• Christopher F Dowrick, John G Hughes, Julia J Hiscock, Mark Wigglesworth, and Thomas J Walley.
• Division of Primary Care, University of Liverpool, Whelan Building, Liverpool, UK. cfd@liverpool.ac.uk
• Bmc Health Serv Res. 2007 Apr 30; 7: 64.

BackgroundSystematic reviews of randomised placebo controlled trials of antidepressant medication show small and decreasing differences between pharmacological and placebo arms. In part this finding may relate to methodological problems with conventional trial designs, including their assumption of additivity between drug and placebo trial arms. Balanced placebo designs, which include elements of deception, may address the additivity question, but pose substantial ethical and pragmatic problems. This study aimed to ascertain views of potential study participants of the ethics and pragmatics of various balanced placebo designs, in order to inform the design of future antidepressant drug trials.MethodsA qualitative approach was employed to explore the perspectives of general practitioners, psychiatrists, and patients with experience of depression. The doctors were chosen via purposive sampling, while patients were recruited through participating general practitioners. Three focus groups and 12 in-depth interviews were conducted. A vignette-based topic guide invited views on three deceptive strategies: post hoc, authorised and minimised deception. The focus groups and interviews were tape-recorded and transcribed. Transcripts were analysed thematically using Framework.ResultsDeception in non-research situations was typically perceived as acceptable within specific parameters. All participants could see the potential utility of introducing deception into trial designs, however views on the acceptability of deception within antidepressant drug trials varied substantially. Authorized deception was the most commonly accepted strategy, though some thought this would reduce the effectiveness of the design because participants would correctly guess the deceptive element. The major issues that affected views about the acceptability of deception studies were the welfare and capacity of patients, practicalities of trial design, and the question of trust.ConclusionThere is a trade-off between pragmatic and ethical responses to the question of whether, and under what circumstances, elements of deception could be introduced into antidepressant drug trials. Ensuring adequate ethical safeguards within balanced placebo designs is likely to diminish their ability to address the crucial issue of additivity. The balanced placebo designs considered in this study are unlikely to be feasible in future trials of antidepressant medication. However there remains an urgent need to improve the quality of antidepressant drug trials.

24 keywords...

hide…