MAINTENANCE STRATEGIES FOR UNIPOLAR DEPRESSION - AN OBSERVATIONAL STUDY OF LEVELS OF TREATMENT AND RECURRENCE

Background: This paper analyses data from a large observational study of the course of affective illness to provide insight into the duratio n and dose of effective maintenance therapies. Methods: The data are 2 36 unipolar patients who had received antidepressants during recovery and were followed for affective recurrence for up to 5 years. Using da ta on the naturally selected somatic treatments, we have conducted ana lyses that adjust for the potential confounding effects of prognosis a nd treatment intensity to estimate the causal effect of level of medic ation on the course of recurrence. Results: The results of these analy ses show that it is important for patients to remain on the level of s omatotherapy used to treat the acute episode for the initial 8 months after symptoms have abated. After that time, the rate of recurrence fo r patients with fewer than five previous episodes is approximately 1% per week or less at all levels of medication (including discontinuatio n). Patients who had experienced more than several recurrences are at greater risk of recurrence and continue to benefit from any level of m edication during the first year after recovery. Conclusions: The CDS a nalyses reported here suggest that effective maintenance strategies fo r all but highly recurrent patients may be a middle road, opting for f ull-dose strategies of limited duration. These results have implicatio ns at both the policy and the clinical level, given the need to consid er both monetary and nonmonetary costs (side-effects) associated with continued pharmacotherapy during remission. Limitations: The observati onal design of the CDS limits the degree to which cause and effect rel ationships can be inferred from the observed associations. (C) 1998 El sevier Science B.V.