No Excess Mortality in A Population-Based Danish Rheumatoid Arthritis Cohort from The Late Pre-Biologic Era

Ann Rheum Dis 2016;75:197 doi:10.1136/annrheumdis-2016-eular.2092

Background

Patients with rheumatoid arthritis (RA) from hospital-based series have an increased mortality, in particular from cardiovascular diseases. Estimates of standardized mortality ratios (SMR) from population-based inception cohorts have been inconsistent and mortality in RA patients may be modified by treatment and preventive strategies. Furthermore, human senescence in the general population has been delayed since 1950.(1) We therefore wanted to investigate the all-cause SMR in a Danish population-based inception RA cohort from the late pre-biologic era seen at a regional hospital with documented focus on intention-to-treat (2, 3) and cardiovascular screening (4).

Objectives

To investigate SMR in an unselected inception cohort of Danish RA patients from 1995 to 2002.

Methods

The case base was the population of the County of South Jutland according to the Danish Civil Registration System (CVR). Incident RA cases were identified at the referral center for rheumatic diseases in the County where the observed completeness of registrations from the case base was 97%.(5) The registry was scrutinized for all RA patients fulfilling the 1987 American College of Rheumatology criteria for RA from 1995 to 2002. Linkage of cases to CVR and the Danish Register of Causes of Death was based on unique personal identification numbers. Cases were followed from inclusion until emigration, death or end of follow-up (31.12.2013). Statistics Denmark provided mortality rates for the total Danish population from 1995 to 2013.

Results

A total of 509 cases were identified (Females 68%; mean age 61 years; rheumatoid factor positive 77%; erosive 23%). During 6079 person-years (py) of follow-up 200 (39%) patients had died (Females 32%; males 55%). The main causes of death are seen in table 1. There were no statistically significant differences in the causes of death according to sex (Chi-squared test=12.2, degrees of freedom: 7, p=0.092). The median follow-up time was 13 years and the average mortality rate 33/1000 py (Females 26/1000 py; males 50/1000 py). The overall SMR was 1.04 (95%>confidence interval (CI): 0.91–1.19), in females 0.96 (95%>CI: 0.79–1.43) and in males 1.16 (95%>CI: 0.95–1.43).

Conclusions

In this population-based study there were no significant excess mortality compared to the general population. The predominant causes of death were cardiovascular-, neoplastic- and pulmonary disease.

References

1   Vaupel JW. Nature 2010;464(7288):536–42.

2   Hetland ML, et al. Arthritis Rheum. 2006 May;54(5):1401–9.

3   Hørslev-Petersen K, et al. Ann Rheum Dis. 2014 Apr;73(4):654–61.

4   Primdahl J, el al. Ann Rheum Dis. 2013 Nov;72(11):1771–6.

5   Pedersen JK, et al. Rheumatol Int 2009;29(4):411–5.

Acknowledgement

Supported by the Region of Southern Denmark and the Danish Rheumatism Association.