Collaborative Research: III: Small: Data-driven and AI-powered Innovations for Tackling the Nexus of Poverty-driven Food Insecurity and the Opioid Crisis

About This Grant

Today, society faces critical issues of poverty-driven food insecurity and the opioid epidemic: in 2023, 36.8 million Americans lived in poverty, 47.4 million experienced food insecurity, and 8.9 million misused opioids. As these crises operate synergistically, addressing poverty-driven food insecurity while mitigating opioid-related harms presents a pressing and complex societal challenge. Although efforts have been made to tackle these issues separately and to explore their interconnections, research on developing effective, integrated interventions tailored to affected populations is lacking. To bridge this gap, by harnessing the big data revolution and advancing artificial intelligence (AI) technologies, the goal of this project is to design and develop a data-driven, AI-augmented paradigm to investigate the intersection of poverty-driven food insecurity and the opioid crisis and develop integrated, personalized interventions for affected individuals to address the intertwined challenge, and thus help enhance national public health, safety, and welfare. The project outcomes (e.g., open-source code, benchmark data, models, and findings) will be made publicly accessible and broadly distributed through demos, publications, and media presses, etc. This project will integrate research with education, including novel curriculum development, student mentoring, professional training and workforce development, and K-12 outreach activities. Tackling the nexus of poverty-driven food insecurity and the opioid crisis is an urgent societal priority. To achieve this goal, this project consists of three coherent research objectives. First, although the U.S. food assistance system (with 211 food banks and 26,000 pantries) serves millions, the specific distributed foods and their nutritional value remain unclear. To address this, the team will develop an adaptive multi-agent framework powered by large language models (LLMs) to automate analysis of free food supplies and reveal their nutritional contributions. Second, a critical gap exists in understanding how poverty-driven food insecurity and opioid misuse reinforce each other, and what the specific nutritional needs of vulnerable populations are. To fill the gap, the team will build an integrated graph from multi-source data across social, food, health, and nutrition domains, and advance graph prompt learning and graph retrieval augmented generation (GraphRAG) techniques to develop a novel causal analysis method that explores their intersection and informs targeted food demand strategies. Third, with the analyzed food supplies and informed demand strategies, optimizing personalized, food-secure, and nutrition-adequate interventions for affected individuals remains a key objective. To achieve this, the team will develop a novel multi-armed bandit algorithm integrating free food access, user budgets, and nutritional needs to close the supply-demand gap and enable effective, integrated interventions. The suite of novel AI-driven techniques developed in this project will benefit research communities in information integration and informatics (III). This AI-augmented paradigm can also be adapted to other crises - such as substance abuse, educational deprivation, and suicide risk in impoverished communities - and will benefit fields including economics, epidemiology, policy, and social sciences. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.