Symposia
Health Psychology / Behavioral Medicine - Adult
Jasper S. Lee, M.S.
Predoctoral Fellow
Harvard Medical School / Massachusetts General Hospital
BOSTON, Massachusetts
Sierra Bainter, PhD
Assistant Professor
University of Miami
Coral Gables, FL
Alexander Tsai, MD, PhD
Associate Professor
Harvard Medical School
Boston, Massachusetts
Jessica F. Magidson, PhD
Assistant Professor
University of Maryland
College Park, Maryland
Lena S. Andersen, Ph.D.
Senior Research Officer
University of Cape Town
Cape Town, Western Cape, South Africa
John Joska, MBChB, FC, MMed, PhD
Professor
University of Cape Town
Cape Town, Western Cape, South Africa
Conall O'Cleirigh, Ph.D.
Director, Behavioral Medicine
Harvard Medical School/ Massachusetts General Hospital,
Boston, Massachusetts
Steven A. Safren, Ph.D., ABPP
Professor of Psychology
University of Miami
University of Miami, Florida
Introduction: South Africa has the highest number of HIV cases globally. There is also a high prevalence of both comorbid and synergistic mental health and structural epidemics (i.e., syndemics) that act as barriers to HIV care. This study was the first to assess the extent to which HIV is syndemic with mental health problems among people with HIV (PWH) receiving care in South Africa, and the first to compare additive and interaction effects approaches to modeling syndemics.
Method: We enrolled 194 PWH at public primary care clinics in Khayelitsha, a peri-urban informal settlement near Cape Town, and assessed for the presence of 10 different psychosocial and structural problems, including depression, substance use, and food insecurity. HIV viral load and antiretroviral therapy (ART) adherence (pharmacy refill) were extracted from medical records approximately 12 months later. We used network analysis to examine interrelationships of syndemic problems, and to empirically specify interaction terms among syndemic problems. Regression models systematically compared additive and interaction effects (based on network analysis results) approaches to modeling the effects of syndemic problems on viral load and ART adherence.
Results: The results of the network analysis are presented graphically. The most central variables in the network were depression, substance use, and food insecurity. Results models examining ART adherence suggested higher number of syndemic problems may be associated with worse ART adherence via a statistical trend (b= -2.71, 95%CI [-5.92 – 0.50], p = .098). In assessing for potential interactive relationships, we estimated a statistically significant interaction of depression and food insecurity (b=0.04, 95%CI [0.001–0.08], p = .030), such that at high levels of food insecurity, depressive symptoms did not have an effect on ART adherence, but at lower levels of food insecurity, increasing severity of depressive symptoms was associated with worse ART adherence.
Discussion: This study was the first to: explore the effects of psychosocial and structural syndemic problems on HIV outcomes in South Africa; utilize an empirically specified interaction effects approach to examine syndemic problems; and systematically compare additive and interaction effect approaches. Findings support the use of interaction terms in modeling syndemics. Findings may inform intervention development in South Africa, via a transdiagnostic and potentially multilevel intervention.