For most adults, a fever starts at a core (rectal) temperature of about 100.4°F (38.0°C), but the exact cutoff depends on where you measure it and what time of day it is. For children, most pediatric guidelines also use 100.4°F rectally as the threshold, though the numbers shift slightly by age and measurement site. These thresholds sound simple, but the reality is messier than a single number suggests, because “normal” body temperature is a moving target that changes throughout the day, varies between individuals, and reads differently depending on the thermometer you use.
The Standard Thresholds and Why They Vary
There is no single universally agreed-upon number that defines a fever. The widely cited 100.4°F (38.0°C) measured rectally serves as the default in most hospital and clinical guidelines. But when you look at how major medical references define fever, the picture gets more complicated. According to a review of fever definitions, Harrison’s Principles of Internal Medicine defines fever as a rectal temperature of 99.5°F to 100.9°F, while setting the oral threshold at above 99°F in the morning or above 99.9°F in the late afternoon. The Merck Manual, by contrast, draws the line at an oral temperature above 100.0°F or a rectal temperature above 100.8°F.1PubMed Central. Defining Fever
These aren’t contradictions so much as reflections of how messy human temperature regulation actually is. Your body doesn’t maintain a fixed 98.6°F all day. That iconic number comes from a 19th-century German study and has since been shown to overestimate the modern average. Body temperature naturally dips to its lowest point in the early morning, around 5 to 7 AM, and peaks in the evening between roughly 8 and 10 PM.2IntechOpen. Human Body Temperature Circadian Rhythm in Health and Disease – Section: Diurnal variation of human body temperature – A manifestation of our circadian rhythm That swing can be a full degree Fahrenheit in healthy people, which is why Harrison’s gives separate morning and afternoon oral thresholds.
In practical terms, a reading of 99.5°F at 6 AM is more concerning than the same reading at 8 PM, because it represents a bigger departure from where your temperature should be at that hour. Most people checking a thermometer at home don’t factor in time of day, and most clinical settings default to the simple 100.4°F rectal cutoff for convenience.
Measurement Site Matters More Than You Think
The gap between thermometer types is large enough to change whether you “have a fever” or not. A rectal reading is considered the closest to true core body temperature. Oral readings run lower, axillary (armpit) readings lower still, and the various infrared devices (ear and forehead) fall somewhere in between but with wider variability.
In a study of emergency department adults, the average difference between rectal and oral temperatures was about 1.1°F, but individual readings could be as much as nearly 3°F apart. Oral thermometers had a strong positive predictive value for fever (97%), meaning that when an oral thermometer said you had a fever, it was almost always right. Tympanic (ear) thermometers performed much worse, with a positive predictive value of only 55%.3PubMed Central. Oral and Tympanic Membrane Temperatures Are Inaccurate to Identify Fever in Emergency Department Adults
A systematic review comparing thermometer types found that temporal (forehead) and tympanic infrared thermometers were specific but not particularly sensitive for detecting fever. That means a below-threshold reading on a forehead scanner doesn’t reliably rule out a true fever, because many peripheral thermometers miss fevers that a rectal thermometer would catch.4PubMed Central. The diagnostic accuracy of digital, infrared and mercury-in-glass thermometers in measuring body temperature: a systematic review and network meta-analysis Among infrared devices tested in an outpatient clinic, temporal artery thermometers performed best, with sensitivity and specificity both around 88%, while contactless forehead thermometers had a sensitivity of only 13%.5PubMed Central. Are all thermometers equal? A study of three infrared thermometers to detect fever in an African outpatient clinic
If you’re using an armpit thermometer on a child, the general adjustments don’t always hold up neatly. In a pediatric study, axillary temperatures were on average about 0.25°C lower than oral and about 0.43°C lower than rectal. To best detect a true fever of 38.0°C rectally, the armpit threshold needed to drop to 37.2°C (about 99°F).6PubMed Central. Axillary, Oral, and Rectal Routes of Temperature Measurement During Treatment of Acute Kawasaki Disease So the common advice of “add a degree to an armpit reading” is roughly in the right ballpark, but it’s an approximation that can miss real fevers or flag false ones.
Fever Thresholds in Children by Age
The American Academy of Pediatrics and similar European guidelines define fever in children as a temperature above 38.0°C (100.4°F) for all ages, with rectal thermometers recommended for children under three.7PubMed Central. Threshold for defining fever varies with age, especially in children: A multi-site diagnostic accuracy study That 100.4°F figure is the standard clinical threshold, but in very young infants the picture is more granular.
