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.
翻译:许多试验的目的是在随机化之后的预定时间或前后收集结果,实际上,实际评估参与者的时间可能有很大的变异,因此,治疗效果的驱动因素可能是评估的时机,而不是治疗对参与者基本结果轨迹的影响,解决这个问题的方法之一是集中注意每个(固定的)定点评估时间的治疗效果,为此,需要无法验证的假设,因此,通过敏感性分析,评估偏离这些假设的推论将如何改变。我们在此制定这种敏感性分析方法,同时采用半参数、基于影响功能的估算方法。我们运用我们的方法研究患有无节制哮喘的低收入参与者,并在现实的模拟研究中评估我们程序的表现。