This report originally appeared in the March 2011 issue of DOTmed Business News

Many workers in Manhattan complain about the grinding brutality of their commute. But Christopher Moore has them beat: he has to take a plane to get to his office.

Moore makes his home in a small Oregon town, but travels once a month to New York, where he works in Midtown, about a stone’s throw from the Empire State Building. There, he runs a group purchasing organization, one of the middlemen most hospitals use to reach deals with vendors for buying medical supplies, drugs and even the food they serve.

Moore’s GPO, Magnet, has carved out a niche for itself: it mostly deals with capital equipment and other big-ticket purchases. But virtually every hospital belongs to a GPO, with reports of hospital involvement ranging from 90 to 99 percent. In fact, most belong to between two and four, according to data provided by the GPOs.

Why so popular? GPOs are seen as a way to drive down supply costs while ensuring hospitals don’t experience the headaches associated with wrangling deals out of suppliers or paying huge salaries to professional bargainers.

They’re also huge — according to a Government Accounting Office report from last summer, the six biggest GPOs accounted for more than $108 billion in goods purchased through GPO contracts in 2008 alone.

It’s no surprise then, that over the past decade GPOs have gained their share of skeptics, who see them not as a relief to the country’s skyrocketing health care bill, but as fuel for it. Anti-GPO reports have steadily trickled out. A whistleblower lawsuit brought by a major GPO’s ex-employee, alleging corrupt practices with vendors, is wending through the courts. And hostile websites have sprung up, including the seemingly homespun Gporeform.org, with its American flag banner and freewheeling, early-days-of-the-Internet vibe.

“American taxpayers and the federal government needlessly pay higher prices for health care, while our citizens are forced into a system where innovation is stifled and patient safety is an afterthought,” the site darkly warns, summing up the critic’s case.

But GPOs, rightly, point out that medical device companies – with whom they are supposed to be negotiating lower prices – have been behind many of the reports. And the relationship between the two industries ought to be, at least in principle, “adversarial,” in the words of Curtis Rooney, spokesman for the Health Industry Group Purchasing Association, a GPO lobby.

“The only people who would benefit from changing the current system would be the medical device manufacturers,” Rooney tells DOTmed News. “Unless they’re willing to take a blood oath that prices would be reduced, I have serious doubts any good would come of it.”

But increasingly, the federal government has also turned a suspicious eye toward the multibillion dollar industry. The Senate launched two preliminary investigations, in 2002 and 2009. And last year, Sen. Herb Kohl (D-Wisc.), chair of the Senate Finance Committee, hinted he would haul GPO executives before Congress for questioning. And the office of Charles Grassley, an Idaho Republican, budget hawk and famed fraud buster, prepared a 15-page report last fall that concluded “empirical data [were] lacking” to support GPO claims that they save the public money.

True, those congressional hearings were scrapped – a Senate aide told DOTmed News it was, supposedly, because the fall elections threw out space on the calendar. But GPOs aren’t out from under federal scrutiny yet. DOTmed News has learned from well-placed sources that the Federal Trade Commission is planning a major investigation this year into whether GPOs violate antitrust practices.

Even if the feds’ efforts vindicate the GPOs, as the organizations and some economists expect, or simply peter out, as critics fear, there are other forces at work. The pressures of health care reform, the rise of technologies enabling larger hospitals to more easily manage vendor negotiations directly, and even competition among the GPOs leading them to diversify their services – all these could make the GPOs of the future different from the ones we have now.

“My opinion is,” Moore says on the phone from Oregon, “in 10 years, I don’t see the GPOs there in the same form.”

Strength in numbers
The industry claims GPOs go back in one form or another almost a century, and the idea is simple: strength in numbers. Hospitals might not be able to drive much of a bargain with a supplier on their own, but grouped with a bunch of other hospitals, they can enjoy the benefits of lower prices thanks to much higher volumes.

At first, GPOs were largely paid for by membership dues from the hospitals, but in 1986, a change took place: Congress gave GPOs safe harbor exemptions from the Anti-Kickback Statute. This law let GPOs pocket administrative fees from vendors, to the tune of 3 percent of the volume contracted. It’s estimated that GPOs net at least half their revenue through these payments. And here’s where the controversy starts.

