April 2008 FEATURE ARTICLES
Health Plans and Technology
Measuring Physician Performance
Building an effective physician performance system starts with transparency.
By Jeff Hanson
Measuring physician performance is right up there with religion and politics as a topic to avoid at cocktail parties. It's sure to raise a few hackles. On one side, health plans, employers and government agencies are concerned about the alarming cost of healthcare and want to ensure they are paying for cost-effective, high-quality healthcare. On another side, physicians are concerned with rating systems that seem to be more about cutting costs than raising quality. These paint-by-number approaches to measuring performance don't adequately account for a patient's level of sickness or whether patients comply with treatment plans. These approaches also often rely on unreliable data, that vary from health plan to health plan, and which don't teach anybody anything.
However, like politics and religion, measuring physician performance is here to stay. It has been going on for well over a decade and its use is expanding. If this effort is to bear fruit over the long term, stakeholders will have to agree on a methodology that effectively and fairly assesses and improves the delivery of healthcare. In the meantime, as we work toward that day, it's essential that the metrics used to evaluate physician performance are entirely transparent.
What's the Motivation?
As healthcare costs rise, the organizations that pay for that care — employers, health plans and governments — are driven to evaluate what they are getting for their money, and to look for ways to control costs and improve quality.
I understand this drive. When I was northeast regional healthcare manager of benefits strategy and design at Verizon Communications, I experienced the frustration of payer organizations seeking some way to justify the rising cost of healthcare with no apparent improvement in the health of employees. At the time, Verizon spent $3.5 billion on healthcare, yet absenteeism and chronic disease persisted.
Employers formed programs, such as The Leapfrog Group and Bridges to Excellence, to influence the affordability and quality of healthcare. Health plans responded by developing incentive programs to reward effective and efficient care. In its "Incentives and Rewards Compendium," The Leapfrog Group lists more than 50 such programs for physicians and 30 for hospitals. Increasingly, health plans create tiered networks that peg some healthcare providers as "preferred," and modify benefit plans to drive patients to these doctors and hospitals. The country's largest payer, the Centers for Medicare & Medicaid Services (CMS), initiated the Physician Group Practice Demonstration to explore the effectiveness of financial incentives for high-quality, cost-effective care.
Long-established physician evaluation programs, such as continuing education and recertification, are evolving and converging with evidence-based medicine measurement systems...which are based on physicians’ actual practices and the disease burden of their patients.
Consumers, like payers, also want information about who will provide them the best care. They have always wanted that information, which they have pieced together from family or friends, referral services or other practitioners. Now that many health plans require high deductibles, consumers want to know where their money is going and whether they are paying for the best care. In addition, more consumers have grown used to having information instantly available to them on the Internet. If they can compare car prices and shop for real estate on a computer, then they expect to find information on the quality and cost of hospitals and physicians there, too.
Governments, both state and federal, have weighed in on making healthcare cost and quality information available. President Bush's Executive Order No. 13410, issued in August 2006, instructs federal departments to implement more transparent and high-quality healthcare by demanding that their contracted hospitals, physicians and other providers adhere to "four cornerstones" of healthcare improvement: the use of interconnected, interoperable information technology; common measures to evaluate performance and cost; transparency through distribution of this information to consumers; and, payment reform.
Performance Measurement
Measuring physician performance is not a new
idea. Physician organizations have developed ways to monitor and
improve performance, including continuing medical education,
which has evolved into "continuous physician professional
development," and board recertification, now referred to as
"maintenance of certification."
The Physician Consortium for Quality Improvement — convened by
the AMA and representing nearly all the national specialty
societies — is dedicated to developing and testing
evidence-based measures of physician performance. The Physician
Consortium has released an initial set of measures endorsed by
the National Quality Forum (NQF) and is currently working to
develop measurement sets for various sub-specialties and
diseases.
Whether pay-for-performance is an effective tool to improve physician performance has yet to be determined. At the moment, there is not much peer-reviewed research on P4P, however, there has been quite a lot of media attention with accompanying commentary and criticism from doctors.
Long-established physician evaluation programs, such as continuing education and recertification, are evolving and converging with evidence-based medicine measurement systems that require life-long learning and which are based on physicians' actual practices and the disease burden of their patients.
