The analysis of randomized controlled trial data with more than one follow-up measurement. A comparison between different approaches

When more than one follow-up measurement is analysed in a randomized controlled trial, there is no consensus how to analyse the overall intervention effect in a proper way. Mostly, longitudinal analysis of covariance is used, because with this method a correction is made for possible regression to the mean. However, in this paper it is shown that this method (mostly) leads to an overestimation of the intervention effect. A possible solution is the use of autoregression, although this does not seem to be the best solution, because it leads to an overcorrection. Due to these flaws, in this paper a new approach is introduced in which a correction for the baseline value is made for the first follow-up, but no correction is made for the remaining follow-up measurements.