Semi-Parametric Sensitivity Analysis for Trials with Irregular and Informative Assessment Times

Many trials are designed to collect outcomes at or around pre-specified times after randomization. In practice, there can be substantial variability in the times at which participants are actually assessed. As a result, treatment effects can be driven by the timing of assessments, rather than the effect of treatment on participants' underlying outcome trajectories. One way to deal with this problem is to focus on the treatment effect at each of the (fixed) targeted assessment times. For this, untestable assumptions are needed, and it is therefore important to assess how inferences would change under departures from these assumptions via sensitivity analysis. We develop such a sensitivity analysis methodology here, along with a semi-parametric, influence function-based estimation approach. We apply our method to a study of low-income participants with uncontrolled asthma, and we evaluate the performance of our procedure in a realistic simulation study.