OBJECTIVE: The role of Helicobacter pylori in nonulcer dyspepsia is controversial. Speculation has arisen that only strains of H. pylori carrying the CagA virulence factor are important in the development of dyspepsia. The objective of this study was to determine whether nonulcer dyspepsia correlated with CagA-positive H. pylori infection. METHODS: A total of 435 healthy blood donors and 102 general medicine clinic respondents completed the Bowel Disease Questionnaire and the PRIME-MD survey, a validated screen for common psychiatric disorders. Subjects were classified as cases of nonulcer dyspepsia if they reported pain in the upper abdomen more than six times in the previous year and denied a past or current history of peptic ulcer disease. Study participants were tested for IgG antibodies to H. pylori and the CagA protein. RESULTS: Clinic respondents were more likely than healthy blood donors to meet the case definition for nonulcer dyspepsia (34% vs 13%, p < 0.001), to be seropositive for H. pylori (54% vs 18%, p < 0.001), and to be CagA seropositive (41% vs 10%, p = 0.01). Logistic regression identified CagA seropositivity (p = 0.03), race (p = 0.001), and positive screens for depression (p = 0.007) or somatization (p < 0.001) as variables independently associated with nonulcer dyspepsia. CONCLUSIONS: Infection with a CagA-positive strain of H. pylori is associated with a clinical diagnosis of nonulcer dyspepsia. However, nonulcer dyspepsia was also strongly and independently associated with positive screens for depression or somatization disorder as well as with ethnicity. These potential sources of variance should be considered in the design of future studies evaluating nonulcer dyspepsia. (C) 2000 by Am. Coll. of Gastroenterology.