Comments at the Oregon Health Forum
John A. Kitzhaber, M.D.
June 21, 2018
Let me start by revisiting the policy concepts that guided the creation of the Coordinated Care Organizations (CCOs) in 2011. The original vision was threefold. First, we wanted to create a new kind of organization that was community-based with local providers, local citizens and a local governance structure. Second, we wanted an organization that would move away from the narrow medical model and focus more broadly on community health. And finally, this organization would operate within a global budget indexed to a sustainable growth rate (3.4% per member per month) and be required to meet rigorous metrics around health outcomes, quality and patient satisfaction.
Over the past five years, the CCOs have operated within the constraints of the per capita growth cap, enrolled over 385,000 more people under the ACA Medicaid expansion; met the outcome and quality metrics stipulated under the waiver; and Oregon has realized a cumulative total funds savings of over $1 billion. It is important to remember these accomplishment and to make sure that the effort to design CCO 2.0 does not undermine the principles that led to this remarkable success—which is unparalleled in any other state.
I think CCOs face three major and interrelated challenges over the next five years: (1) reestablishing a true global budget; (2) defining the appropriate role and scope of the CCOs in addressing the social determinates of health; and, (3) doing a better job integrating behavioral health. Let’s start with the global budget.
A central tenant of the CCO model was a global budget that allowed the flexibility to invest in things that were not “medical” but which had an impact on community health (e.g. an air conditioner for a woman with CHF in the summer; housing, job support and other social services). This is a critical point. The financing model for CCOs was to be based on a budget, not a rate. The CCO would receive a fixed amount of money and a set of outcomes and would then be responsible and accountable for meeting those outcomes within the constraints of the growth cap. The current rate setting process undermines this basic principle.
Establishing what “actuarially sound rate” at which to pay CCOs creates a disincentive for exactly the kind of investments needed to begin to address the social determinants of health. Determining actuarial soundness involves analyzing claims, then trending the cost forward—which simply embeds all the silos, inefficiencies and perverse incentives that we have in the current system, into the CCOs going forward–which is exactly what we were trying to avoid.
Furthermore, there are activities that yield a significant health benefit, but for which there are no claims or codes. We started our work around CCO’s understanding there was no code or accounting for buying a window air- conditioner, for example, let alone for investing in housing. Unfortunately, we have spent the better part of past five years trying to figure out how to count these things rather than on how best to deliver those services. The result? There is no incentive for efficiency or for making the kinds of investments that make the CCO different from a traditional medical provider. CCOs were intended to move us away from the narrow medical model and focus more broadly on community health. Rate setting moves us in the opposite direction.
On the question of the social determinants of health and equity, let’s be clear that this is not solely the purview of CCOs although they certainly have an important role-play. All the efforts around Kindergarten readiness, for example, are also focused on addressing adverse childhood experiences (ACEs) and other social determinants. The central problem is that we lack an effective early childhood delivery system that can get the right services to the right children and families at the right time, and for long enough to make a difference. This is a longer conversation, but if we can’t collectively solve this problem, investments by CCO’s or by anyone else in the social determinants of health will not produce the outcomes we seek.
With that in mind, there are at least two areas where CCOs should logically be involved. The first is during the first 1,000 days of life (from conception to age two) because it is during this time that the foundations of optimum health, growth, and neurodevelopment are established. Furthermore, we know from epigenetics that stress during a woman’s pregnancy can alter genetic expression, leading to problems in the emotional, behavioral and physical development of the child.
Since the health care system sees almost all children well before they ever reach school age or interact with any other system—with the possible exception of child care—this is where the initial ACE/social determinants risk screening should be done, starting with One Key Question. If a woman is planning to get pregnant within the next year and her risk profile reflects the kinds of stress factors that can impact child development during pregnancy or the first 1,000 days, she should be referred to an effective early childhood delivery system. Women who are not captured during the One Key Question interview should receive the risk screening at the first prenatal visit.
The second area where CCOs can have a huge impact is in behavioral health. We know that the presence of untreated mental illness and/or substance abuse disorder in a family is a potent adverse childhood experience. Indeed, nearly two thirds of children who end up in foster care do so because of parental substance abuse. Furthermore, for many CCOs, opioid dependence is the single most expensive diagnostic code on a per member per month basis, straining the global budget.
Thus, better integration of behavioral health must be a top priority for CCOs. Doing so requires a clear-eyed recognition that our current community mental health system is appropriately much more focused on social services than simply on medical treatment, creating a cultural challenge. Furthermore, the behavioral health system has been chronically underfunded. Reimbursement for mental health is far lower than for physical health; and funding for substance abuse disorder treatment is even lower than for mental health.
Addressing these challenges will require a number of things, including: clarifying and agreeing upon what should be treated in the physical health system versus the behavioral health system; better care coordination and communication between the two systems, including improving interoperability of the EHR; establishing reimbursement parity between behavioral health and physical health; and directly confronting the social stigma that persists around mental health and substance abuse disorders.