Changes in vital signs as predictors of bacterial infection in home care: a multi-center prospective cohort study

Objectives: To explore whether the combination of changes in heart rate and body temperature can predict bacterial infection in home care patients. Methods: This multicenter, prospective cohort study was conducted in Japan from March 2012 through December 2013 and involved three clinics. The study population comprised all patients who received regular home visit services for at least 3 months and met one of the following inclusion criteria: 1) fever over 37.5 degrees C at home visit, 2) physician's clinical suspicion of fever, or 3) physician's suspicion of bacterial infection. We collected temperature and heart rate data on the day of enrollment, and determined the probable causes of fever after treatment of febrile episodes. We defined the combination of changes in heart rate and body temperature as delta HR/BT. We calculated two types of delta HR/BT, averaged and assumed, using different baseline values for heart rate and body temperature. Results: A total of 124 patients were enrolled and 194 episodes of fever were analyzed during the study period. The sensitivity, specificity, positive predictive value, and negative predictive value for the average delta HR/BT with a cut-off >= 20 were 20.4% (95% CI, 16.7-23.3), 84.2% (95% CI, 75.2-91.0), 75.7% (95% CI, 61.8-86.2), and 30.6% (95% CI, 27.3-33.0), respectively. The sensitivity, specificity, positive predictive value, and negative predictive value for the assumed delta HR/BT with a cut-off = 5 was 91.2% (95% CI, 89.2-94.0), 8.9% (95% CI, 4.1-15.7), 70.9% (95% CI, 69.3-73.0), and 29.4% (95% CI, 13.6-51.8), respectively. Conclusions: The combination of changes in heart rate and body temperature could help physicians determine whether home care patients have bacterial infections.