By Eric Pfeiffer

Case Study: Kaiser Permanente

Corporate Headquarters | Oakland, Calif.
Year founded | 1945
CIO | Clifford Dodd
IT employees | 5,000
Business units | 29 medical centers, 423 medical offices, 11,345 physicians
Revenue | $19.7 billion in 2001, up 11.3% from $17.7 billion in 2000
Net income | $681 million in 2001, up 16.6% from $584 million in 2000
IT budget| 5% to 7% of operating revenues

For CEO David M. Lawrence, 1997 was the year to seriously entertain the notion of staging a palace coup. Nothing seemed to be going right for his nonprofit, Kaiser Permanente, the largest HMO in the United States. For the first time in its 52-year history as a healthcare innovator, Kaiser was bleeding money: In 1998, it would post a $288 million net loss on sales of $15.5 billion. One reason was that the HMO was forced to send many patients to costlier, non-Kaiser specialists because it had underestimated demand.

Yet like the rest of the hypercompetitive healthcare industry, Kaiser also was suffering from red tape, mind-boggling complexity, rising drug and care costs, and excessive overhead: Of the $1.2 trillion Americans spend annually on healthcare, some $250 billion to $450 billion is spent on administrative costs.

Kaiser was also stymied by its far-flung and decentralized culture, which stretched over a geographical area of some 10,000 square miles, from Washington to Honolulu. Its three units—comprising 29 medical centers, 11,000 doctors and the insurance company that pays for it all—were, for the most part, calling their own shots, without any central oversight regulating the purse strings. "The [competitive] pressures of healthcare became so great that the organization panicked," says Robert Pearl, CEO of The Permanente Medical Group, the unit of Kaiser that represents the HMO's doctors. "Everyone closed wagons around each of the separate parts. The entire fabric of the organization got torn asunder."

CEO Lawrence saw the need to give Kaiser emergency tech care. Patient records were still being kept on paper and carted around in fleets of trucks from one doctor's office to another. IT, Lawrence believed, would not only digitize those records, but also streamline time-consuming administrative procedures, cut error rates and greatly enhance the quality of Kaiser's medical care. Keeping doctors more up-to-date on the latest medical information via the Web, he believed, also would give companies—and consumers—more of a reason to pick Kaiser over other HMOs.

There was just one problem: Kaiser's IT department was as balkanized as the rest of the organization. Each of the nonprofit's seven regions—California, Colorado, Georgia, Hawaii, the Mid-Atlantic, the Northwest and Ohio—had its own CIO and its own IT agenda. Adding to those layers of red tape, the nonprofit also had 27 different financial ledgers and 13 different data centers. Even Kaiser's most important IT initiative—a long-standing project aimed at digitizing all 8.2 million patient medical records—did not have a coherent project, vendor strategy or implementation plan. Each region was spending millions to devise its own digital records system, paying no heed to what anyone else was doing. "I began to get really nervous about the amount of money each region was spending on this," says Lawrence. "Everyone was doing their own thing. I needed more control over spending—and direction."

Bitter Medicine

Bitter Medicine

What he needed was to dismantle the regional IT offices—fast. So Lawrence did something dramatic and rather unorthodox in the conservative, consensus-driven healthcare culture: He announced that IT at Kaiser would be centralized. Goodbye, regional control and cultural autonomy. Each unit would now report to a national CIO. And local executives who refused to go along? They'd get no more IT funding. Lawrence appointed Tim Sullivan, the ex-CIO of First Interstate Bank, to be Kaiser's first national CIO. The immediate goal: savings in the millions and a break in the cultural logjam surrounding IT at Kaiser. "If you look at the history of IT rollouts," Lawrence says, "you see moments where the leadership had to act this way to get anything done."

Kaiser, five years and some $2 billion in IT investments later, is still trying to make progress, and the cultural uproar triggered by Lawrence's no-holds-barred ultimatum still rankles. For some, Lawrence's decision was akin to a declaration of war. "Up until the time of his decision," says Henry Neidermeier, a Kaiser IT executive in charge of Web development, "Kaiser was run as a collegial organization, not like a corporation. Physicians had been making all the decisions." Lawrence, a former physician himself, says it was the toughest management decision of his career. "It was a unilateral action," he says. "I was deeply criticized and resented. I had several very, very uncomfortable conversations with physician leaders."

