States and policy makers have expressed a strong interest in using Medicaid to address social determinants of health (SDOH). While this approach holds promise for improving outcomes and reducing costs, using Medicaid to pay for services outside the medical system creates challenges. This article examines efforts to address SDOH in Oregon, which, as part of its 2012 Medicaid waiver, incorporated health-related services that lacked billing or encounter codes and were not included in Oregon's Medicaid state plan as a strategy to improve outcomes and control costs. We examine the varieties of health-related services that were used and describe the specific challenges in deploying and paying for these services. We conclude with lessons from Oregon that can help states and the federal government as they work to address SDOH.
To improve outcomes and control costs among Medicare and Medicaid members, state and federal agencies have begun placing greater emphasis on addressing social determinants of health (SDOH). SDOH include factors like employment and income, access to food and transportation, and the safety of home and neighborhood environments (Artiga and Hinton 2018). These factors—and the policies used to address them—may be particularly important for the low-income and vulnerable populations of Medicaid beneficiaries.
Using Medicaid as a mechanism to coordinate and pay for services that address SDOH may hold promise for improving outcomes and controlling costs for at least three reasons. First, Medicaid now provides coverage for more than 70 million Americans (CMS n.d.), representing a low-income, vulnerable population where SDOH are likely to be substantial factors in overall health. Second, state contracts with Medicaid managed care organizations (MCOs), which managed benefits for over two-thirds of Medicaid members as of 2016 (CMS 2018a), create an opportunity for states to embed incentives or requirements to address SDOH. Third, independent of state contracts, MCOs may have internal incentives to address SDOH if such investments help them reduce spending on medical services and capture a larger share of capitation payments from state Medicaid programs.
However, important challenges exist to addressing SDOH through Medicaid. These include defining the kinds of SDOH investments that should be targeted, establishing sustainable funding for SDOH investments given federal rules for setting MCOs' capitation rates, and evaluating the impacts and returns on SDOH investments. Failure to address these issues may create barriers to addressing SDOH on a large scale through Medicaid.
The State of Oregon, an early pioneer in efforts to address SDOH through its Medicaid program, offers several lessons for state and federal policy makers. Under a 2012 Medicaid demonstration waiver, the state created 16 coordinated care organizations (CCOs) to manage care for the majority of its Medicaid members. CCOs shared features of Medicaid MCOs and accountable care organizations (ACOs) (McConnell et al. 2014). Like other kinds of MCOs, they contracted with and managed provider networks and received capitated payments from the state's Medicaid program. Like ACOs, CCOs were designated as the point of accountability for the health care access and quality of a defined population of members and could receive financial incentives for performance. Each CCO received a global budget that covered physical, behavioral, and oral health care for Medicaid members in a given geographic area, and was responsible for coordinating and integrating all services covered by the global budget.
A noteworthy feature of Oregon's 2012 Medicaid waiver was that it directed CCOs to use “flexible services” to replace or reduce the need for medical services. The waiver defined flexible services as low-cost, health-related services not covered by Oregon's Medicaid program (CMS 2013). A 2016 update to Oregon's administrative rules clarified that flexible services were services that lacked traditional billing or encounter codes and were likely to be cost-effective alternatives to covered benefits. These services could be provided at the individual or community level, and thus could extend to support a farmer's market in a “food desert” or classes on healthy meal preparation. The rules required CCOs to work with Medicaid members and their care teams to determine the flexible services members should receive, and required CCOs to create formal policies on how they would work with health care providers to deliver flexible services (55.2 Or. Bull. 537).
John Kitzhaber, Oregon's former governor and a key figure in the development of the 2012 Medicaid waiver, frequently described the potential for flexible services to improve care and reduce costs with the following hypothetical: Imagine a beneficiary with congestive heart failure who experiences increased pain and difficulty breathing during a heat wave. The traditional Medicaid program would pay for an unlimited number of ambulance transports and emergency department visits for this condition, with some potentially leading to inpatient admissions. However, all of this utilization could be avoided though the purchase of a $200 air conditioner (Lehman 2012). Provisions for flexible services in Oregon's waiver were intended to encourage CCOs to think broadly and to seek out and pay for cost-effective alternatives to medical services, such as the air conditioner.
Oregon's experience with flexible services provides insight into the opportunities and challenges that MCOs and states face as they seek to address Medicaid members' SDOH. In this article, we draw on interviews with CCO informants to describe the kinds of flexible services CCOs provided and the challenges they experienced with deploying and paying for these services. We highlight several lessons from Oregon, including the need to define and provide guidance on SDOH investments, address challenges related to federal rules, and support data collection and evaluation.
