Population Health Management and ACOs: Will They Achieve Their Goals of Better Health and Lower Costs?

What are the elements of a successful population health management program? What results have been reported for ACOs to date?

Having been in practice for over 36 years, I have seen many technological advances in the diagnosis and management of diseases; much has been taken for granted. We have seen the dissolving of a code of medical ethics that opposed advertising and solicitation of patients, replaced by a system of so-called free-market competition, clinical integration, and the right of insurers and hospital systems to choose their own providers. During this managed care revolution, we have seen reforms that have actually increased costs and are responsible for much of the frustration and anxiety that is so prevalent today among practicing physicians and many patients as well. One thing that remains constant is the patient in need of a competent and compassionate physician who can spend the time to listen to their complaints.

We are entering another era of change in medicine reimbursement as the concepts of population management and Accountable Care Organizations (ACOs) enter the picture. Many feel this is a continuum of what started with the HMOs in the 1990s failing to control costs. The U.S. has by far the highest health care cost of any Western industrialized nation without any discernible improvement in life expectancy or other measures of health care. Many say this is the result under managed care with the formation of hospital and insurance company monopolies, which like all monopolies, bring increased costs.¹

Whether we like it or not, we are entering this phase and must attempt to improve the quality of care and reduce cost at the same time. I will define what population management is and the desired role of ACOs in trying to achieve the goal of delivering “value-based payment (VBP).”

Value-based care (VBC) or VBP has been broadly adopted in the private market by both commercial insurance payers and employers. According to the Catalyst for Payment Reform’s 2014 National Scorecard on Payment Reform,² about 40% of payments by private plans to health care providers were based on quality and performance measures. This represented a significant increase in adoption from 2013, when a scorecard found that only 11% of payments were value-based. Assuming that regulations under the Medicare and CHIP Reauthorization Act (MACRA) regulations continue unchanged, VBP will soon be implemented in some form for nearly all Medicare reimbursement, regardless of what happens to the Affordable Care Act.

MACRA has two payment tracks, MIPS and the advanced alternative payment models (APMs). Both tracks meet the broad definition of VBP. This means that by the end of 2017—absent regulatory changes due to administration or congressional action—all Medicare Part B services will be reimbursed through value-based payment of some sort. By the end of 2018, the Department of Health & Human Services has a goal of devoting 50% of Medicare spending toward value-based reimbursement. However, Kaiser reports most physicians are not participating in risk-based contracts, but I suspect this will increase as the number of ACOs grows. Medicare payment reform gives preferential treatment to the APMs created by the ACA Innovation Center and the Centers for Medicare and Medicaid Services (CMS), who have made changes to allow more participants to qualify as advanced APMs.

The development of ACOs in response to the transition of payments to the VBP model is the central piece necessary for population health management to be implemented. The Medicare Shared Savings Program (MSSP) was established by the ACA to facilitate improved care and reduce unnecessary costs.

There are three types of ACOs: physician-led, hospitalled, and hybrids, which include both physician and hospitals. A study published in the New England Journal of Medicine, as well as one from KPMG, found that ACOs led by independent primary care groups typically saw greater savings than hospital-led ACOs.³ “ACOs have shown an ability to reduce hospital admissions and health care spending, which is why Congress specified ACOs to be part of MACRA and why CMS has given them a prominent role in achieving its overall goals,” says Tim Gronniger, deputy chief of staff at CMS.

The ACO program has evolved in the last five years, and all the changes have been based on feedback resulting in moving from 6 models to 10 in MACRA-proposed rules for 2018.

Defining Population Health Management

Health care reform is feeling the shift away from fee-for-service models toward pay-for-performance, VBP paradigms. And in order for health care organizations to successfully transition, there is an acute need for actionable analysis of data from individual patients and populations.

Multiple definitions for population health management abound. Some feel population health management should be defined the same way public health was defined years ago by C.E. Winslow, founder of the Yale Department of Public Health, as “the science and art of preventing disease, prolonging life, and promoting health through the organized efforts and informed choices of society, organizations, public and private communities, and individuals.”⁴

Eighty percent of what affects health outcomes is associated with factors outside the traditional boundaries of health care delivery—health behaviors (tobacco use, sexual activity), social and economic factors (employment, education, income), and physical environment (air quality, water quality).⁴ When health care delivery systems expand their interactions with patients to these territories—now the purview of the public health system—outcomes will improve.

Population health management (PHM) is the aggregation of patient data across multiple health information technology resources, the analysis of that data into a single, actionable patient record, and the actions through which care providers can improve both clinical and financial outcomes. The ultimate goal is to improve the health outcomes of the group by monitoring and identifying individual patients within that group, using a business intelligence tool to aggregate data and provide a comprehensive clinical picture of each patient. This allows the tracking of these patients and focusing on the sickest (such as those with diabetes, vascular disease, COPD, hypertension, heart failure or coronary disease), which hopefully will improve the clinical outcomes of these patients while lowering the cost.

