Improving the Coordination of Services for Adults with Mental Health and Substance Use Disorders: Profiles of Four State Medicaid Initiatives. IV. Cross-State Themes and Observations


Each state implemented a different approach to improving care coordination for Medicaid beneficiaries with behavioral health conditions. Despite these diversities, state Medicaid officials, managed care representatives, providers, and consumer representatives in all states identified similar facilitators, challenges, and lessons learned for improving coordination. In this chapter, we present these cross-state themes and observations, organized according to our five research questions.

A. To What Extent Does the Strategy Being Implemented in Each State Involve Changes in the Financing, Delivery, and Scope of Services Available to Adult Medicaid Beneficiaries?

These states are moving toward greater reliance on Medicaid managed care entities to coordinate care. Louisiana and Tennessee hope that new Medicaid managed care arrangements will facilitate improved care coordination. Tennessee currently contracts with three MCOs that now carve in behavioral health services. Louisiana has a single MBHO responsible for all specialty behavioral health care (physical health services continue to be managed by separate MCOs). Louisiana also has expanded the scope of mental health and substance abuse services available to Medicaid beneficiaries, particularly those having SMIs.1 Both states required care coordination as part of their managed care contracts. Such coordination typically includes telephone referrals and follow-up, and when necessary, more intensive support for consumers and providers. In addition, these state Medicaid programs cover in-person case management as part of their managed care arrangements.

Rather than alter benefits at the plan level, Illinois and Massachusetts are experimenting with mechanisms to reimburse in-person care coordination. Illinois is using state funding to support local CCEs (which operate outside of capitated managed care arrangements) to spur development of care coordination models. Although the CCEs do not operate at full risk for their enrolled clients, they are responsible for coordinating services across multiple providers in a manner similar to MCOs. In addition, the state hopes that CCEs will market their care coordination services to MCOs in the future. State officials believe that CCEs can provide more intensive in-person care coordination, compared with the more traditional telephonic support the MCOs provide. In Massachusetts, the state Medicaid program reimburses CSPECH care coordination services as part of the MBHO's capitated rate. CSPECH is currently available only to Medicaid beneficiaries enrolled with the MBHO for their behavioral health services. However, the state Medicaid program is developing some incentives to encourage other MCOs to adopt the CSPECH model.

Reimbursement for case management or care coordination is a key element of these strategies. State officials, managed care representatives, providers, and consumer representatives in all states stressed the importance of reimbursement for case management or care coordination services. Providers in all states noted the value of dedicated case managers or care coordinators responsible for connecting individuals with serious behavioral health challenges to health and behavioral health providers, and helping them follow through with treatment plans. Physicians and behavioral health specialists said that case managers and care coordinators provide critical information to inform treatment planning. For example, a CCE provider in Illinois noted that the care coordination team provides important contextual information about their clients, including any home-related barriers to care and whether patients are receiving care elsewhere. One provider said, "I can prescribe medication and I can order a test, but it's the social issues, the financial barriers, the home issues that need to be addressed. I need someone else to address those issues so that I can treat their health issues effectively." Likewise, CSPECH providers in Massachusetts report that care coordination services are so critical to their clients that they often continue to provide such support without reimbursement, even when individuals lose CSPECH eligibility.2

Consumer representatives also noted the importance of having dedicated case managers and care coordinators to help them navigate services. Stakeholders reported that case management and care coordination would be severely limited if the financing for these services was not directly reimbursable or built into the capitated rates. Providers and state officials in particular were concerned that such services would cease if financing was not available because some behavioral health providers do not have the capacity to take on intensive case management in the absence of funding.

States have taken different approaches to changing billing processes and procedures to coordinate care. Providers and officials in all states noted the importance of thoughtfully engineered and executed billing policies and processes. Providers and state officials reported that billing delays have serious implications for providers, who often operate without much financial cushion and have difficulty in absorbing payment delays. For example, some of the providers interviewed in Louisiana expressed frustration regarding challenges stemming from the implementation of a new billing system and revenue cuts, which they attributed to the implementation of the managed care arrangement. These billing and revenue changes resulted in some providers closing, reducing staff, or refusing to provide care to Medicaid beneficiaries. Likewise, several providers in Tennessee reported that they initially struggled to align internal billing policies to accommodate the different billing processes of multiple MCOs. This was a particular challenge for some community mental health centers (CMHCs), which had little experience in contracting with multiple private insurers. In both states, state officials and managed care entities responded to these concerns by offering provider education and ongoing support to help ease the transition in payment practices.

