In Reply.—The idea proposed by Dr Baker of using fever in a family member as a predictor for the treatment of febrile infants is noteworthy. My colleagues and I had planned to use illness or fever in a family member or close contact as an independent variable in our study. Unfortunately, its presence or absence was not consistently recorded in the charts. Retrospective clinical studies can provide some useful information but were hampered by the limitations of the medical record in this case.
The data shared by Dr Baker are interesting and lend preliminary support for testing the reliability of family illness or fever as a predictor of benign illness in febrile infants. It may well prove to be a valuable additional factor in treating a problem that continues to perplex clinicians.