Never before have multicenter trials been so widespread, nor so contentious. Barnett points to the remarkable progress in stroke prevention, and argues that it largely flows from the results of multicenter trials. Marshall agrees that clinical trials rather than clinical impressions are the appropriate method of evaluating therapy but maintains that one of the basic tenets of a controlled trial, ie, precise definition, is seldom adhered to because of the multiple causes and clinical diversity of stroke. Marshall also questions the ability of randomization to eliminate the effects of variables that cannot be controlled or even recognized. He advocates small trials of well-defined pathophysiological categories.
Small trials are feasible and desirable in conditions that have a high mortality or a high probability of a clinically meaningful event. The much-touted first modern clinical trial of streptomycin in primary pulmonary tuberculosis may have been obviated if streptomycin had been used initially in
Hachinski V. Multicenter Trials in Stroke. Arch Neurol. 1990;47(4):445. doi:10.1001/archneur.1990.00530040101025
Customize your JAMA Network experience by selecting one or more topics from the list below.