Objective
There is a perception that many drug usage evaluations do not widely influence prescribing behaviour. The aim of this study was to critically evaluate recent journal articles which fit the Medline definition for Drug Utilization Review (DUR) and which also cover multiple healthcare sites.
Methods
PubMed (National Library of Medicine, NLM) (2003, 2004) was searched using the MeSH topic ‘drug utilization’. Retrieved studies were evaluated to ascertain those describing a DUR (measuring drug use against specific criteria). These were subdivided according to whether the DUR was conducted at one site or across many. The multi-centre DURs were critically reviewed, including evaluating whether all phases of a quality cycle were completed and determining aspects of design such as whether the study was prospective or retrospective, any interventions conducted and provision of feedback.
Results
A total of 646 unique articles were retrieved. Of these, 495 (77%) did not meet the definition for DUR, while 151 (23%) articles did. Thirty-five (5%) described English language multi-centre DURs; ethics approval was obtained in ten of these and 18 were carried out retrospectively. In all 35 studies some comparator or standard was used, but only eight conducted an intervention and only three provided feedback to the prescribers.
Conclusion
Most DURs were not conducted across a number of centres. Of the recent published multi-centre DURs most presented only an initial audit and did not complete the quality cycle with feedback, intervention and re-audit. To widely influence prescribing behaviour, the full cycle is required with involvement of as many sites as possible to achieve improvements across many jurisdictions.