Contrast therapy: what the research actually supports

What it beats, and what it does not

Start with the comparison that gets skipped. Almost every favourable summary of contrast therapy is comparing it to sitting still, and on that comparison it does fine. The 2013 meta-analysis of 18 randomised trials in 356 people found contrast water therapy reduced soreness against passive rest at every timepoint measured, with standardised mean differences from roughly 0.62 under six hours to 0.40 at 72 hours, and improved strength recovery at 24, 48, and 72 hours (Bieuzen et al., 2013).

The same paper is where the story gets complicated, and its authors say so themselves. Compared against other active recovery methods rather than rest, the differences disappear: no significant difference versus cold water immersion at any timepoint for soreness or power, and none versus active recovery, stretching, or compression. Their conclusion is worth quoting rather than paraphrasing. There is "little difference in recovery outcome when CWT is compared to other popular recovery interventions," and "overall the study quality in this review was low," with most studies at high risk of bias.

The Cochrane review reaches the same place from a different direction. Five of its trials compared cold water immersion against contrast immersion directly, and pooled data for pain showed no evidence of a difference between the two at any of four follow-up times (Bleakley et al., 2012). We are not going to put an effect size on that comparison, because the review does not publish pooled estimates for it in its text and we could not retrieve the underlying data tables. "No significant difference at any follow-up time" is the honest ceiling on what we can say.

Two recent network meta-analyses keep the trend going. A 2024 review of 57 studies ranked contrast water therapy best for creatine kinase, a blood marker of muscle damage, while cryotherapy beat it for soreness and for jump performance (Chen et al., 2024). Read that carefully: its best result in a friendly review is on a number in a blood sample, not on anything you would notice. And a 2026 network meta-analysis of 51 trials found contrast therapy's effects on soreness, creatine kinase, IL-6, and CRP were all statistically non-significant immediately after exercise, and had it underperforming cold water immersion alone on most metrics (Wu et al., 2026).

In our view the fair reading is that contrast therapy is a reasonable thing to do and an unremarkable one. It is not the upgrade over a cold plunge that the equipment marketing implies.

Does the order matter? Nobody has tested it

This is the question everyone asks first, and it has no answer in the literature. We looked specifically for it: all 46 papers PubMed returns for the exact phrase "contrast water therapy", the full 29-paper extraction table in the 2021 scoping review, and the protocol tables in both recent network meta-analyses. No study manipulated the sequence. Not hot-first against cold-first, not ending-hot against ending-cold.

Two papers look like they might be the exception and are not. A 2020 study on "timing-sequence recovery" in distance runners varied when after exercise the treatment was applied, at zero, 24, 48, and 72 hours, not the internal order of the hot and cold phases. And a 2025 pilot on repeated hot and cold stimulation studied neck and shoulder stiffness from phone posture in 20 people, against a control, with no order manipulation and no exercise.

Meanwhile the scoping review whose entire job was to map this literature never raises sequence as a variable at all. It identifies the absence of protocol standardisation as the central gap and does not list order among the parameters anyone has varied (Greenhalgh et al., 2021).

So where does "always finish cold" come from? Every version of it we could trace runs back to a commercial source: cold plunge retailers, sauna manufacturers, and recovery studios, several of which sell the equipment being recommended. That does not make it wrong. Ending cold may well be better, and there are plausible mechanisms to argue it. It means the rule is convention, and anyone presenting it as an established finding is telling you something the research has not established.

The version you are actually doing has not been studied

Here is the gap that matters most for anyone reading this on a site about home recovery gear. Contrast therapy research is almost entirely water research. Of the 46 papers under that exact phrase, none uses a sauna. Of the 29 papers in the scoping review, none used a sauna or any dry heat: only three used anything other than water at all, and those were hot packs, hydrocollator packs, and ice massage.

Applying that literature to a sauna-and-plunge routine is an extrapolation, and it is not a safe one, because the two situations differ in ways the studies cannot speak to. Water immersion adds hydrostatic pressure, and the 2013 meta-analysis raises the possibility that some share of any benefit comes from that pressure rather than from the temperature swing. A dry sauna does not reproduce it. Heat moves out of 40 C water and 85 C air at very different rates. And one study in the scoping review found contrast therapy delivered with packs could not produce a significant change in intramuscular temperature a centimetre below the fat layer after twenty minutes, with the authors noting that if the effects depend on tissue temperature change, contrast therapy would need reconsidering as a modality.