We conducted semistructured interviews with CCO informants in March and April 2016 as part of an evaluation of Oregon's 2012 Medicaid waiver for the Oregon Health Authority (OHA), Oregon's Medicaid agency (Kushner et al. 2017). Informants were identified by each CCO's leaders as being knowledgeable about their CCO's flexible services program. We carried out 14 interviews, with one to four informants in each interview. (Informants at two interviews provided information for two CCOs each because these CCOs shared common administrative staff.) The roles of informants were diverse, including CEOs, COOs, and CFOs; medical officers and case management staff; financial and analytics staff; and other kinds of staff. We reviewed data on flexible services spending reported by CCOs to OHA in quarterly financial reports and asked informants about their process for populating the reports and the completeness of the reports. In addition, we held informal discussions with OHA staff and reviewed publicly available documents about flexible services. Our study protocol was approved by Oregon Health & Science University's Institutional Review Board.
We developed a preliminary list of thematic codes based on flexible services documentation and preliminary discussions with OHA staff. The study team coded a sample of interview transcripts as a group, modified the code list until agreement on codes and coding practices was achieved, and coded all transcripts using the finalized list. We then extracted and summarized themes from the coded transcripts, conferring as needed to ensure agreement on interpretation. Transcripts were coded using Atlas.ti Version 8.0.
What Kinds of Flexible Services Did CCOs Provide?
Table 1 displays examples of flexible services provided by CCOs.
Informants provided multiple examples of flexible services they believed had improved health and lowered health care costs:
Temporary Housing. “We had an individual who had a very severe intestinal disease, and she was homeless and living in her car. . . . Because she had this intestinal disease, it was difficult for her to keep herself clean. She couldn't get evaluated by a doctor, and she also couldn't then get the treatment that she needed because in order for the treatment to take effect, she had to remain clean. She was ending up in the hospital ED . . . so we put her up in the hotel, we got her cleaned up, we got her visit to her doctor taken care of, and she began the treatment that she needed, and then went into the facility.”
Exercise Equipment. “We gave [the member] an exercise bike so that she could use that at home, because she has been relatively house-bound and not able to get somewhere like a gym. . . . She's been able to exercise a few times a day for short periods of time, has lost weight, and has actually been able to cite some better emotional regulation, and actually through that then has opened up to other flexible services, and one is a living well class.”
Environment Improvements. “A patient I've known a long time . . . lived in a mobile home. She's quite obese, and she's got diabetes and a number of other chronic conditions. She has enough trouble getting around anyway, but getting in and out of her mobile home was fairly dangerous. I think she even had a fall. . . . Somebody was able to hire a handyman to build a nice, stable set of stairs with a handrail so she can get in and out of her place. . . . I think that just allowed her to be more mobile, get in and out of her place more safely, get lots of doctors' appointments, and she's actually pretty active socially with her family, so it's allowed her to be out in society a lot more, as well as getting to her doctor's appointments.”
Care Coordination or Case Management. “We . . . have overall programs that could be considered flex services. One is our community health worker hub and how we serve and help navigate for members. We also have a wellness center that we've invested in that does behavioral-based psychoeducation around pain management as well as other self-management for chronic diseases. We support a lot of peer services. . . . I think we have a broad range of programs that could generally be considered flex services.”
Programs to Improve Community Health. “We funded a number of grant projects and these grant projects are available to providers in the community . . . and their projects are all focused on community health improvement and focused on improving the CCO quality metrics. . . . We consider those as flexible services and we'll be reporting those in 2016 financials as expenses.”
How Much Did CCOs Spend on Flexible Services?
Beginning in 2014, OHA required CCOs to report the number of members who received flexible services and spending on flexible services in their quarterly financial reports. Table 2 compares reported flexible services spending to total medical spending from 2014 to 2015. Reported spending on flexible services remained relatively low throughout this period, although interviews with CCO informants suggest that CCOs' financial reports may have underrepresented actual spending on flexible services. Due to lack of clarity about the definition of flexible services (described below), some CCOs omitted spending on specific types of services that would meet the definition of flexible services in Oregon's waiver and administrative rules. However, even if actual spending on flexible services was twice that reported by CCOs, the overall amount would be relatively low.
Challenges with Providing Flexible Services
CCO informants described a variety of factors that may have inhibited the use of flexible services: (1) a lack of definitions and guidance about flexible services; (2) concerns about the sustainability of flexible spending, which were linked to federal rules; and (3) data and evaluation.
