Clinical predictors of incident somatic morbidity in a sample of depressed patients: A 16–30 years follow-up study

Background and objectives: We investigated the independent associations between various characteristics at trial entrance and subsequent development of somatic morbidity in patients participating trials on antidepressants. Methods: 338 in-patients diagnosed with major depression who had participated in trials on antidepressants conducted between 1983 and 1994 were followed for up to 30 years in Danish registers. By applying a Cox regression model with incident diagnoses of somatic disease as outcome, explanatory variables such as age at first episode, duration of index episode, bipolarity and scores on the Hamilton Depression Scale and subscales hereof, were investigated. Results: Cardiovascular diseases were associated with increasing number of previous depressive episodes at baseline (HR 1.06, 95% CI (1.00–1.11)). The risk of diabetes was associated with increasing duration of index episode (HR 1.01, 95% CI (1.00–1.01) as was respiratory disease (HR 1.00, 95% CI (1.00–1.01)). Diagnoses of infection were associated with increasing score on HAM-D₆ (HR 1.11, CI 95 % (1.01–1.22)). Conclusions: The association between number of previous depressive episodes and CVD is in line with previous results. The findings of associations between the psychometric measures and specific diseases should be interpreted with caution, as well as the associations between duration of episodes, higher severity and higher number of previous episodes, and increased risks of somatic morbidity, albeit these are in line with previous evidence.