A study of rectal temperatures in healthy infants under three months found that a single 100.4°F cutoff classified about 6.5% of perfectly well babies as febrile. The researchers found that the threshold should actually shift slightly with age: 100.4°F (38.0°C) for newborns up to 30 days old, 100.6°F (38.1°C) for one-month-olds, and 100.8°F (38.2°C) for two-month-olds. Normal infant temperatures also ran higher in summer than in winter.8PubMed. What is fever? Normal temperature in infants less than 3 months old
The stakes of a fever are different for a newborn than for a toddler. In infants under about two months, any fever at all warrants medical evaluation because their immune systems are immature and infections can progress quickly. In older children and adolescents, the threshold and level of concern align more closely with adult guidelines, and a low-grade fever with no other worrisome symptoms can often be monitored at home.
Fever in Older Adults Often Looks Different
If the standard threshold for a fever is 100.4°F, older adults frequently fail to reach it even when seriously ill. In elderly nursing home residents, the mean oral baseline temperature was only about 97.4°F, well below the traditional 98.6°F. During documented infections, nearly half had a “blunted” fever response, with a maximum temperature that never reached 101°F. About a quarter of those blunted cases actually showed a meaningful rise from their own low baseline but still didn’t hit the standard fever cutoff.9PubMed. Fever response in elderly nursing home residents: are the older truly colder?
Across studies of elderly patients, the classic fever response is absent or blunted roughly 20% to 30% of the time during infections.10PubMed. Fever in the elderly This is why geriatric guidelines sometimes use a lower fever threshold, such as 99°F oral or a rise of 2°F or more above the person’s known baseline, whichever comes first. For an older adult, waiting for the thermometer to hit 100.4°F before suspecting infection can mean dangerously delayed care.
What Fever Actually Is (and What It Isn’t)
Fever isn’t your body malfunctioning; it’s your body deliberately turning up the thermostat. When your immune system detects an infection, inflammatory signals reach the brain and trigger the production of a chemical messenger called prostaglandin E2 in a region of the hypothalamus that acts as the body’s temperature control center.11PubMed Central. Neural Mechanisms of Inflammation-Induced Fever That molecule resets the target temperature upward, and the body then works to reach the new set point through shivering, blood vessel constriction, and reduced sweating. You feel cold during a rising fever not because your body is cold, but because your brain has decided the current temperature is too low.
This is fundamentally different from overheating due to heat stroke, prolonged exercise, or drug reactions. In those cases, the body’s thermostat hasn’t changed; the body is simply overwhelmed by external or internal heat it can’t shed. Fever involves a regulated rise that the body actively defends, while unregulated hyperthermia involves a temperature climbing beyond the body’s ability to control it.12PubMed. Fever versus hyperthermia That distinction matters because treatments are different: antipyretics like acetaminophen and ibuprofen work by blocking prostaglandin production, which lowers the set point. They are effective for fever but do very little for heat stroke, which requires aggressive external cooling.
The Menstrual Cycle and Other Baseline Shifts
Women in the second half of their menstrual cycle (the luteal phase, after ovulation) run roughly 0.3°C to 0.5°C warmer than they do in the first half. This shift is linked to progesterone, the hormone that rises after ovulation. In one study, baseline core temperature in the luteal phase averaged about 37.08°C compared with 36.66°C in the follicular phase.13PubMed. Estrogen modifies the temperature effects of progesterone Another study found a moderate correlation between progesterone levels and basal body temperature, with temperature plateauing once progesterone rose above a certain concentration.14PubMed. Descriptive analysis of the relationship between progesterone and basal body temperature across the menstrual cycle
This means a woman in her luteal phase might register 99.3°F and be perfectly healthy, while that same reading in the early follicular phase could represent a mild fever. Hormonal contraceptives containing progestins can produce a similar warming effect. It’s one more reason why knowing your own personal baseline is more useful than comparing against a population average.
Should You Treat a Fever or Let It Run?
Fever has a genuine biological purpose. At elevated temperatures, immune cells move faster, kill pathogens more efficiently, and produce more of the signaling molecules that coordinate the immune response. Febrile temperatures boost the activity of white blood cells, enhance interferon responses that fight viruses, and make rapidly dividing bacteria more vulnerable to destruction.15PubMed Central. Let fever do its job The meaning of fever in the pandemic era – Section: HOW FEVER WORKS—EXPERIMENTS AND THEORY Across animal species, fever appears to confer a survival advantage during infection.16PubMed Central. Fever and the thermal regulation of immunity: the immune system feels the heat
None of that means you should tough out every fever without medication. The main reason to treat a fever in practice is comfort. A child who is miserable, refusing fluids, or unable to sleep benefits from having the temperature brought down, even if the fever itself was doing some immune work. The American Academy of Pediatrics frames fever treatment as being about the child’s overall comfort and hydration, not about chasing a number on the thermometer.