Challenge #1: Lack of Definitions and Guidance. Confusion about the state's definition and intent for flexible services may have inhibited CCOs from spending more on flexible services. CCO informants described lack of clarity about the definition of flexible services and how flexible services spending should be reported to OHA. They received little guidance about flexible services before the state updated its administrative rules in 2016, which may have led to differences in the kinds of flexible services CCOs provided and reported. For example, some CCOs believed flexible services could be provided only to individuals, while others believed flexible services could also be provided at the community level. Overall, CCOs described OHA's 2016 rules update as providing helpful definitions and guidance. However, lack of clarity about whether certain kinds of health-related services would meet the state's definition of flexible services remained even after the new rules went into effect. Some informants indicated that they did not believe flexible services encompassed services with billing codes (e.g., a yoga class, which could be billable for some specific medical conditions but not billable for others) or services not tied to a medical diagnosis (e.g., shoes), while other informants indicated that they did include these kinds of services in their definition of flexible services. Other areas of confusion included care coordination and disease management programs and spending that could not be attributed to individual members, such as a grant for a local homelessness shelter that served both CCO members and others in the community.
Challenge #2: Capitation Rate Setting and Sustainability. CCO informants expressed considerable concerns about the federal rules that set parameters for how states develop the capitation rates paid to Medicaid MCOs. Federal rules require states to set rates based on reasonable costs of providing services in MCOs' contracts, including the costs of medical services covered by a state's Medicaid program and specific operating costs, such as administration, taxes, and risk margin. States must obtain historical data on these costs and trend these data forward by an appropriate factor. The data “must be in accordance with actuarial standards for data quality” (42 CFR secs. 438.4–5). Changes to federal managed care rules that went into effect in 2016 required states' rate-setting processes to depend more heavily on historical managed care claims or encounter data (Mendelson, Goldberg, and McConnell 2016).
While Medicaid MCOs may pay for any service they voluntarily agree to provide, federal rules generally preclude states from using services outside the Medicaid state plan to set rates (42 CFR sec. 438.3). These rules may create barriers to spending on services to address members' SDOH for several reasons. First, MCOs that use flexible services to address SDOH today may incur the risk that such spending will not “feed back” into tomorrow's capitation rates, thereby reducing resources available in the future to address SDOH. Second, efforts to address SDOH today may actually reduce future capitation rates if such spending reduces the use of medical services on which rates are based. Moreover, federal rules about the kinds of data that states can use to set capitation rates may deter incorporation of SDOH spending into the rate-setting process if standardized, high-quality historical data on health related services are unavailable to state rate-setting authorities.
Concerns about federal rules for capitation rate setting are likely to have inhibited CCO spending on flexible services. CCO informants expressed frustration about the inability to “get credit” for spending on flexible services in the same way as spending on medical services. Several CCOs had hoped that Oregon's 2012 Medicaid waiver would create a path for greater incorporation of flexible services spending into capitation rate setting. As one informant said, “The dream was flexible services would be built into your rates. Well, no. Not unless they meet two fairly restrictive CFRs. CMS has been clear that mostly it will be paid out of any surplus dollars. That just seems so fundamentally different to me from what the state was advocating, but I really want someone to circle back to the state level and address this with us, but that hasn't happened yet.” Another CCO described the state as “trying to figure out where do these [flexible services] go and how do we account for them? How do we work them into the rates or just the overall global budget?”
In addition to concerns about the relationship between flexible services and rate setting, CCOs also described confusion about the relationship between flexible services spending and medical loss ratios (MLRs). While federal rules generally preclude states from using spending on SDOH to set capitation rates, they may allow states to count such spending toward MLRs, defined as the minimum share of premium revenue that MCOs must spend on health care services, if SDOH spending meets specific criteria. For example, SDOH spending may count toward MLRs if it increases the likelihood of desired health outcomes in measureable ways; is grounded in evidence-based medicine, widely accepted best clinical practice, or criteria issued by nationally recognized health care quality organizations; and is designed to achieve one of five specific goals listed in federal rules (45 CFR sec. 158.150). This option—to count SDOH spending toward MLRs—could, theoretically, encourage MCOs to invest in SDOH. However, CCO informants described confusion about the different treatment of flexible services spending for capitation rate setting (i.e., it could not be included in the same way as medical spending according to federal rules) and MLR calculation (i.e., it could count toward MLRs in the same way as medical spending according to the rules).
