Mind & Body

A Normal Temperature Isn't Really 98.6 Degrees Fahrenheit

You're shivering, achy, and hot to the touch. You're almost positive you have a fever. But when the thermometer beeps, you're dismayed to see the reading: 99 degrees Fahrenheit (37.2 degrees Celsius). If you don't have a fever, why do you feel so lousy? The answer: You very well might have a fever; it's the medical definition of a normal body temperature that's wrong. It turns out that the old 98.6-degree rule of thumb is a result of antiquated science that nobody thought to question — until relatively recently.

Hot Blooded, Check It and See

It's easy to imagine that doctors have been taking temperatures and using stethoscopes since the age of Hippocrates. But in fact, the idea of taking scientific measurements to diagnose illness didn't come around until the 19th century. That was when physicians stopped trying to fit individual symptoms to abstract philosophical principles and started actually observing large populations of patients with the most cutting-edge equipment of their day to arrive at conclusions supported by the evidence. That was also when a German medical student named Carl Reinhold August Wunderlich was trained as a physician.

Wunderlich wrote several books in his lifetime, but none left as large a mark on the medical community as his 1868 tome, "Das Verhalten der Eigenwärme in Krankheiten," or "The Course of Temperature in Diseases." In it, he analyzed a dataset so massive that it has never been equaled before or since: several million temperature records of around 25,000 patients. His reported analysis of this huge swath of data resulted in this historic statement: "... when the organism is in a normal condition, the general temperature of the body maintains itself at the physiologic point: 37°C = 98.6°F."

Wunderlich also concluded that the upper limit of the normal body temperature was 100.4 degrees Fahrenheit (38 degrees Celsius), and anything above that counted as fever. If you've ever wondered where we got 98.6 degrees as the normal temperature of the human body and 100.4 degrees as fever, this is it. Every check of a thermometer, touch of a burning forehead, and prescription of fever-reducing drugs hearkens back to this 1868 book.

Wunderlich's findings were so celebrated — and his dataset so massive — that hardly anyone thought to test their validity until around a century later. When they did, the numbers turned out to be wrong.

Cool Reception

The first modern volley against Wunderlich's rule of thumb came in 1950 when three researchers from the University of Pennsylvania made their own observations of 54 people. The average oral temperature they got was 97.8 degrees Fahrenheit (36.5 degrees Celsius) in the morning and 98.2 degrees Fahrenheit (36.8 degrees Celsius) in the evening. These lower temperatures were even more puzzling, considering that Wunderlich's temperatures were taken from the armpit, which generally runs cooler than the mouth.

Decades later for a study published in JAMA in 1992, physician Philip Mackowiak and his team at the University of Maryland took the oral temperatures of 148 people multiple times a day to get a total of 700 temperature recordings. They arrived at an average temperature of 98.2 degrees Fahrenheit (37 degrees Celsius) and an upper limit of 99.9 degrees Fahrenheit (37.7 degrees Celsius). With their results and the results of other researchers, the team concluded that the old 98.6-degree rule of thumb should be abandoned entirely, and that "several of Wunderlich's other cherished dictums should be revised."

With so much data, how was Wunderlich so wrong? Well, for one thing, it's nearly impossible that he analyzed all of the data he collected, much less with proper statistics. "Wunderlich's lack of access to computer technology makes it inconceivable that he could have analyzed more than a small fraction of the total data set," wrote Mackowiak in a 1994 article published in Clinical Infectious Diseases. "In addition, although the principles of statistical analysis were known as early as the 1830s, these did not find their way into general use until the 1890s. There is no evidence that Wunderlich had knowledge of such principles."

Same goes for his thermometer: He took his temperature readings before the thermometer scale was standardized, something that only happened a decade after his death. In fact, Mackowiak actually ran tests with Wunderlich's thermometer, which was on display at the Mutter Museum in Philadelphia. It registered temperatures 2.9 to 3.2 degrees Fahrenheit (1.6 to 1.8 degrees Celsius) higher than the digital thermometer he compared it to, and even higher than that when compared to readings from other thermometers of the same era.

This isn't to say that Wunderlich was a bad scientist; just that even the most respected science still needs to be double-checked. Wunderlich changed medicine for the better by establishing fever as a symptom of a disease, not a disease in its own right. (Although, as Mackowiak noted, a year before his death came the advent of fever-reducing drugs, a fact that totally disregarded the scientist's most important lesson).

Despite its lack of evidence, the definition of normal body temperature as 98.6 degrees refuses to die. But when it comes to your own health, don't worry if your "normal" temperature registers a degree or so below 98.6, and go ahead and take precautions if you're feeling feverish without a 100.4-degree thermometer reading (the CDC, after all, says it counts as a fever if someone "feels warm to the touch or gives a history of feeling feverish"). And above all, remember that science is only valuable with a system of checks and balances. No result is above a second look.

Get stories like this one in your inbox or your headphones: Sign up for our daily email and subscribe to the Curiosity Daily podcast.

History is full of medical errors — and some of them are downright hilarious. Check out "Quackery: A Brief History of the Worst Ways to Cure Everything" by Lydia Kang and Nate Pedersen for more. The audiobook is free with an Audible trial. We handpick reading recommendations we think you may like. If you choose to make a purchase, Curiosity will get a share of the sale.

Written by Ashley Hamer July 12, 2019

Curiosity uses cookies to improve site performance, for analytics and for advertising. By continuing to use our site, you accept our use of cookies, our Privacy Policy and Terms of Use.