In contrast, because the initiative in Massachusetts did not involve changes at the health plan level, it built on the existing billing structure. CSPECH providers noted that being able to bill for services under a pre-existing service category minimized disruption in payment. In addition, the MBHO's requirement that providers of CSPECH services already be part of its provider network ensures a minimal need for billing-related training and assistance. (That is, providers must already provide other Medicaid-reimbursable services.)

States are making efforts to ensure that billing codes and service eligibility criteria facilitate ongoing team-based care. State officials, managed care representatives, and providers noted two issues that have had an impact on their ability to provide team-based coordinated care. First, the reimbursement mechanism must offer flexibility. For example, several providers in Tennessee reported that they prefer daily or case rates for case management (such as those used in the Massachusetts and Illinois initiatives). They perceive these rates as offering more flexibility for case management services compared with fee-for-service reimbursement. Second, providers and consumers noted that the service reauthorization processes and medical necessity criteria for services can provide barriers to care or disrupt recovery. These stakeholders stressed that individuals need continued access to services and team-based care coordination that enable functioning above the level of medical necessity. However, when an individual exceeds medical necessity criteria, he or she will lose benefits and access to care. One provider expressed frustration regarding service reauthorizations and said that clients must "fail in order to succeed again."

B. In What Ways Do These Strategies Seek to Improve Care Coordination by Fostering New Partnerships and Developing New Care Coordination Mechanisms?

States are seeking to foster and capitalize on local partnerships between providers. Although each state adopted a different strategy and formed various partnerships at the state level, all four recognized that the success of their initiatives depended on the strength of provider buy-in and local partnerships. For example, Illinois and Massachusetts officials positioned their initiatives within local community-based organizations that understand the needs of their communities and are familiar with local resources. Providers and consumers also noted that case managers and care coordinators in these states often live in the same communities in which they work, and have shared experience with the clients they serve. Likewise, the success of the initiatives in Tennessee and Louisiana depend on building local provider networks and relying on them to identify community resources. Both states hope that the plan-level changes they have adopted will ultimately encourage local collaboration, including co-location of physical and behavioral health services.

Strong communication among providers, MCOs, state agencies, and consumer organizations facilitates smooth implementation of care coordination strategies. Stakeholders in these states noted the importance of careful planning and transparency when making changes in reimbursement and implementing different approaches to care coordination. For example, providers in Massachusetts and Illinois saw a genuine sense of partnership with the state Medicaid agency, which allowed for flexibility in developing localized coordination models. The CCEs in Illinois were particularly reliant on open and direct communication with the state Medicaid agency--they were led by community-based providers with limited experience and capacity for creating the necessary infrastructure associated with establishing a new entity (such as developing contracts, data-sharing agreements, and other legal documents).

In Tennessee and Louisiana, the managed care entities and state agencies provide opportunities to receive input from providers and consumer representatives, often through advisory board meetings and dedicated provider or consumer relations specialists. Providers and consumer representatives in these states noted the importance of offering more opportunities for dialogue and ensuring that changes in policies and payment practices represent their interests and priorities. Providers in these states also stated the importance of having the opportunity to communicate with managed care entities and state agencies regarding contract language and service definitions.

State Medicaid, mental health, and substance abuse agencies have played different roles in the development and implementation of these initiatives. Given the differences in the service delivery and financing reforms across these states, the role of the state agencies also has differed. Tennessee and Louisiana implemented substantial reforms in their Medicaid benefits, reimbursement strategies, and managed care arrangements that required strong partnerships between the state Medicaid program and behavioral health agencies, and continued involvement of these agencies in implementing the programs. In contrast, the involvement of state agencies in Illinois and Massachusetts has been somewhat more limited. In Illinois, the state Medicaid agency likens its role to that of a venture capitalist--providing CCEs with an initial investment (auto-enrolled clients, claims data, and technical assistance with establishing contracts and legal arrangements). The agency sees its role after that as supportive only, encouraging CCEs to seek their own path toward sustainability. Likewise, the Massachusetts Medicaid agency has been supportive of the MBHO's implementation of CSPECH and has encouraged other managed care entities to adopt the model; the agency itself has not been integrally involved in CSPECH's implementation, however.

