WASHINGTON -- Managed care plans that publicly and consistently report performance measurements are outperforming plans that don't, according to the National Committee for Quality Assurance.

"Accountability really will drive toward better performance," said Margaret E. O'Kane, president of the Washington-based managed care watchdog group.

The NCQA last week released its second annual "State of Managed Care Quality" report, which assesses managed care plan quality by 10 clinical measures of care and by member satisfaction.

Among the report's key findings in analyzing the 1997 and 1996 data, according to the NCQA, are:

* Health plans that submitted data for public release performed substantially better on clinical measures and attained significantly greater member satisfaction than plans that submitted data on condition that they not be identified.

* Health plans that submitted data for two consecutive years, covering performance for 1996 and 1997, reported higher scores and higher overall improvement on eight out of 10 performance measures than the industry as a whole.

* NCQA-accredited plans outperformed non-accredited plans on all clinical and satisfaction measures.

"Our mission at the NCQA is to put this information out there to work with those in the marketplace that are selecting health plans -- purchasers and consumers -- and help them get the information they need to reward quality performers," Ms. O'Kane said.

All plans that publicly report data deserve recognition, she said.

The managed care industry saw gains of 1% or less between 1997 and 1996 in terms of most of the 10 clinical performance measures.

One performance measure that showed a higher level of improvement, however, was the percentage of smokers or recent quitters age 18 and older who received advice to quit from a health plan professional. The NCQA noted that smoking is the leading preventable cause of death in the United States.

The national average for adult plan participants receiving this advice was 64%, compared with 61% last year.

Among plans reporting two consecutive years, the average in 1997 was 66.3%, up 4.5% from 1996.

There aren't many other interventions that have such a significant effect on reducing people's risk of dying within the next decade, said Dr. Jeffrey Harris, director of the division of prevention research and analytic methods for the U.S. Centers for Disease Control and Prevention in Atlanta.

Although the improvement among repeat participants may appear modest, it translates into 1,521 lives saved, Dr. Harris said. If average performing plans reached the level of the highest performing plans on this measure, the impact would be quite dramatic, he said.

The national averages for other clinical measures were: breast cancer screening, 71.3%; Cesarean sections, 20.7%; childhood immunizations, 65.4%; diabetic retinal exam, 39.0%; follow-up after hospitalization from mental illness, 67.3%; prenatal care in the first trimester, 83.1%. Those numbers reflect the percentage of the pertinent health plan population that have received the particular screening, immunization or treatment from their health plans.

The report was based on NCQA's Quality Compass 1998 database, which includes 1997 Health Plan Employer Data and Information Set submissions and accreditation information from 292 managed care plans and identifies them by name. Another 155 plans submitted data for NCQA to use in calculating national averages, but they would not allow their identity to be publicly released. The total of 447 plans contributing data represent 65 million enrollees, or about 80% of the 80 million Americans enrolled in HMOs. Member satisfaction information was publicly reported by 240 plans.

"In many cases, the difference between plans that disclose their performance and plans that don't is dramatic," the report states.

For example, the percentage of children who received appropriate immunizations before age 2 was 69.3% on average among publicly reporting plans, and 57.7% for those not publicly reporting the data.

An average of 76.5% of patients in publicly reporting health plans received treatment with Beta blockers after an acute heart attack, while 65.4% of patients in non-publicly reporting plans received such treatment. "Beta blockers are a remarkably effective, low cost drug that have been shown to significantly reduce morbidity and mortality associated with heart disease, as well as reduce the chance of a second heart attack," according to the NCQA.

Among publicly reporting plans, an average of 73.2% of women between ages 52 and 69 received at least one mammogram within the last two years, while 67.7% of women in that age range in non-publicly reporting plans received mammograms.

The 252 plans that submitted performance data for two consecutive years outperformed national averages on every measure of clinical quality.

For example, the percentage of members receiving Beta blocker treatment after a heart attack was 78.3% for plans reporting for two consecutive years, compared with a national average of 73.8% for plans overall.

In addition, the percentage of women who received at least one Pap test for cervical cancer screening in the past three years was 73.7% for plans reporting for two consecutive years, while the overall national average for cervical screening rates was 71.3%.

Plans reporting for two years also showed greater year-to-year improvement than plans overall. For example, adolescent immunization rates averaged 57.8% for repeat participants, an increase of 4.2% over the previous year. The national average for plans overall was 52.2%, up 0.7% from last year, according to the report.

Also, the increase in cervical cancer screening rates was 2.2% for repeat participants, compared with an increase of 0.7% for all plans.

Wide variations among regions were found, with New England health plans having the highest average on 10 out of 11 clinical performance and service measures. For example, the Beta blocker treatment rate among New England plans was 90.1%, compared with 73.2% in the Mountain region and 60.3% for the South Central region.

Wide variations also can exist within plans in a region, noted Dr. Cary Sennett, NCQA executive vp, at last week's news conference. Looking at three plans in the Mid Atlantic region, breast cancer screening rates ranged from 60% to 80%. If the plan with the lowest rate attained the 80% rate, based on the expected number of breast cancer cases for the plan size, an additional 20 cases of breast cancer could be detected each year, he said.

The NCQA data is valuable to employer purchasers, employees and providers, said Bruce Bradley, managed care director for General Motors Corp. in Detroit. The automaker provides its salaried employees and retired salaried employees financial incentives in terms of lower out-of-pocket costs to join plans that are high performers on quality measures such as HEDIS and satisfaction surveys. "We've seen massive migration from poor performing plans to good performing plans," he said.

GM also uses the data in its discussions with health plans about quality.

"We're seeing behavior changes on the part of providers" because the data is out there, Mr. Bradley said. When health plans see that the purchaser cares about quality, they're more willing to participate with the employer in ensuring that it is delivered, he said.

A spokesman for the American Assn. of Health Plans in Washington said the report clearly shows that the "act of measuring and reporting spurs improvement."

However, she said that the decrease in the number of plans that would provide the NCQA data for public reporting is a concern. "I think that what that tells us is that a lot needs to be changed in the environment," she said.

She attributed the reluctance of some HMOs to publicly report to the lack of data about the fee-for-service sector of the industry and concerns about irresponsible use of data, despite NCQA's efforts to promote responsible use of the information. Both those factors can prevent the data from being seen in it its proper context, she said.

There is a tendency to overemphasize "looking at the measures as a moment in time and underemphasize the role it plays in improvement over time," she said.

The regional variation in plan performance is an issue that relates to the entire health care system, not just managed care, she said. "Managed care organizations are trying to address this very directly," she said, adding that performance measures are a key part of solving the problem.

The "State of Managed Care Quality" is available on the NCQA's web site, http://www.ncqa.org. The Quality Compass 1998 CD-ROM costs $2,500. The electronic data files -- two of which contain plan-specific data; two of which contain summary statistics -- cost between $2,500 and $5,000. Both the CD-ROM and the electronic data files are available through NCQA's Publications Center at 800-839-6487.