A reading of 97.1 °F (36.2 °C) falls comfortably within the normal range for an adult measured orally. The old benchmark of 98.6 °F dates back to the mid-1800s, and modern research consistently finds that the true average oral temperature is closer to 97.9–98.2 °F, with healthy adults routinely registering anywhere from the mid-96s to the upper 98s depending on the time of day, how the measurement was taken, and individual factors like age and sex. So if you saw 97.1 on the thermometer and wondered whether something was wrong, the short answer is almost certainly no.
Where 98.6 Came From and Why It Is Too High
The figure everyone grew up with traces to Carl Wunderlich, a German physician who published a landmark study in 1868 based on roughly a million armpit readings from about 25,000 patients. He declared 37.0 °C (98.6 °F) the average human body temperature, and that number stuck for over a century. It was not seriously challenged until 1992, when a study at the University of Maryland measured oral temperatures in 148 healthy adults and found that the actual mean was 36.8 °C (98.2 °F), not 37.0 °C. The same study placed the upper boundary of normal at 37.7 °C (99.9 °F), a full half-degree Fahrenheit below Wunderlich’s threshold of 100.4 °F.1JAMA. A critical appraisal of 98.6 degrees F, the upper limit of the normal body temperature, and other legacies of Carl Reinhold August Wunderlich
Some of Wunderlich’s error may have been methodological. His thermometers were early instruments, less precise than what came later. He measured exclusively in the armpit, which tends to read lower than oral and can vary with how tightly the arm is held. But the discrepancy is probably not just about old equipment.
Human Body Temperature Has Been Dropping for Generations
A large study published in eLife analyzed more than 677,000 temperature measurements from three U.S. cohorts spanning nearly two centuries. The researchers found a steady decline in body temperature by birth decade: roughly 0.03 °C per decade for both men and women. Men born in the early 1800s ran about 0.59 °C warmer than men born in the late 1990s, and women showed a comparable drop of about 0.32 °C between the 1890s and the 1990s.2eLife. Decreasing human body temperature in the United States since the Industrial Revolution The trend held across racial groups and could not be explained by changes in thermometer technology alone, since the decline continued even within cohorts measured with the same instruments.
The leading hypothesis is that reduced chronic inflammation is cooling us down. Before antibiotics, widespread sanitation, and dental care, low-grade infections like tuberculosis, periodontal disease, and chronic wound infections were far more common. Persistent inflammation raises your metabolic baseline and, with it, your temperature. As the population burden of chronic infection fell over the twentieth century, average resting temperature appears to have followed. Changes in climate-controlled living, diet, and physical activity levels may also play a role, though these are harder to isolate.
The practical takeaway is that 98.6 was probably a reasonable average for people living in the 1860s. It is not a reasonable average for someone born in the 1980s or later. A reading of 97.1 today does not require the same explanation it might have needed 150 years ago, because the population average itself has shifted downward.
What Counts as the Normal Range
A systematic review pooling data from 36 studies across multiple countries found that normal oral temperature, defined as the mean plus or minus two standard deviations, spans 35.7 to 37.4 °C, or about 96.3 to 99.3 °F.3PubMed Central. Normal Body Temperature: A Systematic Review At 97.1 °F (36.2 °C), you are sitting well inside that window. A separate systematic review that emphasized studies with strong or fairly strong evidence found an even wider oral range of 33.2 to 38.2 °C, though the low end of that range reflects occasional outlier readings rather than typical values.4PubMed. Normal oral, rectal, tympanic and axillary body temperature in adult men and women: a systematic literature review
These ranges are wide enough that two perfectly healthy people can differ by more than a full degree Fahrenheit at the same time of day using the same thermometer. Your own baseline temperature is more informative than any single population average, and that baseline can shift from day to day for reasons as mundane as how much sleep you got or what you ate for breakfast.
Time of Day Makes a Bigger Difference Than Most People Realize
Your body temperature is not a fixed number. It follows a circadian cycle, dipping to its lowest point in the early morning, roughly two hours before you wake up, and peaking in the evening about two hours before you typically fall asleep.5PubMed Central. Human Body Temperature Circadian Rhythm in Health and Disease The 1992 Maryland study measured this swing at about 0.9 °F on average, with nadir around 6 AM and peak between 4 and 6 PM.1JAMA. A critical appraisal of 98.6 degrees F, the upper limit of the normal body temperature, and other legacies of Carl Reinhold August Wunderlich
If you took your temperature first thing in the morning and saw 97.1, you caught yourself near the bottom of the daily curve. The same person might read 97.8 or 98.0 in the late afternoon and both numbers are equally healthy. People who check their temperature only when they feel off tend to compare their reading against 98.6 without accounting for the time of day, which creates unnecessary anxiety in the morning and can mask a mild fever in the evening.
