@ Mathematica. Progress Together Understanding the Experiences of Dually Eligible Beneficiaries Enrolled in Ml Health Link Report Summary December 2022 Danielle Whicher, Madelaine Spiering, Diane Beaver, Faye Miller, and Rachel Gringlas MMCO Dually Eligible Individuals Summary of the Michigan study Overview This brief report describes the results of our research on beneficiaries' experiences with care coordination and access to health care services, home- and community-based services (HCBS), and social services in the Medicare-Medicaid Financial Alignment Initiative (FAI) demonstration in Michigan, known as MI Health Link.! Michigan, the Centers for Medicare & Medicaid Services (CMS) and participating Medicare-Medicaid plans (MMPs)- referred to as integrated care organizations (ICOs) in Michigan- entered into three-way contracts (TWCs) that govern the demonstration. Though the ICOs are required to deliver care coordination and other services to enrolled beneficiaries, Michigan retained its existing carve- out for Medicaid behavioral health (BH) services. The state contracts directly with regional Prepaid Inpatient Health Plans (PIHPs) for the provision of Medicaid-covered treatments for substance use disorder, care for mental illness, and certain HCBS waiver services.? However, PIHP coordinators are responsible for participating in the beneficiary's Integrated Care Team and coordinating Medicaid BH services with other services outlined in the MI Health Link enrollee's Individual Integrated Care and Supports Plan.3 The results of this research are based on our interviews conducted from April to June 2022 with ICO enrollees who are dually eligible for full Medicare and Medicaid benefits. Though this research was conducted through a CMS contract, none of the findings are indication of compliance (or lack thereof) with the TWC. Methods We asked the state to provide contact information for about 500 MI Health Link beneficiaries eligible for recruitment into the study. Using this information, we randomly selected beneficiaries with diverse demographic characteristics to reflect the following sampling subgroups: those with multiple chronic conditions (MCCs), users of BH services, users of HCBS, or some combination of these. We focused on these higher risk subgroups as these individuals generally have greater care coordination needs. Given the importance of improving health equity, we also oversampled African American and Hispanic/Latino/a/e MI Health Link beneficiaries to understand if these groups have distinct experiences with their MI Health Link plan*. We also included beneficiaries from the three distinct areas of the state where MI Health Link operates: (1) southeastern Michigan, the region encompassing the urban Detroit area; (2) the Upper Peninsula, a primarily rural environment; and (3) southwestern Michigan, a region mixed with smaller cities and rural areas. This would help us to begin to understand if there were different experiences of beneficiaries living in urban versus rural areas, and potential barriers to health equity among those subpopulations. ! Information on MI Health Link can be found at https://www.cms.gov/Medicare-Medicaid-Coordination/Medicare- and-Medicaid-Coordination/Medicare-Medicaid-Coordination-Office/Financial AlignmentInitiative/Michigan. 2 For more discussion about administering this carve-out, please see an independent evaluation of the demonstration which can be found at: https://innovation.cms.gov/data-and-reports/2022/fai-mi-secondevalrpt 3 See section 2.4.1.6 in the MI Health Link three-way contract available at: https://www.cms.gov/medicare- medicaid-coordination/medicare-and-medicaid-coordination/medicare-medicaid-coordination- office/financialalignmentinitiative/michigan 4 While we attempted to conduct interviews with beneficiaries who speak Arabic and Spanish, we were unable to recruit Spanish-speaking or Arabic-speaking participants. Mathematica® Inc. MMCO Dually Eligible Individuals Summary of the Michigan study We then conducted 30 semistructured telephone interviews with higher-risk beneficiaries enrolled in MI Health Link ICOs for at least 12 months.> After receiving their consent to participate, we asked MI Health Link beneficiaries about their experiences with: (1) health risk assessments and care planning, (2) care coordination services offered by the plan, and (3) accessing needed medical and mental health, HCBS, and social services.