Rx for sharing health records

Architects of the NHIN must figure out how to build a network that lets rival hospitals freely share patient data while competing with one another for market share.

Architects of the National Health Information Network (NHIN) must figure out how to build a network that lets rival hospitals freely share patient data while competing with one another for market share. Health care information technology experts say failure to tackle this apparent paradox will perpetuate health care industry flaws, such as mass duplication of records, lengthy hospital stays and a worsening epidemic of medical errors.

NHIN, which will knit together thousands of electronic health record (EHR) systems, is a bold solution to this problem. Experts say it promises to reduce health care costs by 10 percent, lower the estimated 45,000 to 98,000 deaths caused annually by medical errors and cut the barrage of paperwork facing many hospital patients.

Without a system that can share records across hospitals, physicians must create original and often dangerously incomplete medical histories for patients, even when data might be available in the same town.

Yet despite widespread agreement about NHIN's benefits, health experts say the network faces a gargantuan obstacle: the ancient but natural lack of trust among competing providers.

"The first thing you run into whenever you start talking about taking information from various hospitals is trust," said Mike Sommers, chief information officer at the Cook County, Ill., Bureau of Health Services, which has installed a countywide EHR system and wants to join a Chicago-area health information organization.

"You might have two hospitals, and they are literally a block apart," he said. "You think they trust each other? Of course not. So the challenge is going to be to get people who summarily don't trust each other to be able to work with it and support it."

Out of chaos, 'chaorder'

Convincing hospitals to trust one another won't be easy, Sommers said. "I think it's going to need a push on the side of the government or the payers to make that happen and then a methodology or architecture so that they feel that their data is secure and they're not giving up the cookie jar."

In considering a solution, one group of health care planners sees a parallel between the market today and the financial services industry of the late 1960s. That is when Bank of America introduced the first plastic credit card, spurring a rival group of California banks to form a partnership under the MasterCard banner.

The ensuing retail credit war was so heated that the banks started mass-mailing credit cards, even to high-risk consumers such as college students, a strategy that increased banks' costs and nearly caused the industry to implode.

That's when a young banker named Dee Hock stepped onto the playing field. He introduced a strategy for the fledgling credit industry based on what at the time were unorthodox business principles.

Among them, he asserted that an organization's success has more to do with shared purpose than its advantages in assets, expertise or management. A corollary was equally striking: As in biological systems, organizations need to maintain a delicate balance between competition and cooperation to succeed.

"Neither competition nor cooperation can rise to its highest potential unless both are seamlessly blended," Hock wrote in "Birth of the Chaordic Age," a book on his management principles. "Either without the other swiftly becomes dangerous and destructive."

Hock gathered these principles together in a strategy he dubbed "chaorder." According to his strategy, organizations need to build nonhierarchical networks made up of equal parts cooperation and competition. Power is pushed out, and authority rises from the bottom. Using those principles, Hock established Visa International, a competitive but cooperative network that forms the backbone of the retail credit market today.

In its workings, Visa is a clearinghouse of consumer credit information held by different banks. By swiping their cards, consumers authorize their banks to release only financial information that is relevant to the transaction. In the Visa model, there is no central strongbox of consumer credit information. Visa is not an entity that resides anywhere in particular. Instead it relies on maintaining chaorder in the financial services industry.

That model can apply to the health care services market, proponents say. "If you change the word 'bank' to 'hospitals' and change 'account information' to 'medical records information,' the Visa architecture and the philosophy is something that makes impeccable sense," Sommers said. "I don't think the answer is to have big monolithic computers in every major city or contained in St. Louis.... I don't ever see that happening. I think the Visa model by far is the most eloquent and proven."

Overcoming hurdles

The Patient Safety Institute, a nonprofit organization based in Plano, Texas, is incorporating the Visa model as the basis of its strategy to enable health care information sharing among hospital systems, clinics and physicians who otherwise compete for clients.

With Hock and other health care and financial services leaders on its board of directors, PSI is eyeing several communities to test its concept, including Chicago; Knoxville, Tenn.; and Denver. It has already conducted a successful demonstration of its method in Seattle.

The institute hopes the tests will show that the principles of chaorder will work in health care. "Here are the choices," said Johnny Walker, chief executive officer of PSI. "You can do nothing, and in a few years, the public outcry becomes so big we're forced into a single-payer system to remedy the cost problem. Or you finally adopt a voluntary system like a Visa [system], or you have mandatory collaboration that's forced by the government."

Regardless of the path, a lack of technology will not be the reason why hospitals fail to share records, Walker said. "This has nothing to do with hospital systems not being standardized or being incompatible," he said. "It has to do with one hospital having no desire for the other to see who their very lucrative heart patients are or to see things that may not have been done very well from a patient safety viewpoint. For competitive reasons, there's no desire to share data. It's not that they couldn't or shouldn't."

PSI's strategy to overcome this hurdle is to build an information clearinghouse modeled on the Visa network to buffer information among competitive health care providers. The scheme obviates the need for a massive vault of data outside hospitals' firewalls.

"No consumer wants that," said Jane Delgado, CEO of the National Alliance for Hispanic Health (NAHH), which comprises national and local health care organizations that serve the Hispanic community. "So this system is a hunt-and-seek system, and that works better. People ask, 'Well, aren't there going to be mistakes?' I say, 'You know what? There'll be much fewer than there are now and more confidentiality than we have now.' So is it going to be a perfect system? Of course not. It was developed by humans. But it will be better than what we have now."

Beyond the savings inherent in such a network, consumer and community groups such as NAHH are starting to see its benefits for their constituents, many of whom face cultural or educational challenges in navigating the health care maze.

"There are people who don't understand why an organization concerned with Hispanic health would be so interested in this technology," said Delgado, who serves on PSI's board. "Our community includes some people who do not speak English, so trying to say your medicines and knowing all of them may be difficult. This makes it easier. Also, people may not have health records themselves. This pulls it together. Better health care is good for everyone but especially for those people who may not be the most active health consumers."

In endorsing the idea, Rep. Sam Johnson (R-Texas), who serves on the House Ways and Means Committee's Health Subcommittee, wrote, "The health care system needs a viable organizational model wrapped around a flexible, interoperable network that can accommodate the individual needs of every user or community, regardless of vendor, application or platform.... This will shift the competitive landscape away from competing on which organization has patient data to how well each provider uses that information to improve patient care and convenience."