Repeated measures in clinical trials: Analysis using mean summary statistics and its implications for design

摘要: This paper explores the use of simple summary statistics for analysing repeated measurements in randomized clinical trials with two treatments. Quite often data each patient may be effectively summarized by a pre-treatment mean and post-treatment mean. Analysis covariance is method choice its superiority over analysis means or changes quantified, as regards both reduced variance avoidance bias, using model structure between time points. Quantitative consideration also given to practical issues design measures studies: merits having more than one measurement are demonstrated, methods determining sample sizes designs provided. Several examples from presented, broad recommendations made. The support value compound symmetry assumption realistic simplification quantitative planning trials. makes no such assumption. However, allowance alternative non-equal correlation structures can should made when necessary.