Advisory Committee Chair
Emily B Levitan
Advisory Committee Members
Eric Chamot
E Michael Foster
German Henostroza
Stewart E Reid
Document Type
Dissertation
Date of Award
2013
Degree Name by School
Doctor of Philosophy (PhD) School of Public Health
Abstract
Zambia is a country in Southern Africa with dual HIV and tuberculosis (TB) epidemics. The annual TB incidence is 444/100,000 and 64% of TB patients are co-infected with HIV. Accurate screening and diagnosis for TB is challenging since it often presents with non-specific signs and symptoms and the available diagnostics are neither sensitive nor specific. Papers one and two of this dissertation focus on the derivation of TB screening rules to more accurately identify high-risk persons in two Zambian populations: inmates and HIV clinic patients. Among inmates, being HIV positive and having a body mass index < 18.5 most strongly predicted HIV disease. Chest pain was also strongly associated, but this may have been a chance finding. For HIV clinic patients, we developed four rules with increasing complexity of clinical tools used for screening. Sensitivity was very high with all rules; the addition of vital signs and pulmonary exam findings to symptom-based screening rules greatly increased the rules' specificity. Furthermore, a rule incorporating pulmonary exam findings performed similarly to a rule with chest radiography suggesting pulmonary exam may be an inexpensive alternative to x-ray. The final paper looks at TB incidence, mortality and intermediate clinical outcomes in a cohort of HIV clinic patients with and without TB during their first 12 months of HIV care. TB incidence was very high during this time period, despite thorough screening at baseline. Mortality was elevated in both culture-confirmed and clinically-diagnosed TB patients, with much of the elevated risk concentrated in TB patients who had culture-confirmed but clinically-unapparent disease. Looking at intermediate outcomes among patients who started anti-retroviral therapy, the change in CD4 cell count was similar for patients with and without TB. TB patients started off with lower mean body weights and hemoglobin concentrations but also showed greater increases over 12 months, thus `catching up' with the TB-negative patients. If these findings are replicable in other cohorts they could have substantial impact on TB screening and diagnostic procedures in resource limited, high burden countries like Zambia.
Recommended Citation
Harris, Jennifer, "Tuberculosis Screening and Treatment Outcomes in Zambia" (2013). All ETDs from UAB. 1879.
https://digitalcommons.library.uab.edu/etd-collection/1879