Setting the pay-for-performance bar: simple is better
From the October ACP Observer, copyright © 2005 by the American College of Physicians.
By Janet Colwell
With pay-for-performance around the corner or already a reality, physicians are under pressure to set practice standards they can track and measure. But out of the host of national guidelines and health plan measures, how should physicians choose which standards to track? And how do they reconfigure their workflow to make sure standards get implemented?
If you're just getting started, it's wise to begin with just a couple of disease areas and simple goals, said Steven Dosh, FACP, an internist with OSF Medical Group in Escanaba, Mich. His 18-physician practice—part of the Peoria-based OSF Healthcare System—chose to focus on diabetes and asthma. For the past three years, the group has documented whether its physicians meet minimum standards of care, such as when diabetic patients receive HbA1c tests, foot and eye exams, or education about home management.
"We picked minimum standards that everyone could agree on," said Dr. Dosh, who said the practice has not yet set specific outcome measures to track, such as whether a patient's HbA1c is below a certain level. "We decided not to make it too threatening to start out."
'Physicians for the most part are adverse to change and they must be convinced of the need for it.'
—F. David Winter, FACP
In fact, choosing what to measure is often the easy part, said F. David Winter, FACP, outgoing Governor for the College's Texas Northern Chapter and president of the 28-physician MedProvider, a division of the Dallas-based HealthTexas Provider Network, a 375-physician group affiliated with Baylor Health Care System. More difficult, but crucial, may be getting consensus on change in general and how it will benefit the practice.
"The first thing you must do is agree on the concept of standards and on the reasons to move forward," said Dr. Winter, a HealthTexas board member and chair of its quality committee. "Change is the issue: Physicians for the most part are adverse to change and they must be convinced of the need for it." Even then, Dr. Winter added, "it is often very painful to go through."
To start off, pick clearly defined diseases or targets where "there is not a lot of gray," Dr. Winter said. Decide to track how well physicians adhere to adult immunization guidelines, for instance, or care for patients following a heart attack.
Instead of reinventing the wheel, use standards and measures already created by professional societies, health care organizations, nonprofits and government. Look at what's out there—and decide what to keep and what to change to suit your practice needs. ("Standards" state appropriate ways to treat, whereas "measures" are based on standards but also contain operational elements such as what patient populations were included or excluded or coding information.)
This year, for instance, the Ambulatory Care Quality Alliance—a consortium of quality improvement organizations, health plans and national medical groups, including ACP—developed a starter set of clinical performance measures that focus on cancer screening, immunizations and common chronic conditions. (See "Performance measures: What should you track?") Those measures have been endorsed by the National Quality Forum, a nonprofit charged with creating a national strategy for quality measuring and reporting.
"Looking at national standards and evidence-based medicine gives you a starting place, a template," said Dr. Winter. "Then you look at how you can meld those standards into one." Even when the choice of a simple standard seems obvious—because it's been endorsed by quality groups or is already used routinely by physicians—it's important, he said, for groups to go though a collective vetting process.
"Start off with that procedure," Dr. Winter said, "so you get buy-in from everyone."
Or look to local standards and measures. Dr. Dosh's group, for instance, decided to track diabetes standards set by the Michigan Peer Review Organization, the state's designated quality improvement organization for Medicare. The practice also follows asthma standards established by the Michigan Quality Improvement Consortium, a collaborative effort set up to develop a core set of clinical practice guidelines and performance measures.
You can also look to measures from local health plans. Greenhouse Internists in Philadelphia, for example, a general internal medicine practice with four physicians, decided to prioritize national standards by checking with major area insurers—Independence Blue Cross and Aetna US Healthcare—to see what standards and measures the plans would be using in pay-for-performance programs.
"Money depends on their standards," said Greenhouse's Richard J. Baron, FACP. "We knew that any standards they had would be nationally recognized. We wanted to make sure we could document that we met their standards so we could retrieve information for the things they would audit us for."
According to Dr. Winter, that strategy is a safe starting point for smaller groups, although larger groups, like his own, often use more sophisticated standards.
"Health plans usually send us an abbreviated form of what we're already doing," he said. "Their standards never conflict with ours because they tend to be very conservative, non-controversial guidelines taken from national standards."
