Paying for low-value care

2018 Report to the Congress

MedPAC - Medicare Payment Advisory Commission, June 2018

Chapter 10: Medicare coverage policy and use of low-value care

Chapter summary (excerpts)

Some researchers contend that a substantial share of Medicare dollars is not spent wisely. Many new services disseminate quickly into routine medical care in fee-for-service (FFS) Medicare with little or no basis for knowing whether or to what extent they outperform existing treatments. In addition, there is substantial use of low-value care—the provision of a service that has little or no clinical benefit or care in which the risk of harm from the service outweighs its potential benefit.

In this chapter, we review the coverage processes used in FFS Medicare and by Medicare Advantage (MA) plans and Part D sponsors. FFS Medicare covers many items and services without the need for an explicit coverage policy. When an explicit coverage policy is required, some services do not show that they are better than existing covered services. Coverage policies are often based on little evidence and usually do not include an explicit consideration of a service’s cost-effectiveness or value relative to existing treatment options.

MA plans are generally required to provide the same set of benefits that are available to beneficiaries under FFS Medicare. However, MA plans are permitted to use tools that are not widely used in FFS Medicare, such as requiring providers to obtain prior authorization to have a service covered and controlling utilization through the use of cost sharing. Part D plan sponsors are responsible for creating and managing formularies, which are lists of drugs their plans cover. By contrast, Medicare FFS lacks the flexibility to use formularies for drugs that Part B covers.

We also review the literature on low-value care, which reveals that such care is prevalent across FFS Medicare, Medicaid, and commercial insurance plans. Evidence suggests that the amount of low-value care in a geographic area is more a function of local practice patterns than payer type.

We analyzed selected low-value services in FFS Medicare using 31 evidence-based measures. In 2014, there were between 34 and 72 instances of low-value care per 100 beneficiaries, depending on whether we used a narrow or broad version of each measure. Between 23 percent and 37 percent of beneficiaries received at least one low-value service, and annual Medicare spending for these services ranged from $2.4 billion to $6.5 billion. The spending estimates are conservative because they do not reflect the downstream cost of low-value services (e.g., follow-up tests and procedures).

We discuss six tools that Medicare could consider using to address the use of low-value care.

* Expanding prior authorization, which requires providers to obtain approval from a plan or payer before delivering a product or service, could help reduce the use of low-value care.

* Implementing clinician decision support and provider education could decrease low-value care.

* Increasing cost sharing for low-value services has the potential to reduce their use.

* Establishing new payment models that hold providers accountable for the cost and quality of care—such as accountable care organizations (ACOs)—creates incentives for organizations to reduce low-value services.

* Revisiting coverage determinations on an ongoing basis has the potential to both decrease use of low-value services and result in the development of more rigorous clinical evidence.

* Linking information about the comparative clinical effectiveness and cost- effectiveness of health care services to FFS coverage and payment policies has the potential to improve the value of Medicare spending. For most items and services, Medicare lacks statutory authority to consider evidence on cost-effectiveness in either the coverage or the payment process.

Chapter 10:

Fact sheet on the 2018 MedPAC Report to Congress:



By Don McCanne, M.D.

When concerns are expressed about the high costs of health care, frequently the excess use of low-value health care is cited as a major contributor to these high costs. MedPAC (Medical Payment Advisory Committee) in its 2018 report to Congress discusses its perception of the extent of the problem and potential policies that might reduce the overuse of low-value care. We should examine not only MedPAC's views but also the larger perspective of cost containment in relation to improving health care value.

First of all, how do you define low-value care? Most would agree that care that is of no value whatsoever should not be covered by public or private insurance programs. But low-value? That seems to imply that the care may have very limited value, but does that mean that it should not be covered if the patient has medical needs and there is no better care available? Where do you draw the line?

