Abstract
Setting: A high tuberculosis (TB) incidence, resource-limited urban setting in Zimbabwe. Objectives: To compare treatment outcomes among people initiated on first-line anti-tuberculosis treatment in relation to age and other explanatory factors. Design: This was a retrospective record review of routine programme data. Results: Of 2209 patients included in the study, 133 (6%) were children (aged <10 years), 132 (6%) adolescents (10-19 years), 1782 (81%) adults (20-59 years) and 162 (7%) were aged ⩾60 years, defined as elderly. The highest proportion of smear-negative pulmonary TB cases was among the elderly (40%). Unfavourable outcomes, mainly deaths, increased proportionately with age, and were highest among the elderly (adjusted relative risk 3.8, 95%CI 1.3-10.7). Having previous TB, being human immunodeficiency virus positive and not on antiretroviral treatment or cotrimoxazole preventive therapy were associated with an increased risk of unfavourable outcomes. Conclusion: The elderly had the worst outcomes among all the age groups. This may be related to immunosuppressant comorbidities or other age-related diseases mis-classified as TB, as a significant proportion were smear-negative. Older persons need better adapted TB management and more sensitive diagnostic tools, such as Xpert(®) MTB/RIF.