Abstract

This research examines the impact of alcohol consumption on immune function in the presence and absence of HIV infection. It is composed of three studies using data collected prospectively by the Multicenter AIDS Cohort Study between 1984 and 2009. With a sample of over a thousand men providing multiple visits spanning several decades, with each visit including in-depth behavioral, clinical, and laboratory follow up, this dataset is sufficiently and uniquely powered to gain new insight into immunologic function. Participants of this study provided information on a biannual basis with the primary exposure assessed by interview and the outcome of interest obtained from multiplex assay measurement of peripheral blood serum samples stored in a repository. Following a repeated measures design, we incorporate a range of parametric and non-parametric statistical methods including ANOVA, mixed models, multivariable linear regression, generalized estimating equations, factor analysis, and structural equation models. Our results indicate that a subset of inflammatory biomarkers is linked to alcohol consumption. Alcohol consumption is most strongly associated with deceased concentrations of several soluble cytokine receptors as well as the acute phase effector C-reactive protein after adjustment for important confounders including age, race, income, BMI, smoking, CD4+ count, and hepatitis infection. This research may be used in the further development of a conceptual model in which latent chronic inflammation, expressed as an inflammatory risk score, can be used to monitor modifiable risk factors for disease outcomes and to delay premature decline in immune function. We conclude that moderate alcohol consumption appears to reduce chronic low-grade inflammation in the presence of HIV-infection with no detectable viral load.