NCQA to Add More Measures of the Quality of Cancer Care to its HEDIS Performance Dataset

FORT LAUDERDALE, Fla—For 10 years, the National Committee for Quality Assurance (NCQA) has been overseeing the quality of managed care organizations. “NCQA is the leading accreditor of HMOs, and, through HEDIS (the Health Plan Employer Data Information Set), our impact has stretched beyond accreditation,” Dr. Cary Sennett said at the Fourth Annual Conference of the National Comprehensive Cancer Network (NCCN).

However, he noted, “our work in the area of cancer is truly just beginning.” Currently, NCQA looks only at certain cancer prevention measures—mammog-raphy rate, cervical cancer screening, and rates of smoking cessation counseling—“but we are committed to developing a broader set of measures,” he said.

As executive vice president for NCQA, Dr. Sennett shares responsibility with NCQA’s president for overseeing efforts to help purchasers and consumers make value-based choices between and among managed care plans. NCQA, he said, is a not-for-profit, private sector quality oversight body. “Our mission is to improve the quality of care and service,” he said, “and we do this by providing information about health care organizations.”

He said that NCQA has launched a Cancer Measurement Advisory Panel to assist in the effort to develop measures of quality of care in cancer and incorporate those measures into HEDIS. Such new measures may include colorectal cancer screening, follow-up after mammog-raphy, and stage of diagnosis.

Dr. Sennett said that the lead time from launching a committee to developing measures and deploying those measures in HEDIS is 2 to 3 years, based on work with other advisory panels in areas such as asthma, diabetes, and heart disease.

Changing the Rules

He noted that intense price competition is driving the market for managed care, but that this competition “doesn’t drive organizations to provide higher quality or higher value care; rather, it may be the single greatest impediment to efforts to improve the quality of care.”

Thus, the NCQA’s mission “is really no less than to try to change the rules of engagement in a very fundamental way in the health care marketplace, that is, to try to enable the marketplace to move from one that is price competitive to one that is value competitive.” And to do this, he said, will require information about quality. “Without this information, there is little capacity or possibility of competing based on value.”

He noted that NCQA provides information through two streams. The first is its accreditation program, which assesses the health plan’s organization and how it operates. “It’s a look at the key systems that we believe are essential to providing high quality care and service, but it is a systems view rather than an outcome or results view,” he said.

Historically independent and complementary to the accreditation program is HEDIS, the statistics that NCQA collects to give information on outcomes or results.

‘Rigorous and Discriminating’

Dr. Sennett called NCQA’s accreditation program a “rigorous and discriminating process.” The process is voluntary, and only about half of the HMOs in the United States have asked to be evaluated by NCQA.

When Dr. Sennett started at NCQA 4 years ago, one of every seven HMOs that were reviewed failed outright. “At this point, about 60% of the plans that are evaluated receive our highest award, which we call full accreditation,” he said. “Thus, we have seen some improvement over time. We think that that reflects the seriousness with which HMOs are taking our program and the investments that they have made in improving systems.”

Although the program is voluntary, increasingly the marketplace is mandating that it be done. “Many of the Fortune 500 firms make it a condition of participation that a health plan be NCQA accredited, and beginning in July of this year, we are going to be bringing HEDIS performance data explicitly into the accreditation program. So beginning this year results will count.”

He pointed out that HEDIS is a very broad set of measures, including not only measures of clinical effectiveness but also measures intended to address the full range of issues that are important to consumers and to the purchasers of care such as access and service. “We look at patient satisfaction, financial stability, and other things that matter,” he said.

He called HEDIS “a rapidly evolving work in progress.” It began in 1993 and was adapted to the Medicaid population in 1995. In 1996, it was reconstituted into a single set of measures (HEDIS 3.0) that could be applied across insured populations of Medicaid and Medicare, as well as those of commercial payers.

HEDIS undergoes annual enhancement, he said, and is named by the year of revision—HEDIS 99, HEDIS 2000, etc. The process is an incremental one, but one that is broadly participative, he said. A committee on performance measurement (CPM) sets priorities and establishes measurement policy. Functioning as working subcommittees of the CPM are the measurement advisory panels (MAPs).

Lessons Learned

Dr. Sennett described some “lessons learned” by NCQA in its 10 years of operation. “First of all,” he said, “it may not come as any surprise to you, but it comes as a surprise to many, that there is a huge variation with respect to the performance of managed care organizations, and this variation has a huge impact on the well being of populations covered by managed care organizations.”

In addition, he said, NCQA has learned that there is a growing demand for information about this variability. As an example, he cited the increased interest in statistics on the use of beta-blockers after a heart attack.

The mean rate of beta-blocker use for about 400 plans analyzed was around 61%, he said, “meaning that about two of every five patients for whom a beta-blocker might have been indicated were not receiving that therapy.” An important point, however, is that in some managed care organizations, rates of use approach 100%, while in others rates are far below the mean.

Mammography Rates

More relevant to the NCCN audience, he said, is the HEDIS measurement of mammography rates in commercially insured populations (women over the age of 50). This shows regional variations, ranging from under 70% in the south central states to 76% in New England. “The national average is 71%,” he said. “Again, as with beta-blocker use, there is a tremendous opportunity for improvement in mammography rates.”

These regional variations, he said, “tell us something about the health plans and something about the environment in which the health plans operate.” An analysis of a subset of plans in the same region (south central, the region with the lowest mammography rates), showed substantial variations, ranging from 65% to about 85%.

The implication is that regional factors, while important, do not explain all the variation in performance. An 85% mammography rate, he said, “would put this health plan among the top 10% of all plans in the country. It has managed to succeed despite the fact that it is located in a part of the country where mammog-raphy rates generally tend to be lower.”

Report Cards for the Public

 Public interest in the quality of health plans is, in some respects, at an all time high, Dr. Sennett said, but he also mentioned that the public has a long way to go before it really understands the issues and applies public pressure for change and improvement.

NCQA, he said, is working toward increased public awareness of health care quality. “We produce annual statistics that describe the performance of the industry and make that data available to consultants, purchasers, and brokers through a dataset called Quality Compass, and through the work that we do with the media and with states to produce report cards.”

The media has “grabbed this and run with it,” he said, noting that US News and World Report has, for the last 3 years, produced rankings of HMOs based on NCQA data.

Another public awareness effort involves the Maryland report card, produced for Maryland residents using NCQA data. “This uses a consumer reports type of format that consumers can understand and includes detailed information on a number of preventive care measures for all the HMOs operating in the state,” he said.

Outstanding Issues

Dr. Sennett briefly discussed three important outstanding issues in evaluating managed care. First, he said, the ability to measure managed care is critically limited by the amount of data required to support these efforts and the cost of acquiring such data. “There is a need to profoundly enhance data capture and data management systems,” he commented, “to link measurement to improvement or, looked at another way, to link guidelines to measures. Measurement, I think, creates opportunity and also creates pressure for improvement.”

Second, he said, is the need to accelerate the rate at which measures are developed and deployed. Finally, he emphasized, there is a need to extend and coordinate measurement at other levels of the system.

“The delivery of care rarely takes place at the managed care organization,” Dr. Sennett said, “and it is at the point at which care is delivered that the greatest opportunity for improvement exists. So we need to extend our measurements to those other levels of the system, and we need to do so in a way that is coordinated, logical, coherent and efficient.”