Importance: Although oral temperature is commonly assessed in medical examinations, the range of usual or “normal” temperature is poorly defined.

Objective: To determine normal oral temperature ranges by age, sex, height, weight and time of day.

Design: We applied a filtering algorithm (LIMIT) to 10 years of outpatient temperature measurements. LIMIT iteratively removed encounters with primary diagnoses overrepresented in the tails of the temperature distribution, leaving only those diagnoses unrelated to temperature. Mixed effects modeling was applied to the remaining temperature measurements to identify independent predictors of normal oral temperature and to generate individualized normal temperature ranges.

Setting: Single large medical care system, divisions of Internal Medicine and Family Medicine.

Participants: All adult outpatient encounters that included temperature measurements, April 2008 - June 2017.

Exposures: Primary diagnoses and medications, age, sex, height, weight, time of day and month, abstracted from each outpatient encounter.

Main outcomes and measures: Normal temperature ranges by age, sex, height, weight, and time of day.

Results: From 618,306 encounters, 36% were removed by LIMIT because they included diagnoses or medications that fell disproportionately in the tails of the temperature distribution. The encounters removed due to overrepresentation in the upper tail were primarily linked to infectious diseases (76.81% of all removed encounters); type 2 diabetes mellitus was the only diagnosis removed for over-representation in the lower tail (15.71% of all removed encounters). Prior to running LIMIT, the mean overall oral temperature was 36.71°C (±0.43); following LIMIT, the mean temperature was 36.64°C (±0.35). Using mixed effects modeling, age, sex, height, weight and time of day accounted for 6.86% (overall) and up to 25.52% (per subject) of the observed variability in temperature. Mean normal oral temperature did not reach 37°C for any subgroup; the upper 99th percentile ranged from 36.8°C (tall underweight old men in the morning) to 37.9°C (short obese young women in the afternoon).

Conclusion and relevance: Normal oral temperature varies in a predictable manner based on sex, age, height, weight and time of day, allowing individualized normal temperature ranges to be established. 

From the Stanford Research Repository (STARR, also known as STRIDE), we identified all adult outpatient encounters that included temperature measurements from April 2008 through June 2017 in the Divisions of Internal Medicine and Family Medicine within Stanford Health Care (Stanford, CA). Oral temperature, the date and time of the temperature measurement, age, sex, weight, height, body mass index (BMI), primary reason for the visit, prescribed medications, and all visit ICD-10 codes were identified from each encounter. An individual patient could have multiple encounters.

After identifying ineligible encounters in STARR, we applied a filtering algorithm, LIMIT, which iteratively removed encounters with primary diagnoses overrepresented in the tails of the temperature distribution, leaving only those diagnoses unrelated to temperature.

Three subsets are specified:  (1) "Baseline" or STARR/STRIDE encounters (those excluded from consideration by LIMIT due to extreme or missing variables, or highly skewed values (temperatures less than 34°C (93.2°F) or over 40°C (104°F), age less than 20 or over 80 years, BMI less than 10 or over 50, height less than 1.37m (4’6”) or greater than 2.13m (6’10”), weight less than 27.22 kg (60.0lbs) or greater than 181.44 kg (400lbs)). (2) The "outlier" set of encounters removed by LIMIT. (3) The "analysis" set of encounters with usual or "normal" temperature values as determined by LIMIT.

Three files describing code are included with the data and the README file: Section A SAS; Section B Temperature modeling (in R); Section C LIMIT (in R). The dataset anon_AnalysisDataV2.csv includes a reduced set of variables to meet the guidelines for properly anonymized human subject data. Empty cells are intentional, representing no data in the modified dataset, due either to values removed for anonymization purposes or to no actual value.

The datafile is a compressed CSV file.