Data monitoring committees for multicenter clinical trials sponsored by the National Institutes of Health. I. Roles and membership of data monitoring committees for trials sponsored by the National Eye Institute.
The operational aspects of data monitoring committees for multicenter clinical trials have received little attention in the published literature, in contrast to a wealth of publications concerning the statistical aspects of data monitoring. Occasional anecdotal reports suggest that the role of these committees is complex and that responsibilities and decisions extend beyond choice of a stopping rule. In an effort to gain a better appreciation of the domain of responsibilities shared by all or most of the data monitoring committees (DMCs) for the 20 multicenter studies (30 clinical trials) sponsored by the National Eye Institute, one of the National Institutes of Health, published information was studied and key members of the DMCs for these studies were surveyed. Information about the membership of these important groups also was sought in order to guide those with responsibility for appointing future DMCs. Three tasks--all directly associated with monitoring accumulating data for evidence of benefit or harm to trial patients--were reported unanimously to be the responsibility of the DMCs. In addition, many responsibilities concerned with study oversight rather than data monitoring were assigned to the DMC in 75% or more of the trials. The domain of responsibilities common to a majority of the DMCs suggests that these groups function in most NEI-sponsored multicenter trials as combined data monitoring and policy boards, even when not so designated. The median size of the 20 DMCs was 10 members, with six appointed and four ex officio members. Ophthalmologists accounted for 38% of all DMC positions and statisticians for 33%.