UCLA

Background: Despite significant public health advancements in prevention, screening and treatment of tuberculosis (TB), mortality with TB remains unacceptably high. The epidemiology of TB mortality is not well understood, and therefore urgent investigation of potentially modifiable aspects of TB care that could inform public health service delivery and improve patient outcomes is needed. The primary aim of this study was to evaluate the relationship between supervision of TB treatment and risk of TB death during TB treatment.

Methods: A retrospective cohort study was conducted to estimate the effect of health provider supervision on the risk of all-cause TB mortality during TB treatment among clinically and/or laboratory verified active TB cases alive at time of diagnosis and who started treatment in Los Angeles County during 2010-2014.

Results: Of the 3016 TB cases included in the analysis, 283 (9.4%) died. TB treatment was solely supervised by a provider outside of the health department for 508 (19.2%) case-patients. Bivariate analysis revealed significant differences in age, TB disease characteristics and distribution of social, behavioral and common comorbid conditions between patients with TB treatment supervision solely by providers outside the health department (HD), compared to patients who had any HD supervision during their TB treatment. After adjustment for potential confounding factors, patients managed solely by providers outside the HD were found to have increased risk of all-cause mortality during TB treatment, compared to patients who had any HD involvement in the supervision of their TB treatment (adjusted risk ratio [aRR] 2.33, 95% CI 1.80-3.01).

Conclusion: Patients who did not have any HD involvement in their TB treatment supervision experienced over twice the risk of death as those patients with any HD supervision in their TB treatment. Further investigation of TB treatment practices and case management strategies by providers outside the HD is warranted to understand potential mechanisms for the observed increased risk of TB mortality.