Objective: The most common way to evaluate the effect of an intervention is to compare the intervention and nonintervention groups regarding the change in the outcome variable between baseline and follow-up; however, there are many different ways to define "changes." The purpose of this article is to demonstrate how different definitions of "change" used in the analysis can influence the results of a study.
Study design and setting: Two different randomized controlled trials were used as examples.
Results: The results of the analyses showed that for continuous outcome variables, analysis of covariance seems to be the most appropriate because it corrects for the phenomenon of regression to the mean. For dichotomous outcome variables, multinomial logistic regression analysis with all possible changes over time as outcome seems to be the most appropriate, especially because of its straightforward interpretation.
Conclusion: A different definition of "change" can lead to different results in the evaluation of the effect of an intervention.