In our last Road to Hellth post, we looked two of the four ”value-based payment” demonstrations whose results were recently reported by the Congressional Budget Office (CBO).
These demonstrations are important because they represent the best thinking that Medicare can muster with respect to how to reduce healthcare costs and make things all better on the American healthcare front. In particular, they represent the Affordable Care Act’s only real hope for stemming the more than one-trillion dollar hole on the federal budget that it’s going to create over the next ten years. So much is riding on these results that it would be a crime to simply acknowledge the findings with a headline and blithely launch into yet another series of demonstrations. After all, our national objective should be to find and disseminate what works, and discard what does not work. In our previous post we looked at the Physician Group Practice Demonstration and the Premier Hospital Quality Incentive Demonstration and were frankly disappointed with the results. How did the next two fare?
The Medicare Home Health Pay-for-Performance Demonstration lasted for two years, but the CBO only had the results of the first year available to analyze. As to the design:
“The Medicare Home Health Pay-for-Performance Demonstration was initiated by CMS to test the effects of providing financial incentives to home health agencies (HHAs) to improve quality of care. Under the demonstration, HHAs received their standard payments from Medicare and were eligible to receive additional payments if the participating HHAs in their region achieved Medicare savings and if they met certain criteria regarding the quality of care…
For each of four regions, CMS estimated the Medicare savings attributable to the demonstration and distributed the entire amount of those savings to HHAs in the treatment group whose quality scores were in the top 20 percent among demonstration participants in their state or whose improvement in quality scores was in the top 20 percent.23 Thus, although it was anticipated that the demonstration might lead to improvements in the quality of care that could reduce Medicare expenditures, the demonstration was not intended to yield net savings for the Medicare program. HHAs in regions that did not achieve savings did not receive any additional payments, regardless of their quality scores.”
Once again this study hopes to demonstrate that: (1) “high quality (at least as Medicare measures it) leads to low costs”; and (2) clever payment, leads to high quality. In order to estimate the impact of the payment scheme on its expenditures, Medicare compared the actual expenditures for patients of the HHAs in the treatment group with an estimate of what they would have been without the demonstration. The result:
“During the first year, the demonstration had no overall effect on the seven measures of the quality of care that were used to determine HHAs’ eligibility for bonuses and had little or no effect on the other reportable measures…
An analysis of the effects of the demonstration on Medicare expenditures showed that HHAs in three regions achieved combined savings of about $15 million during the first year; expenditures were about $9 million higher than projected in the fourth region.25 However, the estimated savings were very small as a proportion of projected spending—ranging from 0.4 percent in one region to 1.2 percent in another region—suggesting that those differences between actual and projected expenditures could reflect random variation in Medicare expenditures rather than actual savings. The absence of a notable effect on Medicare expenditures is consistent with the finding that the demonstration had no discernible effect on quality of care—particularly in the number of hospitalizations and in the use of emergency care.”
Here we see some of the same lessons that we should have learned from the Premier Hospital Quality Incentive Demonstration. In this study as in that one, all of the participating organizations were volunteers. Who volunteers for a project that offers to reward home health agencies that meet certain performance criteria? Why, agencies who already know how. So what has Medicare proven by rewarding them for what they already do? Nothing. Absolutely nothing. How much did Medicare save as a result of this innovation? Statistically, nothing. Absolutely nothing.
Who designs these studies?
But there’s no use crying over spilled taxpayer dollars, especially when they’re borrowed from China. Let’s look for some good news from the fourth and last “value-based payment” trial, the Medicare Participating Heart Center Demonstration (MPHCD).
This project was a little different from those that we’ve discussed thus far. Instead of trying to achieve a financial healthcare miracle by rewarding people for hitting “quality” metrics and saving money through more efficient and effective care, the MPHCD represents an effort to save money by “bundling” services. Bundling simply says that, instead of paying for each thing doctors and hospitals do to produce a successful coronary artery bypass operation, Medicare would instead negotiate an up-front one-price-for-everything fee to be paid for each bypass patient treated. If the total cost is higher than the negotiated price, the hospitals and doctors lose money. If it’s lower, they get to make a profit. Simple, eh? Here’s CBO’s description:
“CMS selected the 7 hospitals from 27 applicant institutions.30 CMS’s criteria for selecting participants for the demonstration included the projected Medicare savings implied by their proposed bundled payment rates, the quality of care at the hospital, and the volume of bypass surgeries performed there…
Applicants were required to propose separate rates for two DRGs in which bypass surgery is performed.32 During the demonstration, the bundled rates were updated annually using the payment updates for hospitals and physicians in Medicare’s fee-for service program. According to the government’s estimates before the demonstration, the expected savings at the four original hospitals for services covered under the bundled-payment arrangement varied from about 10 percent to 30 percent, with an average of about 20 percent. For each of the three later-starting hospitals, the expected savings were about 7 percent to 8 percent.
Competitive pressures prompted the hospitals to apply for the demonstration. All of the hospitals that applied anticipated that being named a Medicare Participating Heart Bypass Center could help boost their volume of bypass surgeries. The hospitals also were concerned that Medicare could implement a national policy requiring bundled payments for bypass surgery and contract with a limited number of providers; they expected that their participation in the demonstration might improve their chances of inclusion in such a national program.”
