Interpreting National Health Expenditure Projections: Issues And Challenges

Health Affairs today published the projections for health spending over the next decade from the Centers for Medicare and Medicaid Services (CMS) Office of the Actuary. The top line estimate is that health spending will grow at 5.5 percent per year through 2026. This rate is about halfway between the pre-recession rate of 7.3 percent and the exceptionally low rate (3.8 percent) experienced during the recession and immediate aftermath. This projected spending growth is 1 percentage point above expected gross domestic product (GDP) growth, a smaller gap than for almost any 10-year period since 1990. These non-partisan, thorough projections are a valuable benchmark for all stakeholders anticipating the fiscal footprint of the health care system on the economy, but there are several important issues to keep in mind.

Modeling What Spending Would Be Under Current Law, Not What It Will Most Likely Be

First, these projections are predicated on “current law”. The authors are not trying to predict what spending will most likely be. Such a prediction exercise would require assessments of how policy may change. For example, will the low-fee trajectories called for by the Medicare Access and CHIP Reauthorization Act (MACRA) and the Affordable Care Act (ACA) productivity adjustments be realized? What will be the future of the ACA? The authors here don’t attempt to incorporate the answers to these questions. They assume that the current law will persist. Last week’s budget legislation, which included the repeal of the Medicare Independent Payment Advisory Board and other health care changes, illustrates this point because it came too late to be included in these projections. This highlights how quickly and unpredictably law and policy can change.

An Uncertain Environment

Second, as the authors recognize, there is considerable uncertainty around these projections. During the recession, we experienced a dramatic slowdown in the rate of growth in “use and intensity” of care, a catch-all phrase capturing more physician visits, hospital stays, lab tests, etc. As the ACA was implemented, use and intensity rebounded, capturing both a return to a more common rate of growth and an increase attributable to coverage expansion. (If there is one thing we know well it is that greater coverage generates greater utilization.)

The assumption moving forward is that use and intensity will revert to historical patterns, affected a bit by benefit design changes. The impact of payment reforms, in both the public and private sector, is largely not reflected in these projections. That assumption is certainly reasonable given the modest impact of such changes to date, but payment systems continue to change and their impacts may grow, suggesting that perhaps use and intensity growth will be lower than projected.

Of course, uncertainty is not one sided. Other hypotheses, such as greater introduction of new technologies or weakening commitment to controlling utilization, would yield higher spending projections.

The Policies We Choose Matter

Third, and perhaps most importantly, the actual rate of health spending growth that we experience over the next decade, and beyond, depends on what we do. Health spending is not a natural phenomenon to be predicted like the tides. Our fate is not sealed. Our actions matter. We should not ask whether health spending growth will accelerate (or not). Instead we should ask if we will let it accelerate (or not). This requires complex choices.

It is tempting to read the projections such as those by Gigi Cuckler and her CMS colleagues with alarm. The notion that health care will consume almost 20 percent of the economy in 2026 is legitimately concerning because of the implications for future taxation or borrowing to maintain publicly financed health insurance programs. In fact, Kate Baicker and Jon Skinner predicted back in 2011 that if public health spending growth consistently exceeded national income growth by 1 percentage point and was financed by taxes (increased proportionately on all income groups), the tax rate for the upper income bracket in 2060 would need to rise to 70 percent. This is unlikely to happen, but we must act to make sure it does not.

The Dual Nature Of Health Spending: Consumption And Investment

The challenge, of course, is that health spending, on average, improves health. Therefore, actions to restrain resources devoted to health—including restricting access to health care or health insurance—run the risk of slowing or reversing improvements in health. Moreover, apart from the obvious benefit associated with access to health care services, many people rely on the health care sector for jobs. Heath care is in many ways both a tapeworm on the American economy and a Keynesian stimulus— cutting health spending will have economic consequences. Moreover, while creating jobs is not a justification for waste, lower spending growth can imply lower revenue growth for health care stakeholders, which presents a significant political problem.

These two perspectives are not as hard to balance as one may think. We do not need to cut heath care spending below current levels, just slow the rate of its growth. Moreover, resources not absorbed into the health care sector can be put to valued activities. (If there are not more valued activities for these incremental resources, then more spending on health care is justified).

Beyond How Much We Spend: Does It Bring Value And How Will We Finance It?

Finally, all of this highlights the heterogeneity in value associated with greater health spending. We currently pay higher prices in the United States than in other countries. While health care price inflation slowed between 2014 and 2016, the projections assume that price growth in health care will exceed general inflation. While we cannot conclude that prices in the United States should match those in other countries (there are many differences in our economies), much of the reason for high and growing prices in the United States is a lack of competition associated with consolidation in the health care sector (largely in the commercial sector) and other institutional features. Paying too much for care or other services not only distorts behavior, but also represents a transfer from the broad population to the health care sector. Policy actions to address this issue should be high on the agenda.

Use and intensity also varies in value. While innovation and delivery of appropriate care is the centerpiece of a high-value health care system, our system too often provides care that offers little or no health benefit. Great strides have been made to quantify low value care, through initiatives such as Choosing Wisely, but a lot remains unmeasured. Eliminating such care is hard because the value of care depends on patient traits and delivery system reform requires motivation of influential and invested stakeholders, which does not occur rapidly. Nevertheless, efforts to move in this direction are important.

Ultimately, the question we should ask when we ponder projections of higher health spending is this: Will we get enough value for the added spending? We should also ask how we should finance this spending: Any way we do so—through taxation, higher premiums or cost sharing at the point of service—has dramatic distributional and moral implications.

There are no simple answers to these questions, politically or operationally. One thing is clear: We cannot continue to publicly finance health spending if it grows 1 percentage point faster than GDP. We are not in crisis yet, but at this rate, eventually we will be. Yet, changing health care takes time. Many innovations in both payment and benefit design show promise, but success is uncertain. With luck (or more importantly dedication and hard work) we will be able to spend less than these projections suggest and maintain, or even improve, the care delivered to Americans. Unless we keep trying, however, we will fail.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s