With employers and HCFA watching health plans like a hawk, it's crucial for HMOs and providers to compile evidence of their effectiveness in tracking and treating plan members.
But the darndest things can work against providing an accurate picture of how health plans are doing at one service or another. For instance:
A heart-attack inpatient gets a prescription for a beta-blocker drug but doesn't get to the pharmacy until several days after returning home.
An older adult gets a postcard from his HMO doctor telling him to come in for a flu shot, but during his next visit to the town's senior-citizen center a public-health nurse gets to him first.
Such random turns on healthcare routines don't affect the big picture of health promotion and illness prevention. The people who need flu shots and heart medication still get them.
But in a world where payers want proof of value for their healthcare dollar, anything that happens outside the vigilance of a measurement process amounts to a penalty score for health plans and their providers. That's the effect of not being able to tally the full number of enrollees getting the care for which plans are being judged under the industry-standard Health Plan Employer Data and Information Set.
Under the measurement rules laid down in the original draft of a new and expanded version of HEDIS developed by the National Committee for Quality Assurance (July 22, p. 2), the normal behavior of enrollees in the above examples would have confounded the best efforts of health plans to prove they're doing the job.
But in the final version released late last month, small but significant changes in methods of measuring the incidence of flu-shot and beta-blocker administration gave providers of those services an even break, said Carolyn Cocotas, NCQA assistant vice president for new-measure development.
Cocotas said the changes were typical of the scores of compromises struck throughout the HEDIS process. On one hand, the focus had to be narrow enough to satisfy employers' interests in differentiating among healthcare companies. But the measurement changes also had to be broad enough to keep the ultimate goal in mind-a thorough provision of quality health services.
In the cardiac-care situation, the aim was to measure what percentage of patients hospitalized for acute myocardial infarction were given a prescription for a beta blocker to reduce the risk of a recurrence. Allowances were made to exclude patients who may suffer side effects if they take the drug.
The original stipulation was that a beta-blocker prescription had to be filled within two days of discharge from a hospital where a heart attack was treated. That was to be certain the drug was for cardiac follow-up care and not some other ailment.
But that would put the heat on hospitals to get their mainly elderly patients over to a pharmacy right away or risk a black mark if the patient was a day late. The requirement was "way too restrictive," Cocotas said.
The revision allows the prescription to be filled within seven days after discharge, and it also gives credit for a prescription filled up to 30 days before discharge from the hospital.
Some patients may have a prior prescription that hasn't run out, or a physician may hand out samples to patients while they're hospitalized, Cocotas said. Those arrangements for the needed medicine should be taken into account in judging how well a plan covered the bases on follow-up care, she said.
In the case of the flu-shot measure, the task of capturing the total population of immunized older adults was like hitting a moving target.
Although a principal purpose of HEDIS measures is to detect a direct correlation between health plan actions and value for the premium dollar, the measure was a prime example of a preventive action that can get results outside the formal reach of the plan, Cocotas said.
The usual combing of administrative or medical-record data won't pick up those shots delivered at the senior center or at many other public places that enrollees may go after an initial notice from their health plan puts the thought in their heads.
"These days you can walk into a gas station and get a shot," Cocotas said, citing one example she's come across. Some seniors "will never walk into a doctor's office to get a shot."
The NCQA decided that health plans should collect the flu-shot information using patient survey data rather than administrative or medical records. That will reduce the likelihood that the measure will be underreported.
The new HEDIS already calls for an annual enrollee satisfaction survey, and a question about flu shots will be added to it, Cocotas said.
The measure applies to health plan enrollees who are 65 or older as of the first of the year and who get a flu shot in the last four months of the year.
For that measure, the affected group could be determined simply by age. One other measure involving flu shots-for high-risk adults-had to be pulled back for more work because experts in the measurement field couldn't settle on a way to consistently identify the high-risk population.
And the measure of heart-attack treatment at least included basic agreement on the benefit of the medication to be used. A measure dealing with quality of preventive care for asthma was returned to the HEDIS shop partly because of disagreement over whether the specific medications to be reported in the measure are effective in the treatment of all asthmatics (Nov. 4, p. 16).