So let’s get this straight. Medicare went to a bunch of hospitals and said directly or indirectly: “Hey, we might give a few hospitals around the country an effective monopoly on Medicare bypass surgeries. If you guys want to have a chance of bein’ one o’ da lucky winners, we want ya to quote us a bundled price for bypass surgery that’s lower than what we’re currently paying ya. If not, well golly gee. It would be too bad if’n ya lost the ability ta do bypass surgery on Medicare patients, especially since da majority of bypass patients are on Medicare. That would kinda wipe out yer heart surgery program, wouldn’t it? But hey, you guys make up your own minds about whether ya wanna take us up on our offer. But don’t wait too long to decide…”
Now here’s the good news. According to the CBO, this strategy worked!
“The evaluation showed that the demonstration reduced Medicare expenditures on services furnished during hospital stays for bypass surgery by about 10 percent. The estimated savings were in the range of about 5 percent to 10 percent for five of the seven hospitals and about 20 percent for the other two hospitals. Those savings reflect the estimated difference between the bundled payments and the amounts that Medicare would have spent for services provided to those bypass patients in the absence of the demonstration…”
Did this bundling reduce the true cost of taking care of those patients? It’s not really clear. It may have in three hospitals, but not a fourth, and there was no direct comparison with comparable cost data in other hospitals. Was there any change in patient outcomes? Nope.
The immediate lesson? If you’re the government and a monopsonist (i.e., pretty much the only buyer of a specific set of goods and services), and you beat up your suppliers by threatening them in a nice way, you can get them to lower their prices. Unfortunately that approach is only going to work for so long. It’s not the financial miracle the Affordable Care Act is going to need in order to keep healthcare from bankrupting the country.
And with that, let’s get back to the original question posed by these posts: will our unelected leaders at Medicare and the Department of Health and Human Services bother to learn anything from the substantial amount of time and money spent on these demonstrations? First let’s review what we believe to be the important take-home lessons of these four “value-based payment” studies:
- It’s time to get real with respect to saving American healthcare by use of Accountable Care Organization. Far from proving a boon to the healthcare system in general and Medicare in particular, the results of the PGP demonstration show that this complex, top-heavy strategy has little to offer other than an increase in healthcare infrastructure and administrative expense. Yes, two of the ten PGPs involved secured bonuses in all five years of the demonstration and “saved” money relative to projections. If Medicare specifically wants to study those programs in an effort to understand what made them special, that’s swell. But when the pilot project is a failure, most responsible people would think twice before launching it on a widespread basis as if nothing happened. As they continue to move forward with ACOs, Medicare and the Department of Health and Human Services (HHS) seem strangely indifferent to the results of their own five-year study.
- If Medicare wants to prove that paying doctors and hospitals to improve quality measures is a good investment of time and capital, it’s time to design a decent study to prove it. Imagine this scenario. You want to prove that students who are paid to study hard and do their homework will improve their grades and incur fewer education costs than students who aren’t paid. To prove that the payments make a difference, who are you going to recruit into the study: (a) students who already study hard, get top grades and incur few education costs; or (b) students who don’t study as hard, don’t get top grades, and incur substantial education costs?
Yes, we would have picked (b) as well, but this subtlety is apparently lost on America’s best and brightest at HHS. Who are the people who approve these demonstrations, anyway?
Just as students who have already achieved the desired end point in the absence of payment aren’t going to improve much even if we pay them, hospitals and home health agencies that are studly enough to volunteer for Medicare’s pay-for-performance studies aren’t going to change their behavior if they’re already meeting the desired quality criteria. Nor will they prove anything for Medicare by virtue of their participation. Yet this is exactly the approach that Medicare took in both the Premier Hospital Quality Incentive and Home Health Pay-for-Performance (P4P) demonstrations. If you want to show that improvements in quality can be created by a P4P program, common sense dictates that you test your hypothesis on hospitals and home health agencies that have lots of room for improvement, and that you choose the participants on that basis. Not the best, probably not the worst, start out somewhere in the middle. If they show improvements in quality metrics and the level of overall spending after (and only after) you launch the P4P program then you’ve proven something. If not, drop the idea and move on. And if you’re not willing to do the appropriate study,
- Proving that you can beat up your suppliers doesn’t mean that you’ve hit upon a beneficial, or even sustainable, strategy for reducing healthcare expenditures. Since the Medicare Participating Heart Bypass Demonstration is the only “value-based payment” demonstration to document savings, it might seem logical to conclude that forcing doctors and hospitals to negotiate bundled Medicare payments that are lower than what they’d normally get from fee-for-service billing is a brilliant idea. However that depends upon who gets stuck with the bill. Medicare and Medicaid have a long history of cost shifting onto patients who self-pay or have private insurance, thus creating yet another hidden tax for them to pay on behalf of the federal government. Not only is this type of cost shifting unsustainable, but it’s exactly the type of approach that is most likely to destroy the entire healthcare system in the long run.
We’ll be the first to admit that these lessons will be hard for anyone at Medicare to accept. They’re grim at best because they say that Medicare’s current cost-saving doctrines are mostly wrong, unproven or unsustainable. But Americans are far better served by facing this reality than continuing to live in a fantasy world where hiring more “quality care” administrators is the best way to reducing the cost of healthcare. The overall message of the CBO to everyone involved in healthcare is pretty clear: it’s time to take a radically different approach to finding economic salvation.