Exploring individual- and community-level predictors and mediators of suboptimal HIV primary care appointment adherence: the importance of place

Abstract

The advent of highly active antiretroviral therapy (HAART) for the treatment of human immunodeficiency virus (HIV) has provided prolonged viral suppression and an extended quality of life for HIV-infected persons. Research focusing on factors associated with suboptimal HIV medication adherence has been the principal antecedent to more contemporary research suggesting that absolute adherence to HIV primary care appointments is vitally important to achieve successful clinical management of the disease, as well, and missed visits have been independently associated with HIV virologic failure when other factors are controlled. Generally speaking, research examining the role of community correlates in shaping behaviors is somewhat scant in the HIV treatment literature. Existing research suggests that HIV-infected persons may experience unique patterns of negative health outcomes, such as quality of life, access to and retention in treatment, and morbidity. It stands to reason that these patterns may be the result of certain characteristics of many communities that are conducive to poor health, in general, such as low high school graduation rates, high rates of un- and underemployment, substandard housing, and extensive poverty. This study, conducted in two distinct phases, involves secondary data analysis of individual-level factors collected through a prospective cohort study (the 1917 Clinic Cohort) that includes HIV positive individuals who receive primary and sub-specialty medical care at the University of Alabama at Birmingham (UAB) 1917 Clinic. The second phase involves investigation of community-level data (census tract- or county-level) for each participant's residence address at the time of initial presentation to HIV primary medical care. The mediation effect of community-level variables on the association between individual-level predictors and linkage to and retention in care outcomes was then investigated. A non-probability, convenience sample of HIV-infected adults initially presenting for HIV primary medical care at the UAB 1917 Clinic was selected from larger cohort studies, including the CFAR Network of Integrated Clinical Systems (CNICS) and Project CONNECT, a local UAB 1917 Clinic new patient orientation protocol which provides systematic linkage to HIV medical care. Study eligibility included age 19 or over, an HIV diagnosis with a Project CONNECT interview during the study period (7/01/09-6/30/11), normal cognitive functioning, ability to speak and communicate in English, and self-reported residence within the State of Alabama. The principal exposures for this study included three measures of linkage to care (a general linkage to care measure, effective linkage to care, and efficient linkage to care) and five measures of retention in care (appointment no show, visit adherence, visit constancy, gaps in care, and the HRSA-HAB). All measures are widely used in the literature. Bivariate logistic regression was used to determine relationships between individual- and community-level independent variables and the study outcomes. To determine mediation, bivariate logistic regression and Ordinary Least Squares (OLS) regression was used to identify associations between individual- and community-level independent variables and measures of strain. The study replicated existing findings in the literature that have established relationships between individual- and community-level independent variables and linkage to and retention in HIV primary medical care. Additionally, findings suggest that some concepts of community-level strain--specifically, aggregate community measures of age structure, crime, and family structure--may also predict principal outcomes. While a measure of composite strain was not found to mediate the relationship between independent variables and the study outcomes, two aggregate strain measures (age structure and family structure) were found to mediate the relationship between self-reported trouble remembering and HIV transmission risk, respectively, and HIV primary medical care visit adherence. The findings are highly relevant to social work practice, research, and policy in the HIV/AIDS linkage to and retention in care field, and they suggest a number of future directions to further investigate the role of the community--and community strain specifically--in facilitating and/or preventing optimal HIV treatment.