Morphine is a preferred narcotic since meperidine forms toxic metabolites. Determinants of meperidine use have been poorly described. The objective of this study is to explore factors associated with the ordering of meperidine versus morphine. Retrospective chart review of adult patients, randomly selected based on orders for morphine or meperidine. 1552 orders were written for 670 patients. Of these, 36% were for meperidine. In multivariable analysis, the ordering of meperidine was associated with the following variables in decreasing order of importance: physician specialty, total doses received, hospital location, patient race, age and insurance, and physician gender. More orders for meperidine were written for those receiving fewer doses. Though meperidine has little role in the routine management of hospital pain, we found it continues to be used frequently. Importantly, meperidine is ordered more frequently for patients who receive shorter courses of narcotics. Our study suggests that interventions targeted at more appropriate use of meperidine rather than complete elimination might be more acceptable to physicians while minimizing the risk of toxicity.