That said, when temperatures climb above roughly 104°F (40°C) in adults or 105°F in children, the risk of cellular damage increases. At extreme levels, sustained high temperatures can cause organ dysfunction through a combination of direct cellular injury, metabolic stress, and circulatory strain.17PubMed Central. The pathophysiological basis and consequences of fever Temperatures in that range warrant urgent medical attention regardless of the cause.
Sponging, Medication, or Both
Tepid sponging is a time-honored home remedy for fever, and it does produce an immediate temperature drop by pulling heat out through the skin. But the effect is shallow and temporary. A meta-analysis found that children who received only sponging were 75% less likely to be fever-free two hours later compared with children who received acetaminophen (paracetamol).18PubMed Central. Is tepid sponging more effective than paracetamol at relieving fever in febrile children in hot tropical climates? a mini review Another trial found sponging lowered temperature within the first 30 minutes but produced no clinically meaningful reduction beyond that, and children remained febrile. A significant proportion of the sponged children also experienced shivering and crying.19PubMed Central. Comparison of Cold Water Sponging and Acetaminophen in Control of Fever Among Children Attending a Tertiary Hospital in South Nigeria
A Cochrane review of physical cooling methods confirmed that shivering and goose pimples were about five times more common in children who were sponged compared to those given medication alone.20Cochrane Database of Systematic Reviews. Physical methods for treating fever in children Sponging works against what the brain is trying to do: the hypothalamus wants the body warmer, so cooling the skin triggers compensatory shivering that can actually increase discomfort and metabolic demand. Medication addresses the root cause by lowering the hypothalamic set point. Sponging can be a helpful supplement while waiting for medication to take effect, but it’s not a substitute.
Fever Phobia in Parents
Parents consistently overestimate how dangerous fevers are. In a well-known survey, nearly half of parents considered temperatures below 100.4°F to be fevers, over 40% believed temperatures below 104°F could be dangerous, and 15% thought untreated fever could climb to 107.6°F or higher on its own.21Pediatrics. Parental Fever Phobia and Its Correlates More recent studies have found the anxiety persists: in one survey, more than 90% of mothers favored giving antipyretics for any fever, and fears of seizures and brain damage were the most common concerns.22PubMed Central. Why Fever Phobia Is Still Common? In some settings, more than 70% of parents give antipyretics even when the temperature is below the treatment threshold.23Patient-Oriented Medicine and Pharmacy. Parent’s fever phobia: a new look at an old problem
The fear of febrile seizures drives much of this anxiety, and understandably so. Febrile seizures affect roughly 2% to 5% of young children, typically between the ages of six months and five years.24PubMed Central. A Review of Febrile Seizures: Recent Advances in Understanding of Febrile Seizure Pathophysiology and Commonly Implicated Viral Triggers They are alarming to witness but overwhelmingly benign in outcomes. In the vast majority of cases, they last under 15 minutes and do not cause lasting brain injury.25Jurnal Kesehatan Pasak Bumi Kalimantan. Differences in Risk Factors between Simple Febrile Seizures and Complex Febrile Seizures Importantly, giving antipyretics to a feverish child has not been shown to prevent febrile seizures. The seizures appear to be triggered by the rapid rise in temperature, not by how high the temperature ultimately gets, so by the time you notice the fever and reach for the medicine, the vulnerable window has typically already passed.
Behavioral Fever in Cold-Blooded Animals
Fever isn’t limited to mammals and birds. Fish, reptiles, and amphibians can’t raise their own body temperature internally, but when they’re infected, they seek out warmer environments to achieve the same effect. This phenomenon, called behavioral fever, produces measurable immune benefits. In a recent study of Nile tilapia infected with a bacterial pathogen, fish that moved to warmer water showed improved T cell function and better survival. The researchers concluded that the link between fever and adaptive immunity is an ancient strategy that predates the evolution of warm-blooded animals.26PubMed Central. Cold-blooded vertebrate utilizes behavioral fever to alleviate T cell apoptosis and optimize antimicrobial immunity This behavioral fever response has been documented across a wide range of ectothermic vertebrates.27PubMed. Behavioral fever in ectothermic vertebrates
The fact that animals who can’t generate internal heat will actively expend energy and risk predator exposure to warm themselves during infection tells you something about how valuable fever is as a defense mechanism. It evolved hundreds of millions of years ago and has been conserved across vertebrate lineages, which is a strong signal that, on balance, the benefits of a moderate fever outweigh the metabolic costs. That context is worth keeping in mind the next time you reach for the medicine cabinet at the first sign of 100°F on the thermometer.