Beyond concerns about the treatment of flexible services spending for capitation rate setting and MLR calculation, some CCOs expressed concerns about perceived expectations that they fund large-scale investments in SDOH. The CCO model emphasized addressing local population health and community needs, and CCOs were required to carry out community health assessments and develop community health improvement plans (McConnell et al. 2014). CCO informants recognized the importance of large-scale investments in SDOH, such as housing and health care workforce, to achieve these goals. However, they expressed concerns about the adequacy of CCOs' global budgets to fund these investments and reflected on the need to fund these investments with resources from outside the health care system. One informant captured the spirit of CCOs' concerns, referring to “growing expectations at the community level” and the need for OHA and the State Legislature to “[be] thoughtful about putting forward expectations” for large-scale investments outside the health care delivery system given the expectations for CCOs to transform the health care delivery system itself.
Challenge #3: Data and Evaluation. State and federal rules in effect at the time of Oregon's 2012 Medicaid waiver could be interpreted as requiring CCOs to rigorously evaluate the effects of flexible services. For example, Oregon's administrative rules described flexible services as “likely to be cost-effective alternatives to covered benefits” and “likely to generate savings,” suggesting that CCOs would need to collect data and evaluate their effects on costs (55.2 Or. Bull. 537). Similarly, federal rules stated that health-related services must “increase the likelihood of desired health outcomes in ways that are capable of being objectively measured and of producing verifiable results and achievements” in order to be counted toward an MCO's MLR (45 CFR sec. 158.150). Generally, CCO informants said that Oregon's 2016 administrative rules required additional effort to justify and track flexible services. One informant stated that the CCOs were “expected to prove cost effectiveness” under state and federal rules.
CCO informants described evaluation of flexible services, which they viewed as required by OHA and CMS, as a challenge. CCOs varied widely in their capacity to track and report on members' use of flexible services; while some used custom-built software to track orders and pay vendors for flexible services, others simply used spreadsheets. Most CCOs were unable to link data on members' use of flexible services with data on outcomes, such as health care spending from medical claims, because utilization and outcomes data were stored in separate data systems. In addition to data collection, CCO informants described the need for more longitudinal data and difficulty finding appropriate comparison groups as barriers to confidently identifying the effects of flexible services. As one informant said, “It's hard to find that homeless person that has all the same health conditions that didn't get a hotel room and compare their outcomes with somebody who did get a hotel room.”
Lessons from Oregon
States and the federal government have expressed interest in using public programs to address SDOH. A growing number of states have begun requiring MCOs to address SDOH as part of their contracts (Artiga and Hinton 2018). In addition, CMS recently announced plans to expand the range of benefits that can be covered by Medicare Advantage plans to include non medical services that improve health or health-related quality of life (CMS 2018b). Moreover, Health and Human Services Secretary Alex Azar recently suggested that CMS may grant health care organizations greater flexibility to pay for housing, nutrition, and other social needs (Azar 2018). However, additional evidence is needed about best ways to implement and fund SDOH investment, and about the impacts and returns from such investment (Gottlieb et al. 2016, 2017). Although Oregon's CCOs differ from traditional MCOs, lessons from the CCO experience are directly relevant for engaging MCOs to address members' SDOH in other states. We outline four generalizable lessons from Oregon that can help states work with MCOs on SDOH:
Lesson #1: Provide MCOs with Detailed Definitions and Guidance Up Front on Health-Related Services They Are Expected to Coordinate and Pay For. An extremely wide variety of services—from exercise equipment and gym memberships, to cooking classes and peer support, to affordable housing and parks—may conceivably be health related. Flexibility to pay for a variety of services may help MCOs respond to the unique health-related needs of their members and local communities. However, the experience of Oregon's CCOs suggests that states should provide detailed definitions and guidance on what kinds of health-related services MCOs are expected to pay for—that is, what kinds of health-related services are “in-bounds” and “out-of-bounds”—if they expect MCOs to deploy health-related services in order to achieve specific objectives of their Medicaid programs.
In our interviews, CCO informants reflected on the advantages and drawbacks of detailed definitions and guidance: While some described the new, more detailed administrative rules that went into effect in 2016 as limiting their ability to flexibly meet members' needs, most described the clarification and certainty provided by the new rules as helpful. In particular, our interviews suggest that states should provide guidance to MCOs regarding services like care coordination or navigation and services with billing codes that are not covered by state Medicaid programs, which straddle the line between medical and health-related services and may cause confusion among MCOs. Providing detailed definitions and guidance up front can help MCOs to get health-related services “out the door” and deploy their resources efficiently.
