Answers are grounded in this paper and cite their sources.
Loss to follow-up is often hard to avoid in randomised trials. This article suggests a framework for intention to treat analysis that depends on making plausible assumptions about the missing data and including all participants in sensitivity analyses
No takes yet. Share an insight, caveat, or question.
White et al. (2011) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: