We tested the degree to which longitudinal observations fit two hypotheses of psychiatric co-morbidity in DSM-IV major depressive disorder (MDD) among adult patients: (1) Axis I co-morbidity is dependent on major depressive episode (MDE) course, and (2) Axis I co-morbidity is independent of MDE course.

In the Vantaa Depression Study (VDS), 269 psychiatric secondary-care patients with a DSM-IV MDD were evaluated with the Schedules for Clinical Assessment in Neuropsychiatry (SCAN) at intake and at 6 and 18 months. Three evaluations of co-morbidity were available for 193 out of 259 living patients (75%). A latent curve model (LCM) was used to examine individual-level changes in depressive and anxiety symptoms across time. Outcome of MDD was modeled in terms of categorical DSM-IV diagnosis and Beck Depression Inventory (BDI) and Hamilton Depression Rating Scale (HAMD) scores, and co-morbidity in terms of categorical DSM-IV anxiety and alcohol use disorder (AUD) diagnoses and Beck Anxiety Inventory (BAI) scores.

Depression and anxiety correlated cross-sectionally at baseline. Longitudinally, changes in depression and anxiety correlated in both the 0–6 and 6–18 months time windows. Higher baseline depression raised the likelihood of an AUD at 6 months, and patients with more depressive symptoms in the 0–6 months time window were more likely to have had an AUD at 6 months, which further linked to less improvement in depression symptoms in the 6–18 months time window.

Longitudinal and individual-level courses of both internalizing and externalizing disorders in adult patients with MDD seem to be dependent, albeit to differing degrees, on the course of depressive symptoms.