 |
 |
 |
 |
 |
 |
 |
“P4P is ridiculous. It’s horrible. It’s expensive, it’s not productive and it’s potentially dangerous.”
“The end result is better care for the patient.”
Doctors have other concerns as well. Some are based on hearsay: P4P programs entice doctors to cherry-pick for healthy patients; the systems are unfair for physicians who work in low-income areas where conditions such as diabetes and obesity are more common; and physicians will be motivated to churn patients through their offices to meet P4P goals. But many physicians’ fears are based on fact—no one knows for sure whether P4P programs improve outcomes or reduce costs, and for the most part, a practice absolutely must be using a solid, reliable electronic medical record system to manage the programs.
Then, there are the philosophical issues. Studies often judge P4P programs based on a few “quality” measures—events that the practice has handled in a favored manner. But are quality measures the same as actual “quality?” If P4P programs cause physicians to be motivated only by cold hard cash, will quality actually go down?
There are no easy answers. P4P programs are evolving, changing every year. And new studies are emerging to shed light on some of the unknowns. Especially during this era of looming healthcare reform, it’s only normal to be concerned that chaos, complexity and unintended consequences might undermine any positive impact of P4P programs. But with the knowledge of its pitfalls and perquisites, doctors can better deal with P4P—or avoid it entirely.
An Imperfect System
Pay-for-performance programs were born out of the desire to
improve medical care. But many physicians see it as “pay-for-reporting”
or “pay-for-process.” That’s because the vast majority of health
plan P4P programs center around a tool designed by the National
Committee for Quality Assurance. Known as HEDIS, the tool is a
set of 71 measures spread across eight domains of care, including
measures involving diabetes care, the control of high blood pressure
and breast cancer screening, along with five other areas. Using
HEDIS, physicians report their actions on each of these measures,
and health plans use the data to identify the high achievers who
perform favored actions and allocate dollar incentives among them.
Indeed, early studies were inconclusive as to whether P4P was really increasing the real quality of healthcare. Recently, however, studies have begun to emerge showing that there is a quality advantage to be gained from P4P, at least judging by physicians ’adoption of favored actions for specific quality measures.
Certainly, studies show that medical groups that implement P4P programs increase their use of specific, good healthcare practices, such as breast cancer screenings and pap smears. The Integrated Healthcare Association P4P program, which is the largest non-governmental physician program in the United States, includes eight large health plans such as Aetna, Anthem Blue Cross, and Blue Shield and more than 225 physician groups representing 35,000 California physicians. In late September, it released 2008 results from the P4P program, showing steady improvements in clinical quality for all measures but one—and that one showed only a marginal decrease of 0.4 percent.
A much broader December 2007 study by UCLA School of Public Health researchers finds similar improvements in 11 quality measures in claims data concerning 222,213 patients who saw one of about 2,000 doctors. Published in the journal Health Services Research and conducted by Antonio Legorreta, Amanda Gilmore and collegues, the study gathered data from a non-profit PPO in Hawaii for six years, and compared quality measures for patients seeing P4P-participating physicians and patients with non-participating physicians. Patients who saw P4P doctors received recommended care significantly more often, as defined by each of the 11 quality measures. And if good medical care can really be measured by how many doctors do the favorable thing in quality-measurable situations, then these patients received better care.
But even proponents admit the programs are not perfect. Lytton Smith, MD, a specialist in family and geriatric medicine whose group St. Jude Medical Heritage Group in Yorba Linda participates in P4P programs says, “P4P came out of the movement by health plans trying to figure out a way to measure quality. They were looking for simple, reproducible methods where service codes could be extracted from their databases and analyzed. You can’t measure intellectual activities, such as sitting down with a patient and diagnosing a complex condition, but you can measure treatments such as immunizations, pap smears and mammograms.”
Opponents argue that P4P systems are simply a means of providing a bribe or a tip to a physician for providing good healthcare. “P4P is really unethical,” says Melvyn Sterling, MD, an internal medicine and palliative care specialist in Orange. “When a patient comes to see me with high blood pressure—or any condition— I do my best to control it with the best medication. Patients have the right to expect the physician to do the best for them.”
Danger, Will Robinson!
But even if the programs are not perfect, what is there to
worry about, if studies are showing positive quality improvements?
According to opponents, plenty.
Some worry that P4P programs will change physician behavior negatively. Physicians in P4P may begin to cherry-pick healthier patients in order to artificially inflate their performance scores, for example. Other opponents worry about the opposite effect— that they may get an outlier patient with an extreme illness will cause an overall drop in their performance score.
Still others vastly dislike the idea of reporting patient data to a health plan. “You are required to report patient data to the health plans,” says Dr. Sterling. “But I’m uncomfortable with that. I know third-party payers have wide access to patient data, but I don’t have to be an active part of it.”
