Quality Data on Physician Compare Debut to Poor Reviews

February 28, 2014

Healthcare consumerism passed a milestone on February 21 when, for the first time, Medicare posted the performance scores on quality measures for 141 accountable care organizations (ACOs) and 66 group practices on its Physician Compare Web site.

Americans now can see the rate at which clinicians in these organizations prescribe aspirin to patients with diabetes and ischemic vascular disease, for example.

Officials with 2 associations for group practices, however, told Medscape Medical News that it will be several years before the quality data on Physician Compare truly helps consumers compare — and select — their physician.

"For year one, I would take any use of the numbers and comparisons with a big grain of salt," said Jerry Penso, MD, chief medical and quality officer for the American Medical Group Association (AMGA), which represents some of the nation's largest group practices.

"This information can be misleading for Medicare beneficiaries," added Allison Brennan, senior advocacy advisor for the Medical Group Management Association (MGMA), geared more toward small and medium-sized groups.

In other words, there's a quality problem with the quality data, according to Brennan and Dr. Penso. The measures are too few and overly focused on diabetes, the data is dated, and the scores may reflect poor documentation rather than poor patient care. Furthermore, Medicare beneficiaries are likely to find the Web site hard to navigate.

It's reasonable to assume that Medicare's use of quality data will improve over time, given feedback from groups such as AMGA and MGMA. Organized medicine hopes the improvement will come sooner rather than later. Next year Physician Compare, a creation of the Affordable Care Act (ACA), will begin publishing performance data on individual physicians derived from the Physician Quality Reporting System (PQRS).

The Centers for Medicare & Medicaid Services (CMS) acknowledges that what it's doing is a work in progress.

"This is an important first step in publicly reporting quality measures on Physician Compare," said Patrick Conway, MD, the agency's chief medical officer and deputy administrator for improvement and quality, in a news release last week.

The Learning Curve

Prior to last Friday, Physician Compare provided visitors with the kind of information that physicians publish about themselves on their own practice Web sites — training, board certification, hospital affiliation, contact information, and whether they accept Medicare assignment. CMS also noted on Physician Compare whether physicians participated in PQRS and the incentive programs for electronic prescribing and meaningful use of an electronic health record (EHR) system.

Organized medicine has been leery about CMS adding performance scores on quality measures to Physician Compare until it has fixed some problems with the Web site. Last summer the American Medical Association complained to CMS about the site's glitchy search function, inaccurate physician data, and how it can take up to 6 months to correct mistakes.

The complaints persist, but CMS nevertheless has pushed forward with publishing quality data on Physician Compare. ACOs and group practices now can be compared on 5 metrics:

  • Percentage of patients with diabetes who had their hemoglobin A1c under 8

  • Percentage of patients with diabetes who had their blood pressure under 140/90 mm Hg

  • Percentage of patients with diabetes who do not use tobacco

  • Percentage of patients with diabetes and ischemic vascular disease who are prescribed aspirin

  • Percentage of patients with coronary artery disease and diabetes or left ventricular systolic dysfunction who have been prescribed angiotensin-converting-enzyme inhibitor or angiotensin receptor blocker therapy

Performance scores, hailing from 2012, vary widely. Eighty-eight percent of patients with diabetes in Allina Health's ACO had their blood pressure under control compared with 9% of such patients in the John Muir Health Medicare ACO. On the group-practice side, clinicians at DuPage Medical Group in suburban Chicago, Illinois, prescribed aspirin to 100% of patients with diabetes and ischemic vascular disease, while their counterparts at Sutter Pacific Medical Foundation in and around San Francisco, California, posted a 31%.

A visitor to the Web site can draw conclusions from these divergent scores, but Dr. Penso advises caution. After all, physicians in a particular medical group might be prescribing aspirin as they should, but nobody's recording it consistently in the medical record. Or, if they are, the group may be struggling to submit the information to CMS through its "Group Practice Reporting Option" (GPRO) Web site. "That's another learning curve," said Dr. Penso.

Only when a group masters documenting its performance on a quality measure and submitting it successfully to CMS will Physician Compare show a true baseline of clinical quality, he said. Getting to that point of clarity may take only a few years, he added.

Like Rome, Not to Be Built in a Day

CMS intends to add more quality measures to Physician Compare, and again, that can't happen too soon for the medical profession. A patient with prostatitis who's in the market for a new physician might not find the current 5 measures, so heavily skewed toward diabetes, a helpful guide, said MGMA's Allison Brennan.

Dr. Penso at the AMGA said Physician Compare eventually ought to report performance on measures relating to the patient's experience of care.

"It's important to understand if they felt their needs were met," said Dr. Penso. "Was there care coordination between the doctor's office and the hospital and back again? Did they receive education and support?"

