Factors determining the outcome of treatment of adult smear-positive tuberculosis cases in The Gambia

SETTING: Health centres in The Gambia, West Africa. OBJECTIVES: To identify factors determining the outcome of treatment of adult tuberculosis cases in a Tuberculosis Control Programme using directly observed treatment. DESIGN: Information on the outcome of treatment was collected on all tuberculosis cases registered with the Tuberculosis Control Programme in 1994 and 1995 and treated under supervision by tuberculosis control staff, nurses or village health workers. Treatment outcome was recorded as cured, completed treatment, failed, defaulted or died. Transferred-out patients were traced and their treatment outcome recorded at the health centre where they had last been seen. RESULTS: Data were analysed for 1357 adult smear-positive tuberculosis cases. Sputum smear conversion 2 months after the start of treatment was observed in 90% of smear-positive cases and was more likely to occur if the initial bacterial load in the sputum was low. The total cure rate was 74.6%. Female tuberculosis patients were more likely to achieve cure than males. Adjusting for sex, the cure rate was higher when treatment was provided by tuberculosis control staff in the main health centres rather than by nurses or village health workers at the peripheral level (odds ratio [OR] = 1.60, 95% confidence interval [CI] 1.23-2.09). The absence of sputum smear conversion after 2 months of chemotherapy was associated with defaulting later during treatment (OR = 2.0, 95% CI 1.15-3.57). Adjusting for age and sex, the death rate during treatment was higher in human immunodeficiency virus (HIV) positive than in HIV-negative tuberculosis patients. CONCLUSION: Directly observed treatment is an effective intervention for improving adherence of tuberculosis patients to treatment in a resource-poor country, provided that drugs are effectively delivered to the most peripheral level, and that health staff are adequately trained and regularly supervised. Patients with high bacterial load in initial sputum smears need to be closely supervised, as they are more likely to default from treatment.