Comprehensive Analysis of Temperature in Hospitalized Patients

Human body temperature is believed to be linked to clinical diagnoses. However, most of the available data stems from healthy individuals, with no large-scale studies addressing body temperature in the inpatient setting, which is the focus of our study.

This is a retrospective analysis of a total of 695,107 temperature readings from 16,245 patients hospitalized over a 1-year period at a tertiary medical center, ages 0-105 years, 50% female, with rectal, monotherm, axillary, oral, temporal and tympanic measurement sites. The average temperature (T_ave) per patient and per measurement site was used in all calculations. Descriptive statistics, Student's t-test, and Pearson's correlation were used, where appropriate, with statistical significance set at P < 0.05.

T_ave from all measurement sites was 98.13 ± 0.48(SD)F(36.74 ± 0.27°C). T_ave varied by the site of measurement, in decreasing order highest-to-lowest being rectal, monotherm, axillary, oral, temporal, and tympanic, all of which were higher than the available reported averages for healthy subjects. T_ave decreased as patients' age increased. There was only slight and likely clinically insignificant difference between the sexes. There were differences in T_ave between the intensive care units (ICUs), listed from highest-to-lowest: Neuro ICU, Pediatric ICU, Surgical ICU, Cardiac ICU and Medical ICU. However, there was no difference between all ICU and non-ICU patients.

Our inpatient data demonstrate that previously identified body temperature trends among healthy subjects are preserved, to an extent, in the inpatient setting. To our knowledge, ours is the first study that evaluates the temperatures of all hospitalized patients at a large tertiary medical center.