Data Analysis Can Show Whether Bills for Treatment Exceed the Average : Health-Care Expert Helps Companies Save Millions in Insurance Costs
Robert Chernow is a man who thinks about medical treatment in terms of averages.
“I’m not trying to be cruel about this, but I’m forced to look at things in large numbers,” he said.
Chernow is the founder and president of Corporate Health Strategies, a New Haven, Conn.-based firm that helps corporations control the costs of employees’ health coverage.
“This is a new industry. There was always an industry on the other side to help the hospitals, but until recently there really hasn’t been anything to assist the corporation,” Chernow said. He estimated that CHS has only “three or four” competitors in the field of health-care data analysis.
Started Four Years Ago
Chernow started CHS four years ago, and sold it in 1982 to Metropolitan Life Insurance. As a subsidiary of the giant insurance company, it works on both Metropolitan projects and as a consultant to a number of other corporations.
When a company hires CHS, Chernow’s staff begins by sifting through data provided by the insurance carrier, trying to determine how much it pays for different types of medical treatments.
“We have one of the largest proprietary hospital-care data bases in the country, with over 50 million claims, updated every year,” Chernow said. “That allows us to take your corporation and compare it--having adjusted for age, sex and type of disease. We can pinpoint where you’re out of line with what’s anticipated.”
That initial process is, Chernow said, “a pain in the neck.” Getting all the numbers in a form the computer can handle involves a good deal of translation and consolidation.
“Probably one time in 10 the insurance company submits the data as we’d like,” Chernow said. Medical codings vary from one insurance company to another, and often one corporation will use several carriers. Treatment of a single patient probably will show up in several different files, since the doctors and the hospital bill separately.
“If you’re hospitalized and I don’t know what happened in the hospital, I’m in trouble,” said Chernow. “I’d like to know about that as well as what the physician did and what he charged.”
Analysts Take Over
Once the numbers are in place, a team of analysts takes over. The groups at CHS include a computer specialist, a doctor or nurse and a public health expert “with a minimum of a master’s degree, although PhDs in public health are easy to get these days,” said Chernow, who studied for his own public health doctorate at Yale.
The analysis pinpoints areas where a corporation’s health costs are out of line with average bills for similar treatment.
“For the same process, the difference in what physicians will charge, and hospitals will charge, is astonishing,” Chernow said. “You can see swings of 50%.”
Once a company knows where its costs depart from national norms, “there are 100 different techniques” to use in attacking the problem, Chernow said.
One method is to require second opinions. If urinary procedures in one area appear to cost much more than average, it may be because the most widely used local surgeon is overcharging, Chernow said.
“The recommendation might be that certain urinary procedures should go on a second surgical opinion system.”
Team of Consultants
CHS provides a team of consultants who visit doctors and hospital administrators who appear to be special problems. Often, the consultants will use computer programs to give physicians on-the-spot estimates of how their own charges for different procedures differ from their peers’.
If the doctor argues that his patients are older or sicker than the norm, the computer can quickly factor in those differences.
The surplus of doctors in many parts of the country has helped CHS persuade providers to pay attention.
“In the old days, the doctors didn’t have to be bothered. They’d say it was none of your business,” Chernow said. “They used to be guaranteed a living. But in certain metropolitan areas I don’t think that’s true any more.”
Once CHS has visited a physician, it sends him or her new information every three to six months on treatment costs in the area.
“We let them know how things stand. They practice in isolation--they don’t know,” Chernow said.
The computer systems and database also allow CHS to provide companies with models of their health-care plan. Corporate executives can fiddle with them to their hearts’ content to see how changes might affect the bottom line.
Chernow recalled one large company that was insisting, during labor negotiations, that it wanted to increase the amount of medical costs workers had to pay themselves. They wanted to increase it to $1,500 from $1,000.
Workers Opposed
The workers were so vehemently opposed to this, they were willing to strike rather than agree.
“Our model indicated they’d save only about 1% to 1.5% in costs,” he said. “They were shocked. They realized it just wasn’t worth it.”
CHS bases its analysis of hospital costs on Diagnosis Related Group technology, the same system instituted recently for reimbursement of Medicaid and Medicare payments.
DRG analyzes the average costs of a given treatment based on an array of different medical and demographic factors.
Some health-care professionals complain that the DRG system encourages hospitals to release patients before they are ready to leave, since those whose costs fall under the average rate are rewarded.
Chernow, who worked on one of the early DRG studies while he was at Yale, is a supporter of the concept.
“The system is far from perfect. But I’ve seen more happening in one year than in the previous 10 years,” he said. “The average length of hospital stay has already dropped one day.”
Forced to Leave
There have been instances, he admitted, of patients who were forced to leave while they still needed hospitalization.
“We’re finding horror stories, but you’re forgetting the middle ground. So many people were staying longer than necessary and a hospital is not a healthy place to be. Instead of looking at the exceptions, I’m looking at the middle. And things go very well in the middle.”
One of Chernow’s favorite methods of controlling hospital costs is the preferred provider organization--a program it developed with Metropolitan Life.
The idea is to identify the most cost-efficient hospitals in a metropolitan area, and give insured families an incentive to use them.
“In Dade County, we were able to identify the hospitals that were most and least efficient for the county school board,” Chernow said. “They signed contracts with nine of the 45 hospitals in the area.”
Although the school board had a fairly high deductible and required a substantial co-payment from its employees, Chernow said, it waived both for patients who used the PPO hospitals.
“They’re so much more efficient on average, we can waive that large amount of money and still come out way ahead. And the market share for the preferred hospitals has more than doubled.”
Natural Response
Chernow sees the PPOs the DRGs, and all the alphabetic attacks on health costs as a natural response to a health-care system that has ballooned on federal funds since the implementation of Medicare in the 1960s.
“More care is being provided by almost all standards than is medically necessary,” he said.
Metropolitan’s PPO plan requires doctors to get approval from the insurance company for any elective surgery.
“It’s the sentinel effect--the doctor knows that before he gets paid he has to talk to a clinician,” Chernow said. “A big component of the review program is to make sure people really require hospitalization and couldn’t be treated as out-patients.”
One way hospitals try to get around new reimbursement regulations, Chernow said, is to “give you a big discount, then admit twice as many patients as before.”
The health-care industry is gradually abandoning the practice of requiring hospitalization for any form of surgery, Chernow said.
“But we have to create the right incentives. If Fred has a bunion and the only way I can get paid for the treatment is to hospitalize him, I’ll hospitalize him.”
Chernow is irritated by long lists of figures that show one patient after another hospitalized for exactly the length of time a benefit plan pays the bill.
“The plan pays for four weeks for alcoholism treatment, and we’d see this person hospitalized 28 days. Wouldn’t it be better to get them quickly stabilized and into an out-patient center?”
But as the nation’s experience with deinstitutionalization of mental patients suggests, the clinics, visiting nurses and outpatient care facilities needed to follow through on such ideas often never get established.
“In many cases we don’t have the backup facilities,” Chernow admitted.
