Time to pursue patient-centered payment models designed by doctors

The physician community has advocated for the implementation of a diversity of payment models that fit various specialties and practice venues, remove barriers to innovation in care delivery, predictably reward quality and value, and attribute outcomes to those who can control or influence them. 

Stakeholders have expressed frustration that after front-line physicians with experience of the barriers to innovation in care delivery have put in years of work to develop patient-centered alternative payment models (APMs), implementation by government and commercial insurers has been slow.

While the COVID-19 pandemic has not led to more opportunities for participation in APMs, it did result in remarkably accelerated uptake of one care delivery innovation—telehealth. In 2020, physicians and health systems quickly deployed and expanded technologies to diagnose, treat, and counsel millions of patients through virtual encounters using telehealth technologies.

The rapid adoption of telemedicine illustrates the critical role of payment policy as both a barrier and potential catalyst for the uptake of care delivery reforms with known potential to improve value. While many medical specialties had been practicing telemedicine and developing an evidence base to support its use for decades, the 2020 expansion in access was made possible only when long-standing payment barriers were removed. 

Interestingly, this widespread, rapid, and impactful innovation in care delivery resulted not from any of the CMS value-based payments or APM models, but from changes to the Medicare fee-for-service physician payment schedule.

Patients needed access to telehealth services long before the pandemic began, but even most APMs hadn’t resolved the barriers keeping physicians from providing those services. This left many patients unable to follow up using digital technologies with physicians who knew them and felt that the appropriate standard of care could be delivered without in-person encounters—leading to delays in needed care and unnecessary cost and travel time.

Even CMS’ largest APM, the Medicare Shared Savings Program (MSSP), had not enabled physicians in most accountable care organizations (ACOs) to be paid for telehealth visits. Since most of Medicare’s other APMs are episode-based—triggered by delivery of a procedure or a major diagnosis—they also lagged in advancing access to telehealth, especially as nonemergent procedures were halted early in the pandemic.

The response by physicians to the removal of telehealth payment barriers was swift. Medicare alone spent $1.8 billion on physician telehealth services from March 16–June 30, 2020, compared with only $14 million in the first ten weeks of the year. Uptake was rapid because the need among patients for the services was real, and because physicians are quick to adopt innovations when there is an evidence-base and a path to support implementation.

Telehealth is by no means the only high-value, evidence-based, patient-centered service affected by payment barriers from Medicare and other health plans. For example, outpatient palliative care for many patients with serious illnesses remains inadequately supported despite its capacity to improve quality of life and prevent ED visits and hospitalizations.

Another example is intensive post-operative home rehabilitation, which is too often not supported under current payment systems even when it could avoid the need for more expensive inpatient rehabilitation.

In many cases, when physicians invest in implementing new systems to help patients stay healthy and avoid visits, procedures or hospitalizations, they face two financial repercussions. Not only are the innovative interventions unreimbursed, but for most interventions, the physician practice takes a second hit from the loss of revenue for avoided services. While rent, labor, equipment and other practice expenses typically don’t go down when innovating, inflexible payment systems compound the disincentive by penalizing revenue. These problems can occur even under shared-savings models.

Given the enormous chronic disease burden in the U.S., payment models must remove barriers to the innovation required to address the multiple epidemics of diabetes, hypertension, substance use disorder, and mental health diagnoses. While this has been the impetus for APMs, existing models implemented by insurers simply haven’t been designed in ways that work for most front-line physicians and their patients.

In Medicare’s population-based and episode- or condition-based APMs, the overwhelming drivers of payment adjustment have been shared-savings bonuses and shared-loss penalties based on spending targets, not quality of care. The high and increasing risk-sharing requirements have not only slowed participation in ACOs and other APMs, but—paired with a lack of up-front payments for innovative high-value services—have made it difficult for physicians to make significant changes in the way they deliver patient care under Medicare APMs.

