After the novelties of a merger or acquisition wear off, the realities of systems integration come into focus
When Aetna, Inc. went public with news last September that a snafu related to its integration of U.S. Healthcare, Inc. would cause a third-quarter shortfall totaling as much as $105 million, its stock plunged 20 percent in the span of a week. The dive was the beginning of a decline that bottomed out in late December when the stock hit $66--a 44 percent decline from the 52-week high of $118 it had reached in early August.
Though Aetna, Hartford, Conn., is one of the largest healthcare companies to recently experience the highs and lows that growth through acquisition can bring, it isn’t the only one. Throughout the industry the message is clear. Please stand by: We’re experiencing technical difficulties.
Consolidation in the market is creating new technical problems and compounding old ones, says Jennifer Carr, senior analyst with GartnerGroup, Wakefield, Mass. One reason is that when merger-minded companies are signing deals, they are thinking of increased market share, not systems integration nightmares, she says.
"It seems so easy and so glorious at the time--’we’re going to merge and offer all these products and have a wider service area’--but you do need to be able to support them," Carr says. "If you can’t, you’re going to lose the members you hoped to gain."
It’s all right if a systems integration doesn’t happen overnight, Carr says. In the interim, there are relatively easy and inexpensive ways to combine limited amounts of information from two different systems such as creating a data warehouse or creating a common provider or member database.
Increasingly, though, companies are opting to consolidate their offerings by phasing out products and converting members into one plan, according to Carr. While converting members to one plan eliminates the maintenance of multiple systems, many companies such as Harvard Pilgrim and PacifiCare have found that the process has its pitfalls. Others, including Aetna, are grappling with old problems that have been compounded by the sheer enormity of their organizations, Carr says.
Costs of change
Aetna U.S. Healthcare, Aetna’s managed care subsidiary, took a $103 million charge in the third quarter of 1997 because a claims backlog masked a spike in medical expenses. But according to R. Max Gould, vice president of operations and technology, the backlog was the result of a human, not a system error.
As part of a post-acquisition streamlining effort, Aetna U.S. Healthcare embarked on a plan to reduce the number of its customer service centers from 44 to 12, says Gould. "Over much of 1997 we shut down 20 centers. That process caused us to build a backlog. It really wasn’t an IT problem, it was more a matter of moving from a trained staff at one office, a smaller office, to a less-trained staff at a bigger office," he says. "When we built that claims backlog, it caused our actuaries to misjudge the medical costs by about a percent and a half. But when you pay out $7 billion to $10 billion a year in medical costs, a percent and a half translates to $105 million."
By mid-February, Aetna U.S. Healthcare had the backlog under control, Gould says. To prevent another backlog as the company moves ahead with its plan to consolidate its service centers, it will slow down the pace of closures, hire more people to staff the new centers and extend the training time, he says.
By contrast, other integration efforts at Aetna U.S. Healthcare have gone smoothly, according to Gould. In an 18-month period ending last January, the company was able to convert its HMO membership from all but three states, where they are awaiting regulatory approval, to one system. "We consolidated 14 different HMO systems onto the one U.S. Healthcare platform," Gould explains. "That’s given us a tremendous ability to more easily develop products, more easily service national accounts with the same set of plans, some efficiencies in terms of systems and, from our customer viewpoint, the ability to create very national data."
The company has installed an in-house pharmacy system that does online real-time processing of all of its pharmacy transactions, an auto-adjudication system for its indemnity and PPO claims and an Oracle financial system, Gould says. "So we’ve made good progress."
But several big projects lie ahead. Aetna U.S. Healthcare’s members are still on two different systems: indemnity/PPO members are on one, HMO members are on another. Although the ultimate goal is to integrate the systems, Gould says, in the near-term the company is using a data warehouse and common lookup databases to gather certain data elements from both systems. After the 1996 acquisition, the company chose to keep the members on separate systems because each of the platforms was so strong and the company didn’t want to take on too much change at once, according to Gould.
Migrating to one system
Harvard Pilgrim Healthcare, Inc. in Brookline, Mass., also has maintained two systems since the merger that formed it three years ago. Harvard Pilgrim was created by the 1995 merger of Harvard Community Health Plan, Inc., Brookline, Mass., and Pilgrim Healthcare, Inc., Norwell, Mass.
After the merger, officials at Harvard Pilgrim chose to continue offering a Harvard health plan and a Pilgrim health plan. And, with each plan came the legacy system to support it. So, for three years, Harvard Pilgrim has laboriously maintained two separate information systems. Why?
"Well, there’s this concept of operational readiness," says CIO Debra Speight, who joined the company in April 1997. "I question whether we would have been prepared as a company to move any faster. I think now that we have gone through this, if we had to do it again we would do it differently. But I think we took the best approach at the time."
