Tuberculosis treatment delays and associated factors within the Zimbabwe national tuberculosis programme

Abstract

Background Delayed presentation of pulmonary TB (PTB) patients for treatment from onset of symptoms remains a threat to controlling individual disease progression and TB transmission in the community. Methods A structured questionnaire was administered at 47 randomly selected health facilities in Zimbabwe by trained health workers to all patients aged ≥18 years with microbiologically confirmed PTB who were started on TB treatment and entered in the health facility TB registers between 01 January and 31 March 2013. Multivariate logistic regression was used to calculate adjusted odds ratios (aOR) and 95% confidence intervals (CIs) for associations between patient/health system characteristics and patient delay >30 days or health system delay >4 days. <h3 id='results'>Results</h3> Of the 383 recruited patients, 211(55%) were male with an overall median age of 34 years (IQR, 28-43). There was a median of 28 days (IQR, 21-63) for patient delays and 2 days (IQR, 1-5) for health system delays with 184 (48%) and 118 (31%) TB patients experiencing health system delays >30 days and health system delays >4 days respectively. Starting TB treatment at rural primary healthcare vs district/mission facilities (aOR 2.70, 95% CI 1.27-5.75, p = 0.01) and taking self-medication (aOR 2.33, 95% CI 1.23-4.43, p = 0.01) were associated with encountering patient delays. Associated with health system delays were accessing treatment from lower level facilities (aOR 2.67, 95% CI 1.18-6.07, p = 0.019), having a Gene Xpert TB diagnosis (aOR 0.21, 95% CI 0.07-0.66, p = 0.008) and >4 health facility visits prior to TB diagnosis ((aOR) 3.34, 95% CI 1.11-10.03, p = 0.045). This research was supported by the UK Department for International Development’s Operational Research Capacity Building Programme led by the International Union Against TB and Lung Disease (The Union)