Factors Associated with Default from Treatment among Tuberculosis Patients in Nairobi Province, Kenya
Tuberculosis is caused mainly by Mycobacterium tuberculosis.' It affects all tissues and organs except hair teeth and nails. Over 2 billion people were estimated to be infected with the tubercle bacilli in 2005. The immune system is able to contain the bacillus. Only 10% of infections progress to clinical disease. Over 90% of global TB cases and deaths occur in the developing countries. The WHO estimates an incidence of 207,311 new cases in Kenya and 44,576 tuberculosis related deaths annually. Kenya is ranked io" among countries with high tuberculosis burden. Tuberculosis treatment requires use of combination of drugs for 6-8 months. Adherence is vital for successful cure and prevention of drug resistance and treatment failure. In the year 2005, 7.6% of patients defaulted from treatment nationally. The high default rate in the country impedes the achievement of the global target to successfully treat 85% of detected TB cases. Treatment adherence is a complex issue and improving treatment outcomes for tuberculosis requires a full understanding of the factors that prevent people from taking medicines correctly and those that help them complete their treatment. Determination of predictive factors for default was thus justified for early interventions and for policy and strategy formulation to address non-compliance of TB treatment. Default from treatment enhances risk of developing multi-drug resistant tuberculosis, lowers treatment success rates and increases cost of treatment. The objective of this study was to determine factors associated with default from tuberculosis treatment in Nairobi. A Case-Control study was used. Defaulters formed the case and the successfully treated the control group. Secondary data from conveniently sampled treatment facilities was used. Further, cases and controls were traced and interviewed using a structured questionnaire. The response variable was default outcome. Independent variables included drug side-effects, knowledge on TB, access to health care, stigmatization, HIV co-infection and demographic and socioeconomic factors among others. Data was analyzed using SPSS and Epi Info statistical software. Descriptive statistics and analyses of contingency tables to determine association were used. Chi-Square, Fishers exact tests and confidence intervals were used to establish significance. Multivariate logistic regression modeling of associated factors and Kaplan-Meier method to determine probability of staying in treatment over time were employed. Results revealed a 16.7% prevalence of treatment default in Nairobi. Default occurred most frequently during the initial three months of treatment. Among defaulters who were AFB smear positive at initiation of treatment, 47.7% defaulted before conversion was confirmed. Major reasons for default included ignorance, traveling, feeling better, side effects, opting for herbal medication, alcohol use, inadequate food, poor facility factors and stigma. Factors independently associated with default included HIV co-infection (OR 1.56, P<O.OOl), the male sex (OR 1.43, P<O.OOl), history of previous default (OR 2.33, P=0.017), herbal medication use (OR 5.7, P=0.017), low income (OR 5.57, P=0.04), inadequate knowledge on TB (OR 8.67, P=0.017) and alcohol use (OR 4.97, P=0.007). Findings from this study indicate that enhanced health education on TB, pre-treatment counseling, advocacy on treatment compliance, social support and integration of TB and HIV services should be prioritized by MOH so as to address tuberculosis treatment default.