Question
Is it true that cold plunges are debunked?
Short answer
Cold-water immersion eased soreness after exercise in small, low-quality trials, and the mood evidence leans towards benefit without establishing one. It can blunt strength gains after lifting, and cold shock is a real risk. For people of African descent, cold plunges are untested.
What this is based on
- Ice baths eased muscle soreness after exercise in small, low-quality trials that did not report ethnicity. No trial has tested them in people of African descent, and cold injury was far more common in African American soldiers in one British army clinic. Cold shock is a real risk.From the card's plain answer
- Who was tested: 366 participants in 17 small trials, mainly athletes and active adults (not selected by ancestry). What was found: Compared with rest, cold-water immersion reduced muscle soreness at 24, 48, 72 and 96 hours (for example standardised mean difference -0.55 at 24 hours). The authors rate study quality as low; the trials varied in temperature, duration and exercise, and most did not report active surveillance of adverse events.Tested Bleakley 2012 Study page
- Who was tested: 21 physically active men (12-week study) and 9 men (muscle biopsy study). What was found: Strength and muscle mass increased more with active recovery than with cold-water immersion, and several markers of muscle growth signalling were lower after cold-water immersion.Tested Roberts 2015 Study page
- Who was tested: 13 studies, 4,105 adults (cold showers, immersion, winter swimming). What was found: Pooled estimates generally favoured cold exposure for depressive symptoms, perceived stress, wellbeing and mood, but confidence intervals were wide, several estimates shrank in controlled-only analyses, and most studies had serious or critical risk of bias. The authors state the evidence does not establish that cold-water exposure improves mood or stress and that it should not be regarded as a standalone psychological treatment. Anxiety showed little consistent benefit.Tested Ngandeu 2026 Study page
- Who was tested: Studies of cold-water immersion for recovery or mood in people of African descent. What was found: My search found no trial of cold-water immersion for recovery or mood in people of African descent, and no study of skin pigmentation and cold-water immersion outcomes. The ClinicalTrials.gov query returned unrelated studies.Tested PubMed 2026
- Untested: ice baths, cold plunges or winter swimming in people of African descent for any outcome; whether skin pigmentation affects cold response; dose (temperature, time) for safety in Black adults; interaction with sickle cell trait; long-term effects. Kox 2014 (PMID 24799686) tested a 10-day meditation, breathing and cold-exposure training in 24 healthy volunteers against a control group and an endotoxin challenge; I read the abstract only and it does not state ethnicity. V2: Burgess 2009 (PMID 19817081) abstract re-read: the paper itself says "African Americans" (objective, results, conclusion), "ethnic origins" in the title, "Gurkha ethnicity"; it does not say how ethnicity was assigned. The card uses the paper's term. RNLI cold-water-shock page returned 403; the Met Office page quoting RNLI and RLSS UK was opened instead.Not tested, and gaps (from the card)
From the card
Ice baths eased muscle soreness after exercise in small, low-quality trials that did not report ethnicity. No trial has tested them in people of African descent, and cold injury was far more common in African American soldiers in one British army clinic. Cold shock is a real risk.
Where this comes from
- Bleakley 2012 (Study page) Open source
- Roberts 2015 (Study page) Open source
- Ngandeu 2026 (Study page) Open source
- Maley 2014 (Study page) Open source
- Burgess 2009 (Study page) Open source
- Barwood 2024 (Study page) Open source
- PubMed 2026 Open source
- Met Office blog, "Cold water shock: what it is and how to stay safe" (quotes RNLI and RLSS UK) Open source
Read the full entry: Ice baths and cold exposure
Last checked against its sources: . Next check due: 16 May 2027.
This is information, not medical advice. If you are worried, see a doctor or pharmacist. How we grade the evidence.