Background Increased mortality among rheumatoid arthritis (RA) patients has been recognized for many decades, and some studies suggest a widening mortality gap between RA patients and the general population. The excess mortality has often been linked to a higher incidence of cardiovascular disease (CVD), but in several studies the increased mortality risk from CVD equalled the overall mortality, which suggests that other factors are important.

Objectives To analyse cause-specific mortality data from a recent nationwide population-based RA cohort.

Methods Through linkage of nationwide individual-level administrative registers we identified the entire Danish population alive on January 1, 1995 and aged >18 years. The cohort was followed to time of death, emigration or December 31, 2011. RA diagnosis were defined by a combination of relevant discharge and outpatient codes (The National Patient Register) and initiation of DMARD therapy. Primary cause of death was available from the National Cause of Death Register. Mortality rate ratios (MRRs) were calculated by poisson regression controlling for sex, age and calendar time.

Results From a total of 5.3m Danish residents included in the cohort, 25 104 developed RA during follow-up (mean age at diagnosis 58.5 [SD 14.9] years; 69.9% women). Table 1 shows mortality data for RA patients compared to the general population for main causes of death. All-cause mortality in RA patients was increased by 39%. Deaths from CVD and cancer were common and conferred an almost identical impact on death rates, but with lower MRRs than the overall RA MRR. Conversely, deaths due to infection were more than twice as common among RA patients than in the general population causing a relatively high number of excess deaths (224) compared to CVD (279) and cancer (271) despite 3-4 times lower incidence. Generally, cause-specific MRRs were higher in younger patients, however, significant interaction with age were only noted with lung- and cancer-related deaths. No significant differences in MRRs between sexes, calendar time or follow-up were found.

Conclusions In this nationwide Danish contemporary cohort of RA patients followed for up to 17 years, we found the mortality to be increased by 39% compared to the general population without indications of changes over time. As in the general population, cancer- and CVD-related deaths accounted for the largest proportion of deaths in RA patients but lung disease and, in particular, infections contributed relatively more to the increased mortality. These results indicate that the higher mortality in RA patients is not particular driven by CVD or cancer, but may reflect a general increased biological ageing and possibly treatment related disease.

Disclosure of Interest None declared