We could probably agree that we should exclude from coverage care that is confirmed to have no benefit at all, care that might have very limited benefit but clearly always has greater detrimental effects, or care that clearly is always inferior to other less costly care. But in many instances, care that might be considered of low value should probably be covered. Often we do not have enough data to know the precise value of care, but we should avoid moving into the treacherous zone of micromanaging every health care decision. Regardless, it is much more difficult to objectively and fairly exclude low-value care from coverage than many in the policy community suggest. This is true not only in the United States but in all other health care systems as well.

So what policies does MedPAC suggest to reduce the use of low-value care? Let's take a look.

* They suggest expanding prior authorization. Rather than being decided by those most acutely involved - the patient and the attending health care professionals - the decision is moved to a financial intermediary whose role is more often to protect the pooled health care funds that it is to improve resource allocation to benefit the patient. Not only does prior authorization waste administrative resources, it also results in a definitive decision when the potential benefit lies somewhere along a linear scale. What point do you select in that scale to come to an all or none conclusion on health care? That point should be pretty close to zero which then makes most of the prior authorization process a bureaucratic boondoggle. Besides, prior authorization is a significant source of patient discontent and a major contributor to health professional burnout - neither of which should be part of a high performance health care delivery system.

* They suggest implementing clinician decision support and provider education. More valid information is always helpful. Accumulating data and developing and improving patient care algorithms are always helpful. Some criticize this as "cookbook medicine," but that is nonsense. The more information the better. These protocols and algorithms are most appropriately used as helpful guidelines rather than as dictates of a rigid bureaucracy. This is what the science and art of medicine always has been and should continue to be in the future as we constantly refine our standards.

* They suggest higher patient cost sharing for low-value care in order to deter the patient from receiving such care. However it would be very difficult to quantify just what the cost sharing should be since low value lies along a linear scale rather than occupying one point. More importantly, cost sharing impairs access to care and if care is appropriate, even if of low value, it should be covered. Financial barriers should be removed so patients can receive the care they should have. Other more effective methods of cost containment that are more patient friendly should be used instead of cost sharing (e.g., reduction of administrative waste, administered pricing, etc.).

* They suggest establishing new payment models that hold providers accountable for the cost and quality of care, such as accountable care organizations (ACOs). We now have fairly extensive experience with accountable care organizations and other alternative payment models that demonstrate they they are not particularly effective. They have had only a minimal impact on spending, sometimes to the detriment of patients, and the quality parameters are too inadequate to objectively demonstrate any significant systemic quality improvement. More voices are now calling for an end to this failed experiment.

* They suggest revisiting coverage determinations on an ongoing basis. As new information regarding the value of care becomes available, continuing reevaluation of current practices is certainly in order so that changes can be made in recommendations that improve care and its value. If low-value care proves to be no value care then it should be eliminated from coverage, and vice versa.

* They suggest linking information about the comparative clinical effectiveness and cost effectiveness of health care services to FFS coverage and payment policies. This is where we have been too protective of the medical-industrial complex. We should increase our research in comparative effectiveness and use that information to improve patient care. We should also demand that high priced care and products be repriced at a level deemed to be cost effective. Although the level may be relatively arbitrary, it's clear that we shouldn't be paying half a million dollars for a health outcome gain that some may judge to be worth only ten cents. Cost effectiveness determinations should be used to terminate outrageous pricing.

So we can conclude that care that is beneficial, even if only modestly so, should be covered, and it should be priced right.

That leaves us with a decision of how we could accomplish the goal of improving value in health care without exposing patients to potential financial hardship though the misguided application of consumer-directed care. Single payer advocates know the answer. Instead of reducing beneficial health care services through financial barriers, we should recover the large amount of waste inherent in our health care financing system: profound administrative excesses, and excessive pricing.

By changing to a single payer national health program - an improved Medicare for all - we could recover a few hundred billion dollars in administrative waste. Since our health care prices are often too high, we could reduce them over time to levels that would provide adequate value while ensuring fair margins for the health care delivery system. The funds recovered would be enough to ensure that all of us would receive the health care services that we should have - better care and better value for everyone.

Stay informed! Visit to sign up for daily email updates.