Clinical relevance of depressive symptom improvement in bipolar I depressed patients

Gaps remain between rating scale changes obtained in a clinical trial and what those results mean in clinical practice.

To better understand the relevance of results from a clinical trial we examined the relationship between rating scale measures and the clinicians' assessment of illness severity.

Data from a randomized double-blind 8-week study of bipolar I depression were examined post hoc in patients who received placebo (PLA, n = 355), olanzapine (n = 351) (OLZ, 5 to 20 mg/d), or olanzapine-fluoxetine combination (n = 82) (OFC, 6 and 25, 6 and 50, or 12 and 50 mg/d). Principal components analysis identified related symptoms (factors) from Montgomery-Asberg Depression Rating Scale (MADRS) item scores. Regression analysis examined baseline to endpoint changes in factor scores and Clinical Global Impression (CGI) scores. Mixed-effects model repeated measures analysis assessed differences between treatment groups.

MADRS factors identified were: sadness, negative thoughts, detachment, and neurovegetative symptoms. Factor and CGI scores were significantly reduced from baseline to endpoint (LOCF) in the combination therapy group as compared with placebo (p < .01). Changes in factor scores were highly correlated (p < .001) with changes in the CGI. Over 80% of this treatment effect was attributable to indirect effects of improvements in the MADRS factors, the remaining difference could not be explained even when changes in the YMRS and HAMA scores were included in the analytical model.

The changes in MADRS factors were closely aligned with the clinician's assessment of overall depression severity, which may suggest a high degree of clinical relevance for differences observed between treatments.