When patients need to pick a doctor, there's not much out there for them to compare. They can check board certification, see malpractice claim payouts, and count years of experience, information available on many state medical board websites.
But a RAND study from Massachusetts says that the limited amount of publicly available information doesn't correlate with higher quality practitioners, at least according to one set of measurements commonly used. The report is published in Monday's Archives of Internal Medicine.
"Our big take home message is that the types of physician characteristics that are widely available don't predict which physicians will deliver high quality, evidenced-based care," says lead author Rachel Reid, a medical student at the University of Pittsburgh School of Medicine. "What this says is that you can't judge a book by its cover."
She called for more publicly available, accessible and understandable measurement tools the public can access, such as those used by health plans to reward physicians, or those used by HEDIS, TheHealthcare Effectiveness Data and Information Set.
Reid and colleagues used 124 process measures, called RAND's Quality Assessment tools, to score each of the 10,408 Massachusetts physicians based on claims for 1.13 million patients in that state. The process measures included such checklists as whether a pregnant woman being seen for prenatal care is given a blood test to check for anemia, or whether a doctor orders an INR blood coagulation test before giving a patient warfarin.
The patients were enrolled in one of four Massachusetts commercial health plans from 2004 to 2005. And the physicians were evaluated o the care provided for 22 common medical problems, such as diabetes and heart disease.
Then they paired those tools with how those doctors look on the Massachusetts Board of Registration in Medicine's website.
They were surprised, however, that there was no consistent association between the number of malpractice payments or disciplinary actions against physicians and the quality of care they provided, based on the RAND measurement tools. Likewise, they did not find any difference between physicians' years of experience and quality.
"Thus, there is little evidence to suggest that a patient will consistently receive higher quality care by switching to a physician with these characteristics," said the authors wrote.
While there was some correlation between board certification and quality, it was very small, Reid said.
"Few characteristics were consistently associated with high quality care and those we did find were so small in magnitude that they are not significant in a practical sense," said study co-author Ateev Mehrotra, an assistant professor at the University of Pittsburgh School of Medicine and a RAND researcher.
Overall, the physicians studied provided about 63% of the recommended care, according to the RAND process measure tools. Average performance varied by condition, ranging from 31% for cataract care to 68% for care of congestive heart failure.
Reid acknowledges that the study results might be different if done today, when more practitioners have access to electronic medical records and decision support tools on the Internet. More health plans are requiring checklists for a variety of process measures in exchange for better pay.
And, she says, the RAND process measure tool is only one type of tool, and does not measure actual outcomes, or mortality, or whether the patient actually recovers from the illness that prompted the trip to the doctor.