Skip to main content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.


The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Innovative Medicaid Managed Care Coordination Programs for Co-morbid Behavioral Health and Chronic Physical Health Conditions: Final Report

Publication Date

Printer Friendly Version in PDF Format (36 PDF pages)


The prevalence of co-morbid chronic physical and behavioral health conditions is an established driver of spiraling costs and poor health outcomes among Medicaid recipients. States are increasingly deploying new managed care models to help control Medicaid costs and promote improved health outcomes. This study examines innovations in the coordination of care for Medicaid covered individuals with chronic physical and behavioral health conditions. Six geographically dispersed Medicaid managed care health plans serve as detailed case studies, and findings from these are reviewed.

DISCLAIMER: The opinions and views expressed in this report are those of the authors. They do not necessarily reflect the views of the Department of Health and Human Services, the contractor or any other funding organization.









AmeriHealth Caritas Pennsylvania
BlueCare Tennessee
Community Health Plan of Washington
Hudson Health Plan
Rocky Mountain Health Plan
Cenpatico/Sunshine State Health Plan







We would like to thank Garrett Moran, Ph.D., Howard Goldman, M.D., Ph.D., Benjamin Druss, M.D., M.P.H., Pamela Greenberg, M.P.P., and Roger Kathol, M.D., C.P.E. for serving on our Advisory Board and providing valuable feedback throughout the development of this report. We would also like to thank the project sponsors at the Office of the Assistant Secretary for Planning and Evaluation (Kirsten Beronio, J.D., and Joel Dubenitz, Ph.D.). Most importantly, we would like to thank all of the health plans and staff that talked with us, hosted site visits, and otherwise supported the development of this study.



The following acronyms are mentioned in this report.

ADHD Attention Deficit Hyperactivity Disorder
CAD Coronary Artery Disease
CAG Complexity Assessment Grid
CHPW Community Health Plan of Washington
CHW Community Health Worker
CMSA Case Management Society of America
COPD Chronic Obstructive Pulmonary Disease
CQM Clinical Quality Measure
EHR Electronic Health Record
FQHC Federally Qualified Health Center
GAD-7 Generalized Anxiety Disorder-7 screen
HEDIS Healthcare Effectiveness Data and Information Set
ICM Integrated Case Management
IMPACT   Improving Mood - Promoting Access to Collaborative Treatment  
MCO Managed Care Organization
MHIP Mental Health Integration Program
MLR Medical Loss Ratio
PAM Patient Activation Measure
PCP Primary Care Provider
PHQ-9 Patient Health Questionnaire-9
PMPM Per Member Per Month
PRIME Payment Reform Initiative for Medicaid Expansion
RMHP Rocky Mountain Health Plan
SHAPE Sustaining integrated Healthcare Across Primary care Efforts
SMI Serious Mental Illness



The prevalence of co-morbid chronic physical and behavioral health conditions is increasingly a driver of spiraling costs and poor health outcomes among Medicaid recipients. As states and others seek to lower health costs and improve outcomes, managed care plans are now being deployed to improve the quality of care and the coordination of services for Medicaid covered beneficiaries, along with new innovations in care coordination. Examining these innovations provides guidance for states, health plans, and provider systems on new best practices and their effect on improved care.

This study examines innovations in the coordination of care for individuals with chronic physical and behavioral health conditions. Six health plans serve as detailed case studies, and findings from these are presented. The sites selected for this review are geographically dispersed and include AmeriHealth Caritas Pennsylvania (Pennsylvania); BlueCare Tennessee; Community Health Plan of Washington*; Hudson Health Plan* (New York); Rocky Mountain Health Plan (Colorado); and Cenpatico/Sunshine State Health Plan (Florida). In-person visits were conducted at four of the sites, and two (*) were telephonic reviews. This study presents a plan overview for each site, discusses the challenges of care coordination identified by each plan, and reviews the strategies that address these challenges through the health plans' innovations in care coordination.

Overall, six strategies to meet the care coordination needs of members have been identified, along with health plan innovations that support improved health outcomes. These strategies and innovations include: (1) using information technology to identify at-risk members and stratify their needs for care coordination; (2) supporting practice-based change for improved care coordination; (3) using financial incentives and payment reform to support enhanced care coordination; (4) implementing information technology to enhance care planning and shared clinical practice coordination; (5) reaching out to and engaging with covered beneficiaries in their communities; and (6) coordinating physical and behavioral health care management services within plan operations. Details of the health plans' innovations are included in case studies as well as the health plan's measurement of the effects of these innovations.



Medicaid currently provides health and long-term care coverage for more than 66 million low-income beneficiaries (Smith, Gifford, Ellis, Rudowitz, & Snyder, 2013). This accounts for one in six dollars spent on all health care in the United States. Managed care and other care coordination programs are increasingly being used to improve care, manage costs, and improve quality. The role of managed care plans in Medicaid has been expanding over recent years, and states are increasingly relying on these programs to provide quality health care to their beneficiaries. Between 2001 and 2011, the proportion of beneficiaries covered by Medicaid health plans expanded from 37 percent to 51 percent (America's Health Insurance Plan Center for Policy & Research, 2013).

Care coordination is becoming a universal attribute across most state Medicaid programs (Smith et al., 2013). In a review of state plans for Medicaid transformation, the Kaiser Commission on Medicaid and the Uninsured's 50-state Medicaid budget survey for state fiscal years 2013 and 2014 found that 40 states reported new or enhanced care coordination activity or initiatives. Strategies for care coordination at either the managed care organization (MCO) or provider levels are common, and principal goals include reducing fragmentation and improving coordination across behavioral and physical health care. Peer support services are increasingly being provided by managed behavioral health care organizations to promote engagement and improve coordination during transitions in levels of care (Association for Behavioral Health & Wellness, 2013).

States are increasingly contracting with managed care plans to improve both the physical and behavioral health care services and outcomes for their Medicaid covered beneficiaries. The prevalence of co-morbid behavioral health and chronic physical illnesses is high. Evidence suggests that individuals with physical illness who also have mental health and substance use conditions are at a higher risk of experiencing functional disabilities and a poorer quality of life compared to those with only physical conditions (Miller, Paschall, & Svendsen, 2006).

Health plans that serve Medicaid beneficiaries work within a spectrum of different state regulations and managed care arrangements. In some states, the plans have responsibility for physical health, while behavioral health is carved out to another vendor or system of care. In other states, the plans may have full managed care responsibility for all health conditions. Almost universally, Medicaid managed care plans have difficulty identifying those with co-morbid chronic conditions, providing outreach and engagement resources, and supporting integrated approaches to care that foster better self-care and improved health outcomes.

Care coordination has been identified as an important role for managed Medicaid plans to help promote high quality care for their beneficiaries, control costs, and reduce unnecessary health services. These services are defined as: "Care coordination is the deliberate organization of patient care activities between two or more participants (including the patient) involved in a patient's care to facilitate the appropriate delivery of health care services. Organizing care involves the marshalling of personnel and other resources needed to carry out all required patient care activities, and is often managed by the exchange of information among participants responsible for different aspects of care" (McDonald et al., 2007). Five key elements of care coordination are identified and include: (1) numerous participants who are typically involved in care coordination; (2) coordination is necessary when participants are dependent on each other to carry out disparate activities in a patient's care; (3) in order to carry out these activities in a coordinated way, participants need adequate knowledge about their own and others' roles and available resources; (4) to manage all required patient care activities, participants rely on an exchange of information; and (5) the integration of care activities has the goal of facilitating appropriate delivery of health care services.

McDonald and colleagues also make the key observation that care coordination looks different depending on whether one's perspective is that of the patient or family, the health care provider, or a representative of the health care system. Care coordination is particularly essential at times of transition and information transfer among people and entities involved in planning or delivering care, over time across the lifespan, and with changes in the status of illnesses or conditions (McDonald et al., 2007). As a part of this study, a comprehensive environmental scan of how care coordination is used to improve care for those with chronic physical and behavioral health conditions was conducted. This full report is included in Appendix A (separate document).



This report describes some innovative practices that are currently being developed and deployed by Medicaid managed care plans to support improved engagement, activation, and health outcomes for covered beneficiaries who have co-morbid physical and behavioral health conditions. Six health plans that provide services to Medicaid covered beneficiaries were reviewed for this study.

An initial list of more than 20 health care plan candidates, prepared in consultation with an expert advisory panel and a series of other national health care organizations, was considered for this study. Twelve were telephonically screened, and six selected. Plans were considered based on the nature of their innovations in care coordination for physical and behavioral health needs of their Medicaid beneficiaries, geographic diversity, and other factors. Six Medicaid plans were selected for this study, and four day-long site visits and two telephonic site reviews were conducted. The health plans selected were AmeriHealth Caritas Pennsylvania (Pennsylvania); BlueCare Tennessee; Community Health Plan of Washington; Hudson Health Plan (New York); Rocky Mountain Health Plan (Colorado); and Cenpatico/Sunshine State Health Plan (Florida).

Case studies for each of these plans provide examples of how they are providing care for Medicaid covered beneficiaries and how their innovations are supporting care coordination for those with co-morbid physical and behavioral health conditions. The core research questions also guided this study. Case studies include an overview of each plan; challenges identified by the plan for coordinating care for physical and behavioral health; the types of innovation and their descriptions; and measures of the impact of the innovative care coordination programs. The details provided for each of the case studies are influenced by the nature of the innovations, the specificity of the program's design, and the tracking of their outcomes.



AmeriHealth Caritas Pennsylvania

Plan Overview

AmeriHealth Caritas operates in 15 states and the District of Columbia and serves more than 4.8 million Medicaid, Medicare, and Children's Health Insurance Program members through its integrated managed care products, pharmaceutical benefit management services, behavioral health services, and other administrative services. Headquartered in Philadelphia, AmeriHealth Caritas is a mission-driven company with 30 years of experience serving low-income and chronically ill populations. AmeriHealth Caritas also has a wholly owned behavioral health management subsidiary, PerformCare, which the company partners with for Medicaid plans when state contracts allow. The company is owned by Independence Blue Cross and Blue Cross Blue Shield of Michigan, and is for profit but not publicly traded.

Challenges Identified for Coordinating Care for Physical and Behavioral Health

AmeriHealth Caritas has identified a number of challenges in serving Medicaid populations. The company notes that the transient nature of Medicaid covered populations sometimes makes it difficult to locate program participants. This places added importance on obtaining the most recent and valid contact information for participants. Further, member engagement is dependent on establishing community partners, collaborating with essential community providers, and completing health risk assessments and in-home evaluations.

The availability of covered beneficiary or health plan member data and the ability to link the data to care coordination resources is noted as very important for tracking and monitoring outcomes. Identifying and stratifying the health risks of members requires balancing statistical significance and applying practical utility to available data sources. Also, because each state's Medicaid program is unique, it is difficult to develop a universal data analytics system that can be applied across all health plan members in every location.

State policy is not always aligned with integrated care, causing separations between physical and behavioral health management, and this sometimes restricts innovation. Some states provide the plan with specific algorithms to use for the identification of the population they are to target for care coordination. Other states may just identify target health status characteristics, such as key chronic health conditions. AmeriHealth Caritas notes that it is important for the plan to reach a clear level of understanding of the state's expectations in each market.

Innovation Type and Description

Innovations in Data Analytics, Care Coordination, and Managed Care Infrastructure

AmeriHealth Caritas is innovating in three areas: data analytics; coordination of care; and managed care infrastructure.

AmeriHealth Caritas has placed importance on using data analytics to best identify and stratify members who will benefit most from its integrated program. The company is in the beginning stages of developing, testing, and refining its custom algorithms. The company uses data to support its programs and help influence the direction of state policy. The advance analytics team partners with each strategic initiative project lead to ensure timely and accurate measurement of the quantitative impact of each program. Together they develop project tracking reports to measure the program's progress and the statistical assessments that measure its impact.

A pilot predictive modeling project has been developed to identify current members with a strong likelihood of future high physical and behavioral health needs. Decision tree logic models and other methods are used to identify high-risk members, including predictive risk scores, health risk assessments, and referrals from providers. Another model in development is focusing on identifying super-utilizers (the top 5 percent of members responsible for 50 percent of overall claim costs).

AmeriHealth Caritas uses a four-quadrant model (high/low behavioral acuity and high/low physical acuity) to identify the population it wants to serve. This program focuses primarily on: (1) populations in quadrant II (high behavioral health and low physical health needs) to provide specialty behavioral health programs integrated with physical health co-morbidity medical management programs; and (2) populations in quadrant IV (high behavioral health and high physical health needs) to provide fully integrated physical health and behavioral health medical management programs. High-cost medical cases are also followed for their individual needs. The program uses a multidisciplinary care management team to coordinate care for members.

For the coordination of care, AmeriHealth Caritas uses an innovative hub model that embeds staff in the community and recruits members most in need of care. The program uses "community care connectors" who are familiar with the local resources and can interact well with members; these connectors are knowledgeable about locally available social services, share language and life experiences of the members served, and help members sort through complex and competing needs, many of which are not just medical. Community care connectors are provided internal training on care coordination and health-related competencies and also work closely with the care coordinator to ensure that the member is getting the services he/she needs.

The community care management program is being piloted in four communities that have concentrated numbers of high-risk members, as demonstrated by high-cost utilization and unmet social and health care needs. The program relies on a community-based care management team that includes physician oversight, professional care managers, registered nurses, social workers, a community care connector, and an office coordinator. AmeriHealth Caritas recognizes that members' involvement is essential to their recovery. The integration program ensures that members and their families (if requested) are involved in developing and carrying out the care plan.

Finally, for its system of care program in New Jersey, AmeriHealth Caritas has developed a state-of-the-art health management information system to coordinate care for the children it serves, offering a single point of access and care coordination for children, youth, and young adults up to age 21. Through AmeriHealth Caritas's coordinated information system, multiple medical, community, and social service provider organizations are able to share data. This approach promotes better care planning and coordination and has almost eliminated the use of out-of-state behavioral health residential placements and decreased the use of in-state residential care. In addition, a high percentage of youth being discharged from residential treatment in the community are remaining in the community. The information system platform allows the full continuum of medical and social service providers, care coordinators, and members/families to enter their care planning information into one care coordination system. This resource is separate from each agency's individual case records and allows community-based agencies to share care planning information and avoid duplicate or unnecessary services.

Measuring Effects and Using Data

It is difficult for AmeriHealth Caritas to measure outcomes across its various Medicaid programs, as designated populations and systems of care vary from state to state. In some states behavioral health benefits are carved in while in others they are carved out. Therefore, there are no markets in which AmeriHealth Caritas serves identical populations, covers the same benefits, or has consistent contractual requirements.

As an example of how state populations differ, in Indiana, behavioral health is carved in, and the covered population is nearly 100 percent eligible for Temporary Assistance for Needy Families. Therefore, the percentage of individuals with serious mental illness (SMI) is less than it is in other states such as Louisiana, Pennsylvania, or the District of Columbia, where behavioral health benefits may or may not be carved in. The aged, blind, and disabled population in Pennsylvania and South Carolina is 30 percent; however, one state carves out behavioral health management and the other does not. This makes it difficult to measure outcomes uniformly across the company's various state markets.

AmeriHealth Caritas measures outcomes both in the short term and in the long term. Short-term measurement involves studies that track data to evaluate the impact of engagement in care management for the first 3 months. Short-term data analysis is used when there is high beneficiary turnover or insufficient longitudinal data to measure statistically significant outcomes. Data are observed at 30, 60, and 90 days after intervention, with the recognition that each member may engage at different points in time. Despite this, the data are aligned for each member to observe possible trends. This approach can be useful for creating a timeline and tracking interventions by date, including monthly tracking reports that observe per member per month (PMPM) utilization and cost.

Long-term evaluation is possible when there are ample data and the plan is able to conduct a controlled statistical test. For example, the plan will obtain pre-measures and post-measures, comparing intervention group utilization against a propensity matched control group and conducting tests of significance. Depending on the state and program, the plan collects data beyond 12 months when it is able to, but the data have the possibility of rapidly becoming limited.

Internal and external reporting is done through the plan's quality improvement process. Reporting requirements vary across the state contracts. However, the majority of the reporting is generally described as tracking and trending reports. None of the states require reports that involve significant statistical tests or comparison groups. Reporting requirements generally are related to demographic information such as the number of behavioral health case managers, number of total case manager contacts, and number of members engaged. Overall, the states served are not requiring sophisticated statistical reports, and there is a significant amount of variation in what metrics each state requires. Most states require standard Healthcare Effectiveness Data and Information Set (HEDIS) metrics and some standardized state metrics (e.g., number of people contacted or seen in case management).

Internally, the plan examines a range of data to evaluate the process and outcomes of care coordination activities. These include: PMPM utilization; evaluating services by claim type; numbers of claims; the average cost of these claims; the number of inpatient stays or days used; and similar types of service review data. The plan is also currently tracking specific initiatives such as pharmacy utilization. For example, the plan is reviewing anti-depressant usage and observing proportion of days covered prior to case management and post case management, and whether these services are having a statistically significant impact on medication adherence over time.

The next level of outcome evaluation for the plan includes data mining techniques to better understand the key drivers and predictors of successful care coordination and care outcomes. The plan is seeking to better understand and tailor its analytics program to be more effective in identifying populations in need of care coordination and assessing the outcomes of services provided. The plan is looking to advance the current analysis from retrospective review to prospective or predictive modeling.

It should also be noted that the corporate leadership culture of AmeriHealth Caritas plays an important role in encouraging and promoting innovation efforts of its staff. The company cites this as an organizational commitment to continually improve programs and services for the populations it serves.

BlueCare Tennessee

Plan Overview

BlueCare Tennessee is an independent licensee of the BlueCross BlueShield Association. Founded in 1993, the Chattanooga-based company focuses on managing care and providing quality health care products, services, and information for government programs. The plan is currently serving the western and eastern regions of the state, but effective January 1, 2105, the plan is providing care to the entire State of Tennessee. The two regions the plan serves are very diverse in terms of topography, ethnicity, income, and common physical ailments. Due to this variety, the way the plan approaches members varies by region. In eastern Tennessee, which is characterized by an Appalachian cultural group and topography that features numerous hills, hollows, and foothills, there is a large methamphetamine problem. In western Tennessee, which includes Memphis, the landscape is relatively flat, and beneficiaries have more eclectic health care needs. Plan representatives also note that the mental health issues seem to be similar and do not appear to vary by region. BlueCare Tennessee also offers TennCare select, which is a MCO for foster children, children receiving Supplemental Security Income, and children under 21 in a nursing facility or intermediate care facility for individuals with mental retardation.

Challenges Identified for Coordinating Care for Physical and Behavioral Health

BlueCare Tennessee notes that the challenges for coordinating care for individuals with physical and behavioral health conditions is difficult and requires collaboration between the plan and community providers and facilities. Initially, BlueCare Tennessee's own health plan systems were separated, and case managers were experts either in behavioral health or in physical health. Behavioral health was carved out by the plan to an external vendor, and this contributed to the lack of coordination across physical and behavioral health. Recently, the company has decided to incorporate behavioral health management back into the plan, and this is in progress. The plan has recognized that effective care coordination requires the integration of physical and behavioral health and analytics that can identify individuals at greatest risk for poor health outcomes. BlueCare Tennessee has introduced the integrated case management (ICM) approach from the Care Management Society of America, and this has helped to re-engage care coordinators in an integrated approach.

Innovation Type and Description

Data Analytics to Support Care Coordination

BlueCare Tennessee has built its population health management program around data provided by its predictive analytics team. The population health management model is built around data derived from its "custom 360 platform." The predictive analytics team is able to merge internal data, which includes administrative medical claims, pharmacy claims, episode treatment groups, risk data, and health risk assessment data, with external member-specific data purchased from Experian, a third party, personal data collection and credit agency. Some of the external data elements include expanded demographics information such as Census data on median income and wealth ratings; housing and real estate information; household members; life events data, including new parent status or housing relocation; automotive ownership; summarized credit data; lifestyle profiles, including hobbies, smoking, and investment preferences; and, transactional data that includes recent catalog, retail, and other web-based purchases. These data provide a 48-month history on members.

A microsegmentation approach provides lifestyle and clinical clusters divided into five categories for each area. As the lifestyle and clinical categories were developed, the plan was careful to adopt culturally appropriate labels. This approach provides a combined member profile that helps assign three levels of risk -- no risk, low/moderate-risk, and high-risk -- and customize care coordination activities and interventions. The plan notes that this customization is important, as the needs of two individuals with a chronic condition like diabetes will vary by lifestyle, clinical profile, and other demographic characteristics.

Based on the microsegmentation categories, BlueCare Tennessee can decide how to best reach out to these members. For instance, those in the "Value Seekers" microsegment had a low web presence and preferred traditional print media. This contrasts with the "Suburban Achievers," who are web savvy and prefer email communication. Armed with this information, BlueCare Tennessee is able tailor its messaging to members' preferences, improve health literacy, and in turn achieve a higher engagement rate.

BlueCare Tennessee is using the ICM system, an approach developed by the Case Management Society of America (CMSA). Case managers are trained not only on behavioral and physical health conditions but also on how to build a relationship with the member. The program's goal is to help the members in the "no risk" category stay healthy. Members in the "low/moderate-risk" category were encouraged to manage any health risk they might have, and those in the "high-risk" category received the most assistance to help them manage their complex health conditions. As a result of the microsegmentation program, the plan has demonstrated improved member engagement and satisfaction.

