Advanced Primary Care: A Foundational Alternative Payment Model (APC-APM) for Delivering Patient-Centered, Longitudinal, and Coordinated Care A Proposal to the Physician-Focused Payment Model Technical Advisory Committee From the American Academy of Family Physicians April 14, 2017 AAFP Contact: Kent Moore Senior Strategist for Physician Payment American Academy of Family Physicians 11400 Tomahawk Creek Parkway Leawood, KS 66211 Phone: 800-274-2237, extension 4170 Email: [email protected] April 14, 2017 Physician-Focused Payment Model Technical Advisory Committee c/o Assistant Secretary of Planning and Evaluation Office of Health Policy U.S. Department of Health and Human Services 200 Independence Avenue S.W. Washington, DC 20201 Proposal – Advanced Primary Care: A Foundational Alternative Payment Model (APC- APM) for Delivering Patient-Centered, Longitudinal, and Coordinated Care Dear PTAC Committee Members: The American Academy of Family Physicians (AAFP) is fully supportive of the Physician- Focused Payment Model Technical Advisory Committee’s (PTAC) role in evaluating physician-focused payment models (PFPMs) and making subsequent recommendations about those models to the U.S. Department of Health and Human Services (HHS). The AAFP believes that to be truly successful in improving care and reducing cost, PFPMs and alternative payment models (APMs) need a strong foundation of primary care. Therefore, on behalf of the AAFP, which represents 124,900 family physicians and medical students, I am particularly pleased to submit the following proposal—Advanced Primary Care: A Foundational Alternative Payment Model (APC-APM) for Delivering Patient-Centered, Longitudinal, and Coordinated Care. We request that the PTAC review the model, provide feedback to the AAFP on it, and promptly recommend it to HHS for approval and nationwide expansion. Family medicine plays a critical role in delivering care to Medicare beneficiaries in every community across the country. The AAFP’s Advanced Primary Care-Alternative Payment Model (APC-APM) proposal is an opportunity for the Centers for Medicare & Medicaid Services (CMS) to make advanced APMs broadly accessible to Medicare beneficiaries—and to impact quality and spending in other parts of the health care system. The foundational role of family medicine in care delivery is clearly illustrated by the following: • Family physicians are the most visited specialty—especially in underserved areas. Family physicians conduct approximately one in five office visits. This represents more than 192 million visits annually, which is 48 percent greater than the next most visited medical specialty.1 Family physicians provide more care for America’s underserved and rural populations than any other medical specialty. • Strengthening primary care is critical to driving greater value for beneficiaries, payers, and communities. Transformation cannot be overly complex and burdensome to operationalize. However, there is not a one-size-fits-all solution, as patient panels, populations, and primary care practices vary. There is an emerging consensus that strengthening primary care is imperative to improving individual and population health outcomes, as well as to restraining the growth of health care spending. • The complexity of care provided by family physicians is unparalleled in medicine. Data show that family physicians address more diagnoses and offer more treatment plans per visit than any other medical specialty. Furthermore, the number and complexity of conditions, complaints, and diseases seen in primary care visits is far greater than those seen by any other physician specialty.2 CMS and private payers must make new investments in primary care to truly capture and realize the value proposition of family medicine and primary care. • Primary care is particularly affected by longstanding inequities in payment that must be corrected if it is to be the foundation of a transformed, patient-centered health system. Research shows that fee-for-service “(FFS) is not only flawed for its strong incentives to increase volume, but also in its disproportionate reimbursements for procedural rather than cognitive care.”3 Payment experts offer similar assessments of the problems with testing and building value-based payment models on a flawed physician fee schedule. Drs. Robert Berenson and John Goodson wrote in the New England Journal of Medicine, “If the foundation of Medicare’s fee schedule isn’t sound, these systems will be unstable.”4 According to the 2016 Medicare Payment Advisory Commission (MedPAC) report, compensation continues to be much lower for primary care physicians than for physicians in subspecialty disciplines.5 • Distinct Advanced Alternative Payment Models (AAPMs) must be made available nationally to all primary care physicians. Though primary care oriented AAPMs will continue to clinically coordinate with other payment models, primary care AAPMs must be distinct from bundled payment models to maximize support for the delivery of continuous, longitudinal, and comprehensive care across settings and providers. Including primary care in bundled payments will not provide the support our health system needs to increase value and strengthen primary