EMR systems currently on the market are designed for a fee-for-service world, running entirely on encounter-based medicine. This makes it difficult to manage the health of populations of patients—and difficult to understand the cost of care. Fundamentally, in a population health environment, a health system is managing to margins on a per-member, per-month (PMPM) basis. And in this environment, everyone has to be aware of the cost of care at the point of care. This is not possible without major changes to current EMR software.

Elements of a Successful Population Health Management Program

A successful and well-developed PHM program will give real-time insights to both clinicians and administrators that allow them to identify care gaps within the patient population. This is the key to better outcomes in cost savings, especially in the population with chronic disease. A critical component of PHM is care management; while the objectives of care management can vary from organization to organization, they tend to revolve around improving patient self-management, improving medication management, and reducing the cost of care—such as admission rates to hospitals. In VBP, there is greater financial reward in preventing illness than there is in treating disease. Reducing expenditures can be accomplished by identifying preventable hospitalizations, ED visits, and preventable readmissions. This idea was recently put to the test in the Pioneer ACO program, which scored high for preventative health care measures, generating $147 million in total savings for these initiatives.

All of this leads to population health management in new form. To get there, the industry has to be smart about accelerating development of the right areas and manage expectations about what can be achieved.

For one, the industry needs to do a better job of collecting true patient outcomes data, rather than proxies for care. For example, it inherently does not matter if a patient with diabetes has had a foot exam—but it matters very much if that foot exam discovers an open wound that will not heal or a severe neurovascular deficit identified to work on ways to prevent the ulcer from developing.

Additionally, organizations must also understand cost at a granular, patient level instead of guessing at costs by looking at the average cost for all patients. That approach is almost meaningless when managing margins.

The Population Health Equation and Return on Engagement

The fundamental, traditional Health Care Value (HV) equation is defined by the Quality of Care (QoC) plus Experience (E) divided by the Cost of Care (CoC), or HV = (QoC+E)/CoC. The fundamental equation of population health is the calculation of Return on Engagement (ROE): the Clinical Outcome Achieved (COA) divided by the Total Patient Investment in a Patient’s Health by the Health Care System (TPI), or ROE = COA/TPI.⁴

Described otherwise, “How much does it cost our population health management system to increase a patient’s clinical outcome by one unit of measure?” This is why the understanding of costs and patient outcomes is so fundamentally critical to the success of population health.

The motive behind a care management system—services and software—is to reduce the investment (TPI) necessary by the health care system to achieve a unit of improvement in clinical outcomes (COA) by engaging patients in both the numerator and denominator of their own health.

As provider groups assume more financial risks, there are ACO constraints by the Medicare legislation which limit the ability for patients to choose to exclusively use an ACO, as well as what can be done with regard to benefit redesign. A rogue patient that smokes and eats whatever one wants and does not care about BMI or blood sugar basically ties the physician’s hands when it comes to quality measures. Since a Medicare ACO beneficiary is “not captive” as in private insurance, they can choose to seek care anywhere outside their assigned ACOs. This places that ACO at the mercy of overspending at another facility not involved in the ACO. Until patients have a monetary stake in their health or penalties if they go outside the ACO system, ACOs will be unable to maximize their results in quality and cost savings.

Evaluating ACO Success

Whether ACOs are living up to the lofty expectations of health care executives and federal policymakers is a matter of rising debate. Saving money while raising the quality of health care has been the central focus of the health care industry—from providers to health insurance carriers—since the arrival of the ACA. Yet, the initial costs of these programs were substantially more than the government anticipated. Kaiser Health News reported that 45% of ACOs cost more than anticipated, and in 2014, MSSP had a net loss of $2.6 million.⁵ Some 40% of the participating sites ended their involvement in the program after the first year, they said.

Proponents contend that the early evidence points to increased care quality. But detractors counter that ACOs have failed to produce adequate cost savings and efficiencies, according to commentaries published in the Journal of the American Medical Association.

Kevin Schulman, MD, of Duke University School of Medicine and Harvard Business School, and Barak Richman, of Duke University School of Law, said they aren’t impressed with the early findings, pointing to a trio of published evaluations that suggest less-than-ideal efficiencies generated by ACOs.⁶

The first, a comprehensive evaluation of Pioneer ACOs, showed savings in the cost of care to the CMS in 2012 of 1.2%. But after accounting for bonus payments, net savings to CMS were a paltry 0.4%, and 40% of the participating sites ended their involvement in the program after the first year, they said.

