How Long Should You Use a Cold Compress for a Fever?

Most clinical guidance recommends applying a cold compress for roughly 10 to 20 minutes at a time, then removing it for at least the same duration before reapplying. But that number is less settled than it sounds. A scoping review of calf-compress studies found that cooling temperatures, application durations, and even the definition of fever itself vary widely across the research, with no single protocol emerging as standard.¹ The honest answer involves understanding when a cold compress actually helps, when it backfires, and how to use one without making the person more miserable.

Why a Cold Compress Can Work Against You

When you have a fever, your brain has deliberately raised its temperature set point as part of an immune response. Your body is actively trying to maintain that elevated temperature. So when you slap something cold on your skin, two things happen almost immediately. First, the blood vessels near the surface constrict. This is a reflex designed to conserve heat. Sympathetic nerves trigger a rapid decrease in skin blood flow, boosting the skin’s insulating capacity and reducing heat loss from the body.² In practical terms, your body fights back against the cold compress by locking heat inside.

Second, if the cooling is aggressive enough or lasts too long, shivering kicks in. Shivering is your body’s emergency furnace. In patients with systemic inflammation who were surface-cooled, oxygen consumption jumped by about 58% and blood pressure rose roughly 15% during the cooling process.³ That metabolic surge can actually raise core temperature, the opposite of what you wanted. Physical cooling may increase both discomfort and metabolic stress in someone with a true fever.⁴

This is why duration and intensity matter so much. A brief, moderate application can draw some heat away before the body’s defensive reflexes fully kick in. A prolonged or ice-cold application gives vasoconstriction and shivering enough time to overwhelm the cooling effect. The 10-to-20-minute window is essentially a compromise: long enough to provide some comfort and modest temperature reduction, short enough to avoid triggering a full counterattack from the thermoregulatory system.

Where You Place It Matters More Than You Think

The default instinct is to put a cold cloth on the forehead. It feels soothing, and it has generations of tradition behind it. But the forehead is not where you get the best heat transfer. The traditional advice to target areas with major blood vessels near the surface, like the neck, armpits, and groin, makes physiological sense because blood flowing close to the skin can carry heat away from the core.

There is a more surprising option. A randomized trial compared standard cold-pack placement (neck, armpits, and groin) to packs placed on “glabrous” skin, which includes the palms of the hands, soles of the feet, and cheeks. These hairless areas have specialized blood vessels that are particularly efficient at heat exchange. The glabrous-skin approach cooled core temperature at about 0.30°C per 10 minutes, compared to 0.17°C per 10 minutes for traditional placement, a statistically significant difference.⁵ That study was done in people with exercise-induced hyperthermia rather than infection-driven fever, so the results do not translate perfectly, but the anatomy of heat exchange is the same. If you are going to use a cold compress, holding cool cloths against the palms or placing them on the cheeks and soles of the feet could transfer heat faster than the classic forehead approach.

Cold Packs vs. Tepid Sponging

A cold compress is not the only physical cooling option, and it is not always the best one. Methods that use convection and evaporation, such as tepid-water sponging, tend to be more effective than conduction-based methods like ice packs for managing hyperthermia.⁶ A sponge dampened with lukewarm water and drawn over the skin encourages evaporation, which pulls heat away continuously rather than in a localized spot.

A randomized trial in ICU patients compared sponge cooling to ice packs over three hours. Both methods reduced temperature to similar levels (mean temperatures of about 37.7°C for sponging and 37.8°C for ice packs, with no significant difference between them). But the sponge-cooling group reported fewer chills and less discomfort.⁷ The practical takeaway is that if you have the option of gently sponging with lukewarm water instead of holding an ice pack against the skin, the sponge route may be equally effective and more comfortable.

The key word there is “lukewarm.” Cold water or ice-water sponging is more likely to trigger the vasoconstriction and shivering problems described earlier. Tepid water, just slightly below body temperature, is enough to promote evaporative cooling without shocking the system.

