Morphine use in hospitalized children in the United States: a descriptive analysis of data from pediatric hospitalizations in 2008

Abstract

Background: Morphine is among the top 10 medications given to children in the inpatient setting. It is not labeled for any pediatric indication, making it one of the drugs most widely used off-label in pediatrics.

Objectives: The aims of this study were to describe the epidemiology of morphine use in pediatric inpatients in the United States and to describe the characteristics of patients and hospitals in hospitalizations with morphine use.

Methods: Deidentified data from the Premier Perspective Database (2008) were analyzed. Morphine use was defined as any morphine administration during the hospital stay and estimated by patient age in years, sex, race, and type of insurance; and hospital bedsize, teaching status, setting (urban or rural), and geographic location. Proportions (95% CI) were calculated for the entire population and for individual strata. The estimate was applied to national data to calculate the number of pediatric hospitalizations with morphine use in the United States in 2008. Logistic mixed-effects modeling was used to calculate the probability of morphine use by hospital after controlling for hospital and patient effects.

Results: The database contained records from 877,201 pediatric hospitalizations and 423 hospitals in the United States. Morphine was administered in 54,613 of pediatric hospitalizations (6.2%). Use was higher in boys than girls (6.4% and 6.1%, respectively) and in blacks compared with whites or other racial groups (7.5%, 6.7%, and 5.0%). Use increased from 1.6% in children aged <2 years to 27.4% in those aged 12 to 17 years. Based on these data, morphine may have been administered in 476,205 pediatric hospitalizations in the United States in 2008. The 2 diagnoses most frequently associated with morphine use were appendicitis (14.4%) and fracture (11.1%). On logistic mixed-effects modeling for patients with appendicitis and for patients with fractures, there was hospital variation in morphine use after controlling for variables in the model.

Conclusions: Based on the data from this analysis, morphine was used in hospitalized children in all age groups, despite the lack of pediatric labeling. Common conditions such as appendicitis and fracture were leading diagnoses associated with morphine use.