A second analysis by CMS showed smaller increases in total Medicare expenditures for ACO-aligned beneficiaries, but the result was substantially reduced by the second year of the program and didn’t include incentive costs to ACO participants.

The third report, an evaluation of the CMS Medicare Shared Savings Program, found that the 2012 ACO cohort showed a lower cost of care but triggered CMS bonus payments that exceeded those savings. Then, the 2013 cohort didn’t save money at all.

Furthermore, none of those evaluations, Schulman and Richman said, considered the costs to implement the ACO. They concluded that of the several iterations of ACOs, none have meaningfully reduced the cost of care, which begs the question whether the ACO concept itself is sound. This is an important consideration, they said, because ACOs carry a substantial cost to those outside Medicare.

Since this results in a lack of negotiating power on the part of private health plans, Schulman and Richman said that two strategies have emerged to maintain affordability in the health insurance market: high-deductible health plans, and narrow-network health plans with significant cost sharing. Co-payments for care under these offerings are often unaffordable, leading the authors to conclude that the ACO model may have weakened the financial protection that is the core purpose of health insurance, while undermining the ACA’s goal of expanding coverage.

Schulman and Richman suggested paying physicians and physician-led groups to keep patients away from and out of hospitals, away from costly facilities and tests, and use inpatient services only when other low-cost mechanisms are not effective.

In their view, the ACO model has also accelerated the trend of hospitals acquiring physician practices, enhancing their market power. Consolidation is responsible for sharp price increases across markets within states, they argue, with monopoly hospitals, those that dominate a local market with no real competition, showing 15.3% higher prices than hospitals in competitive markets.

Seeing the matter differently are Zirui Song, MD, of the Department of Medicine at Massachusetts General Hospital, and Elliott Fisher, MD, of the Dartmouth Institute for Health Policy and Clinical Practice. With limited evidence of the effects of ACO contracts, and most studies containing fewer than two years of follow-up, they said, reversing course so early in ACOs’ lifespans would be a mistake—particularly since evidence on quality has been promising.⁷

Quality improved in both process and outcomes in the Blue Cross Blue Shield of Massachusetts Alternative Quality Contract, they said, with similar gains for Medicare beneficiaries in ACOs. For example, process measures included adherence with guidelines for screening for colorectal or breast cancer, and follow-up testing for patients with diabetes or coronary artery disease; outcome measures included the proportion of patients with diabetes achieving target blood pressure, lipid levels, or hemoglobin A levels. In both populations, bonuses for ACOs were highly contingent on achieving gains in quality.

Medicare beneficiaries also reported better access to care, and for those with multiple chronic conditions, much higher subjective ratings of the overall quality of care. That evidence, Song and Fisher contend, is grounds for “cautious optimism.”

What’s Next?

CMS is developing “refinements” to overcome these ACO early failures, culminating with the new APM models of higher levels of financial risk and reward to the providers. They maintain that opportunities for making the model more attractive involve supporting a gradual transition to reasonable risk-bearing that offers greater financial rewards for physicians. I would advocate a transition to enterprise liability, wherein the ACO, not the individual physician, would be liable for malpractice claims.

More clarity is needed to determine ACOs’ efficacy, be it failure or success, and the only thing that can deliver that clarity is time.

These models focus not on historic performance but more on current performance efficiency at managing care, not for “picking low-hanging cost-cutting fruit.” Physician behavior must change in this approach to care, and hopefully not physician practice to deliver the best of care.

Will these “modern” pay-for-performance models support previous failed studies in England and the U.S. that cost, efficiency, and quality are not improved? Are they the same old wine in new bottles? Will they be rejected again as the “gatekeeper” perception was in the 1990s? Will the Medicare Advantage Plans continue to be more costly and more cuts to providers occur in the future as the insurance companies continue to take their 15% cut? Should the patient be part of “cost responsibility” when they do not lose weight, continue to smoke, fail to take medication, ignore ACO guidelines in utilization of the ACOs services, or utilize their PCP first before needless ER visits?

The Physician’s Foundation survey⁸ continues to show physicians are demoralized and dispirited in a dysfunctional health care delivery system. Policymakers might benefit from consulting and listening to those who actually treat the patient, the practicing physician, and not solely to those like the Dartmouth Group.

Biography

J. Collins Corder, MD, FACP, MSMA member since 1980, has been practicing internal medicine for 36 years at Missouri Baptist Medical Center in St. Louis and is affiliated with BJC Medical Group. He was the 2017 president of the St. Louis Metropolitan Medical Society.

Contact: jcorder20@aol.com.

Reprinted with permission from St. Louis Metropolitan Medicine.