Baseline measurements can be omitted in urethral and anal pressure crossover studies: post hoc analysis of a randomized, placebo-controlled study

Objectives To explore the impact of baseline measurements on variance, the precision of treatment estimate, and sample size in crossover studies using the urethral pressure and anal acoustic reflectometry methodologies. Material and Methods This was a post hoc analysis of a randomized, double‐blind, placebo‐controlled, crossover study of the effect of imipramine on urethral and anal opening pressure. We applied three analysis-of-covariance models that include baseline measurements in the three most common ways and performed sample size calculations for future crossover studies based on the within-subject variance from the three models. Results The model which ignores the baseline measurement provided the lowest variance and thus the highest precision of treatment estimate and the smallest sample size whereas the model that incorporates baseline measurements as a change from baseline analysis provided the largest variance, lowest precision of treatment estimate, and largest sample size estimation. Conclusions Our findings suggest that it is not beneficial to include baseline measurements in crossover studies with urethral pressure and anal acoustic reflectometry.