Abstract
OBJECTIVE: Whether comorbid depression increases mortality in patients with rheumatoid arthritis (RA) is unknown. Our objective was to determine whether the presence of depression predicted mortality in patients with RA. METHODS: We followed 1290 consecutive outpatients with RA who met our stringent inclusion criteria during an 18-year observation period. Since 1981, demographic, clinic, and self-report data were collected and entered into a computer database at the time of each clinic visit. The comorbidity data were consistently recorded beginning in 1991. Our primary independent variable was the mean of the Arthritis Impact Measurement Scales (AIMS) depression scores during the first 4 years of entry into the clinic cohort (average 4-year depression). Data were analyzed using Cox proportional hazard models. RESULTS: After adjusting for covariates, the hazard ratio (HR) for each unit increase in the average 4-year depression score on mortality was 1.14 (p < 0.0001). Using only the data obtained from 1991 to 2003, the mortality risk was slightly increased (HR 1.35, p < 0.0001). To reduce residual confounding due to RA disease activity and/or comorbid medical conditions, we then excluded deaths during the first 2 years after study onset. With this method, the HR for the average 4-year depression remained significant (HR 1.35, p < 0.0001). Because an AIMS depression score > or = 4 is consistent with clinical depression, we analyzed the dataset using the average 4-year depression score as a dichotomous variable (score < 4 or > or = 4). The HR of clinical depression on mortality was 2.2 (95% CI 1.2- 3.9, p = 0.01). CONCLUSION: Depression increases the risk of mortality in RA. Our study highlights the importance of comorbid depression in patients with RA.