Lawrence says he remembers more than one shouting match with physicians he had known for years, including several calls by some physicians for his ouster. Andy Wiesenthal, a doctor in Kaiser's Colorado region recalls: "There was open dissent…it was a very difficult time." Almost overnight, Wiesenthal, who was critical of Lawrence's mandate at first, suddenly found he no longer had an IT department to generate his analytical reports—it had been dismantled and most of the IT workers were relocated to Kaiser operations in California. "All the people were gone. The rooms were empty. The mainframe was removed," Wiesenthal says. "The message was impossible to miss."

While he understood the need for a national CIO, Wiesenthal, like many other doctors at Kaiser, also worried that centralization would cause service to suffer: In 1997, an error made by an IT programmer caused Kaiser to inadvertently send confidential health information for 858 patients to the wrong e-mail addresses, fueling charges of privacy violations that made headlines and only bolstered resistance among many Kaiser physicians to Lawrence's technology mandate.

His decision also spurred a flurry of discussion outside the company. For years, the medical industry, considered a technology laggard, had tried but failed to digitize operations. Like Kaiser, most have had to struggle against cultural barriers, administrative complexity, legacy systems, cost constraints and regulations governing the exchange of patient information to make any progress—and usually came up far short: According to Gartner Inc., fully 85 percent of hospitals and other healthcare providers say they still don't know if they have the equipment or infrastructure to accommodate e-health, and budgets are too slim to realize change anytime soon. "Anyone who tries to do the right thing IT-wise in healthcare today is running huge cultural and financial risks," says Uwe Reinhardt, a health economist at Princeton University.

Indeed, the Kaiser experience shows how difficult IT reform can be. Since embarking on its Net effort in 1997, the HMO has fallen nearly two years behind schedule. It now expects to have back-office, insurance and patient records fully digitized by 2005 or 2006, rather than by the end of this year. Cultural battles have triggered long delays in the launching of some pilot programs, doctors in many regions still resist change, and spirited arguments continue over which technologies should be purchased—and from whom. Says John Glaser, vice president and CIO of Partners HeathCare System Inc. in Boston: "When I first heard about Kaiser's move to nationalize the IT department, I knew it would be tricky." Dozens of doctors resigned rather than work through the changes. "We may have underestimated the complexity of this undertaking," Lawrence admits.

Still, Kaiser is starting to see results. Among them: a 20 percent drop in drug costs at some clinics, thanks to an in-house network that doctors tap into from keyboards and flat-panel screens attached to the wall in each examining room that suggests cheaper drug alternatives. In addition, a digital records system in Colorado has cut the number of emergency-room visits by up to 10 percent because nurses staffing call center desks are able to summon updated patient records in seconds, discouraging unnecessary visits which, in the past, might have been based solely on incomplete knowledge of the caller.

Kaiser also has started to reform purchasing, building separate sites so doctors and administrators can order everything from bandages to CAT-scan machines over the Net for volume discounts. Waste and errors are also on the mend: In Kaiser's Northwest region, for example, doctors used to have to repeat 10 percent to 15 percent of patient tests because of lost paperwork or illegible doctors' handwriting. Not anymore. Says Lawrence: "We're just now seeing the first glimmers of the impact IT is going to have on the delivery of healthcare."

But cultural challenges remain—and not just for Kaiser, but throughout the healthcare industry. Technology progress, Lawrence insists, cannot remain the sole call of physicians—a traditionally independent, authoritative group at continuous cross-purposes with IT projects aimed at creating tighter accountability, efficiency, and caps on doctor fees and drug prices. "You can't have IT unless you have strong leadership," says Keith Fraidenburg, vice president of education at CHIME, the College of Healthcare Information Management Executives. "Many hospitals and health systems use IT, but their effectiveness is often hampered by dated technology, disparate systems that can't be integrated to achieve efficiencies, and people and processes that undermine IT's potential. This is an industry steeped in tradition and known for physicians with strong resistance to change."