Some opponents’ concerns are now starting to gain heft. For example, it’s long been thought that physicians working in lowerincome areas would receive less monetary compensation for an equal amount of work, since patients in those areas tend to have higher incidences of uncontrolled chronic diseases. And often, those patients tend to be less compliant than those in higher income areas.
Newly released results from the IHA’s P4P 2008 program hint that this may be true. The Association tracked composite clinical scores for eight California regions. Of those, Los Angeles and the Inland Empire ranked 6th and 7th, respectively. As a result of the finding, the IHA and Lawrence Casalino, MD, from the Cornell Medical College secured a grant from the Robert Wood Johnson Foundation to address the issue of whether physician organizations located in more affluent areas score higher on clinical quality.
Another concern that is starting to come to fruition is that health plans may be enticing physicians to switch patients’ prescription medications to generic equivalents. In one memo Southern California Physician received, a large medical group in Northern California noted that it had received over $350,000 in incentive payout from one insurer in the past year, simply for increasing prescriptions for generics. While generics are chemically equivalent, switching can introduce the patient’s system to new impurities and different rates of absorption, while confusing patients with different dosing.
More startlingly, the memo stated that to further increase performance payouts in the coming year, the group would begin testing therapeutic drug switching, in which a patient is prescribed a different drug of the same class that is not chemically equivalent. Therapeutic substitutes are frequently older and less expensive, and can have unintended side effects for patients. One recent brandname target of pay-for-performance incentives is the cholesterollowering drug Lipitor. Some California insurers are using financial incentives to encourage doctors to switch patients to simvastatin, a similar—but not chemically equivalent—drug that was sold under the brand name Zocor and is now available as a generic.
Although the drugs are similar, there are very real risks, both that simvastatin will not work as well as the brand name drug, and that patients may suffer a life-threatening reaction to it. A 2007 study published in the British Journal of Cardiology found that patients who were switched from Lipitor to simvastatin experienced a 30-percent increase in risk for major cardiac events or death.
The memo did infer, however, that it was the physician’s decision whether or not to prescribe a theurapeutic substitute. And, in speaking to physicians who either previously have participated or are currently participating in a P4P program, all agreed that the individual physician is still in charge of the patient’s care and makes the decision about what is best.
Getting Up and Running
Another issue for solo and small-group practitioners
is the cost to participate in a P4P program. It doesn’t cost anything
to enroll, but because the reporting requirements are so
demanding, a good EHR system is essential for success. This
is one of the main reasons that large medical practices and
hospitals are the main P4P participants. “Good IT is absolutely
required to move ahead with P4P,” says Leroy Ounanian, MD,
who is the medical director at Beaver Medical Group in Redlands.
“I am in a multi-specialty group with a very robust IT
infrastructure and we still find it challenging,” he continues.
Without a good EHR system and IT support, reporting becomes prohibitively cumbersome. For example, while P4P programs center around HEDIS data, each program offers other incentives. The IHA program, for example, also measures patient satisfaction and IT systems enablement, factoring that into the payouts. To be successful, a physician needs not just to track the clinical data, but to know and track the additional criteria required by each program in which he participates.
Additionally, health plans can use varying standards to gauge success. For example, one health plan may require that a blood pressure reading to be less than 130/90 while another requires it to be 120/80. And to make things even trickier, P4P programs are a moving target, changing each year to add new criteria or updating older criteria.
Even if a small group were to invest in an IT system capable of handling the nuances of reporting, their financial gains may not cover the cost. Volume is one issue—it takes a lot of patients to make real money from the incremental gains of 1–8 percent of fees for properly handled quality measures. In addition, the total number of dollars available for distribution is not adequate to justify the devotion of time, money and effort it would take for many medical facilities to implement a good electronic medical record system.
The Aye Vote
Proponents of P4P programs, however, argue that the goals are
good. “There is nothing wrong with the goals of providing better
diabetic care, colorectal cancer screening, mammograms and
pap smears, for example, so we support P4P programs,” says Dr.
Ounanian. Adds Dr. Smith, “We interview patients and encourage
them to get mammograms and all of the other measured treatments,
as necessary. We think it is simply good medical practice.”
And they argue that medical groups set up the programs internally to be relatively transparent to physicians. For example, Dr. Smith’s medical group receives about $1 million per year in P4P bonuses. But it is the group managers who make calls to patients when they are due for certain tests. “These measures are almost peripheral to me,” says Dr. Smith. “The patient doesn’t have to see me to get a mammogram. The organization encourages them to and we get the reports, following up as necessary.”
Kaiser Permanente has performed extremely well on P4P measurements, even though the group does not participate in any external programs. When a patient comes in, a complete medical history is printed. If there are any overdue tests they are highlighted and the physician responds.
But are the P4P programs pushing extra healthcare on the patient? Dr. Smith doesn’t think so. “You could wait for the patient to request a test,” he says, “but more and more, medical ethics are calling for physicians to do the outreach.”