Substance abuse services are regarded as critical but are coordinated in different ways. Underlying all four strategies is the belief that there must be coordination with substance abuse services. However, the level of coordination with state-level and local-level substance abuse service partners varies. In Louisiana, the MBHO is at-risk for both mental health and substance abuse services. The state began consolidating its own substance abuse and mental health clinics before implementing the managed behavioral health arrangement, with the goal of creating integrated care settings to address co-occurring disorders. In Tennessee, Medicaid mental health and substance abuse services are similarly covered under the managed care arrangement, although providers and consumer representatives mentioned significant limitations in the availability of such services. The initiatives in Illinois and Massachusetts promote coordination with substance abuse services through local provider partnerships. For example, the solicitation for the CCEs in Illinois required inclusion of local substance abuse service providers; the state agency responsible for overseeing these services was not a partner in developing the initiative, however. In Massachusetts, substance abuse services are available through all CSPECH sites, either directly, through the lead CSPECH agency, or through a partnership with a local provider.

Partnerships with criminal justice system services remain somewhat limited. Although all states recognize that a sizable percentage of clients with behavioral health needs have experience with the criminal justice system, partnerships with justice counterparts vary. Some state initiatives have involved strengthening partnerships between state-level Medicaid, behavioral health, and criminal justice agencies, whereas others have focused on strengthening such linkages at the local level. However, stakeholders in every state reported the need for stronger collaboration with the criminal justice system and noted some achievements in developing strategies to help individuals with mental health conditions transition from incarceration to the community. For example, the Louisiana OBH (which oversees the LBHP) and the Department of Corrections have established procedures to facilitate continuous access to medications when individuals transition from prison to the community. Although the Illinois and Massachusetts initiatives do not involve state-level partnerships with the criminal justice system, state officials, providers, and consumers noted the importance of local partnerships. Care coordination team members in these states developed relationships with local prisons and jails as well as jail diversion programs, and reported that they often help clients obtain housing and other supports to allow the individual to remain in the community after incarceration.

Managed care entities and provider networks offer data-sharing opportunities to enhance coordination. State officials, managed care representatives, and providers all cited the importance of data-sharing to facilitate care coordination. Each state has different data-sharing capacities. The MCOs in Tennessee, which are responsible for physical and behavioral health benefits, are encouraged to develop data systems that integrate information on physical and behavioral health service utilization. The MCOs can use this data to alert providers of transitions in care, such as when a beneficiary may need follow-up after an ED visit. Providers can also request information on a client's service utilization to inform treatment planning. In an effort to improve coordination with physical health services, the MBHO recently began to receive data on Medicaid pharmacy and physical health claims, and is planning to exchange encounter data with physical health plans. Behavioral health providers in Louisiana can contact the MBHO to request information on a client's service utilization.

Illinois and Massachusetts have somewhat limited data-sharing arrangements. The CCEs in Illinois receive Medicaid claims data only; there is limited data-sharing between providers within a CCE, which several providers identified as a barrier to coordinating care effectively. Many CCE care coordination teams have developed their own systems for tracking service use and patient encounters. However, these systems are not integrated with those used by the individual CCE provider agencies. Low client enrollment and uncertainty regarding the long-term sustainability of CCEs limit providers' willingness to invest significant resources in developing a common data platform to use across all CCE providers. CSPECH providers in Massachusetts receive claims data for behavioral health services to help inform care coordination decisions; they may also have access to information about physical health care services if a client signs a release of information. Because the CSPECH partnerships include far fewer organizations than the CCEs (some CSPECH sites consist of a single agency that provides both behavioral health and housing support services), data-sharing and access was less of a concern.

C. How Do Various Stakeholders Perceive These Strategies as Influencing the Accessibility and Quality of Care, and Health Care Utilization and Costs?

Stakeholders across states were cautious in drawing conclusions about the impact of these care coordination strategies. Stakeholders in Illinois and Louisiana expressed concern that more time was needed for implementation before drawing conclusions. Although the program in Tennessee is more established, respondents were hesitant to make assertions regarding outcomes. Likewise, the CSPECH program in Massachusetts first was piloted in 2006, but stakeholders were hesitant to draw conclusions on its generalizability to broader patient populations or other contexts, given that it serves a relatively narrow population. Although most stakeholders were cautious in drawing conclusions, they did offer their perceptions of whether their programs were moving in the right direction and identified several mechanisms that might facilitate or impede success, as described below.

