ABSTRACT
Single imputation approaches replace each missing value with a single value based on the sample mean, a regression-based prediction, the last observation carried forward, or other strategies. Clinical trials have been broadly classified as pragmatic or explanatory trials. Schwartz and Lellouch probably first used the terms “explanatory” and “pragmatic” to differentiate trials. The term “explanatory” was used for trials that test whether an intervention works under optimal situations that control heterogeneity as much as possible to isolate the intervention effect. The most distinctive features of pragmatic clinical trials are that they compare clinically relevant alternative interventions and include a broadly diverse population of study participants with few eligibility criteria. Pragmatic and explanatory trials are not distinct concepts since trials may incorporate differing degrees of pragmatic and explanatory components. A trial can be designed to have some aspects that are more pragmatic than exploratory and vice versa.
