All-cause and cause-specific mortality in rheumatoid arthritis, psoriatic arthritis and axial spondyloarthritis: a nationwide registry study

Abstract

Objectives: To explore mortality and causes of death among Norwegian patients with RA, PsA and axial spondyloarthritis (axSpA) compared with the general population by conducting a nationwide registry-based cohort study.

Methods: Patients with RA, PsA and axSpA were identified from the Norwegian Patient Registry based on ICD-10 codes between 2008 and 2017. Using age as the time variable, all-cause and cause-specific mortality were estimated between 2010 and 2017 with the Kaplan-Meier estimator and the cumulative incidence competing risk method, respectively. Sex-, education level-, health region- and age group-adjusted hazard ratios (HRs) for mortality were estimated using Cox regression models.

Results: We identified 36 095 RA, 18 700 PsA and 16 524 axSpA patients (70%, 53% and 45% women, respectively). RA and axSpA were associated with increased all-cause mortality (HR 1.45 [95% CI: 1.41, 1.48] and HR 1.38 [95% CI: 1.28, 1.38], respectively). Women but not men with PsA had a slightly increased mortality rate (HR 1.10 [95% CI: 1.00, 1.21] among women and 1.02 [95% CI: 0.93, 1.11] among men). For all patient groups as well as for the general population, the three leading causes of death were cardiovascular diseases, neoplasms and respiratory diseases. RA patients had increased mortality from all of these causes, while axSpA patients had increased mortality from cardiovascular and respiratory diseases.

Conclusion: Even in the era of modern treatments for IJDs, patients with RA and axSpA still have shortened life expectancy. Our findings warrant further attention to the prevention and management of comorbidities.

Keywords: cardiovascular disease; epidemiology; inflammatory joint diseases; mortality.

Fig. 1

Kaplan–Meier survival curves for overall survival Kaplan–Meier survival curves showing overall survival rate among RA, PsA and axSpA cohorts as well as the general population, with age at inclusion to the study (in years) as the time scale. axSpA: axial spondyloarthritis.

Fig. 2

Hazard ratios from Cox proportional hazard regression models for all-cause and cause-specific mortality Sex-, education level-, health region- and age-group-adjusted HRs with 95% CI from stratified Cox proportional hazard regression models (with age as the time scale) for all-cause and cause-specific mortality among RA, PsA and axSpA patients compared to the general population. axSpA: axial spondyloarthritis; HR: hazard ratio.