Cohort analysis of treatment outcomes from facility-based and community-based tuberculosis care in Botswana: a cross-sectional survey

Abstract Background Community-based directly-observed treatment short course is a recent addition to the strategies for delivering anti-tuberculosis (TB) treatment in Botswana. The Botswana National Tuberculosis Program aims to have at least 75% of all patients with TB on community TB care (CTBC). However, prior to this study, the treatment outcomes had not yet been evaluated. This study aimed to compare the treatment outcomes for patients on CTBC with those on facility-based TB care (FTBC). Methods A retrospective review of TB registers was conducted for a cohort of patients receiving TB treatment between Jan 1 and June 31, 2012, in 14 districts. Data were abstracted on sputum results, age, sex, HIV status, treatment outcome, and options. A sample size of 1000 patients was obtained. The outcome of interest was treatment outcome. It was a binary variable with good (cured, completed, or success) and poor (died, defaulted, or failed) as the outcomes. Data were analysed using STATA SE/12. Findings The study had 511 men and 489 women. Mean age was 35 years (SD 2·7). FTBC accounted for 496 patients, while 504 were on CTBC. Of the patients who received FTBC, 113 (22·8%) were cured, 277 (55·8%) completed treatment, 390 (78·6%) had treatment success, 21 (4·2%) defaulted, 11 (2·2%) failed, and 74 (14·9%) died. Of the patients who received CTBC, 118 (23·4%) were cured, 333 (66·1%) completed treatment, 451 (89·5%) had treatment success, ten (2·0%) defaulted, seven (1·4%) failed, and 36 (7·1%) died. Logistic regression showed that patients on CTBC were 2·3 times more likely to have good treatment outcomes compared with those on FTBC. Older age and HIV-positive status were associated with poor treatment outcomes (odds ratio 0·9752, 95% CI 0·9633–0·9872, p=0·001; and 0·4071, 0·2874–0·5765, p=0·001, respectively). Interpretation CTBC offers better treatment outcomes than FTBC care in the Botswana setting. Results show the urgent need to expand CTBC so as to improve coverage rates. Improvements in TB case finding and stigma reduction in the Botswana context are still needed. Funding This research was funded through the I-TECH Botswana/HRSA Budget.