Another complaint about Physician Compare is that the data is 2 years old. "The information needs to be up-to-date as possible," said Brennan. Dr. Penso said shortening the lag time to as little as 3 to 6 months "is reasonable and achievable."

Then there is the Web site's user friendliness. Brennan noted that visitors wanting to study an ACO's performance data can look up these organizations only by name, not by region. How likely is it that the average person will know the name of a local ACO? In contrast, a visitor can look up a group practice by region and compare several side by side, but the information displayed does not include the performance scores (it's coming, according to CMS). Right now, the only way to review the quality data for group practices is by opening a spreadsheet under "Download the Physician Compare Database." Will the average visitor be able to find it? "I don't think so," said Brennan.

However, Physician Compare, like ancient Rome, won't be built in a day.

"I think CMS will learn over time to improve the system," said Dr. Penso.

Group-Practice Performers at the Top and Bottom

Table 1. Physician Compare Quality Measures for Group Practices

Percentage of Patients With Diabetes Controlling Hemoglobin A1c

Top 5 Groups

Medical Associates Clinic Dubuque, IA 86%
The Longstreet Clinic Gainesville, GA 85%
McFarland Clinic Ames, IA 83%
Mayo Clinic Health System — Eau Claire Clinic Eau Claire, WI 83%
Scott and White Clinic Austin, TX 82%

 

Bottom 5 Groups

University of New Mexico Medical Group Albuquerque, NM 60%
UT Physicians (University of Texas Medical School at Houston) Houston, TX 60%
North Shore Long Island Jewish Health Care New York, NY 55%
Ohio State University Medical Center, Dept of Internal Medicine Columbus, OH 51%
Bassett Healthcare Cooperstown, NY 48%

 

Percentage of Patients With Diabetes Controlling Blood Pressure

Top 5 Groups

Mayo Clinic Health System — Eau Claire Clinic Eau Claire, WI 88%
Sanford Clinic North Fargo, ND 86%
Medical Associates Clinic Dubuque, IA 84%
HealthTexas Provider Network Dallas, TX 81%
Cleveland Clinic — Wooster Wooster, OH 81%

 

Bottom 5 Groups

University of Texas Medical Branch Faculty Group Practice Galveston, TX 60%
Scott and White Clinic Austin, TX 60%
George Washington University Medical Faculty Associates Washington, DC 58%
UT Physicians (University of Texas Medical School at Houston) Houston, TX 58%
UNC Physicians and Associates Chapel Hill, NC 56%

 

Percentage of Patients With Diabetes and Ischemic Vascular Disease Prescribed Aspirin

Top 5 Groups

DuPage Medical Group La Grange, IL 100%
Mayo Clinic Rochester, MN 99%
Mayo Clinic Health System — Franciscan Medical Center La Crescent, MN 97%
Virginia Mason Medical Center Seattle, WA 97%
Sanford Medical Center Fargo Fargo, ND 97%

 

Bottom 5 Groups

Northwestern Medical Faculty Foundation Chicago, IL 47%
Sutter Gould Medical Foundation Stockton, CA 43%
Sutter Medical Foundation Fairfield, CA 40%
Sutter East Bay Medical Foundation Orinda, CA 39%
Sutter Pacific Medical Foundation San Francisco, CA 31%

 

Percentage of Patients With Diabetes Who Do Not Use Tobacco

Top 5 Groups

Virginia Mason Medical Center Seattle, WA 92%
HealthTexas Provider Network Garland, TX 92%
Cleveland Clinic — Weston Weston, FL 92%
DuPage Medical Group La Grange, IL 91%
Palo Alto Center Palo Alto, CA 91%

 

Bottom 5 Groups

University Physicians (University of Missouri Healthcare) Columbia, MO 68%
Beaver Medical Group Redlands, CA 68%
Southern Illinois University Healthcare Foundation Salem, IL 46%
University of Arkansas Little Rock, AR 45%
Dignity Health/St. Joseph's Hospital and Medical Center Phoenix, AZ 29%

 

Percentage of Patients With Coronary Artery Disease and Certain Other Conditions Prescribed Medicine to Improve Heart's Pumping Action

Top 5 Groups

Dignity Health Medical Foundation (Formerly Catholic Healthcare West) Folsom, CA 100%
Cleveland Clinic Cleveland, OH 97%
Virginia Mason Medical Center Seattle, WA 95%
Loyola University Medical Center Maywood, IL 95%
University of Arkansas Little Rock, AR 95%

 

Bottom 5 Groups

HealthTexas Provider Network DDA Dallas, TX 64%
Trinity Clinic Tyler, TX 63%
North Shore Long Island Jewish Health Care New York, NY 63%
New Lexington Clinic Lexington, KY 59%
Ohio State University Medical Center, Dept of Internal Medicine Columbus, OH 53%

Source: Centers for Medicare & Medicaid Services

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....