Measuring Effects and Using Data

BlueCare Tennessee continually strives to evaluate its care coordination and clinical outcomes. The company has a designated analytics unit that conducts analyses and reports findings internally to the quality improvement and other stakeholder committees. Because BlueCare Tennessee's approach to microsegmentation is new, the company is beginning to assess and evaluate how these programs affect clinical outcomes and utilization trends. Through its custom 360 profile, the company is able to compare claims and other information to determine how best to engage members and promote the necessary health services for improved outcomes, as well as compare the impact of member profiles with patient engagement outcomes. BlueCare Tennessee has piloted this with a range of health conditions, including ADHD, asthma, coronary artery disease (CAD), diabetes, hypertension, well child and immunization, and others. The result is improved engagement in care coordination programs.

The plan has also developed a clinical registry product, now being piloted with provider groups, that tracks individual cases and monitors care coordination progress. BlueCare Tennessee is working to supply providers with timely and actionable information that can support improved outcomes for beneficiaries with chronic health conditions.

At the state level, the plan has requirements for standardized reporting, largely based on established HEDIS measures. The plan and the state have been looking at the new care coordination approaches to better understand how these may inform routine reporting. In addition, since the plan uses the CMSA's ICM platform, there are some built-in metrics that are evaluated as part of this system.

Community Health Plan of Washington

Plan Overview

Community Health Plan of Washington (CHPW) is the only plan in the state founded by local community health centers. CHPW's statewide, comprehensive network of clinics often provide translation, transportation, dental, and mental health services in addition to primary, preventive care. CHPW combines highly personalized services and integrated care that treats the whole person, not just the isolated symptom. CHPW provides affordable comprehensive coverage to more than 300,000 individuals and families throughout the state. The plan began with a pilot in Pierce and King Counties that was originally designed to serve unemployed adults with short-term disability due to behavioral health problems. The program has now expanded statewide. In 2012, the program expanded to include disabled enrollees, and in 2013, it expanded to include the dual eligible (Medicare and Medicaid) population. The program now includes more than 100 community health centers and 30 community mental health centers.

Challenges Identified for Coordinating Care for Physical and Behavioral Health

CHPW was formed in 1992 by community health centers across Washington, which now total 21 member community health centers that govern CHPW and provide the majority of the primary care network across the state. CHPW is currently the sole not-for-profit health plan that serves the Medicaid population in Washington (four others are for profit). Due to the relationship with the community health centers, CHPW has been focused on advocating for a variety of coverage options for individuals within those safety-net clinics. Thus, CHPW has been focused on Medicaid expansion and benefits for some of the underserved populations.

CHPW has experienced some challenges with the mental health benefit because Washington has carved out services for its covered Medicaid population with SMI. The managed care plans have responsibility for all other behavioral health benefits. CHPW does not use one centralized vendor to identify individuals who may need behavioral health care services. The plan has established local care coordination within the primary care practice system and serves members with co-morbid physical and behavioral health conditions that are not covered by the state's SMI carve-out. In cases where specialty behavioral health care is not required for patients with SMI, the plan manages this care within its primary care network. CHPW has also negotiated a shared savings arrangement with the state.

Innovation Type and Description

Evidence-based Practice Improvement Utilizing the IMPACT Collaborative Care Model

The mental health integration program (MHIP) model is a stepped care treatment program that emphasizes integrated and evidence-based services provided mostly in primary care clinics and includes specialty mental health services when indicated. CHPW uses the Improving Mood - Promoting Access to Collaborative Treatment (IMPACT) model for depression management and follows the model's full evidence-based protocol, embedding care coordinators in its primary care practices. These practices are trained in team care and are involved in a practice transformation initiative. All patients are screened for depression (PHQ-9) and receive stepped care per the model. Psychiatric consultation is provided through a service contract with the University of Washington's Department of Psychiatry. Consultation is generally provided virtually to practices that use telemedicine technology, and patients are tracked and monitored for care coordination through an electronic registry developed by the University of Washington. Both the practice and the plan have access to the care registry platform.

The CHPW program emphasizes that care for depression and co-morbid physical health should be a dynamic process. An individual referred into specialty mental health care does not necessarily require this level of service indefinitely; once goals are met within this system, the individual would return to the primary care clinic. The program is "boots on the ground" in the delivery of community-based care coordination. It recognizes that the community understands the population best and has a more direct relationship than the plan's case managers.

The care coordination model that CHPW has developed calls for the creation of treatment teams that do not always exist in primary care and also requires providers to work in a manner that may be new to some. Further, since practice change is required to implement this model, CHPW supports organizational team building and additional training. CHPW works with the primary care practices' organizational leadership to review the core principles of the model to ensure the practices are comfortable with this new approach. The model also includes a checklist of the processes required by practices to implement this new type of work.

CHPW provides the necessary training to implement this new team care approach. Primary care providers (PCPs) only receive general information focused on how to efficiently utilize the new resources in the clinic. However, the behavioral health care managers require more extensive training, as this is often a new role for them. If a care manager's background is in mental health, he or she must learn to work differently in a primary care practice setting, and if the care manager's background is working in a traditional therapist role, he or she must learn to use the care registry and become more attuned to care and population management. Once training is complete, the team can commence services, and CHPW monitors ongoing progress.

CHPW has developed systems to identify high-risk individuals through plan data. The risk score is based on data received from the state as well as from internal claims. These data are used to identify individuals for the care coordination program. Additionally, CHPW uses several outreach and engagement strategies, including plan-based screening, where the Medicaid population is scanned using a CHPW-developed screening tool and prioritized for care coordination. Based upon these scores, individuals are referred to the MHIP.

CHPW's maintains a registry as a clinician's support and a reporting tool. The registry monitors scores on the PHQ-9 and other screeners, as well as the health progress of members, and informs care management and multiple providers. The registry also supports web-based consultation. The primary care practice behavioral health care coordinator is the primary user of the registry and is responsible for documenting and coordinating the information in the registry and the electronic health record (EHR). The consulting psychiatrist meets weekly with the care coordinator to review the caseload. The registry tracks who is improving over time and who is not as measured by the PHQ-9 score for depression or the GAD-7 for anxiety. Additional tracking elements are included in the registry and support the integration of health measures and behavioral health information. The care coordinator distributes the outcome of the consultation to the relevant members of the care team and PCP. Clinical care documentation occurs through the EHR.

Measuring Effects and Using Data

CHPW actively reviews its care coordination innovations. The results of these innovations helped turn the plan's two-county pilot into a statewide program, based on the findings of reduced inpatient admissions and reduced inpatient psychiatric costs. At a statewide level, CHPW reports that when the program expanded, it achieved hospital savings of over $11 million in the initial 14 months of the program, equating to savings of about $17 PMPM. Being able to demonstrate inpatient reductions and admissions in the two-county pilot program was critical to scaling the program statewide. Further, in the year following implementation, the average time to achieve the targeted level of improvement in depression scores was cut in half. There was also reduction in the variation across CHPW clinical sites for meeting clinical quality improvement aims.

