Quality with teeth: Washington medical groups join with Premera Blue Cross to quantitatively measure quality and quality improvement

Health Management Technology, March, 2005 by Robin Blair

The Big Splash

That lasted about a year, after which the participating clinics decided they had more to gain than to lose by sharing information with each other. "At the end of the first year, we decided to share data unblended," says Sollek. "That way, clinics could be public with each other about excellence. If a practice was knocking patients' socks off with high satisfaction levels, other practices could learn from that. That was what I wanted them to decide," he says, stressing that it was a decision the clinics had to make for themselves.

Word spread, and soon more physicians and practices came on board. The project is now up to about 12 participants, after three years in existence. "We created a Quality Metric Forum in Seattle," says Sollek, "but other health plans weren't too interested, so at the time, the physician practices involved in the project agreed that we should publish its results" to date.

"Publishing" meant issuing a press release that identified the project and its initial participants. That in itself indicates what the participating members felt was the project's significance. "We didn't make the score card the big splash," says Sollek. "The big splash was that we all worked together."

Participating medical groups and those involved in issuing Premera's first public statement about the project include The Rockwood Clinic, Physicians Clinic of Spokane, Wenatchee Valley Medical Center, Everett Clinic, Virginia Mason Medical Center, Pacific Medical Centers and The Polyclinic.

Overcoming a Sizable Hurdle

As commendable as health plan/physician practice collaboration is, what makes the project especially noteworthy from an IT standpoint is that the Quality Score Card team wasn't dealing with HMO data. That would be too easy

"Washington is a PPO world," says Sollek, "and that makes it tougher to assign patients to a specific practice." Just to assemble accurate data for analysis required a system of determining and validating which patients "belonged" to which practices, and when they were seeing a provider in that practice versus seeing another network provider but not the primary provider of care. For example, a female patient might have two or three appointments in a year with her primary care physician. She might also have two appointments with her OB-GYN and one screening appointment for a mammogram at a women's health clinic. Yet, she had to be "attributed" to one practice for purposes of data analysis.

The project team devised a validation model founded upon "two or more" visits being an attribution measure, and then tested their patient attributions by validating them with individual clinics and practices. "We went to the large clinics and said, 'We think these people are your patients. What do you think?'" An overwhelming majority proved that the attribution technology was correct.

Measurement, Achievement and Improvement

Quality healthcare is not a stand-alone commodity; it requires collaboration among physician, patient and health plan. While the initial purpose of the Quality Score Card was to measure how often an individual practice delivered healthcare services in accordance with national standards that indicate a best practice, the underlying objective also was to deliver care that would result in improved health status for the patient. It's not just a matter of, "Did you do the test or give the med?" says Sollek, but rather, did it achieve the desired results?