Most stakeholders perceived that the strategies being implemented in their states will ultimately lead to improvements in the quality of care. In Tennessee, several providers and consumer groups reported that introducing multiple MCOs and managed care contracting has encouraged competition among providers, which in turn could lead to improved efficiency and innovation. State officials and managed care representatives in Louisiana believe that stronger provider credentialing and defined coordination requirements for the managed care entity will lead to improvements in the quality of care. Some Louisiana providers, however, expressed fear that the MCOs' required service authorizations could restrict access to care or delay receipt of services. Providers in all four states noted the value of data-sharing to drive quality improvement. However, as mentioned above, these state programs have very different capacities for data-sharing. Finally, providers and consumers in Illinois and Massachusetts believe that the local partnerships developed through their state initiatives have increased providers' knowledge of resources available in their communities, which will lead to improvements in connecting clients with services.

Stakeholders reported that the strategies used in their states may improve access to care, but not without some challenges. Provider and consumer stakeholders in both Louisiana and Tennessee raised concerns about access to care in the short term. In Louisiana, several stakeholders observed that revenue cuts and reimbursement delays (which they attributed to the new managed care arrangement and changes in state funding) may have led some providers to close, reduce staff, or refuse to serve Medicaid beneficiaries with severe behavioral health issues. (This study could not verify such assertions.) Although providers in Tennessee reported that competition between them may ultimately lead to innovations in the delivery of care, some also expressed concern that certain providers have found it necessary to merge and affiliate to increase their negotiating power with the MCOs--thus potentially limiting the number of independent providers. Some also said that the multiple-MCO structure in Tennessee results in frequent changes to provider networks, which may create confusion for consumers trying to access care.

Providers and consumers emphasized the importance of in-person care coordination and case management as essential for improving access to care. Providers and consumer representatives in Illinois and Massachusetts believe that the care coordination initiatives in their states are making services accessible to individuals who previously had been isolated from health and social service systems. Likewise, property managers in Massachusetts noted that CSPECH provides the support necessary for clients to remain housed while connecting them with services.

All states expect to experience improvements in service utilization patterns and reductions in costs. Respondents in Louisiana and Tennessee were unable to say whether the plan-level changes have impacted service cost and use, although their expectation is that costs would decrease over time. Respondents in Illinois and Massachusetts were more willing to identify changes in service use, particularly a decline in the use of crisis intervention services. Staff associated with the two CCEs in Illinois, for example, reported an immediate reduction in ED use among enrolled clients. Providers associated with the CSPECH program in Massachusetts reported similar reductions in the use of crisis services once individuals are housed and receiving CSPECH services.

D. How Do These Strategies Provide or Coordinate with Housing and Other Social Services?

State initiatives are coordinating with housing providers and fostering local partnerships between behavioral health and housing providers. Each of these states is seeking to improve coordination of housing supports and services for Medicaid beneficiaries with behavioral health needs (see Table IV.1 for a description of the housing component of each state initiative). State Medicaid and behavioral health agencies are also forging partnerships with housing stakeholders at the state and local levels.

The partnership structures in the four states differ, based on local contexts and funding mechanisms. In some states, the reimbursement mechanisms grant providers the flexibility to develop collaborations that take advantage of existing local housing infrastructure. For example, in Illinois, the Medicaid agency encouraged the inclusion of housing providers and other local housing resources as part of the CCEs. In Massachusetts, the behavioral health entity has structured the CSPECH benefit flexibly enough to enable different partnership structures: three of the eight CSPECH providers are partnerships between a CMHC and one or more housing providers, two are led by a federally qualified health center that has partnered with one or more local housing providers, and three are led by a single organization that provides both behavioral health services and housing. Similarly, behavioral health providers in Tennessee either own and operate a group home themselves or contract with local housing providers who own and operate the housing and support services. Louisiana's housing program, which was built with new housing subsidies, created new formal partnerships between state housing and behavioral health agencies, the managed care entity, local behavioral health providers, landlords, and property managers.

