The comparative risk and predictors of adverse gastrointestinal events in rheumatoid arthritis and osteoarthritis: a prospective 13 year study of 2131 patients

J Rheumatol. 2000 Jul;27(7):1668-73.


Objective: It has been suggested that rheumatoid arthritis (RA) itself may be a risk factor for adverse gastrointestinal (GI) events, but this hypothesis has not been studied in a large sample, nor has the effect of time on risk factors been studied. We investigated rates and risk factors for GI events in RA and osteoarthritis (OA) and assessed the additional risks conveyed by having RA.

Methods: A prospective study of patients with OA and RA from a single center was undertaken using questionnaires mailed at 6 month intervals. The relationship between drug therapy and GI events was assessed in the same 6 month time frame. Over 13 years of biannual assessments, 2,131 patients were studied for serious GI events and adverse GI symptoms during 9,621 patient-years of observation.

Results: The incidence rate (IR) for GI hospitalization was 1.56 and 1.28 per 100 patient-years, and for GI bleeding or perforation was 0.50 and 0.58 for RA and OA, respectively. After controlling for age, sex, nonsteroidal antiinflammatory drug (NSAID) and steroid use, the incidence rate ratio (IRR) for RA versus OA did not differ for hospitalization [IRR 1.07 (95% CI 0.66, 1.74)] or for bleeding or perforation [IRR 0.63 (95% CI 0.29, 1.35)]. In multivariate analyses for both groups combined, the IRR was 2.95 (2.05, 4.24) for prednisone use, 1.41 (1.08, 1.85) for NSAID use, and 1.46 (1.22, 1.74) for every 10 year increase in age. In additional multivariate models, Health Assessment Questionnaire disability was also a significant risk factor. During the study period, the odds of NSAID use decreased 2.94 times per 10 year period, while the odds of prednisone use increased by 1.49. Dysphagia [IRR 1.11 (1.00, 1.24)], anorexia [IRR 1.13 (1.03, 1.23)], nausea [IRR 1.13 (1.04, 1.25)], heartburn [IRR 1.12 (1.05, 1.19)], vomiting [IRR 1.20 (1.02, 1.42)], peptic ulcer symptoms [IRR 1.20 (1.11, 1.29)], and abdominal pain [IRR 1.11 (1.01, 1.22)] were associated with NSAID use, but not with steroids.

Conclusion: Patients with RA and OA do not differ in the rates and risk factors for GI hospitalizations and symptoms after controlling for age, steroid use, NSAID use, or (for OA) body mass index. Prednisone is a more important risk factor among patients with RA than NSAID.

Publication types

  • Research Support, Non-U.S. Gov't
  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Adrenal Cortex Hormones / adverse effects
  • Aged
  • Anti-Inflammatory Agents, Non-Steroidal / adverse effects
  • Arthritis, Rheumatoid / complications*
  • Arthritis, Rheumatoid / drug therapy
  • Arthritis, Rheumatoid / epidemiology
  • Causality
  • Female
  • Gastrointestinal Diseases / diagnosis
  • Gastrointestinal Diseases / epidemiology
  • Gastrointestinal Diseases / etiology*
  • Hospitalization / statistics & numerical data
  • Humans
  • Male
  • Middle Aged
  • Multivariate Analysis
  • Osteoarthritis / complications*
  • Osteoarthritis / drug therapy
  • Osteoarthritis / epidemiology
  • Prospective Studies
  • Risk Factors
  • Time Factors


  • Adrenal Cortex Hormones
  • Anti-Inflammatory Agents, Non-Steroidal