Drug selection in French university hospitals: analysis of formularies for nine competitive pharmacological classes

WHAT IS ALREADY KNOWN ABOUT THIS SUBJECT • Variations in hospital drug formularies have been reported in Germany and the Netherlands for competitive classes. • Since a spill-over effect on the community may be expected, assessment of hospital formularies is needed. • No study has yet explored French hospitals' drug formularies. WHAT THIS STUDY ADDS • Large variations in the number and the nature of selected drugs were found among French university hospitals for competitive classes. • The wide range in the number of selected drugs suggests an unaccomplished rationalization process by some hospitals. AIM To give a panorama of the selectivity and agreement of French university hospitals' drug formularies (HDF) for nine competitive classes. METHODS All university hospitals were asked to send their HDF and selection criteria as of January 2009 for nine competitive pharmacological classes (proton pump inhibitors, serotonin antagonists, low molecular weight heparins, erythropoietins, angiotensin converting enzyme inhibitors, angiotensin II receptor antagonists, statins, α-adrenoreceptor antagonists and selective serotonin re-uptake inhibitors). Selectivity of HDF was estimated by the percentage of drug entities selected by the hospital within the pharmacological class. Agreement between hospitals was assessed with modified kappa coefficients for multi-raters. RESULTS Twenty-one out of the 29 hospitals agreed to participate. These hospitals selected between 34% and 63% of the drug entities available for the nine classes, which represented 18 to 35 agents. Regarding the nature of chosen drug entities, the overall level of agreement was ‘fair’ and varied with pharmacological classes. Selection criteria were sent by only 12 hospitals. The technical component was the most important element in all hospitals. The weight of the economic component varied between 20% and 40% in the tender's grade. DISCUSSION Large variations were seen in the number and nature of drugs selected by university hospitals which can be attributable to two successive decision-making processes (evaluation by the Drug and Therapeutics Committee followed by the purchasing process).