Dr. Smith claims that most physicians in his group earn between $8k - $10k per year in bonuses based on P4P performance. But different groups distribute the bonuses in different ways. Dr. Ounanian’s medical group—which receives between 1 percent to 2 percent of gross revenues in P4P bonuses each year— rewards physicians based solely on patient satisfaction ratings, not on P4P measures. Dr. Smith says that his group will begin measuring patient satisfaction as part his group’s incentive structure.
The Medicare Mess
The U.S. Centers for Medicare and Medicaid Services set up
its own Medicare P4P program called the Physician Quality
Reporting Initiative, which was mandated by the 2006 Tax
Relief and Healthcare Act. It has been—to put it bluntly—a
nightmare for many physicians. It’s not the idea that is onerous
to physicians, but the implementation.
Take, for example, the experience of Jeffrey Kaufman, MD, a urologist in Santa Ana. When PQRI was first proposed he was then chair of the health policy committee for the American Urologic Association’s western section. As such, he was involved in identifying the urology criteria that PQRI would measure. “Physicians were afraid of PQRI,” he says. “So I put together a nutsand- bolts article on how to participate, including information on the five or six most appropriate criteria, what the codes were, and how to do the reporting. It was a simplistic no-brainer approach.”
With Dr. Kaufman’s system, a physician would only have to report on three or four of the criteria to hit Medicare’s 80-percent goal. And better, it didn’t require an expensive EMR system— he created a checklist form that a coder could use when submitting reimbursements to Medicare. “There’s no penalty for over-reporting,” he says. “So if I see Mrs. Smith three times for incontinence, I report it three times. Let CMS throw it out two of the times.” Despite the guidelines, a small percentage of physicians participated in the voluntary program.
Then the problems started. Physicians were supposed to get feedback on how they did in February 2008 for the 2007 period, and then get paid that spring. But it wasn’t until August of that year that physicians even found out if they had met their goals. Worse, if a physician had not met the goal, they were not notified at all.
Dr. Kaufman claims that out of about 101,000 participants, only half had qualified for payments. “Think about it,” he says. “Those participants were the most motivated and aware of the system,” he says. They were the best of the best. It’s amazing that only half qualified for payments.” When the payments finally went out, most physicians found themselves pocketing only a few hundred dollars for their efforts, although some large groups did quite well.
Physicians who had not qualified for payments still wanted feedback, so that they could perform better in 2008. But it was already well into 2008 by the time CMS set up an electronic system for feedback, and the agency did not include an appeals process. Worse, the electronic system was so arcane and burdensome that many physicians simply gave up. Dr. Kaufman—who had been denied payment despite his own calculations showing that he had overshot the goal—made his way slowly and frustratingly through the system, only to find a screen stating, “No data available.” Contacting the carrier didn’t help solve the mystery, since it had changed from NHIC to Palmetto GBA during the process.
Physicians protested to CMS. The agency agreed to review denials. The catch? Most physicians still have not been paid. And that’s for 2007. Most still do not even know why they had not qualified in the first place.
Even worse, CMS had set aside a certain amount of money in 2007 to be distributed according to a volume-based formula. Any additional distributions for those physicians wrongly denied payment will need to come from some other pool, thus diluting payments for physicians in that year even further. “Now the PQRI program isn’t functioning as it should or could,” says Dr. Kaufman. “CMS asked physicians to perform and held us to a high standard but they haven’t performed very well themselves.”
So if the PQRI process has been so painful and the rewards so few, why has Dr. Kaufman stuck it out? “I’m the point guy for many U.S. urologists,” he says, “so I needed first-hand experience. Urologists see a lot of Medicare patients.” Dr. Kaufman also believes that PQRI will be extended beyond its 2010 cut-off date and that it will be an important part of future CMS healthcare endeavors.
Where’s It Going?
Dr. Kaufman isn’t the only one trying to guess what the future
holds. Some believe that the carrot-and-stick approach is inevitable,
where “good” performance is financially rewarded and “poor” performance
punished. Others take a more optimistic approach, believing
that the programs will evolve to pay-for-outcome or help lower
healthcare costs. “Moving away from pay-for-process and more
toward pay-for-outcome makes sense,” says Dr. Ounanian.
It is clear, however, that P4P programs are evolving. In 2007, IHA health plans added financial incentives for groups that showed the most improvement year-over-year. Health plans are also moving toward even more standardization of P4P criteria, with extra incentives centering on the utilization of those criteria. Others look to Great Britain and deduce that performance measurements will became a substantial part of how a physician is compensated.
It is clear, however, that P4P programs are evolving. In 2007, IHA health plans added financial incentives for groups that showed the most improvement year-over-year. Health plans are also moving toward even more standardization of P4P criteria, with extra incentives centering on the utilization of those criteria. Others look to Great Britain and deduce that performance measurements will became a substantial part of how a physician is compensated.