To be sure, at Kaiser, some doctors and nurses continue to balk. When a pilot patient Web site launched in Portland, Ore., Shirley Corbari, a registered nurse and consultant, feared that a patient somewhere with a 103-degree fever or shortness of breath would send a "nonurgent" e-mail message to a Kaiser nurse asking for advice—when what he or she really should have been doing was calling for an ambulance. Specialized training of nurses and doctors on the systems has diminished such fears, along with new Web site designs that promote electronic messaging only for members interested in preventive care. But hurdles remain. Says Lawrence: "Much of this IT rollout has had to go one doctor at a time."

The IT staff also has had to take its lumps. CIO Sullivan had to impose strict project management controls on a crew that—whether because of the cultural resistance to change or in spite of it—had been largely held unaccountable for cost and time overruns, a situation not unlike other IT operations at other medical centers. Says Princeton's Reinhardt: "IT folks have always so overpromised what IT will do, that everyone feels let down. It is really a story of a great dissonance in the health field between promise and delivery. Healthcare IT people tend to be like drunken lovers at a bar: big talk, modest follow-through."

Life Support

Life Support

Lawrence's attempt at shock therapy was an effort to put some action behind the big talk. His move, he says, triggered "a very serious conversation" within Kaiser over what the HMO's goals should be—and how the organization should be governed.

To centralize and legitimize decision-making, he immediately set up a Kaiser-wide leadership committee, called KPPG—which stands for Kaiser Permanente Partnership Group—with membership drawn from administrators, doctors and IT staff within Kaiser's three central divisions. The group was given four broad initiatives: create a centralized digital medical records system; construct a single consumer Web site; build a system by which medical supplies, from bandages to heart monitors, could be purchased online; and consolidate back-office operations—then more than 50 different payroll and HR systems—onto a single IT system.

Lawrence gave the group a budget, which at first, fell way short of the mark. ("We under-resourced this project in the early stages," he says now.) He also gave Sullivan and the KPPG team a deadline of 2003 to complete most of the changes (which Kaiser says it will miss by one to three years). And he told them the ultimate goal was to digitize patient records. "You can't cut costs and keep care from falling without effective use of IT and digital patient records," he says.

Automating the patient records system had been a goal of Lawrence's since 1984, when he cut short his tenure as a county public health official in Oregon to enroll in Harvard's MBA program. There, he heard some of the nation's most forward-thinking healthcare administrators of the day talk about the then-fledgling managed care movement and also about how automation could help the patient care industry save billions. The advent of the Internet nearly a decade later only stoked his zeal. Lawrence knew that the challenge could be overwhelming. "The digital patient record is extremely expensive and complex," he says. Problems with incompatible legacy systems, the sensitivity of medical data and other administrative headaches had also discouraged others from trying reforms. But Lawrence was convinced it was the only way to keep Kaiser in the game for decades to come. "It was a huge, bet-the-farm decision," he says.

To pull it off, though, would require not only technical integration, but management savvy. Shortly after issuing his IT ultimatum, Lawrence initiated his 80-20 Rule for managing change—namely, a strategy to identify the most important and useful features in an IT application and then drive their adoption while disregarding the less important ones. Says Lawrence: "Many IT strategies fail because they try to cast a net so wide, trying to get everyone's needs met, that it all gets way too expensive and slows down the process."

Case in point: Even the task of choosing which digital records system to start rolling out nationally brought major pain to Kaiser. The KPPG team first narrowed its choice to competing pilot projects from Kaiser's Colorado, Oregon and Southern California regions. But a heated debate ensued: Each region felt its solution was the best one, and each project had advantages—and allies on the KPPG team—that the others did not.

After a two-year standoff, Lawrence once again had to play hardball. Colorado's project would get the nod, he decided—but Kaiser lost two years of potential cost-savings in the process. "In retrospect, I should have moved faster," Lawrence says. The lesson? "Any major [technology] change has to be made by strong leaders, not by lower levels of the organization." And Lawrence says IT reforms in the healthcare industry can't wait for consensus.

Consider Kaiser's ex-CIO Sullivan. After two years, sources inside Kaiser say, the combination of cultural warfare and tight resources finally got to him. Sullivan resigned two years ago and declined to comment for this story. Quipped one source: "He thought the infighting was worse than in Congress."