Lesson #2: Clarify How Health-Related Services Spending Will Be Treated for Capitation Rate Setting, and Identify a Path to Funding These Services. The lack of a clear path to funding flexible services through capitation rates and confusion about the treatment of flexible services spending for rate setting and MLR calculation may have inhibited CCOs from spending more on flexible services. Resolving issues around rates and funding is particularly challenging because it hinges on federal rules and CMS interpretation of those rules, which are beyond the control of states. At a minimum, states should explain and provide guidance on how health-related services spending will be incorporated into their rate-setting processes given the constraints of federal rules. This can give MCOs confidence to plan and allocate funds for SDOH.
Beyond providing guidance and clarification, options exist for providing MCOs with Medicaid resources to fund SDOH. These include adding certain types of health-related services to a state's Medicaid plan so they can be included in rate setting; executing waivers that provide Medicaid funding for SDOH spending; mandating the use of value-based payments to providers that are not linked to specific medical services, and that can be used flexibly to pay for health-related services; and rewarding MCOs that achieve lower costs through SDOH spending with higher profit margin or risk margin (Bachrach et al. 2018). Most of these options would depend on federal approval and support for experimentation by states. CMS has shown interest in granting states flexibility with other aspects of their Medicaid programs, and the recent announcement that Medicare Advantage plans could expand the health-related benefits they offer may signal willingness to provide flexibility with SDOH spending among other programs.
Lesson #3: Delineate MCOs' Responsibility for Large-Scale Investments in SDOH. Some CCOs expressed concerns about perceived expectations that they fund large-scale investments in SDOH. To alleviate such concerns, states should clarify roles and expectations for MCOs with regard to large SDOH investments. For example, given the high need for affordable housing among Medicaid members and the scale of spending needed to meet that need, states may need to clarify the role of MCOs in facilitating access to affordable housing and how MCOs' performance in this area will be evaluated. Should use of MCO resources be limited to collecting data and reporting to local housing organizations on the housing needs of members, or to working with local housing authorities to refer and place members in housing? Should MCOs be encouraged to pay for tenancy supporting services, such as housing search, temporary rent assistance, and training on dealing with landlords? Should MCOs be encouraged to spend a substantial portion of their capitation payments on large housing investments—such as funding for a homeless shelter or affordable apartment building—or should state guidance actively steer them away from such investments? The answers will depend on state priorities and the resources available to MCOs, state programs, nongovernmental organizations, and local communities, but answering these questions up front will help MCOs plan SDOH investments and focus their resources effectively.
Lesson #4: Specify How MCOs' SDOH Spending Will Be Evaluated, and Support MCOs with Evaluation. States should establish realistic and uniform standards for evaluating SDOH spending among MCOs, including detailed standards for purposes of rewarding MCOs with higher margins or special payments outside of capitation rates. To build the evidence base regarding the kinds of SDOH spending that improve outcomes among different Medicaid populations, states should require MCOs to collect person-level data on SDOH spending in a uniform reporting format that can be linked to data on outcomes, including health care utilization and spending from claims and encounters. Person-level data collection and reporting will be challenging for some MCOs, and states should recognize that data quality will improve over time as MCOs' data collection capacity evolves. In the meantime, states may need to relax their expectations for rigorous evidence about the effects of SDOH investment and accept descriptive statistics as evidence of MCOs' efforts in this area. In addition, states may need to work with CMS on setting realistic standards for SDOH utilization and spending data, since it may be difficult to meet “actuarial standards for data quality” given the heterogeneous and nonstandard nature of SDOH spending.
Outlook in Oregon Going Forward
Oregon learned from its experience with flexible services and incorporated multiple provisions to encourage SDOH investment by CCOs into a new Medicaid waiver executed in January 2017. First, Oregon clarified the kinds of SDOH investment CCOs are required to consider using. It defined a new category of SDOH investment called “health-related services” that encompasses flexible services, defined as cost-effective services offered as an adjunct to medical services, and community benefit initiatives, defined as community-level interventions focused on improving population health and health care quality. The 2017 waiver (CMS 2017), and a subsequent update to Oregon's administrative rules (Or. Admin. R. 410-141-3150), aligned the definition of health-related services with federal rules for MLR calculation. Second, the 2017 waiver allowed CCOs to calculate MLRs for a given year on a three-year rolling basis (i.e., using data from the previous three years), meaning that a CCO with an MLR below 85% in a given year can “catch up” by spending more on health-related services in the next year. Third, the waiver directed Oregon to set capitation rates with a “profit margin” for high-performing CCOs (i.e., those that show quality improvement and cost reduction) in order to offset potential decreases in per capita capitation payments caused by the use of health-related services that reduce members' use of Medicaid benefits. We expect that these changes will encourage CCOs to use health-related services under the 2017 Medicaid waiver and sustain these investments over time.