Such collaborations are particularly helpful for identifying limited housing resources, since Medicaid covers only physical and behavioral health services. Local behavioral health providers are building relationships with housing providers, property managers, and developers, educating them about the supports their tenants will receive, and making themselves available to address the physical health, mental health, and psychosocial barriers that often place individuals at risk of losing housing. From the perspective of housing providers, these state initiatives may be an opportunity to further their own housing goals. For example, one CCE housing provider in Illinois noted, "When thinking about the future of health care, housing must be part of the equation and [the CCE initiative] is an opportunity to have housing recognized, even if related [housing] services are not yet being covered [by the CCE's PMPM fee]."

States are encouraging managed care plans to focus on the coordination of behavioral health and housing services. Three of the states--Louisiana, Massachusetts, and Tennessee--have required or allowed their Medicaid managed care plans to develop and reimburse care coordination or case management strategies specifically focused on beneficiaries with housing needs. The managed care entities in these states have a financial incentive to keep individuals housed and control costs, and they are well positioned to use their data to monitor service use and link data on behavioral health and housing participation (as the MBHO in Louisiana has done).

TABLE IV.1. State Strategies to Coordinate Services with Housing
ILLINOIS: Medicaid-funded CCEs foster partnerships between local health and housing providers. Although CCEs primarily focus on the coordination of behavioral and physical health care, the CCEs include housing partners, which often employ coordination team members with housing experience. The state Medicaid agency auto-enrolled Medicaid beneficiaries with complex behavioral health needs, some of whom were homeless. Although CCEs can use the PMPM to cover non-Medicaid-reimbursable costs, such as housing-related expenses, they are unlikely to do so unless enrollment can be scaled up. For housing providers whose services are neither Medicaid-reimbursable nor covered by the PMPM, CCE membership is an investment and part of a long-term strategy to be in on the conversation regarding the evolving health service system. The CCE initiative has not created new housing options or supports; rather, CCEs facilitate access to existing housing and subsidies through their housing provider members.
LOUISIANA: Management of PSH program services is incorporated into the MBHO contract. Louisiana's PSH program subsidizes approximately 3,300 private rental units and offers Medicaid-covered behavioral health and long-term care services. The PSH program originated in 2005, when the state received federal funding to rebuild affordable housing after devastating hurricanes. In 2013, funding for the behavioral health services provided to individuals in PSH shifted from time-limited community development block grants to Medicaid (and specifically into the MBHO contract). This shift was driven by the need for a sustainable funding source. The MBHO also became responsible for screening PSH applications for Medicaid eligibility, coordinating applications and housing placements with the state housing agency, and managing tenant-landlord relationships, in part due to the state's desire to have a single entity centralize these processes. PSH participants must qualify for the federal housing subsidies, Medicaid, and the 1915(i) State Plan Amendment, which authorizes intensive behavioral health services. Individuals not eligible for 1915(i) services, including those with an SUD but not an SMI, are no longer eligible for the PSH program.
MASSACHUSETTS: The Medicaid MBHO covers care coordination for chronically homeless individuals in PSH. The Massachusetts initiative included in this review is fully rooted in a Housing-First initiative; thus, the CSPECH services are available only when the individual's housing needs have been met. Coordination support, provided by a CSW, is reimbursable beginning 90 days before an individual is housed, an arrangement that makes it easier for CSWs to help homeless individuals secure housing. Services continue as long as the individual remains housed in the subsidized unit. CSWs work with a caseload of up to 12 clients, as required by the MBHO, allowing them the time needed to provide intense coordination support to clients. To be eligible for CSPECH, individuals must meet the U.S. Department of Housing and Urban Development's definition of chronic homelessness and elect to enroll in the MBHP plan. The availability of CSPECH services is also constrained by the limited supply of affordable housing and housing vouchers; receipt of services is guaranteed only after a housing unit and subsidy are available.
TENNESSEE: Supervised group homes are integrated into the managed care arrangement. The state's Medicaid program includes a supported housing benefit, which covers support services provided in supervised group housing facilities staffed by mental health providers around the clock and with fewer than 16 beds. The benefit is intended to be relatively temporary and serve as a bridge for individuals coming from institutions and other restrictive settings into more independent living in the community. To be eligible for group home placement, individuals must meet medical necessity and level of care standards, and require services and supports in a highly structured setting. The supported housing benefit thus offers limited coverage to a narrow population. While in a group home, individuals can receive various services covered under the supported housing benefit. For example, through one of the managed care plans, group home residents work with mental health professionals to create individualized and detailed care and discharge plans that address mental health, physical health, and substance abuse conditions. The discharge plans include a housing transition plan that specifies possible housing options for the client, along with next steps for choosing 1 of those options.

