Three Major Depression Scales Can Be Harmonized for Bipolar Disorder

A study found that harmonizing depression scales into ordinal categories was feasible and did not lose statistical power among older patients. These findings were published in the Journal of Affective Disorders.

Data for this study were sourced from the Global Aging & Geriatric Experiments in the Bipolar Disorder Database (GAGE-BD) which contains pooled data from multiple archival studies about bipolar disorder (BD) and aging. This study sample was derived from 8 studies conducted at 6 sites. Hamilton Depression Rating Scale (HAM-D), Montgomery-Åsberg Depression Rating Scale (MADRS), and Center for Epidemiological Studies Depression scales (CES-D) scores from 582 older adults with BD were harmonized and compared with Global Assessment of Functioning (GAF) scores.

The total sample had a mean age of 57.0 (SD, 12.1) years, 59.5% were women, 65.5% had BD I, disease had onset at a mean age of 27.1 (SD, 14.0) years, lifetime psychiatric hospitalizations was 4.0 (SD, 6.3), 53.6% currently used antipsychotics, and 25.6% currently used lithium.

HAM-D score was 8.6 (SD, 8.1) points, MADRS score was 18.5 (SD, 9.3) points, and CES-D score was 15.4 (SD, 9.7) points.

The relationship with GAF with HAM-D as a continuous (adjusted R², 0.369) and categorical (adjusted R², 0.311) classification of depression, MADRS as a continuous (adjusted R², 0.525) and categorical (adjusted R², 0.386) classification, and CES-D as a continuous (adjusted R², 0.122) and categorical (adjusted R², 0.101) classification indicated worse depression correlated with poorer functioning.

Stratified by depression status, GAF scores for individuals with no, mild to moderate, and severe depression were 68.31, 56.88, and 49.33 points for HAM-D, respectively; 75.23, 58.96, and 49.04 points for MADRS, respectively; and 67.70, 60.84, and 57.85 points for CES-D, respectively.

The major limitation of this study was that not all studies used all instruments to evaluate patients.

Study authors concluded, “The current study illustrates the potential utility of harmonizing different depression rating scales, without losing statistical power. By successfully harmonizing these different depression measures, we found that more depressive symptoms were associated with worse functioning in older age patients with BD. Since the older age patients with BD population is growing at a population level, it is essential to gain more knowledge about this vulnerable group.”