® We also asked about how the COVID-19 public health emergency (PHE) affected participants' access to services. After conducting the interviews, the study team conducted a thematic analysis of structured interview notes from 29 interviews." We developed a codebook and used NVivo data analysis software to analyze the data using a qualitative research design framework called a "grounded theory approach." We coded the structured interview notes to systematically identify key themes related to beneficiaries' experiences with care coordination and receipt of services and developed user personas and journey maps. Results Developing the user personas Through our analysis, we observed important differences in care coordination experiences for each of the three higher-risk subgroups that we identified for recruitment purposes: 1. Individuals with multiple chronic conditions (the "MCC group™), 2. Individuals with BH conditions who may or may not have multiple chronic conditions (the "BH group"), and 3. Individuals receiving HCBS who may or may not have multiple chronic conditions (the "HCBS group"). Most participants in the MCC and HCBS groups reported a relationship with an ICO care coordinator, although beneficiaries in the BH group generally reported that they did not feel connected to their ICO care coordinator. Additionally, more beneficiaries in the BH and HCBS groups reported unmet social needs compared to those in the MCC group. Based on these observations, the study team developed fictitious user personas that summarize the experiences of each beneficiary subgroup: e "Steve" personifies an individual with multiple chronic conditions who represents the themes that emerged from interviews with the MCC group. s "Robert" represents a person with BH conditions who typifies the concepts and concerns that surfaced during interviews with the BH group. e "Sara" personifies an individual receiving HCBS who represents the themes emerging from interviews with the HCBS group. 5 An Integrated Care Organization is Michigan's term for a Medicare-Medicaid Plan. 6 Questions about needed social service centered on beneficiary access to food, housing and transportation services, including non-emergency medical transportation (NEMT). 7 We excluded one interview because the interviewee provided contradictory information during the interview, suggesting that the interviewee may have misunderstood the questions. Mathematica® Inc. MMCO Dually Eligible Individuals Summary of the Michigan study Thus, the three user personas that follow do not represent actual people who participated in this study. Rather, the content of the user personas represents the key themes that came to light during our thematic analysis. We defined the demographic characteristics (gender, geographic location, and medical conditions) of each user persona by examining the most common characteristics among our subgroups.® We did not identify significant differences in the experiences of beneficiaries from different racial and ethnic minority groups. However, identifying differences based on race/ethnicity was challenging, as race/ethnicity was closely correlated with geographic location. Specifically, people in the UP are predominantly White and non-Hispanic, whereas there are significantly more African American and Hispanic/Latin(o/a/e) people in the other parts of the state. 8 We found that majority of those in the MCC and BH groups reside in the mostly rural Upper Peninsula (UP), whereas a majority of those in the HCBS group reside in the southwestern part of the state. The majority in the MCC group are White, non-Hispanic and male. The majority in the BH group are African American, non-Hispanic and male. The majority of those the HCBS group are African American, non-Hispanic and female. Mathematica® Inc. MMCO Dually Eligible Individuals Summary of the Michigan study Fictitious user persona for beneficiaries with MCCs Multiple Chronic Conditions N GV[= BACKGROUND ® He has MCCs. e Steve is a beneficiary who lives in a rural part of Michigan. e He js satisfied with his plan's care coordination services, but still finds it difficult to locate specialists within close traveling distance of his home. Steve typically speaks to a care coordinator from his Ml Health Link plan every other month on the phone. In between those calls, he is able to contact his care coordinator if he needs specific help. Steve remembers his care coordinator conducting a health risk assessment. After the assessment, he received a care plan in the mail that included information about services he needs and health goals related to weight management and exercise. Steve found the care plan helpful and likes that it provides accountability. It is typically updated at least annually with new services, medications, and goals. Steve schedules his own medical appointments but his care coordinator sends him lists of providers when needed. Even with help from his care coordinator, he finds it difficult to find dentists and an ear, nose, and throat specialist in his gecgraphic area. His care coordinator also supported him in receiving a blood pressure cuff and cane, and provided him with information about NEMT services, which he uses to attend medical appointments. The care coordinator is particularly helpful in navigating insurance issues such as preauthorization for his medications. EXPERIENCES WITH MEDICAL AND COMMUNITY SERVICES Steve finds it difficult to schedule appointments with specialists, including in-network dentists. He travels far away for some appointments due to clinic closures in his area. Besides that, he is generally satisfied with his providers. Doctors and providers have been helpful in scheduling, referrals, and obtaining prescriptions and other services. Steve uses the NEMT service