All Content

Global Perspective


We’ve all heard the stereotypes: Other countries have socialized medicine, rationed care, endless lines, and little incentive for innovation. OK, there might be a grain of truth to the wait times. But healthcare in other developed nations is surprisingly varied in its mix of public and private providers, and it yields high-quality outcomes for a far better price than in the U.S. And yes, international innovation is alive and well.

Head-to-head comparisons can only go so far, with many countries using vastly different metrics to measure quality and efficiency. Nevertheless, the examples of bundling, reference pricing, and patient-reported outcomes offer a glimpse of how large-scale initiatives can help improve outcomes and bottom lines in the hospital and beyond.

The numbers suggest that the tests ordered, the drugs prescribed, the devices implanted, and other medical services offered are driving up costs.

Just the Facts

Last November, the nonprofit Commonwealth Fund in New York funded an analysis of healthcare data from the Organisation for Economic Co-operation and Development (OECD), which explains just how expensive healthcare is here.

In 2008, the U.S. spent roughly $7,500 per person on healthcare, an astonishing 50% more than the next closest country: Norway, at about $5,000 per person. And yet we lag behind Norway and almost all of our other peers in mortality rates. Cathy Schoen, senior vice president for policy, research, and evaluation at the Commonwealth Fund, says the OECD statistics also say something about how we use hospitals. Compared with its peers, the U.S. actually spends a smaller fraction of money on hospital care. We’re also on the low end of the number of acute-care hospital beds and hospital discharges per 1,000 people, and below average on the typical length of stay for acute care, at about 5.5 days. “So we’re not using the hospital more, and we’re not staying in it longer,” Schoen says. “Nor do we have way more beds, so it’s not an occupancy issue that’s driving this.”

Instead, the numbers suggest that the tests ordered, the drugs prescribed, the devices implanted, and other medical services offered are driving up costs, at least in part. So what can other countries tell us? As policymakers here debate how to bundle more healthcare payments around episodes of care (see “A Bundle of Nerves,” November 2010, p. 1), European countries including Germany and the Netherlands already are using the payment initiative on a national level to create efficiencies around hospital-based care. And they’ve done it with an American innovation: diagnostic-related groups, or DRGs. Bundling around a hip replacement, for example, includes the cost of the implant, the surgeon, and all of the hospital care. “It gives the hospital overall and all of its physicians an incentive to say, ‘If we could buy supplies cheaper, let’s do it,’ ” Schoen says.

An eye-popping 2007 McKinsey & Company study documents the relative cost of hip and knee replacement surgeries for five countries. In 2004, U.S. doctors performed just over half as many hip replacements per 100,000 people as their German counterparts. Yet the cost of each hip prosthetic averaged more than $4,800 per patient in the U.S.—four times higher than the $1,200 cost in Germany and the $1,400 cost in the United Kingdom.

Part of this difference, Schoen says, is due to supply chain management and involving doctors in the decision-making process. Many countries (and a few integrated health systems in the U.S.) are asking surgeons to help select just one or two prosthetic implants, negotiate for bulk volume pricing, and then track the clinical outcomes of those devices to flag poor performers, she says.

Setting the Bar for New Drugs

Drugs are another big-ticket item, and the U.S. pays almost twice as much per capita as the OECD average. To keep their prices lower, Schoen says, many European countries have information systems that track the relative clinical effectiveness of pharmaceuticals. “And they’re using it to inform the way they cover drugs: not to exclude them from the list of what’s covered, but to do something in Europe that’s called reference pricing,” she adds.

Let’s say a new drug costs 50% more than an older one with roughly equivalent efficacy. Under reference pricing, a doctor can still prescribe the new drug, but the patient must pay all or most of the difference. Such benchmarking has fueled an interesting dynamic. “The brand names that are coming in and want to get some market share will price themselves lower, because if they’re priced really high compared to the reference price, the chances are they just won’t ever get a market share,” Schoen says. As a result, drug prices stay lower.

The concept, although discussed in the U.S., has yet to be widely implemented here. A new study in the April issue of the Journal of Managed Care Pharmacy, however, could cause some cash-strapped governments to take a closer look.1 In the study, the Arkansas State Employee Health Plan used reference pricing for proton-pump inhibitors, using the cost of generic omeprazole as its reference point. Over the 43-month reference-pricing period, net plan costs for the drugs dropped by 49.5% per member per month.

Patient Feedback

A third lesson is that constructive feedback on quality can improve performance, even if no money is attached to outcomes. Like the U.S., Germany is placing a high priority on metrics that evaluate hospital quality. Schoen says the German performance improvement initiative is identifying outliers and providing them with feedback and technical support, but it is not built into the payment system. “They’ve had pretty rapid improvement out of that,” she says, “and I would say we’re learning the same thing in the U.S.”

Initially, Medicare data posted on its Hospital Compare website ( showed a wide hospital-to-hospital variation in mortality rates for pneumonia, heart attacks, and congestive heart failure. But since then, Schoen says, most outliers on the low end have improved dramatically, even though the only payment incentive was to encourage reporting. In fact, CMS is dropping some core measures from its hospital value-based purchasing program.

Public reporting of quality measures, especially mortality rates, is certainly not without controversy. But Schoen says that if handled properly, disseminating information that suggests a facility’s performance is subpar can tap into the professionalism of its staff and create a strong incentive among them to do better. “That’s something true both internationally and in the U.S.,” she says.

In the U.S., Schoen says, the basic questions asked of patients in the HCAHPS (Hospital Consumer Assessment of Healthcare Providers and Systems) portion of Medicare’s new VBP system represent a good start. But the experiences of other countries, she says, suggest that patient reporting should be directed more at outcomes, similar to a proposal left out of last year’s healthcare reform bill that would have created a feedback system for patients receiving implantable medical devices.

Even so, hospitals like Dartmouth-Hitchcock Medical Center in Lebanon, N.H., are instituting patient feedback systems on their own, and a National Institutes of Health (NIH) initiative called PROMIS (Patient Reported Outcomes Measurement Information System) is gaining traction. “It’s less blaming, and it’s more informing,” Schoen says. TH

Bryn Nelson is a freelance medical writer based in Seattle.


  1. Johnson JT, Neill KK, Davis DA. Five-year examination of utilization and drug cost outcomes associated with benefit design changes including reference pricing for proton pump inhibitors in a state employee health plan. J Manag Care Pharm. 2011;17(3):200-212.

Recommended Reading

Into the Night
The Hospitalist
ONLINE EXCLUSIVE: Quick fix eliminates indigent discharge problems
The Hospitalist
ONLINE EXCLUSIVE: Listen to experts discuss drug shortages
The Hospitalist
ONLINE EXCLUSIVE: Subsidy or Investment?
The Hospitalist
ONLINE EXCLUSIVE: Hospitalists discuss strategies for indigent transitions
The Hospitalist
   Comments ()