Quality measures implemented in the second year of the program focused on process and clinical outcomes. These measures were developed with the support of the plan's steering committee, a cross-sector group of representatives from CHPW, Washington State, University of Washington, and primary care and specialty mental health providers. Measures were developed in response to the observed trends occurring in the clinical sites and focused on follow-up and psychiatric consultations for those patients who were not improving. These process measures are tied to reimbursement, and 25 percent of the annual funding given to the primary care practices is tied to providing the process measures and ultimately achieving the clinical outcome measures. CHPW incentivized follow-up and consultation with the consulting psychiatrist for patients who were not improving as measured by the PHQ-9 and the GAD-7.

CHPW uses a web-based registry that tracks and monitors outcomes in real time and over time. For example, CHPW monitors whether PHQ-9 scores are improving by viewing a dashboard in the registry, allowing for the plan to target individuals who need psychiatric consultation. A consulting psychiatrist enters his or her notes into the registry. From these data CHPW is able to monitor the percentage of a given care coordinator's caseload that is not improving and track the psychiatric consultation notes. Based on the quality reports generated from the registry, CHPW can set a threshold for achieving process measures and can promote continuous process improvement.

CHPW had some reporting requirements to the state in the contract for the two-county pilot. However, now that the program has been implemented, the state's current Medicaid contract is not specific to the MHIP program and is focused on standard HEDIS measures. The state is familiar with the outcomes used by CHPW through participation on the steering committee and through the plan's reporting on some clinical outcomes.

Over the past 4 years, CHPW has implemented a quality incentive program, in addition to the quality program, that focuses on tying payment to achievement of its priority HEDIS measures. The incentive program is in place with the CHPW network statewide but is not aligned with the state's prioritized measures, although there is some overlap. A separate quality council that includes internal CHPW staff and representatives from community health centers monitors these measures as part of the incentive program.

Hudson Health Plan

Plan Overview

Hudson Health Plan is a community-based not-for-profit health care organization that provides state-sponsored Medicaid managed care, Child Health Plus, and Family Health Plus insurance coverage to 120,000 members in New York's Hudson Valley. Hudson improves the health of its members and the communities in which they live through its innovations in care coordination and by supporting more than 5,000 local health providers. In both 2009 and 2010, Hudson achieved the highest quality incentive scores of any Medicaid plan in New York State, and in 2010, it received the highest overall performance rating of any Medicaid plan in the Hudson Valley. It also has earned the highest ratings in overall satisfaction among Medicaid managed care members in the Hudson Valley region every year since 2003. Hudson contracts with Beacon Health Strategies, a carve-out behavioral health care organization, for managed behavioral health care services.

Hudson Health Plan identified a number of challenges for the coordination of care for members with physical and behavioral health conditions. New York is moving the behavioral health benefit into managed care as part of its overall approach to transition from fee-for-service to population-based PMPM reimbursements tied to cost and quality outcomes. This integration is scheduled for January 1, 2015, in New York City and will include the remainder of the state in July 2015. The state is also launching a health home project with a goal of creating a primary care system based on this model.

The Beacon Health Strategies program uses its own proprietary care management platform. Beacon care managers are located in the same space as the Hudson care coordinators, and they work collaboratively. Hudson's care management nurses are able to examine the case manager's notes in the Beacon system but cannot enter information. Integrated care coordination rounds are conducted on site with the Beacon and Hudson care coordinators. This approach helps coordinate care across the physical and behavioral health services for members with chronic conditions.

Beacon does not have a role in Hudson's health home initiatives. Beacon's involvement is currently more centralized at the plan level, while care coordination in the health home program is localized. The health plan contracts with the health home directly and provides its own care coordination. When there is an inpatient behavioral health admission, the plan, Beacon, and the health home are involved in care coordination.

Innovation Type and Description

Co-location of Behavioral Health Carve Out Care Coordination Staff WithHealth Plan Care Coordinators

The Hudson Health Plan has established co-location of its care coordination activities with Beacon Health Strategies. Care coordination staff work side by side and collaborate on cases. They work primarily with two different information systems. Both staffs are trained in the CMSA tools and also use Beacon's proprietary systems.

Hudson and Beacon's closely knit collaboration of more than 14 years has produced a number of projects and innovations, including the Westchester Cares Action Program, Beacon's Integrated Partner Model, and other coordinated care initiatives. From the partnership, Beacon pioneered its Integrated Partner Model, a co-located model that has both clinical and administrative staff on site to coordinate care.

The Hudson and Beacon collaboration utilizes the CMSA ICM model. The ICM-Complexity Assessment Grid (CAG) tools cover the domains of physical health, mental health, psycho-social issues, and navigating the health care system. The ICM-CAG is also used for case stratification and helps organize care coordination priorities. This tool is the primary focus for all care planning activities.

The state and plan are expanding into health home options where care coordination will be financed by the state. The health homes and Hudson will work together to share information. Hudson has developed a new care coordination technology tool to help care coordinators' track information and case documentation. Currently, this platform has been rolled out to at least 16 agencies in Hudson Valley and the health homes in upstate New York. The platform allows care managers to be more efficient and to take on more cases than in the past. At the moment, there is no plan for Beacon to be involved in this new program.

Measuring Effects and Using Data

The results of the coordinated care management program are internally tracked and reported through both the Hudson and Beacon organizations. Beacon staff participate in the Hudson committees, and this supports integrated reporting. Outcomes are reported through tracking and trending utilization data, and specific projects are informed by the results of the interventions.

The Hudson Health Plan also uses the CMSA's ICMs. This approach supports standard reporting for care coordination activities and the tracking of utilization outcomes. The plan notes that as the state's health home initiative evolves, it will be developing new reporting systems and coordination across the plan, the health homes, and the state.

Rocky Mountain Health Plan

Plan Overview

Rocky Mountain Health Plans (RMHP) is an independent, not-for-profit MCO that has provided health insurance to Colorado for the past 35 years. RMHP is the only health plan provider in Colorado that serves every market segment, including employers, individuals, Medicare, Medicaid, and Child Health Plan Plus beneficiaries. RMHP serves the western part of Colorado.

Challenges Identified for Coordinating Care for Physical and Behavioral Health

RMHP has developed care coordination programs that are practice-based and funded through payment reform initiatives. RMHP reports that sustainability of the program is a significant challenge since many of the programs are grant funded. The concern is that after the grants end, the plan or the practices will be unable to support the program. RMHP hopes that if it can help the primary care practices build the infrastructure for care coordination, these provider systems will be able to keep the program operational.

RMHP has also cited a challenge with determining the appropriate payment allocation on a PMPM rate for care coordination. The plan has been working with Colorado's Department of Health Care Policy and Financing and actuaries to develop the appropriate rates based on a shared data set and overall projection. The company notes that as it compared its independent analyses, the rates calculated differed by less than 0.3 percent. RMHP is working diligently with the state to meet its implementation date.

Innovation Type and Description

Innovative Payment Reform

RMHP is currently involved in two pilot programs that support this innovation. These are: (1) Medicaid PRIME, funded by the State of Colorado under the Accountable Care Collaborative Payment Reform Initiative - H.B. 1281 Proposal; and (2) a payment reform pilot, Sustaining integrated Healthcare Across Primary care Efforts (SHAPE), which is funded by the Colorado Health Foundation with collaborative support from the Department of Family Medicine at the University of Colorado Denver and the Collaborative Family Healthcare Association. RMHP is working with primary care practices in its service area (western Colorado) to develop primary care practice initiatives that foster care coordination.

