Assessing risk of bias in randomized controlled trials.

Abstract

Even though randomised controlled trials are the design of choice for evaluating the efficacy of health care interventions, they are not immune to bias that may affect research process and validity of results. In the present paper we discussed how trial quality may be appraised considering both whether a clinical trial is reported in a comprehensive and complete way (consistently with what had been declared in the study protocol), and whether the characteristics of the trial itself are associated with risk of bias.