Efficiency of a third serial sputum smear examination in the diagnosis of tuberculosis in Moldova and Uganda

SETTING: Twenty-four and 30 tuberculosis (TB) microscopy laboratories in Moldova and Uganda, respectively. OBJECTIVE: To estimate the workload required to identify one additional case of TB with a third serial sputum smear examination. METHODS: Retrospective laboratory register study to determine the prevalence and the incremental yield of TB cases from a third serial sputum smear examination among suspects in Moldova and Uganda, with the reciprocal of the product of these two fractions providing the number of examinations required to identify one additional TB case. RESULTS: In Moldova, 9% (1141/12 525) and in Uganda 200% (7280/36 054) of suspects met the TB case definition with at least one positive sputum smear. The incremental yield from the third examination was 4% in Moldova and 3% in Uganda. To detect one additional TB case on a third smear, 273 examinations (95%CI 200-389) in Moldova and 175 (95%CI 153-222) in Uganda were thus required. This corresponded to an average of I I days (8-16) and 7 days (6-9), respectively, to diagnose one additional case of TB. CONCLUSION: In both countries, the third serial sputum smear examination was inefficient in diagnosing sputum smear-positive TB.