Smoking, nonsteroidal anti-inflammatory drugs, and acetaminophen in gastric ulcer. A study of associations and of the effects of previous diagnosis on exposure patterns.

A study was conducted in Sydney, Australia to investigate whether the use of some nonaspirin nonsteroidal anti-inflammatory (anti-arthritic) drugs was associated with gastric ulcer in patients who had had neither a gastric nor a duodenal ulcer diagnosed previously (i.e., in new cases of peptic ulcer), and whether the use of anti-arthritic drugs, aspirin, acetaminophen, and cigarettes was influenced by the diagnosis of a peptic ulcer--either gastric or duodenal. Bleeding of gastric ulcer in new cases was studied in relation to smoking, use of aspirin or anti-arthritic drugs, and age. The 417 patients in the study lived in Sydney and were interviewed between 1982 and 1985 after diagnosis of a gastric ulcer; 192 patients had had a gastric or duodenal ulcer diagnosed previously (i.e., were recurrent cases of peptic ulcer), and the remaining 225 patients were new cases to whom 411 community controls were matched on sex, age, and prestige of area of residence. In new cases, odds ratios of gastric ulcer were 5.0 for daily aspirin use and 2.3 to 5.4 for daily anti-arthritic drug use. Fewer recurrent cases than new cases had used aspirin daily (odds ratio (OR) = 0.5) or anti-arthritic drugs daily (OR = 0.5), and more recurrent than new cases had used acetaminophen daily (OR = 2.5). Among new cases, bleeding of gastric ulcer was less common in smokers (OR = 0.6), and more common in daily users of aspirin (OR = 2.1) and, to a lesser extent, in daily users of anti-arthritic drugs (OR = 1.5), and in patients aged 60 or more years (OR = 2.3) independent of usage of the above drugs. It is concluded that 1) the use of anti-arthritic drugs increases gastric ulcer risk; 2) diagnosis of a gastric or duodenal ulcer causes a decrease in the use of aspirin and anti-arthritic drugs and an increase in acetaminophen use, but does not change smoking habits; 3) aspirin use or advancing age may add to the risk of bleeding ulcer although the role of anti-arthritic drugs is less clear.