medication for fever Archives

Reviewed and rewritten 29 August 2026. The original 2011 version of this post has been replaced. The clinical guidance below reflects current recommendations.

This post first went up on 28 February 2011. That same day, the American Academy of Pediatrics published its clinical report on fever and antipyretic use in children (1). I did not plan that. But the report says the thing I have spent years trying to say to worried parents, and it says it in the first paragraph: the goal of treating a febrile child is to improve the child’s comfort, not to make the number on the thermometer smaller.

Fifteen years later, that is still the least popular sentence in pediatrics.

The number is the least useful thing in the story

Fever is a symptom. It is the immune response doing what it evolved to do, and by itself it tells you very little about how sick a child is.

A garden-variety roseola can run 104. A serious bacterial infection in a young infant can run 100.6, or no fever at all. The height of the fever does not grade the illness, and I have watched that fact fail to land over and over, because it is genuinely counterintuitive and because everyone has an aunt with a story.

What actually grades the illness is how the child looks and behaves. A child of two with a temperature of 103.5 who is drinking, making eye contact, complaining, and playing with something on the floor between complaints is a child I am not very worried about. A child of two with a temperature of 100.8 who is limp, mottled, will not drink, and does not track you across the room is a different situation entirely, and the thermometer is the last thing I care about.

Look at the child. Not the number.

Fever phobia has a name and a date

Here is my one digression, and I think it earns its place.

In 1980, Barton Schmitt surveyed parents in a pediatric clinic and found that most of them believed untreated fever could cause brain damage, and a majority were treating temperatures below 100 (2). He called it fever phobia, and the name stuck.

Twenty years later, Crocetti and colleagues repeated the survey to see whether two decades of patient education had moved anything. They found that 91 percent of caregivers still believed fever could cause harm, and a quarter were giving antipyretics for temperatures under 100 (3).

Two more decades on, in a job where I read the chief complaint before I ever see the patient, I would not bet on the number having improved. This is one of the most durable beliefs in medicine, and no amount of handout has dented it. I mention it because if you are the parent reading this and feeling slightly told off, you are in enormous company, and the belief was handed to you by people who meant well.

Antipyretics do not prevent febrile seizures

This is the single most useful thing on this page, and it is the one almost nobody has been told.

The fear underneath most aggressive fever treatment is the febrile seizure. Parents medicate at 100.5 because they are trying to head one off. It is a completely understandable thing to do and it does not work.

Rosenbloom and colleagues pooled the randomized trials in a systematic review and meta-analysis and found that antipyretics did not reduce the recurrence of febrile seizures (4). In the trials, the drugs did not even reliably lower the temperature during the episodes in which a seizure happened. The AAP report says the same thing plainly (1).

Why this is so is not fully settled. Susceptibility looks largely genetic, and by the time anyone notices a fever and reaches for a bottle, whatever sets the seizure off has usually already happened. What is settled is the practical part, which is that the drug does not stop it.

So if you have been setting alarms overnight to dose a child who is otherwise sleeping comfortably, in the belief that you are preventing a seizure, you can stop. You are not preventing anything, and you are waking a sick child who needs the sleep.

Febrile seizures are frightening to watch and are, in the ordinary simple form, benign. That is a strange pair of facts to hold at once. It does not make them less frightening.

When fever genuinely is an emergency

All of the above is about the well-appearing older child. There are situations where fever is an emergency and the reassurance does not apply.

Any infant under three months with a rectal temperature of 100.4 F (38.0 C) or higher needs to be seen the same day. Not tomorrow. That is a hard rule and I do not soften it. Young infants have immature immune responses, they do not localize infection well, and serious bacterial infection can be present in a baby who looks unremarkable. The AAP’s 2021 guideline builds its whole framework on that 100.4 threshold and separates management into three age bands, 8 to 21 days, 22 to 28 days, and 29 to 60 days, with the youngest group getting the most aggressive workup (5).

Do not give the baby acetaminophen before you go. It does not help and it makes the assessment harder.

At any age, these need urgent attention regardless of the temperature: a rash that does not blanch when you press a glass against it, difficulty breathing or fast breathing, a stiff neck or severe headache with photophobia, a child who cannot be roused or who is confused, no wet diaper in eight hours or no tears when crying, a seizure lasting more than five minutes, or a first seizure of any kind. Fever persisting past five days deserves a look for Kawasaki disease.

An immunocompromised child, or a child with sickle cell disease, is a different conversation and those families already know it.

Dosing, and the schedule I will not write

If you are treating, treat for comfort.

Acetaminophen is 10 to 15 mg/kg per dose, every four to six hours, and no more than five doses in twenty-four hours. Ibuprofen is 5 to 10 mg/kg per dose, every six to eight hours, and not in an infant under six months. Dose by weight, never by age, because the age bands on the box are a rough proxy and children of the same age vary enormously. Use the syringe that came in the package.

