Evaluating Portfolio Interventions for HIV Incidence Reduction in the United States: Development of a Novel Agent-Based Decision-Analytic Model for Dynamic Evaluations of Interventions

点击翻译按钮获取中文摘要

Project Summary/Abstract The number of people living with the human immunodeficiency virus (HIV) in the United States has been gradually increasing from 800,000 in the late 1990's to 1.2 million by 2011. This is partially due to HIV-infected persons living longer and closer to normal life years on highly-active antiretroviral therapy (ART) treatment for HIV. However, the number of persons becoming newly infected with HIV has not decreased in recent years, it has been stable at almost 50,000 persons each year since the late 1990's. While mortality due to acquired immune deficiency syndrome (AIDS) related causes have been decreasing because of effective ART treatment, it is estimated that persons with HIV could be at a higher risk of certain non-communicable diseases that could lead to mortality. Further, the lifetime costs of treating an HIV-infected person on ART is very high, ranging from USD 250,000 to USD 400,000. Thus, it is important to identify optimal investment strategies for the prevention of new infections, which will reduce future HIV-related disease burden and costs. The overall goal of the project is to identify population-specific cost-effective combinations of care and behavioral intervention measures (intervention portfolios) that would help reduce new infections. The U.S. National HIV/AIDS Strategy (NHAS), 2015, proposes a goal of a 25% reduction in new infections by year 2020 compared to 2010. The analyses from this proposal would inform the development of a national strategy for achieving the NHAS goal for 2020 and similar such strategies in the future. It specifically proposes to advance theoretical concepts for the development of novel structure and algorithms for individual-level simulation of contact dynamics for disease spread (Aim 1), construction of a new agent-based decision-analytic model for dynamic evaluations of HIV interventions in the US (Aim 2), and development and implementation of new algorithms for evaluation and identification of optimal intervention portfolios for HIV prevention in the US (Aim 3). In this age of `big data' and computational power for analyzing these data, development of innovative methodologies for simulating the complicated dynamics of disease spread, and integrating disparate data sources to derive significant information that otherwise cannot be inferred through any of the data sources independently, could significantly improve use of decision-analytic models for evaluation of national strategies for disease prevention. Models can also further inform data collection for more accurate design of models and intervention analyses in the future. The theoretical knowledge gained through Aim 1 could also be foundational for the development of new methodologies for real-time decision-analyses during outbreak of emerging infectious diseases, similar to previous disease outbreaks such as Ebola-virus Disease or the Middle-Eastern Respiratory Syndrome.