We searched for a sauna-to-plunge recovery trial directly and found none. The closest published work measures thermal and cardiovascular strain rather than recovery, including a 2026 study of 16 adults doing two 15-minute 85 C sauna bouts with cooling breaks, with and without a 90-second immersion (Owen et al., 2026). That is a useful paper about heat strain. It is not a recovery trial, and we are not going to present it as one.

Stating a negative search result is not a rhetorical move here, it is the finding. If you have both machines, you are doing something that is physiologically reasonable and empirically unstudied.

A number that is circulating and does not exist

While researching this we kept meeting a specific claim: that a 2022 study in the Journal of Strength and Conditioning Research found roughly 30% faster lactate clearance with a post-sauna cold plunge. It appears across at least five cold plunge and sauna retail sites, usually with no author and no title.

We cannot find it. It does not appear in PubMed, it has no traceable authors, and every instance we followed led to another vendor blog rather than to a journal. We are treating it as fabricated. Worth knowing because search engines have started repeating it back as though it were established, which is how a phantom citation becomes furniture.

If you have the actual paper, we would genuinely like to see it and will correct this section. Until then, do not let that figure into a decision.

What the protocols in the literature actually look like

There is no validated contrast protocol. There is a cloud of protocols that happened to get published, and the spread inside it is wide enough to matter.

The scoping review found a nominal consensus on 10 to 15 C for the cold phase and 38 to 40 C for the hot, but reported actual studies running 8 to 15 C and 35.5 to 45 C, with total durations from 6 to 31 minutes and hot-to-cold ratios anywhere from 1:1 to 10:1. The 2013 meta-analysis found cold phases of one minute in nearly every study, hot phases of one to three minutes, three to seven cycles, and totals of 6 to 24 minutes. Note the range on the hot side: 35.5 C is a warm bath and 45 C will scald some people. Those are not variations on one protocol.

The most defensible single recommendation comes from the 2013 water immersion review, which suggests equal time hot and cold, individual immersions of about a minute, and a total of up to roughly 15 minutes, while explicitly doubting that a dose-response relationship exists at all (Versey et al., 2013). That is one review team's synthesis, not a tested optimum, and it is a narrative review rather than a meta-analysis. We are passing it on with that label attached.

If you want a number to start from, that one is defensible. What is not defensible is the precision you will see elsewhere. Three rounds is not a finding. Ninety seconds is not a finding.

Safety, and the part the safety literature does not cover

The real mechanism of concern with the cold half is autonomic conflict: cold skin drives the heart rate up while facial immersion and breath-holding drive it down, and the two opposing signals arrive together. In healthy young people arrhythmias occur in roughly 2% of free-breathing head-out immersions, but in 62 to 82% when submersion is combined with breath-holding, mostly within ten seconds of releasing the breath, with risk rising below 15 C water (Shattock and Tipton, 2012). The groups that paper names as higher risk are long QT syndrome, coronary artery disease, and hypertension or cardiac hypertrophy.

One honest limitation, and it is a real one for this guide: that paper does not address immersion after heat exposure. The best mechanistic source on cold water risk was not written about the hot-then-cold case, so we are borrowing it, and you should know we are borrowing it.

On the sauna half, the base rate is reassuring. A Finnish forensic series covering 1990 to 2002, in a country where sauna use is close to universal, found fewer than two deaths in sauna per 100,000 inhabitants per year, with alcohol involved in half of all cases (Kenttamies and Karkola, 2008). Alcohol is the modifiable factor that shows up in the data, not duration.

For practical handling the British Sauna Society's guidance is the most directly applicable we found, and it cites Tipton's work: enter the water gradually, especially after heat, keep your face out until your breathing settles, do not breath-hold, and do not do it alone (British Sauna Society).