Rx for Change

Rx for Change

Whatever the cause, Lawrence says he's more sober about the challenges ahead but no less committed. New CIO Clifford Dodd, a former software systems chief for American Express and ex-CIO of Qwest Communications, will not face all of the problems Sullivan did. Kaiser now has one public Web face, down from dozens, enabling it to better control its brand. The HMO, as of December, had fully rolled out its digital patient record system in Hawaii. In Colorado, pilot projects are under way. Kaiser doctors in both states now look at 1 million digital medical records and write 200 to 400 progress notes in cyberspace per month.

Dodd says he will continue working to bring the other regions online and build an IT infrastructure to electronically link its hospitals and clinics with Kaiser's doctors and technical, administrative and clerical employees.

Lawrence is optimistic. He hired Dodd for speed: As Qwest's CIO, Dodd moved two huge back-office systems onto a single platform in six months. But Dodd acknowledges that the cultural roadblocks facing him at Kaiser are considerable. "Success for IT people at Kaiser is never 100 percent," says the HMO's Neidermeier, adding that even Kaiser's push to cut its data centers from 13 to two has been flawed by the fact that there are still a couple of mini-data centers operating in Oregon in violation of the nationalized IT operation. "You decide what fights you are going to fight," Lawrence says, citing his 80-20 Rule and applying it to management in this case. "Most of the switch to centralization has been made."

But Dodd says he's battle-ready for whatever lies ahead. Since coming on board in January, he has formed a half-dozen cross-cultural software development teams within the IT organization. Each team is charged with helping to build consensus from the ground up for a set of technology priorities going forward. "IT can go a long way to knit together warring camps," Dodd says. The carrot: new performance review processes that make cooperation a virtue and reward the goal of measurable ROI over cultural squabbling.

But can Kaiser change fast enough? Wiesenthal, though an IT convert, is less sanguine. Sure, Kaiser's digital revolution could be just what the doctor ordered on the cost-cutting and care front, but the patient—Kaiser, in this case—will need to stay on life support for a while longer. With Kaiser's number-one IT champion, Lawrence, set to retire in June, it's not clear whether the HMO will keep taking its medicine. "It's important that we keep focused on IT," Wiesenthal says. "I don't think anything will happen to distract us, but without [Lawrence], it's still possible we could lose our way." Not a hopeful prospect for an HMO—and a CEO—so keen on recovery.

ERIC W. PFEIFFER is a consulting editor for Forbes ASAP and a contributor to Red Herring magazine. CIO Insight researcher Kathleen Paton contributed to this story.


's Technology Rx">

Kaiser's Technology Rx

Kaiser Permanente is using the Net to improve healthcare while cutting costs


Problem: Prescription drug costs are rising 19% annually.

Solution: Kaiser's in-house intranet suggests cheaper drug alternatives or less dangerous medications to doctors who prescribe drugs to patients.

Payoff: Drug costs have dropped 20% at some clinics, legal costs are down, and incorrect prescriptions have been reduced.


Problem: Fees for services vary widely across the nation.

Solution: The company's physician-care network taps into a database that suggests standard treatments aimed at containing excessive charges.

Payoff: Healthcare costs have been reduced by 5% to 19% across Kaiser's operations.


Problem: Between 10% and 15% of patient tests must be repeated because of lost or misread records.

Solution: Kaiser creates digital records that can easily be accessed by authorized medical personnel.

Payoff: Duplicate tests have been virtually eliminated in some clinics, for a cost-savings of about $1 million in Kaiser's Northwest unit last year.


Problem: Emergency-room visits, among the costliest in the system, are on the rise.

Solution: A digital medical records system is deployed in a Colorado pilot project.

Payoff: The number of emergency-room visits handled through Kaiser's Colorado region decreased by 5% to 10% last year, saving the company millions.




The Strategic Application of Information Technology in Health Care Organizations
By John P. Glaser
John Wiley & Sons Inc., January 2002

Strategies for Healthcare Information Systems
By Robert Stegwee and Ton Spil
Idea Group Publishing, 2000


The College of Healthcare Information Management Executives is a professional organization of healthcare CIOs.

The Healthcare Information and Management Systems Society is a nonprofit organization dedicated to promoting a better understanding of healthcare information and management systems.

This article was originally published on 04-15-2002