Where You Measure Changes the Number
Not all thermometer placements give you the same reading, and confusing one site’s value with another is one of the most common mistakes people make at home. Rectal temperatures run higher than oral readings, which in turn run higher than armpit readings. A study of emergency department patients found that mean oral temperature was 98.3 °F while mean rectal temperature was 99.4 °F, a gap of about 1.1 °F on average. More troublingly, the individual spread was wide: oral readings could be nearly 3 °F lower or 0.7 °F higher than the rectal value in the same person.6PubMed Central. Oral and Tympanic Membrane Temperatures Are Inaccurate to Identify Fever in Emergency Department Adults
Axillary (armpit) temperatures tend to be the lowest of all. One study of healthy young adults at rest measured an average armpit temperature of 36.4 °C (97.5 °F), which would correspond roughly to a higher reading if you moved the thermometer to the mouth or ear.7PubMed Central. Assessment of axillary temperature for the evaluation of normal body temperature of healthy young adults at rest in a thermoneutral environment The systematic review mentioned earlier placed the normal axillary range at 35.0 to 36.9 °C (95.0 to 98.5 °F), confirming that armpit readings can run a degree or more below oral.3PubMed Central. Normal Body Temperature: A Systematic Review
What does this mean for your 97.1 reading? If that was an oral measurement, you are squarely normal. If it was an armpit reading, your core temperature is likely a bit higher than 97.1. And if it was a forehead or temporal scan, keep in mind that these devices can be thrown off by sweating, ambient temperature, and even blood flow to the skin. A systematic review found that oral temperature readings are themselves influenced by ambient air temperature, probe placement, and recent fluid intake, so precision from any single measurement has limits.8PubMed Central. Is oral temperature an accurate measurement of deep body temperature? A systematic review
Ambient Temperature Can Nudge Your Reading
If you just came inside from a cold walk and immediately took your temperature, the number may be artificially low. A study measuring the effect of temperature extremes on oral and tympanic readings found that oral temperatures dropped by up to 0.5 °C after cold exposure and rose by up to 0.4 °C after hot exposure. These shifts were transient but lasted long enough that readings taken within about 20 minutes of coming indoors could be misleading.9PubMed. The effect of ambient temperature extremes on tympanic and oral temperatures The fix is straightforward: wait at least 15 to 20 minutes after coming inside, drinking something hot or cold, or exercising before taking an oral reading.
Age Pulls Temperature Downward
Older adults tend to run cooler than younger adults, which makes a reading like 97.1 even less remarkable in someone over 65. A study of 150 adults with an average age of about 81 found that 6 AM mean temperature was 97.3 °F in nursing home residents, and even midday readings in community-dwelling older adults averaged just 97.7 °F. At every time point, the vast majority of readings fell below 98.6: 97% of morning readings, 94% of afternoon readings, and 90% of midday readings in community dwellers were under the traditional benchmark.10PubMed. Older is colder: temperature range and variation in older people
This cooling with age is not just a curiosity. Research on aging and cognition has found that healthy aging is associated with core temperatures in the lower range of normal, while certain neurological conditions like Alzheimer’s disease may be associated with temperatures slightly above normal.11PubMed Central. Body Temperature Is Associated With Cognitive Performance in Older Adults With and Without Mild Cognitive Impairment: A Cross-sectional Analysis Running a bit cool in later years appears to be the physiological norm, not a warning sign, though persistent temperatures well below 96 °F can indicate hypothermia and deserve medical attention.
There is, however, a clinical wrinkle for frail older adults. A study of older emergency department patients found that frailty is associated with lower body temperatures, likely because of declining metabolic reserves, and those low-temperature frail patients had higher mortality risk.12PubMed Central. Low body temperature and mortality in older patients with frailty in the emergency department The concern here is not that 97.1 is dangerous but that a very low baseline can mask a fever. A frail person whose usual temperature is 96.8 might be mounting a significant immune response at 98.5, a number that would look “normal” to an ER triage nurse comparing against 98.6.
