Cold water and immunity, honestly

You've heard the claim. Cold plunges boost your immune system. It shows up in ice bath ads, recovery podcasts, and every comment thread where someone explains they haven't been sick since they started plunging. The claim is so ambient now that almost nobody asks what it's built on.

I sell ice baths. The claim doesn't hold up.

What sits underneath it is more interesting than the myth. There is real data here. It just doesn't say what the marketing says it says.

The 29% that everyone quotes

Nearly every "cold boosts immunity" claim traces back to a single trial.

Buijze and colleagues (2016, PLoS ONE) randomised 3,018 Dutch adults to finish their normal hot shower with 30, 60 or 90 seconds of cold water, or to shower as usual. Thirty consecutive days, then sixty more days of cold showering at their own discretion. Large, pragmatic, competently run. The headline result is real. A 29% reduction in sickness absence from work in the cold shower groups (incident rate ratio 0.71, p = 0.003).

Here's the part that gets left out. The trial measured two things. Sickness absence was one. Illness days was the other.

Illness days did not differ significantly between groups. IRR 0.89, 95% CI 0.77 to 1.01, p = 0.073.

Put those two results side by side and the meaning changes completely. People taking cold showers reported the same amount of illness. They took fewer days off work for it.

That's a behavioural finding. Nothing in the data suggests the cold shower group caught fewer infections, cleared them faster, or mounted a better antibody response. They got sick about as often and went to work anyway.

Two further details should keep you cautious. There was no dose-response — 30 seconds performed the same as 90 (p = 0.98 across the three shower groups), and genuine physiological effects usually scale with dose. Every outcome was also self-reported through web surveys, in a trial where participants obviously knew which group they were in. The authors flag it directly: participants "could not be blinded for the intervention nor for their own outcome assessment."

So the honest statement is this. Cold showering was linked to fewer self-reported sick days in a large group of healthy Dutch volunteers. Everything past that sentence is someone's inference.

What cold actually does to your immune cells

Now the mechanism, because this is where the "boost" language comes from and where it goes wrong.

Get into 14°C water and your sympathetic nervous system fires hard. Šrámek and colleagues (2000, European Journal of Applied Physiology) measured a 530% rise in plasma noradrenaline during an hour of head-out immersion at that temperature, alongside a 350% increase in metabolic rate. Catecholamines at those concentrations do things to white blood cells. They shear marginated cells off vessel walls and push them into circulation.

So yes, your white cell count goes up after a plunge. Brenner and colleagues (1999, Journal of Applied Physiology) found leukocytosis, granulocytosis, and a rise in natural killer cell count and activity following cold exposure. Rose and colleagues (2022, European Journal of Applied Physiology) found CD8 T cells and cytotoxic NK subsets elevated after three minutes at 9°C.

That is cell trafficking. Cells relocating from one compartment to another, measured in the compartment we happen to sample with a needle. No new immune cells are produced, and nothing about the movement demonstrates better defence against a pathogen. The Rose paper calls it "redistribution" and "transient mobilisation," which is the accurate language. Hard exercise does the same thing.

The longer-term picture is smaller than you'd hope. Janský and colleagues (1996, European Journal of Applied Physiology and Occupational Physiology) immersed people in 14°C water three times a week for six weeks. A single immersion did close to nothing. After six weeks there were small increases in monocyte proportion, IL-2 receptor expressing lymphocytes, and plasma TNF-α. Total leucocytes, granulocytes, neutrophils, immunoglobulins and complement components were unchanged. The authors concluded that repeated cold immersion "activated the immune system to a slight extent" and that "the biological significance of the changes observed remains to be elucidated."

Thirty years on, it still hasn't been.

Pool the trials properly and the effect disappears entirely. Cain and colleagues (2025, PLoS ONE), a systematic review and meta-analysis from the University of South Australia covering 11 randomised trials and 3,177 participants, found no significant effect of cold water immersion on immune function immediately after exposure (SMD −0.16) or one hour after (SMD −0.18). What they did find was a significant increase in inflammatory markers immediately post-immersion and at one hour.

The Wim Hof study says the opposite of what you think

This one earns its own correction, because it gets cited constantly as proof that cold trains immunity.

Kox and colleagues (2014, PNAS) trained 12 healthy men for ten days in meditation, cyclic hyperventilation with breath retention, and cold immersion. Those men and 12 untrained controls were then injected with bacterial endotoxin. The trained group produced more of the anti-inflammatory cytokine IL-10 and less TNF-α, IL-6 and IL-8. They reported fewer flu-like symptoms.

That is immune suppression, stated plainly in the paper's own title: "attenuation of the innate immune response in humans." Potentially useful for inflammatory and autoimmune conditions. The opposite of a strengthened defence against pathogens.

Then the same lab took the intervention apart. Zwaag and colleagues (2022, Psychosomatic Medicine) randomised 48 healthy men to cold exposure training alone, the breathing exercise alone, both, or no training, then ran the same endotoxin challenge. Cold exposure on its own did not relevantly modulate the inflammatory response (F(8,37) = 0.60, p = .77). The breathing exercise did (p = .002).

The cold wasn't the active ingredient. The hyperventilation was.

The version that's actually defensible

Strip the immunity claim out and something real remains.

A cold plunge is a controlled, voluntary, high-intensity stressor with a defined start and a defined end. You get in, your breathing rate spikes, and you deliberately slow it back down. You stay for two minutes when everything in you wants out at fifteen seconds. Do that three times a week and you get measurably better at tolerating it. Cold-trained participants in Zwaag's work rated ice water as less painful than untrained participants (2023, Journal of Pain Research).

