CONTENTS

Why “Pay for Performance” Hasn’t Worked

The most common approach to value-based payment has been “pay for performance” (P4P). Under this approach, a healthcare provider may receive bonus payments if it performs better than other providers on a set of payer-defined measures of quality, utilization, or spending, and it may have to pay penalties or have its payments reduced if its performance is lower than what other providers achieve on these measures. Most physicians, hospitals, and other healthcare providers now participate in at least one pay-for-performance system because of the P4P programs created in Medicare, such as the Merit-Based Incentive Payment System (MIPS) for physicians, the Hospital Value-Based Purchasing Program, and the Skilled Nursing Facility Value-Based Purchasing program. The Centers for Medicare and Medicaid Services (CMS) describes this as “Category 2: Fee for Service with a Link to Quality and Value” (in contrast to “Category 1: Fee for Service - No Link to Quality & Value”).

This approach has been unsuccessful in improving the quality of care for a number of reasons:

  • There are no new fees for services that could improve the quality of care. Physicians, hospitals, and other healthcare providers are still paid only for the same types of services they are paid for under the standard fee-for-service system, and they still lose money if they deliver care in different ways or help patients stay healthier.
  • Payments fall short of the cost of delivering quality care. Even if a healthcare provider qualifies for a performance-based payment, the amount is typically too small to make up the shortfall between fee-for-service payments and the cost of delivering high-quality care. Moreover, the administrative burdens associated with quality measurement can cause a provider’s costs to increase more than the additional revenue it receives from performance-based payments.
  • The measures used do not accurately or completely assess the quality of care delivered. Quality is assessed based on whether the care for a patient met a general standard of quality, even if meeting the standard would have been undesirable or harmful for that particular patient. Moreover, because quality measures are only applicable to a narrow range of health conditions and services, there is no measure of quality at all for many types of health problems and patients.
  • A healthcare provider can be penalized for reasons outside of the provider’s control. For example, if a patient is unable or unwilling to use all of the services needed to achieve a good result on the quality measure (e.g., the patient cannot afford the medications needed to treat their diabetes), the provider will be scored as having failed on the measure for that patient and the provider’s payments may be reduced, even though the provider had no control over the factors affecting that patient’s adherence. In many cases, a provider who treats a patient for one health problem can be penalized based on the quality of care that unrelated providers deliver for completely different health problems.
  • There is no assurance that each patient will receive high quality care. A healthcare provider is paid for delivering an inappropriate or poor quality service to a patient regardless of how the provider scores on quality measures. In fact, since performance-based payments are based on the percentage of patients whose care met the standard in the quality measure, a provider would be paid more for delivering poor-quality care to an individual patient if the provider has higher-than-average quality scores for its other patients.
  • The payments discourage collaboration in care improvement. In many P4P programs, such as Medicare’s Merit-Based Incentive Payment System (MIPS), a provider can only receive a bonus payment for good performance if other providers have been penalized for poor performance. This discourages collaborative efforts to improve care, because if a high-performing provider helps other providers to improve, the high-performer will receive a smaller bonus.

It is often suggested that P4P systems have been unsuccessful because the “incentives aren’t large enough,” but the real problem is that P4P doesn’t actually solve the problems with fee-for-service payment. Moreover, the simplistic quality measures used in these systems can discourage providers from delivering care to disadvantaged patients who have complex needs or difficulty adhering to standard approaches to treatment.

Fortunately, there is a better approach to value-based payment: Patient-Centered Payment.

Back to Top © Center for Healthcare Quality and Payment Reform www.CHQPR.org

The Failure of Shared Savings and Shared Risk

Value-based payment systems have increasingly focused on reducing healthcare spending, rather than improving the quality of care. The most common approach has been “shared savings” payment models. In these programs, a healthcare provider continues receiving standard fees for the services they deliver, but the provider is eligible to receive a bonus payment if the payer determines that total spending on the provider’s patients has decreased or if spending has increased less than the payer expected. A growing number of these models now require “downside risk,” i.e., they require the provider to pay penalties if the payer’s total spending increases or increases by more than the payer expected.

Most of the “alternative payment models” created by the Centers for Medicare and Medicaid Services (CMS) are designed this way, including the Medicare Shared Savings Program, the Bundled Payments for Care Improvement (BPCI) program, and the Comprehensive Care for Joint Replacement program. Providers participating in these programs continue to receive standard fees for individual services, so the payment systems are really just a variation of the pay-for-performance concept. The primary difference from typical P4P programs is that the focus is on reducing spending rather than improving quality, and the amounts of the bonuses and penalties are proportional to changes in spending. This is why CMS describes these payment systems as “Category 3: Alternative Payment Models Built on Fee-for-Service Architecture.”

In general, shared savings and downside risk models have failed to significantly reduce healthcare spending or improve the quality of care because they do not make any fundamental changes in the way healthcare providers are paid:

  • There are no payments to support the delivery of new or different services that could reduce the use of more expensive services. There are many opportunities to reduce healthcare spending in ways that do not harm patients or even improve outcomes. In most cases, this can only be done by delivering a different set of services to patients, but those services cannot be provided today because there are no payments to support them under current fee-for-service systems. Since there is no change in the way providers are paid in shared savings and downside risk programs, they cannot afford to do what is necessary to achieve savings without harming patients.
  • Payments fall short of the cost of delivering quality care. Even if a healthcare provider qualifies for a shared savings payment, the amount may be too small to cover the cost of the services needed to achieve the savings. Moreover, hospitals and other types of providers incur significant fixed costs to deliver essential services, and as a result, a reduction in the number of services delivered reduces their revenues more than their costs. The shared savings payments may be too small to offset the resulting losses.
  • There is no assurance that each patient will receive high quality care. A healthcare provider continues to be paid for delivering a service that is inappropriate or of poor quality. There is no penalty for low-quality care other than a reduction in the bonus payment if savings are generated.
  • It is still impossible to compare providers based on the cost of treating health problems. Although shared savings and downside risk payment systems typically calculate an “expected” amount of spending for a procedure or for all of the services delivered during a particular period of time, these calculations are generally made after services are delivered, rather than in advance. Even if the expected spending is defined in advance, the complex calculations needed to determine the bonuses and penalties for providers can generally only be performed after services have been delivered, so the total amount that will actually be paid will not be known in advance.

