Want to buy a quality measure?January 29th, 2014 by
“Psst … Hey buddy! Want to fix a measure in the Quality Game finals? The coach for NQF team #22 will roll for $11 million; you’ll make a killing!”
Fraudulent billing schemes, kickbacks for referrals, fixing ball games, bribing politicians, insider trading, performance-enhancing drugs – we are all used to these all-too-familiar human failings. Competition or greed leads many astray, but scandal surrounding a Patient Safety Quality Measure was a new twist.
I almost overlooked the Modern Healthcare news bulletin about a DOJ settlement with CareFusion Corporation and the corresponding National Quality Forum’s statement. It took a couple of minutes for the implications to sink in.
According to these sources, Dr. Charles Denham, an internationally respected leader in healthcare quality, accepted $11.6 million to participate in something that sounds a lot like a “product placement scheme.” According to the settlement, while Dr. Denham was chairing a NQF Committee that wrote Safe Practice Quality Measures, his foundation awarded contract grants to the NQF to support the committee’s work. Allegedly, Carefusion Corp. made payments to Dr. Denham’s foundation that were not “fair market value,” and were apparently intended to subvert the integrity of the NQF standard development process. The result: a specific Carefusion product, CloraPrep, was written into the draft quality standard. To its credit, NQF identified the conflict through its standards vetting process, and the specific product recommendation was eliminated.
Why is this important?
This proves pay-for-performance works (perhaps too well);
Measures (e.g. definitions in performance standards) can drive payments;
Provider behavior will follow measure specifications;
Special interests have recognized the critical role played by a measure’s performance specs; and
This case demonstrates measure development may not be an entirely objective process.
Some of us who are involved with quality measures/measurement will smugly blow off this incident as yet another example of commercialism’s evils. I see it differently; to me, it is just the most vivid, and understandable, example of what’s wrong with our current quality strategy.
Our rush to pay-for-performance in two critical aspects of American society — education and healthcare — can lead to perverse results. In some processes, quality goals are easy to agree on and subject to unambiguous improvement. It’s easy to applaud reducing infection rates in the ICU or surgical suite; they are pretty straightforward. Those locations are in closed systems (e.g. a specific part of a hospital) with a precisely defined population.
There are plenty of open system (the community) quality goals that make sense – increasing the childhood immunization rates, reducing premature births, increasing school attendance rates, etc.
All of these effective QI initiatives have some things in common – a clearly defined population, a relatively easy to quantify performance goal with wide professional endorsement, and existing reporting tools.
Some providers are at risk of serving populations with adverse risk profiles. A good measure incorporates risk rate adjustments for specific populations that are seen as reasonable to those being measured. In healthcare you’d consider age, problem/acuity, exposure time, etc. Hospital quality measures largely incorporate these features. Some providers are still adversely affected, but there is near universal attention paid to the quality measures CMS mandates.
The physician world has far less respect for CMS’s quality schemes (e.g. PQRS). Only 25% of all physicians attempted to participate in 2012. Why?
I’m arguing that the initial success of specific quality strategies led to an erroneous assumption that the process could be generalized.
The architects of our new payment reform schemes cite the success of these QI programs as proof that pay-for-performance is a valid idea. Today, most major healthcare payors are employing pay-for-performance schemes. All those programs employ quality measures.
How do those measures come into existence? The scandal with NQF measure #22 is educational. CMS and payors want a set of measures that span the payment spectrum. Private organizations like NQF are chartered to act as stewards of the measures. NQF and other groups establish work groups to develop measures.
People or organizations with specific points of view “self-select” to work on measure development. Only someone with a deep interest will invest the requisite time to go through this process. NQF #22 shows how a commercial entity may have tried to hijack that process. But, what if measures were written to enshrine a particular process, consultation protocol, etc.? Many measures are quite specific in the definitions of “best practices” that constitute “performance met.”
If you work with the chronically ill older adult population, it is apparent that almost no existing PQRS measures are applicable. A recent Health Affairs blog supports this opinion.
It’s apparent to me that CMS needs to call a timeout and reconsider its quality strategy. We need better measure development strategies, not more measures and steeper penalties. Physicians and other outpatient providers work in open systems with heterogeneous patient populations. We need to begin with defining populations, and only then can we develop population specific quality measures. This doesn’t require throwing out all of our existing measures, but defining “performance met” in terms of specific populations.
Also, if the populations are well defined, the provider community can identify appropriate measures based on their patients. That makes it much easier to see where a special interest may have put their finger on the scale during the measure development process.
I fear CMS will ignore this message, and make the measure development process even more complex in reaction to the Measure #22 scandal.