The appendix provides a brief description of the 13 States participating in the initial Learning Network, highlighting many of their successes and lessons learned.
These States are among the leaders in their field of care management, and through their openness and willingness to share lessons learned and productive failures, they have provided the foundation for the material discussed throughout this Guide. While each State developed its care management program to match its unique needs, through the collaboration within the Learning Network, each State has incorporated significant program improvements.
This appendix outlines each State's strategies and lessons learned regarding:
This Guide and appendix reflect current programs and trends. As care management programs evolve to meet the changing needs of their populations, States will continually modify their programs to ensure that they are effectively impacting their populations. This appendix conveys the experiences of States to date. Future editions of the Guide are expected to communicate States' experiences as they implement new program models.
Arkansas: Antenatal and Neonatal Guidelines, Education and Learning System (ANGELS)
Arkansas began its ANGELS program in February 2002. The program currently targets pregnant women in primary care case management (PCCM) and fee-for-service (FFS) with a focus on high-risk obstetrics and neonatology.
Arkansas' Lessons Learned: Gaining Stakeholder Support
The Medicaid agency began planning ANGELS in partnership with the University of Arkansas for Medical Sciences (UAMS) after determining that when high-risk babies are born at UAMS, they tend to have fewer complications.
The Medicaid agency, in partnership with UAMS, designed the ANGELS program, which involves the State's obstetrical providers in developing best practices for high-risk cases and helps providers transfer pregnant women with extremely high-risk cases to UAMS before giving birth. The ANGELS program targets Temporary Assistance for Needy Families (TANF) mothers and babies in the FFS and PCCM programs.
To develop and share clinical guidelines, Arkansas hosts a weekly teleconference focusing on high-risk obstetrics for which physicians may receive Continuing Medical Education credit for participation. An average of 20 to 40 physicians participates each week at 20 teleconference sites.
The ANGELS program offers a call center for physicians and patients, transportation for pregnant women, and physician guidelines.
- Call Center. The ANGELS call center operates 24 hours a day, 7 days a week, for physicians and patients. Primary care providers can consult with specialists regarding patient management issues, and patients can find support related to their pregnancy, labor, and delivery, as well as postpartum care. Additional call center functions include maternal-fetal medicine consults, transportation arrangements, continuing education, advanced practice nurse consultants, telemedicine consults, patent education, patient referrals to community supports, and follow-up calls.
- Transportation. The ANGELS program arranges transportation to UAMS for women with particularly high-risk obstetric cases. The call center coordinates the transportation and can arrange for ambulance pickup or a helicopter for more serious cases. While a patient remains at UAMS for care, her local physicians receive regular reports from UAMS specialists.
- Guidelines. More than 80 finalized evidence-based guidelines have been written with physicians. ANGELS staff work with a group of physicians to adapt existing national guidelines to meet Arkansas' specific needs, especially around issues of cost, time, research, and clinical expertise.
- Provider activation. State staff employ a variety of strategies to engage providers in the ANGELS program. They circulate guidelines, for example, to help providers recognize and treat symptoms and conditions such as postpartum depression. In addition, they work with the Arkansas Foundation for Medical Care to market directly to providers and meet with neonatologists to discuss Level III neonatal intensive care unit (NICU) admissions.
Arkansas is contracting with the University of Alabama at Birmingham to conduct an evaluation using Medicaid claims data. In addition, Arkansas Medicaid's External Quality Review Organization, Arkansas Foundation for Medical Care, conducts the Healthcare Effectiveness Data and Information Set (HEDIS) and Consumer Assessment of Healthcare Providers and Systems (CAHPS®) surveys for Arkansas. The program has two years of data for high-risk pregnancy and neonatology. Finally, UAMS is working with birth certificate data and Medicaid claims data to analyze ANGELS program effectiveness on decreased NICU admissions and complications.
Arkansas Medicaid Web site: https://www.medicaid.state.ar.us/.
