In the February issue of Patient Education & Counseling professor Anders Bærheim at the University of Bergen criticizes a previous work by Fossli Jensen et al in which a crossover randomized controlled trial was used to demonstrate the effect of a course in clinical communication. His point is that a crossover study is inadequate as this type of study requires that the situation at the start of the second intervention should be equal to the situation at the start of the first intervention. Read our response here.