GPOs say the payments are necessary to run large organizations dedicated to the complicated business of tracking and negotiating prices from different vendors, and that smaller hospitals couldn’t afford to get involved if they had to pay hefty membership dues.

But opponents view it differently. In essence, they see a conflict of interest: why provide the best price when it seems to be in your interest to get the highest price?

“If you made more money based on the revenue, where is the incentive to become more efficient, if efficiency is defined as lower costs?” asks Prakash Sethi, an economist with Baruch College in New York, and author of the 2009 book “GPOs: An undisclosed scandal in the U.S. “

And observers also worry that a lot of power and influence is concentrated among a few key organizations. While the number of GPOs is famously controversial, with estimates at about 600, most are small players. According to the Government Accountability Office, the six largest national GPOs accounted for about 90 percent of all hospital purchases nationwide in 2007. The industry lobby HIGPA, on its Website, says the nine companies that make up the organization were 80 percent of market.

Of course, how GPOs pay for their operations, and how few of them control the market, wouldn’t really matter if they save hospitals and, by extension, the whole health care system money. And this leads to the main question, one debated in dueling studies, GAO reports, congressional inquiries and the verdicts of lawsuits. And the answer, it turns out, depends on whom you ask.

Yes, GPOs save the health care system money
GPO proponents are quick to point out that the bulk of investigations have been in their favor and that they have often been vindicated in court. A federal court decision last summer where a hospital sued the device manufacturer C.R. Bard over sole-sourcing contracts – the practice whereby GPOs source products through one producer – ruled in favor of Bard, saying that hospitals ultimately had freedom to buy outside the contract, and that the evidence suggests they save about 16 percent on their products by using GPOs.

And a report prepared for HIGPA by Eugene Schneller, a professor and supply chain expert at Arizona State University, directly surveyed hospitals and found big savings. In Schneller’s 2009 survey of 429 hospitals, he estimated annual savings from GPOs at around $36 billion, with between 10 and 18 percent price discounts on products.

And proponents also point out their popularity with hospitals. In fact, their chief complaint with GPOs is that they don’t work with them enough, according to Schneller’s report. He found that hospitals were happiest with GPOs when they used more of them, and almost 90 percent of hospitals wanted GPOs to have deeper contract penetration in most markets.

GPO advocates say many studies finding against GPOs don’t take into account the big picture. For instance, a 2002 pilot study by the GAO looked at hospitals in one city buying pacemakers and safety needles. It found GPO prices were often higher than prices paid when hospitals negotiated directly with the suppliers.

But critics of the study pointed out that it was limited to a single city. And the more important criticisms were that the study misunderstood how hospitals use contract prices: they don’t always take them as the final price, but rather as a starting-point for future negotiations.

The analysis also failed to take into account significant potential savings provided by GPOs in personnel costs. At the time of the GPO study, it was estimated that it would cost a hospital $155,000 a year to pay for staff to do what a GPO does. And in Schneller’s more recent report, he estimated that by using GPOs, hospitals saw $1.8 billion in savings on personnel costs, with possibly more than $2 billion in savings if whole health systems were taken into account. GPOs are, in effect, he said, “buffering hospitals from the need to comprehensively carry out strategic sourcing, contracting and other key GPO activities for inpatient pharmacy, general medical products, orthopedic products, other clinical products and housekeeping products.”

These findings are echoed by managers of some smaller hospitals, who reckon they would lose money if forced to bring the GPO functions in-house. Bradley LeBaron, the CEO of Uintah Basin Medical Center, a 49-bed hospital with a 90-bed long-term care center in deeply rural Roosevelt, Utah, said he likely couldn’t afford to pay employees to take on a GPO’s role.

“I don’t know that I could hire enough people to do all that minutiae in regards to all that negotiations to have it ever pay back,” LeBaron tells DOTmed News.

This is also part of the larger argument for GPOs – that too many of the criticisms center on the price of goods rather than the global costs associated with buying and maintaining those goods. As most buyers know, the price of a car is not just what you pay to get the keys in-hand so you can drive off the lot. It’s the service, insurance and other costs that must be factored in when looking at the lifetime expense.

