Computerized physician order entry system combined with on-ward pharmacist: analysis of pharmacists' interventions.

Rationale, Aims And Objectives: To analyse pharmacists' interventions in a setting where a computerized physician order entry system (CPOE) is in use and a pharmacist works on the ward.

Method: A prospective cohort study was conducted in seven wards of a French teaching hospital using CPOE along with the presence of a full-time on-ward pharmacy resident. We documented the characteristics of pharmacists' interventions communicated to physicians during the medication order validation process whenever a drug-related problem was identified. Independent predictors of the physician's acceptance of the pharmacist's intervention were assessed using multiple logistic regression analysis.

Results: The 448 pharmacists' interventions concerned: non-conformity to guidelines or contraindications (22%), too high doses (19%), drug interactions (15%) and improper administration (15%). The interventions consisted of changes in drug choice (41%), dose adjustment (23%), drug monitoring (19%) and optimization of administration (17%). Interventions were communicated via the CPOE in 57% of cases and 43% orally. The rate of physicians' acceptance was 79.2%. In multivariate analysis, acceptance was significantly associated with the physician's status [higher for residents vs. seniors: OR = 7.23, CI 95 (2.37-22.10), P < 0.01], method of communication [higher for oral vs. computer communication: OR = 12.5, CI 95 (4.16-37.57), P < 0.01] and type of recommendation [higher for drug monitoring vs. drug choice recommendations: OR = 10.32, CI 95 (3.20-33.29), P < 0.01].

Conclusions: When a clinical pharmacist is present on a ward in which a CPOE is in use, the pharmacists' interventions are well accepted by physicians. Specific predictors of the acceptance by physicians emerge, but further research as to the impact of CPOE on pharmacist-physician communication is needed.