Clinical staging of major depressive disorder: an empirical exploration

Objective: Clinical staging has been proposed to supplement psychiatric diagnoses. We examined the construct and predictive validity of a clinical staging model for major depressive disorder (MDD) that distinguishes 8 consecutive stages (0, 1A, 1B, 2, 3A, 3B, 3C, 4) based on symptom severity (Inventory of Depressive Symptomatology [IDS]) and duration (Life Chart Interview) and number of episodes (Composite International Diagnostic Interview [CIDI] based on DSM-IV criteria). Method: This study is a secondary data analysis based on baseline data (collected 2004-2007) and 2-year follow-up assessment (collected 2006-2009) from the Netherlands Study of Depression and Anxiety. 2,333 baseline participants were assigned to the 8 stages of the clinical staging model for MDD, and 2,012 participants were followed up after 2 years. For construct validity, differences between stages in clinical characteristics (eg, severity [IDS], age at onset [CIDI], and comorbid anxiety [CIDI]) were studied. Predictive validity was measured by the extent to which baseline stages predicted 2-year follow-up outcomes (eg, MDD presence [CIDI]). Results: Later stages scored significantly poorer than early stages on most clinical characteristics and follow-up outcomes (all overall P values < .001), confirming validity of the model. This pattern was especially evident in mostly preclinical stages (0, 1A, 1B, 2). Among clinical stages (2, 3A, 3B, 3C, 4), stages characterized by long-lasting symptomatology had at baseline similar and the highest scores (IDS scores: stage 3A = 37.1; stage 4 = 39.0; disability scores: stage 3A = 36.7; stage 4 = 40.6) and compared to preclinical stages had increased probability of having MDD at 2-year followup (stage 3A: OR = 4.3; 95% CI, 2.8-6.7; stage 4: OR = 8.1; 95% CI, 5.7-11.5). Conclusions: This study showed reasonable validity for an MDD staging model that based its stages purely on clinical characteristics. Results suggest that, contrary to the number of episodes, duration of exposure to the depressed state best characterizes clinical (later) stages of MDD. Future studies should test whether modifications to these clinical stages improve the validity of the model.