We thank Patra & Subodh for their interest in and their very thorough reading of our study. ^{Reference Shimazu, Shimodera, Mino, Nishida, Kamimura and Sawada1} Most of their questions are factual ones and we are grateful that we have been given an opportunity to clarify them. First, whether to call further treatment of patients in partial or full remission after the fully syndromatic episode, as in our study, continuation/maintenance treatment is a terminological issue and not a medically substantive one. And we think our usage of the terms is in consonance with the majority of psychiatrists of the world, as for example done by Paykel et al in their famous study of cognitive therapy to prevent relapse after acute episode of major depression. ^{Reference Paykel, Scott, Teasdale, Johnson, Garland and Moore2}

Second, of the 57 patients who entered the trial, 27 had had a single episode of depression and 30 recurrent major depression. Those with their first lifetime major depressive episode received continuation treatment after their acute phase treatment. The comorbidities of our sample included three individuals with substance use disorder, ten with anxiety disorders and one with personality disorder. It is true that we did not methodically check the intellectual or health status of the primary family members but none apparently suffered from serious dysfunction in these regards. We regret the fact that we did not mention the ethics review committee’s approval in the published paper. It had been reviewed and approved by the ethic review committee of Kochi Medical School in 2003, prior to the commencement of the study. In addition, it is not true that four caregivers had their psychoeducation sessions individually: only 4 out of 16 sessions for these four family members were conducted individually either in the hospital or at home. Altogether, 4 out of 216 sessions (1.9%) were individually delivered. We reasoned that in this first-ever efficacy trial of family psychoeducation for major depression, it would be better for us to deliver the best therapy possible. Our definition of remission reflected some arguments that the accepted threshold might be too high to define true remission. ^{Reference Zimmerman, Posternak and Chelminski3,Reference Furukawa, Akechi, Azuma, Okuyama and Higuchi4} Changing the threshold for remission to 7/8, instead of 6/7, made little change to the findings.

Last, we would like to clarify that there were six (25%) families with high expressed emotion in the intervention group and ten (33%) in the control group.