APA Urges Creation of Payment Codes Specific to Collaborative Care Model

Appropriate payment codes need to be developed to reflect the work of physicians—including psychiatrists—participating in collaborative care models treating patients with mental illness in primary care settings, said APA CEO and Medical Director Saul Levin, M.D., M.P.A., in a letter last month to the federal Centers for Medicare and Medicaid Services.

The 18-page letter was in response to CMS’s Proposed Rule for Medicare Program; Revisions to Payment Policies Under the Physician Fee Schedule and Other Revisions to Part B for CY 2016. The letter addresses a range of issues raised in the proposed rule—including improving payment accuracy for primary care and care management services, chronic care management and transitional care management services, the Physician Quality Reporting System, electronic health records, and the Medicare Access and CHIP Reauthorization Act.

But the bulk of APA’s comments in Levin’s letter focused on payment and coding issues around reimbursement for the collaborative care model (CoCM). The model, developed by leaders in integrated care at the University of Washington, involves a primary care physician, a care manager, and a psychiatric consultant.

“The lack of reimbursement for key components of this model has been the principal barrier to its widespread implementation,” Levin stated. “Although there may be other treatment models that engage primary care clinicians and behavioral health specialists, the specific Collaborative Care Model [referenced in the proposed rule] is the only model that has compelling scientific data supporting its effectiveness. Over 80 randomized, controlled trials have shown the CoCM to be more effective than care as usual. … In addition to the robust research evidence for the value of collaborative care, there is also substantial practice experience with this model of care from the Medicaid-funded Mental Health Integration Program in Washington State, the commercially funded DIAMOND program in Minnesota, and similar programs in several other states.”

The letter continued, “[T]he development of codes and requirements to be used for collaborative care for behavioral health conditions must be specific to the CoCM to enable its clinical approach and processes. To create codes that would facilitate any or all of the clinical roles or transactions embedded in the model (such as co-location of a care manager and screening mechanisms) without being tied to the other elements of the model (such as measurement-based care maintained in a registry with psychiatrist oversight) will not realize the substantiated results of the model’s utilization: that is, better quality patient care and outcomes and cost efficiencies.”

Specifically, Levin said that coding developed for the CoCM can facilitate appropriate valuation of the services furnished if it explicitly incorporates the clinical approach and processes required for an effective implementation of evidence-based collaborative care and if the coding accurately describes the work entailed in each of the explicit functions of each of the key members/providers on a collaborative care team.

Levin noted that APA’s review of current CPT and other HCPCS codes “did not yield any that accurately describe either the functions and services of a care manager or the services provided by the psychiatric consultant for the CoCM. Moreover, their functions and services far exceed the level of service imbedded in most of the current care coordination codes. … Therefore, we think one or two new codes need to be developed to properly value the services provided within the model.”

He said the care-management code or component should include the functions of the behavioral health care manager, such as patient education, outcomes tracking, coordination of care with primary care providers and psychiatric consultants, support of medication management, and provision of evidence-based psychosocial interventions. These interventions include brief counseling or psychotherapy, facilitation of specialty referrals as needed, routine evaluations of patient outcomes using validated outcome measures such as the PHQ-9 for depression, and the use of information technology such as a case registry to track clinical outcomes for all patients in care.

The second code or component would cover the consultation services provided to the primary care practice by the psychiatric consultant. These would include regular (usually weekly) review of all patients treated in primary care who are not improving, diagnostic and/or treatment recommendations to the primary care team, and availability for curbside consultations to primary care providers (PCPs) during work hours, Levin stated.

The APA comments also addressed how reimbursement for CoCMs would work in tandem with existing requirements for quality reporting. “For patients with depression (the predominant behavioral health diagnosis in primary care), the performance of CoCM programs should be monitored via at least two quality measures: a measure showing that PCPs are identifying people with depression via regular/universal screening, and clinical assessment of people who screen positive, using a standardized/validated instrument; and a measure showing that people with depression are remitting/recovering.”

He noted that as of 2015, CMS already requires both types of measures for depression as part of its Quality Measures and Performance Standards for the Accountable Care Organization Shared Savings Program.

“It is important that PCPs use both of these measures, rather than just one or the other. Prior research indicates that screening and case identification are not sufficient to improve patient outcomes,” Levin wrote. “If a PCP adopts the outcome measure but not the screening/assessment measure, then he/she may tend to avoid identifying people with depression and focus mainly on patients with good prognosis.”

He said that for other behavioral disorders treated in the CoCM, analogous measures should be used to ensure population-based identification of patients with a given condition and to track rates of remission/recovery. ■