Medical Device Daily Washington Editor

WASHINGTON — The history of electronic health records (EHRs) is short and replete — replete with stories of frustration which that repetitive rock-roller Sisyphus could easily relate to. And a briefing yesterday at the National Press Club highlighted the rise and fall of one of the more ambitious such undertakings.

The Santa Barbara, California, regional health information organization (RHIO) commenced in 1998, but thanks to the convergence of several factors, the underwriters pulled the plug on the effort at the end of last year.

The story of this RHIO's development from blueprint to graveyard may serve as a guidebook on how not to build such a system. And several speakers at the briefing, sponsored by the publishers of Health Affairs, said that the Santa Barbara RHIO experience may yield insights that might not otherwise be understood by the organizations that foot the bills.

Robert Miller, PhD, a professor of health economics at the University of California at San Francisco who was commissioned to examine the effort prior to last year's collapse, told the audience that Santa Barbara "was once the most ambitious and publicized" RHIO in the U.S., "and many thought that the Santa Barbara exchange was successful."

However, little data was ever actually exchanged.

Miller said that Santa Barbara seemed ideal because "it's an uncomplicated market" with only three major hospitals and one analytical lab. The idea was to provide a system that would allow providers to "go to one local web portal with one user sign-on, a password and a uniform graphic user interface to obtain all healthcare data for a patient across all community healthcare facilities," Miller said.

As the CEO of Care Management Science, David Brailer, who also spoke at the session, decreed the project feasible and informed several organizations of his findings, including the Santa Barbara Regional Health Care Authority (Goleta, California), a Medicaid plan, and the California Healthcare Foundation (CHCF; Oakland), a healthcare philanthropy. The idea was that the system could be built with off-the-shelf software using peer-to-peer networks rather than with a centralized repository for data storage.

Miller said, "Technology development was the main constraint on development" between 2000 and 2005. "The biggest technology barrier was the interfaces," he said, noting that the difficulty of getting the data from one user to another forced planners to opt for an intermediary program to translate the data into a format useable for each provider.

This and other technical fixes, however, boosted costs and drove a wedge between stakeholders.

A sharp disagreement also arose over which organization would assume legal liability for data errors as well, a conflict never resolved.

Miller said CHCF "played the role of venture capitalist," but its board soon "realized that the exchange rewards would not be captured" by those making the investment.

"The overall problem with the money is that demonstration projects distort incentives" in that — in the words of one of the participants — "foundation money polluted the process," according to Miller.

Interest in participation soon waned and "at shutdown," Miller said, "only two organizations were providing data."

"Underlying all this was the lack of a compelling value proposition," he said. Santa Barbara "had an uncomplicated market, which made it attractive as a demonstration project," but the limited size of the market meant that there was limited need for such a data exchange.

"Not surprisingly, there wasn't much use of the exchange for the limited time" it was in operation. "Without adequate reward, even a small amount of risk … looks like a big deal."

Miler said, "To speed investment, it's important to understand value propositions," noting that some services that are needed early will not have a great value proposition because much of that early value "accrues to other investors," such as improved health to the patient that benefits only the patient and his/her employer.

As a consequence, "grant money is important" in getting a RHIO online, and "leadership is critical" in maintaining momentum.

"It's a lot easier to get stakeholders to pay for the marginal cost of development than to pay for the infrastructure," Miller said.

David Brailer, MD, founder and chairman of a California-based VC firm Health Evolution Partners (San Francisco), said, "What I learned in Santa Barbara was nearly the polar opposite of everything I thought I knew about IT." He found especially that what one does "as a business enterprise" does not strongly resemble what one does "as a social enterprise."

Brailer, best known for serving as the first national coordinator for healthcare IT at the Department of Health and Human Services, said it was clear to him in 2002 that the Santa Barbara project "had run its course."

One of the problems was "the orthodoxy … that you do a top-down project" that controls resources.

"That's exactly what we tried to do in Santa Barbara," he said, but he noted that this turned out to be an utterly backward approach.

"You simply have to sell this to people" he said, because buy-in is nearly impossible otherwise. "Until the bottom-up efforts tell us what we need to know," any such effort is doomed.

He said that even though the value model "was not discussed at all in 1998, "we believed that value propositions would reign." Nonetheless, over time, "it became very clear that … the rate-limiting step is the business model" itself because there are too many incentives in healthcare provision that "are perverse."

The leadership demonstrated early on by CHCF led others to take a passive stance, Brailer said. The contractor "took on a large software development project on a fixed budget" and was too optimistic on the delivery date and oversold the project, which led to public skepticism.

In conclusion, Brailer said that "the pace of RHIO development will continue to be slow," and the interface between systems will never be easy to design.

Serious funding is essential, he said, naming healthcare firms and pharmaceutical companies as possible sources, and that even advertising on the RHIO site could generate funds.