Not only have Medicare’s APMs failed to support many of the innovative services that are needed to reinvent care delivery to improve value, but they haven’t even brought about the consistent and substantial cost savings that policymakers had hoped for. Some have even resulted in increased spending.

Some believe that APMs are falling short because physicians have too little financial risk, too few mandates to participate, or too many models. The experience of front-line physicians, however, is that APMs have demanded substantial financial risk while failing to remove the barriers to better care that exist in current payment systems.

Every day, front-line physicians experience frustrating examples of payment policies that stand in the way of improvements they know could reengineer care to improve quality and value. To implement the evidence-based, patient-centered innovations that are needed, we need to stop creating complex and punitive incentives and instead take a radically different approach to payment reform.

Successful patient-centered models focused on higher-quality, lower-cost care need these three key components.

Flexibility to deliver patient-centered care. To focus on outcomes, including avoidable future costs of care for individual patients and populations, physicians need flexibility to deliver a broad variety of evidence-based services to address patient needs. Payment gaps and prior-authorization requirements often stand in the way of services that would benefit patients and help reduce avoidable spending. 

There are extensive opportunities around telemedicine, care coordination, preventive care outreach, shared decision-making about treatment choices, teams to support disease management and avoid chronic disease flares and hospitalizations, outpatient palliative care, and nonhealth service provision such as transportation.

Adequate payments to support the costs of high-value care delivery. Most practices cannot assume population-based risk sharing with substantial possible penalties, so many APMs will need to involve episode- or condition-based payments, bundles or warranties to reach the largest number of patients. If practices participating in these models successfully keep their patients healthier and reduce the need for treatments, there ultimately will be fewer high-cost procedures and hospital admissions. 

However, the payment for those less frequent services will need to be increased—or the shared savings bonus made far more substantial—so that the practice still receives sufficient revenue to cover the fixed costs of treating the patients who do get sick. Those payments also must be reasonably predictable to allow investments in personnel and new workflows, and risk-adjusted to incent the care of patients who are sicker, have complex social determinants of health, and face other barriers to accessing care.

Accountability for delivering high quality services and avoiding unnecessary services. If physicians receive predictable, adequate payments that enable the delivery of high-value services, they will also accept accountability for the aspects of spending and quality they can control. However, they cannot be held accountable for things that are not dependably measured and attributed to the proper individuals or entities, or that they cannot affect—such as rising drug prices, or whether a patient has a caregiver who can help reduce nursing costs. 

Physicians are intrinsically motivated to achieve quality improvement, and desire information systems that put valid, actionable information about their performance in front of them at the point of care. They particularly respond to transparent measures that address areas of substantial harm or waste and were developed by peers, rather than black-box metrics delivered months or years after care has been provided.

The lack of front-line physicians in the development process of Medicare APMs has contributed to an absence of these three components necessary for progress and success.

For over a decade, the AMA has been advocating for the creation of more patient-centered payment models. We supported the creation of the CMS Innovation Center in the Affordable Care Act as a way of developing innovative approaches to payment, and we welcomed the early efforts by the resulting Center for Medicare and Medicaid Innovation (CMMI) to support physician-led efforts to redesign care delivery. 

We developed educational materials and held seminars all over the country to educate physicians about the opportunities to improve patient care under well-designed payment models and to help physicians understand how to design effective APMs.

Congress recognized that to be successful, APMs need to be designed by physicians working on the front lines of care. That is why it created the Physician-Focused Payment Model Technical Advisory Committee (PTAC) process in 2015 as part of the Medicare Access and CHIP Reauthorization Act (MACRA). The AMA also supported MACRA’s APM option based on the expectation that physicians would be able to play a leading role in the creation of appropriate APMs.

Physicians responded immediately to the opportunity to design APMs that would support better patient care. Over the past four years, more than 30 patient-centered payment models have been developed by physicians and submitted to PTAC, including APMs to improve care for patients with asthma, cancer, chronic kidney disease, inflammatory bowel disease, and other health problems and to deliver emergency care, home care, inpatient care, long-term care, palliative care, primary care, outpatient specialty care, and surgery to patients in higher-quality, lower-cost ways. 

