For many years, states have worked to ensure that Medicaid participants have access to patient-centered medical homes (PCMH). In 2010, the Affordable Care Act (ACA) authorized state Medicaid programs to create health homes to improve and better coordinate the full range of services needed by Medicaid beneficiaries with chronic health needs. Capitalizing on these opportunities, Tennessee almost simultaneously implemented both a PCMH program to meet the primary care needs of all Medicaid beneficiaries and a health home program to coordinate behavioral and physical health services for beneficiaries with high behavioral health needs (Tennessee Health Link or THL). The PCMH program began in January 2017 and THL began in December 2016.
TennCare (Tennessee’s Medicaid agency) recognized both PCMH and THL providers needed comprehensive information about the service needs and services received by their Medicaid patients to be most effective. However, in the absence of a state health information exchange (Tennessee’s statewide HIE disbanded in 2012), there was no ready source of that data nor was there any existing organization seeking to turn data into the information providers could use to enhance performance. To meet this need, TennCare developed a care coordination tool (CCT) that marshalled data from diverse sources to help both PCMH and THL providers meet the primary and behavioral health care needs of their Medicaid patients. This brief is based on research and an interview with TennCare representatives.
What is the CCT?
The CCT is a cloud-based application that draws from Medicaid claims data, encounter data from TennCare’s contracted managed care organizations (MCOs), immunization data from the Tennessee Department of Health’s (TDH) statewide immunization information system (TennIIS), and information about admissions, discharges and transfers (ADTs) provided by hospitals via data exchange partners, Tennessee Hospital Association (THA), East Tennessee Health Information Network (etHIN) and Community Hospital Systems (CHS). This information, as presented by the CCT, helps providers identify and fill gaps in patient care. It also speeds notification of hospitalizations and emergency department visits enabling providers to more quickly provide transitional and follow-up care when these events occur. Specifically, the types of information in the CCT include:
- Name and available demographic information for Medicaid beneficiaries who are members of a provider’s panel
- Claims based clinical data drawn from both claims and encounter data, including medication and diagnosis history, as well as records of visits to providers.
- ADT information from emergency departments and hospitals.
- Immunization data for children under 2 years and those 9-13 years of age.
- Each provider’s performance on the quality measures to which PCMH and THL value-based payments are tied.
- Care alerts that identify ‘past due’ services needed by an individual patient such as, mammogram, follow-up to a hospital discharge, or an update to the patient’s comprehensive person-centered care plan.
PCMH and THL providers are eligible for bonus or outcome payments based on each organization’s performance on total cost of care, and a set of efficiency and quality measures. The method of calculating efficiency and the specific quality metrics included in the set varies between THL and PCMH providers—and among PCMH providers by the mix of children and adult Medicaid beneficiaries attributed to the practice. Although the CCT is an optional tool for providers participating in the THL and PCMH programs, it provides a platform for providers to act on near real-time data and shows the impact of their work on the quality and efficiency measures by which provider performance is measured. (More information on payment can be found in TennCare’s PCMH and Health Link operational manuals.)
Multiple data sources produce more useful data
Designed for PCMH and THL providers, the CCT ultimately offers a single consolidated source of information about providers’ panel members, even when members are enrolled in different MCOs. The CCT includes information to help providers identify their members’ primary care needs, any services that have been delivered by other providers, hospital and emergency department ADTs, and more. TennCare obtains much of this information from the claims and eligibility data contained in the agency’s Medicaid Management Information System (MMIS). However, TennCare recognized there are still challenges which exist with incorporating multiple data sources for the CCT. One challenge is the sometimes significant lag between when a service was delivered and when data about that service appears in claims data. Another common barrier for providers and other users of the tool is the lack of connectivity between the CCT and their electronic health records (EHRs), which may mean users need to enter the same information in both the CCT and EHR. Both challenge providers’ ability to use the CCT to obtain the information they need to care for their patients.
TennCare’s partnerships with other organizations that maintain other data sources is a large factor in the agency’s success. The THA became the agency’s first partner when, in 2017, TennCare began contracting with the association to obtain near real-time ADT data to populate the CCT. Because TennCare needed complete ADT information it then worked with the THA to draw in those hospitals that were not already submitting data to the THA. For example, TennCare issued guidance to hospitals requiring them to submit ADT data to qualify for the directed payment of funds that were formerly distributed as unreimbursed hospital cost pool payments. 100% of hospitals in Tennessee now submit ADT data to the THA. Thus, the THA was able to create a more complete data set by partnering with TennCare.
In 2019, TennCare also began contracting with the Tennessee Department of Health (TDH) to incorporate specific immunization data from the statewide immunization information system (TennIIS) into the CCT. This data is used to supplement claims data which populates performance on specific Healthcare Effectiveness Data and Information Set (HEDIS) immunization measures and to identify patients in need of immunizations. Like the THA, the TDH benefits from the contract, as it commits TennCare to working with the TDH to improve the quantity and quality of data submitted to the registry by providers and furthers the shared work of both TennCare and the TDH of improving immunization rates for children in Tennessee.
Although these specific data sources may not be available in other states, it is likely that others are. TennCare representatives advise thorough planning is necessary to identify the information needed in the tool—both at launch and in the future. Identifying the partners and resources needed to access data outside of the Medicaid agency’s control also requires thorough planning. Further, developing mutually rewarding partnerships (and contracts) takes time. It is important to have a long-term plan in place early in development.