After gathering standards to use or modify, you then need to establish a formal process for how to consider and adopt standards on an ongoing basis. HealthTexas Network, for instance, has a standing committee that considers new protocols or any change to existing procedures, Dr. Winter said.
Creating a flow sheet
The next step is to make sure the standards you choose are met and appropriately documented. One approach is to create separate flow sheets for disease areas you want to track or customize a general flow sheet to include things like immunizations or smoking cessation counseling. Flow sheets, checklists and tracking sheets are all low-tech tracking devices.
Dr. Baron and his partners once relied on a standardized paper flow sheet—but have since moved to an electronic health record (EHR). One of the EHR's major benefits, he said, is being able to customize flow sheets for different patient populations.
The EHR's "default view," which group physicians can also customize, is the general internal medicine flow sheet with basic checks such as colonoscopy, mammogram, cholesterol, Pap smear, height and weight. The doctors have developed another view for patients with diabetes and expect to create similar views for patients with other diseases prevalent in the practice, including asthma and HIV.
"Flow sheets are how you operationalize standards," he said. "Everyone has a collective commitment to populating the flow sheet accurately as they practice."
You can create a simple flow sheet without an EHR, said Dr. Dosh, who used his practice's billing software to identify patients in different disease groups based on claims data. (See "Patient registries: a key step in quality improvement" in the September ACP Observer.) His staff randomly review patient charts and pull data on diabetes and asthma standards into quarterly reports that show how well physicians are meeting those standards.
A medical records staff person uses the data to identify diabetic patients and inserts diabetic flow sheets into appropriate patient charts. A nurse then enters baseline data into the diabetic flow sheet and tells the physician what needs to be updated during patient visits.
For the physician, updating the flow sheet—noting whether the patient has had a foot or eye exam in the past year, for example—takes only two or three minutes, Dr. Dosh said. And having the nurse look at the diabetic flow sheet before a patient visit saves time because the nurse can alert him to any needed tests. If a diabetic patient hasn't had an HbA1c level checked in the last six months, for instance, the nurse can prepare a lab order in advance for him to sign.
The new workflow is designed to minimize physician time, he said, but may increase the workload on other staff members.
"I can't function with a single nurse—I now have two medical assistants who help me keep patient flow going," said Dr. Dosh. The medical assistants document the patients' flow sheets, fill out any needed orders and gather the appropriate patient handouts so Dr. Dosh is free to focus on the patient.
Without documentation, the flow sheet can't be useful. But it's important not to burden the most highly trained people on staff with what may be clerical tasks.
When Philadelphia's Greenhouse Internists had paper flow sheets, for example, the physicians could easily fill those in as part of a visit or when reviewing correspondence.
Whey they got their EHR, however, documentation got more complicated. Some data, such as lab values from their interfaced lab or office procedures like flu and pneumococcal vaccinations, automatically populate the electronic flow sheet.
But other data—including mammogram and colonoscopy reports that aren't electronically linked to the EHR—must be entered manually, a laborious process. Physicians grew tired of having to click through a series of screens before they could post that information.
"It took us a while to redesign our workflow so the physicians were responsible for the clinical work of reviewing reports, while the secretary was trained to be responsible for the clerical work of posting the results to the EHR," Dr. Baron said. To successfully delegate data entry, the practice had to break down documentation into several different steps and decide who would do what.
While it's important to delegate, it's also crucial to have capable staff well-versed in correct coding. In California, where many practices participate in a pay-for-performance initiative sponsored by the nonprofit Integrated Healthcare Association, coding errors can sometimes snowball into very costly mistakes.
At Affinity Medical Group, for example, an independent practice association (IPA) based in San Francisco's East Bay that links about 200 mostly small primary care groups and 600 specialists, one practice was surprised to receive very low scores last year due to a coding error.
After the practice switched labs, data on the LDL cholesterol levels of patients at risk for heart disease and HbA1c tests for diabetics did not get transferred to the group's tracking system, said medical director James L. Naughton, ACP Member, a general internist with the IPA's Pinole, Calif.-based Alliance Medical Group. "If the lab results are missing," he explained, "it was considered not done by the health plan."
And if a physician incorrectly codes for diabetes, he could inadvertently create a whole new group of patients who appear to be getting inadequate care. If a doctor orders tests on a non-diabetic patient with a family history of diabetes without using the appropriate family history code, the patient will show up in the tracking system as a diabetic missing many required tests.
