California consumers rate physician groups on quality of care in pioneering survey
California Physicians Survey
September 30, 1997
A groundbreaking consumer survey that attempts to measure quality of care where many experts say it counts—at the physician-group level—is being released this month by two California-based organizations. Project leaders say that if physician groups are to compete on the basis of quality in the future, this pioneering survey points to the need for a bigger investment in better information systems (IS) by providers. Many physician groups had difficulty supplying the enrollment data needed to get the study off the ground, organizers say.
Fifty-five West Coast physician groups participated in the Physician Value Check Survey (PVCS). A random sample of 1,000 managed care enrollees from each of the 55 participating groups received a four-page survey with about 50 questions. (A comparison sample of 4,000 non-managed care patients was also surveyed.)
The survey was conducted by the Pacific Business Group on Health (PBGH), the largest health care purchasing coalition in the country, and the Medical Quality Commission, which accredits physician groups. Survey results were scheduled to be released to the public Sept. 16.
Cheryl Damberg, director of research and quality for PBGH, said the business group has been measuring health plan performance for the past half-dozen years. However, because plans frequently contract with the same medical groups and IPAs, the overlap in networks makes it difficult to differentiate quality among plans.
"We wanted to move down to the level at which care is actually delivered," Ms. Damberg said. "This is the first attempt to push measurement down to that level."
In California, she noted, most people select a provider group before they select a health plan.
The PVCS evaluated physician-group performance in four key areas based on the responses of patients in the practices:
• patient satisfaction with physician care,
•longitudinal change in self-reported health and functional status;
•self-reported receipt of selected preventive services; and
•self-reported processes of care received for high blood pressure and high cholesterol.
Many of the questions on the survey were drawn from other instruments like the SF-12, the GHAA Patient Satisfaction Survey, HEDIS, the RAND Health Insurance Experiment and PBGH’s own questionnaires. Several questions had to be adapted to measure group performance and, in some cases, new measures were created, Ms. Damberg explains.
Patients were surveyed by mail, with three follow-up reminders, including a final phone contact, to increase response. Overall response was 54%, just short of the targeted 60% response rate, but a good return nonetheless, according to Ms. Damberg. Patients will be surveyed again in 1998, two years after the first survey.
Information systems
The greatest technical challenge surfaced up front, when physician groups were asked to supply enrollment files containing demographic and administrative data in order to do the sampling.
"Many of these groups have very primitive data systems," Ms. Damberg said. "A lot of groups have difficulty telling you who is enrolled with them, where these people live and how old they are." When those data files were checked, there were huge variations in accuracy among groups, said Lori Bloomfield, the Medical Quality Commission’s chief operating officer. "Some groups had eight bad addresses out of 1,000 and others had 70." A contractor cleaned those files by checking addresses against those from the master files of the Postal Service.
Nancy Oswald, president of the National IPA Coalition (IPAC), said that inadequate information systems are a significant problem for physician groups, particularly the more loosely structured IPAs. Cost is the greatest barrier; many groups simply lack sufficient capital to invest in state-of-the-art computer systems. "The physician sector is really going through a revolution in terms of information," noted Ms. Oswald. "Getting physicians incentivized financially, administratively and conceptually to report the data and gather it and make it accurate and complete is a challenge." IPAC, which represents about 185 physician organizations in 22 states, and the American Medical Group Association, which also represents physician groups, helped with the study design for the project and encouraged groups to participate in it.
Ms. Damberg said that most physicians readily accepted the study design after the investigators described their methodology, case mix adjustment and plans on how they would present the data. Because of issues surrounding patient recall, the researchers, at some point, may conduct a validity study, she adds.
When asked to participate, many physician groups were skittish at first, partly because of the "very public nature" of the study, she said. Ms. Bloomfield said the groups that participated were motivated to compare their performance with that of their peers and to understand any differences. While physician groups often receive feedback from health plans about how they are rated by the health plan’s consumers, results for a given group often vary among plans because of different survey methodologies.
As for the survey results: "We did find variation among groups—sometimes it was significant and sometimes it was not," Ms. Damberg reports. "Clearly, groups, in the minds of the consumers that they serve, do perform differently."
Detailed reports on the survey results have already been provided to purchasers and participating physician groups. A "more digested version" of that report will be used by employers to share with their employees during open enrollment.
More informed choices
Ms. Damberg said she hopes consumers will be able to use the data to make more informed choices when they select a physician. "We’re really trying to increase consumer awareness of quality-of-care information—not only at the plan level, but now at the provider level."
Meanwhile, Ms. Bloomfield said, many groups are already using the data as a basis for internal quality improvement initiatives. Overall response, she said, has been "very positive." According to Ms. Damberg: "Most of the groups were totally stunned to be able to see very detailed information about their performance relative to their competitors in the market place. This has been a very powerful tool for them."
Ms. Damberg noted that most health plans already collect their own performance data on contracting groups. The PVCS reports may not carry much weight with plans initially, although Ms. Damberg predicts that they will in the long run.
"Many of these groups have very primitive data systems. A lot have
difficulty telling you who is enrolled with them, where these people live and how old they are."—Damberg
Project participants express high hopes that purchasers will recognize the value of these data—and use them eventually to reward high performers. "Studies like this dramatize the importance of allowing physician organizations to compete on quality as well as on price," Ms. Oswald said. "They need the incentives from the entire industry to do that."
She adds that the PVCS initiative is a step in the right direction. The groups that decided to participate, she said, understand that in order to get to the place where we want to be—to compete on quality and design interventions at the appropriate site of care—these organizations have to step up and do the best that they can." That’s not easy, she said, but, in the long run, "it’s worth it."
—Mary Darby
Contact Ms. Damberg at 310-396-7036, Ms. Bloomfield at 562-936-1100, ext. 230 and Ms. Oswald at 510-267-1999.
California consumers rate physician groups on quality of care in pioneering survey
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content
