Temperature Measurement at Well-Child Visits in the United States

Objective To determine the frequency and predictors of temperature measurement at well-child visits in the US and report rates of interventions associated with visits at which temperature is measured and fever is detected. Study design In this cross-sectional study, we analyzed 22 518 sampled well-child visits from the National Ambulatory Medical Care Survey between 2003 and 2015. We estimated the frequency of temperature measurement and performed multivariable regression to identify patient, provider/clinic, and seasonal factors associated with the practice. We described rates of interventions (complete blood count, radiograph, urinalysis, antibiotic prescription, and emergency department/hospital referral) by measurement and fever (temperature >= 100.4 degrees F, >= 38.0 degrees C) status. Results Temperature was measured in 48.5% (95% CI 45.6-51.4) of well-child visits. Measurement was more common during visits by nonpediatric providers (aOR 2.0, 95% CI 1.6-2.5; reference: pediatricians), in Hispanic (aOR 1.9, 95% CI 1.6-2.3) and Black (aOR 1.5, 95% CI 1.2-1.9; reference: non-Hispanic White) patients, and in patients with government (aOR 2.0, 95% CI 1.7-2.4; reference: private) insurance. Interventions were more commonly pursued when temperature was measured (aOR 1.3, 95% CI 1.1-1.6) and fever was detected (aOR 3.8, 95% CI 1.5-9.4). Conclusions Temperature was measured in nearly one-half of all well-child visits. Interventions were more common when temperature was measured and fever was detected. The value of routine temperature measurement during well-child visits warrants further evaluation.