Factors associated with tuberculosis treatment outcomes among patients newly diagnosed as Mycobacterium tuberculosis sputum smear positive in Nairobi County

Abstract

Tuberculosis (TB) is one of the infectious diseases of public health concern globally. In 2017, it is estimated that 10 million people developed TB globally. More than 1.3 million of the TB cases were notified in the African region. During the same period, Kenya reported 85,188 cases of TB. The research aimed at determining factors associated with TB treatment outcomes among patients newly diagnosed as Mycobacterium tuberculosis sputum smear-positive within Nairobi County. A prospective cohort study of 291 patients from 25 health facilities in Nairobi county was conducted between December 2014 and July 2015. Purposive sampling was applied to include facilities with the highest caseloads of TB. The facilities included public, private and faith-based offering either TB treatment only or both TB diagnosis and treatment categorized as level II, III and IV according to Kenya Essential Package for Health classification. The allocation of the number of study participants to the facilities was done using probability proportional to size (PPS). All patients who consented to the study were included in the study for the six month treatment period. Questionnaires were administered within the first three weeks of treatment and after twelve weeks of treatment. After six months of treatment, TB registers were reviewed to collect information on treatment outcomes. Questionnaires were administered to the facility in-charges once during the study period. Double entry of all the data collected was done. There was validation to check the concordance of the two data sets. A descriptive analysis of the data was undertaken. Bivariate analysis of patientlevel factors, institutional-level factors and treatment outcomes was conducted using ChiSquare and Fisher’s exact test. Kaplan-Meier estimator was used in determining the median time to treatment interruption. Survival was analyzed using the Kaplan-Meier probability of failure estimate. The test for the equality of the survivor functions for the level of education, use of alcohol, smoking,perceived availability of sufficient Health Care Workers and nature of facility was done using the log-rank test. Cox regression hazard analysis was undertaken to determine the predictors of treatment interruption. Statistical significance was determined by considering a nominal p-value of less than 5% (P< 0.05) with a 95% confidence level. The highest level of education, affliction with another chronic disease, access to information on TB, nature of the facility, and level of the facility according to Kenya Essential Package for Health classification, all exhibited a statistically significant association with TB treatment outcomes (P<0.05, Fisher’s exact test). Patients who indicated secondary level as the highest level of education posted lower treatment success rates when compared with their counterparts who had achieved primary level education. Cases afflicted by other chronic disease had lower treatment success rates when compared to those who were not affected. Access to TB information showed an association with positive treatment outcomes. Patients treated in private-for-profit and faith-based institutions showed better treatment outcomes compared to those treated in public facilities. Patients treated in Level II facilities (dispensaries) posted positive treatment outcomes when compared to those in Level III facilities (Health Centers). A total of 19 (6.5%) treatment interruptions were observed. The median time to default was 56 [95% CI, 36-105] days. Treatment in a non-public facility [AHR=0.253, 95% CI (0.0585-1.097)] and facilities perceived to have an adequate number of health care workers to offer Directly Observed Therapy (DOT) [AHR=0.253, 95% CI (0.0919- 0.697)] showed a lower hazard for treatment interruption. Attainment of secondary level education [AHR=3.42, 95% CI (0.99-11.815)] exhibited a higher hazard rate of treatment interruption when compared to patients who attained a primary level education. Patientlevel and institutional-level characteristics that exhibited a significant association with treatment outcomes in this study, should be factored into the treatment plans for new SM+ TB patients in Nairobi County to achieve higher treatment success rates. These variables should be considered predictors of treatment outcome during TB treatment in Nairobi County. Non-clinical aspects of health care service delivery influence patient adherence to TB treatment. Subsequently, the health-seeking behavior of groups considered to be at high risk for treatment interruption should be incorporated into the design and delivery of TB treatment.