Prevalence of Loneliness and its Predictors and Outcomes in People with HIV

Rationale: Over half of people with HIV (PWH) report loneliness (i.e., perceived social isolation) in their lifetime. PWH may be at increased risk of loneliness compared to people without HIV (PWoH) due to stigma, discrimination, and objective social isolation. Depression can be a consequence of loneliness, which is problematic given its high prevalence in PWH. This dissertation aimed to address gaps in the literature related to loneliness in PWH by examining: 1) the prevalence of loneliness and psychosocial factors that may mitigate it, 2) associations between loneliness and systemic inflammatory biomarkers, and 3) cross-sectional and longitudinal impacts of loneliness on depression. Design: Retrospective data from the Multi-Dimensional Successful Aging Among HIV-Infected Adults Study were extracted. Participants with UCLA Loneliness Scale-Version 3 data were included in analyses (N=202; PWH n=120, PWoH n=82). Studies 1 and 2 used multivariable regressions, and Study 3 used multilevel modeling. Results: Study 1 (cross-sectional; N=202) 41.7% of PWH reported clinically significant loneliness, exceeding rates for PWoH (29.3%; p=.049). Resilience, personal mastery, and affective wisdom were negatively associated with loneliness (ps<.05). PWH were more vulnerable to loneliness than PWoH when cognitive wisdom and emotional support were lower. Loneliness explained significant variance in depressive scores after accounting for demographic, psychological, and social variables. Study 2 (cross-sectional; N=82) In PWH, specific biomarkers of inflammation (D-dimer, CCL2/MCP-1, and sCD14) predicted loneliness after accounting for relevant covariates (ps<.05). There was also a combined association of loneliness and inflammation with depressive symptoms whereby higher D-dimer was associated with increased depressive symptoms only when loneliness was higher (p=.0497). Study 3 (longitudinal; N=139) Individuals who experienced more loneliness than their typical average reporte