For adults, a fever is generally considered “high” once it reaches about 103 °F (39.4 °C), while in children the threshold is lower and depends heavily on age. A baby under three months old with any temperature above 100.4 °F (38 °C) needs prompt medical evaluation regardless of how the child looks, while a toddler between three months and three years crosses into high-fever territory around 101.3 °F (38.5 °C). These numbers shift depending on where on the body the temperature is taken and even what time of day it is, which makes the question less straightforward than a single cutoff on a thermometer might suggest.
The Numbers by Age Group
A multi-site diagnostic accuracy study that stratified fever thresholds by age offers a useful set of benchmarks. For infants three months and younger, any reading above 99.4 °F (37.4 °C) on an ear thermometer displaying an oral-equivalent value is flagged as a fever, with no separate “high fever” category because any confirmed fever at that age warrants attention. For children between three months and three years, fever begins above 99.6 °F (37.6 °C), and high fever is defined as anything above 101.3 °F (38.5 °C). For anyone older than three years, including adults, fever starts above 99.9 °F (37.7 °C), and the high-fever threshold is above 103.0 °F (39.4 °C).1PubMed Central. Threshold for defining fever varies with age, especially in children: A multi‐site diagnostic accuracy study
Most emergency departments and pediatric guidelines use 100.4 °F (38.0 °C) rectally as the universal starting line for “fever” in all ages, and that number is a reasonable rule of thumb. The confusion usually begins above it: parents and patients want to know when a fever stops being routine and starts being alarming. For a school-age child or adult, a temperature hovering around 101 °F is uncomfortable but rarely dangerous on its own. Once it climbs past 103 °F, the risk of complications rises and most clinicians recommend closer monitoring or an evaluation, especially if the fever persists beyond a day or two.
Where You Take the Temperature Changes the Answer
The thermometer you use can easily shift a reading by a degree or more, which matters when you are trying to decide whether a number qualifies as “high.” Rectal readings are the clinical gold standard, particularly for young children. An emergency-department study found that the average difference between rectal and oral temperatures was about 1.1 °F, but individual readings varied widely: an oral reading could be as much as roughly 2.9 °F lower than the rectal one.2PubMed Central. Oral and Tympanic Membrane Temperatures Are Inaccurate to Identify Fever in Emergency Department Adults That gap means a person could have an oral reading of 100 °F while their core temperature is actually closer to 103 °F.
Ear (tympanic membrane) thermometers sit somewhere in between. A separate study comparing oral, tympanic, and temporal artery readings against rectal temperatures found that about a third of tympanic and temporal artery readings differed from the rectal value by half a degree Celsius or more. Tympanic readings were the most precise of the non-rectal options, but their sensitivity for catching a true fever was only about 68 percent when using the standard 38 °C cutoff. Lowering the tympanic threshold to 37.5 °C pushed sensitivity above 90 percent with reasonable specificity.3PubMed. Temperature measurement in the adult emergency department: oral, tympanic membrane and temporal artery temperatures versus rectal temperature
The practical takeaway: if you are using an oral or forehead thermometer at home and the number looks borderline, the true core temperature could easily be higher. When in doubt with a young child, a rectal reading gives the most reliable answer. For adults, oral thermometers are fine for everyday use, but keep in mind that drinking cold water, breathing through your mouth, or taking the reading right after exercise can all skew the number downward.
Normal Body Temperature Is Not One Fixed Number
The old “98.6 °F is normal” figure comes from a 19th-century study and has been revised downward by more recent data, but even the revised averages mask substantial individual variation. A telemonitoring study tracking core temperature during daily routines found that individual baseline averages ranged from 36.6 to 37.5 °C (about 97.9 to 99.5 °F), and the daily swing in temperature varied more than fourfold from person to person.4Frontiers in Physiology. Telemonitored Human Circadian Temperature Dynamics During Daily Routine Someone whose resting temperature typically sits at 97.8 °F might actually be running a fever at 99.5 °F, a number that would be dismissed as normal in most settings.
Body temperature also follows a predictable daily rhythm, peaking in the evening and dipping in the early morning. A population-based study confirmed that temperatures tend to be lowest between about 8 and 10 a.m. and highest between 8 p.m. and 3 a.m. The pattern varies by age too: average temperatures rise from birth through age two, then gradually decline with older ages.5PubMed Central. Diurnal temperature variation and the implications for diagnosis and infectious disease screening: A population-based study A child who hits 102 °F at 7 a.m. is probably sicker than one who hits the same number at 10 p.m., because the morning reading is working against the body’s natural low point.
Febrile Seizures in Children
One of the biggest fears parents have around high fevers is seizures. Febrile seizures occur in roughly two to five percent of children between six months and five years of age, and they are the most common cause of seizures in early childhood. Research shows that the height of the temperature, rather than how quickly it climbs, is the strongest risk factor for a first febrile seizure. In general, the higher the temperature, the greater the chance a seizure will occur.6PubMed Central. Febrile seizures: an overview
Interestingly, the relationship between peak temperature and seizure recurrence goes in the opposite direction. One study found that children whose first febrile seizure happened at a lower peak temperature (around 101 °F) had a higher recurrence risk at one year (about 42 percent) than children who seized at a peak of 105 °F or above (about 12 percent).7PubMed Central. Febrile seizures That seems counterintuitive, but the explanation is that a child who seizes at a relatively low temperature likely has a lower personal seizure threshold and is therefore more vulnerable to repeat events.
