Designing Effective P4P Programs: It’s About Appropriateness
June 2nd, 2008
 by
Howard Beckman
|
Editor’s Note: Today, The Health Affairs Blog concludes a series of four posts on trends in performance measurement and performance-based payment in health care. The series focuses particularly on the increasing emphasis being placed on measuring and rewarding cost-efficiency. Howard Beckman (below) and Arnold Milstein contribute posts today. Last Thursday, James Robinson and Tom Williams weighed in. See also Howard Beckman and coauthors’ recent Health Affairs article Beyond The Efficiency Index: Finding A Better Way To Reduce Overuse And Increase Efficiency In Physician Care.
The pay-for-performance experiment has taught us a few important things. For one, the belief that we can identify the “best” physicians to whom we should preferentially send patients has not worked. Our attempt to define the attributes upon which physicians can reliably be rated has focused more on what can be measured than on what defines the best doctor.
Measures to date focus on what the physician and patient should complete together, rather than defining the core attributes of practices to whom we should send patients. For example, mammography rate is a common measure. But this is as much a patient’s decision as a physician quality issue. Even more troubling is the fact that the use of efficiency indexes to represent a physician’s effectiveness in spending health care dollars has simply not been validated. We are judging and reporting physician’s cost efficiency without reliable confidence in the scores generated. In fact, there is growing concern that this global measure is confounded more by “noise” in the measurement model than actual utilization decisions.
To me, current “value-based purchasing programs,” which use these modest and insufficient measures to judge a physician’s value, are simply a façade for attempting to moderate cost, and they are poorly designed for that task as well. In a nation in the midst of a significant shortage of primary care physicians, trying to shift patients from one physician to another is the wrong strategy. Rather, we need to help each practice be as effective and efficient as possible.
As an example, consider Dr. Leslie Brewster, a busy geriatrician. So what makes her a good doctor? Is it because she sees patients within 15 minutes of their scheduled appointments (a common measure on patient satisfaction surveys), or is it because in the course of a 15-minute scheduled appointment she hears her patient implicitly say he is afraid of dying and explores that with him (which is not a question on satisfaction surveys)? Is it because she has an electronic record in the office, or is it that when a patient finds her speech slurring during the night, the patient knows Dr. Brewster or her coverage team will be reliably available? Is it because her rate of colonoscopies is greater than the 90th percentile nationally, or because she chooses to discuss the risks and benefits of screening for colon cancer with her 70- to 90-year-old patients to try to reach an informed agreement with them about what should be done? In truth, we have a long way to go to know enough about a physician’s practice effectiveness to use copay differentials and public humiliation to disrupt long-standing physician-patient relationships in order to encourage a patient to move to a “more effective” physician.
Of course, each physician should strive to be excellent at all of these dimension of care. However, my work as medical director of a physician organization has shown me that if we want to rely on the doctors that do it all, there will be a very long line outside the doors of a very few physicians. Our data suggest that most physicians — like other professionals — are better at some things and less effective at others. The fantasy that there are a large number of physicians who are head and shoulders above the rest is simply that — a fantasy. My conclusion from nine years of working in P4P is that using reporting programs to identify and promote the best doctors is terribly flawed. With most medical care providers somewhere in the middle, shifting patients between physicians amounts to unprincipled tampering for no proven benefit.
What Is P4P Good For?
So, if P4P is not good enough yet to determine whom a consumer should see professionally, what is it good for? Well, it turns out that based on the public reporting of results and modest financial incentives, physicians will set their sights on improving their performance on a few key measures each year. Remember that most practitioners work in small offices of one to four doctors, that many are not facile with computers, and that many employ office staff who have been with them for many years. And realistically, they don’t have the resources to hire an IT person to create a registry of patients with specific diseases, to set up a prompting system when a patient checks in, or to develop a reminder system that makes sure each patient in need of a preventive service is called or a way to easily check for drug interactions without assistance on their own.
To engage physicians in the transformation of practices to meet the needs of this next decade, a few requirements must be met. First, P4P and other reporting programs must select measures for which solid medical evidence suggests that a requested intervention substantially improves patient outcomes. Second, variation analysis should suggest that there is sufficient room to improve to make each practice’s effort worthwhile. To get from 83% to 88% adherence in mammography screening may not be worth the effort, while improving blood pressure control from 65% to 85% may prevent many strokes. Third, for each intervention selected, the financial incentive for improving performance should at least offset the cost of the intervention. For example, it takes little cost to change a prescribing pattern. One only needs to write a different prescription. But if the intervention is seeing a patient at night to avoid an emergency department visit, the cost of coming in to open the office has to be covered.
In creating meaningful P4P and other reporting programs, the payer’s or insurer’s responsibility is to formulate clear goals for the program and then determine the three areas most in need of improvement. Then, an appropriate reward and reporting program to encourage that improvement can be designed. Using this model, practitioners will come to understand the need for continuous improvement and will be incented to do so.
Often, Improvement Is About The Whole Team, Not Just The Physician
This brings me to the next important observation. As a medical director charged with leading a P4P program from 1999 through 2006, I’ve spent a lot of experience talking with physicians about what they need to do differently to succeed in their improvement efforts. It turns out that often what is needed is a redesign of the way the practice is organized. While the physician is a critical part of the practice’s success, to be successful in the new world of population-based care, the entire practice team has to work together to ensure that practice goals are set and performance against those measures is reported regularly. However, this transition has proven quite difficult for the many physicians who were accepted into medical school because they functioned well as individuals, not team members. In my training, the definition of a “team” was a group of people who did what I told them.
To succeed in the P4P world, a practitioner has to effectively care for a population of patients and make sure that his or her practice population receives all the care that they should, and no care that is unnecessary. In our IPA and in other programs around the country, “higher performing groups” have figured out how to involve practice staff in the creation of solutions to improving performance scores. The physician was not viewed as the solution to every problem or need.
For example, the receptionist can create a database that identifies patients with targeted chronic diseases, or the medical assistants can review vaccination history and give flu shots or a tetanus shot before the physician even sees the patient. This transformation to creating and managing an effective, resourceful, committed team is the task at hand. Antiquated notions of what constitutes a “good” doctor are irrelevant, and programs that attempt to punish the low performers simply make matters worse. Lower-performing physicians and groups need more support and assistance to adjust to the changing demands of patients and payers. Developing ways to provide that assistance is an important challenge for us all to address.
What is needed nationally is a clear and unequivocal commitment to improving quality of care for all Americans. The action plan involves nationally determining what constitutes appropriate care and then helping each practice to get there. The responsibility resides with each community, not physicians alone. Based on principles of honesty, transparency, accountability, and respect, a number of communities have already begun to work together to identify what is needed to improve care and offer assistance to those attempting to make the effort. In Rochester, NY, the RHIO (regional health information organization) project is an effort to share patient information with all those treating a patient, to reduce redundancy and improve each practitioner’s understanding of all the care any one patient has received, regardless of the source. In California, Puget Sound, Wisconsin, and Massachusetts, organizations have formed to collect and share information, in order to promote improvement, not judge and punish.
We have an opportunity now to redesign the delivery of care to match the expectations patient-consumers have and deserve. The effort won’t succeed if we pit high performers against low performers. It requires a collaboration of partners committed to improving care and being held accountable for that improvement. I, for one, look forward to that effort.
