Like many cities in the United States, Louisville, Kentucky is experiencing a growing housing insecurity challenge. From 2018 to 2021, the city saw a 41 percent increase in the number of people without stable housing.¹ This instability has negative implications for health outcomes: when people lack secure housing, it makes it difficult for a person to engage in preventive health care, such as managing chronic conditions or attending regular primary care visits.

Substance use disorders (SUDs) are also prevalent in Louisville, which reflects broader statewide trends. Kentucky has one of the highest rates of opioid-related deaths in the country, and Louisville has one of the highest prevalence rates for opioid use disorder in Kentucky.²

As part of its population health strategy, Kentucky Medicaid prioritizes addressing members’ unmet health-related social needs, including initiatives that connect members to housing, food, and transportation supports.³ The state also has goals to improve health outcomes for people with chronic conditions and behavioral health needs, including SUD.⁴ To help support these efforts, between April 2024 and December 2025, two Kentucky managed care organizations (MCOs) piloted an initiative in Louisville offering financial incentives and social support to Medicaid members facing housing insecurity, with the goal of increasing engagement with care management and improving health outcomes. The pilot was supported by The Leona M. and Harry B. Helmsley Charitable Trust through the Medicaid Innovation Collaborative.

Implementation Approach

To identify potential members for the pilot, the participating MCOs, Aetna and Humana, used multiple data sets including claims, health risk assessments, the Kentucky Homeless Management Information System, and other population health data. Members were included if they had indicators for housing insecurity as well as type 1 diabetes or other chronic conditions, or SUD. While Type 1 diabetes was an initial population of interest, the MCOs were unable to engage this group because they could not identify enough individuals to achieve an adequate sample size. The MCOs shared lists of eligible members with their care managers and community health workers for outreach. In addition to MCO outreach, Samaritan — a company offering a tech-enabled member engagement and incentive platform — partnered with community-based organizations (CBOs) to further support member engagement.

Interested members were enrolled in the pilot where Samaritan provided an engagement infrastructure that helped turn care plans into concrete, action steps and sustained follow-through over time. The first step was a health risk assessment to identify members’ needs. Care managers then worked with participants to understand their priorities and develop individualized care plans with tailored action steps. Action steps supported participants’ health and social needs, ranging from visiting a primary care provider to accessing resources at a local food bank.

Upon completing each action step, members received varying financial incentives, with a maximum of $40 per month, that are set by the case manager to align with the patient’s goals. Incentives were distributed through Samaritan-issued debit cards that functioned like traditional debit cards, but restricted use at businesses that primarily sell substances, like smoke shops or liquor stores, and prohibited cash withdrawals or money transfers. In addition to incentives, a network of supporters sent encouraging messages — primarily via text — to motivate members to stay engaged and complete action steps. This communication was important for helping participants feel supported through the pilot.

Impact

The goal of this pilot was to help people facing housing insecurity engage with the health care system by incentivizing behavior change, with the goal of increasing preventive care, reducing avoidable service use, and reducing costs. Humana and Aetna aim to measure pilot impact by assessing key measures, including increased primary care use, HEDIS quality measures related to primary care and prevention, and increased social services engagement. They are also seeking to assess whether the pilot reduced emergency department visits, inpatient use, and total cost of care, where clinically appropriate. While the sample size was small and the claims data is early, preliminary data suggests the pilot had positive impacts on health care use and costs.

Member and community health worker experiences also highlight the pilot’s impact. For example, the pilot helped one member complete job applications, attend interviews, and take steps to pursue his GED. This member is also being assessed for a housing voucher. In a post-pilot survey, another member described that the program provided both financial support and emotional support from the care team.

Community health workers at CBOs also reported that the pilot supported their engagement efforts. For example, one community health worker described how the pilot helped her stay in contact with members and helped members feel accountable for their goals. She also reported that three members she worked with during the pilot found stable housing or employment.

Implementation Lessons

• Leveraging existing CBO networks supports success. Both the MCOs and Samaritan had an existing network of CBO partnerships and leveraged these networks to engage hard-to-reach members in the pilot. The CBOs supported outreach both to members on MCOs’ eligibility lists and used their community-level work to recruit additional people to the pilot who were not originally identified through MCO data. Involving the CBOs expanded the outreach efforts because they were already deeply connected to the community.

• State involvement is important for gaining buy-in and addressing challenges. Kentucky Medicaid was highly supportive and engaged in design and implementation of the pilot, regularly joining planning meetings. Medicaid agency involvement also helped streamline program launch, for example, by easing pilot approvals and helping address implementation challenges.

• Lack of CBO access to MCO enrollment data was a barrier. Because CBOs are not medical providers, they are unable to access Medicaid enrollment data to check a person’s eligibility for the pilot. As a result, CBOs had to verify eligibility through the MCOs, which delayed pilot engagement.

• Implementation takes time. Contracting and implementation processes are complex. At MCOs, this often means involving a wide range of departments, such as policy, legal, care management, and data teams, among others. This complexity is further compounded when additional partners are involved, such as multiple CBOs. It takes significant time and effort to get the right details in place to be successful, like ensuring member privacy, creating efficient workflows, training internal and external partners, and establishing evaluation mechanisms. Allowing sufficient time for this type of planning is critical.
  Samaritan also reflected that additional time and resources for engaging with CBOs early, including through in-person sessions, may be a helpful strategy for achieving success.

Looking Ahead

As a result of their positive experience with the pilot, Samaritan, Aetna, and Humana sought funding from the Kentucky Association of Health Plans (KAHP) to continue and expand the pilot. KAHP is providing funding for all five Medicaid MCOs in Kentucky to work with Samaritan through 2026. The expanded pilot will support 500 people across the five MCOs. Participation across all five Kentucky Medicaid MCOs will make it easier for CBOs to engage and enroll Medicaid members, as eligibility will not be limited to a select few MCOs. The statewide pilot will also support scaling engagement infrastructure across payers — offering lessons and a potential foundation for other states and regions seeking to adopt a similar model.

All partners are looking forward to expanding the pilot and continuing to support positive outcomes for Medicaid members facing housing insecurity.

1. Coalition for the Homeless. (2022). Addressing Urgent Needs: A 2022 Analysis of Homelessness in Louisville. https://louhomeless.org/wp-content/uploads/2022/04/Addressing-Urgent-Needs-2022-Analysis-of-Homelessness-in-Louisville.pdf
2. Thompson, K., Barocas, J. A., Delcher, C., Bae, J., Hammerslag, L., Wang, J., Chandler, R., Villani, J., Walsh, S., & Talbert, J. (2023). The prevalence of opioid use disorder in Kentucky’s counties: A two-year multi-sample capture-recapture analysis. Drug and alcohol dependence, 242, 109710. https://pmc.ncbi.nlm.nih.gov/articles/PMC9772240/
3. Kentucky Cabinet for Health and Family Services. (n.d.).  Equity and Determinants of Health Branch. https://www.chfs.ky.gov/agencies/dms/dpqo/EDOH/Pages/default.aspx
4. Kentucky Department for Medicaid Services. (2024). Driving Quality in Kentucky’s Healthcare System. https://www.chfs.ky.gov/agencies/dms/dpqo/mco-qb/Documents/Quality%20Slides.pdf.