RESUMO
BACKGROUND: We sought to investigate the relationship between tuberculosis (TB) treatment outcomes and its predictors in the KwaMashu region in KwaZulu-Natal (KZN). This area is currently a hotbed for TB and human immunodeficiency virus (HIV) co-infection. METHOD: A retrospective study design was adopted to characterise adult patients diagnosed with Gene Expert (GXP) positive pulmonary TB from 01 January 2016 to 31 December 2017. Tuberculosis treatment outcomes were assessed after two months and five months according to the standard World Health Organization (WHO) criteria. Multiple logistic regression analysis was used to calculate the odds ratio (OR) of the possible determinants associated with unsuccessful treatment outcomes. RESULTS: Amongst the 596 patients diagnosed, 57.4% (95% confidence interval [CI]: 53.3-61.4; 342 of 596) had successful treatment outcomes. Of these reported cases, 88.89% (85.1-92.0; 304 of 342) were cured. For the unsuccessful treatment outcomes, 52.4% (46.0-58.6; 133 of 254) patients were lost to follow-up, 20.9% (16.0-26.4; 53 of 254) failed treatment, 1.2% (0.2-3.4; 3 of 254) died and 25.6% (20.3-31.4; 65 of 254) of the patients could not be accounted for. Patients with unknown HIV status were more likely to have unsuccessful treatment outcomes (adjusted OR [aOR] = 4.94 [1.83-13.36]). Patients who had sputum conversion at 2 months (aOR = 1.94 [1.27-2.96]) were significantly more likely to exhibit unsuccessful treatment outcomes. CONCLUSION: Treatment success rate was 57.4% which was below the target set by the WHO. This underscores the urgent need to strengthen treatment adherence strategies to improve outcomes, especially in high HIV burden settings.
