Fix For A Sick System

It's not good business to pay a supplier that misses deadlines and ships incomplete orders the same fee that you'd pay one with a sterling reputation for on-time delivery and accurate service. But that's the approach the federal government takes in reimbursing health-care organizations for their treatment of Medicare patients. The fee is the same whether a person leaves the hospital on his own two feet or in, shall we say, a more horizontal position. In Medicare's traditional fee-for-service system, hospitals and doctors are even penalized for doing a good job, because healthier patients mean fewer visits and lower reimbursements.

Something has to change. The federal government's Centers for Medicare and Medicaid Services, known as CMS, has begun a data-analysis project with Premier Inc., a health-care services company. The project is aimed at boosting reimbursements to hospitals that show they offer better patient care and reducing payments to those that fall below certain thresholds. The Medicare Prescription Drug, Improvement, and Modernization Act of 2003, signed into law last week by President Bush, also emphasizes the value of data reporting and analysis in arriving at best-care practices. It offers higher Medicare reimbursements to hospitals that agree to share with CMS data about the outcomes of their heart-attack, heart-failure, and pneumonia cases.

Transforming the almost 40-year-old Medicare program, which accounts for more than half of some hospitals' revenue, is as much a financial necessity as a health-care imperative. Medicare has grown as a percentage of total health expenditures, which added up to $1.4 trillion in 2001, from 10.5% in 1970 to 17% in 2001, says The Commonwealth Fund, a health- and social-research foundation. The Medicare budget is projected to more than double from $239.5 billion in 2000 to $599.1 billion in 2025, according to a January 2001 report from the Urban Institute, a nonpartisan research organization.

The Hospital Quality Incentive Demonstration Project was launched last summer by CMS and Premier's Healthcare Informatics group, which offers data-analysis services to support hospitals' quality- and cost-improvement efforts. The project likely will cost the government between $7 million and $8 million in financial incentives over the three years of the pilot. Two-hundred-eighty hospitals, including Premier affiliates and non-Premier care providers that subscribe to its services, are participating in the program. Collectively, Medicare sends billions of dollars annually in reimbursements to those hospitals, says Mark Wynn, a CMS research analyst with Medicare's demonstration programs group. The potential to save billions of dollars by promoting best practices that reduce patient complications and readmissions more than makes up for paying out a few million in incentives, he says.

"If you reward for quality, everyone has an incentive to improve quality, and that helps keep costs down in the bigger picture," says Larry Grandia, Premier's chief technology officer.

Some health-insurance companies have similar ideas and are starting to analyze patient outcomes at member hospitals. But "no one insurance company has the volume to drive massive changes in behavior of health-care providers," says Ed Howe, president of Aurora Health Care, a not-for-profit provider in eastern Wisconsin that's participating in the program. "The government does have that volume to drive change," not just in Medicare reimbursements but in the industry at large, he says.

Participants in the CMS project began providing data in five clinical areas--pneumonia acquired outside the hospital, coronary artery bypass grafts, heart attacks, heart failures, and hip and knee replacements--in October and will continue to do so for a year. At that point, CMS will analyze the first year's data and set performance benchmarks for the remaining two years of the pilot. It's too soon to tell whether this will result in the rollout of a national incentive program, but there's strong support behind the initiative, starting at the top with the Bush administration, CMS's Wynn says. Premier is owned by 200 not-for-profit hospitals and health-care systems that run 1,500 hospitals and health facilities nationwide. All hospitals in the pilot already have some historical data stored in Perspective, the database service available from Premier's Healthcare Informatics group. Each month, the 280 hospitals electronically send Premier discharge and billing data with details such as the drugs, procedures, and lab tests patients received in the hospital and the time those drugs and tests were ordered.

Premier uses four Oracle9i databases to manage the data. Once the data is mapped and validated, it's transferred via an extraction, transformation, and loading tool from Ab Initio Software Corp. into an IBM RedBrick data warehouse that already has 2 terabytes and 5 billion rows of data. The hospitals can access this data, to compare their performance with peer groups', using MicroStrategy Inc.'s business-intelligence tools. The information pertaining to the five clinical areas under consideration for the Hospital Quality Incentive Demonstration Project also is sent to CMS, which maintains a proprietary data warehouse and evaluates the data on 34 industry-accepted measures, including process steps such as prompt administration of beta blockers and outcomes such as mortality rates.

Depending on the results and the performance benchmarks that are set, participating hospitals have a lot to gain, or lose. Of the participating hospitals, the top 10% of performers in the five clinical areas will be eligible for a 2% increase in Medicare payments for services related to those health problems. The next 10% of best performers will receive a 1% increase in reimbursements. During the first year, thresholds will be established based on outcomes of the lowest 20% of performers. In the third year, hospitals that perform at or below the lowest threshold levels will be penalized by having their reimbursements reduced by 1% or 2% in those five clinical areas, Wynn says. If hospitals fall out of the top-performing slots in any clinical areas in the second or third year of the program, their reimbursements will fall accordingly.

"This program has risk and reward," says Aurora's Howe. But joining the program was voluntary, and Howe says the risk is worth it if it means better care.

Jim Reid, assistant VP of IS change management at St. Joseph Health System, agrees. Three of St. Joseph Health System's hospitals in Orange County, Calif., are part of the pilot. "My own feeling is, if you are taking efforts to provide better care, then incentives can provide funds to do that," Reid says. "The health-care industry, like many others, is under some stress. The ability to have capital available to buy technologies that can improve care is important." St. Joseph is deploying physician order-entry, clinical documentation, and digital radiology systems to improve data collection and better measure quality of care.

But there's one area of weakness when it comes to evaluating how well the data reflects whether a hospital is providing the best care, CMS's Wynn admits. With the exception of knee-replacement patients, the project doesn't collect or take into account data about what happens to a patient once he or she has been discharged from a hospital. Some follow-up is possible, though. Because CMS ultimately gets bills for all Medicare patients, it can track a heart-failure patient who was discharged from a hospital participating in the project and subsequently checked into a nonparticipating hospital with the same problem.

The Leapfrog Group, an organization of more than 145 public and private firms that provide health-care benefits, calls the project a step in the right direction to counter financial disincentives for quality. "Measuring the quality of health care is an imperfect science," says Suzanne Delbanco, Leapfrog's executive director. "We can certainly criticize different quality-measurement methods because they aren't comprehensive enough, but the Premier pilot is one of the most-sophisticated sets of measurements that is feasible today."

After years of haggling over the idea of rewarding quality, Grandia says, the time had come to take action. "When CMS joined the movement, it gave the argument the biggest boost in credibility," he says. And changing behavior to improve quality of care is the real reward.

-- with Beth Bacheldor