care. Primary care is the primary access point to the health care system for millions of Americans across a diverse range of communities. The AAFP is pleased to present the APC-APM proposal to the PTAC to ensure that more Medicare beneficiaries have access to care delivered under advanced APMs. We feel this will help achieve the goals of improving overall health outcomes of Medicare beneficiaries and the health of communities, as well as bring stability to the Medicare program. Thank you for your time and consideration of this proposal. For any questions you might have, please contact Mr. Kent Moore, AAFP Senior Strategist for Physician Payment, at (800) 274-2237, extension 4170, or [email protected]. Sincerely, Wanda D. Filer, MD, MBA, FAAFP Board Chair TABLE OF CONTENTS I. Background and Model Overview ............................................................................. 1 II. Scope of Proposed PFPM (High Priority Criterion) ................................................ 1-4 III. Quality and Cost (High Priority Criterion) ............................................................... 4-6 IV. Payment Methodology (High Priority Criterion) .................................................... 6-12 V. Value over Volume ............................................................................................. 12-13 VI. Flexibility ............................................................................................................. 13-14 VII. Ability to be Evaluated ........................................................................................ 14-15 VIII. Integration and Care Coordination ..................................................................... 15-16 IX. Patient Choice ......................................................................................................... 17 X. Patient Safety .......................................................................................................... 17 XI. Health Information Technology .......................................................................... 17-18 Advanced Primary Care: A Foundational Alternative Payment Model (APC-APM) for Delivering Patient-Centered, Longitudinal, and Coordinated Care ABSTRACT The Advanced Primary Care-Alternative Payment Model (APC-APM) embodies the principle that patient-centered primary care is comprehensive, continuous, coordinated, connected, and accessible from the patient’s first contact with the health system. The APC-APM aims to improve clinical quality through the delivery of coordinated, longitudinal care, and uses the approach to deliver care that improves patient outcomes and reduces health care spending. The APC-APM is envisioned as a multi-payer model that builds on concepts already tested through the Comprehensive Primary Care (CPC) and CPC Plus (CPC+) initiatives. The APC- APM would be open to almost 200,000 primary care physicians and potentially impact more than 30 million Medicare patients. Based on available evidence, additional spending on primary care is projected to be more than offset by savings elsewhere in the health care system, resulting in a net savings to the payers involved. Each APC-APM entity will be evaluated based on reporting six measures, with one being an outcomes measure in order to align with the Medicare Access and CHIP Reauthorization Act’s (MACRA’s) Merit-based Incentive Payment System (MIPS) reporting requirements. These measures will come from the core measure sets developed by the multi-stakeholder Core Quality Measures Collaborative to ensure focus, alignment, harmonization, and the avoidance of competing quality measures among all payers. These measure sets include patient experience measures, and all but one of the core measures are also measures under the MIPS. The APC-APM would create a new payment structure for participating primary care practices consisting of a combination of four mechanisms: • A prospective, risk-adjusted, primary care global payment for direct patient care; • Fee-for-service limited to services not included in the primary care global fee; • A prospective, risk-adjusted, population-based payment; and • Performance-based incentive payments that hold physicians appropriately accountable for quality and costs. Other features of the model require that physician practices be: • Fully flexible to accommodate differences in clinical settings and patient subgroups covered by primary care; • Able to be fully evaluated for quality and cost at the model and APM entity levels; • Reflective of the Joint Principles of the Patient-Centered Medical Home (PCMH) and the five key functions of the CPC+; • Attribute patients based primarily on patient choice; and • Adopt, and ultimately use, interoperable, certified health information technology, with the expectation that at least 50% of qualifying participants will use certified electronic health record technology (CEHRT). Page 1 of 20 April 14, 2017 I. Background and Model Overview The APC-APM is built on the principle that patient-centered primary care is comprehensive, continuous, coordinated, connected, and accessible from the patient’s first contact with the health system. While the APC-APM aims to improve clinical quality