Illinois: Disease Management Program
Illinois' Disease Management Program started in November 2006, focusing on three populations: aged, blind, and disabled (ABD) members; persistent asthmatics; and emergency room (ER) users who have visited the ER more than six times in the last fiscal year without a hospital admission. The program covers all conditions of the eligible populations, with special emphasis on five disease categories for the ABD population: asthma, diabetes, coronary artery disease (CAD), congestive heart failure (CHF), and chronic obstructive pulmonary disease (COPD). Within Illinois' populations, the program excludes dual eligibles, home- and community-based waiver clients, and members enrolled in Medicaid managed care.
In-Person Care Management
Illinois' vendor has divided the State into 24 catchment areas staffed with 170 "feet-on-the-street" workers. These staff members can be lay health workers, social workers, or nurses.
The vendor also will place nurses in 10 high-volume hospitals and lay health workers in 10-12 high-volume clinics to assist with discharge planning, program outreach, and followup.
The goals of Illinois' disease management program are to improve health outcomes, decrease inappropriate use, and reduce costs. The program was implemented with an understanding that disease management can improve health outcomes for members with chronic conditions.
Illinois contracted with a vendor to implement its disease management program. Illinois' vendor assumes responsibility for improving cost and quality outcomes for all members who fit into the three population categories. Eighty percent of the vendor's risk is based on financial savings across all populations, and 20 percent is based on improvement in clinical indicators in the five targeted diseases.
Illinois and its vendor are working together to engage providers. Building on a long history of working with providers, the State has reached out to physician organizations, nurse organizations, behavioral health providers, and their sister health agencies. The program's Medical Director is the most recent past president of the State pediatric society and is associated with the family practice and medical associations in Illinois. Physicians will receive no additional payments for participating in the disease management program.
All members receive an introductory letter and educational materials. Illinois' vendor then performs targeted case management, including assessments and action plans for higher-need members. Moderate-level members receive quarterly contact, and high-level members receive telephone calls and in-person visits.
The State is currently implementing its program evaluation strategy, which includes establishing the baseline for the financial and performance indicators.
Illinois Medicaid Web site: http://www.hfs.illinois.gov/dm
Indiana: Indiana Chronic Disease Management Program
Indiana established the Indiana Chronic Disease Management Program (ICDMP) in 2003 for the State's primary care case management population. The program members were primarily in the ABD aid category. Diseases covered included asthma, diabetes, CHF, cardiovascular disease, and chronic kidney disease. Indiana "assembled" its program by partnering with local vendors to provide services, including a call center, nurse care management, and program evaluation.
Building on experience from the original program and successes in other States, Indiana created a new program, Care Select, for its ABD population. In November 2007, Indiana signed contracts with two vendors to provide medical homes, utilization management, prior authorization, and care management services as appropriate to approximately 70,000 members. Indiana seeks to provide comprehensive care coordination to this previously unmanaged population.
In November 2002, as part of the early conceptualization for ICDMP, Indiana's Medicaid Director, the Indiana Department of Health Director, and two State legislators attended a National Governors Association Policy Academy on disease management and met with Dr. Ed Wagner, who is the director of the Macoll Institute for Healthcare Innovation and lead developer of the Chronic Care Model. The Policy Academy provided the necessary impetus for program development, including legislative buy-in.
Indiana decided to assemble its ICDMP program by partnering with local agencies to provide chronic disease management services. The main components of the program are as follows:
- Program Management. Medicaid and the Department of Health jointly assume responsibility for managing the program.
- Primary Care. Members are assigned a primary care provider who serves as the focal point of patient care.
- Care Management. Members have access to care management via in-person nurse care managers or the call center based on stratification.
- Patient Data Registry. An electronic data registry is available to Medicaid providers and care managers.
- Measurement and Evaluation. Indiana conducted a statewide evaluation and a randomized controlled trial.
Indiana also partnered with AmeriChoice for the call center, the Indiana Primary Health Care Association for nurse care managers, and the Regenstrief Institute for help with member stratification and program evaluation.
Approximately 30,000 members have received disease management services through ICDMP. Eligible members, who are identified through Medicaid claims data, are stratified into either high-risk or low-risk groups, but they can move in and out of high-risk and low-risk management. Indiana developed its patient stratification methodology internally with assistance from the Regenstrief Institute. Factors that drive patient stratification include historical claims data, referrals (by patient, physician, or call center), new costs (e.g., hospitalization), and pharmacy utilization.