Oregon's 2017 Medicaid waiver provides a framework for building incentives to invest in SDOH into CCOs' contracts. Oregon will execute new five-year contracts with CCOs in 2020. In January 2019, OHA released a request for applications (RFA) for organizations to bid on the new contracts. The RFA set forth new requirements for CCOs based on input from CCO stakeholders, policy experts, and the public, which OHA (2018) calls “CCO 2.0.” These include spending a portion of CCOs' net income or reserves on SDOH and health equity through a process that involves the CCOs' community advisory councils (CACs) and partner organizations, and prioritizing spending on housing-related services and supports (Oregon Health Authority 2019b). In the initial years of the new contracts, OHA will provide bonus payments to CCOs that achieve SDOH or health equity milestones, and require CCOs to reinvest these payments in SDOH or health equity priorities. Before the new contracts take effect, OHA will adopt a package of administrative rules that support CCO 2.0. The rules will require CCOs to set SDOH spending priorities using input from CACs and community health improvement plans; enter into formal agreements with partner organizations that receive SDOH investments; and report detailed information about health-related services spending and projected return on investment back to OHA (OHA 2019a). Together, the contract requirements and administrative rules will provide new incentives and accountability for SDOH investment. As with changes in the waiver, we expect that these tools will encourage CCOs to make new SDOH investments and sustain these investments over time.
This study has three primary limitations. First, our study was restricted to one state. Differences in Medicaid populations, MCOs, and state Medicaid programs may mean that other states would encounter different opportunities and challenges as they deploy services to address members' SDOH. Oregon demonstrated a particular willingness to experiment with its Medicaid program even before the 2012 waiver (e.g., with the use of a prioritized list of health services beginning in 1994). The challenges of extending Medicaid beyond its usual confines may be greater for states with less experience experimenting with their programs. However, the federal government has expressed interest in the link between SDOH and Medicare and Medicaid, and a growing number of states have begun addressing SDOH through MCO contracts. While state Medicaid programs vary, federal laws and rules that provide a framework for state Medicaid programs—such as federal rules for setting capitation rates—are uniform across states. Given increased interest and activity in addressing SDOH and a common framework of pertinent federal laws and rules, Oregon's experience should provide useful lessons for other states.
A second limitation of our study is that the roles of CCO informants we interviewed varied widely. As a result, interview results may not be directly comparable across CCOs. A third limitation suggested by our interviews is that flexible services use and spending was not always completely or uniformly reported across CCOs. Clarification of definitions and more detailed reporting requirements for health-related services under Oregon's new Medicaid waiver and CCO contracts may result in more complete and uniform data on these investments in the near future. As a result, it will be valuable to continue monitoring these data to understand how Medicaid MCOs and their partners respond to SDOH investment policies.
Using Medicaid as a mechanism to address SDOH has the potential to improve health and wellness at relatively low cost. Under a 2012 Medicaid demonstration waiver, Oregon's CCOs experimented with a wide variety of flexible services to address members' SDOH. Oregon's experience with flexible services indicates that states will need to provide clear definitions and guidance on SDOH investment, define how SDOH investment fits into capitation rate setting, and support data collection and evaluation as they work with MCOs to address members' health-related needs. Particularly difficult challenges will include delineating the kinds of SDOH investments that Medicaid programs should target, given the extremely wide variety of services that may affect members' health; navigating federal rules for rate setting and their interpretation by CMS, which are largely out of states' control; and rigorously evaluating SDOH investments given the challenges presented by nonstandard data and a lack of feasible comparison groups. As Medicaid investments in SDOH become more common and well known, challenges may arise relating to fair and equitable provision. Despite these challenges, using Medicaid's potential to address SDOH can be a key component in the effort to move health systems toward more prevention and better outcomes at the same or lower cost.
This research was funded by the Oregon Health Authority, the National Institute of Mental Health (R01MH1000001), the National Institute on Minority Health and Health Disparities (R01MD011212), and the Silver Family Foundation. We thank Sonya Howk, Edgardo Peteros, Ruth Rowland, and Zoe Velie for contributing to data collection and analysis.