he is referred to by the health plan and is generally satisfied with it. Steve generally finds out about social services from friends or community organizations. He receives food assistance through the Meals on Wheels program. PERCEPTION OF HEALTH PLAN Steve is generally satisfied with his health plan and feels his care coordination services and communication efforts meet his needs and conditions. However, he finds it difficult to adjust when the plan assigns him a new care coordinator. UNFULFILLED NEEDS Most of Steve's needs are met, but he would like to have easier access to specialists in his area. Mathematica® Inc. MMCO Dually Eligible Individuals Summary of the Michigan study Fictitious user persona for beneficiaries who receive BH services Rober BACKGROUND Behavioral Health Services e Robert is a beneficiary who lives in a rural part of Michigan. ® He receives BH services and has MCCs. ® He needs a replacement walker and requires transportation to get to his behavioral health appointments. Robert typically speaks to a care coordinator from the M| Health Link plan once every six months by phone. Robert does not feel connected to his care coordinator and cannot remember his coordinator''s name. He recalls his care coordinator asking him questions about his health and needs, but he does not recall ifa care plan was made. Robert's care coordinator provides him with information about NEMT and the process for requesting durable medical equipment. Robert tries to use the information provided by the care coordinator to find needed services, but he is unsuccessful. When Robert calls the NEMT service, he is told there is a shortage of drivers. While his care coordinator has suggested that Robert ask friends to drive him to appointments and have the plan reimburse the mileage, Robert is unable to rely on friends because his appointments are far away from his house. Robert also has difficulty accessing the durable medical equipment he needs, such as a walker, through his health plan. Robert finds the logistical steps of obtaining and submitting referrals very frustrating. Eventually he "gives up" and pays for the medical equipment out of pocket. EXPERIENCES WITH MEDICAL AND COMMUNITY SERVICES Rebert reports having a good relationship with his primary care doctor and his psychiatrist. He relies on their office staff to help him make follow-up appointments. Robert uses some food stamps and other services to access affordable food that he learns about from family and friends in the area. UNFULFILLED NEEDS PERCEPTION OF HEALTH PLAN Robert is generally satisfied with his health plan due to the services covered and because he does not have co-pays. He likes the doctors in the network. However, he does not think the care coordination services offered by the plan help him to meet his needs. Robert believes most of his needs have been met but wishes his care coordinator could help him access the resources he needs and could refer him to more reliable NEMT services. Mathematica® Inc. MMCO Dually Eligible Individuals Summary of the Michigan study Fictitious user persona for beneficiaries who receive HCBS S a ra Home and Community-Based Services BACKGROUND e Sara is a beneficiary who lives in a medium-sized town in the Southwestern corner of Michigan. ® She receives HCBS through her Mi Health Link plan. ® She fs satisfied with her plan's care coordination services but requires transportation to get to her appointments and could benefit from housing services in her community. Sara typically speaks to her care coordinator from the M| Health Link plan about once every three months by phone. Sara feels connected to her care coordinator who she calls whenever she has issues or guestions. She recalls her coordinator conducting a health risk assessment and remembers receiving a care plan in the mail after this conversation. However, she does not find her care plan that helpful and does not refer to it when seeking care. Sara's care coordinator helped her secure a home care aide who she relies on as a key part of her daily life. The care coordinator has arranged for needed modifications to her home such as grab bars in the shower and a cane to help her move around her house. Sara also uses the resources provided by the care coordinator te access needed medical services. In addition to the care coordinator, Sara speaks to her home care aide and community members to identify other services she may need, such as food assistance and transportation. EXPERIENCES WITH MEDICAL AND COMMUNITY SERVICES Sara has a good relationship with her doctors and her home care aide. Her doctor's office helps her make appointments and her home care aide helps her keep track of these appointments. Sara requires food and transportation assistance, which her home care aide helps her access. She has tried to access housing assistance and has even asked her care coordinator about these services but has not been able to navigate the system. She finds that the NEMT service she was referred to by the plan was unreliable and has turned to using family or her home care aide instead. PERCEPTION OF