July 14, 2010

Bias and Trials Stopped Early for Benefit

JAMA. 2010;304(2):156-159. doi:10.1001/jama.2010.933

To the Editor: The apparent purpose of the analysis in the review by Dr Bassler and colleagues1 was to argue that trials terminated early will overestimate treatment benefit, thereby misleading physicians and patients. It is true that a trial terminated early for benefit will tend to overestimate true effect; this happens because there always is variability in estimation of true effect, and when assessing data over time, evidence of extreme benefit is more likely obtained at times when the data provide a random overestimate of truth. However, statistical methods are available to adjust for this random-high bias,25 although they are not applied as often as they should be.

First Page Preview View Large
First page PDF preview
First page PDF preview