How Our Obsession with “Quality Metrics” Drives Overuse

Written by Stephen R. Smith, M.D., M.P.H., NPA Founding Board Member & NPA-CT Steering Committee Member
Like nearly all clinicians, I get an annual “report card” on my clinical performance based on a set of “quality metrics” with data gleaned from our electronic health records. While many will attest to being driven crazy by these metrics, they may also drive overuse.
An award-winning abstract presented at the recent Lown Conference by Drs. Alan Drabkin, Ron Adler, and NPA activist Stephen Martin, and medical student Courtney Scanlon showed that only 20 out of 65 “quality” metrics included in Medicare’s Merit-based Incentive Payment System met criteria for appropriateness.  Compliance with inappropriate quality measures, the researchers suggest, can lead to unnecessary medical services that can harm patients, waste resources, and get in the way of delivering the right care.
One way these often-inappropriate metrics can drive overuse is to create a defensive mindset in clinicians who are called on the carpet for not doing some clinical test or procedure frequently enough. When I was a medical student on rounds, the attending physician would grill me on which tests I had ordered on the patient we were seeing. He would inevitably come up with a test I didn’t order and then proceed to humiliate me in front of the other team members for my negligence. This practice is known as “pimping.” In a vain attempt to prevent a repeat of the humiliation, I would order every conceivable test on my patients whether or not they were truly needed.
Now we have “electronic pimping” emanating from these poorly conceived “quality metrics.” The result is the same defensive behavior on the part of the clinicians subjected to exhortations to do better on their report cards.
How many of these metrics look at common problems of overuse such as inappropriate use of antibiotics or unnecessary imaging? None that I’ve seen. The wizards in business intelligence should certainly be able to come up with ways to measure overuse starting with the Choosing Wisely lists. These should be matched one-for-one with a reduced set of current “quality” metrics that are firmly based on solid evidence. Perhaps this rebalancing will help convince clinicians that we need to address both underuse and overuse with equal measure.