The investigation by ProPublica earlier this year of the dialysis industry pointed to federal health officials holding back on releasing outcomes data. But can it be used to 'cherry pick' the best patients to treat?
A recent editorial published on Sept. 20 in The Wall Street Journal entitled "More transparency, better health care," suggested that the health care industry needs to open its medical records for scrutiny, not only for physicians but for patients who deserve to know who provides the best care.
Written by a trio of authors led by George Schultz, former secretary of labor (1969-70), secretary of the Treasury (1972-74) and secretary of state (1982-89), the editorial argues that "fantasy baseball managers have far more data to evaluate players for their teams than patients and referring doctors have in matters affecting life and death." They suggest that one easy step to reform the U.S. health care system is for federal officials to release claims data from the Medicare program. "Controlling costs without compromising quality will require multiple scalpels rather than one blunt instrument," the authors wrote. "Public reporting of performance measures could provide those scalpels by allowing the public to compare doctors and hospitals based on cost and clinical results."
The Affordable Care Act includes a provision that requires that Medicare release claims data—with full protection of patient privacy—to qualified organizations that can analyze the data and publicly report on hospital and doctor performance, the authors say. "This one step could greatly motivate doctors and hospitals to improve the quality and affordability of their care and substantially lower health-insurance premium growth from its current ruinous trajectory."
It’s also an argument made by the reporters at ProPublica, an investigative news group that did a three-part series earlier this year in The Atlantic about poor outcomes in the End-Stage Renal Disease Program. One part of the series discussed the lack of openness by CMS in releasing complete data on how dialysis clinics performed. CMS acknowledged it had the data but said it was reluctant to release it in raw form because it might be misinterpreted.
Some of the 19 online comments on the Journal article echoed the same concern. And the authors said they understood that. "Some physician leaders worry that public reporting based on health-insurance claims data may be inaccurate, inconsistent and misunderstood by the public. To be sure, there is great need in the measurement process for integrity and fairness to physicians." But, they added, "the legislation has provisions to give doctors ample time to review their records and correct errors before the results are released. It also requires full disclosure of, and opportunities to critique, measurement methods."
Among the commentators on the article was nephrologist Richard Amerling. He said, "Assuming health outcomes can be accurately assessed (and this is a huge assumption), the greatest factor affecting them is the underlying health status of the patient. Outcome reporting, and ultimately "pay-for-performance" will inevitably lead to cherry picking of healthier patients by doctors, hospitals and all the new Accountable Care Organizations created by the administration." Amerling is the Director of the Association of American Physicians and Surgeons, which has sued the Department of Health and Human Services over the health care reform law.
Another letter writer also worried about cherry picking: "…what the proposed system fails to make clear is that if physicians are going to be graded, they will alter their behavior. This means they will be more selective in the patients they treat. The obese, hypertensive, diabetic patient who has a higher risk of complications based on pre-existing problems, and who also is less likely to be compliant, will be left out in the cold."
If that writer, Claudio Vincenty, is right, patients with kidney failure, many who have those comorbid conditions, could be the victims of too much data. The ESRD Program entitles all individuals with kidney failure with access to dialysis treatment and transplantation. What if the pressures of pay-for-performance and the new clinical performance measures on their way in the Quality Incentive Program paved the way for dialysis providers to ‘select’ the best patients who generated the most profits?
Like any profession, physicians need to be held accountable for the way they practice. Data on performance – how well patients do under their care – should be made public, particularly in a publically funded program. But let’s use that data to improve outcomes for all patients. That’s the key.