Abstract
OBJECTIVE: To study the association between anti-tumor necrosis factor (anti-TNF) therapy and mortality in a national cohort of patients with rheumatoid arthritis. METHODS: We prospectively followed up 12,672 patients who were beginning anti-TNF therapy and 3,522 biologic-naive patients receiving disease-modifying antirheumatic drugs (DMARDs) until either July 31, 2008, or death, whichever occurred first. Notification of death and cause of death was received from the UK National Death Register. Mortality was compared using Cox proportional hazards models. Inverse probability of treatment weighting was used to adjust for the confounding effects of baseline differences between groups, including age, sex, disease severity, disability, and comorbidity. Missing baseline data were accounted for using multiple imputation. RESULTS: When compared with the DMARD cohort, the anti-TNF cohort was younger (median age 57 years versus 61 years), had greater disease activity (median Disease Activity Score in 28 joints 6.6 versus 5.1), and had greater disability (median Health Assessment Questionnaire score 2.1 versus 1.6). Patients in the DMARD cohort were more likely to have a history of myocardial infarction (4.8% versus 3.1%) and chronic obstructive pulmonary disease (8.1% versus 4.8%) but were less likely to have had depression (16.5% versus 18.9%). There were 9,445 and 50,803 person-years of followup in the DMARD and anti-TNF cohorts, respectively, during which time 204 DMARD-treated and 856 anti-TNF-treated patients died. The weighted mortality hazard ratios in the anti-TNF cohort were as follows: all-cause 0.86 (95% confidence interval [95% CI] 0.64-1.16), circulatory disease 0.73 (95% CI 0.44-1.23), neoplasm 0.65 (95% CI 0.39-1.09), and respiratory disease 0.81 (95% CI 0.36-1.83). CONCLUSION: Our results indicate that, compared with standard DMARD therapy, treatment with anti-TNF therapies was not associated with an increase in mortality.