HEALTH PLAN Sara is generally happy with her M| Health Link plan and the medical services she receives. She feels that her care coordinator listens to her needs and helps her address them. UNFULFILLED NEEDS Sara finds that most of her needs are met by the health plan, but she could use some more assistance accessing housing assistance and more reliable transportation. Mathematica® Inc. MMCO Dually Eligible Individuals Summary of the Michigan study Key Findings Care coordination Most beneficiaries receiving HCBS reported a strong relationship with their MI Health Link care coordinator, and most beneficiaries in the MCC group reported feeling satisfied with their care coordinator. A couple of those in the BH group reported receiving care coordination services through their regional PIHP. However, most beneficiaries with BH needs either reported that they did not receive care coordination through their MI Health Link plan or that they were frustrated with the care coordination services they received. They cited insufficient support from care coordinators such as requesting assistance with o "If | need something, my coordinator is going to make sure I get it." -Beneficiary receiving HCBS in southwestern Michigan P "When | really needed the help, [my care coordinator] couldn't get it done." -Beneficiary receiving BH services in the Upper Peninsula accessing Durable Medical Equipment (DME), such as a walker, and/or getting help with coordinating appointments with providers. Several beneficiaries from all three subgroups reported turnover (frequent change), of their assigned care coordinators. These beneficiaries expressed frustration with having to reiterate their health conditions and needs to multiple new people and would have preferred one consistent care coordinator.' Access Most beneficiaries across all three higher-risk groups reported that they were satisfied with their access to medical care and the medical services they received. A majority of beneficiaries in the HCBS group reported that they could access the home-based care services they needed. Challenges Beneficiaries in the rural areas of the Upper Peninsula were more likely to report challenges with accessing medical specialists than beneficiaries in other parts of the state. o "Everything | can't do, they give me someone [home health aide] that can do it for me." -Beneficiary receiving HCBS in southwestern Michigan P "l was disappointed because | went to a [specialist] 30 miles from me, but | got a notice last summer that he left the area and the clinic closed. | had to go 60 miles away for a new [specialist]." -Beneficiary with MCCs in the Upper Peninsula 9 Notwithstanding these findings, the demonstration's Contract Management Team reported that ICOs have been able to sufficiently recruit care coordinators to meet coordinator-to-enrollee ratio requirements established in 2022. Mathematica® Inc. MMCO Dually Eligible Individuals Summary of the Michigan study Some beneficiaries reported challenges with accessing reliable NEMT services. ' Interviewees from the Upper Peninsula were more likely to report that NEMT was unavailable, while interviewees in the southern part of the state mentioned NEMT was sometimes unreliable. A few beneficiaries with BH needs said they experienced challenges with accessing DME, such as a i walker, chair lift, or mobility scooter. A couple of "I need transportation to go to physical beneficiaries in the BH subgroup said they spoke with therapy.... The problem for me is getting their care coordinator about needing certain DME, but . /v ¢ the people [from NEMT] initially their care coordinators were not able to help, and some . v /0. i1 g what I need [like] the reported having to pay for DME out of pocket. doctor's name and address and all Many beneficiaries across all three subgroups reported that. o It's my fault probably for not learning about social services such as food assistance ~ following up as QOOd-"I am not good on from friends, community members, and the state-not phones or compulter. from their care coordinator. -Beneficiary receiving BH services in the Upper Peninsula Impact of the COVID-19 PHE Across all higher-risk groups and geographic areas, most beneficiaries reported that they continued to receive their medical services, social services, and HCBS without major disruptions during the COVID-19 PHE. Several beneficiaries from different regions of the state said they began to use telehealth for medical appointments during the PHE. Most of these beneficiaries found telehealth easy to access. One person from the Upper Peninsula described how her health plan facilitated continued care during the PHE by providing a laptop to use for telchealth appointments. Possible Next Steps Based on the most significant issues raised in the interviews, we identified the following opportunities for improving care coordination and access to services for MI Health Link beneficiaries: Use outreach to build stronger relationships between care coordinators and beneficiaries. Beneficiaries who receive BH services in particular may benefit from enhanced care coordinator