Jack Huguenard hears a common excuse when he tells some doctors that they order more tests or perform more expensive procedures than their peers:
“Every doctor thinks they have the sickest patients,” said Huguenard, a physician and corporate medical director of Alliance Blue Cross/Blue Shield in St. Louis.
That excuse is wearing thin, he said.
Sophisticated new computer models–some developed by the same organization that mapped U.S. nuclear strategy in the 1960s–let Huguenard and other managed-care executives analyze millions of patient records, helping them determine the cheapest and most effective way of treating everything from appendicitis to depression.
It’s valuable information for health maintenance organizations and hospital companies, which operate on the premise that patient costs can be analyzed in much the same way as the costs of running an assembly line.
And it’s one reason that the largest U.S. hospital company, Columbia/HCA Healthcare Corp., is paying $1.12 billion for Value Health Inc., whose Value Health Sciences data unit analyzes prescription and hospital data.
Drug companies also are paying big money for the information because they need to know how their drugs are used in a managed-care setting–and whether they’re more cost-effective than the drugs made by rivals.
“We can do some pretty direct analysis of how one company’s drug compares to a competitor’s,” said Robert Navarro, head of the pharmacoeconomics division of Practice Patterns Science Inc. in Maryland Heights, Mo. The company is a fast-growing unit of Express Scripts Inc.
Practice Patterns, formed two years ago, already is “very, very profitable,” said chief executive Douglas Cave, although he declined to provide figures. Practice Patterns’ customers include drugmakers Bristol-Myers Squibb Co. and the Wyeth-Ayerst unit of American Home Products Corp.
Practice Patterns and Value Health Sciences are owned by prescription-benefit companies, which give them access to millions of records showing how drugs are prescribed and used. To that they add millions of patient records turned over by their HMO customers.
Then they run this mass of data through software programs that help sort it into identifiable categories, such as the total annual cost of treating a patient with diabetes or the average number of tests ordered during a routine pregnancy.
Practice Patterns built its software around systems developed by physicians and economists at the University of Washington, while Value Health’s system was developed at Rand Corp., the well-known California think tank. Both organizations were trying to figure out ways to compensate doctors so they’re paid enough to treat a disease without giving them an incentive to scrimp on care or dump their sickest patients.
Less sophisticated systems can penalize doctors by finding that they should be paid the same amount for treating a routine case of pneumonia, for example, as for a pneumonia case associated with lung cancer.
“We want to give them appropriate incentives to care for all types of patients,” said Doug Conrad, an economist with the University of Washington Department of Health Services.
Of course, the new models also help HMOs and other companies determine whether physicians are performing too many procedures. Value Health Sciences, for example, recently performed a study for a large physician group and found that women in New England were receiving “five or six times the number of ultrasounds for a normal pregnancy.”
“They used the data and went back to their own members and said `Look guys, here’s an opportunity to be more compliant with what is generally accepted practice and save money,’ ” said Jackie Kosecoff, president and co-chief executive of Value Health Sciences.
Practice Patterns took a closer look at another frequent target of cost-cutters, expensive new anti-depressants such as Eli Lilly & Co.’s Prozac. While the drugs can cost twice as much as older anti-depressants such as benzodiazepine, Practice Patterns found that they’re associated with lower overall medical costs because fewer patients show up in emergency rooms.
“It turns out the main cost driver (with mental illnesses such as depression) is emergency-room visits,” said Practice Patterns’ Navarro.
While the drug studies are valuable, HMOs still primarily use the data generated by Practice Patterns and other companies to weed out expensive and inefficient physicians. That task has never been easy–doctors are notoriously resistant to being told what to do–but having more exact information helps, experts say.
“Physicians don’t need to be hit over the head,” said Dr. Ron Leppke, chief medical officer at PhyCor Inc., a Nashville company with more than 3,400 physicians under contract. “What they need is valid data and seeing someone else within the family they trust is doing it better.”
