In the recent HPTN 052 trial, antiretroviral treatment (ART) dramatically reduced HIV transmission at the level of serodiscordant heterosexual couples, resulting in tremendous enthusiasm for HIV treatment as prevention (TasP) to reduce HIV incidence at the population level. Despite the strong biological plausibility of TasP, behavioral and social factors may limit its true public health impact. To ensure success of TasP, valid, reliable metrics for forecasting HIV incidence and monitoring TasP effectiveness are critically needed. Currently, the US National HIV/AIDS Strategy recommends "community viral load" (CVL) as a metric for this purpose, and the US Centers for Disease Control and Prevention have issued guidance on the collection of CVL and similar aggregate measures of viral load. Unfortunately, these recommendations are based on just a few ecological studies, and aggregate viral load measures have several potential limitations that may interfere with their effective use to predict HIV incidence and reflect TasP impact. To date, these measures have not been subjected to careful evaluation and validation.
The aims of the proposed research are to: 1) Assess the performance of existing and novel TasP metrics as predictors of HIV incidence under a range of scenarios in a hypothetical HIV epidemic among men who have sex with men (MSM), and 2) Assess the performance of existing and novel TasP metrics as markers of TasP effectiveness. For both aims, we will develop a deterministic mathematical model describing HIV transmission, HIV progression, and movement through the HIV care cascade (including testing, entry into care, and receipt of ART). To evaluate the performance of existing metrics in a range of plausible circumstances expected to affect their predictive power, reliability, and validity, we will manipulate the form of TasP, the association between risk behaviors and health- seeking behaviors, the effect of TasP interventions on risk behavior, and the extent of behavior change due to other forces. We will propose and assess several novel metrics in the same way. For the Aim 1 analyses, we will estimate the cross-correlation between each metric and HIV incidence to assess the predictive power of the metric in a given scenario, and we will compare the cross-correlations across scenarios to assess the metric's reliability as a predictor of HIV incidence. For the Aim 2 analyses, we will estimate the change in HIV incidence per unit change in each metric for a given scenario, and we will compare this association across scenarios characterized by varying degrees of confounding and selection bias. This modeling approach will allow the systematic examination of a range of plausible scenarios, as well as the comparison of observable and "true" relationships among TasP, incidence, and metrics in ways that have not been possible in observational studies. This research will represent the first formal evaluation of existing and novel metrics'validity and reliability, allowing researchers, policy makers, and public health officials to make more informed decisions about their use in the critical tasks of predicting HIV incidence and monitoring TasP impact.

Public Health Relevance

Antiretroviral treatment (ART) holds great promise as a means of preventing onward HIV transmission, but its potential impact on HIV incidence at the population level is difficult to quantify. As the use of ART for public health purposes has continued to increase, several metrics for monitoring the population-level effectiveness of ART have been proposed and widely promoted, but these metrics have important potential drawbacks that have not been carefully evaluated. By systematically examining the validity and reliability of these metrics, as well as developing and testing new metrics, this work will allow HIV researchers and policy makers to make more informed decisions about the use of these metrics in monitoring the impact of ART on HIV spread.

Agency
National Institute of Health (NIH)
Institute
National Institute of Mental Health (NIMH)
Type
Small Research Grants (R03)
Project #
1R03MH100987-01
Application #
8541340
Study Section
AIDS Clinical Studies and Epidemiology Study Section (ACE)
Program Officer
Gordon, Christopher M
Project Start
2013-04-01
Project End
2015-03-31
Budget Start
2013-04-01
Budget End
2014-03-31
Support Year
1
Fiscal Year
2013
Total Cost
$76,000
Indirect Cost
$26,000
Name
University of North Carolina Chapel Hill
Department
Chemistry
Type
Schools of Arts and Sciences
DUNS #
608195277
City
Chapel Hill
State
NC
Country
United States
Zip Code
27599
Hurt, Christopher B; Powers, Kimberly A (2014) Self-testing for HIV and its impact on public health. Sex Transm Dis 41:10-2
Miller, William C; Lesko, Catherine R; Powers, Kimberly A (2014) The HIV care cascade: simple concept, complex realization. Sex Transm Dis 41:41-2