How Physician Quality Measures Are Developed
Quality Feature – April 2013
Tex Med. 2013;109(4):59-63.
By Amy Lynn Sorrel
If physicians measure the care they deliver, the improvements will come. So goes the thinking when it comes to reforming the health care system. But who actually builds those doctor quality measures, and how?
As it turns out, practicing physicians themselves largely drive the metrics they are accountable for as health care inches toward a value-based payment system that factors in both costs and care improvement. Keeping a seat at the quality table will be even more important as measures become more sophisticated and less voluntary, Texas Medical Association leaders say.
"The biggest misconception most doctors have is that the government, and CMS [Centers for Medicare and Medicaid Services] specifically, is the one that makes up these measures, and they do not. The measures come from the experts themselves, usually via the specialty societies," said Ronald S. Walters, MD, a Houston oncologist and member of TMA's Council on Health Care Quality. "These things applied to physicians, or by which we are measured downstream, are what our individual specialty societies thought were very important to measure. And the message there is: If you don't like the measures, get involved with your society. That's where they come from."
Dr. Walters also serves on a measures review committee of the National Quality Forum (NQF). Federal lawmakers created that public-private collaborative to endorse and recommend nationally recognized quality measures for Medicare quality reporting and performance-based programs.
NQF also helps oversee the National Quality Strategy outlined in the Patient Protection and Affordable Care Act (PPACA). A consortium of 52 organizations, called the National Priorities Partnership, carries it out. It includes the American Medical Association and its measure-development organization, the Physician Consortium for Performance Improvement (PCPI), and the Council of Medical Specialty Societies, in addition to representatives of hospitals, employers, health plans, and patients, among others.
Quality improvement is nothing new to doctors, but the effort thus far is largely voluntary, says Robert B. Morrow, MD, TMA's PCPI representative. Over the last decade, and particularly with the passage of PPACA, momentum also shifted from simply improving performance to tying it to payment.
For example, the health system reform law for the first time specifically instructs cancer centers to define and report on quality measures. In 2015, incentives under federal programs like the Physician Quality Reporting System (PQRS) will turn into penalties for nonparticipation. And come 2017, Medicare will use a value-based modifier to adjust physician payments for the quality and cost of care they provide.
Those kinds of mandates now "turn the carrot into a stick and put some muscle behind the need to make sure these are valid measures," said Dr. Morrow, medical director of quality for the Rural and Community Health Institute at Texas A&M Health Science Center.
It Starts With the Evidence
For physicians, that process starts with the medical evidence. Then, organizations like PCPI – composed of physicians across specialties – turn that evidence into a measure they hope ends with gold star recognition from NQF.
To determine whether a measure is needed, "we look at the disease burden: how many patients are affected and how much we are paying as a nation," said Ghassan F. Salman, MD. The Austin internist was TMA's representative to PCPI from 2008 to 2012 and is a liaison to TMA's Council on Health Care Quality.
Most of the early measures, for example, focus on chronic illnesses such as heart disease, obesity, diabetes, and cancer – high-impact and high-cost areas also identified in the national quality strategy.
Once PCPI identifies a need for a measure, the next steps include:
- Reviewing the evidence: What do existing guidelines say? How strong is the evidence? Where are the gaps or variations in care?
- Determining how to measure it: What's needed to reach an outcome? Which patients should be included or excluded? Are there unintended consequences to a measure?
- Deciding where the information for the measure would come from: The chart? The patient?
- Testing the measure on a small scale: Is it reliable? Is it valid? Can it be reproduced?
- Checking for validity: Is the measure accomplishing its intended goal?
After clearing those hurdles, PCPI reaches out to its physician members to test the measure on a larger scale "to see if it did work in real life and if physicians can live with it," Dr. Salman said.
If the answer is yes, the measure heads to NQF for approval, which may adopt it as a national standard. Roughly 60 percent of the physician measures used in Medicare's PQRS program came from PCPI recommendations. While PCPI generally leads the development of physician quality reporting measures, individual medical specialty societies, the National Committee for Quality Assurance, and state-based quality improvement organizations also contribute.
Once at NQF, the Measures Application Partnership (MAP), where Dr. Walters serves, again sifts through the data and "spends a great deal of time looking at measures to see if they are doing what they are supposed to do. We have to make sure the measures are in accordance with the priorities in the National Quality Strategy. Are they relevant? And how are we going to know we are getting better?"
