As we get closer to HIMSS13, I think back to some of my key takeaways from HIMSS12. One of the things I heard over and over again last year is that healthcare is in a bubble. Going into HIMSS13, with Bill Clinton as the keynote, it seems worthwhile to ask: Are we in a health care bubble? If so, is it showing any signs of popping or at least deflating?
“To my mind … healthcare is in a bubble. When prices rise to a point where NOBODY can afford the product, the end is always the same. You have a crash, and that is the likely outcome in healthcare.”
“We don’t have a budget problem, we have a Medicare and Medicaid problem.”
Certainly healthcare spending looks like a bubble, the rise in prices has outpaced inflation since the 80s. But, in order to decide if we’re in a bubble, let’s take a look at what a bubble is and how we would know one if we saw one.
What is a bubble? Well, there really is no clear-cut definition, and there’s some debate in economic circles about whether you can have such a thing, but Bubbles trade in products or assets with inflated values. Normally you can only identify whether prices are inflated retrospectively after prices drop. The hard part is determining what “inflated” means when you are currently paying those prices.
What drives healthcare prices? In a word, the government. Total government spending is close to half of all health care spending, and Medicare reimbursement acts as a benchmark for prices across all of health care. Healthcare is not a traditional bubble because it’s not a true marketplace, it’s not a real economy. We don’t do much comparison shopping in healthcare based on price; a single payer sets the prices as opposed to the consumers of those products and services as in a traditional supply-and-demand marketplace. Indeed, Nate Silver recently showed in a recent fivethirtyeight post…”soon, we may cross an important symbolic threshold: when the overall majority of government expenditures are spent on, essentially, insurance programs.”
What causes (traditional) bubbles? Worthwhile reads on economics, bubbles, statistics and epistemology are Silver’s The Signal and the Noise: Why So Many Predictions Fail – but Some Don’t, and Nassim Nicholas Taleb’s (if you can get past the first third of the book and Taleb’s ego), Antifragile: Things That Gain From Disorder. I’ll go into more depth on these two in future posts, but to summarize, when bubbles happen, and when prices don’t reflect true value, bubbles can be linked to a few major problems:
1. Information asymmetries. In other words, people don’t have access to the same information. This is true in health care. Just try to find out how much a procedure costs, or get an itemized bill from your last hospital visit. This is hopefully changing, however, with things like All Payer Claims Data, Blue Button and the overall digitization of health info, making access to information and predictions based on that information easier.
If it does continue to change, it could drop prices long term through “health promotion, prevention, patient engagement, and coordinated care rather than the current emphasis on high-cost, sporadic treatment of disease events that are largely preventable.” – Ralph Syderman, MD, chancellor emeritus at Duke University and immediate past CEO/president, Duke University Health System
2. Mistaking uncertainty for risk. Risk is when you can (confidently) give a number for the likelihood of some event happening, uncertainty is when you can’t. Bubbles are often built by mistaking the two. While Silver might argue that we can understand real risk, Taleb might argue that we are always operating under uncertainty (so bet that bad things will happen and you’ll win).
The basic idea is that we can come up with algorithms that might lead us to predict the chances of something happening in the future, but if we’ve left out some important variables, we are actually operating under uncertainty. Of course, it’s difficult to tell if we’re missing an important variable in predicting future health care prices, but that’s pretty much the point, you can never know for sure. This could work in either direction, to modify future prices.
My own sense is that quality measures and consumer HealthIT might be more effective than we think at controlling prices, but that EHRs will make hospitals more “effective” at billing and driving up costs, at least in the near term.
3. More people have a stake in prices going up than in keeping them down. Incentives are aligned toward increased prices rather than “true” values. That was the case in the real estate bubble and financial crisis, and it’s almost certainly true in health care. Consumers are far removed from exerting any real price pressure on the system, although copays are going up and this will likely continue. I’d like to see more incentives offered to consumers to stay healthy, and I suspect we will as private insurers continue testing the consumer incentive waters.
Are prices set to drop significantly?
While there is some evidence that we may need to fix quite a few things to get prices aligned with value in health care and get away from less “inflated” prices, we likely won’t have a real market in health for one simple reason: There’s nothing on the horizon to change the fact that there are only a few purchasers of healthcare that can determine prices.
