Pay for performance is one of those concepts people either love or hate, but at least they no longer seem to question whether it will take root in the U.S. health care system. Dozens of smaller pay-for-performance programs are already in place nationwide, larger programs are springing up in states such as California and Massachusetts, and now the large shadows of Medicaid and Medicare pay-for-performance programs are looming.
A Commonwealth Fund study published earlier this year predicted that nearly every state will have a Medicaid pay-for-performance program in the next five years, and the first national Medicare program will start this month.
Now the focus of the discussion can shift to the anticipated effects of pay-for-performance systems. Will they improve the quality of care that providers deliver, or will insurance companies use them simply to cut costs? Will they result in more money for providers, or will they simply reshuffle the dollars that are already in the system?
And what about the potential for gaming the system? Gaming — or manipulating elements to increase your return — has been around for as long as there have been incentive schemes.
Now many people are wondering what impact it will have on health care programs as an unintended consequence of pay for performance.
There are two main reasons for gaming any pay-for-performance program, said Twila Brase, president of the Citizens Council on Health Care. Some people simply want to squeeze out more money for themselves, but others have somewhat more altruistic motives.
“Gaming may be the only way physicians can take care of their patients and get paid for it,” she said. “If a difficult patient doesn’t come in as often as pay-for-performance requires, or someone doesn’t take their aspirin as often as they should, the doctor may falsify [performance] results by getting the patient to say what’s needed.”
Gaming benchmarks Pay-for-performance programs haven’t existed long enough in the United States to draw many conclusions about what is happening or could happen here, so the best source on the potential for gaming is a study of a national pay-for-performance program in the United Kingdom.
After a series of smaller projects throughout the country, the National Health Service included pay-for-performance incentives in a new contract it signed with the country’s general practitioners in 2004. Doctors had to report performance against 146 quality indicators.
In the program’s first year, those physicians received an average of $40,000 each in pay-for-performance bonuses — an unexpected level of performance that sent the service’s budget into a deficit. Officials concluded that either they had set the performance targets too low or physicians had gamed the system.
An analysis of the second year’s results indicated that at least some physicians were indeed gaming the system by increasing the rate at which they excluded certain patients from their reporting.
Under the program’s rules, there are legitimate reasons for so-called exception reporting, including the patient’s refusal to submit to treatment and tests that would be inappropriate because of special circumstances such as terminal illness or extreme frailty. But at least a small number of physicians excluded patients from their reporting because they had missed their performance targets, which would have lowered their scores.
However, François de Brantes, national coordinator of Bridges to Excellence, a nonprofit organization that creates incentive programs for physicians, cautions against drawing grim conclusions about the future of pay for performance.
Although it’s hard to pinpoint the percentage of people who spend their time gaming the system, it certainly is not the majority. De Brantes said fewer than 10 percent of physicians were involved in gaming the U.K. incentive program.
“In the four years that [Bridges to E
xcellence] has been working in different regions of the country, we’ve never once encountered a situation where a physician was trying to look good to meet a certain performance threshold,” de Brantes said.
Gaming increases as the stakes increase, he said, and it is prevalent when there are low levels of professionalism and few, if any, controls.
“In health care, there is a high degree of professionalism,” he said, “which means that, on average, you’re likely to see less gaming than in other activities.” Effect of gaming Even as the adoption of pay-for-performance programs accelerates, it could still be some time before the potential effect of gaming becomes clear. In the United States, the industry is still heavily reliant on insurance claims data to assess quality of care, said Dr. John Tooker, executive vice president and chief executive officer of the American College of Physicians.
All that data goes directly to the insurance plans, and they have complete control of it, he said, so it will be a while before the country moves to clinical-based outcome measures.
“We are not close to where the U.K. is on outcomes-based evidence, so it will be years before we know how much of a problem [gaming] is going to be,” Tooker said. “It may become more evident once Medicare starts using claims data for its reporting, but it will be years before we can say the same for pay for performance.”
The extent of gaming’s impact could take some time to reveal itself, said Dr. Doug Allen, medical director of Greater Newport Physicians, an independent California health practice. That is because physicians’ slow adoption of health information technology means many providers aren’t ready for the move to pay for performance.
Will EHRs enable gaming? Ironically, electronic health records systems could make it easier for physicians to outwit pay-for-performance programs.
“Medical assistants or the physicians themselves can type information into the medical record as easily as they can write it on the patient’s chart,” Allen said. “They can forcibly enter something into the record as being done when it hasn’t been, and it will show up in the performance record as something that’s just as legitimate as a real procedure.”
For example, physicians could say they spent more time with patients than they actually did, he said. It would be hard for anyone to dispute it unless they were in the physician’s office at the time.
“That’s why quality submission control will be so important,” Allen said. “And that’s where auditing comes in. It will be the only way to catch discrepancies like that.”
There are several simple kinds of controls that could catch attempts at gaming, de Brantes said. Spot audits, for instance, let participants know that they run the risk of having their records checked, he said. Experience has shown that the approach limits the temptation to commit fraud.
Organizations could also develop strategies for spotting abnormal behavior. Most of the time, a systematic approach to making numbers look good creates a pattern that is easy to spot, de Brantes said. “Just ask the fraud-prevention folks in Las Vegas,” he said.
“So is gaming possible?” he asked. “Yes. Is it controllable? Yes. Is it widespread? No, and it’s likely not to be as long as we keep the controls in place.”
From the battlefield to the home front: Managing medical data
Government Health IT presents Col. Claude Hines Jr., program manager for the Defense Health Information Management System, in this recent InSight eSeminar. Col. Hines discusses the health information technology and tactical challenges faced by the military medical community in Iraq, Afghanistan and other areas of conflict. In doing so, he describes the current information technology solutions for transferring clinical data between battlefield care givers to health care personnel at military treatment facilities worldwide.