The reimbursement mechanisms that facilitate care coordination or case management are critical for helping individuals obtain and maintain housing. Although consumers might access supportive services and housing subsidies through other means, representatives from consumer groups, behavioral health providers, and housing providers consider these state strategies to improve care coordination as critical for housing individuals with severe needs. These stakeholders believe that, without intensive, coordinated support, maintaining housing would be a considerable challenge for many of the individuals they serve. As a behavioral health provider in Louisiana explained, "Getting someone into PSH and keeping them there is largely dependent on the relationship they have with the supports that keep them [housed]." Some providers in Massachusetts and Louisiana said that linking behavioral health and psychosocial services with housing connects consumers to needed services they would not otherwise access. Some consumers who formerly were chronically homeless in Massachusetts regard CSPECH as a "life saver." A few managers of homeless shelters reported that the availability of CSPECH has also made property managers more comfortable with renting to tenants who need supportive services.

E. What Data are Available in Each State to Potentially Facilitate the Monitoring of These Efforts and Future Evaluations?

Given that each state is implementing a different strategy, the data available for monitoring or evaluating these efforts vary across states. In this section, for each state, we first identify data sources available for a potential evaluation and then describe notable strengths and limitations associated with these sources.

1. Illinois

Physical and behavioral health service data. CCE participants are identifiable in the state fee-for-service Medicaid claims data, which the state uses to track patient outcomes, paying particular attention to hospitalizations and ED use. The Department of Healthcare and Family Services (HFS), which oversees the CCE initiatives, processes all claims related to physical and mental health, and the Department for Alcohol and Substance Abuse Services processes substance abuse service claims.

Provider and consumer experience data. HFS also tracks consumer and provider grievances, which the CCEs receive and HFS submits and handles. Although some CCEs collect information on consumer satisfaction, there are no common or comprehensive surveys conducted across all CCEs.

Housing and social service data. Each CCE maintains its own process for tracking participant-level data, with many using a client database or electronic health record (EHR) system. The data captured and tracked vary by CCE but may include a range of information, including care coordination encounters; housing status and use of housing subsidies; receipt of homelessness prevention services (typically through access to Homeless Management Information Systems [HMIS] by one of the CCE partners); use of other services not reimbursable by Medicaid, such as transportation or employment services; use of other safety net benefits, such as Supplemental Security Income (SSI), Social Security Disability Insurance, and the Supplemental Nutrition Assistance Program (SNAP); education; employment; criminal involvement; and consumer satisfaction with care. Some of this information is collected through assessments that the care coordination team conducts; CCE providers compile other information. Although some CCEs collect a wealth of data on clients served, these data elements are not consistent across CCEs and they are not submitted to HFS.

Data strengths and limitations. Although it is possible to identify CCE participants in the state Medicaid claims data, these data cannot be used to observe the delivery of CCE care coordination services because those services are not billed separately. Although CCEs track their encounters with patients, service use, and some limited patient outcomes, no consistent method is used across CCEs. Rather, each CCE care coordination team has developed its own mechanism for tracking client data and encounters, with several adopting a client database or EHR system. The types of data tracked through these efforts vary greatly by CCE. Each CCE consists of individual pre-existing provider organizations that often maintain their own distinct EHR or client-tracking systems. For use of all services not reimbursable by Medicaid, including housing-related supports and subsidies, care coordination team members must rely on data tracked separately by their CCE housing and social service partners. Some CCEs have made an effort to enter this information manually into their EHR system but again, the level of data integration varies widely. No CCE has yet invested in developing an EHR system that can be used or accessed by all the participating provider organizations as well as the care coordination team.