When it comes to purchasing, many GPO contracts reflect “lifespan” costs, such as warranties and replacing out-of-date goods, advocates maintain.

“We tend to look at price, not total cost of ownership,” Schneller tells DOTmed News. “In those GPO contracts are things like returns and what happens if a product becomes obsolete.”

No, GPOs don’t save you as much as they could
But for others, the answer is unequivocal. “The reality is, nothing in life is free, and while the hospitals may not pay for these GPOs directly, they’re paying [for them in] inflated prices,” Mark B. Leahy, president and CEO of the Medical Device Manufacturers Association, a medical device lobby that has funded studies examining GPOs, tells DOTmed News.

An economist who conducted one of those studies, Hal Singer, now a partner with consulting firm Navigant Economics, says many of the studies demonstrating value for GPOs have a fundamental flaw. They always start off asking the wrong question, he says.

“The GPO proponents are good at misfocusing or misdirecting the argument [about] whether they generate savings. Of course they generate savings relative to a world in which there are no GPOs,” Singer says. “The interesting question is whether the GPOs could do a better job than they do today if they were financed differently.”

It comes down, in part, to the agency problem or principal-agent problem, which Singer says has bedeviled economists for decades. “How do you incentivize an agent to operate in your best interests?” he asks. “If you’re letting him be paid by some third-party for the transaction, that means he’s beholden to the third-party, and not to you.”

In the report prepared for MDMA and released last fall, Singer and his colleague Robert E. Litan were able to address some of these issues through what Singer calls a “natural experiment.”

That “experiment” involved the 10-year-old company MEMdata LLC., a service that works with hospitals trying to buy capital equipment. In essence, MEMdata requests that hospitals provide information on the products or services they’re looking to buy and quoted offers they receive; the service then attracts vendors and has them bid to compete against the earlier offers in so-called aftermarket auctions. In these auctions, GPO-contract vendors often bid with contract prices against other vendors.

“When [hospitals] go to purchase something, they simply forward a copy of their quotation to us, so we know how it’s priced and how it’s configured,” explains Bob Yancy, CEO of MEMdata, based in College Station, Texas. “We analyze based on that; we benchmark that quotation against what we have in our database. Once we have ascertained what the price point of the incumbent’s quotation is, we issue an e-RfP to the entire applicable market. There’s an average of 4.2 manufactures in every capital equipment category. MEMdata issues an e-RfP to those 4.2 manufacturers, we assemble those into a report, post it into an online account, and the customers are free to pick and choose who they want to go with.”

Singer says MEMData makes money generally as a percentage of the savings realized, not a percentage of the transaction. “Here you have a ‘GPO’ whose incentives are not potentially distorted,” he argues.

In the report, the researchers said that on average, through the 8,100 auctions studied, the winning bid was 10 to 14 percent less than the original GPO price over the decade’s worth of data they had available. The average GPO price was $81,436, but the average winning bid was $73,990, according to the authors. In 2010, the most recent data available, savings reached 18 percent, MEMdata said.

"Indeed, in [more than] half of all auctions in the transactions database, incumbent device makers on the GPO contract were induced to lower their own prices for the same product to the same hospital, and did so by approximately 7 percent on average," the report said.

Relying in part on figures like this, GPO critics estimate repealing the anti-kickback statute would reduce federal health care spending by $11.5 billion and save hospitals and providers up to $37.5 billion, based on 2010 data. Total private health care savings could exceed $25 billion.

Of course, these findings are far from final. HIGPA says the methodology used in the study is suspect. And further, GPO proponents say it’s not clear that MEMdata’s figures can be extrapolated to the whole of health care. MEMdata traffics in capital equipment, which, with so-called physician preference items, like pacemakers and orthopedic implants, GPOs have less of a stake in. In Schneller’s survey, only about one-third of hospitals, and half of health systems, said they completely relied on GPOs to contract prices for capital equipment. It’s important to note, majorities of both said they wanted to “improve” GPO contract penetration in this area.