ICDMP interventions include a call center, care management, provider collaboratives, and provider toolkits. The call center monitors patient status and follows up based on established protocols. Call center staff assume responsibility for:
- Letters to patients and physicians.
- Outbound calls to assess, inform, and motivate.
- Patient education materials.
- Inbound calls.
The nurse care managers provide more intense followup and support to high-risk members. The care management intervention typically lasts 4 to 6 months. During this time, nurse care managers provide disease education and help patients set self-management goals. They also help foster the patient's relationship with his or her primary care provider. After 4 to 6 months of care management, patients "graduate" from the program and receive followup calls every 3 months from the call center.
Indiana's Lessons Learned: Evaluation
For providers, Indiana offers ongoing education, training, toolkits, and nurse care manager support. In addition, at the beginning of the program, Indiana conducted provider collaboratives, including three learning sessions followed by action periods that allowed for implementing new practices. Teams implemented practice-site improvements and reported results to the State.
The State legislature mandated a program evaluation when Indiana created the program. To ensure the study's legitimacy, the State decided to use an outside evaluator, the Regenstrief Institute. The study included a random control trial within the Indianapolis population and a time-series evaluation comparing patient care in different parts of the State. Data was collected from:
- Collaborative learning sessions.
- Medicaid administrative claims.
- Electronic medical records (Central Indiana only).
- Care management vendors.
Evaluation results found that the program reduced expenditures for patients with CHF but might increase expenditures modestly for diabetics.
Indiana Medicaid Web site: http://www.indianamedicaid.com/ihcp/index.asp
Indiana Care Select Web site: http://www.indianamedicaid.com/ihcp/HoosierHealthwise/rbmc_index.asp
Indiana Chronic Disease Management Program Web site: http://www.indianacdmprogram.com/
Iowa: Care Management Program
Iowa's care management program, established in July 2005, covers members with asthma, diabetes, and CHF, as well as high-utilization and high-cost members. The Iowa Foundation for Medical Care (IFMC) operates the program as part of a larger contract with Iowa Medicaid.
Electronic Medical Records
Iowa has developed an in-house online health information tool, the Iowa Electronic Medical Records System for physicians and hospitals. The system contains claims and pharmacy data updated weekly. Iowa worked closely with its provider community to test the system and gather feedback.
Iowa developed its program to provide optimal care to all Medicaid members. Program goals include:
- Improving access to care and eliminating unnecessary care.
- Increasing member involvement in care through self-management skills.
- Using community resources efficiently.
- Improving clinical outcomes.
- Saving program money.
Iowa's Medicaid program is operated through the Iowa Medicaid Enterprise, a collection of nine vendors that collaborate with the State to accomplish program goals. The vendors work in the same building with State staff and strive to provide Medicaid services seamlessly. As one of these vendors, IFMC runs the State's care management program. In 2003, before the creation of the Iowa Medicaid Enterprise, IFMC operated a State diabetes pilot program that provided lessons about operating and evaluating care management programs.
Iowa's asthma program launched in July 2005, its CHF program in October 2006, and its diabetes program in December 2006. The State chose to implement the asthma program first because staff believed asthma would render the best initial return on investment. The program is opt-in, which has made enrolling members difficult due to trouble contacting them.
Using claims data, Iowa identified 1,312 asthmatics with high costs and inappropriate use patterns for program outreach. The State attempted to reach this group through telephone calls but, after repeated attempts, had reached only one-third of the asthmatics. Iowa then sent letters to the identified group, but enrolled only 17 members through this method. Program staff realized that to enroll their target population of 250 members, they would have to open enrollment to the entire population of asthmatics. This decision led to their enrolling 266 members.
Iowa also has also worked to involve providers in the program, with engagement activities such as creating a Clinical Advisory Committee of physicians throughout the State and extending outreach to provider organizations. The standing Clinical Advisory Committee includes nine members who represent primary care providers throughout the State. Responsibilities of the Clinical Advisory Committee include the following:
- Assess member use of services.
