As health care and financing systems become more sophisticated, health care systems are increasingly using a process known as "risk tiering" to group patients with similar degrees of need for health care and care coordination services. Payers are becoming part of this conversation, which for children is in its very early stages. Families and care providers of children with chronic and complex conditions should understand the risk tiering process, as it may affect access to services these children need.
This February 28 conversation was led by three of the authors of a new report, Aligning Services with Needs: Characterizing the Pyramid of Complexity Tiering for Children with Chronic and Complex Conditions. Speakers described current tiering practices and uses, and made recommendations for policy and research, particularly as the process relates to children with special health care needs.
- Christopher Stille, MD, MPH - University of Colorado School of Medicine, Children’s Hospital Colorado
- James Perrin, MD - Harvard Medical School, Massachusetts General Hospital for Children
Our biggest challenge is with payment for care coordination and having the staff to support care coordination from discharge to outpatient care. What financial resources are available to provide team-based care? What shared care plan templates are available that we can use to help facilitate care, communication?
Chris: Payment for care coordination is a very local thing - some practices have partners (hospitals, state Medicaid, other payers) to help address this. There are lots of shared care plan templates out there. Epic users are working on several tools in their “healthy planet” suite, notably the LPoC (longitudinal plan of care) tool.
When you mention social determinants, are you referring to adverse childhood experiences (ACES) or more general measures of social determinants of health?
Chris: Broader. ACES are part of it, but most of the social determinants of health we mentioned are more immediate, like food insecurity, housing instability, financial problems, and family instability.
Are there tiering tools that outpatient health systems could use for children with medical complexity?
Chris: Yes, there are several out there - there is a brief outline in the report (pages 6-8).
Who is the best team member to administer a social determinants of health screening tool to obtain the most accurate information?
Chris: This varies – some places have the medical assistant do it, other have a nurse or social worker, others a provider. More important is having a workflow and resources/connections to handle different types of positive screens and being able to refer them to appropriate services.
Are there ways to use screening tools for risk stratification or does it rely on a blend of screening and administrative data? If so, which tools?
Chris: A few centers are integrating social determinants of health (SDH) tools with diagnosis-based risk stratification, in different ways. For example, some centers use SDH data to determine tiers, while others connect patients to a separate package of resources (e.g. social work). You need data from both administrative sources and targeted screening to provide a complete picture of risk.
You report that risk tiering is not yet used to inform payment except in a few localities but that using tiering for payment is on the horizon. What are the cautions that states should be mindful of in using tiering for payment?
Jim: It is critical that states know whether the tiering methods they use have any validity – i.e., will they work efficiently and fairly to allocate resources? As an example, one state recently determined eligibility for a major program for children with special health care needs by using a single question for parents that could (and did) lead to major misclassifications. Using tiering methods to understand population differences makes much more sense than using tiering methods to determine eligibility for services.
How do you define complex needs for the children you are discussing?
Chris: Different systems define complexity differently. Most risk-tiering systems think of the top 5-6% of children with the most complex needs, which is the population identified by the Seattle Pediatric Medical Complexity Algorithm and the 3M Clinical Risk Group categories 5b-9.
How do we address the problem of transition when CSHCN “tiers” do not match up with tiering activities in the adult world? Adult population health programs are often focused on specific conditions (e.g. diabetes, congestive heart failure, chronic kidney disease) and as a result, the needs of children with complex conditions may not be addressed adequately by the system.
Chris: We need a separate system of tiering that is specific to children – this is outlined in the report (Recommendations, page 11).
Do you see a tiering system facilitating increased immediate access to specialists such as pediatric neurologists and physiatrists?
Chris: It is possible, if systems prioritize access to specialists for more complex kids. Most places now do it on a case-by-case basis.
Are there standardized tiering criteria and tools that can be used in practice, instead of us developing our own?
Chris: There is nothing standardized yet, but the Children’s Hospital Association Center for Medicare and Medicaid Innovation Coordinating All Resources Effectively (CARE) project used standardized Clinical Risk Group scores to identify the initial population for contact.
How would a health care system affect social and behavioral determinants of health in the community?
Chris: In a lot of ways. Providing connections to community services and resources related to a family’s needs is one first step.
How can clinicians tier their populations if they do not have access to administrative claims data?
Chris: Some do it by just naming the kids that are complex in their eyes – this works pretty well, if you have a relatively small group of patients.
Dr. Perrin stated that we need to think about public resources as one resource to improve children's health over time -- this was a very helpful perspective. What would be, or is there, a parallel argument for private payers?
Jim: Many of the services and programs that influence child health and well-being are not found within the health sector. Programs and communities that aim to improve children’s well-being need to coordinate health with other community services (education, financial support, housing, etc.)
Much of what was found in terms of tiering could be applied to any child population, not just CSHCN. We know that a number of parental mental health and other family social needs put "apparently healthy children" at significant risk of later behavioral, mental health, and social challenges. What are the risks and benefits of thinking about tiering in terms of CSHCN vs. "apparently normal children"?
Chris: Risk is very important - that’s why including social determinants of health, as well as revising criteria/updating data frequently, is important for kids.
Is there a role for validated screening tools like the ASQ, SWYC, MCHAT, Edinburgh, PSC, and PHQ-9 to help with tiering?
Chris: Yes, these are all good tools. Some systems are now integrating results from these tools into their tiering systems.