Harvard Pilgrim officials spent most of 1997 getting a new system ready so that by open enrollment in January they’d be able to begin moving Harvard members and Pilgrim members into one, new integrated plan, Speight says. The new system uses an Amisys software product made by HBOC in Atlanta and will replace the Harvard and Pilgrim legacy systems.
By mid-February, about half of all 1.2 million Harvard Pilgrim members had been moved into the integrated plan. "We have approximately another 700,000 left to go," Speight says. "By the end of the year we expect to have everybody migrated into the new product."
Though the migration has gone well so far, its success has come at a price, Speight says. Employees can only bear the crushing workload associated with a systems integration for so long. It’s affected the company’s turnover rate, she says. "I do know some of the cause of why people are leaving is that we have put the organization under a tremendous amount of stress."
At just below what she calls a "national standard" of between 20 and 25 percent, Harvard Pilgrim’s IT turnover rate is higher than Speight would like. And, with the year 2000 problem looming large, employees with legacy expertise are more valuable than ever, she says. So this year, retaining IT employees is a priority, Speight says, and that boils down to two things: money and opportunity.
"I clearly think it’s naive to say people want a nice environment. Money talks," Speight says. "Also, for IT people, you have to be able to create opportunity for them."
This year many of those opportunities will lie in a major effort to streamline operations at Harvard Pilgrim. "We have got to become a little more operationally efficient. But I think one of our issues, again, is we just haven’t had enough people to do that because we’ve been so busy trying to migrate our members and in the conversion of our system."
Officials at PacifiCare Health Systems, Inc. in Santa Ana, Calif., know a little something about complexity and scale. Using an aggressive growth-through-acquisition strategy, PacifiCare quadrupled its size by gobbling up 14 companies over the last five years. In the process, officials have learned a lot about systems integration. They’ve also learned that one of the problems with taking big bites is that sometimes they can be difficult to swallow.
That’s been the case with PacifiCare’s most recent acquisition, FHP International, Fountain Valley, Calif. PacifiCare’s financial performance has fallen far short of analysts’ expectations in every quarter since the February 1997 acquisition. The company took a $14 million charge in the second quarter because of problems with a systems conversion and upgrade in Washington state. And it capped off the year with the unhappy announcement that it would be taking a fourth quarter pre-tax charge to write off good will and restructure the company’s operations, a charge that could total as much as $145 million. It also braced shareholders for a fourth quarter loss and the possibility of its first-ever annual loss.
The bulk of PacifiCare’s problems with the FHP acquisition can be summed up in one word: Utah.
Flashback to a little over a year ago. In a bid to boost membership, FHP kept premium prices low and signed contracts with several new provider groups. But medical costs from the new provider groups were higher than anticipated. So high, in fact, that the commercial product was priced too low to cover actual costs.
PacifiCare planned to correct the problem by increasing premium prices, recontracting with physicians and recontracting with hospitals, says Susan Whyte Simon, a PacifiCare spokeswoman. But things quickly moved from bad to worse.
FHP contracted with Paracelsus Regional Hospital and Medical Center in Salt Lake City on a capitated basis and sent most of its members there when they needed hospital care. But the hospital’s unexpected closure last May forced PacifiCare to reroute patients to hospitals with per diem, not capitated, contracts.
"There aren’t a lot of hospitals in Utah and we didn’t have favorable contracts with any of them," says Whyte Simon. The remaining hospitals were owned by two chains, she says: Columbia/HCA, Nashville, Tenn., which was dealing with its own problems at the time, and InterMountain Healthcare Corp., Salt Lake City, which saw no reason to revise the existing contract, she says.
At the same time PacifiCare administrators were frantically scrambling to recontract with provider groups and hospitals, they also were finishing a systems conversion that FHP had started. During the two months that it took to complete the conversion, provider claims were processed before the once very predictable hospital claims. It wasn’t until the conversion was complete that PacifiCare officials discovered how out of control medical costs actually were. The plan was hemorrhaging red ink.
In late November, PacifiCare announced its intention to divest itself of the money-losing plan. "Operationally we couldn’t fix the problem," says Whyte Simon. By the end of 1997, PacifiCare had lost about $50 million on its doomed entry to the Utah market.
PacifiCare officials say the systems conversion played a minor role in the disaster. "All the integration did was delay our realization of how bad the problem was," says Whyte Simon. "Had we known we were going to have all these problems, we may have delayed it."
In other former FHP markets, things have gone more smoothly, says Jim Williams, PacifiCare’s CIO. In California, for example, PacifiCare converted 800,000 FHP members to PacifiCare plans in about seven months. "I’m happy to say that particular conversion is going per our expectations," Williams says. "It’s on schedule and within budget and we intend to complete the conversion by late spring of this year."
