The health care costs of nonsteroidal anti-inflammatory drug-related gastric ulcer in Olmsted County, MN, 1987-1990

The objective of this cross-sectional descriptive study was to estimate the costs incurred in the management of gastric ulcers (GU) related to use of nonsteroidal anti-inflammatory drugs (NSAID). Using a population-based data resource which ensures nearly complete ascertainment of all medical care interactions for county residents, we identified all Olmsted County residents with a previous diagnosis of osteoarthritis or rheumatoid arthritis who were treated, 1987-1990, for a confirmed (by X-ray or endoscopy) GU and had documented NSAID use at GU diagnosis. All charges (in 1990 dollars), for health services rendered in the care of a GU episode were obtained, except those for outpatient medications. Services were classified as those utilized in the management of the ulcer or its complications, in concurrent arthritis care, in general health maintenance or in the treatment of comorbid medical conditions. The 102 subjects identified had a mean age of 68 years (32-95 years) and 80% were female. Fifty were treated as outpatients, 44 as medical inpatients and 8 as surgical inpatients. The median charge incurred in the care of NSAID-associated GU in these patients was $3479. While 75% of all charges were related to GU management per se, 18% were related to the treatment of comorbid conditions, the remainder being equally distributed between arthritis care and general health maintenance. Unadjusted univariate analyses suggested the following associations: evidence of bleeding and advanced age with higher GU-related charges; multiple NSAID use with higher arthritis-related charges; advanced age with higher health maintenance related charges; and corticosteroid use with higher comorbidity-related charges. Stepwise multiple linear regression identified three patient characteristics which were independently associated with higher charges: evidence of bleeding, multiple NSAID use and advanced age. Assuming these data can be generalized nationally, they suggest yearly expenditures for NSAID-associated GU of approximately $300 million, of which $100 million can be attributed to NSAID use. These data emphasize the importance of research into the prevention of NSAID-associated GU.