Improving Health and Housing Outcomes through a Simulation and Economic Model (iHOUSE Model)

University Of Colorado Denver, Aurora CO

Investigators

Abstract

Homelessness increased by 45% from 2020 to 2021, largely from economic hardship caused by the COVID-19 pandemic and has reached all-time highs across the US. More than 1.25 million people experience homelessness in the US at some point in a year, with some groups being affected more heavily. Denver and San Francisco are two cities in which homelessness is at crisis levels. Homelessness increased by more than 30% since 2022 in Denver and San Francisco has the highest prevalence of homelessness in the US. Homelessness can lead to and is associated with profound health effects. People who experience homelessness die on average 30 years earlier than other Americans and have increased risk of substance use disorders and incident HIV than people who are stably housed. It is critical that we develop feasible, effective, and cost-effective tailored approaches to improve health among all people in the US, specifically with regard to life expectancy, overdose, and HIV. Yet, while effective, evidence-based solutions that improve health outcomes exist, they are far from ubiquitously implemented. Homelessness and the health issues among people experiencing it are heterogeneous, driven by locale-specific factors. Thus, locale-specific solutions are needed urgently. Simulation models can quickly fill knowledge gaps by serving as laboratories for testing hypotheses in real-time. Models that simulate the housing continuum can be important in augmenting randomized trials that provide evidence on system innovations, projecting the impact on health and costs. Our goal is to provide an evidence base for the prioritization and optimization of strategies to improve health outcomes of people experiencing homelessness including decreasing HIV incidence, reducing fatal and nonfatal overdose, and increasing life expectancy. In Aim 1, we will employ a Group Model Building (GMB) approach to engage stakeholders in the scientific process and inform the development of agent-based models that simulate health outcomes along the housing and homelessness continuum of care. In Aim 2, we will develop agent-based models simulating the dynamic processes contributing to overdoses, HIV, and life expectancy among people along the housing continuum of care in Denver and San Francisco. In Aim 3, we will simulate and compare HIV and substance use service delivery programs implemented along the housing continuum versus housing-centered programs. We will assess changes in HIV and overdose rates among the general population experiencing homelessness. We will also examine the potential of these programs for specific population health outcomes as well. Through this proposal, we will develop a national resource that generates the scientific knowledge needed to improve health among people experiencing homelessness, decrease overdose, and end the HIV epidemic.