Here are some great examples.

Project Sonar, developed by a gastroenterologist to reduce hospitalizations for patients with inflammatory bowel disease, was the first payment model recommended by PTAC four years ago. When adopted by a commercial insurance plan, the model improved patient quality of life, reduced hospitalizations, and decreased spending. 

While some commercial payers have expanded this model to create “specialty medical homes” for several serious chronic diseases managed by specialists, Medicare patients have not benefited from this improved way of delivering care because CMS hasn’t implemented an APM or made the necessary changes in the payment schedule to support it.

The Acute Unscheduled Care Model is another approach recommended by PTAC. It was developed by the American College of Emergency Physicians to increase the number of patients who could be sent home after an ED visit with enhanced care coordination rather than being admitted to the hospital. While similar pilots funded by grants have successfully reduced hospital admissions and repeat visits to the ED, Medicare has not adopted this APM so the approach cannot be replicated widely.

Making Accountable Sustainable Oncology Networks (MASON) is another APM recommended by PTAC. MASON was developed by an oncologist to support enhanced services to cancer patients that reduce the frequency of ED visits, hospital admissions and other services to treat complications of chemotherapy.  This was designed to build upon a CMS-funded project that successfully reduced Medicare spending while improving the quality of life for cancer patients, but most oncologists haven’t been able to replicate this success because CMS hasn’t adopted the APM.

PTAC has recommended 19 APMs for further development, testing, implementation or other action by Medicare. To date, however, CMS has not implemented a single one of the PTAC-recommended APMs, including those that previously demonstrated success using short-term CMS grant funds.

This has prevented millions of Medicare patients from receiving higher quality care and has caused Medicare spending to increase more than necessary. Moreover, most of these APMs would also have provided physicians with the flexibility they needed to deliver care in different ways during the COVID-19 pandemic.

Some have argued that Medicare has too many APMs, and that large, population-based, total-cost-of-care models should be the focus moving forward. Many diseases and patient populations are not directly addressed by these models, however, and most physicians practice in groups of 10 or fewer.

With important quality improvement work focusing on management of specific conditions and patients, and so much medical care continuing to be delivered outside of large, multispecialty health systems that can bear substantial financial risk, we believe that a one-size-fits-all approach will fail to maximize quality and cost improvement. 

It is certainly important that we focus the work of physicians, hospitals and insurers on a reasonable set of APMs that address diseases and populations where care delivery innovations can meaningfully improve quality and value. But thus far, we still lack models in which most physicians can meaningfully participate. 

More than five years after Congress passed MACRA, a large proportion of primary care practices in the country still cannot participate in Medicare primary care medical home payment models, and there are no Medicare APMs to support specialty care for most patients with chronic conditions. The episode-based APMs in Medicare have focused almost exclusively on inpatient procedures, even though most procedures are performed in outpatient settings. Frequently performed procedures such as cataract surgery, endoscopies, colonoscopies, and spine and joint injections still lack APMs.

While focusing improvement efforts on meaningful gaps is important, we need payment models that support innovation in ambulatory care as well as hospital-based procedures, diagnostic work as well as treatment, small practice settings as well as large ones, and primary care as well as specialty care.  Patient-centered APMs should not be viewed as competitors to ACOs that will fragment care, but as complements with the potential to remove barriers to care coordination.

In order to achieve higher-quality, more affordable care while addressing our nation’s chronic disease epidemics and unacceptable health disparities, we need to accelerate efforts to remove the barriers created by our current payment systems. 

The rapid expansion of telehealth access in the wake of COVID-19 showed us that the health system can innovate and adopt new systems rapidly when barriers are removed. Implementing a focused but comprehensive set of patient-centered, physician-designed payment models would be a win-win-win—delivering better care for patients, reducing spending for Medicare and other payers, and maintaining financially viable physician practices and hospitals to expand access to care.