ABSTRACT
Numerous studies have evaluated the effect of nonsteroidal antiinflammatory drugs (NSAIDs) on acute gastrointestinal (GI) problems, specifically, upper GI bleeding and the incidence of esophageal and duodenal ulcers. But, although these studies investigated the impact of NSAID use on acute GI problems, none investigated the association between NSAIDs and generalized GI distress. The population was then reduced by eliminating patients with a history of GI upset. To determine whether patients had a prior history of GI upset, diagnostic codes indicative of GI upset and bills submitted for histamine antagonists or antacids during the first 90 days of observation for each patient were used. Since the period of time a specific recipient could be “observed” was variable, and since patients who exhibited GI upset were deleted from further observation, selection bias might occur if the length of observation was not controlled.
