The declining incidence of nonsteroidal anti-inflammatory drug gastropathy in rheumatoid arthritis patients: rates and reasons

Background

Nonsteroidal anti-inflammatory drug (NSAID) gastropathy is recognized as a major cause of hospitalizations and deaths, with more than 100,000 annual hospitalizations and more than 10,000 annual deaths in the United States alone. Trends in this epidemic over time have not been examined.

Objective

We studied the possibility that new preventive approaches to NSAID gastropathy may have reduced the magnitude of this epidemic.

Methods

We studied 5598 patients with rheumatoid arthritis from longitudinal data banks previously employed to help establish the epidemiology of NSAID gastropathy. Consecutively enrolled patients were followed with bi-annual Health Assessment Questionnaire assessments and medical record audits between 1981 and 2000. Annual rates of hospitalization involving gastrointestinal (GI) bleeding, obstruction, or perforation were calculated and curves fitted using spline regression.

Results

Annual GI hospitalization rates rose from 0.6% in 1981 to a peak of 1.5% in 1992, then declined again to 0.6% again in 2000. The period of rising incidence was associated with an increase in patient age and GI risk propensity score. The period of declining incidence was correlated with use of lower doses of aspirin and ibuprofen, a decline in use of 'more toxic' NSAIDs from 52% to 42%, a rise in use of 'safer' NSAIDs from 19% to 48%, and increasing use of proton pump inhibitors. The decline was not associated with changes in age, NSAID exposure, or GI risk.

Conclusions

The risk of GI hospitalization for NSAID gastropathy has declined by 60% in these cohorts since 1992. We estimate a 16% decline attributable to dose reductions, 10% to use of proton pump inhibitors, and 10% to use of less toxic NSAIDs. Given continued trends in these three factors, these declines are likely to continue.