outreach to establish rapport and identify unmet health and social needs. Take steps to enhance consistency in care coordinator assignments. Our findings suggest that beneficiaries value a personal connection with their care coordinator and benefit when their care coordinator has firsthand knowledge of their health needs and conditions as they change over time. Improving care coordinator consistency and retention could improve beneficiary satisfaction. Leverage care coordinators to help beneficiaries access social services. Beneficiaries could benefit from additional support from care coordinators to find and access the social services they need. For example, based on needs identified in health risk assessments, care coordinators could provide lists of 10 This challenge is not specific to Michigan. Medicaid beneficiaries in many states have reported challenges with their NEMT benefits. (Medicaid and CHIP Payment and Access Commission (MACPAC). "Report to Congress on Medicaid and CHIP." June 2021. Available at https://www.macpac.gov/wp-content/uploads/2021/06/June-2021- Report-to-Congress-on-Medicaid-and-CHIP.pdf. Accessed May 3, 2022.) Mathematica® Inc. MMCO Dually Eligible Individuals Summary of the Michigan study available government-provided and community-based services and/or refer beneficiaries to specific agencies for social services. o TImprove access to transportation. Beneficiaries could benefit from additional transportation resources, such as information about public transit and local programs that provide free or low-cost transportation, or by utilizing ride-sharing companies to alleviate transportation access issues. Beneficiaries also could benefit from other resources that alleviate the need for transportation services in rural areas, such as telehealth opportunities or providers who travel to local communities. Limitations of Qualitative Research Due to the qualitative nature of the research and small sample size, this study has the following limitations: e Additional themes may have emerged from a broader or different sample of research participants. e Beneficiaries who chose to participate in this study may have felt more strongly about their interaction with their MI Health Link ICO (either positively or negatively) than beneficiaries who chose not to participate. The views of nonparticipants regarding care coordination or other matters may differ those of research participants. e The demographics of study participants did not completely align with those of the broader population of beneficiaries enrolled in an ICO; study participants were more likely to be African American or Hispanic/Latin(o/a/e) due to our intentional oversampling of these populations. e We were unable to determine how the mean age of those enrolled in MI Health Link ICOs overall compared with the average age of study participants as we did not receive the latter from interviewees. e The findings regarding lack of contact and care coordinator support for the BH group should be interpreted with caution, given the structure of Michigan's carve-out for Medicaid BH services and the potential for some participant confusion around the source of their care coordination services: PIHP or ICO. o Although the study team asked questions designed to elicit participants' recall of care coordination experiences, some beneficiaries may not have remembered a call from a care coordinator or receiving a care plan in the mail, even if a care coordinator had reached out to them and/or if they received a care plan. We based our findings only on what beneficiaries reported in their interviews. e The approach we used for defining beneficiaries at higher risk is different from the approach that ICOs use to identify enrollees requiring more frequent follow-up from a care coordinator. It is possible that research participants included those identified by their ICO as moderate or low risk, therefore requiring less frequent follow-up. e Although the findings of the report reflect the views of the beneficiaries interviewed, some of the themes discussed are not unique to MI Health Link ICOs. For example, issues with NEMT are common across many state Medicaid programs, and retention of care coordinators is difficult for programs across the country.!! 11 Retention of care coordinators was explored in previous research conducted around MI Health Link. (RTI International. "Early Findings on Care Coordination in Capitated Medicare-Medicaid Plans Under the Financial Alignment Initiative." Baltimore, MD: Center for Medicare and Medicaid Innovation, March 2017. Available at https://innovation.cms.gov/files/reports/fai-carecoordination-issuebrief.pdf. Accessed August 22, 2022.) Mathematica® Inc. | Mathematica Inc. Princeton, NJ * Ann Arbor, MI « Cambridge, MA " Mathematica. Chicago, IL « QOakland, CA * Seattle, WA w Woodlawn, MD ¢ Washington, DC Progress Together mathematica.org website EDI Global, a Mathematica Company Operating in Tanzania, Uganda, Kenya, Mozambique, and the United Kingdom Mathematica, Progress Together, and the "spotlight M" logo are registered trademarks of Mathematica Inc.