In this paper the methodological problems that might arise in a randomized clinical trial carried out in primary care and involving a complex intervention aiming at behavioural change, are discussed. The design of a clinical trial carried out among patients with irritable bowel syndrome is briefly described, in which the intervention has been applied by general practitioners. Part of the standard criteria developed by the Cochrane Collaboration on Effective Professional Practice Working Group to assess the methodological quality of (randomized) clinical trials included in their reviews are used as a framework for discussion. Problems that might arise when trying to meet these criteria are discussed, both in general and in relation to study design. Possible solutions are discussed. Items that require special attention in the case of a trial with a complex intervention carried out by the care-providers in daily practice, are (1) random allocation at the level of the (group of) care-providers; (2) blind or objective assessment of primary outcome measurements and (3) the prevention of contamination. We also discuss three additional problems, namely the adequacy of the control condition, the standardization of the intervention and effectiveness versus efficacy. Alternative designs for studies with complex interventions to be applied by care-providers are discussed.
|Number of pages||6|
|Journal||Huisarts en Wetenschap|
|Publication status||Published - 1 Dec 1999|