BACKGROUND: The Health Resources and Services Administration's Ryan White HIV/AIDS Program provides services to more than half of all people diagnosed with HIV in the United States. We present and validate a mathematical model that can be used to estimate the long-term public health and cost impact of the federal program. METHODS: We developed a stochastic, agent-based model that reflects the current HIV epidemic in the United States. The model simulates everyone's progression along the HIV care continuum, using 2 network-based mechanisms for HIV transmission: injection drug use and sexual contact. To test the validity of the model, we calculated HIV incidence, mortality, life expectancy, and lifetime care costs and compared the results with external benchmarks. RESULTS: The estimated HIV incidence rate for men who have sex with men (502 per 100,000 person years), mortality rate of all people diagnosed with HIV (1663 per 100,000 person years), average life expectancy for individuals with low CD4 counts not on antiretroviral therapy (1.52-3.78 years), and lifetime costs ($362,385) all met our validity criterion of within 15% of external benchmarks. CONCLUSIONS: The model represents a complex HIV care delivery system rather than a single intervention, which required developing solutions to several challenges, such as calculating need for and receipt of multiple services and estimating their impact on care retention and viral suppression. Our strategies to address these methodological challenges produced a valid model for assessing the cost-effectiveness of the Ryan White HIV/AIDS Program.
BACKGROUND: The Health Resources and Services Administration's Ryan White HIV/AIDS Program provides services to more than half of all people diagnosed with HIV in the United States. We present and validate a mathematical model that can be used to estimate the long-term public health and cost impact of the federal program. METHODS: We developed a stochastic, agent-based model that reflects the current HIV epidemic in the United States. The model simulates everyone's progression along the HIV care continuum, using 2 network-based mechanisms for HIV transmission: injection drug use and sexual contact. To test the validity of the model, we calculated HIV incidence, mortality, life expectancy, and lifetime care costs and compared the results with external benchmarks. RESULTS: The estimated HIV incidence rate for men who have sex with men (502 per 100,000 person years), mortality rate of all people diagnosed with HIV (1663 per 100,000 person years), average life expectancy for individuals with low CD4 counts not on antiretroviral therapy (1.52-3.78 years), and lifetime costs ($362,385) all met our validity criterion of within 15% of external benchmarks. CONCLUSIONS: The model represents a complex HIV care delivery system rather than a single intervention, which required developing solutions to several challenges, such as calculating need for and receipt of multiple services and estimating their impact on care retention and viral suppression. Our strategies to address these methodological challenges produced a valid model for assessing the cost-effectiveness of the Ryan White HIV/AIDS Program.