Because the market is not driving prices, it’s difficult to say we’re in a traditional bubble. A better way to look at it is whether there is a high likelihood for a significant drop in prices, and what constitutes a “significant” drop.
It’s established that health care prices are unsustainable. For most consumers of health care, it’s completely unaffordable if left to pay for alone. For a quick check of what the world was like when people had to pay for their health care risk, check out Nobel-prize winning paper “The Market for Lemons” (.pdf) on information asymmetries and how they lead to uncertainty from 1970. Some highlights on health care pricing:
“Generally speaking policies are not available at ages materially greater than sixty-five. . . . The term premiums are too high for any but the most pessimistic (which is to say the least healthy) insureds to find attractive. Thus there is a severe problem of adverse selection at these ages.”
“a 1956 national sample survey of 2,809 families with 8,898 persons shows that hospital insurance coverage drops from 63 per cent of those aged 45 to 54, to 31 per cent for those over 65.”
Can you imagine living in that world? I suspect that those over the age of 65 with insurance would be closer to 10% today, but it’s impossible to say what would have happened without Medicare-controlled pricing.
While we don’t have a true marketplace in health care and we won’t if we wish to keep (at least) 69% of our grandparents alive, the Affordable Care Act is attempting to make it more of a marketplace.
While still an artificial market, we’re slowly working toward a system where prices are more closely linked to quality. Beyond the ongoing experiment with ACOs, the most critical example of how changes are starting to affect the market is Medicare reimbursements for certain DRGs. In October 2012, Medicare started dropping reimbursement rates for hospitals that don’t meet certain quality benchmarks, “excessive readmissions” for heart attacks, heart failure and pneumonia. Some hospitals that are likely underperforming are already starting to feel the pinch and blaming Obamacare for layoffs. The penalties are reportedly designed to force low-performing hospitals that are unwilling to change out of business. That’s certainly a bubble bursting for those hospitals.
We’re going to hear a lot about layoffs from these hospitals in the coming months and years, but we’ll hear less often that the layoffs are happening because of poor quality and an inability to change to new incentive pricing models. If that’s popping the health care bubble, while there may be some pain in certain areas, particularly among hospitals performing poorly on readmissions, it may not be a bad thing for the overall health care system.
Does that count as a “significant” drop in prices? Depends on who you ask, but it will likely be small for the health care system as a whole, but life or death for the underperforming hospitals. Whether a bubble is bursting depends a lot on whether you stand on the inside or the outside.
Some early evidence suggests that attempts to control costs are already working, suggesting Medicare spending per capita is flattening, and that the aging population will do more to increase overall spending than per capita increases.
A traditional bubble? No
Based on this, I don’t think we’re in a traditional bubble. If we are in a bubble, it’s a Medicare-created and controlled bubble that’s fundamentally different than traditional market bubbles. Prices may start changing direction, but they are highly unlikely to drop quickly. Therefore, the “shock” (if we are fortunate enough to contain spending) won’t be as dramatic as a traditional bubble, and, if we’re lucky, we’ll slowly deflate back to the ground of more affordable care as opposed to popping. Again, it depends on where you stand.
To do this right and create a soft landing while still reigning in prices, institutions will need to continue focusing on:
- Aligning toward quality care for both providers and patients.
- Opening up data stores through more data and digitization, more adaptive architectures and improved patient engagement to fix information asymmetries.
- Giving patients the tools they need to get and stay healthy under new incentives.
- Focusing less on billing codes and more on quality of care provided.
Meanwhile, Medicare/Medicaid will need to keep working on aligning incentives toward quality.
Final Note: It’s also important to mention that we may have a health care jobs bubble. Some are arguing that Meaningful Use and health care reform have caused an unwarranted increase in the labor market for healthcare. While this might be a similar kind of spending bubble, I think it’s different than the overall health care bubble we are in, although changes here, such as removing incentives that are in place, might be an even more severe shock to the market. ♦
Latest posts by Leonard Kish (see all)
- Data Liberación Gets Personal, EXTREME Interoperability - June 23, 2015
- Health Data Planes, Trains and Automobiles: Think Like a Retailer - May 5, 2015
- The Death of Interoperability: Is it Time for One Record? - March 25, 2015