We applaud the effort of Drs Philipp and Fickinger that clearly underscores the point made in our report: varying definitions of response or remission will lead to quite different lengths of episode and illness-free periods. The careful follow-up study of 66 psychiatric inpatients described above clearly demonstrates that an episode can be anywhere from 1 to more than 6 months in length depending on the particular criteria set for defining remission. We concur that a broad consensus on operational definitions of change points would lead to a more interpretable literature on depressive illness. We look forward to seeing the authors' demonstration that the cutoff 3/4 leads to the most valid definition of remission, since the validity of various cutoffs must be established before consensus definitions can be promulgated appropriately.
Supported in part by a grant from the John D. and Catherine T. MacArthur Foundation Research Network on the Psychobiology of