through the delivery of coordinated, longitudinal care—assessed through the Core Quality Measure Collaborative measure sets—the broader goal of the APC-APM is to use this approach to deliver care in a manner that improves patient outcomes and reduces health care spending, such as through decreased inpatient and emergency department visits. As illustrated in our proposal, the APC-APM would accomplish this through a prospective, risk-adjusted primary care global payment for direct patient care, a population-based payment covering non-face-to-face patient services, along with fee-for-service (FFS) payments, which are limited to services not otherwise included in the primary care global payment. These payments are coupled with prospective performance-based incentive payments that hold physicians appropriately accountable for quality and costs by rewarding practices based on their performance on patient experience, clinical quality, and utilization measures. Supporting information about the value of primary care to patients and payers in terms of its positive effects on costs, access, and quality, as well as policy details on how the APC-APM would advance these goals are described in the AAFP’s position paper, “Advanced Primary Care: A Foundational Alternative Payment Model (APM) for Delivering Patient-Centered, Longitudinal, and Coordinated Care.”6 A copy of this position paper is found in Appendix A. In it, we present a transformational, primary care focused, and patient-centered model, including: • The definition and recognition of an APC-APM participating physician; • An appropriate, four-step methodology to attribute patients to the APC-APM; • How global and performance-based incentive payments should be structured and made; • Reporting quality measures and the calculation of value-based payments; and • Financing for the model. II. Scope of Proposed PFPM (High Priority Criterion) In this section, PTAC seeks input on ways the APC-APM would broaden or expand CMS’ APM portfolio by either addressing an issue in payment policy in a new way or including APM entities whose opportunities to participate in APMs have been limited. Goals of PFPM. The AAFP appreciates that the PTAC and Center for Medicare and Medicaid Innovation (CMMI) are working to increase the number and variety of models available to ensure that a wide range of clinicians, including those in small practices and rural areas, have the option to participate in an Advanced Alternative Payment Model (AAPM) under the Medicare Access and CHIP Reauthorization Act (MACRA). While the AAFP fully supports that CMMI tested the CPC initiative and is currently testing the CPC+ model, as well as other primary care transformative models, those models are limited to specific markets in certain geographical regions. Patients not in those regions are unable to benefit from the same improved access to primary care and more coordinated care in regions where the CPC and CPC+ models are being tested. By recommending to HHS that the APC-APM be Page 2 of 20 April 14, 2017 implemented nationally, the AAFP is hopeful that all patients would benefit from primary care’s positive effects on access, quality, cost, and health promotion. The APC-APM concept is being actively tested through the multi-payer CPC and CPC+ models that promote longitudinal, comprehensive, and coordinated care with primary care teams. Early CPC evidence shows that participating practices: • Continue to make progress in how they deliver primary care functions and are advanced in risk-stratified care management. • Generate improved patient experiences among attributed Medicare beneficiaries. • Reduce emergency department visits and Medicare Parts A & B expenditures. • Reduce total monthly Medicare expenditures not including care management fees. • Are more advanced in other aspects of care delivery than comparable practices. The feasibility of the CPC and CPC+ models indicates that the APC-APM could be implemented across a diverse set of family medicine settings that are committed to practice transformation. Physician Practices. The AAFP envisions the APC-APM to be available to all physicians with a primary specialty designation of family medicine, general practice, geriatric medicine, pediatric medicine, or internal medicine. The AAFP also envisions the APC-APM to be designated as an AAPM under MACRA. Based on Physician Compare data, there are approximately 195,000 such primary care physicians that could practice within a designated AAPM entity. Given the evident merits of the model, the push from CMS to tie more Medicare payments to quality and value, and the current small number of AAPMs under MACRA, we anticipate that many of these physicians would express interest and willingness to participate in the model if it is approved and expanded to scale. The APC-APM is equally applicable to physicians who are employed or independent, which is especially critical for increasing participation in AAPMs among rural and/or small practice physicians. The AAFP has supported member recruitment and education related to the CPC and CPC+ models and found widespread interest in participation, which we believe will lead to broad interest in participating in the APC-APM. To the