Of greater concern, however, is that this approach to payment has the potential to harm the quality of care for patients, make it more difficult for high-need patients to access care, and force small healthcare providers out of business:

  • Providers are rewarded for withholding services the patient needs. Under a shared savings payment model, if a physician does not order a test, procedure, or medication for a patient, that is considered “savings” regardless of whether the patient needed the service or not, and the physician can receive a bonus payment. The small number of simple quality measures used in these payment models do nothing to protect against most potential types of undertreatment.
  • Providers can be penalized if they care for higher-need patients. Patients with higher needs will generally need more services and more expensive services, and the risk adjustment systems used in shared savings models do not adequately address this. The more high-need patients that a provider cares for, the higher the spending that will be assigned to that provider, which will reduce the likelihood of receiving a shared savings bonus and increase the chances that the provider will be financially penalized.
  • Providers can be penalized for factors beyond their control. The spending measures used in these payment models generally include spending on all services the patients receive, not just services delivered or ordered by the providers who are at risk, and those providers may have little or no ability to influence whether and how the other services are delivered. Moreover, spending can increase due to increases in the prices of drugs and other services that are beyond the control of the providers in the shared risk payment model. Small physician practices and hospitals do not have adequate financial reserves to pay for these costs and may be forced out of business.
  • Money is spent on avoiding losses instead of delivering patient care. Providers participating in downside risk models have to spend large amounts of money on data systems, stop-loss insurance, risk score maximization, and other activities designed to increase the chances of receiving bonuses and reduce the likelihood of paying penalties, rather than on improving care for patients.

Fortunately, there is a better approach to value-based payment: Patient-Centered Payment.

Back to Top © Center for Healthcare Quality and Payment Reform www.CHQPR.org

The Problems With Population-Based Payment

“Population-based” payment systems do not pay fees for individual services, but instead pay a healthcare provider a fixed amount each month to provide a patient with as many services as the patient needs. The payments are referred to as “population-based” because the total amount of revenue a provider receives depends on the number of patients the provider is caring for, not how many services or what types of services are used to treat the patients.

Population-based payments are fundamentally the same as traditional capitation payments except that (1) the payment amounts may be higher for individual patients who have more chronic conditions, and (2) the average payment amounts may be adjusted up or down based on measures of quality or spending similar to pay-for-performance and shared savings programs.

A single monthly payment for each patient gives a healthcare provider greater flexibility to deliver different services and a more predictable revenue stream than paying fees for each individual service delivered, and it also makes it easier for a patient or payer to predict how much they will need to pay for care. However, this approach fails to solve all of the problems with current fee-for-service payment systems, and it also creates new problems that do not exist under fee-for-service:

  • There is no assurance that patients will receive care when they need it. A strength of fee-for-service payment is that a provider is only paid if they deliver a service to a patient. In contrast, under population-based payment, the provider is paid regardless of whether a patient receives the services they need.
  • Payments may not be sufficient to cover the cost of delivering high-quality care. Most proposals for capitation and population-based payments set the payment amounts at levels designed to generate the same amount of revenue the provider has been receiving from fee-for-service payments, and in some cases less. This means that if the fee-for-service revenue was inadequate to cover the cost of high-quality care, it is likely that the capitation payments will also be inadequate.
  • Providers are rewarded for withholding services the patient needs. Capitation payments are the same regardless of how many services are delivered or what types of services are delivered. However, the provider will incur higher costs when more services are delivered, and the provider will save money when fewer services are delivered, so even if the payment is large enough to support all of the services the patient needs, the provider’s profits will be higher if fewer services are delivered. The small number of simple quality measures used in these payment models does nothing to protect against most potential types of undertreatment.
  • Providers can be penalized for providing care to higher-need patients. One of the strengths of fee-for-service payment is that providers will be paid for providing additional services to patients who have more healthcare problems. In contrast, typical population-based payment systems make no adjustments in payment for patients with new chronic conditions, for patients with multiple acute problems, or patients who need more or different services because they face non-medical barriers to care. This can penalize providers who care for high-need patients.
  • Providers can be penalized for factors beyond their control. In “global” or “comprehensive” population-based payment systems, a physician practice or health system is expected to pay for all of the services a patient needs from the fixed population-based payment. The provider may have little or no ability to control many of the factors that affect the cost of those services, such as increases in the prices of drugs or services that have to be delivered by other providers or health systems. This can further discourage caring for patients who have higher needs and create even stronger incentives to withhold necessary but expensive treatments.
  • Investors, vendors, and financial intermediaries can profit at the expense of patient care. Because of the high levels of financial risk associated with population-based payment systems, healthcare providers are forced to pay vendors and consultants to help them identify ways to maximize profits and minimize losses, which reduces the revenues available for healthcare services to patients. Private investors and financial intermediaries can make bigger profits on the fixed capitation payments if they can find ways to cherry-pick patients and increase the risk scores assigned to patients, rather than by improving the quality of healthcare services.

Fortunately, there is a better approach to value-based payment: Patient-Centered Payment.

Back to Top © Center for Healthcare Quality and Payment Reform www.CHQPR.org