A successful screening programme may lead to a diminishing yield over time, at least if the risk group is a fixed population. Over time, changes in the background burden of TB and changes in the profile of TB patients in the community (for example, a trend towards fewer patients with symptomatic TB, fewer cases of sputum smear-positive TB and decreasing TB mortality) can lead to a reduction in the yield from screening, an increase in the NNS, a reduction in cost–effectiveness and a change in the ratio of benefits to harm. Successful programmes that facilitate access to care may also lead to diagnosis of more people with TB through screening than via the patient-initiated pathway. Trends in all these indicators should be monitored, and the priority of risk groups, the choice of screening approach and the screening interval should be reassessed regularly. Criteria for stopping screening should be established before a screening initiative is implemented.