extent that a large portion of the services provided will be capitated through the global primary care payment and population-based payment, the APM entity and its eligible clinicians will bear risk for performance related to those services. Additionally, the APM entity and its eligible clinicians will bear performance risk through the performance-based incentive payments, since, as noted elsewhere, failure to meet agreed upon benchmarks would involve an APM entity repaying all or part of their incentive payments (depending on the level of performance). Since the APC-APM requires participants to only assume performance risk, we believe the model is feasible for small practices. Page 3 of 20 April 14, 2017 Patient Population. Based on the number of Medicare patients seen by primary care physicians in 2014, we estimate that more than 30 million Medicare patients would be impacted if the APC-APM were implemented nationally. To the extent the APC-APM is a multi-payer model, the actual number should be substantially more than that. The AAFP has experience working with commercial payers on multi-payer models (such as CPC and CPC+), and meets regularly with the largest national commercial health insurers on a variety of issues, including payment reform. APM entities would be responsible for reporting and performance on selected performance measures in the Core Quality Measures Collaborative’s Patient-Centered Medical Home (PCMH)/Accountable Care Organizations (ACO)/Primary Care Core Set. The collaborative’s measures are found in Appendix B. We believe that these performance measures will help ensure that patients receive necessary care and are not harmed by efforts to achieve savings. This is of particular concern given that APM entities with performance that does not meet agreed upon benchmarks face recoupment of their incentive payments and potential expulsion from the APM. We also think the APC-APM model supports patient safety by making patient choice the primary attribution methodology. Patients who do not believe that they are receiving the care they need may elect to leave an APM entity. Since the risk-adjusted, capitated primary care global fee and population-based payment (which will comprise a significant part of an APM entity’s revenue stream) follow the patient from practice to practice, APM entities will have an incentive to treat patients appropriately and deliver high-quality, coordinated care. Spending. As noted elsewhere in this proposal and based on research, the AAFP recommends that the percentage of total health care dollars spent on primary care be doubled.3 At the same time, based on experience in Rhode Island7,8 and other demonstrations,9 there is evidence that such an increase can be accomplished without an increase in overall spending on health care. In fact, evidence indicates increased spending on primary care will lead to a decrease in overall spending on a per-patient basis. For instance, using a simulation model, Reschovsky, et al. projected that a permanent 10% increase in Medicare fees for primary care ambulatory visits would result in a six-fold annual return on lower Medicare costs for other services, primarily inpatient and post-acute care.10 Accordingly, the AAFP estimates the overall anticipated impact on spending to be a net savings to the payers involved. The AAFP believes spending on primary care under the APC-APM should be increased from current levels, given the evidence that access to primary care is associated with improved individual and population-health outcomes and reduced costs. Today, primary care only represents approximately 6-8% of total spending on health care.11 We, and others, believe this should be increased to at least 12% of total spending.9 The AAFP believes that such an increase can be accomplished without an increase in the overall spending on health care. In fact, evidence indicates increased spending on primary care will lead to a decrease in overall spending on a per-patient basis.9 Page 4 of 20 April 14, 2017 This belief is rooted in the experience of other Organization for Economic Cooperation and Development (OECD) countries. Most of those countries have health care systems where primary care is foundational, and their spending per capita is well below that of the United States. Within the U.S., Rhode Island mandated an increase in primary care spending from 5.4% to 8% between 2007 and 2011.12 The Rhode Island Insurance Commissioner reported a 23% increase in primary care spending was associated with an 18% reduction in total spending—a 15-fold return on investment.12 Lastly, Portland State University completed a 2016 study of Oregon’s Patient-Centered Primary Care Home (PCPCH) program and found that every $1 increase in primary care expenditures as part of the PCPCH model resulted in $13 in savings in other health care services, including specialty, emergency room, and inpatient care.13 Cross-Payer Impacts. The APC-APM model, similar to CPC and CPC+, can be a multi- payer APM. We believe this is a strength of the proposal, as it can improve care quality and reduce costs for other patient populations. It may also help advance the movement to “Other Payer Advanced APMs” in 2021. III. Quality and Cost (High