RESUMO
SETTING: St. Francis Hospital in Katete District, Eastern Province, Zambia. OBJECTIVE: To compare the incremental cost-effectiveness of examining serial sputum smears for screening suspects for pulmonary tuberculosis at a rural district hospital in Zambia. DESIGN: An incremental cost-effectiveness analysis of serial sputum smear examinations for diagnosing pulmonary tuberculosis based on laboratory results collected during 1997 and 1998 in a rural district hospital in Zambia. The cost analysis took a health service provider perspective, and used the ingredients approach. The cost-effectiveness is expressed in terms of the incremental cost per tuberculosis case diagnosed. Relevant information was obtained from various sources, including administrative records, interviews and direct observation. RESULTS: Of a total of 166 acid-fast bacilli positive suspects who had three sputum smears examined sequentially, 128 (77.1%) were found on the first smear, a further 25 (15%) on the second smear and 13 (7.9%) additional cases were identified on the third smear. The economic analysis shows that the incremental cost of performing a third test, having already done two, increases rapidly with only a small gain in terms of additional cases of tuberculosis identified. CONCLUSION: A policy of examining two samples should be considered in resource-poor settings, if the remaining steps of the national diagnostic algorithm can be adhered to with respect to smear-negative suspects.