Combining Physical Cooling With Fever Medication

Using a cold compress or tepid sponge alongside acetaminophen or ibuprofen tends to produce a faster temperature drop than either approach alone. A study of 73 children with acute febrile illnesses found the greatest temperature reduction in the group that received both acetaminophen and sponging, compared with either sponging alone or acetaminophen alone.⁸ This makes intuitive sense: the medication tells the brain to lower the thermostat set point, while the compress or sponge helps move heat out of the body. When the brain is no longer defending the elevated temperature, physical cooling works with the body instead of against it.

Without medication, the body perceives external cooling as a threat to its new set point and responds by generating more heat. With medication on board, the set point has already started drifting down, so removing heat no longer triggers the same shivering and vasoconstriction. If you are going to use a cold compress for a fever (as opposed to for comfort alone), giving an appropriate dose of acetaminophen or ibuprofen about 30 minutes beforehand is a practical way to get better results with less discomfort.

How This Changes for Children

Children are where the cold-compress question gets the most emotionally charged. Parents watching a feverish child often feel compelled to do something physical right away. A survey of caregivers presenting to emergency departments found that about three-quarters had what researchers classified as “fever phobia,” with roughly half reporting a high level of concern about fever and about 62% holding at least one incorrect belief about what fever could do to their child. Among these caregivers, about 66% reported sponging their child with water as a fever management practice.⁹

The evidence does not strongly support physical cooling as a go-to strategy for children with ordinary fevers. A Cochrane systematic review of physical methods for treating fever in children found that mild adverse events like shivering and goose pimples were about five times more common when physical cooling methods were used, compared with medication alone.¹⁰ Febrile children treated with tepid-water sponging plus antipyretic drugs were more uncomfortable than children treated with antipyretics alone, even though they showed slightly faster temperature drops.⁶

An integrative review of nonpharmacological fever management in children noted that where physical cooling shows any benefit, the effectiveness tends to be limited to the first 30 minutes after application.¹¹ After that initial window, the temperature-lowering effect flattens out while the discomfort continues. For a child who is already miserable with a fever, adding shivering and goose pimples on top of that misery is a tough sell when the medication alone would have gotten the temperature down nearly as fast.

That does not mean cold compresses are never appropriate for kids. If a child is very hot and finds a cool cloth soothing, a brief application on the forehead or the back of the neck is unlikely to cause harm. The problem arises when parents pursue aggressive or prolonged cooling, especially ice water or alcohol-based sponging (which carries toxicity risks and should never be used), out of fear that the fever itself is dangerous. Most pediatric fevers below about 40°C are uncomfortable but not harmful, and the primary goal of treatment is comfort, not hitting a particular number on the thermometer.

Protecting the Skin From Cold Injury

It sounds improbable that an ice pack could cause a burn, but it happens more often than people realize. A case report described a 59-year-old woman who suffered a superficial partial-thickness burn covering about 1% of her body surface area after applying an ice pack to her calf.¹² The injury looked and behaved like a burn from heat, with blistering and tissue damage, because the mechanism is similar: extreme temperature causes cell death regardless of which direction the extreme goes.

Ice pack burns are most likely when the pack is applied directly against bare skin without a cloth barrier, when it is left on too long, or when the person falls asleep with the pack in place. For fever management specifically, a few precautions help:

• Use a barrier: Wrap the cold pack in a thin towel or pillowcase. Direct skin contact with ice or a frozen gel pack concentrates the cold too aggressively.
• Set a timer: If you are drowsy from the fever, it is easy to lose track of time. A 15-minute timer prevents extended exposure.
• Check the skin: If the area under the compress turns white, bright red, or numb, remove the compress immediately. Numbness is the dangerous one because it means you have lost the warning signal that tells you the tissue is too cold.
• Avoid gel packs straight from the freezer: Reusable gel packs can reach temperatures well below 0°C. Letting them sit at room temperature for a few minutes before applying brings them into a safer range.

