Predictors of Treatment Failure among Patients with Pulmonary Tuberculosis Attending Public Health Facilities in Nairobi County, Kenya

Abstract

TB treatment failure is defined as the persistence of a positive sputum smear or culture at five months or later during treatment, or the need to change treatment due to lack of clinical or bacteriological response. Tuberculosis (TB) is a major public health challenge in Kenya, which remains among the high TB burden countries despite steady progress in control efforts. In 2023, Kenya notified over 97,000 TB cases and achieved a treatment success rate of approximately 88–89% among patients started on anti-TB therapy, with an estimated 25% HIV co-infection rate among TB cases. However, a proportion of patients still experience poor outcomes, including treatment failure, relapse, loss to follow-up, and death. Identifying factors associated with treatment failure is critical to strengthening TB control and reducing morbidity, mortality, and transmission in Kenya. This study aimed to identify socio-demographic, behavioral, and clinical factors associated with TB treatment failure among TB patients in Nairobi County, Kenya. An unmatched case–control design was utilized, involving 81 TB patients with treatment failure (cases) and 162 TB patients who successfully completed treatment and were declared cured (controls). Controls were selected using simple random sampling from TB registers within the same health facilities as the cases, at a ratio of 1:2.Data was collected using structured data abstraction from patient interviews and medical records, focusing on socio-demographic, behavioral, and clinical characteristics. Statistical analysis was conducted using SPSS version 23. Univariate, bivariate, and multivariate analyses were performed, with 95% confidence intervals (CIs). Odds ratios (ORs) were calculated using unconditional logistic regression to determine factors independently associated with treatment failure. Patients with TB treatment failure were significantly more likely to have a history of prior exposure to first-line anti-TB drugs (OR = 85.0; 95% CI: 29.7–243.3; p < 0.0001). A positive sputum smear at two months of treatment, indicating delayed sputum conversion, strongly predicted treatment failure (OR = 20.63; 95% CI: 5.42–78.41; p = 0.0021). Poor adherence to treatment, shown by missed doses or appointments, was also significantly associated with treatment failure (OR = 4.7; 95% CI: 2.1–10.6; p = 0.004). Conversely, HIV-positive status (OR = 0.34; 95% CI: 0.1–0.9; p = 0.025) and participation in the Directly Observed Therapy (DOT) program (OR = 0.23; 95% CI: 0.1–0.6; p = 0.002) were associated with a reduced risk of treatment failure, suggesting a protective effect. Ethical approval was obtained from the Jomo Kenyatta University of Agriculture and Technology (JKUAT) Board of Postgraduate Studies and the Kenyatta National Hospital–University of Nairobi Ethical Review Committee. Permission to conduct the study was granted by the Nairobi County Health Department and the National Tuberculosis, Leprosy and Lung Disease Program. Written informed consent was obtained from all participants, and confidentiality of patient information was strictly maintained. In the Kenyan context, TB treatment failure is significantly associated with retreatment history, delayed sputum conversion at two months, and poor adherence to therapy, while engagement in DOT and HIV-positive status were protective against failure. To address these challenges, routine bacteriological monitoring (e.g., sputum microscopy or GeneXpert) at two months of treatment should be strengthened, along with enhanced implementation of DOT and comprehensive contact tracing for patients at risk of treatment failure. These measures are critical to improving treatment outcomes and advancing Kenya’s progress toward ending TB by 2030.