eleven studies, 3,177 participants, water at 7 to 15 degrees, exposures from thirty seconds to two hours, mostly in baths.

The findings were mixed in a way that the enthusiasts and the debunkers have each quoted selectively. Inflammation markers rose significantly immediately after immersion and an hour later, which is the opposite of the anti-inflammatory claim and is what one would expect from a physical stress. Stress scores fell significantly at twelve hours after immersion, but not immediately, not at one hour and not at twenty-four. The pooled analysis found no significant effect on immune function. The narrative synthesis, where the trials were too different to pool, pointed to improvements in sleep quality and quality of life and none in mood, and leaned on the Dutch sick-day result for its longer-term claim. The authors were careful: the effects appear to be time dependent, the trials are few and small, and most involve people who are already fit.