Scary as they look, simple febrile seizures (lasting under 15 minutes, affecting the whole body, and occurring only once in 24 hours) are generally benign. They do not cause brain damage or increase the risk of epilepsy. Giving fever reducers preemptively has not been shown to reliably prevent them, which is another reason aggressive treatment of moderate fevers is not always necessary.
Why Fever Under Three Months Is Treated Differently
Newborns and very young infants are a special case. Their immune systems are still immature, and a fever can be the only outward sign of a serious bacterial infection like a urinary tract infection, bloodstream infection, or meningitis. In a cohort of 500 febrile infants under three months of age, about 32 percent were diagnosed with a serious bacterial infection, most commonly a urinary tract infection.8PubMed. FeBRILe3: Risk-Stratification and Diagnosis of Serious Bacterial Infections in Febrile Infants Less Than 3 Months Old A larger study of over a thousand febrile infants found a similar pattern, with serious bacterial infections diagnosed in about 22 percent of cases.9Scientific Reports. Febrile infants risk score at triage (FIRST) for the early identification of serious bacterial infections
These rates are high enough that most guidelines recommend a full workup, including blood tests, urine analysis, and sometimes a spinal tap, for any infant under three months who presents with a rectal temperature of 100.4 °F or above. For these babies, there is no waiting-and-watching period. This is the one scenario in which the question “how high is the fever?” matters less than the question “is there a fever at all?”
Older Adults and the Blunted Fever Response
At the other end of life, the problem reverses. Older adults tend to run lower baseline temperatures and have a harder time generating a robust fever when they are genuinely sick. A systematic review concluded that older adults are less able than younger adults to mount an effective febrile response, likely because of age-related changes in immune signaling and blood vessel regulation.10PubMed Central. Altered Febrile Responses in Older Adults: A Systematic Review A study of nursing home residents infected with SARS-CoV-2 confirmed this in real time: both baseline and maximum temperatures decreased with advancing age.11PubMed Central. The effect of age on fever response among nursing home residents with SARS-COV-2 infection
This means a reading of 100 °F in an 85-year-old can represent a more serious infection than 103 °F in a 30-year-old. Emergency physicians often use a lower threshold, sometimes 99 °F or even just a rise of about 2 °F above a person’s known baseline, to flag a potential infection in the elderly. A study of elderly patients in an emergency department found that rectal thermometry identified fevers in roughly 15 percent of patients who appeared afebrile by oral measurement.12The Journal of Emergency Medicine. A comparison of oral, tympanic, and rectal temperature measurement in the elderly If you are caring for an older person who seems unwell but whose thermometer reads “normal,” trust the clinical picture.
People on Chemotherapy
Cancer patients undergoing chemotherapy are another group where a fever that would be unremarkable in a healthy person can signal a medical emergency. Chemotherapy can wipe out the white blood cells that fight infection, a condition called neutropenia. A single temperature of 100.4 °F (38.0 °C) or above in a neutropenic patient is treated as febrile neutropenia, and clinical guidelines recommend that the first dose of antibiotics be given within one hour of arrival at the hospital.13PubMed. Outpatient Management of Fever and Neutropenia in Adults Treated for Malignancy: American Society of Clinical Oncology and Infectious Diseases Society of America Clinical Practice Guideline Update For these patients, the “high fever” threshold is irrelevant. Any fever is urgent.
Fever Versus Hyperthermia
Not every dangerously high body temperature is a fever. Fever is a regulated process: the brain deliberately raises the thermostat in response to infection, and the body’s heating and cooling systems work together to reach and maintain the new set point. That is why you shiver at the onset of a fever — your muscles are generating heat to bring your core temperature up to where the brain wants it.14PubMed. Fever versus hyperthermia
Hyperthermia is a different situation entirely. In hyperthermia, the body’s cooling mechanisms fail or are overwhelmed, and the temperature rises uncontrollably. Heatstroke is the classic example: the body cannot shed heat fast enough, and the temperature can soar past 106 °F. The brain’s thermostat is not the problem; the cooling system is. This distinction matters because aspirin-like drugs lower a fever by resetting the thermostat, but they do nothing for hyperthermia. Trying to treat heatstroke with ibuprofen is ineffective and wastes critical time.15DeckerMed Medicine. Hyperthermia, Fever, and Fever of Undetermined Origin
Certain medications can also trigger dangerous hyperthermia. Neuroleptic malignant syndrome, for instance, is a rare reaction to antipsychotic drugs that produces extreme temperatures along with muscle rigidity, altered consciousness, and unstable vital signs. The temperature rise results from both blocked heat-loss pathways in the brain and the massive heat generated by rigid muscles.16PubMed Central. Drug-induced hyperthermia in critical care Any temperature above 104 °F (40 °C) that comes on in the absence of obvious infection, especially alongside confusion or stiffness, should prompt a trip to the emergency department rather than a dose of Tylenol.