What This Means for Defining a Fever
The traditional fever threshold of 100.4 °F (38.0 °C) was calibrated to an era when 98.6 was considered the average. If the true average is closer to 98.2 or even lower, then insisting on 100.4 as the cutoff for fever may miss genuinely elevated temperatures in some people. A pilot study on cancer patients at risk for neutropenic fever illustrates the problem. The researchers found that patients’ average baseline temperature was about 36.7 °C (98.1 °F), and a temperature would have to climb more than four standard deviations above that baseline to reach the standard 38.0 °C fever threshold.13PubMed Central. Personalized Cutoffs for the Diagnosis of Neutropenic Fever Based on Patients’ Baseline Body Temperature: A Retrospective Pilot Study In plainer terms, using the conventional cutoff would fail to flag most people who were running significantly warmer than their own normal.
A related analysis of the same question estimated that at two standard deviations above the average baseline, only about 3% of patients would reach 38.0 °C, and at three standard deviations, only 20% would get there.14Open Forum Infectious Diseases. 1022. The Role of Baseline Body Temperature in Neutropenic Fever This is why some clinicians are now advocating for personalized fever thresholds, especially for patients with cancer or other immune-compromising conditions. If you know your baseline is 97.1, a jump to 99.5 might be more meaningful for you than it would be for someone who normally sits at 98.4.
For everyday purposes, a common clinical rule of thumb is that oral temperatures above 99.5 to 100.0 °F should prompt attention, especially if accompanied by symptoms. But the underlying point is that a single number does not capture the full picture. Your own pattern matters more than a one-size-fits-all line on the thermometer.
Sex Differences and Other Individual Variation
Women tend to run slightly warmer than men. The 1992 Maryland study confirmed this, as did the systematic review of body temperature ranges, which found somewhat broader oral temperature ranges in women (33.2 to 38.1 °C) compared with men (35.7 to 37.7 °C).4PubMed. Normal oral, rectal, tympanic and axillary body temperature in adult men and women: a systematic literature review Menstrual cycle phase drives some of this: body temperature rises by roughly 0.3 to 0.5 °C after ovulation and stays elevated through the luteal phase. A woman who normally reads 97.1 in the first half of her cycle might read 97.5 or 97.8 in the second half without anything being wrong.
Body composition plays a role too, though its effects on resting temperature are more subtle than people assume. Higher metabolic rate, which correlates with muscle mass and physical activity, can raise resting temperature slightly, while lower metabolic output can pull it down. Early-life health history may also leave lasting marks: animal and human research suggests that neonatal infections can alter thermoregulatory programming in ways that persist into adulthood.15PubMed. Neonatal bacterial infection alters fever to live and simulated infections in adulthood
Medications That Shift Your Thermostat
Certain drugs can nudge your resting temperature in one direction or another, particularly under heat stress. A systematic review and meta-analysis found that medications with strong anticholinergic effects raised core temperature by about 0.4 °C when air temperatures exceeded 30 °C, primarily by suppressing sweating. Non-selective beta-blockers, adrenaline, and anti-Parkinson’s medications also produced small but measurable increases.16PubMed Central. The effect of prescription and over-the-counter medications on core temperature in adults during heat stress: a systematic review and meta-analysis Common antidepressants, diuretics, and drugs with weaker anticholinergic activity did not alter core temperature significantly in the same analysis. If you are on a medication that suppresses sweating or alters vascular tone, your temperature readings might be slightly higher in warm environments but would not explain a reading of 97.1.
On the other end, anti-inflammatory drugs (like ibuprofen or acetaminophen) can lower a mildly elevated temperature toward baseline. If you took something for a headache and then checked your temperature, the reading might be pushed slightly downward. These effects are usually small, on the order of a few tenths of a degree, and are unlikely to produce a reading outside the normal range in someone who was not already running warm.
Continuous Monitoring and the Future of Temperature Tracking
Wearable devices that track skin temperature around the clock are beginning to change how researchers and clinicians think about body temperature. A study on continuous wrist-worn temperature monitoring found that most of the variation in a person’s temperature trace came from the expected daily rhythm, with clear lower temperatures during the daytime and higher temperatures overnight. When participants reported fevers, the continuous data captured elevated nighttime peaks that aligned with symptom onset, a pattern that single spot-checks would likely have missed.17Scientific Reports. Feasibility of continuous fever monitoring using wearable devices
This matters because the traditional way of checking temperature, a single reading at a single point in time, captures only a snapshot of a moving target. Two people with the same single reading of 97.1 might have very different temperature curves over 24 hours. One might be cycling normally between 96.5 at night and 97.8 during the day. The other might normally peak at 98.4 and is reading low because they are about to develop an illness. Continuous monitoring makes these distinctions visible. For most healthy adults, this level of surveillance is unnecessary. But for people managing chronic illness, undergoing chemotherapy, or recovering from surgery, a running picture of temperature trends is more useful than any single number on any single morning.