Tuberculosis retreatment outcomes and associated factors: a mixed-methods study from Puducherry, India.

Setting: Four Tuberculosis (TB) Units in the Union Territory of Puducherry in India. Objectives: To identify factors associated with unfavourable outcomes among retreatment TB cases in Puducherry. Design: The study had a mixed-methods design in which routinely reported TB data from retreatment TB cases registered during 2014 and 2015 were used for the quantitative part; the qualitative phase comprised interviews and focus group discussions with health care providers. Results: Among the 392 retreatment cases, 297 (75.8%) had favourable outcomes. Outcome for previous treatment such as loss to follow-up (LTFU) (adjusted relative risk [aRR] 1.6, 95%CI 1.1-2.4, P = 0.001), treatment failure (aRR 1.7, 95%CI 1.04-2.8, P = 0.03) and pre-treatment weight <40 kg (aRR 1.8, 95%CI 1.3-2.5, P = 0.001) had increased risk for unfavourable outcomes. Health care providers reported that alcoholism, lack of family support, job-related issues and lack of access to trained staff for injections were some of the reasons for unfavourable outcomes. Providing incentives, nutritional supplements and early retrieval of the LTFU cases were some of the suggestions to reduce unfavourable outcomes. Conclusion: Outcome of previous treatment and low pre-treatment weight of the patient affected retreatment outcomes. Health professionals and workers highlighted social and health system-related factors. Commitment at all levels on the part of health care providers and addressing their concerns can improve retreatment outcomes.