After 14 acquisitions, one thing PacifiCare officials have learned is that maintaining separate systems can be problematic and costly. "We had acquired a couple of plans in the past and taken a ’let’s leave them alone’ mentality and that was a major failure for us," Williams says. "We weren’t able to create the integration of the two companies that we really needed."
In addition, it’s less costly in the long run to operate one system. "We see it as a sustainable, competitive advantage because it gives us a low-cost computing environment that we can build on." It’s an advantage that will be more important than ever as the new millennium approaches, he says.
Companies that choose not to integrate their systems will have to spend huge amounts of human and financial capital to upgrade multiple systems to address the year 2000 problem, Williams says, "whereas we probably will spend somewhere between $3 million and $4 million to complete the trip. And that’s a mere pittance, on a relative basis, to what some of our competitors are having to spend," he says. "It’s kind of like the Midas muffler commercial: You’re going to pay now or you’re going to pay later."
After the FHP conversion is completed later this year, there will be four states left with FHP systems, Williams says. PacifiCare will convert Colorado and Ohio in April and May of 1999, and Nevada and Arizona in August and September, he says. If those conversions happen on schedule, PacifiCare will have converted the 2 million members it acquired with FHP in two and half years.
Through mergers and acquisitions, PacifiCare, Aetna and Harvard Pilgrim all have increased their market share. But they’ve also experienced growing pains that have taxed their human and financial resources.
According to Williams, careful planning can smooth the path to integration on the technical side. "And it’s nice for a chief information officer to be able to say every once in a while that our problems weren’t system related."
The Human Element
SOMETIMES THE MOST VALUABLE RE-
source a company obtains with an acquisition is the people, not the product, says Russell Lewis, chief technology executive for HBOC in Atlanta. But getting them to work together is sometimes one of the biggest challenges of getting systems to work together.
"If you ever want to go through a character-building exercise, you should acquire your biggest competitor," Lewis says. "People who were sworn enemies are now put in a room together and asked to work together and sing Kum Ba Yah."
Letting down that competitive guard is particularly difficult for IT specialists, Lewis says. IT people tend to develop an intellectual and emotional attachment to their information systems and view those developed elsewhere with a fair amount of skepticism, he says. "There’s this ’not-invented-here’ syndrome." But after a merger or acquisition, it’s best to pick one system and move to it as soon as possible, Lewis suggests. "It’s best not to let them continue to compete."
To make sure things go smoothly after the deal is done, Lewis suggests getting a team of key technical people from both companies involved in the process before the deal is done.
To find the key people, don’t ask company executives because they don’t always know. Instead, Lewis suggests asking 10 technical people for a list of the top five people in the department. "Usually they will list themselves first," he jokes. But some names will appear on everybody’s list.
Team members need to have respect from both sides, so try to choose people with broad technical knowledge and 10 years of real world experience, Lewis says. And when choosing team members, remember what they are trying to accomplish. "Never send somebody to do technical due diligence who will lose their job if you don’t pick a certain product."
Expert Advice from the Trenches
BEFORE JOINING HARVARD PILGRIM, CIO Debra Speight spent years in the aerospace industry where she weathered several mergers of a scale much larger than is typical in the healthcare industry. She offers these suggestions to CIOs who are about to face the daunting task of systems integration:
- Do more sooner rather than later. Getting the integration moving as quickly as possible is crucial to its success and the company’s.
- Define a direction and go after it. Merger partners shouldn’t waste precious time trying to placate each other with each minor decision.
- Don’t underestimate the impact of cultural differences. Cultural differences, including corporate language (you say migration, I say conversion) can create huge obstacles.
- Hire engineers. Technology departments need people with technology degrees.
- Be aware that you probably don’t have enough capacity to handle the new network infrastructure. Build beefing-up time and money into your integration plan.
- Treat the merger of your technology as the major corporate initiative that it is. Get all the major players involved and create a plan, a timeline and a budget.
- Speight says, remember that with mergers one plus one rarely equals two.
- CIO, Jim Williams at PacifiCare, also has some advice for information technology officers facing systems integration in their futures:
- Build an infrastructure that can support rapid growth. Upgrade to a scalable technology, for example.
- State a clear, concise vision right away.
- Get systems people involved as early as the due diligence stage.
- Don’t be afraid to outsource.
- some tasks after a merger or acquisition can alleviate the strain on an already taxed IT staff, Williams says. While officials at PacifiCare were converting FHP systems to PacifiCare systems, they also were divesting plans in New Mexico and Illinois. "So we had to support the de-conversion from the former FHP systems to the acquirer’s systems," says Williams. "It really stretched us very thin."
In addition, Williams hired a consulting firm to take over the maintenance of FHP’s legacy systems so that the IT staff could devote its energy to the systems conversion.
Carla Solberg is a healthcare writer in Minneapolis.