BACKGROUND: Few studies have examined the course of illness among severely depressed patients ascertained at first hospitalization. Using data from the Suffolk County Mental Health Project (SCMHP), we investigated the times to and predictors of the first full remission and the first relapse during a 4-year period in a first-admission cohort with major depressive disorder (MDD) with psychotic features. METHOD: The cohort included 87 county-wide, first-admission patients with a longitudinal consensus diagnosis of MDD with psychotic features who were systematically followed over a 4-year period. We examined the associations of background, clinical and treatment factors, and time-varying indices of antidepressant (AD) and antipsychotic (AP) medication use to time to remission and relapse using Cox regression. RESULTS: By the 4-year follow-up, 60 respondents (69.0%) had achieved a period of full remission (median time of 22 weeks among remitters and 54 weeks in the full sample). In the multivariable analysis, longer time to remission was associated with longer latency between initial episode and hospitalization, lower pre-hospital Global Assessment of Functioning (GAF) score, and lack of insurance, but not use of medication. Twenty-six remitters (43.3%) relapsed (median time of 50 weeks among those who relapsed and 192 weeks among all remitters). None of the risk factors or time-varying medication variables was significantly associated with time to relapse. CONCLUSION: Only two-thirds of the sample had at least one full remission by 4 years, and almost half of them subsequently relapsed. Poorer pre-hospital resources predicted remission but not relapse. Medication use over the follow-up was not associated with remission or relapse.
BACKGROUND: Few studies have examined the course of illness among severely depressedpatients ascertained at first hospitalization. Using data from the Suffolk County Mental Health Project (SCMHP), we investigated the times to and predictors of the first full remission and the first relapse during a 4-year period in a first-admission cohort with major depressive disorder (MDD) with psychotic features. METHOD: The cohort included 87 county-wide, first-admission patients with a longitudinal consensus diagnosis of MDD with psychotic features who were systematically followed over a 4-year period. We examined the associations of background, clinical and treatment factors, and time-varying indices of antidepressant (AD) and antipsychotic (AP) medication use to time to remission and relapse using Cox regression. RESULTS: By the 4-year follow-up, 60 respondents (69.0%) had achieved a period of full remission (median time of 22 weeks among remitters and 54 weeks in the full sample). In the multivariable analysis, longer time to remission was associated with longer latency between initial episode and hospitalization, lower pre-hospital Global Assessment of Functioning (GAF) score, and lack of insurance, but not use of medication. Twenty-six remitters (43.3%) relapsed (median time of 50 weeks among those who relapsed and 192 weeks among all remitters). None of the risk factors or time-varying medication variables was significantly associated with time to relapse. CONCLUSION: Only two-thirds of the sample had at least one full remission by 4 years, and almost half of them subsequently relapsed. Poorer pre-hospital resources predicted remission but not relapse. Medication use over the follow-up was not associated with remission or relapse.