Changing professional behaviour: an updated overview of systematic reviews

Article type
Authors
Grimshaw J, Mayhew A, Stevens A, Graham S
Abstract
Background: The Cochrane Effective Practice and Organisation of Care Group (EPOC) supports systematic reviews of professional, organizational, financial and regulatory interventions to improve healthcare delivery and care systems. It has conducted two previous overviews of systematic reviews of professional behaviour change interventions published in 1998 (n=18) and 2002 (n=41). We are currently updating these overviews and have identified over 150 potentially relevant reviews.
Objectives: To collect and summarize, in an overview, systematic reviews of professional behaviour change.
Methods: Systematic reviews published between 1966 and 2006 were identified from Pubmed and the Cochrane Library by EPOC. Two review authors independently abstracted data on the quality and results of the reviews. Vote counting was used as the common metric for data synthesis. Interventions were classified as effective if more than two thirds of studies showed benefit, mixed effects if between one third and two thirds showed benefit and ineffective if less than one third demonstrated benefit. In addition, we present the median absolute improvement in process of care measures (for example compliance with guidelines) if possible. The results are available in a searchable website at: http://www.cadth.ca/index.php/en/compus/optimal-ther-resources/interventions.
Results: Over 150 potentially relevant reviews were identified and the results of 50 key reviews were analysed for this project. Generally effective interventions included: printed educational materials (median absolute improvement +4%), interactive educational meetings, educational outreach (+5%), local opinion leaders (+10%), audit and feedback (+10%). The results varied considerably within intervention categories. Multifaceted interventions were not more effective than single interventions.
Conclusions: There are a number of interventions that are generally effective for changing professional behaviour. The local applicability of these findings should be informed by considerations of the potential barriers, mechanisms of action and resources available.