Prediction of Antidepressant Efficacy from Response to the First Ect

1. Gupta et al have proposed a 9% depression attenuation separator to differentiate potential responders from potential non responders. This separator is small and is intuitively not logical; for example, a 10% figure would have made m&re sense. The choice of a non-intuitive figure suggests that the separation between responders and non responders was very narrow. In other words, response to the first ECT may not clearly distinguish potential responders from nonresponders. Next, by changing the definition of response, one patient shifted from the good to the poor response category. This altered the statistical significance of the results and confirms that there was poor separation between good and poor responders based on 9% response with the first ECT. Finally, if the authors had extended the ECT course to beyond 6 treatments, some patients classified as poor responders might have responded. These patients would probably have shown less than 9% response to the first ECT. With the addition of these patients to the good response group, the first ECT response construct would have been further weakened.