The reference temperature of 98.6°F (37°C) inadequately describes the variation in normal body temperatures, an observational study found.
After controlling for factors like age and race, every unexplained 0.149°C increase in body temperature was linked to 8.4% relative greater odds of mortality within the year, or an absolute increase of 0.52% (P=0.014) among patients seen at outpatient clinics of a large academic hospital in 2009-2014 (n=35,488).
The average temperature of this study group made up of individuals who were expected to have normal body temperatures, having neither a diagnosis for an infection nor a prescription for antibiotics at the time, was actually 97.9°F (90% range 96.3° to 99.1°F), reported Ziad Obermeyer, MD, MPhil, of Brigham and Women’s Hospital in Boston, and colleagues online in The BMJ.
Factors like age and race only accounted for 8.2% of individual temperature variation, the authors said.
“This fits with a larger body of research showing that reduction in body temperature increases longevity and delays aging in a range of (ectothermic) experimental models, including Drosophila and Caenorhabditis elegans, as well as transgenic (homeothermic) mice engineered to have lower temperatures,” according to Obermeyer’s group.
As for humans, they wrote, “[m]arried couples sitting next to each other in the same room routinely disagree about whether to turn the heat up or down. Individuals have different baseline propensities to feel hot or cold — at any given absolute room temperature. Yet doctors have forgotten this lesson when measuring core body temperature. We would not use absolute room temperature to infer perceived warmth, but we do use absolute body temperature to infer fever.”
In an accompanying editorial, Philip Mackowiak, MD, and Daniel Morgan, MD, MS, both of Baltimore’s University of Maryland School of Medicine, suggested that the 98.6°F used as reference today is of more historical than clinical significance. They recalled that this number came about in 1868, when Carl Wunderlich established 37°C as the reference temperature for humans after studying 25,000 subjects — with imperfect thermometers.
“In healthy young adults at least, 37°C is not the overall daily mean temperature, the mean temperature at any particular time of day, or the most commonly recorded temperature,” according to Mackowiak and Morgan.
The dataset used by Obermeyer and colleagues showed that body temperature differs by demographics, older people being the coldest (-0.021°C for every decade, P<0.001) and African American women the hottest (+0.052°C over white men, P<0.001). Every 1 m/kg2 in BMI was also associated with 0.002°C temperature rise (P<0.001). Additionally, comorbidities had varying relationships with body temperature: hypothyroidism was tied to a 0.013°C drop (P=0.01) and cancer a 0.020°C increase (P<0.001).
Some of this was old news to the editorialists.
“Obermeyer and colleagues used a host of statistical techniques to try to minimize the distorting effects of unmeasured variables, but much of what they discovered simply corroborated Wunderlich’s less sophisticated observations nearly a century and a half ago. Their most provocative finding — that ‘temperature correlates with mortality’ — is unconfirmed, though intriguing enough to merit further study,” the pair commented.
Study participants had a mean age of 52.9 years and were 64% women and 41% non-white. Their temperatures were taken at varying anatomical sites and with different thermometers.
Caveats to the study were that records came from one academic center and one climate zone. It is also possible that some patients had undiagnosed infections at the time they had their temperature taken.
Obermeyer, Mackowiak, and Morgan listed no relevant ties to industry.