Latent structure of the Hamilton depression rating scale in major depressive disorder

Previous research has suggested that the Hamilton Rating Scale for Depression (HRSD) in schizophrenia has a latent structure that includes a Negative Symptom (NS) -like factor, a depression factor and two other factors. In line with our interest in the transdiagnostic nature of NS and the overlap between NS and depressive symptoms (DS), we measured NS and DS in 23 patients meeting DSM-IV criteria for Major Depressive Disorder (MDD) using the Scale for the Assessment of Negative Symptoms (SANS) and the 17-item HRSD. We separated HRSD items which were descriptive of NS (#7, #8, and #13) into a separate scale HRSD-NS. The remaining 14 items were combined into HRSDDS scale. HRSD scores were highly correlated with HRSD-DS scores (Spearman r=.95, p<0.0001) but not with HRSD-NS scores (p=.10, ns). There was no significant correlation between HRSD-DS and HRSD- NS scores (p.17, ns). Spearman correlation analysis also showed no significant correlation between HRSD or HRSD-DS scores and SANS total scores, but showed a significant correlation between HRSD-NS scores and SANS total scores (r=0.80, p<.0001). While there was no significant correlation between HRSD and HRSD-DS scores and any of the SANS subscale scores, there was a significant correlation between HRSD-NS scores and three of the SANS subscale scores: avolition apathy (r=.44, p<0.05), alogia (r-0.58, p<.005), and affective flattening (r=.67, p<.0005). The present study indicates that NS and DS in MDD are separate constructs, and HRSD might have a latent structure in MDD as it does in schizophrenia. Factor analysis of HRSD scores using a larger group of subjects is needed to reveal the full multifactorial structure of HRSD in MDD.