OBJECTIVES: To assess the cost-effectiveness of three tests for HP detection in adults and develop decision analysis models to compare "test - no test" strategies for the treatment of persistent dyspepsia. METHODS: Two decision analytic models were constructed and analysed from the perspective of the health service. The first model was a simple decision tree of three types of HP test, allowing for true and false test results, with the ‘number of true outcomes' as the measure of effectiveness. Tests considered were the serological test, the C-urea breath test and the monoclonal faecal antigen test. The second model was based on published guidelines for managing dyspepsia and procedures in secondary care. Measures of effectiveness for the second model include numbers in each end state, number of endoscopies performed, number of HP eradication treatments given inappropriately and the extent of wasted resources consumed. Data used to furnish the models were gathered from the literature and available published costs. RESULTS: The monoclonal faecal antigen test was the most cost effective solution with an ICER of £2 per additional true outcome, but was highly dependent on the sensitivity and specificity of the serological test. The stool test and the breath test either dominate or are relatively cost effective in relation to the serological test unless the specificity of the serological test exceeds 0.93. The stool test either dominates or is relatively cost effective in relation to serological test up to a cost of £18.67 per stool test. Preliminary results from the second model suggest that the decision is highly dependent on values attached to the variables in the model, especially in relation to costs of treatment and costs associated with malignancy. CONCLUSION: In the UK, the faecal antigen test is a cost effective solution to testing for HP in dyspepsia patients.