A group of Baltimore doctors say the century-old belief that 98.6 degrees constitutes a person's normal body temperature should be discarded because, well, it is wrong.
Researchers from the Baltimore Veterans Affairs Medical Center looked at oral temperatures taken from 148 healthy people and found their average reading to be 98.2 degrees -- two notches below the little red arrow that signals "normal" on a glass thermometer.
More importantly, a reading between 96 and 99.9 degrees would fall into the normal range for most people.
Dr. Philip A. Mackowiak, who headed the study, said the concept of 98.6 as "normal" is so deeply ingrained in American culture that many people conclude they are sick simply because theirtemperature is a few notches above that mark.
"It's almost subconscious, a concept that's worked its way into our thinking," he said.
Dr. Mackowiak said he and Carl Reinhold August Wunderlich, the 19th-century German scientist who established 98.6 as a benchmark, agree on the central facts about body temperature.
For starters, they agree that it fluctuates throughout the day and differs from person to person.
"I don't mean to suggest in any way, based on our observations here, that we should denigrate Wunderlich's contribution to measurements," said Dr. Mackowiak, chief of the medical service at the Baltimore VA. "He knew that temperature [reaches] its nadir at 6 in the morning, and a peak in the afternoon and evening between 4 and 8.
"And he knew that temperature was a very sensitive marker for disease, and taught us that fever was a symptom and not a disease," said Dr. Mackowiak, whose study appears in today's Journal of the American Medical Association.
L But, he said, the German doctor was simply wrong about 98.6.
Dr. Mackowiak, an infectious disease specialist, said he was surprised to discover how little had been written on the topic while doing research on fever. The only large study he could find belonged to Wunderlich, who recorded temperatures of more than 25,000 people over 16 years.
Setting out to do a study of his own, Dr. Mackowiak found a ready-made data bank in the records of the University of Maryland's vaccine development program. There, scientists had recorded 700 oral temperature readings of 148 people who volunteered to take experimental vaccines in the 1980s.
The volunteers were between the ages of 18 and 40 -- an important fact because it means their average temperature cannot be applied to children or the elderly.
