Feb 29,2008 00:00
In 2004, fewer than two-thirds of heart attack patients at Palomar Medical Center underwent a highly recommended emergency procedure to open clogged arteries within two hours of arriving at the Southern California hospital.
A year later, that number had risen to three out of four.
Much of the credit for the improvement was given to an ongoing Medicare experiment that pays extra money to Palomar, in Escondido, Calif., and more than 250 other hospitals for meeting a set of quality standards.
More improvements followed. Now, more patients at Palomar receive antibiotics before surgeries, influenza vaccinations, aspirin for heart problems, advice on smoking cessation and other treatments favored by medical authorities.
The experiment is part of an expanding movement known as pay for performance, which has swept through the nation's health care system in recent years.
In California alone, bonus programs sponsored by seven of the state's largest insurers involve 40,000 physicians and more than 12 million patients. The insurers in 2007 gave doctors about $145 million on top of their basic payments for caring for beneficiaries.
One survey last year identified 148 different programs in the United States, up from 39 in 2003.
Pay-for-performance advocates see the programs as a critical part of efforts to rein in runaway health care costs, reverse declining health trends and reduce medical errors that can be expensive and sometimes life-threatening.
The program in place at Palomar is making a difference, said Opal Reinbold, chief quality officer for Palomar Pomerado Health District. "We wanted to hold ourselves to a higher standard," she said. "What we began to see was that everyone's performance was getting better."
But more than five years into the movement, opinions on the effectiveness of the programs are mixed. Several recent studies have concluded that bonus payments result in only marginal improvements to the health of patients.
A review of 10 pay-for-performance programs by PricewaterhouseCoopers found tremendous variation among how health care providers were evaluated and how bonuses were paid, creating an administrative nightmare for providers participating in multiple programs.
The cost of participating in a program could actually be higher than the value of the bonuses, the report concluded. And while incentive plans for doctors are more fully developed than they are for hospitals, comparing doctors remains difficult, if not impossible, for patients because most pay-for-performance programs keep most of their evaluation results private.
"A report card on (insurer-sponsored pay-for-performance programs) would show mixed performance," the review concluded. "They would earn respectable marks for trying to involve physicians and ... hospitals in this complex and difficult undertaking. However, their performance in other areas would barely receive a passing grade."
Meanwhile, some doctors say recent moves to incorporate cost-efficiency measures into some programs have only heightened their suspicions that the programs could lead to a rationing of health care services and higher profits for insurers.
"I think pay for performance is another bureaucratic scheme that will interfere with the patient-physician relationship," said Dr. Donald Palmisano, a New Orleans surgeon and former president of the American Medical Association.
He said many programs appear to be shifting existing funds among health care providers rather than creating sources of revenue for them.
"What I believe is that all of this is an attempt to save money; it's a rationing of care," Palmisano said. "(Insurers) are going to take it out of the pot. That means other people are going to get less money."
Program advocates note that the bonuses help offset the sometimes considerable cost of acquiring electronic medical record systems that can be used to closely track patient care, schedule regular checkups and better manage the treatment of patients with chronic diseases.
But should cash incentives be necessary when doctors are trained to provide the highest level of care possible?
Palmisano and other pay-for-performance critics say no.
"For somebody to come to me and say, 'Here's the deal: We're going to give you more money if you perform better.' I find that insulting," Palmisano said.
The trend is being fueled by employers and the government, which pay the bulk of the nation's health care bill and are demanding more value for their money.
Health care spending in the United States hit $2.1 trillion in 2006, equaling $7,026 for each person and 16 percent of the country's gross domestic product, according to the federal Centers for Medicare and Medicaid Services. Spending is projected to nearly double by 2016 to $4.1 trillion and equal $12,782 per person and 20 percent of GDP.
Health care providers participating in the pay-for-performance programs are typically evaluated on a range of measures that track the number of times they perform certain functions, such as screening for breast cancer, measuring cholesterol levels and providing detailed instructions when a patient is discharged from a hospital.
The bonuses are paid out in different ways. Some programs pay incentives only to the top-performing health care providers while others distribute the funds in tiers or reward doctors and hospitals that show the most improvement from year to year.
Bonus payments to physician groups are modest, ranging from 0.5 percent to 5 percent of the normal compensation that the doctors receive from the insurers, said Tom Williams, executive director of the association.
In the experiment being run by Medicare for hospitals, the improvements have been equally satisfying, said Stephanie Alexander, senior vice president and general manager of health care informatics for Premier, the San Diego company that is administering the program for the government.
Some hospitals have raised their scores by as much as 80 percent since starting the program, Alexander said. "Their improvements are dramatic."
So far Medicare, the federal health care program for the elderly and disabled, has awarded $8.85 million in incentive pay to the program's top-performing hospitals.
However, researchers reporting in The New England Journal of Medicine in February 2007 found far more modest results. When comparing hospitals in the pay-for-performance program with those that voluntarily reported information about the quality of their care without the benefit of incentives, the researchers found that the pay-for-performance hospitals outscored the others in individual measurement categories by a range of 2.6 percent to 4.1 percent.
Those numbers suggest that improvements generated by cash incentives in the current program are marginal at best, said Dr. Evan Benjamin, one of the report's authors and an associate professor of medicine at Tufts University in Medford, Mass.
"All we can say from our study is that financial incentives may be helpful in changing behavior," Benjamin said. "We speculate that if larger incentives are used, it may be possible to see bigger improvements."
Another problem is accounting for differences in patient populations when scoring and comparing doctors and hospitals. For example, a medical group in a poor community where diabetes is prevalent might score lower on evaluations than another group in a wealthier community where diabetes is less common.
"Physicians who attract more than their share of clinically complicated patients may find it difficult to score well on quality indicators that are based on patient outcomes," researchers wrote in a December Robert Wood Johnson Foundation report that looked at some of the difficulties of assessing the success of pay-for-performance programs.
One possible solution is to let health care providers withhold data on patients who suffer from multiple ailments that distort quality care measurements.
That's already been done in the United Kingdom's pay-for-performance program for physicians. But a study of the program's first-year results concluded that some doctors were inflating the number of patients they excluded in order to collect larger bonuses.
In some cases, British physicians excluded as many as 80 percent of their patients, said University of Minnesota health policy professor Jon Christiansen, who co-wrote the Robert Wood Johnson Foundation report. "Clearly, there was some gaming going on," he said.
Dr. Jerry Penso, associate medical director of Sharp Rees-Stealy Medical Group, said pay-for-performance programs should tread cautiously as they refine their methods and expand the list of measures that they use to grade health care providers."I think we have to be careful that we're looking at all sides of the issue. Are the patients getting all the care they need? Are they getting the quality they need, and are they getting the best value?" Penso said. "Those are very challenging things to measure, but we're moving in that direction."