2. Louisiana

Physical and behavioral health service data. The MBHO's data systems for authorizing services and paying provider claims capture a range of behavioral health service use and expenditure data, including on mental and substance abuse services, psychiatric hospitalization, and ED visits. The MBHO sends these data files to the state, along with client-level data extracted from the MBHO's EHR system and billing platform. Although the MBHO does not systematically collect data on primary care service use, it has begun receiving information about physical health encounters from the state's Medicaid fiscal intermediary.

Provider and consumer experience data. The MBHO conducts consumer and provider satisfaction surveys at least once a year. It administers the provider survey to all network providers who rendered services during the survey period. It administers consumer surveys to a representative sample of members, gathering input on the care received from network providers. As required by the state, the MBHO also has a grievance process that allows consumers or providers to file complaints about the treatment they receive from the MBHO (or its providers); there is a separate process for appealing decisions on service authorizations or payments.

Housing and social service data. The MBHO tracks data related to the PSH program, including on application status, housing status, and supportive service providers. Although the MBHO's PSH records only date back to October 2013, when it began managing these services, it can link PSH data to current and historical behavioral health data for those members enrolled in PSH.

Health and social indicators. The MBHO collects systematic information on health and social indicators through National Outcomes Measures (NOMs) data requirements established by the HHS Substance Abuse and Mental Health Services Administration (SAMHSA) and the Level of Care Utilization System (LOCUS) assessment of behavioral health functioning. The LOCUS assessment is administered to all individuals with SMI applying for 1915(i) State Plan services. Providers may also track health and social indicators through the MBHO's EHR system; however, such data are not available for all individuals served by the MBHO because the state has not required providers to use this system.

Data strengths and limitations. The MBHO and the state did not express any major concerns about the completeness or reliability of the MBHO's systems for capturing behavioral health encounters. Presumably, these data are relatively complete, given that all providers within the state must submit claims to the MBHO to receive reimbursement. Early in the LBHP implementation process, there were reports of claims not being paid because providers were unfamiliar with the new billing systems and the MBHO's new electronic billing system was still being fine-tuned. Thus, some services may have been provided but not captured in the system. However, it is unclear to what extent this continues to be an issue or whether it is possible to identify rejected claims in their data.

The MBHO only began collecting data on participants in the PSH program in October 2013. Data for previous years would need to be obtained from the state agency that previously managed the program.

Although the MBHO's EHR system could include data on health and social indicators, not all providers within the MBHO's network use the system. Providers interviewed conveyed mixed messages about the usability of and the extent to which they use the EHR system and noted that it has evolved over time; this suggests that EHR data are limited. Although the system is probably not a complete and reliable source of data at this time, it has the potential to be useful for evaluation in the future. Health and social indicators that derive from the LOCUS and NOMs assessments are likely to be consistently available; however, as of this report, the LOCUS assessment data are not in a format that can be used for an evaluation.

3. Massachusetts

Physical and behavioral health service data. Individuals who receive CSPECH services are identifiable in the state Medicaid claims data maintained by MassHealth, the state's Medicaid agency. The MBHO is the payer for all mental health and substance abuse service claims associated with CSPECH clients. The Department of Mental Health processes all claims associated with mental health and psychosocial rehabilitation services, whereas providers send physical health claims directly to MassHealth. CSPECH care coordination services are Medicaid-reimbursable and thus identifiable in the claims data. The MBHO reimburses CSPECH providers using a flat per-day rate for each client in the program.

Provider and consumer experience data. The MBHO tracks consumer satisfaction with care, timely access to care, provider satisfaction, and consumer and provider grievances, although none of these indicators is tracked uniquely for the CSPECH program. These data are collected through surveys and the grievance submission process. However, consumers and providers are surveyed on services beyond CSPECH; thus, these data sources would have limited value to an evaluation of the CSPECH program.

Housing and social service data. Each individual CSPECH site--which consists of either a single organization that provides both housing and care coordination services or a health and housing partnership--maintains its own EHR system. The MBHO does not require the CSPECH providers to track and submit data beyond what is required through the claims process. However, some providers do track other client-level indicators, such as physical and mental health status and functioning; criminal justice system involvement; housing status; use of housing-related supports and subsidies; homelessness-related services (many of the providers operate shelters and have access to HMIS); and use of other safety net services and benefits, such as SSI and SNAP.

