Date of Award

Summer 8-11-2020

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Public Health

First Advisor

Matthew J. Hayat

Second Advisor

Matthwe J. Magee

Third Advisor

Kenneth G. Castro


Globally, nearly 3% of the 38 million people living with HIV (PLHIV) died in 2018. Underlying comorbidities in PLHIV present critical challenges to clinical care and contribute to increased mortality in PLHIV. Key infectious disease comorbidities include tuberculosis (TB), hepatitis B virus (HBV), and hepatitis C virus (HCV). Given the rapid expansion of type 2 diabetes mellitus (T2DM) worldwide, T2DM is an emerging non-infectious disease comorbidity among PLHIV. All four of these comorbidities are associated with increased mortality in PLHIV. Currently there is an urgent need to develop clinical guidelines for HIV programs that integrate prevention, screening, and treatment for comorbidities. However, existing data that quantifies the impact of comorbidities on HIV outcomes are limited, especially in resources limited settings. To answer critical questions related to the effects of comorbidities on health outcomes in PLHIV, we conducted three observation studies using the clinical data of PLHIV from Myanmar.

In study 1, we assessed biological interaction between hyperglycemia and low body mass index (BMI) on the risk of TB disease among patients entering HIV care. Cox proportional hazard models were used to estimate the rates of TB disease due to hyperglycemia, due to low BMI, and due to joint exposure (hyperglycemia and low BMI). We used continuous, categorical, and spline measures of hyperglycemia and low BMI to model the dose-response relationship with TB incidence. We performed sensitivity analyses to assess the direction and magnitude of bias in estimation of the association between low BMI and hyperglycemia with TB incidence due to unmeasured confounders, exposure misclassification, and competing risks. We reported a biological interaction between BMI and hyperglycemia on the risk of TB disease [relative excess risk of TB disease due to joint exposure: 0.42, 95% CI:0.07, 0.78].

Study 2 focused on the relationship between TB disease and key clinical outcomes in PLHIV, including all-cause mortality, CD4+T cell response, and virological failure. Study 2 also assessed the role of glycemic status as a mediator in the relationship between active TB and all-cause mortality. This study used log binomial regression, general linear mixed models, and mediation analysis to achieve the analysis objectives. After one-year of follow-up time, we observed an association between active TB with all-cause mortality [adjusted hazard ratio (aHR): 1.75, 95% CI: 1.38, 2.21]. The mean log CD4 cell count during the follow-up was significantly lower in PLHIV with active TB compared to those without active TB [βTB = -0.61, 95% CI: -0.71, -0.51, βTB*follow-up month= 0.04, 95% CI: 0.03, 0.05]. However, we did not observe glycemic level to be a key mediator in the relationship between TB disease and all-cause mortality.

Study 3 examined whether associations between HBV/HCV coinfection and all-cause mortality differed by hyperglycemia status among PLHIV. We used Cox proportional hazards models to compare all-cause mortality rates in participants by HCV/HBV coinfection and hyperglycemia status. We reported coinfections with hepatitis increased rates of all-cause mortality [aHR: 1.25, 95% CI: 1.11, 1.40] for HBV, 1.52, 95% CI: 1.35, 1.71] for HCV, and 2.14, 95% CI: 1.51, 3.02] for HBV/HCV coinfection compared to HIV mono-infection]. The observed associations were greater in HCV and HBV/HCV coinfected patients with hyperglycemia [aHR: 1.85, 95% CI: 1.23, 2.78 for HCV and 4.39, 95% CI 1.51, 12.76 for HBV/HCV coinfection].

Findings from this dissertation will inform clinical care of PLHIV in low- and middle-income countries. Our results suggest that 1) PLHIV with hyperglycemia and low BMI represent a subgroup with high risk of TB incidence, 2) PLHIV with active TB experience higher mortality and lower mean CD4 cell counts during the follow-up, and 3) hepatitis coinfection and hyperglycemia are associated with higher risk of mortality in PLHIV. We highlight in this dissertation the need for guidelines to prevent and manage TB, HBV, HCV and hyperglycemia in PLHIV.


File Upload Confirmation