In the SHAPE program, RMHP is using this payment reform approach to evaluate the application of a global budget model for integrated behavioral health in primary care. The global payments are allocated based on each practice's cost, panel size, panel complexity, and program design. This demonstration allows practices to use the allocated global payments to hire and staff care coordination positions in their clinical sites. The providers interviewed for this study are using this funding to hire care coordination staff to help extend the scope of services provided. These staff are actively helping assess and address behavioral health needs of the Medicaid covered population.

As part of the demonstration program, the practices and RMHP share both the risk and the incentives through quality targets and improvement in certain patient outcomes. RMHP develops quarterly reports that show providers the claims data and expenses associated with their practices and the members they serve, and how they compare to other practices. RMHP has also provided information to these primary care practices about what is being tracked, what outcomes can be anticipated, and how savings are being achieved. RMHP is working hard to build a strong, comprehensive, and quality focused infrastructure for its providers. As a result, some providers are developing their own risk stratification systems to identify potential clients for their care coordinators as well as quality improvement processes to improve outcomes. Providers involved in the process reported appreciation for the plan's transparency and data sharing, noting that in the past they were not given this level of information and did not understand how to interpret it. The plan reports that this transparency has empowered providers to better understand their covered members and assume full responsibility for their engagement, outcome, and costs of services.

RMHP is also a member of the Colorado Beacon Consortium. This consortium focuses on strengthening the local health information technology infrastructure to support improvement in the quality and efficiency of health care and includes a community-wide health information exchange platform and provider education to foster adoption and effective use of health information technology.

The plan has also launched a health engagement team pilot that is an innovative emergency room and hospital diversion program using community health workers (CHWs). The plan has provided funding to the local mental health system to hire CHWs to work with the behavioral health system and primary care to identify and work intensively with beneficiaries to avoid unnecessary emergency department utilization. Preliminary findings support improved care outcomes and provider and member satisfaction.

Measuring Effects and Using Data

The RMHP outcomes and reporting is based on a range of factors linked to current innovations. Since RMHP is using financial incentives as part of its care coordination projects, it is important that there be sound reporting systems in place. Existing targets have been developed by the state for the plan, and by the plan for its providers. The plan's target medical loss ratio (MLR) is 93.5 percent, and if the plan is successfully below this, the plan retains all of the savings up to 85.5 percent MLR. In addition, the plan must satisfy four quality performance measures, including three HEDIS measures (determined by the state) and a patient activation measure (PAM). The PAM project is in development; in year 1 it promotes provider adoption and improvement of screening rates in the attributed population, and in year 2 it focuses on coaching for activation and examining the movement of PAM scores year to year.

For providers, the plan crosswalks each of the HEDIS measures with clinical quality measure (CQM) domains. Each practice has its own targeted approach. The lower functioning practices work with the plan's practice transformation advisors to set initial objectives. Other practices continue longitudinal reporting and progress toward the respective practice's specific targets. The plan uses a close-the-gap type method to analyze process outcomes, and many times the plan will take the HEDIS 90th percentile or the 75th percentile, as applicable, and set a target for the higher functioning practices.

As a shared savings program, there is a pool of savings at the end of the year following the state's assessment of whether the plan has met its HEDIS measures and PAM scores. The plan has licensed the PAM tool and made it available free to its providers, including coaching and training on its use. The goal for the first year is a penetration metric of at least 50 percent of the practices collecting PAM data and representing at least 50 percent of the attributed population. Use of the PAM tool is funded by grants and not the plan's MLR savings. The plan is speculating that this pilot program will be successful and will eventually be able to pay for PAM through its savings. Data collection has commenced, and the plan uses Insignia's reporting tools.

This risk pool is distributed among the practices following the plan's assessment of each practice's participation and pass/fail success in the project. These criteria include the plan's assessment of the volume of patients and the risk of each practice's patients. This assessment is then reported to its executive committee each quarter. The practices submit measures via the plan's web portal; the measures are then aggregated and "rolled up" into a region-wide CQM. The data are then broken out by practice, and the plan provides feedback and reporting of how each practice is doing. Administrative data are monitored throughout the year to determine whether or not the plan is meeting its target.

From a policy perspective, the plan has been able to build this program not on state Medicaid dollars but on its grants and other resources, including its own resources. The state is flexible concerning the plan's reporting process; in its contract with the state, the plan in the future will migrate from HEDIS measures to CQMs as a basis for the shared savings calculation. Some of the challenges that are anticipated include the ability to move away from HEDIS measures and to work with plan-developed measures. In addition, fostering innovation in payment incentive programs will require flexibility to adjust current financial models that are pegged to adjudicated fee-for-service encounters to population health outcomes.

Cenpatico/Sunshine State Health Plan

Plan Overview

Cenpatico began managing behavioral health benefits in Florida in 2009 with its sister company Centene as the Sunshine State Health Plan. Together they serve about 200,000 Medicaid recipients in multiple counties across Florida. Headquartered in Sunrise, Florida, the Sunshine State Health Plan provides health, behavioral, vision, dental, and pharmacy services to members in both urban and rural areas who might not otherwise have access to quality health care.

Challenges Identified for Coordinating Care for Physical and Behavioral Health

Cenpatico has identified a number of challenges for the coordination of care for physical and behavioral health. This includes covered beneficiaries who transition to Medicaid coverage without a complete medical history available, and they can be challenging to locate and identify for appropriate care management programs. Members frequently have gaps in their enrollment periods and are difficult to track because the MCO does not have access to accurate demographic data from the state. Historically, telephonic outreach has proved ineffective and yielded limited engagement rates.

Cenpatico provides outreach and care coordination for its members, and there is occasionally a fine line between care coordination and treatment. Cenpatico Choose Health coaches conduct needs assessments, provide education, and link and coordinate services for the members they serve. They do not provide a diagnosis or psychotherapy and do not dictate treatment. This program is still in the beginning stages, but the initial member response to face-to-face intervention by the Choose Health coach has been positive.

Innovation Type and Description

Expanded Care Management Outreach Efforts

Cenpatico currently uses a telephonic health coaching approach that is loosely based on the IMPACT stepped collaborative care depression care model. Health coaches provide telephonic outreach to members identified via internal health risk screening, Sunshine State Health Plan referrals, and predictive modeling. The Choose Health program utilizes key IMPACT program components such as systemic use of depression symptom scales, behavioral activation, psychiatric consultation as needed, and relapse prevention.

Cenpatico's Choose Health coaches support and collaborate with primary care physicians to ensure that members receive the most effective and efficient resources. This program offers technical assistance to targeted primary care physicians on stepped care and the IMPACT tenet of treating to goal. To the extent they are able, the health coaches work to improve the engagement and follow-through with depression care, including those with co-morbid physical and behavioral health conditions.

