Comparison of false positive and false negative rates of two indices of individual reliable change: Jacobson-Truax and Hageman-Arrindell methods.

Background: Quantification of change is crucial for correctly estimating the effect of a treatment and for distinguishing random or non-systematic changes from substantive changes. The objective of the present study was to learn about the performance of two distribution-based methods [the Jacobson-Truax Reliable Change Index (RCI) and the Hageman-Arrindell (HA) approach] that were designed for evaluating individual reliable change. Methods: A pre-post design was simulated with the purpose to evaluate the false positive and false negative rates of RCI and HA methods. In this design, a first measurement is obtained before treatment and a second measurement is obtained after treatment, in the same group of subjects. Results: In relation to the rate of false positives, only the HA statistic provided acceptable results. Regarding the rate of false negatives, both statistics offered similar results, and both could claim to offer acceptable rates when Ferguson's stringent criteria were used to define effect sizes as opposed to when the conventional criteria advanced by Cohen were employed. Conclusion: Since the HA statistic appeared to be a better option than the RCI statistic, we have developed and presented an Excel macro so that the greater complexity of calculating HA would not represent an obstacle for the non-expert user.