The skin-safety concern is another reason the 10-to-20-minute guideline exists. Even with a cloth barrier, sustained cold application can damage skin, especially in children, older adults, or anyone with compromised circulation.

Fever vs. Hyperthermia and When Cooling Is Actually Urgent

Everything discussed so far applies to fever, which is the body’s deliberate temperature increase in response to infection or inflammation. Hyperthermia is a different situation. In hyperthermia, such as heat stroke or heat exhaustion, the body’s cooling system has been overwhelmed and core temperature is rising uncontrollably. The brain has not raised the thermostat; the thermostat is failing.

Physical cooling is the first-line treatment for hyperthermia, and the rules about duration change dramatically. In heat stroke, aggressive and continuous cooling (cold water immersion if possible, ice packs to the neck, armpits, and groin as a fallback) should continue until core temperature drops below about 39°C. This is not a 10-minute-on, 10-minute-off scenario. The vasoconstriction and shivering concerns still apply, but the risk of organ damage from sustained high temperature far outweighs them. Physical cooling is important when the body’s thermoregulatory mechanisms are overwhelmed.⁴

The distinction matters for practical decision-making. A person who developed a 39°C temperature overnight with body aches and a sore throat has a fever. A brief cold compress for comfort is reasonable. A person who collapsed after running a race in 35°C weather and is confused and flushed has hyperthermia. Aggressive, continuous cooling is potentially lifesaving, and the gentle 15-minute compress approach would be inadequate.

What “Fever Phobia” Gets Wrong

A striking finding in the caregiver survey mentioned earlier is that about 61% of parents held at least one incorrect belief about what fever can do, and roughly 81% reported always giving medication for any fever.⁹ The combination of fear and misinformation drives a lot of unnecessary cold-compress use. Parents who believe that a fever of 39°C can cause brain damage (it cannot, under normal circumstances) are understandably tempted to throw everything they have at it: medication, sponging, cold compresses, and repeat temperature checks every 15 minutes.

The reality is that for a typical infection-driven fever in an otherwise healthy child or adult, the fever itself is doing useful work. It makes the immune system more effective and the environment less hospitable for pathogens. Bringing the temperature down is about comfort, not about preventing some catastrophic outcome. That reframing changes how you think about cold compresses entirely. If the compress makes you feel better, use it briefly. If it makes you shiver, feel worse, or just adds stress to an already unpleasant situation, skip it. The fever will resolve on its own as the infection clears, with or without a washcloth on your forehead.

Practical Timing for Different Situations

Since no single protocol has been established as the clear standard, the best approach depends on the circumstances. Here is a practical framework based on what the evidence supports:

• Mild adult fever with discomfort: A cool (not ice-cold) compress on the forehead, neck, or palms for 15 to 20 minutes at a time, with at least a 20-minute break between applications. This is primarily for comfort. Consider acetaminophen or ibuprofen if the temperature or symptoms warrant it.
• Child with fever: Medication first, if indicated. A brief lukewarm sponge or cool cloth for 10 to 15 minutes if the child tolerates it. Stop if shivering or visible distress begins. Do not wake a sleeping feverish child to apply a compress.
• High fever with medication already on board: Tepid sponging of the arms, legs, and trunk for 15 to 20 minutes can complement the medication’s effect. This window aligns with the period before the medication’s peak effect and when physical cooling has its strongest measurable impact.
• Suspected heat stroke: Continuous aggressive cooling without the on-off cycling. Call emergency services. Apply ice packs to the neck, armpits, and groin. Immerse in cool water if available. Do not stop cooling until professional help arrives or core temperature drops below 39°C.

In all fever scenarios, the compress is a supporting player, not the main act. Its most useful role is in the first 30 minutes of application, with diminishing returns after that. If you find yourself reapplying compresses every 20 minutes for hours on end with no improvement, the compress is not the tool for the job, and it is time to focus on medication or medical evaluation instead.