Connecticut is deploying data science to solve one of public health's most stubborn problems: getting HIV treatment and addiction care to people who use drugs before outbreaks occur. A \$2.2 million grant from the National Institutes of Health will fund the development of a geo-targeting system designed to direct the country's first legalized mobile pharmacy and clinic to the exact neighborhoods in the state where need is most concentrated.
The project, called GEO IMPACT, builds on a Connecticut innovation that already puts it ahead of the rest of the country. In 2022, Connecticut became the first state to legalize mobile pharmacies that can dispense controlled substances including buprenorphine, naloxone, and antiretrovirals directly from a vehicle in communities. The concept was championed by Yale University researchers and advocates who argued that bringing the pharmacy to the patient would reach people who are unstably housed or too distrustful of institutions to seek care in a clinic. The model has been running, but with an unresolved question: where should the vehicle go?
That's what this grant is designed to answer. For the first time, six Connecticut state health agencies will share data across systems that have historically operated in silos — HIV surveillance, overdose data, substance use treatment records, social services — and researchers will use spatial modeling to map where unmet need is highest. Community health workers embedded with the mobile unit will add on-the-ground intelligence that data alone can't capture.
The urgency is real. HIV outbreaks among people who use drugs have become a recurring national crisis since a 2015 outbreak in rural Scott County, Indiana infected 181 people in a county of 24,000. Similar clusters have emerged in West Virginia, Massachusetts, and Philadelphia, driven largely by the shift to injectable fentanyl and stimulant co-use. The federal Ending the HIV Epidemic initiative, launched in 2019 with a goal of cutting new infections 90% by 2030, has struggled to reach people in active addiction. Connecticut has roughly 12,000 residents living with HIV and records more than 1,300 overdose deaths annually, with fentanyl involved in nearly all of them. The cities of New Haven, Hartford, and Bridgeport — likely primary targets for the mobile unit — carry a disproportionate share of that burden despite sitting inside one of the wealthiest states in the country.
Connecticut's small size turns out to be an asset here. As one of the five smallest states geographically, the entire state can be modeled and a single mobile unit can realistically reach most high-need areas, making it an unusually clean test case for what researchers are calling a "learning health system" — one that uses incoming data to continuously improve where and how services are delivered. Similar thinking is driving a Los Angeles study testing whether peer networks can drive HIV prevention and addiction treatment, and New York hospitals are separately testing HIV prevention drugs for patients already in addiction treatment.
The NIH awarded this under an RM1 cooperative agreement, a relatively rare mechanism reserved for high-priority team-science initiatives. The $2.2 million likely covers the first year of what is typically a five-year project, putting potential total funding in the range of $10 to $12 million. Researchers will evaluate outcomes at three levels: state-level HIV and overdose trends, community access to services, and individual patient outcomes among people living with or at risk for HIV. Results from early phases will shape how the mobile unit is deployed in later ones.
