Identification of drug-related problems by a clinical pharmacist in addition to computerized alerts.

BACKGROUND: Both clinical pharmacists and computerized physician order entry systems with clinical decision support (CPOE/CDSS) can reduce drug-related problems (DRPs). However, the contribution of a clinical pharmacist in addition to CPOE/CDSS has not been established in a prospective study.

OBJECTIVE: To determine which DRPs can be identified by a clinical pharmacist in a setting with routine use of CPOE/CDSS.

SETTING: Two surgical and two neurological wards in St. Elisabeth hospital, a 600-bed teaching hospital in the Netherlands.

METHODS: In this observational prospective follow-up study a clinical pharmacist reviewed the pharmacotherapy of patients admitted to surgical and neurological wards to identify DRPs (i.e. medication errors and adverse drug events) and discussed the relevance of identified problems and interventions to resolve these with the responsible physician. Acceptance of the proposed interventions and the presence of alerts in CPOE/CDSS were assessed. Primary outcome was the proportion of DRPs identified by the clinical pharmacist that also triggered a CPOE/CDSS alert. Differences between the DRPs that generated an alert and those that did not were expressed as relative risks or analyzed with Chi square statistics or Mann-Whitney U tests.

MAIN OUTCOME MEASURE: The proportion of drug-related problems identified by the clinical pharmacist that also generated an alert in the CPOE/CDSS.

RESULTS: During 1206 medication reviews, 442 potential DRPs were identified; 286 (65 %) DRPs were considered relevant and 247 (56 %) of the proposed interventions were accepted. A CPOE/CDSS alert was generated for 35 (8 %) of the DRPs the clinical pharmacist identified. The only difference between problems that triggered an alert and those that did not was the class of the DRP (indication 23 vs. 36 %, effectiveness 23 vs. 13 %, safety 23 vs. 10 % and pharmaceutical care issues 31 vs. 42 %, p = 0.02). CPOE/CDSS triggered 623 additional alerts that were handled during routine pharmacy service.

CONCLUSIONS: As most DRPs identified by a clinical pharmacist were not detected in daily clinical practice by CPOE/CDSS, a clinical pharmacist contributes to reducing DRPs. The sensitivity of CPOE/CDSS to detect certain classes of problems should be optimized.