The National Quality Strategy includes six priority areas for measurement and improvement: patient safety, patient engagement, effective communication and care coordination, prevention and treatment practices for the leading causes of mortality, community involvement, and care affordability.
The rigorous process does not stop there. Even after endorsement, MAP continues to evaluate whether physicians use a measure, if it provides value, and if it needs revision. Last December, the committee revised 550 measures.
Dr. Morrow acknowledges criticisms that most quality measures so far tend to focus on care processes rather than care outcomes, for example, whether a physician gave a beta blocker to a patient with heart disease or whether the doctor counseled a patient who smokes on cessation versus whether their condition improved.
While the next generation of quality measures moves in the direction of outcomes tracking, they always will reflect a combination of both elements, he says.
That's because outcomes are difficult to measure, often because of the time it takes to reach that point and the lack of data due to that lag. Process measures, on the other hand, are more attributable to certain parts of the care cycle, and data systems are better equipped to capture that information.
At the same time, the outcome does not always reflect the quality of the actual care, Dr. Morrow added. "You've no doubt heard the phrase: 'The operation was a success but the patient died.' Sometimes people get very good care and have what we all agree is a poor outcome. By the same token, you can have poor care, but a good outcome. We are people, after all."
The Next Frontier
For those reasons, NQF aims for some flexibility, Dr. Walters says. For example, the forum may offer measures for a particular part of care delivery – like a bundle of surgical care improvement measures – and physicians or other health care entities can adopt different combinations to fit their needs.
"Just because we can't measure long-term outcomes doesn't mean we should give up on measuring short-term process measures because some of those we know are correlated with long-term outcomes," he said.
A patient with chest pain is more likely to die if he or she does not get aspirin within an hour. If a surgeon does not clean the surgical area, the patient is more likely to get an infection and stay in the hospital longer.
"Obviously if you don't do something that leads to higher mortality, then the long-term outcome is going to be bad. So what we try to do in the measure world is find examples like that where the train and the linkage of the processes to outcomes is pretty nondebatable," Dr. Walters said.
Other goals for quality measure development include efforts to link them across the continuum of care and allow for patient satisfaction input.
So-called "efficiency" measures are also on the horizon, Dr. Salman says. Translation: costs in relation to quality – measures the federal health reform legislation specifically calls for.
While not the most important outcome, cost is a factor in delivering the best care possible, Dr. Salman says.
It's also a big reason why everyone, from physicians to hospitals to payers, has taken a bigger interest in incorporating quality measurement into a variety of payment- and performance-based programs.
Dr. Salman's practice alone is engaged in half-a-dozen quality incentive initiatives: PQRS, meaningful use of electronic health records, internal clinical quality programs, a disease management registry, and a health plan-sponsored Bridges to Excellence program.
A majority of the measures used in those programs track NQF standards and overlap across initiatives.
Ideally, all players would follow NQF standards, Dr. Salman says. "That would be in the best interest of patients, if they were to ask, and it makes us [physicians] more efficient, which is the goal."
The reality is that's not always the case.
While NQF directly forms CMS measures, private payers and other quality improvement programs are not required to follow them.
That's why TMA continues to advocate that private health plans and other entities employing performance-based programs adhere to evidence-based standards, Dr. Salman says.
Dr. Walters acknowledges physicians' rightful skepticism about the lack of transparency surrounding quality measurement over the years. But he says the problem often stems from not the measures themselves, rather how those using them apply them.
Part of NQF's mission is to overcome the transparency problems and encourage alignment of public- and private-sector performance measurement efforts, he says. And payers, both public and private, are engaging in more coordinated efforts to incorporate quality measurement into payment and performance improvement programs, particularly as more integrated care structures develop.
"There is not an insurance company or employer group or health care organization that is not in these NQF meetings or groups. Everybody is there because everybody recognizes that [improving health care] is a system problem," Dr. Walters said.
The quality measures out there are not perfect, nor are they the same ones that will exist five years from now, he says. But that's not an excuse for physicians to ignore them. It is a reason for doctors to get – and stay – involved in developing them.
Amy Lynn Sorrel can be reached by telephone at (800) 880-1300, ext. 1392, or (512) 370-1392; by fax at (512) 370-1629; or by email.
April 2013 Texas Medicine Contents
Texas Medicine Main Page