PII‐16

BACKGROUND Problems with the dissemination of guidelines are frequently cited as a major reason for failure to influence clinical practice. We report the results of a time‐series impact study of online antibiotic (AB) guidelines on the quality of prescribing for pneumonia in a teaching psychiatric hospital comprising 410 adult beds. METHODS The intervention consisted of embedding our current AB guidelines in our computerized physician drug‐order entry (CPOE) system. In total, 126 consecutive AB orders for pneumonia (63 before and 63 after the intervention) were analysed by the pharmacy department for appropriateness relative to the guidelines. The 3 evaluation criteria used were: choice of the AB relative to the mode of acquisition of pneumonia (community‐ or hospital‐acquired), daily dosage, and duration of treatment. RESULTS The post‐ and pre‐intervention groups of patients were similar for mode of acquisition of pneumonia. The rate of AB orders containing at least 1 reason of inappropriateness tended to decrease after vs. before the intervention: 36.5% vs. 47.6% (p=0.2). The total number of reasons of inappropriateness was significantly lower, by 22.2% (95% CI 5.2–39.2; p=0.012), after vs. before the intervention: 26/63 (41.3%) vs. 40/63 (63.5%), with a trend towards a better adherence to the guidelines for the 3 evaluation criteria used. CONCLUSION Providing therapeutic information at the time of drug order appeared to be moderately contributive to the quality of AB prescribing in our study. Clinical Pharmacology & Therapeutics (2005) 79 , P39–P39; doi: 10.1016/j.clpt.2005.12.141