Measuring clinical outcomes has also evolved. In 1999, the publication of the Institute of Medicine (IOM) report, "To Err Is Human," exposed the extent and human cost of medical mistakes to a wide audience of healthcare consumers. In the years since the report, every component of the healthcare system (hospitals, health plans, physicians and other providers) has come under increased scrutiny. Several initiatives to address the underlying causes of medical errors resulted from that wake-up call and from the recommendations offered in the IOM's next report, "Crossing the Quality Chasm." That IOM report recommended that healthcare system processes be redesigned following certain rules, among them, shared knowledge and the free flow of information, evidence-based decision making and transparency. The authors of the report urged the healthcare system to focus greater attention on the development of care processes for the 15 most common conditions that afflict people.
Following the IOM report, groups like the NQF, AQA Alliance and The National Committee for Quality Assurance began to develop nationally vetted, evidence-based clinical measures for the most common and prevalent diseases. These are the measures on which physician performance will be evaluated. Currently, health plans may customize indicators of performance to accommodate differences in member population, but eventually one library of clinical measures will dominate.
Health plans mine health claims and billing data, using sophisticated programs to analyze which doctors provide effective and efficient care. They have long had the capability to steer patients to care that they rate as both high quality and sensibly priced. Many have developed physician "report cards" and have initiated pay-for-performance (P4P) programs that offer monetary incentives to physicians who provide high-quality, cost effective care. Whether P4P is an effective tool to improve physician performance has yet to be determined. At the moment, there is not much peer-reviewed research on the subject, however, there has been quite a lot of media attention with accompanying commentary and criticism from doctors.
P4P, apparently, is still evolving.
Shortcomings
If the progression of physician performance measurement follows the usual evolutionary path of the survival of the fittest, a nationally accepted system will eventually emerge without the shortcomings of the current tools.
One obvious shortcoming of the current batch is the sheer number and variety of standards. The good news is that evidence-based clinical standards have been developed for a number of common conditions and diseases and eventually there will be one standard set of clinical measures. To be sure, some diseases and conditions will fall outside the standard guidelines. Treatment of depression, for example, is not as clear-cut as treatment of diabetes.
While standard clinical guidelines are emerging, the method for analyzing data is still evolving and a number of methodologies exist. Even now, CMS and the Agency for Healthcare Research Quality are testing these methodologies in different programs and comparing the outcomes. Eventually, like the measures themselves, a consensus will emerge on a standard methodology.
At present, many health plans have physician report cards based on an assortment of rating systems, the variety of which can frustrate physicians who no longer wish to accept 10 different (and often conflicting) report cards from 10 different payers with whom they have contracts. If their performance is to be measured, they want one report card based on one set of indicators. That demand challenges payers to pool information they have never shared before, and, in fact, have closely guarded. However, payers are already joining regional collaboratives, such as those in Massachusetts, California and New York. As physician performance measurement evolves, a single report will be a reality.
Another bump in the evolutionary path to effective and fair physician performance measurement is the patchy capability of collecting all the data points that measurement requires. Adoption of healthcare information technology at the physician practice level is slow, due, in no small part, to its cost. In addition, there is currently no standard coding of patients, physicians and procedures, or clinical status of the patient.
And, some data — such as socio-economic impacts — resist capture all together.
The Right Path
Insistence on a perfect physician performance measurement system should not paralyze us. Rather, we should continue plodding along the bumpy road toward the best solution, doing the hard work and making the compromises that would result in a fair, constructive and useful measurement process.
In the interim, before the industry adopts a standard set of measures and methodologies, data transparency is the key to successful physician performance evaluation. For this to happen, payers and providers must pool data, without compromising patient privacy, to ensure that physicians are evaluated on an adequate volume of results. And, the data must be trustworthy.
Performance measurement is one of the four cornerstones of healthcare system transparency that the President's executive order sets forth as a national priority. It must be a funded mandate. We are not talking about a simple software solution here, but a difficult and complex challenge that requires patience and a robust dialogue among all stakeholders.
Transforming our healthcare system is the right challenge at the right time. It will take a lot of effort. But we have no other choice.
Jeff Hanson, M.P.H., is a vice president with
Thomson Healthcare, which is headquartered in Stamford, Conn.
Contact him at jeffrey
hanson@thomson.com.