What the evidence does not support is a claim in either direction about whether alternating is safer or riskier than cold alone. The one direct comparison we found points mildly toward the combination lowering thermal and cardiovascular strain, but it is 16 people measuring acute strain rather than events. And both the Cochrane review and the 2013 meta-analysis note that the recovery trials did not actively monitor predefined adverse events, so the safety record of this practice in research is largely an absence of looking.

So should you do it?

If you own both and you enjoy it, yes. The evidence for harm in healthy people is thin, the evidence of benefit over doing nothing is real if modest, and enjoying a recovery routine is a legitimate reason to keep one.

What we would not do is buy a second machine to unlock contrast specifically. Nothing in the literature supports the combination outperforming the cold half on its own, and the sauna-plus-plunge version has never been tested. If you are choosing between them, our cold plunge and ice bath comparison and the water temperature guide cover what actually moves the needle on the cold side, and the sauna heat guide covers the risk that scales with the season. If you already have both and are planning where to put them, the backyard setup guide handles the space and electrical side.

Sources

• Bieuzen F, Bleakley CM, Costello JT. Contrast water therapy and exercise induced muscle damage: a systematic review and meta-analysis. PLoS One. 2013;8(4):e62356.
• Bleakley C, McDonough S, Gardner E, Baxter GD, Hopkins JT, Davison GW. Cold-water immersion (cryotherapy) for preventing and treating muscle soreness after exercise. Cochrane Database Syst Rev. 2012;2012(2):CD008262.
• Greenhalgh O, Alexander J, Richards J, Selfe J, McCarthy C. The use of contrast therapy in soft tissue injury management and post-exercise recovery: a scoping review. Physical Therapy Reviews. 2021;26(1):64-72. Free accepted manuscript via the University of Central Lancashire repository; the publisher version is paywalled and the paper is not indexed in PubMed. The authors declare no conflicts of interest, and we note that the work was funded through a Knowledge Transfer Partnership involving Swellaway Ltd, which manufactures a contrast therapy device. The review's conclusions are cautious and unfavourable to contrast therapy, so the funding does not appear to have shaped the result, but readers should know it exists.
• Versey NG, Halson SL, Dawson BT. Water immersion recovery for athletes: effect on exercise performance and practical recommendations. Sports Med. 2013;43(11):1101-1130. A narrative review rather than a meta-analysis; its protocol recommendation is the authors' synthesis.
• Chen R, Ma X, Ma X, Cui C. The effects of hydrotherapy and cryotherapy on recovery from acute post-exercise induced muscle damage: a network meta-analysis. BMC Musculoskelet Disord. 2024;25(1):749.
• Wu J, Wang A, Hu H, Zhang H. Impact of different cryotherapy interventions on post-exercise acute delayed-onset muscle soreness, athletic performance, and inflammatory biomarkers: a systematic review and network meta-analysis. Front Sports Act Living. 2026;8:1819396.
• Shadgan B, Pakravan AH, Hoens A, Reid WD. Contrast baths, intramuscular hemodynamics, and oxygenation as monitored by near-infrared spectroscopy. J Athl Train. 2018;53(8):782-787. Frequently cited in support of recovery claims. It is a descriptive laboratory study in 10 people measuring blood volume and oxygenation in one calf muscle, with no exercise, no muscle damage, and no recovery outcome measured. We cite it here only to note what it does not show.
• Shattock MJ, Tipton MJ. "Autonomic conflict": a different way to die during cold water immersion? J Physiol. 2012;590(14):3219-3230. Concerns cold water immersion generally and does not address immersion following heat exposure.
• Kenttamies A, Karkola K. Death in sauna. J Forensic Sci. 2008;53(3):724-729.
• Owen TF, Giles CJ, Leaney SF, Coombs GB, Oliver SJ. Thermal and cardiovascular responses to sauna are attenuated by adding cold water immersion to cooling breaks. Exp Physiol. 2026. Measures thermal and cardiovascular strain in 16 adults, not recovery outcomes.
• British Sauna Society. Sauna and cold water safety.

RecoveryScored is general information, not medical advice. We score what a device measurably delivers and cite the literature in measured language. Consult a clinician before starting red light, cold, sauna, or similar practices, especially if pregnant, photosensitive, on photosensitizing medication, or managing a condition. Follow the manufacturer's instructions and eye-protection guidance.