Health Insurance Coverage and Health-related Quality of Life among People Affected by HIV

Abstract

Health insurance coverage is one of the most salient determinants of health and health care access in the U.S., especially among people living with HIV. To achieve viral suppression and experience prolonged wellness, people with HIV require consistent sustained access to care. The insured are more likely to adhere to antiretroviral treatment (ART), experience sustained viral suppression, and access mental health care for emotional and behavioral needs—all associated with attenuated HIV disease progression and reduced HIV transmission. However, many people living with HIV in the U.S. are uninsured or underinsured. Opportunities for health insurance are socioeconomically patterned and often influenced by policy. Given that HIV disproportionately impacts populations living in poverty and otherwise marginalized communities, people living with HIV and those at risk of acquiring HIV may experience barriers to securing coverage. Moreover, those who do secure coverage may not have access to resources that address their specific health care needs. Understanding how insurance coverage impacts health is necessary to inform interventions and initiatives to end the HIV epidemic. Previous studies have considered the relationship between insurance coverage and clinical outcomes along the HIV Care Continuum; however, no studies have investigated the effect of insurance coverage on holistic wellness. Optimizing health-related quality of life (HR-QoL) is increasingly considered key target of global strategies to combat the AIDS epidemic. Understanding how insurance coverage affects HR-QoL could help to identify unmet needs under current coverage strategies and guide health policy decision-making. In the absence of this evidence base, we used high quality observational data, robust causal inference methods, and a novel approach to effect heterogeneity to investigate the relationship between health insurance coverage and HR-QoL among people affected by HIV. In Chapter 1, we examined the association between health insurance coverage and health-related outcomes including clinical HIV-related measures (viremia, CD4+ T cell count, and ART adherence) and HR-QoL subscales among youth and young adults in the multi-site PHACS AMP Up series. Our findings suggest that privately insured young adults living with HIV have lower risk of ART nonadherence compared to those with public insurance. Both young adults with private coverage and young adults who were unaware of their coverage, regardless of HIV status, had lower risk of suboptimal role functioning compared to the publicly insured. Lastly, compared to those with public coverage, young adults with private insurance had better health perceptions. For youth, we observed little difference in health-related outcomes across insurance coverage status. These findings suggest that insurance coverage may have meaningful influence on some HR-QoL metrics and highlight the need to further explore the mechanisms through which insurance coverage may shape health. In Chapter 2, we emulated a target trial of health insurance strategies and 4-year HR-QoL among women living with HIV and women at risk of acquiring HIV in the Women’s Interagency HIV/AIDS Study (WIHS). We found little difference in the comparative effectiveness of several strategies compared to the observed, natural course of insurance coverage experience among these women. Sensitivity analyses using a negative control outcome were conducted to investigate the presence of unmeasured confounding by health engagement/health literacy. We found higher risk of our negative control outcome, loss to follow up due to incomplete data, among uninsured women. Though the possibility of unmeasured confounding cannot be excluded, these findings highlight the need for additional research to evaluate how these strategies may impact the health of specific subpopulations to inform coverage provision and optimized allocation of resources. In Chapter 3, we applied a novel risk modeling approach to identify potential heterogeneity in the treatment effects of insurance coverage on HR-QoL. We predicted 4-year risk of low health-related quality of life index (HQLI) using LASSO-regularized logistic regression and conducted risk-stratified comparative effectiveness analyses using the parametric g-formula. The findings suggest that HQLI was qualitatively similar across risk groups after several insurance coverage strategies. As in Chapter 2, our negative control analysis suggested potential unmeasured confounding by health engagement/health literacy. This study demonstrates the merits of applying a risk modeling approach to explore effect heterogeneity as an alternative to “one-variable-at-a-time” analyses. These findings inform ongoing policy dialogue on insurance provision for low-income populations living with chronic disease.