National Health Spending: Faster growth in 2015 as coverage expands and utilization increases

By Anne B. Martin et al.
Health Affairs, January 2017

Total US health care spending increased 5.8 percent and reached $3.2 trillion in 2015…. Following five consecutive years of historically low growth, from 2009 through 2013, health spending growth accelerated in 2014…. The faster growth in 2014 and 2015 occurred as the Affordable Care Act (ACA) expanded health insurance coverage for individuals….

Republicans will soon own the health care mess. Whether they do or don’t “repeal and replace,” they’ll own it. It didn’t have to be this way. They could have worked constructively with Democrats to achieve something resembling health care reform. Instead they chose sabotage as their primary goal, both before and after the enactment of the ACA.

Now, after years of practicing sabotage, Republicans must learn to build. Republicans have still not revealed their replacement plan, so it is impossible to say at this date what they will build. However, given Republicans’ longstanding support for managed care ideology, we may predict they will recommend, or passively accept, some or all of today’s managed care fads, including “accountable care organizations,” “medical homes,” pay-for performance, and possibly bundled payments, plus electronic medical records to facilitate all of the above. They will probably continue the overpayments to Medicare Advantage insurers as well, based on the theory that they know how to “coordinate care”/“integrate care”/do “value-based purchasing”/pick your favorite managed care buzzword.

Republicans have never celebrated managed care ideology as loudly as Democrats have, primarily because the GOP has never supported universal health insurance. (Since the modern health care reform debate began a half century ago, proponents of universal coverage have had to at least appear to have a cost-containment strategy.) But over the last quarter-century, conservative think tanks and Republican legislators, presidents and governors have endorsed, implicitly or explicitly, the fundamental tenets of managed care ideology, namely: Overuse (not high prices and administrative waste) is the primary cause of the American health care crisis; fee-for-service is the cause of that alleged overuse; and the solution is to antidote the fee-for-service incentive by shifting insurance risk to doctors and to authorize third parties to micromanage doctors. The most recent endorsement of managed care ideology by Republicans occurred in 2015 when a large majority of them endorsed MACRA (the Medicare Access and CHIP Reauthorization Act). Like the ACA, MACRA relies on every current managed care fad. It represents end-stage managed-care think even more vividly than the ACA.

So now would be a good time for Republicans to learn a lesson Democrats have yet to learn: Managed care doesn’t cut costs. It’s possible some managed care nostrums, such as “medical homes,” can have some modest positive effects on quality, but even that is not clear because all analyses of the impact of managed care experiments on quality rely on measurements of an infinitesimally small fraction of medical services. We must assume until proven otherwise that “teaching to the test” explains all or a substantial portion of the research that attributes modest improvements in quality to “homes,” ACOs, P4P, etc.

Perhaps the best way to attract the attention of Republicans to my message – managed care doesn’t cut costs – is to remind them why Democrats, from Obama on down, predicted the ACA would cut costs. Democrats did so sincerely and with great confidence. For a good example of such sincere but foolish rhetoric, see this 2010 article in the New England Journal of Medicine in which White House advisors Ezekiel Emanuel and Peter Orszag claimed the ACA “will significantly reduce costs.” Emanuel et al. believed their rhetoric; they really believed the main tenets of the managed care faith – overuse is the problem, and ACOs, etc., are the solution.

The prediction by Democrats that the managed care nostrums in the ACA would cut costs was one of the greatest gifts Democrats ever gave to Republicans. Republicans joyously beat Democrats over the head with that promise in every election since the ACA was enacted, and it paid off, most clearly in the 2010 and 2016 elections. I ask all Republicans: Do you really want to return that favor? As you contemplate your “replacement” legislation, do you really want to endorse any legislation that assumes ACOs or any other managed care nostrum will lower costs?

If I were in charge of teaching Republicans how to think clearly about managed care ideology and the role it played in fooling Democrats into making promises about the ACA they couldn’t keep, my first lecture would review the debate about the ACA’s impact on national health spending. My second lecture would review the evidence indicating ACOs and “medical homes,” the most important managed care nostrums of the dozens written into the ACA, have either raised Medicare spending or have had no effect. In the rest of this comment, I’ll focus on the first issue.

The argument Obama, Emanuel, Orszag, et al. made most often in favor of their claim that the ACA is saving money was that the ACA deserved much of the credit for the slowdown in national health care spending that began in 2009 and ended in 2014. Even by 2010 it was apparent that the main cause of the inflation lull was the Great Recession, which ran from December 2007 through June 2009. The evidence was even more compelling by 2014. The evidence today is overwhelming – the Great Recession was the primary cause of the inflation lull, and the net effect of the ACA has been inflationary, not deflationary.