A more troubling point is made by some critics -- that hospitals might not realize, or care, about the extent of the problem. This point was brought up in a class-action lawsuit, potentially the biggest in history, against the GPO Novation and a network of hospitals. In the qui tam suit, an ex-employee of Novation claims she was fired by the Irving, Texas-based GPO after she allegedly complained that a vendor was not fairly bidding on contracts, but instead hoped to snag them by offering various deals to the GPO.

In a 2007 New York Times article about the case, the paper notes the complicated rebate savings method employed by many GPOs, in which hospitals are given a small percentage, say around 2 percent, of the savings back at the end of the year provided they meet volume purchasing quotas. This, critics say, could allow millions of dollars to get lost in the system, with some hospitals “overstating” their supply costs, thereby “getting larger Medicare reimbursements than they were entitled to," even if it weren't done deliberately.

In fact, this confusion gets to the core issue critics say underlies the whole problem: the lack of transparency. Many of the GPOs are privately held companies (of the big ones, only MedAssets is publicly traded). Although in response to public pressure, the GPOs set up a code of conduct they’re supposed to adhere to, it’s not independently monitored the economist Sethi says, which makes it hard to know what’s really going on.

What’s more, most of the studies – pro or con – on GPOs are based on fragmented data, as most medical device companies try to keep a tight lid on their prices. The prices that are usually shared are list or “sticker” prices, which nobody pays, and often the really expensive deals are protected by nondisclosure agreements with purchasers.

“Everybody thinks they got the best deal in town. You can go to a hospital in New York, and they all think they’ve done great,” Schneller says. “I can show you they pay 30 to 40 percent more for any given item [than the other buyer].”

This means that the data used for studies have serious limitations. “Put it this way, if they say they have great price data and they know what the price points are, I’d love to have access to it,” Schneller says.

GPOs: peeping into the crystal ball
To GPO critics, the hope is simple. “Our objective is to restore the illegality of these kickbacks to GPOs,” MDMA’s Leahy says.

But this practice has become increasingly widespread outside even the United States. Schneller says he recently spent two months in Europe looking at pricing, and found the GPO model – where the organizations were paid a cut of the transactions costs – was common throughout much of the Continent.

“I was at a meeting in Paris, where GPOs from around the world came together. There are GPOs in France, there are GPOs in Italy, there are GPOs in Portugal,” he says. “This is not unusual.”

So what sort of changes will occur? Integrated Delivery Networks, or IDNs, which are networks of hospitals in the same system and which can negotiate with vendors, have grown in popularity, but even these frequently rely on GPOs. And some larger hospitals are finding they can negotiate with vendors directly – for instance, for decades, Duke University’s pharmacy department, because of its massive, multimillion dollar drugs budget, was able to enjoy volume-based bargains from suppliers. And with better software, the process has gotten easier.

When it comes to competition amongst themselves, GPOs do about as well with commodity pricing as each other, according to most observers. “I think what we’re starting to see in the market, is that for everyday commodities there’s hardly a lot of price differences in and among GPOs,” Schneller says. But some have started to distinguish themselves in other fields. Duke, for instance, works with a GPO – not for drug pricing – but for knowledge-exchange and practice-management meetings. (However, a recent survey found many hospitals preferred GPOs for hard pricing metrics, and less for “touchy feely” endeavors.)

In the end, GPOs might also do what much of U.S. business has done – look abroad.

“What you see are many GPOs now engaging in global sourcing, looking toward China and India to be able to global-source their products and as they do, they hope they’ll gain a strategic advantage,” Schneller says.

If anything, new practices and technologies that make it easier to work with vendors could get co-opted by GPOs. For instance, MEMdata and the GPO Amerinet signed a deal several years ago for Amerinet’s members to use a service run by MEMdata called BudgetMatch, which helps plan capital equipment budgeting.

“GPOs have been around for 100 years, and they will continue to provide services to hospitals,” Rooney says. “I think the core function of GPOs is here to stay, and the new dynamics of all the data and technologies out there will be interesting, and they will inform purchasing, but I don’t think it would be unrecognizable.”

In fact, the continued existence, and possibly relevance, of GPOs is one thing friend and foe might agree on. “Nothing has changed,” Sethi says. “I’m convinced nothing will change.”