No aspirin. Not for a child, not for a teenager with a viral illness, because of Reye syndrome. That one has been settled since the 1980s and it still needs saying.

Now the opinion. I do not write alternating acetaminophen and ibuprofen schedules, and I think the practice is one of the worse pieces of advice in general circulation.

The evidence for it is that dual therapy lowers temperature slightly more than either drug alone. Fine. But we have already established that the temperature reading was never the point, so the benefit being demonstrated is a benefit to the thermometer rather than to the child. Set against that, you have a tired parent at three in the morning running two drugs on two different intervals, four to six hours for one and six to eight for the other, and antipyretics are already among the most common medications involved in accidental overdose at home. The AAP’s position is that a single agent at the right dose and interval is usually preferred, and that if a clinician does recommend alternating, the family needs a written schedule and explicit teaching (1).

My rule is simpler. Pick one drug. Dose it correctly. If the child is still miserable, that is information about the illness, not a signal to add a second drug.

What I ask on a video visit

I practise entirely by telemedicine now, so I cannot put a hand on a child’s abdomen or look at an eardrum, and that shapes what I ask for.

I want the child in the frame, awake, not asleep on a shoulder. I want to see them track something, and I want to see them drink. I ask how many wet diapers or trips to the bathroom there have been today, because hydration is the thing I can assess remotely and it drives most of the disposition. I ask what the temperature was, how it was taken, and with what, because a temporal scanner on a squirming toddler and a rectal thermometer in a six-week-old are not the same measurement and only one of them counts under three months.

And I ask the parent directly whether the child seems like themselves. Parents are better at that judgment than they give themselves credit for, and when a parent tells me something is off even though they cannot name it, I take it seriously and I lower my threshold for sending them in.

For parents

Treat your child, not the thermometer. If they are comfortable, drinking, and behaving roughly like themselves, a fever of 103 does not require anything from you except fluids and attention.

You are not preventing a febrile seizure by medicating, so you do not need to wake a sleeping child to dose them.

Under three months, a rectal temperature of 100.4 or above is a same-day problem, and that overrides everything else on this page.

For colleagues

Say the comfort framing out loud at the visit. Parents accept it far better than we expect, and it is the sentence that changes behaviour.

Stop handing out alternating schedules by default. If the child is uncomfortable enough that one correctly dosed agent is not touching it, the useful move is to reconsider the diagnosis, not to add a second drug and a second interval to a household that is already short of sleep.

The under-three-month rule survives contact with a reassuring history. Well-appearing is not a discharge criterion in that age group.

The Bottom Line

Fever is a symptom and its height does not tell you how sick a child is. Treat for comfort, using one drug dosed by weight, and skip the alternating schedule. Antipyretics do not prevent febrile seizures, so there is no reason to wake a comfortable child to medicate. The exceptions that matter are the ones that override all of it: any infant under three months with a rectal temperature of 100.4 F or higher needs to be seen the same day, and so does any child who looks genuinely unwell, whatever the thermometer says.

Sources

1. Sullivan JE, Farrar HC; Section on Clinical Pharmacology and Therapeutics; Committee on Drugs. Fever and antipyretic use in children. Pediatrics. 2011;127(3):580-587. PMID 21357332. https://pubmed.ncbi.nlm.nih.gov/21357332/

2. Schmitt BD. Fever phobia: misconceptions of parents about fevers. Am J Dis Child. 1980;134(2):176-181. PMID 7352443. https://pubmed.ncbi.nlm.nih.gov/7352443/

3. Crocetti M, Moghbeli N, Serwint J. Fever phobia revisited: have parental misconceptions about fever changed in 20 years? Pediatrics. 2001;107(6):1241-1246. PMID 11389237. https://pubmed.ncbi.nlm.nih.gov/11389237/

4. Rosenbloom E, Finkelstein Y, Adams-Webber T, Kozer E. Do antipyretics prevent the recurrence of febrile seizures in children? A systematic review of randomized controlled trials and meta-analysis. Eur J Paediatr Neurol. 2013;17(6):585-588. PMID 23702315. https://pubmed.ncbi.nlm.nih.gov/23702315/

5. Pantell RH, Roberts KB, Adams WG, et al. Evaluation and Management of Well-Appearing Febrile Infants 8 to 60 Days Old. Pediatrics. 2021;148(2):e2021052228. PMID 34281996. https://pubmed.ncbi.nlm.nih.gov/34281996/

Dr. Scott Rennie

Tag: medication for fever

Child Has a Fever? Why You Shouldn’t Rush to Medicate