I reviewed the evidence as of 2014 in a comment posted here in December 2014. At that date the latest data we had on national health care spending was the annual National Health Expenditures Accounts report for 2013, posted online by CMS’s Office of the Actuary in early December. That report, authored by Micah Hartman et al. and published in the January 2015 edition of Health Affairs, indicated that the lull in health care inflation that began in 2009 had continued through 2013. It was too early to say the lull had ended. However, because Hartman et al. attributed the slowdown to the Great Recession, they felt comfortable warning their readers that as the economy recovered the lull would end. [1]

That was not the news managed care advocates wanted to hear. They were sure that at least some of the dozens of managed care fads written into the ACA, particularly the provisions authorizing CMS to experiment with ACOs, would cut health care costs so fast the effects would be visible within a year or two of the ACA’s enactment and would more than offset the inflationary effect of the ACA’s expanded coverage and of an expanding economy. By as late as April 2015 Jason Furman, Obama’s economic advisor, was still claiming (a) the inflation lull had continued beyond 2014 and (b) the ACA deserved much of the credit for this non-fact. As late as August 2016, President Obama was still making that claim. [2]

But Hartman et al. and the numerous other observers who said in 2014 that the slowdown would end as the economy recovered turned out to be correct. The two national expenditure reports issued by the CMS actuaries since the 2014 report indicate the inflation lull ended in 2013. The second of these reports, the paper by Martin et al. quoted above, stated that substantial increases in inflation in total U.S. spending occurred in 2014 and 2015 and, moreover, those increases were due largely or primarily (it is not clear which was intended) to the historic drop in the uninsured rate caused by the ACA (2014 was the first year in which the uninsured could buy coverage on the ACA exchanges).

Here are the inflation rates for the years during and since the 2009-2013 inflation lull as reported by the January 2017 Health Affairs paper:

2009: 4.0%
2010: 4.1%
2011: 3.5%
2012: 4.0%
2013: 2.9%
2014: 5.3%
2015: 5.8%

Obviously the CMS actuaries’ prediction in December 2014 that the inflation slowdown would end as the economy recovered was correct. Obviously the managed care proponents who argued inflation would remain low even as the ACA reduced the uninsured rate and as the economy recovered were wrong.

Orszag, Obama and others have also alleged that a slightly less pronounced inflation lull in Medicare spending (a) has continued beyond 2014 (it hasn’t) and (b) was caused by the managed care experiments foisted on the Medicare program by the ACA (it wasn’t). The Medicare lull also ended in 2014 – inflation in total Medicare spending rose from 4 percent in 2013 to 5 percent in 2014.

Apparently recognizing that they can no longer cite a lull in inflation in total Medicare spending, some ACA proponents, including Orzag and Obama, have focused attention on inflation in per-Medicare-enrollee spending. But the lull in that measure also ended in 2014. Moreover, that lull was probably due largely to the aging of Baby Boomers into Medicare. Younger Medicare enrollees cost less than older enrollees, and the Baby Boom generation is large relative to older cohorts. For some time to come the dilution of the pool of Medicare beneficiaries by “boomers” will put downward pressure on per-enrollee inflation.

Finally, let me comment on one of the few kind words the CMS actuaries had in their latest report for a managed care nostrum. The January 2017 Health Affairs article noted that one of the causes of a slightly reduced rate of inflation in Medicare hospital spending in 2015 was “reductions in hospital readmissions.” As anyone following the health policy debate knows, one of the hottest managed care fads is eliminating “unnecessary” hospital readmissions. The ACA endorsed the evidence-free proposition that if hospitals were punished by Medicare for “excessive” readmissions within the first 30 days after discharge, hospitals would find a cheap or totally free method of keeping people out of hospitals, those methods would do more good than harm, and all the savings from reduced readmissions would be pure gravy.

The January Health Affairs paper noted, however, that “spending growth for nursing home and home health care [Medicare services] accelerated” in 2015, offsetting the reductions in hospital spending. A reasonable inference from these data is that the reduction in hospital admissions was achieved in whole or in part by greater utilization of nursing home and home health services. If that’s true, then the savings to Medicare from reduced readmissions was not pure gravy – it was offset, possibly swamped, by new expenditures on non-hospital services.

End of first lecture for Republicans embarking on their quest to own the health care crisis.

For the last half century, Republicans have been under no pressure to comprehend how little evidence there is to support managed care ideology. Those days are over. Republicans will either comprehend that managed-care think is a form of groupthink, or they will repeat the error Democrats made. They will make promises they can’t keep, and they will pay a price.


[1] Here is how Hartman et al. expressed their belief that the economic  recovery would eventually terminate the inflation lull: “The key question is whether health spending growth will accelerate once economic conditions improve significantly; historical evidence suggest that it will.” (p. 159)

[2] Obama and Forman were not alone in peddling the false claim that the inflation lull had continued beyond 2013 and the ACA was a major reason. Peter Orszag made this astonishing claim in an August 2, 2016 paper  in JAMA: “The conventional wisdom at the time the ACA was enacted was that despite its ostensible dual mandate, the act largely addressed the coverage problem while doing almost nothing to address cost trends. That perspective was flawed and frustrating at the time, but even the most optimistic forecasts were conservative relative to what has since occurred.” The accurate statement about “what has since occurred” is that the ACA’s managed care cost-containment provisions (as opposed to the simple cut to Medicare the ACA ordered) have not cut costs and may well have added to costs, while the ACA’s coverage provisions drove costs up substantially.  Note the title of Orszag’s paper includes the phrase “cost containment.”

[3] The real rates of inflation in total U.S. spending (that is, the nominal inflation rates listed above deflated by the underlying, economy-wide inflation rate) reported by Martin et al. show the same abrupt termination of the lull in 2014.

Kip Sullivan, J.D. is a health policy expert and frequent blogger, and is a member of PNHP Minnesota’s legislative committee. His articles have appeared in The New York Times, The Nation, The New England Journal of Medicine, Health Affairs, the Journal of Health Politics, Policy and Law, and the Los Angeles Times.