A serological test that predicts accurately active Campylobacter pylori infection in the human stomach has been developed and validated by means of serum from 189 patients who were undergoing endoscopy in Sydney. Our enzyme-linked immunosorbent assay (ELISA) has a sensitivity of 100% and a specificity of 94%. An important part of the test is the inclusion of a simple absorption step with C. jejuni for those sera whose results are close to the cut-off point for positivity. This has been shown to be particularly relevant in epidemiological studies on populations that are likely to be exposed to C. jejuni continually. For example, by means of the ELISA test without an absorption step, the infection rate in Papua New Guineans was shown to be 56%, whereas use of the validated test showed the infection rate to be 25%. Compared with an age-matched normal control group of Australians, this still is higher. These studies suggest a need for the reappraisal of some of the reported epidemiological data. The development of a validated diagnostic test that is a highly-accurate predictor of active C. pylori infection will be useful for epidemiological studies, particularly for comparative studies among populations that are known to differ significantly in the incidence of stomach disease. However, routine use of this test in diagnosis in general practice awaits the definition of more precise markers of C. pylori-associated symptomatology and the development of improved regimens for the treatment of C. pylori infections.