New health spending data for 2014 and spending projections over the next decade from the Centers for Medicare and Medicaid Services (CMS) Office of the Actuary were just published in Health Affairs. They show that total growth in health spending picked up in 2014; this was expected given the significant expansion of insurance coverage and the release of expensive new drugs for hepatitis C.¹
But all of the evidence points to continued modest growth in per capita/enrollee spending. This low growth in per enrollee costs is a strong signal that we may be in an era where the “new normal” is more restrained growth in the use of health care services, even as the economic recovery continues and rates of uninsurance decline.
What Are The Estimates?
In brief, the Office of the Actuary estimates that national health spending growth in 2014 returned to a rate of over 5 percent, a growth rate not seen since 2007. However, their longer-term health care spending growth projections continue to be pegged at 1.1 percentage points faster than economic growth, a rate well below the historical average.
In addition, 2014 growth per capita remained below 5 percent, even without accounting for the aging of the population. That’s considerably lower than the historic average of over 7 percent. Keeping age-adjusted per capita growth in line with overall economic growth is the holy grail of health cost control: when the spending growth rate rises above the overall rate of growth in the economy it is termed “excess” cost growth and health consumes an increasing share of national output.
This is not to say that insurance coverage rates are not important. Overall insurance coverage rates and the portions of the population covered by different insurance programs have changed dramatically over the past few years as Medicaid expanded, more people became insured through exchanges, and the baby boomers began aging into Medicare. Given this background, per capita growth—ideally age and health status adjusted—gives the clearest indicator of the growing cost of care delivery and the changes in our health care system.
How Do The Estimates Differ By Payer?
Underneath the overall growth rate there are major differences in growth rates by payer, both overall and per capita. Overall and per capita growth rates by payer from 2012 to 2014 are shown in Exhibit 1 below. Due to population growth, overall growth rates exceeded per capita rates during this period and both were considerably below historic averages. In addition, even in years like 2012 when overall growth rates across all payers were approximately 4 percent per capita, growth rates ranged from CMS’ new estimate of 0.0 percent for Medicare to 3.7 percent for private insurance.
The 2012 flat rate of growth per capita in Medicare was remarkable, even if it was attributable in part to baby boomers pulling down the average age of the population. The most extreme difference between per capita rates and overall rates is seen in 2014 for Medicaid: overall spending grew dramatically because of enrollment growth, but per capita spending growth was actually negative due to a healthier-than-average expansion population.
Exhibit 1: Overall and per capita growth rates by payer from 2012 to 2014
In the absence of detailed adjusted per capita costs, looking at trends in the health care marketplace can explain some of these differences by payer. Exhibit 2 shows long-run trends in actual and projected per capita cost growth by payer from a supplemental table that was also just released by CMS. Three things are striking:
- The sharp decline in growth rates over the past five years.
- The variability in growth rates by payer.
- The CMS projections of converging growth at the rate of roughly 5.5 percent in the future.
That said, there are strong differences by payer. The low per capita rate for Medicaid in 2014 because of healthier enrollees was already noted. For Medicare the main reason for the spike in 2014 is the new hepatitis drugs: without Part D drug spending growing by over 17 percent, the per capita growth rate would have been below the 2013 rate, according to a presentation made at a recent Altarum symposium. Medicare spending growth is expected to be lower next year due to some technical changes in payment rates. In addition, 2015 data from Treasury Monthly Reports indicate a lower rate of spending growth in the first part of 2015.
Exhibit 2: Per Capita Expenditure Growth, By Payer 2009-2014 Actual, 2015-2024 Projected
Over the longer run, however, as the Medicare Trustees report issued on July 22 makes clear, both the effects of specific policy changes and general trends are in play. It states “[t]he methodology for projecting Medicare finances assumes a substantial long-term reduction in per capita health expenditure growth rates relative to historical experience, to which the ACA’s cost-reduction provisions would add substantial further savings.”
Trends in private spending are harder to interpret because they represent averages across many different insurers and benefit plans. Across the board, however, observers—including CMS economists and actuaries—are pointing to increased cost sharing in private plans and narrow networks as trends that will attenuate cost growth over the longer term.
During the decade of the 2000s, real wage growth was wiped out by increasing health care costs. Since the recession there has been faster migration towards these types of cost-containment efforts. We can debate whether or not higher cost sharing and narrow networks place undue burdens on the seriously and chronically ill, but they do represent new modes of cost control and they appear to affect health care costs and cost growth.
Notably, increases in service utilization rates (e.g. hospital stays and physician office visits) have slowed, meaning much of the increase in private spending is due to prices. Indeed, the Health Care Cost Institute’s latest analyses (for 2013) found a relatively low rate of spending growth due to falling utilization rates offsetting much of the observed price increase.
What Else Might Be On The Horizon?
Most economists believe that over the long run health spending growth is driven by new technology. What is interesting about this point in history is that efforts at innovation are more focused on delivering care efficiently than they have been in the past. “Patient-centered” is a new mantra; information technology (IT) products to support accountable care organizations (ACOs) and providers focused on achieving savings under bundled payment are developing; and predictive modeling is being used to enhance adherence to drugs for chronic diseases.
Even within the drug sphere, where many are apprehensive about the costs of specialty drugs and biologics in the drug pipeline, there is new attention to price. Just recently oncologists called for curbs on cancer drugs. And the expensive hepatitis C drugs, which were the top drivers of higher spending in 2014, were debated hotly even as their manufacturers touted them as cost-effective. That the debate over these new drugs even includes a discussion of cost-effectiveness is a sea change.
In the article issued today CMS argues that the greater cost sharing in insurance plans will have a persistent effect on growth rates. In my opinion, this effect could be augmented by more price transparency, greater consumer awareness of price and quality differences between providers, and greater provider awareness of costs of alternative treatments. At the same time, experience suggests that spending growth will rise with projected economic growth—more specifically growth in personal income—a precedent that suggests higher health spending in the future coupled with overall benefits to consumers.
The impact of official budget forecasters taking into account all of these types of changes in assumptions can be seen in this figure from a recent piece by David Blumenthal in the New England Journal of Medicine; the figures shows how slower growth in health spending has led to lower projections of the future costs of the Medicare program. These lower rates of growth and projected growth have had real policy effects: tellingly, Congress is now more focused on health information technology (IT) issues and the Food and Drug Administration (FDA) policy than on cost control, and the sustainable growth rate (SGR)“doc fix” was passed including bonuses for quality.
Some may want to read these projections as the latest installment in the debate over whether the recession or the ACA caused the slowdown in health care spending growth. This is far too simplistic for at least three reasons.
First, it isn’t clear that we would have had the ACA without the rise in the uninsured associated with the recession. Second, the important questions are not about whether the recession contributed to the slowdown, but how much the recession contributed and how permanent the service delivery changes will be. And third, the slowdown affected all payers and all service sectors—albeit to different degrees—indicating that there are multiple underlying factors driving growth.
Finally, no matter what the context, in the world of forecasts predicting turning points is the hardest thing to do. CMS has admirably adjusted their projections to take on new data and new research findings each year that the slowdown has persisted. We all want new cures to be developed while the existing care system delivers better value. The 2014 data and emerging trends described by CMS indicate this may well be happening.
Funding Acknowledgement: Dr. Buntin has a grant from the Commonwealth Fund to examine Medicare expenditure growth.
For clarity, this post refers to 2014 spending data in the past tense, but CMS treats the 2014 data in the Health Affairs article as projected because they do not yet fully incorporate all sources