"If you don't get the code right, you create a huge population of people that look like they have uncontrolled diabetes," Dr. Naughton said, "and you look like a terrible doctor."
Don't expect miracles
Implementing standards is a gradual process that takes efficient teamwork, so expect resistance at first and be patient about getting results.
It took OSF Medical Group about six months to get a flow sheet up and running, Dr. Dosh said. It can take much longer before everyone accepts and uses a new system.
According to Dr. Winter, "it took four to five years before there was an open mind to the issue [of using a flow sheet]." The group relied on a gradual approach over a matter of years—time that groups now may not have.
"Our first tracking sheet took about six months to roll out system-wide," he said. "The first year we strongly recommended it but didn't mandate it; then we mandated it, but still didn't enforce it severely for a while. The process is ongoing and we revisit it every year."
Dr. Winter convinced some network physicians to use a flow sheet to track agreed-upon standards by showing them data comparing the performance of physicians who worked alone vs. those who worked through their staff. In general, physicians who worked as part of a team had much better results for preventive screenings, immunizations and other clinical measures.
Still, expect a complex conversation when you're trying to figure out what can safely be delegated to support staff. "When we started talking about new workflow, we really didn't agree on what could be done by a secretary and what tasks should be done only by a doctor," Dr. Baron said. "Finding those boundaries, especially in the radically altered environment created by implementing an EHR, required a lot of meetings and a lot of give and take, listening to each other's concerns."
Ideally, Dr. Winter pointed out, changing your culture is something you should do before introducing new technology.
"We're in the middle of getting set up for EHRs and what we've learned is that if you think the computer is going to fix it for you, you're wrong," he said. "We're developing the culture of having forms, checklists and reminders without computers, so when the computers do that for us, we'll be used to that."
Before physicians can assess their quality improvement efforts, they first have to agree on which standards or performance measures to track.
Here is a look at some of the starter set of measures approved this year by the Ambulatory Care Quality Alliance, a consortium of purchasers, health plans and national medical societies, including ACP. The full set is online.
- Breast cancer screening: percentage of women who had a mammogram during the measurement year or year prior to the measurement year.
- Colorectal cancer screening: percentage of adults who had an appropriate screening for colorectal cancer.
- Pneumonia vaccination: percentage of patients who ever received a pneumococcal vaccine.
Coronary artery disease (CAD)
- Lowering LDL cholesterol: percentage of patients with CAD who were prescribed a lipid-lowering drug therapy (based on current American College of Cardiology/American Heart Association guidelines).
- Beta-blocker treatment post-myocardial infarction (MI): percentage of patients hospitalized with acute MI who received an ambulatory prescription for beta-blocker therapy (within 7 days discharge).
- Beta-blocker therapy post MI: percentage of patients hospitalized with acute MI who received persistent beta-blocker treatment (6 months after discharge).
- HbA1C management: percentage of patients with diabetes with one or more A1C test(s) conducted during the measurement year.
- HbA1C control: percentage of patients with diabetes with most recent A1C level greater than 9.0% (poor control).
- Lipid measurement: percentage of patients with diabetes with at least one LDL-C test (or ALL component tests).
- LDL cholesterol level (<130mg/dL): percentage of patients with diabetes with most recent LDL-C of less than 100 mg/dL or less than 130 mg/dL.
- Use of appropriate medications: percentage of individuals identified as having persistent asthma during the year prior to the measurement year who were appropriately prescribed asthma medications, such as inhaled corticosteroids.
- Antidepressant medication acute phase: percentage of adults diagnosed with a new episode of depression who were treated with an antidepressant medication and remained on an antidepressant drug during the entire 12-week acute treatment phase. Source: Ambulatory Care Quality Alliance, 2005
Internist Archives Quick Links
Fenway Guide to Lesbian, Gay, Bisexual, and Transgender Health, 2nd Edition
This new edition reflects recent clinical and social changes and continues to present the important issues facing practitioners and their LGBT patients. Read more about the Guide. Also see ACP’s recent policy position paper on LGBT health disparities.
Join Us in Washington, DC for the Most Comprehensive Meeting in Internal Medicine
Register now and enjoy:
Discounted rates, the best national faculty, a wealth of clinical and practice management topics and hands-on sessions! Learn more about the meeting.