SHADAC Advocates a Data-based Approach to Advancing Medicaid and CHIP Access Monitoring Plan (Response to CMS Request for Information)
April 29, 2021:
On February 17, 2022, the Centers for Medicare & Medicaid Services (CMS) released a request for information (RFI) regarding access to coverage and care in Medicaid and the Children’s Health Insurance Program (CHIP).
SHADAC researchers focused our response on Objective 4: Question 1, which asked researchers to consider how CMS might develop a stronger Medicaid and CHIP access monitoring approach through data reporting and analysis, and is copied for reference below.
Portions of SHADAC’s response were also included as part of a larger commentary on each of the RFI’s five total objectives submitted to CMS by President and CEO Richard Besser, MD, of the Robert Wood Johnson Foundation (RWJF) on April 18, 2022.
Centers for Medicare & Medicaid Services (CMS) – Request for Information (2022)
Access to Coverage and Care in Medicaid & CHIP
Objective 4: CMS has data available to measure, monitor, and support improvement efforts related to access to services (i.e., potential access; realized access; and beneficiary experience with care across states, delivery systems, and populations). CMS is interested in feedback about what new data sources, existing data sources (including Transformed Medicaid Statistical Information System [T-MSIS], Medicaid and CHIP Core Sets, and home and community based services (HCBS) measure set), and additional analyses could be used to meaningfully monitor and encourage equitable access within Medicaid and CHIP programs.
1. What should CMS consider when developing an access monitoring approach that is as similar as possible across Medicaid and CHIP delivery systems (e.g., fee-for-service and managed care programs) and programs (e.g., HCBS programs and dual eligibility in Medicaid and Medicare) and across services/benefits? Would including additional levels of data reporting and analyses (e.g., by delivery system or by managed care plan, etc.) make access monitoring more effective? What type of information from CMS would be useful in helping states identify and prioritize resources to address access issues for their beneficiaries? What are the most significant gaps where CMS can provide technical or other types of assistance to support states in standardized monitoring and reporting across delivery systems in areas related to access?
Response from SHADAC
Thank you for the opportunity to help inform the development of an improved Medicaid and CHIP access monitoring plan. These comments are based on SHADAC’s experience providing data and evaluation technical assistance (TA) to states, which has given us a unique view into the day-to-day challenges and barriers that states encounter related to these issues. These comments mirror recent testimony to the Medicaid and CHIP Payment Access Commission (MACPAC) provided by SHADAC’s Deputy Director.
We present five points for your consideration, each of which are guided by the same principle, which is: That any monitoring plan should seek to minimize burden on state agencies.
First, consider an iterative approach that starts with a limited number of measures and expands over time. This will require difficult tradeoffs regarding priorities, but these tradeoffs will be worth it if the first iteration is achievable for a wide range of states. As the plan develops and more complicated measurement concepts are added, it should engage leading states by including them in additional measure selection and incentivizing them to participate in pilots that test the collection and analysis of data – sharing concrete implementation lessons with other states.
Second, support states in efforts to improve existing data with a focus on the ability to disaggregate. There are existing data streams to draw on (as documented by Urban Institute in 2017 and discussed at a recent MACPAC meeting), and while none are perfect, a successful plan should first focus on improving those. One tangible way to improve states’ existing data is to support data disaggregation efforts. There is a renewed focus and energy to promote equity within Medicaid and calls from stakeholders to see data about important groups of interest. This includes an interest in better data by race, ethnicity, sexual orientation, gender identity, disability, and geography. Improving existing data streams to better support disaggregation is a good investment and one that will meet multiple demands.
For example, the Transformed Medicaid Statistical Information System (T-MSIS) will likely play a role in monitoring service use, but there are concerns regarding the existing quality and completeness of the race and ethnicity data. We have worked with several states who are trying to improve the collection of race and ethnicity data in Medicaid—modifying question wording and expanding response options to better represent the populations they serve, making technical changes to better capture the data, and modifying instructional language and scripts for enrollment assisters to make them stronger partners in data collection. And, most importantly, they are doing this with community input. But they continue to face challenges in this work. For example, current OMB standards for demographics are dated and do not align with the most current research. Additionally, rules for collecting race and ethnicity data are not uniform across federal programs. A new access monitoring effort could serve as further impetus for federal agencies to revise the guidance on race and ethnicity and for states to take action to improve existing data collection to address this critical data gap.
Third, states need both direct funding and hands-on technical assistance to support this work. We appreciate CMS’s understanding that states will need assistance to implement the access monitoring framework. However, based on our work with states, we think this needs to go beyond documentation and uniform measure specifications. To do this well, states should have access to experts who can provide practical, hands-on advice that is responsive to their specific needs. The type of assistance needed will vary and needs to be flexible, from states who are further along in their process and may seek help troubleshooting a particular coding challenge to other states whose request might require more fundamental support, such as walking through the requirements to assess staff and training needs, setting priorities, and help developing contract amendments or RFPs. We would also recommend that any TA effort include a forum for states to discuss implementation challenges with their peers.
The funding provided to states for this effort should also be flexible. It should support direct costs like system modifications, but also things like stakeholder engagement, which is critical to the iteration and improvement of the monitoring plan, and related data collection, but will also allow the results of this monitoring to be shared in a meaningful way.
Fourth, some areas of Medicaid access monitoring are best addressed through federal data collection. We believe that the access monitoring effort would benefit from periodic fielding of a 50-state Medicaid Consumer Assessment of Healthcare Providers and Systems (CAHPS) survey aimed at both children and adults and administered by the federal government. The federal government has a long and successful track record of fielding high-quality surveys that produce estimates for all 50 states, including a one-time National Medicaid Adult CAHPS survey.
While this undertaking would be resource intensive, it seems to us to be the most efficient way to collect comparable information on important facets of access, like enrollee experience, and a starting point for comparing individuals enrolled in fee-for-service versus managed care.
Finally, it is critical to treat states as a full partner in this process, including the communication of results. It goes without saying that states should be consulted in the development of the monitoring plan, but they should also be consulted when the data are being released, preferably beforehand. Ideally, they should have access to analytic files so that they can do their own data runs and share customized findings with stakeholders. Finally, once the data are collected, there should be a commitment that the data be published or released in some format in a timely manner.