Stephen J.LurieMD, PhD, Senior Editor
To the Editor: Dr Balk and colleagues1 found no consistent relationship between the size of treatment effects in RCTs and their methodological quality. However, previous studies found that the lack of an allocation system that prevents the investigator from knowing in advance which treatment the next person will get was consistently associated with larger treatment effects.2 Proper concealment of the allocation sequence thus appears to be crucial to prevent selection bias in RCTs.3 Balk et al argue that such quality effects may not be as important as previously thought.
Jüni P, Egger M. Allocation Concealment in Clinical Trials. JAMA. 2002;288(19):2407. doi:10.1001/jama.288.19.2407-JLT1120-4-2