This piece was originally published in Politico Pro on October 10, 2012
Escalating health care costs have appropriately focused the attention of both parties on the quality of care Americans receive - resulting in the increasingly intensive scrutiny of medical treatments and the health care workers who prescribe or deliver them.
We've become especially interested in "quality metrics," as we seek assurance that the provider, hospital, drug or device will perform according to expectations.
As we've pushed harder on quality metrics, however, we've made an important discovery: All metrics aren't created equal, and we need to be sure that a metric is meaningful before we embrace its use.
A good illustration of this challenge, as I recently described, is the increasing emphasis on certification for all health care workers. Scrub nurses, neurosurgeons, physical therapists, physician assistants - everyone must pass through a gauntlet of expensive certification exams, which are often a requirement for employment, and can also directly impact salary and reimbursement as well.
Certification is presumed to be a proxy for quality, and often touted as a symbol of quality.
But is it?
There just doesn't seem to be a lot of data supporting the value of certification. Not only does there seem to be scant evidence correlating certification with performance in most areas of health care, it's not even clear that the instruments used to evaluate certification are always reliable. (To be reliable, they need to yield the same result on successive administrations, independent of who is scoring it.)
And good luck prying loose any data that exist on the reliability and validity of certification tests. These data tend to be held very closely by the professional societies that administer the tests and confer certification.
Fortunately, there's hope. As our health care system becomes increasingly focused on outcomes, we've learned that we don't have to rely on the voice of authority, and have the right (and obligation) to demand data. And while professional societies can use almost any inclusion criteria they choose to select members, the rest of us can be more rigorous.
Specifically, health care worker certification - much like drugs and devices - should only be valued by hospitals and payers if the certification process itself is meaningful, if there is compelling and transparent evidence that the certification test is demonstrably reliable, and the certification correlates in some way with performance.
Given both the difficulty and the national importance of understanding how health care worker certification tests are developed and the extent to which they are validated, Congress may have a role to play here, a unique opportunity here to investigate an important health care challenge in a bipartisan fashion.
After all, we know existing certification processes are extremely costly - for the health care workers who must pay to be tested and for the government and private insurers who pay extra based on the results.
Perhaps it's time to see whether we're really getting quality in return.