Background: Undesirable practice variation remains a major concern with the quality of the healthcare system. While care in pediatric hospitalist systems has been demonstrated to be efficient, neither the quality of care nor determinants of variation in pediatric hospitalist systems are well understood.
Objective: To measure variation in pediatric hospitalists' reported use of common inpatient therapies, and to test the hypothesis that variation in reported use of proven therapies is lower than variation in reported use of unproven therapies.
Design and measurements: We conducted a survey of pediatric hospitalists in the US and Canada. Respondents reported their frequency of using 14 therapies in the management of common conditions. Each therapy was determined to be of proven or unproven effectiveness using published critical appraisals. Variation in reported use of proven and unproven therapies was compared.
Results: 67% (213/320) of surveyed individuals participated. Little variability existed in reported use of albuterol and corticosteroids in asthma (4-6% of respondents reported not often using them) and systemic dexamethasone in bronchiolitis (12% of respondents reported using it more than rarely). Moderate to high variation existed in reported use of all other therapies studied. Variation in reported use of proven therapies was significantly less than variation in reported use of unproven therapies (15.5 +/- 12.5% vs. 44.6 +/- 20.5%).
Conclusions: Substantial variation exists in hospitalists' reported management of common pediatric conditions. Variation is significantly lower for strongly evidence-based therapies. To decrease undesirable variation in care, a stronger evidence base for inpatient pediatric care must be built.
Copyright 2008 Society of Hospital Medicine.