Background: In limited resource settings, sputum smear conversion at the end of the intensive phase of tuberculosis treatment is an indicator not only of patients' response to treatment, but also of anti-tuberculosis program performance. The objective of this study was to identify factors associated to sputum smear non-conversion at the end of the intensive phase of treatment, and the effect of smear non-conversion on the outcome of smear-positive pulmonary tuberculosis patients.
Method: This retrospective cohort study was carried out on data of patients treated in the Diagnostic and Treatment Centre of Baleng, West-Cameroon from 2006 to 2012. Logistic regression models were used to evaluate the association of socio-demographic and clinical factors with delay in sputum smear conversion, and the association of this delay with treatment outcomes.
Result: Out of 1425 smear-positive pulmonary tuberculosis patients treated during the study period, 1286 (90.2%) were included in the analysis. Ninety four (7.3% CI: 6.0- 8.9) patients were identified as non-converted at the end of the intensive phase of treatment. Pre-treatment smears graded 2+ and 3+ were independently associated to delay in smear conversion (p<0.01). Years of treatment ranging from 2009 to 2012 were also associated to delay in smear conversion (p<0.02). Delay in smear conversion was significantly associated to failure [Adjusted Odd Ratio (AOR):12.4 (Confidence Interval: CI 4.0- 39.0)] and death, AOR: 3.6 (CI 1.5- 9.0).
Conclusion: Heavy initial bacillary load and treatment years ranging from 2009 to 2012 were associated to sputum smear non-conversion at the end of the intensive phase of TB treatment. Also, delay in smear conversion was associated to unfavorable treatment outcomes. Patients with heavy initial bacillary load should thus be closely monitored and studies done to identify reasons for the high proportion of non-conversion among patients treated between 2009 and 2012.