Data strengths and limitations. Medicaid claims are probably the strongest data source for evaluating CSPECH client outcomes--particularly service use and costs associated with the program. Because providers are continuously reimbursed for providing the care coordination services that define the CSPECH program, it is possible to identify when an individual enters and exits the program. (However, few clients ever exit the program because receipt of CSPECH services is attached to the housing subsidy.) Although CSPECH providers do track a range of other information on clients, the completeness and reliability of those data vary across providers. For example, the MBHO expects providers to monitor client mental health and physical function, but there is no required method or data collection tool for measuring these indicators; thus, provider-level data may not be comparable. Furthermore, each CSPECH provider site uses its own EHR system, so the way they collect and organize data will also vary.

4. Tennessee

Physical and behavioral health service data. The Bureau of TennCare collects data on service use and expenditures for both physical and behavioral health care, as well as long-term care services for all of its members. Claims and encounter data generally are available going back to 2007, when the state began integrating mental health services into its managed care contracts. The MCOs in the state send encounter data files, to the Bureau of TennCare at the end of each payment cycle. Additional data on service use for mental health and SUDs are available on individuals who receive treatment services funded by the Department of Mental Health and Substance Abuse Services (DMHSAS) through its Behavioral Health Safety Net (BHSN) and various substance abuse prevention and treatment programs. BHSN providers submit fee-for-service claims through an electronic database that DMHSAS maintains. SUD service utilization data are collected through the Tennessee Web Infrastructure for Treatment Services. This is a robust data collection and billing database that contains a wide range of SUD service use, demographic, and other social service use information for individuals who have received SUD treatment services funded by DMHSAS. This may include some TennCare beneficiaries who have maxed out their Medicaid substance abuse benefits and are receiving care through DMHSAS programs.

Provider and consumer experience data. The Bureau requires each MCO that participates in TennCare to be accredited by the National Committee for Quality Assurance (NCQA), which, in turn, requires MCOs to collect and report consumer perceptions of care through Consumer Assessment of Healthcare Providers and Systems. To meet contract obligations and retain NCQA accreditation status, TennCare also requires MCOs to collect and report on Medicaid-relevant Healthcare Effectiveness Data and Information Set (HEDIS) physical and behavioral health performance measures, including, for example, HEDIS antidepressant medication management and follow-up after-hospitalization measures. MCOs must also report to TennCare on consumer and provider grievances and findings from provider surveys. In addition, DMHSAS collects and reports NOMs, which are used for SAMHSA reporting.

Housing and social service data. TennCare collects and maintains claims and encounter data from MCOs for the beneficiaries eligible for and receiving Medicaid-supported housing services. DMHSAS also has housing data on individuals with behavioral health disorders served through the Department's Creating Homes Initiative, which may include TennCare beneficiaries not eligible for Medicaid-supported housing.

Other indicators. The Tennessee Association of Mental Health Organizations (TAMHO) created a unique database for its member organizations, which include CMHCs and other behavioral health providers in the state. The association and its membership felt the need for a single database that collected data on behavioral health services, regardless of payer source; the association worked with a vendor to develop a database that met providers' and advocates' needs. TAMHO currently collects service use, demographics, payer source, employment and veteran status, receipt of housing supports, recent criminal history, and other data from participating providers, thus giving the association a comprehensive picture of individuals served in the public mental health system. Several providers noted that the TAMHO database is the only statewide source that includes such a wide range of data. However, not all community providers are members of TAMHO or submit information, so data are limited to those individuals served by participating member organizations.

Data strengths and limitations. Although a wealth of data is collected on Medicaid beneficiaries, each data source has important limitations. Tennessee's Medicaid claims data are a strong source, but officials note that the data quality has improved significantly since integration began. It may thus be difficult to differentiate between true outcomes and artifacts of poor data quality in the earlier years of the integration efforts. Data related to housing are limited to Medicaid beneficiaries eligible for TennCare's supportive housing benefit; beneficiaries may receive supportive housing services through other state initiatives, but there is no statewide comprehensive data source on such services (besides TAMHO, which has other data limitations). Although the state collects information on health status and functioning for various reporting requirements, it is not clear what data are collected for whom, particularly as part of the state's NOMs requirements. Finally, although TAMHO appears to be a promising data source, it is limited to participating providers and its completeness and reliability are unknown.

View full report


"4Initiative.pdf" (pdf, 488.91Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®