That's tolerance to a stressor, trained the way you'd train anything else. It's also the most plausible reading of the Buijze result. If cold showering makes you better at functioning while uncomfortable, then "I feel rough but I'm going in" becomes a more likely decision on a Tuesday morning. Fewer sick days, same amount of illness. That explanation fits the data at least as well as any immune story, and it doesn't require a mechanism nobody has demonstrated.

The mood and stress findings sit on firmer ground than the immune ones. Cain's meta-analysis found a significant reduction in stress at 12 hours post-immersion, though not at any earlier timepoint, plus improvements in self-reported sleep quality and quality of life.

What you can and can't say

You can say that cold water immersion produces a large catecholamine response, transiently redistributes circulating immune cells, and has been linked in one large unblinded trial to fewer self-reported sick days.

You can't say it boosts immunity, prevents infection, or helps you fight off illness. Nobody has shown any of that. The best pooled evidence shows no immune effect, and the study most often held up as proof points in the other direction.

The folklore fails in reverse too. Johnson and Eccles (2005, Family Practice) chilled 180 people's feet in cold water; 13 of 90 in the chilled group reported cold symptoms over the following days versus 5 of 90 controls. Small trial, symptom-based, no virological confirmation. Still, cold exposure isn't obviously protective against respiratory illness in either direction.

One safety note. If you have a cardiovascular condition, an arrhythmia, or uncontrolled hypertension, speak to your doctor before starting cold immersion. The cold shock response and the diving reflex fire simultaneously on entry, and that combination can provoke arrhythmias in susceptible people (Shattock and Tipton, 2012, Journal of Physiology).

Plunge for what it does to your head, and for the stress and sleep data, which are reasonable. Plunge because getting into 3°C water at 6am is a repeatable, honest test you can pass before most people are awake. Just don't plunge because someone told you it would stop you getting sick.

Enhanced Human makes a drop-in chiller that holds water at 3°C and turns any bathtub into a cold plunge. Probe goes straight into the water, 900 W, no plumbing and no filter box. We built it for the reasons above — the stress tolerance, the sleep and mood findings, the fact that a cold plunge is one of the few genuinely hard things you can do reliably every morning. Not for your immune system. If a brand tells you their ice bath will stop you getting sick, ask them which trial.

References

All citations below were verified against PubMed prior to publication.

Brenner, I.K., Castellani, J.W., Gabaree, C., Young, A.J., Zamecnik, J., Shephard, R.J., Shek, P.N. (1999). Immune changes in humans during cold exposure: effects of prior heating and exercise. Journal of Applied Physiology, 87(2), 699–710. https://doi.org/10.1152/jappl.1999.87.2.699

Buijze, G.A., Sierevelt, I.N., van der Heijden, B.C.J.M., Dijkgraaf, M.G., Frings-Dresen, M.H.W. (2016). The effect of cold showering on health and work: a randomized controlled trial. PLoS ONE, 11(9), e0161749. https://doi.org/10.1371/journal.pone.0161749

Cain, T., Brinsley, J., Bennett, H., Nelson, M., Maher, C., Singh, B. (2025). Effects of cold-water immersion on health and wellbeing: a systematic review and meta-analysis. PLoS ONE, 20(1), e0317615. https://doi.org/10.1371/journal.pone.0317615

Janský, L., Pospíšilová, D., Honzová, S., Uličný, B., Šrámek, P., Zeman, V., Kamínková, J. (1996). Immune system of cold-exposed and cold-adapted humans. European Journal of Applied Physiology and Occupational Physiology, 72(5–6), 445–450. https://doi.org/10.1007/BF00242274

Johnson, C., Eccles, R. (2005). Acute cooling of the feet and the onset of common cold symptoms. Family Practice, 22(6), 608–613. https://doi.org/10.1093/fampra/cmi072

Kox, M., van Eijk, L.T., Zwaag, J., van den Wildenberg, J., Sweep, F.C.G.J., van der Hoeven, J.G., Pickkers, P. (2014). Voluntary activation of the sympathetic nervous system and attenuation of the innate immune response in humans. Proceedings of the National Academy of Sciences, 111(20), 7379–7384. https://doi.org/10.1073/pnas.1322174111

Rose, C.L., McGuire, H., Graham, K., Siegler, J., Fazekas de St Groth, B., Caillaud, C., Edwards, K.M. (2022). Partial body cryotherapy exposure drives acute redistribution of circulating lymphocytes: preliminary findings. European Journal of Applied Physiology, 123(2), 407–415. https://doi.org/10.1007/s00421-022-05058-3

Shattock, M.J., Tipton, M.J. (2012). 'Autonomic conflict': a different way to die during cold water immersion? The Journal of Physiology, 590(14), 3219–3230. https://doi.org/10.1113/jphysiol.2012.229864

Šrámek, P., Šimečková, M., Janský, L., Šavlíková, J., Vybíral, S. (2000). Human physiological responses to immersion into water of different temperatures. European Journal of Applied Physiology, 81(5), 436–442. https://doi.org/10.1007/s004210050065

Zwaag, J., Naaktgeboren, R., van Herwaarden, A.E., Pickkers, P., Kox, M. (2022). The effects of cold exposure training and a breathing exercise on the inflammatory response in humans: a pilot study. Psychosomatic Medicine, 84(4), 457–467. https://doi.org/10.1097/PSY.0000000000001065

Zwaag, J., Timmerman, H., Pickkers, P., Kox, M. (2023). Modulation of pain sensitivity by a hyperventilatory breathing exercise and cold exposure training. Journal of Pain Research, 16, 1979–1991. https://doi.org/10.2147/JPR.S400408