Initiation of symptomatic medication in Alzheimer's disease clinical trials: Hypothetical versus treatment policy approach

In clinical trials in populations with mild cognitive impairment, it is common for participants to initiate concurrent symptomatic medications for Alzheimer's disease after randomization to the experimental therapy. One strategy for addressing this occurrence is to exclude any observations that occur after the concurrent medication is initiated. The rationale for this approach is that these observations might reflect a symptomatic benefit of the concurrent medication that would adversely bias efficacy estimates for an effective experimental therapy. We interrogate the assumptions underlying such an approach by estimating the effect of newly prescribed concurrent medications in an observational study, the Alzheimer's Disease Neuroimaging Initiative.