- Assess new therapies and technologies.
- Review Medicaid policies and recommend changes.
- Support member and provider education.
- Promote preventive services to members and providers.
Although disease-specific interventions vary for asthma, diabetes, CHF, and high utilizers, all members receive telephonic care management and educational materials. Disease-specific interventions include providing peak-flow meters to members with asthma and providing Pharos (a telephonic reporting system) for members with CHF. Iowa is working with Des Moines University to provide the Pharos system. As part of the Pharos intervention, members call the system every morning and answer five questions about their CHF health status (e.g., weight, shortness of breath, swelling). If the member's answers indicate a need for further assistance, a care manager calls the member.
Iowa has completed an evaluation for the first year of its asthma program, including a description of the program and interventions, participating members' demographic data, program costs, and program outcomes. In addition, IFMC creates monthly reports on the care management program that cover enrollment, contact data, and information on services provided to specific patients. Iowa also measures pharmaceutical utilization, emergency room utilization, and inpatient admissions.
Iowa Medicaid Web site: http://www.ime.state.ia.us/
Iowa Diabetes Pilot Evaluations: http://www.ime.state.ia.us/ManagedCare/ManagedCareDocs.html
Kansas: Enhanced Care Management Program
Kansas began enrolling Medicaid PCCM members in its pilot care management program in March 2006. The program serves high-need PCCM members in Sedgwick County, the State's most populous county. Although the program currently has fewer than 200 members, it is expanding. Serving identified high-risk Medicaid beneficiaries, the program provides disease-specific management for asthma, diabetes, CHF, and other conditions.
Because of budget constraints, Kansas' program was nearly discontinued in August 2006. However, response from the community and local physicians created enough support to reverse the decision to cancel. Kansas staff would advise other States to seek and maintain a higher level of visibility for the program early to build a positive reputation within its own agency and the State.
Following a recommendation by the State legislature, the Kansas Medicaid agency decided to implement a care management pilot program as a cost containment and quality improvement measure. Before implementation, Kansas carefully reviewed care management options and data from its population. The State decided to focus on a care management program to develop an administrative program coordinating a broad range of services allowing for Medicaid beneficiaries' "whole health." Kansas chose to pilot the program in Sedgwick County because of its large concentration of patients, established PCCM program, strong legislative support, presence of a viable local vendor—Central Plains Regional Health Care Foundation (Central Plains), and a supportive medical society. The pilot was originally designed to operate for 5 years, but due to budget constraints it will operate for 2½ years.
Kansas partners with Central Plains to deliver care management to PCCM members who choose to participate in the Sedgwick County project. The vendor is a nonprofit organization that also manages Project Access to connect the uninsured with donated community health services. Central Plains' long-term relationship with providers and its connections to the Medical Society of Sedgwick County proved instrumental in helping the State secure provider buy-in for the program.
State staffing limitations compelled Kansas to partner with a vendor. Central Plains' enhanced care management staff includes four nurse care managers for approximately 200 members. Each nurse care manager has a maximum caseload of 60 members. To assist the nurses, the State also employs a disease management specialist nurse, whose maximum caseload is 150 members, and two community resource care managers with a social service background.
Kansas uses the Johns Hopkins Adjusted Clinical Groups Case-Mix predictive model to identify patients and stratify members for the care management program. After patients are identified, Central Plains contacts the potential enrollees for voluntary enrollment. Interventions, which vary, are based on member-identified needs, PCCM-identified needs, and utilization history.
Initially, Kansas faced difficulty recruiting members for the program. Eligible patients first are sent an invitation letter for the program, after which care managers attempt to reach them at least three times by phone.
To increase enrollment, Kansas expanded its criteria for potential members. Medicaid staff have visited the Central Plains office many times to review cases and program operations. Through these site visits and work with Central Plains, the State has been able to encourage a focus on the project's clinical outcome aspects.
Kansas bases its interventions on the use of an interdisciplinary team of nurse care managers and social service specialists. Interventions include in-person and/or telephonic care management, connection with community supports, collaboration with the PCCM program, and provider and patient education materials.