Treating a Fever at Home
The two main over-the-counter fever reducers are acetaminophen (Tylenol) and ibuprofen (Advil, Motrin). Both work, and their safety profiles are similar in children and adults. Head-to-head, ibuprofen appears to be a slightly more effective fever reducer. A meta-analysis of nine trials involving over a thousand children found that ibuprofen lowered temperature more than acetaminophen at two, four, and six hours after dosing, with the difference growing more pronounced at higher ibuprofen doses.17PubMed. Efficacy and safety of acetaminophen vs ibuprofen for treating children’s pain or fever: a meta-analysis A separate meta-analysis covering both adults and children reached the same conclusion: ibuprofen was at least as effective as acetaminophen and in most comparisons slightly better for both pain and fever.18PubMed. Efficacy and safety of ibuprofen and acetaminophen in children and adults: a meta-analysis and qualitative review
For children under two, the evidence is consistent. A systematic review of studies in this age group found that ibuprofen reduced temperature more than acetaminophen at both the less-than-four-hour and four-to-24-hour marks, with no meaningful difference in serious side effects between the two drugs.19JAMA Network Open. Comparison of Acetaminophen (Paracetamol) With Ibuprofen for Treatment of Fever or Pain in Children Younger Than 2 Years Ibuprofen should not be given to babies under six months without a doctor’s guidance, though, and acetaminophen is generally the first choice for very young infants.
One important point: the goal of treating a fever is to make the person more comfortable, not to normalize the number on the thermometer. A child with a temperature of 102 °F who is playing and drinking fluids does not necessarily need medication. A child with a temperature of 101 °F who is listless and refusing to drink probably does. Comfort and behavior matter more than the precise number.
Why Fever Exists in the First Place
Fever is not a malfunction. It is an ancient and highly conserved immune strategy found in mammals, reptiles, fish, and even some insects. The one-to-four-degree rise in core body temperature that accompanies a typical fever enhances several branches of the immune response, from speeding up white blood cell migration to making certain pathogens replicate more slowly. There is evidence that suppressing fever can worsen outcomes. Modeling studies have estimated that using antipyretics during influenza infection correlates with roughly a five percent increase in mortality at the population level, and animal studies found dramatically higher death rates when fever was pharmacologically blocked during viral infection.20PubMed Central. Fever and the thermal regulation of immunity: the immune system feels the heat
None of this means you should never treat a fever. A temperature of 104 °F accompanied by disorientation or rapid breathing demands intervention. But routine moderate fevers in otherwise healthy people are doing what they evolved to do, and the reflexive urge to extinguish them immediately is not always in the patient’s best interest.
Fever Phobia and What Parents Get Wrong
Researchers have been documenting parental anxiety around fever for over four decades, and the phenomenon has its own name: fever phobia. A landmark survey in the early 1980s found that most parents were unduly worried about low-grade fevers. More than half believed a moderate fever of 104 °F or below could cause serious brain damage.21PubMed. Fever phobia: misconceptions of parents about fevers A follow-up study 20 years later found that the problem had not improved: about 56 percent of caregivers were “very worried” about the potential harm of fever, roughly one in five listed brain damage as a possible consequence, and a quarter were giving fever-reducing medications at temperatures below 100 °F. Even more concerning, 44 percent of caregivers were dosing ibuprofen at too-frequent intervals, and 18 percent of those who sponged their children were using alcohol, a practice that can cause dangerous drops in blood sugar or even alcohol poisoning through the skin.22Pediatrics. Fever phobia revisited: Have parental misconceptions about fever changed in 20 years?
The fear that an untreated fever will keep climbing until it causes organ damage is one of the most stubborn myths. In a true fever (as opposed to hyperthermia), the brain’s thermostat is in control and rarely allows the temperature to exceed about 106 °F. The body is not losing control; it is defending a new set point. That does not mean you should ignore a persistently high fever, but it does mean that a temperature of 103 °F in a toddler who is drinking fluids and making eye contact is not an emergency. Watching the child matters more than watching the thermometer.
Post-Vaccination Fevers
Fevers after childhood vaccinations are common and almost always harmless, but they catch parents off guard because the child was healthy an hour earlier. A large retrospective study using app-reported temperature data found that post-vaccination fevers in children typically began between 9 and 14 hours after the shot and resolved within about 20 to 31 hours. Maximum temperatures clustered between 38.7 °C and 39 °C (roughly 101.7 to 102.2 °F), depending on the vaccine. The hepatitis A and influenza vaccines tended to produce the highest and longest-lasting fevers, while the pneumococcal vaccine fever resolved more quickly.23PubMed Central. Postvaccination Fever Response Rates in Children Derived Using the Fever Coach Mobile App: A Retrospective Observational Study Knowing these typical patterns can help parents distinguish a predictable vaccine response from an illness that needs attention. A fever that starts more than 48 hours after a vaccination, or one that persists beyond about two days, is more likely to have a separate cause.