Background Rheumatoid Arthritis (RA) is associated with increased mortality compared to the general population. RA-patients have an increased risk of infections compared with non-RA individuals, but it remains unclear whether RA affects infection prognosis as well. Medical registries may be a valuable source for studying this issue, if diagnoses are coded accurately.
Objectives To examine whether RA is associated with increased mortality in patients who are hospitalized with pneumonia and to assess the validity of RA diagnoses in The National Patient Registry (NPR) in Denmark.
Methods We conducted this population-based cohort study of adults with a first-time hospitalization with pneumonia between 1997 and 2011 in Northern Denmark. Information on RA, comorbidity and pneumonia was obtained from medical databases. To validate the diagnosis of RA in the NPR, we reviewed a sample of 200 medical records on pneumonia patients with one or more previous RA registrations in NPR (100 with one registration, 50 with two registrations, and 50 with three or more registrations). Using three definitions of RA (clinically confirmed RA, fulfillment of the American College of Rheumatology (ACR) 1987 criteria for RA and the 2010 ACR/EULAR classification criteria for RA for patients diagnosed in 2010 or later) a trained rheumatologist confirmed the diagnosis. For the prognostic study, data on mortality was ascertained from the Danish Civil Registration System. Cox regression was used to compute the adjusted mortality rate ratio (MMR) within 30 and 90 days following hospitalization comparing patients with and without RA, controlling for sex, age, level of comorbidity, alcoholism and antibiotics before admission.
Results Overall 88% (95% CI: 0.83-0.92) of the RA-diagnoses in the NPR sample were confirmed according to the medical records. Among patients with three or more registrations in the NPR, 98% of the RA diagnoses were confirmed whereas the percentage of patients with a confirmed diagnosis among those with only one registration was 81%. No patients were diagnosed using the 2010 ACR/EULAR classification criteria for RA (Table).
A total of 80995 patients were hospitalized with pneumonia. Among these 1249 (1.5%) had RA. The 30- and 90-day mortality in pneumonia patients with RA was 14.7 and 21.9%, respectively, compared with 14.0 and 20.4% in non-RA pneumonia patients. Corresponding crude 30- and 90-day MRRs were 1.05 (95% CI: 0.91-1.22) and 1.08 (95% CI: 0.96-1.22) and adjusted MRRs were 0.97 (95% CI: 0.83-1.12) and 0.96 (95% CI: 0.85-1.08), respectively.
Conclusions RA was not associated with increased 30- or 90-day mortality following hospitalization with pneumonia. The positive predictive value of RA diagnoses in the Danish NPR is high and the registry may be a useful tool to study RA outcomes.
Acknowledgements This study was supported by The Danish Rheumatism Association
Disclosure of Interest M. Holland-Fischer Consultant for: Abbvie, Speakers bureau: UCB, MSD, M. Nørgaard: None declared, U. Tarp: None declared, R. Thomsen: None declared