Kansas is contracting with an external evaluator to conduct an evaluation of the care management pilot program. The claims-based evaluation will compare the program to a reference group in a similar Kansas county. When designing the evaluation, the State, external evaluator, and Central Plains met to discuss the evaluation philosophy and goals and to set measures. Central Plains also has implemented its own internal evaluation to assess patient health and program outcomes.
Kansas Medicaid Web site: https://www.kmap-state-ks.us/
North Carolina: Community Care of North Carolina
North Carolina Medicaid operates a statewide enhanced PCCM program, Community Care of North Carolina (CCNC). Implemented in 1998 and built on a traditional PCCM program (Access), CCNC currently has approximately 745,000 Medicaid beneficiaries. North Carolina's program is based on local physician networks responsible for the local leadership of the enhanced care management programs.
North Carolina's Lessons Learned: Provider Engagement
Before the implementation of CCNC, the majority of North Carolina's Medicaid population was enrolled in Access, the State's PCCM program. Although the purpose of Access was to provide every enrollee with a medical home, it was not intended to serve as a holistic care coordination system for a large population. CCNC was developed to help primary care providers manage the Medicaid population's health care needs and improve the quality of their care. CCNC's gradual development allowed the provider networks time to create effective programs and show positive results without legislative scrutiny.
North Carolina's program includes 14 physician networks. Unique in terms of structure, community partners, and project activity, each CCNC network was designed locally, allowing it to best fit the needs of its region. Each network must collaborate (via a business associate agreement) with the local health department, department of social services, and hospital or hospitals. The CCNC networks range in size from 17,000 to 180,000 members and receive $2.50 per member per month from the State for administrative and operation costs. Network physician leaders came to a consensus to concentrate care management and quality improvement efforts initially on asthma, diabetes, high ER utilization, and high-cost patients, based on inpatient hospitalization data and ER utilization. Currently, CCNC is expanding its care management and quality improvement program to cover CHF statewide. Individual networks manage other chronic illnesses, including obesity, attention deficit hyperactivity disorder, COPD, mental health integration, and sickle cell anemia.
To build the CCNC networks, North Carolina relied on the unique strength of its physician community and the appeal of locally run programs. The first step in CCNC's network creation was to garner physician participation and buy-in by sending letters to primary care providers who serve more than 2,000 PCCM patients introducing them to the program's concept and asking them to participate. North Carolina then worked through an informal request for proposal (RFP) process with the interested primary care providers and other community Medicaid providers. Finally, the State partnered with local stakeholders to finalize program implementation.
Networks provide all beneficiaries with a medical home and a toll-free call center and selected beneficiaries with care management. Beneficiaries are selected for care management using claims data stratification (identification of high-cost beneficiaries with chronic conditions) and provider referrals. For high-intensity patients eligible for care management, care managers first call or send a letter introducing themselves. The care manager reviews the patient's chart, conducts a four-page assessment, and talks with the family, especially if the patient is a child. Finally, the care manager develops a plan of care with the patient. Assigned to physician offices, care managers help ensure that patients make and keep their appointments. Care managers work closely with the physician and attend physician office staff meetings to become part of the office team that manages patient care.
Since program inception, CCNC has collected data and monitored financial, quality, and health outcomes. CCNC conducts both claims and chart audits to review outcomes and process data and measures. The CCNC program office assumes responsibility for collecting a range of outcome measures via claims analysis (e.g., inpatient admission rate) and performance and process measures via randomized chart audits (e.g., implementation of an asthma action plan). Each network has a medical committee consisting of participating primary care providers that reviews evidence-based guidelines and Medicaid claims data to make recommendations to the clinical directors. State program staff meet regularly with the network clinical directors in finalizing the performance measures for the program. This process helps ensure physician buy-in and support for the measurement process. North Carolina has contracted with the University of North Carolina's Sheps Center for Healthcare Research and Mercer Consulting to evaluate program outcomes and savings.
North Carolina Medicaid Web site: http://www.dhhs.state.nc.us/dma/
Community Care of North Carolina Web site: http://www.communitycarenc.com
Sheps Center Evaluation: http://www.communitycarenc.com/PDFDocs/Sheps%20Eval.pdf