Priority Criterion) In this section, PTAC seeks input on ways the APC-APM would improve health care quality at no additional cost, maintain health care quality while decreasing cost, or both improve health care quality and decrease cost. The value proposition of the APC-APM is simple—we believe it will improve quality of care and outcomes and reduce overall costs (especially use of acute services), based on our analyses and early CPC results. Care Delivery Impacts. The APC-APM would improve care delivery and achieve savings or improve quality in manners similar to the CPC and CPC+ models, including but not limited to: • Reduced emergency department (ED) visits. Over the first three years of CPC, ED visits for Medicare FFS beneficiaries in CPC practices increased at a slower rate (2% less), relative to beneficiaries in comparison practices. The estimated effect on ED visits was a statistically significant difference.14 • Improved the quality of care among high-risk beneficiaries with diabetes at a statistically significant level.14 Barriers and Risks. Existing regulatory and administrative burdens that public and private payers impose on practicing family physicians are detailed in the AAFP’s Agenda for Regulatory and Administrative Reforms.15 Other barriers in federal laws and regulations that may prevent or discourage needed changes in delivery include, but are not limited to, provisions of the Stark Law and the requirement that patients must have a three-day inpatient hospital stay as a prerequisite for coverage of skilled nursing facility care. We believe that the proposed model can still have an impact, even if present regulatory barriers are not addressed. However, the potential impact will be diminished. Metrics. Each APC-APM entity will be evaluated based on reporting six measures, with one being an outcomes measure in order to align with MACRA’s MIPS reporting requirements. These measures come from the core measure sets developed by the multi-stakeholder Core Page 5 of 20 April 14, 2017 Quality Measures Collaborative to ensure parsimony, alignment, harmonization, and the avoidance of competing quality measures among all payers. These measure sets include patient experience measures. We note that all but one of the core measures are also measures under the MIPS. Thus, the APC-APM meets the quality measurement standard for an AAPM. Since primary care physicians treat a wide range of medical conditions for all patients, regardless of sex, age, or type of condition, the APC-APM will not need to develop any specialty-specific measures or other measures outside of those identified by the Core Quality Measures Collaborative. This approach would lead to streamlined quality measure reporting and assessment, and reduced administrative burden to physicians (especially small practices). Data Issues. The APC-APM embraces the use of data from multiple sources to provide a complete view about quality and cost performance. This ensures that both APC-APM participants and CMS may readily identify numerators, denominators, inclusions, and exclusions contributing to assessments of quality and cost performance, as well as factors likely contributing to outlier quality and cost data. APC-APM participant use of data from multiple sources will be encouraged. It will help identify new insights into potential care interventions for patient populations that offer the greatest potential to maximize efficacy and efficiency of care that result in positive health outcomes. APC-APM participants would work with their health IT vendors to generate timely and clinically actionable reports, including both practice- and provider-level data. Since a key objective of the APC-APM is continuous quality management toward value-based care and because quality improvements may lead to unintended and difficult to identify increases in disparities of care, the APC-APM will encourage participants to use social determinants of health data, to the extent possible. Social determinants of health data that include customized reports, analyses, and visualizations of performance and improvement activities can serve as a means of checks and balances. Data may show the potential for specific-care interventions, improvement activities, or use of technology, such as patient-specific education and secure messaging that could result in unintended consequences. For example, while quality measure scores increase, disparities of care can be seen to increase if social determinants of health data are also present within the same reports or data visualizations. Use of data from multiple sources is encouraged to more readily identify increases in disparities among vulnerable patient populations. Drilling down into all factors may be useful to identify an undesirable increase in the disparity of care. This could enable corrections, which minimize potentially negative patient outcomes and correct what would result in eventual increases in cost of care. Family physicians and other eligible clinicians benefit from timely performance feedback in order to adjust their performance or modify workflows. In this sense, the APC-APM will encourage participants to engage in electronic reporting more frequent than annual reporting.