The impact of a drug bulletin on prescribing oral analgesics in a teaching hospital was studied. Using an abbreviated time-series design, prescriptions for oral analgesics for all inpatients were surveyed one, three, and five weeks before and one, three, five, and seven weeks after the distribution of one of the hospital's regular drug bulletins. This bulletin contained guidelines for the treatment of minor, moderate, and severe pain. The 493 patients included in the study were classified by drug category, pain severity, and timer period. The drug categories were: (1) propoxyphene, (2) aspirin or acetaminophen alone or with codeine 8 mg or less, and (3) codeine alone or in combination products with more than codeine 8 mg, or other oral narcotic drugs. Each patient's pain severity was determined by interviewing attendant nursing staff; the validity of this approach was confirmed by correlating nurse and physician perception of pain at one time point. Multivariate contingency table analysis revealed that the drug bulletin significantly (p less than or equal to 0.05) reduced the prescribing of propoxyphene hydrochloride across both the minor and moderate pain categories. An increased use of aspirin and acetaminophen was significant three weeks after the release of the drug bulletin. This effect, however, was not significant at other time points. Drug bulletins can be used to achieve a temporary change in physician prescribing patterns of oral analgesics. For a sustained effect, alternative strategies are needed.