The plan has currently hired one health coach and soon will be employing a team of health coaches and embedding them in advanced primary care practices. Health coaches also continue to do some telephonic follow-up with members referred internally. In more sophisticated primary care practices, Chose Health coaches provide on-site care coordination. Patients are administered the PHQ-9 screening and followed for depression treatment and care outcomes. The Choose Health coach consults with the Cenpatico psychiatric medical director and together reviews the Florida Best Practice Psychotherapeutic Guidelines for Adults on cases that have not responded as expected to depression treatment. The Choose Health coach provides feedback to the primary care practice for treatment plan adjustment consideration. This program is currently implemented at the Family Care Partners Federally Qualified Health Center (FQHC) clinic sites that proactively screen members for depression using the PHQ-9. Once members are identified, they receive the program information from the PCP and are scheduled for a face-to-face visit in the PCP's office with a Choose Health coach for an initial assessment. The Choose Health coach administers the PHQ-9 every month. If the score has not improved by at least 50 percent after 8-12 weeks, the Choose Health coach consults the Cenpatico psychiatric medical director and together reviews the Florida Best Practice Psychotherapeutic Guidelines for Adults.

The Choose Health predictive modeling technology is used proactively to profile the primary care practice and its covered patients for health risk needs and care coordination opportunities. Centene (Cenpatico's parent company) has developed its own health information care management system that supports common care coordination and utilization management records documentation. Member profiles have been developed so that care managers can track the course of treatment and outcomes of care and share this information with physicians.

Measuring Effects and Using Data

The Choose Health program uses a modified version of the IMPACT model. This is intended to improve the outcomes of care for beneficiaries who are treated for depression. This model has a built-in outcomes assessment system based on individual improvements on the PHQ-9 scores. These are monitored by the care coordination plan and tracked and reported on an individual basis.

Because the innovations in care coordination are relatively new, the plan currently is only able to track and trend individual improvements, but its goal is to expand this tracking and evaluate the successful outcomes of different primary care settings and providers. This evaluation will enable the plan to also monitor utilization of high-cost members and services and use this information to develop provider-focused system improvement initiatives.



Among the plans reviewed in this study, six broad strategies for improving the coordination of care for Medicaid covered beneficiaries were evident. These include: (1) using information technology to identify at-risk members and stratify their needs for care coordination; (2) supporting practice-based change for improved care coordination; (3) using financial incentives and payment reform to support enhanced care coordination; (4) implementing information technology that supports care management through enhanced care planning and shared clinical practice coordination; (5) reaching out to and engaging with covered beneficiaries in their communities; and (6) coordinating physical and behavioral health care management services within plan operations. Some of the plans reviewed had multiple innovations in different categories, and in some cases the innovations reported had overlapping characteristics. A fuller description of the strategies used by the plans in this study to implement these innovations is provided below.

1. Predictive modeling and the use of information technology to identify at-risk members and stratify their needs for care coordination.

Predictive modeling using a four-quadrant model (high/low health, high/low behavioral health) identifies members who are likely to be super-utilizers and can help direct care coordination activities and stratify risk for physical and behavioral health conditions. The triggers for behavioral health stratification include inpatient admissions for behavioral health conditions in the past 6 months, newly diagnosed behavioral health conditions, recent suicide attempt or drug overdose, and care management reported information, including unstable or disruptive behaviors. Physical health triggers include recent inpatient or emergency room visits for target conditions, including diabetes, asthma, COPD, CAD, congestive heart failure, hypertension, and cancer, and other unstable health conditions.

Plans use different approaches for the identification of at-risk Medicaid beneficiaries. They note that it is important that identification models balance statistical versus practical significance. Often plans have contractual obligations with the state for outreach and engagement with newly enrolled members. This outreach approach incorporates a health risk appraisal to determine health care needs and establish a welcoming connection between the plan and new members. However, plans reported that while the information gathered in this process can be informative and helpful in promoting engagement, it is generally not sufficient for population health management. Several examples illustrate how the different approaches that are used by plans fit into their overall care management strategies:

  • Using data analytics can help inform care coordination activities. The identification of Medicaid beneficiaries who are at higher risk of poor health outcomes is a key priority for the plans studied. Plans report that they receive demographic population and historical claims data from the states; however, there was almost universal recognition that these data have severe limitations and are inadequate to identify high-risk populations. A common theme across plans studied was that these data present challenges associated with availability and completeness, the time lag of many data sources, including claims, and the need for long-term trends to build useful models. Additionally, they note that there are gaps in data when behavioral health and other services are carved out from the plan and reported through other sources.

  • Distributing a monthly data feed on the patients who are seen or covered by primary care practices in the plan network allows the practices to conduct their own analysis for quality improvement and care coordination. The data feed, which can be presented in a database format, includes standard utilization data, risk adjustment profiles based on established external models, and benchmarks for effective practice outcomes. This strategy encourages a collaborative partnership between the plan and the providers and supports the effective engagement and coordination of care at the practice site.

2. Health plan activities that support practice-based change for improved care coordination.

Health plans work with practices to promote team-based care and practice change, including implementation of the IMPACT collaborative care model of care for depression in primary care practices. In this model, a patient's primary care physician works with an established treatment plan that is developed in conjunction with an on-site care manager. Outcomes are measured throughout treatment using the PHQ-9 assessment tool; treatment is based on an evidence-based algorithm with the goal of achieving a 50 percent reduction in symptoms within 10-12 weeks. The PCP and the care manager consult with a psychiatrist as needed when progress does not meet the established goal. A coordinated care approach is used to adjust treatment intensity, such as increasing medication doses as needed, or to add additional services such as psychotherapy and other specialty care, as recommended by the consulting psychiatrist.

As patients improve and goals are met, patients are able to continue their care in the primary care system. Plans in this study used the following activities to support this practice change:

  • Providing necessary training to the primary care practices.

  • Using a practice improvement checklist focused on how to implement the model and build necessary resources.

  • Using a telephone-based care coordination program that is conceptually based on the collaborative care model.

  • Using in-person care coordination and on-site health coaches to work within one of the more established FQHC practices and support the stepped care model of care coordination.

  • Providing psychiatric consultation administratively to the PCPs by a health plan psychiatrist, including through telemedicine.

3. Financial incentives to support enhanced care coordination.

Providing funds administratively through global payments to primary care practices enables the practices to employ care coordinators and support other practice improvement activities. These funds can support new capacity to promote outreach, engagement, and follow-up for their patients in a way never before possible.

Additionally, the global payment funds can be used to develop quality improvement programs and support data analytics on service utilization and cost of care outcomes in practices as well enhance improvements in EHRs.

4. Information technology solutions that support care management through enhanced care planning and shared clinical practice coordination.

A comprehensive care coordination software program allows all providers and caregivers, including families, to enter, monitor, and amend care plan activities and services. Information technology solutions also include linking care coordination activities among providers, organizations, and the health plan, taking into account that Medicaid covered populations can be difficult to locate and engage. This is particularly important when there are multiple physical and behavioral health conditions and a full range of services and providers involved in care.

A care registry can support tracking and reporting care coordination tasks, activities, and progress for the patients seen. With the necessary training, providers are able to use the registry's data for quality improvement and their own local analytics. This strategy highlights the importance of shared information, analytics, and technology systems capable of promoting coordination at the site of care.

