Maybe Doctors Could Price Shop Services on Behalf of Patients, Economist Says

In this series, MedPage Today is asking healthcare economists and policy experts the same questions about the high costs of U.S. healthcare. They’ll reveal what they believe is working, what’s not working, and what else can be done to bring costs down.

In this interview, William Schpero, PhD, a healthcare economist at Weill Cornell Medicine in New York City, explains — among other things — why he thinks consumer-focused price transparency in healthcare won’t work, but putting the onus on physicians might.

What has been the greatest contributor to high healthcare costs in the U.S.?

William Schpero: What we as a country spend on healthcare is a function both of how much healthcare we use and how much it costs. In other words, both quantities and prices matter. And I think here, international comparisons are helpful. When it comes to how much care we use, quantities, I think it’s a bit of a mixed bag. We use more of certain types of care and less of other types. For example, we have more c-sections per capita, but fewer hospital stays per capita.

I think on the other hand, with prices, the story is very clear. Prices for healthcare services in the U.S. are consistently higher than for services in peer nations. For example, for knee replacements, we know those cost about two and a half times here in the U.S. then what they do in similar nations abroad.

So then I think the question is, well, what’s the biggest determinant of our high prices? The best place to look is, where do we spend the most money in healthcare? It’s really hospitals. Hospitals are the largest chunk of healthcare spending in the U.S. It’s around a third of overall spending. We know that prices in hospital markets are largely determined by the market power of hospitals and health systems.

When hospitals, when health systems merge, they gain in market power, they gain in market share, and can negotiate higher rates, higher prices from commercial payers. We’ve seen massive amounts of hospital consolidation over the last 20 or so years. It’s for that reason, largely, that hospital prices have outpaced general inflation by a factor of two.

Ultimately, these higher prices that are commanded by hospitals and health systems are passed through to consumers in two different ways. First of all, they’re passed to consumers in the form of higher premiums. The average premium for a family of four with commercial insurance in the U.S. stands at around $27,000. It’s also passed on to consumers in the form of higher deductibles. So we see more and more Americans who face fairly high deductibles as their employers seek to pass on the financial risk of bearing their higher healthcare costs.

Has physician pay contributed to high costs?

Schpero: The Chicago economist Josh Gottlieb and his co-authors have estimated that physician earnings constitute around 9% of overall healthcare spending. So it’s not nothing, but it’s not a large part of the overall pie. We do know that physician salaries in the U.S. are much higher than in peer nations. But we tend to think that’s primarily because of two things.

One, we’re a higher-income country, and so everyone here in any given profession earns more than they do in other countries. Second, in economic terms, the outside options available to physicians are different. So in other words, we tend to think that physicians in the U.S. are paid at a level commensurate with their opportunity costs. That is, the jobs outside of medicine available to folks who become physicians are quite high-paying too. Medicine is a competitive field. If someone is capable of being a physician in the U.S., they’re probably capable of being an effective corporate lawyer or work in finance or some other high-paying profession.

I think the general sense is that physician earnings in the U.S. on average are probably about where they should be given the overall structure of the economy. And I think there’s pretty good work showing that if we were to reduce physician earnings and move physician earnings here to be more at the point in the income distribution where they are in other countries, that wouldn’t have a meaningful effect on overall healthcare spending. It’d be quite minimal.

Have administrative costs contributed to high costs?

Schpero: I think it’s certainly the case that administrative costs are higher in the U.S. The best work suggests that administrative costs here constitute a relatively small amount of spending, maybe 10% or 15% of overall spending, tops. Unlike prices, administrative costs have not ballooned in recent years, so it’s not a primary driver of the growth in spending that we’ve seen.

I think reducing administrative costs certainly could meaningfully lower the level of healthcare spending we’ve seen in the U.S. relative to peer nations, but I think it’s not a panacea. And I don’t think it’s where there’s likely the greatest room for improvement.

What is the best solution you’ve heard for lowering the cost of healthcare?

Schpero: When I think of best, I often think of large-scale redesign. The best solution would be a major redesign of our healthcare system. But I just don’t think that’s realistic given the fiscal and political constraints we face as a country at this point in time. And thus my own bias is that incremental reform is more realistic.

We certainly need to take steps to address consolidation in provider markets and the perverse effect that has on prices. I think this certainly includes increasing antitrust enforcement. Every proposed hospital merger deserves deep skepticism from the Department of Justice, the Federal Trade Commission, and state attorneys general. I think the problem with this solution is that many markets are already really consolidated and it’s very hard to put the proverbial genie back in the bottle.

The other issue is that many of the mergers that occur today involve not horizontal mergers across hospitals, but vertical integration — the acquisition of outpatient practices by hospitals and health systems. The issue is these tend to be smaller transactions that often escape regulatory scrutiny.