The CCT is a large investment, but state costs are offset by federal matching funds
The CCT is a large investment. Its development required leadership and dedication of a wide range of staff. Multiple TennCare departments were involved, including the chief medical office, behavioral health operations, information systems, legal, and others. The CCT’s development required the input and feedback of providers and MCOs that would ultimately benefit from the tool. In addition to the contracts with the CCT vendor (HealthEC), THA and TDH, TennCare information system (IS) contractors were also utilized for design and implementation. TennCare representatives estimate that a group of 15-20 were heavily involved in development, but many others were called in as needed.
Of course, the largest part of the investment is the cost to develop and maintain the CCT. TennCare conducted a procurement to select the first vendor to develop and operate the CCT—this contract was awarded in 2016. The second CCT, launched in November 2020, was procured through a current TennCare IS contractor. TennCare representatives feel the new CCT better meets their vision for the tool and the needs of THL and PCMH organizations.
Based on their experience, TennCare representatives advise states seeking to develop a tool to ensure that the selected vendor has qualifications to accurately provide and process the data as inaccuracies will quickly lead to distrust of information flowing through a tool. TennCare representatives advised that it is important for states to:
- Set basic qualifications for vendors and involve knowledgeable information technology staff in determining whether potential vendors meet those qualifications.Select a vendor already working with large, multi-site locations and processing many different types of data from multiple data sources.
- If state plans include generating HEDIS measures it is important the vendor have a National Committee for Quality Assurance (NCQA) Measure Certification.
- Seek input from the providers who will be using the tool, for example by convening technical assistance groups of providers to help with design or conducting end user testing with providers.
- Conduct detailed reference checks on potential vendors to find out their strengths and weaknesses.
- Have a strategic training and engagement plan for the 12 months post go-live
Although the investment was large, the original CCT implementation qualified for 90/10 federal matching funds under the Health Information Technology for Economic and Clinical Health (HITECH) Act. As part of the Medicaid agency’s MMIS, TennCare receives 75/25 federal matching funds for the new HealthEC CCT. This funding greatly reduced the cost to the state. (Note: HITECH funding is only available through 2021, but other opportunities will remain to secure 90 percent funding for design and development of HIE and HIT.)
There is evidence of success
One measure of success is whether providers use the tool. As of June 2021, all PCMHs, THLs, and MCOs have registered at least one staff person to use the tool and a total of 662 MCO, PCMH, and THL users have registered. TennCare representatives report the enhanced ADT data was a big draw for these providers. TennCare offers monthly trainings to some users and individualized assistance to others. TennCare and its contractor have also developed a learning library that includes recorded trainings and quick reference guides to help providers use the CCT to accomplish tasks such as identifying frequent emergency department users.
The most important measure of success, however, is whether using the CCT produces improved cost and quality outcomes. This question is difficult to answer due to the multiple influences in reducing cost and improving quality. However, TennCare did evaluate both the PCMH and THL programs, and agency representatives believe there are indications in these evaluations that using the CCT did produce improved outcomes. Key findings from these evaluations include the following.
- The state’s evaluation of the PCMH found evidence of improved quality, increased access to primary care, and increased follow-up visits following emergency department and inpatient visits.
- The state’s evaluation of the THL program found indications of improved access to primary care among THL members. It also found the THL members had lower inpatient and emergency department visit rates then a comparison group of similar Medicaid program participants who were not enrolled in the program.
While these changes cannot be directly attributed to use of the CCT, TennCare representatives believe it did play a role by, as designed, helping providers identify gaps in care and providing the information needed to improve follow-up and coordination. This belief is bolstered by findings from focus groups and interviews with PCMH and THL providers. These providers frequently mentioned the importance of the tool for care coordination, especially the ADT data. Also, many of the providers wanted even more information, suggesting they see great value in the information provided by the CCT. There is anecdotal evidence the information in the CCT helped providers identify the specific primary care and behavioral health needs of their patients and facilitate access to care.
TennCare’s experience illustrates how state Medicaid agencies can develop IT tools that successfully help providers improve the care delivered to Medicaid beneficiaries, even in the absence of a statewide HIE. TennCare’s success was rooted in meeting provider needs. The agency considered end user needs and preferences during the CCT’s development, provides accurate and near real-time data to improve care coordination efforts, and focuses on giving providers the information they need to improve performance linked to payment. To accomplish this, TennCare looked to data sources outside those operated by the Medicaid agency to create a more timely and complete set of data and developed a strong partnership with the state’s IS agency. Finally, access to enhanced federal matching funds offset the cost of the investment.
Acknowledgements: The National Academy for State Health Policy (NASHP) would like to thank the state officials from Tennessee who contributed to the brief as well as Health Resources and Services Administration Project Officer Diba Rab and her colleagues for their feedback and guidance. We also thank the state officials from Montana and the staff of the Montana Healthcare Foundation whose interest in improving the care delivered to Montana Medicaid participants led to the creation of this brief. Finally, the author wishes to thank Hemi Tewarson, Kitty Purington, Jodi Manz, and Luke Pluta-Ehlers of NASHP for their contributions to the paper. This project was supported by the Health Resources and Services Administration (HRSA) of the US Department of Health and Human Services (HHS) under co-operative agreement number UD3OA22891, National Organizations of State and Local Officials. The information, content, and conclusions are those of the author and should not be construed as the official position or policy of, nor should any endorsements be inferred by HRSA, HHS, or the US government.