4701007_monash_120291.pdf (7.7 MB)
Download file

Key social and behavioural factors influencing clinical outcomes for people with HIV receiving antiretroviral therapy in diverse settings

Download (7.7 MB)
posted on 28.02.2017, 03:24 authored by McMahon, James Hamilton
Since the advent of combination antiretroviral therapy (ART) in 1996 there has been significant reductions in mortality for patients living with HIV where these medications have been available. Despite this advance, socio-behavioural factors such as socioeconomic status (SES) while on ART and adherence to ART can impact outcomes considerably for these individuals. The overarching goal of this thesis was to identify patient specific factors associated with worse clinical outcomes and the identification of methods to better assess adherence to ART in diverse clinical settings. These factors and estimates of adherence can then allow for the implementation of evidence based interventions or the design of targeted interventions that will maximise clinical outcomes for people receiving ART. The studies that were performed to make up this thesis used multiple techniques to identify assessments of ART adherence and socio-behavioural factors that predict clinical outcomes for patients receiving ART. These include systematic reviews of the literature, analyses of prospective cohorts receiving ART in the North-eastern United States and retrospective cohort studies of a population of HIV-infected individuals receiving ART in Tamil Nadu India. Major findings from this group of studies include the first systematic summary of how ART adherence assessments using routinely collected pharmacy data predict survival, virological and immunological outcomes for patients in HICs and LMICs. In addition a framework to categorise pharmacy adherence measures (PAMs) and recommendations on how to best select PAMs to predict clinical outcomes in patients receiving ART was established. Further systematic reviews established normative rates of virological suppression for individuals in LMICs after 12-months ART, and documented the impact of tracing patients who become lost to follow-up (LTFU) after initiating ART. In addition a systematic review of the effects of physical tracing of patients who are LTFU describes how physical tracing may lead to increased re-engagement of patients in care, rather than just improved classification of outcomes for patients considered LTFU. Analyses of a cohort of individuals followed in the North-eastern United States revealed how different markers of SES such as poverty, education level and housing insecurity predict survival despite the use of ART. An additional study identified that individuals assessed as food insecure on even one occasion over multiple years of follow-up was a potent predictor of immunological decline even in the setting of ART. Studies of a retrospective cohort of HIV-infected individuals initiating free ART in Tamil Nadu, India identified multiple factors predicting virological failure after 12-months of ART. Programmatic factors such as prolonged patient travel time to clinic and individual factors such as patients having busy schedules or reporting a history of alcohol use were identified as factors predicting poor virological outcomes after 12-months ART. Furthermore multiple assessments of adherence to ART using different questions about self-reported adherence or pharmacy data identified that PAMs but not self-report measures were most predictive of virological outcomes in this setting. This body of work is a significant contribution as multiple socio-behavioural factors and adherence measures that predict poor outcomes for people receiving ART were identified. Importantly this group of studies have been performed in different populations affected by HIV using different models of HIV care. This gives a unique insight into how different social, cultural and behavioural aspects of the lives of people living with HIV can influence clinical care. Furthermore, these novel patient specific factors and mechanisms to assess ART adherence that were found to predict clinical outcomes are important because they are amenable to implementation, in both HICs and LMICs to alleviate morbidity and mortality for patients receiving ART.


Principal supervisor

Julian Hedley Elliott

Year of Award


Department, School or Centre

Central Clinical School

Campus location



Doctor of Philosophy

Degree Type



Faculty of Medicine Nursing and Health Sciences