So I think increasing antitrust enforcement is certainly a big part of a first-order solution for reducing healthcare spending. But I think that the returns to that sort of policy solution are limited and it’s likely time for us to think about the trade-offs involved in some form of direct price regulation. For example, there is a team of folks at Dartmouth who have long proposed capping commercial prices … at something like 250% of Medicare rates.

Would having a single payer help control costs?

Schpero: This is difficult to say because the devil, to some extent, is in the details. Single payer could take many forms and any single-payer proposal involves big questions with different trade-offs and potential effects on spending, like what services will be covered, who will be eligible, would there be cost sharing, how would providers be paid, how would prices be set?

The answers folks have for these different questions varies across the dozen or so major single-payer proposals that are out there. For example, I think one big parameter that varies a lot across these proposals that has major implications for spending is coverage of long-term care. Many of these proposals have generous coverage of long-term care, and this would be expensive and actually I think may result in overall spending under single payer that is above where we are now since we really don’t have a functioning market for long-term care coverage in the U.S.

That said, it’s certainly the case most of the proposals that are out there involving setting prices somewhat closer to Medicare rates, and so that would result in a net reduction in prices to the extent we think prices are a large driver of our high level of spending.

Would greater transparency in pricing help bring costs down?

Schpero: For quite some time in the U.S., we’ve experimented with the theory that if consumers shop on price, this will promote competition across providers on price, and that will bring prices down. I think the evidence on this is a little bit mixed, but, the majority of the evidence to date indicates that is unlikely to be borne out in reality.

While I’m not bullish on consumer-facing price transparency as a mechanism to reduce healthcare spending, I’m actually a little more optimistic on provider- or physician-facing price transparency as an opportunity to lower healthcare spending, where the physician in some ways becomes the agent of competition. If physicians within the electronic medical record are given information on prices for, let’s say, an outpatient MRI at point of care, and can take into account how price varies across different imaging centers and how they think quality varies, they can be a good arbiter of quality and price trade-offs and refer patients to an imaging center that they think will deliver sufficient quality while also perhaps reducing out-of-pocket exposure for the consumer.

This requires not only information at point of care, but also a financial incentive for physicians to play this role. Perhaps this is in a model where they’re on the hook for the overall spending that is incurred by this patient in some sort of value-based care arrangement. This type of system is quite common at the moment for prescription drugs. So in many electronic medical record systems, physicians do have information at point of care on how different drug options or generic versus brand name options might lead to different out-of-pocket costs for their patient based on that patient’s specific plan and where they are in their use of the plan over the course of the year.

It is much rarer to see those types of systems for other forms of care, although they exist. If scaled in aggregate, that sort of provider-facing or physician-facing price transparency paired with incentives to use it could reduce spending through the same mechanisms we think about around trying to promote price transparency and shopping among consumers.

Will artificial intelligence (AI) be able to help control healthcare costs?

Schpero: I think there’s reason to be both optimistic and pessimistic about how AI might affect healthcare spending. I think at the moment, we don’t really know which side of the coin will land upright.

On the one end, I think AI intuitively has much potential to reduce spending. So it could make the healthcare system much more productive and sort of reduce the fixed costs of providing care so that clinicians can focus on delivering care and not documentation and things like that.

I think it also could lead to better targeting of care and better utilization management — things like prior authorization. That, I think, will improve our ability to differentiate high- from low-value care provision better, and then we can align reimbursement accordingly.

The last opportunity is perhaps the big one, that there’s great potential for AI to improve medicine, to improve diagnostic accuracy, to lead to earlier detection and treatment of disease. That could meaningfully reduce long-run healthcare spending.

Conversely, AI could meaningfully increase healthcare spending, particularly in the short run. AI could actually induce greater demand for care. AI leads to lots of new cheap diagnostics. You could have lots of folks who are using those services, who are requesting lots of additional follow-on services, that could lead to cascades of testing and screening that leads to a lot more healthcare use, at least in the short run.

Perhaps the biggest short-run concern I have at some level, because it’s already happening in the marketplace, is that many provider organizations and payers are using AI to help with coding. They’re using AI to increase the set of diagnoses they include on claims when they seek reimbursement for care for a given person. In general, if someone is sicker, observably sicker in the claims data, then payers will reimburse for that care at a higher level. So AI could increase coding intensity, and that could lead to increases in spending that I think would not necessarily be particularly high value.

We have very little evidence on all this to date. Lots of folks are studying all of these potential pluses and minuses and we’ll have a lot more understanding very soon.

Other Interviews in This Series:

Price Controls ‘Inevitable’ in U.S. Healthcare, Economist Says

Hospital Prices Drive High Healthcare Costs, Economist Says

Single Payer ‘Hands Down’ the Best Way to Solve High Healthcare Costs, Advocate Says

AI May Drive Health Costs Up, Doc-Economist Says

Please enable JavaScript to view the comments powered by Disqus.