Conflict of interest in psychiatry

Abstract

Aims and method: To study the association between study support and outcome in randomised controlled trials (RCTs) of psychotropic drugs, we reviewed all RCTs published in four psychiatry journals over a 5-year period. Chi-squared tests were used to analyse the association between RCT support and outcome, and logistic regression to determine which variable best predicted outcome. Results: A significantly higher proportion of manufacturer-supported RCTs (125/138, 91%, 95% CI 88-93) had a positive outcome than nonmanufacturer-supported RCTs (39/50, 78%, 95% CI 72-84; P=0.02). Having an employee author almost guaranteed a positive outcome (56/58, 97%, 95% CI 94-99). Clinical implications: Outcomes of drug RCTs have a significant association with support by the manufacturer of the experimental drug. Systematic reviews and meta-analyses based on these RCTs may be biased in favour of newer drugs.