Assessment of clinical pharmacists' interventions in French hospitals: results of a multicenter study.

Background: The development of clinical pharmacy activities in most European countries is underway; however, data on these activities are still poorly reported. Multicenter studies are necessary to standardize and demonstrate the value of clinical pharmacy activities in these countries.

Objective: To document clinical pharmacists' daily routine interventions (PIs) to identify trends of intervention, drugs, and situations most frequently associated with drug-related problems (DRPs) and to estimate physicians' acceptance of PI.

Methods: A prospective study of PIs was conducted in 6 French hospitals. The sample consisted of 300 randomized PIs per hospital, recorded during the medication order validation process when a DRP was identified. We recorded patients' demographic characteristics, drugs involved, wards, DRP description, pharmacists' recommendations, and whether or not the recommendations were accepted by the physicians.

Results: A total of 38,626 medication orders were analyzed by 28 clinical pharmacists, leading to 1800 PIs (4.66 PIs per 100 medication orders). Of the 1800 PIs, 25.9% targeted psychotropic drugs, 15.9% targeted antithrombotic drugs, 15.5% targeted digestive and metabolic drugs, and 15.0% targeted cardiovascular drugs. The most commonly identified DRPs were nonconformity to guidelines or contraindication (21.3%), followed by improper administration (20.6%), supratherapeutic dose (19.2%), and drug interaction (12.6%). Nearly half (42.2%) of the pharmacists' recommendations were related to drug choice (drug switch 22.2%, drug discontinuation 16.3%, addition of a new drug 3.7%) followed by dose adjustment (23.8%), optimization of administration (21.9%; change of administration route 10.3%, administration modalities 11.6%), and need for drug monitoring (12.2%). The rate of physicians' acceptance was 73.4% (15.3% refusals, 11.3% not assessable).

Conclusions: In French hospitals, pharmacists contribute to preventing DRPs during medication order validation. This study suggests that a few types of drugs and errors constitute a substantial proportion of PIs. Knowledge of the most frequent DRPs could significantly increase the efficiency of clinical PIs.