5. Coordinated care initiatives that engage covered beneficiaries in the community.

Plans are able to improve outreach and engagement by embedding in the community staff who are well aware of resources and the social and cultural aspects of the communities in which they work. These staff can meet with covered members in homes, care facilities, restaurants, and other locations and work closely with the plan's care coordinators and providers to build effective communications and plans for the members they serve. Peer support specialists can be good candidates for these roles.

Another initiative is to provide funding to primary care physicians to hire practice-based care coordinators, which promotes the full transition of care coordination to the site of care; these staff report to the practice and not the plan. Additionally, funding to the community mental health system to hire a contingent of CHWs can support the goal of reducing unnecessary emergency room and hospital use.

6. Coordinated physical and behavioral health care management services within health plan operations.

In plans with carve-outs for behavioral health care, co-locating staff from each plan on site helps coordinate care for members with physical and behavioral health conditions. Different funding arrangements and responsibilities for the oversight of care require enhanced coordination among the health plan organizations responsible for these services.

Shared technologies are also required to coordinate care management within health plans that have multiple organizational resources for members with physical and behavioral health conditions.



Based on this review, Medicaid managed care plans are actively developing new programs and resources to improve care outcomes for the members they serve. From the plans included in this study, some principal findings emerge.

Information technology is a key tool for coordination of care by the Medicaid managed care plans studied. A range of approaches were identified: identification and stratification tools for at-risk members and population health; data analytics that are available to provider systems for custom analysis of members; and shared platforms for developing care planning, documenting and tracking care coordination, and monitoring service utilization. Innovative solutions are being developed that include customized population health profiles using new and non-traditional data sources.

By supporting the use of data analytic tools in their primary care practices, some plans are enabling the practices to conduct self-analysis and tracking of their high-risk patient populations. This increased transparency of health information supports new and expanded relationships between providers and health plans. It also fosters funding innovations that support the primary care practice in participating in risk-based payment models and assuming greater population health management roles.

Some of the practices associated with the plans included in this study have developed clinical registries to better assimilate and share care coordination information between plans and providers. However, access to these resources is limited to the practice and plan-based care coordination staff. This information is not generally accessible to other providers, hospital facilities, and community support organizations that frequently have important information related to care planning and coordination activities. Further, when behavioral health services are provided in different systems, the sharing of coordination information is more complicated.

The disconnect between health plan and provider system technology resources has led Medicaid health plans to develop innovative ways to share health information and data between payers, primary care and specialty providers, general and psychiatric hospital facilities, pharmacy and lab services, and others. While Medicaid and health plans ultimately may have access to this information, the timeliness of access to these data makes it difficult to develop and share common care plans, coordinate services across levels of care, and track service utilization and care outcomes. When behavioral health services are provided outside of the continuum of care managed by the health plan and not associated with the primary care system, additional barriers for sharing information are also likely to exist. Medicaid contracts that fragment payment and provider arrangements for physical and behavioral health care make it difficult for primary care practices to fully invest in coordinated care.

Some new initiatives explore ways to support in-network primary care practices to be more successful in fulfilling their care coordination responsibilities. These initiatives include transferring care coordination activities to the sites of primary care practices; supporting the broad inclusion of other specialty providers, hospital facilities, and community-based resources; and providing primary care practices the technical assistance needed to analyze data, inform outcomes, and improve the quality of their care. This transfer of the locus of care coordination activity is occurring through both health plan staff placed in practice settings and resources for practice staff to provide these services. Some health plans are also supporting care coordination services that are community-based and provide liaison between health care and other social services. Some health plans are recognizing that the most effective coordination resources are provided in close proximity to the members and their communities and also support effective use of information and analytics to improve care coordination.

Some health plans are investing in their primary care networks to support the development of technology resources, data analytic capacity, quality improvement infrastructure, and practice redesign. Other health plan investments in primary care include plan-provided training and practice development as well as direct financial resources for providers to invest in resources that specifically meet the needs of their practices.

Some of the health plans included in this study also cited the importance of an organizational culture that supports innovation. They noted that in order to respond to the challenges of coordinating physical and behavioral health care for their covered Medicaid beneficiaries, there needs to be a leadership commitment to new program development and innovation. This cannot be accomplished without thoughtful program development, ongoing quality improvement review, and continuous surveillance of and attention to opportunities for program growth and modification. Some Medicaid contracts persist in fragmenting payment and provider arrangements for physical and behavioral health care, making it difficult for plans and providers to fully invest in coordinated care.

This study features some Medicaid managed care plans that are actively developing innovative solutions for the coordination of care for their covered beneficiaries with combined chronic physical illnesses and behavioral health conditions. Six health plans were reviewed and a variety of care coordination strategies were identified. Many of the innovations had common attributes across the different health plans and were influenced by state Medicaid contracts. The effective coordination of care for Medicaid covered beneficiaries with chronic physical illnesses and behavioral health conditions requires an integrated approach involving state Medicaid authorities, health plans, provider systems, and recipients of care. Therefore, there are a range of responsibilities that must be integrated across the multiple stakeholders and systems that provide coverage and care for this population. This is an evolving process, and continued attention to innovations and implementation of best practices are required.



America's Health Insurance Plans, Center for Policy and Research. (2013). Innovations in Medicaid Managed Care: Highlights of Health Plans' Programs to Improve the Health and Well-Being of Medicaid Beneficiaries. From

Association for Behavioral Health and Wellness. (2013). Peer Support Services A Valuable Component of Behavioral Healthcare. From

McDonald, K.M., Sundaram, V., Bravata, D.M., Lewis, R., Lin, N., Kraft, S.A., McKinnon, M., Paguntalan, H., & Owens, D.K. (2007). Definitions of Care Coordination and Related Terms. In Closing the Quality Gap: A Critical Analysis of Quality Improvement Strategies (Vol. 7: Care coordination). (Technical Reviews, No. 9.7). Rockville, MD: Agency for Healthcare Research and Quality. Retrieved from

Miller, B., Paschall, C.B., & Svendsen, D. (2006). Mortality and Medical Comorbidity among Patients with Serious Mental Illness. Psychiatric Services, 57(10), 1482-1487.

Smith, V., Gifford, K., Ellis, E., Rudowitz, R., & Snyder, L. (2013). Medicaid in a Historic Time of Transformation: Results from a 50-state Medicaid Budget Survey for State Fiscal Years 2013 and 2014. Retrieved from the Kaiser Family Foundation website:

Innovative Medicaid Managed Care Coordination Programs for Co-morbid Behavioral Health and Chronic Physical Health Conditions

This report was prepared under contract #HHSP23320100026WI between the U.S. Department of Health and Human Services (HHS), Office of Disability, Aging and Long-Term Care Policy (DALTCP) and Westat. For additional information about this subject, you can visit the DALTCP home page at or contact the ASPE Project Officer, Joel Dubenitz, at HHS/ASPE/DALTCP, Room 424E, H.H. Humphrey Building, 200 Independence Avenue, S.W., Washington, D.C. 20201.  His e-mail addresses is:

Reports Available

Innovative Medicaid Managed Care Coordination Programs for Co-morbid Behavioral Health and Chronic Physical Health Conditions: Final Report (May 2015)

Innovative Medicaid Managed Care Coordination Programs for Co-morbid Behavioral Health and Chronic Physical Health Conditions: Final Environmental Scan (March 2014)