A few days ago, the medical community received unexpected good news from the government about a “simplification of quality measures:”
Impactful news today: Physicians, CMS and major commercial plans, announce simplification of quality measures. “Huge step forward” #aafp
— Andy Slavitt (@ASlavitt) February 16, 2016
Strictly speaking, and contrary to what Mr. Slavitt’s tweet would lead us to believe, the agreement to the new rules was primarily between commercial insurers and CMS, the Center for Medicare and Medicaid Services. Physicians were not actually party to the deal.
Nevertheless, doctors were expected to greet the news with cheers. As Rich Duszak reported, Adam Slavitt, acting administrator for CMS, also declared that “patients and care providers deserve a uniform approach to measure [sic] quality.”
— Rich Duszak, MD (@RichDuszak) February 18, 2016
Indeed, we all deserve uniform quality measures. Equality in quality!
Taking the pulse of “industry leaders,” Becker’s Hospital Review affirmed they were confident the agreement would help “accelerate the country’s movement to better quality.” And who could possibly be against better quality? Only anti-progress reactionaries, to be sure!
To understand this latest development, the context may be helpful.
Fifteen years ago, the Institute of Medicine, suddenly aware of the calamitous state of American healthcare, issued a call to arms urging everyone to “Cross the Quality Chasm.” And since, in those days, management gurus were repeatedly telling us that “if you can’t measure it, you can’t manage it,” the missionaries of quality dispatched in the wake of the IOM report naturally went on a quality measurement frenzy. How else could they be expected to tame the beast?
But, as Dr. Vinod Seth humbly remarked, if you measure something, is it still a quality?
— Dr Vinod Seth (@askdoctorvinod) February 17, 2016
“Quantifying quality” does sound a tad incoherent, doesn’t it?
So what is Quality? To answer the question, Robert Pirsig felt compelled to combine Zen buddhism, pragmatism, and Peyote. But can we put our finger on the concept without being driven to insanity?
I like to stick to the old school realism of Aristotle, and I think ancient thinking may come handy here.
In the classical sense, a quality of a thing is a characteristic that distinguishes it from another thing: a sound is hollow, as opposed to muffled or thundering. A person is pleasant, as opposed to rough or taciturn. Qualities separate and segregate things into distinct categories according to the subjective experiences one may have of those things. Quantification plays no role in determining qualities.
Now, it is true that a sound can be more or less muffled, but once we focus on how muffled the sound is, we are no longer interested in muffledness as a quality. A slightly muffled sound or a highly muffled one all belong to the family of muffled sounds. They all have the same quality.
The same consideration goes for discriminating between something good and something bad. Goodness and badness are important qualities that express the ultimate reason for approving or disapproving of a thing: We approve of this product because it is of good quality; we disapprove of that service because it is of bad quality.
Note that the stipulation is coincident with the choice: It is not as if we first determine whether something is good or not and then apply our seal of approval. We approve of something because we have experienced it as good (or vice versa: if we experience something as good, we approve of it). That being the case, there is no room for “measuring” good quality. It can only be experienced.
Of course, an immediate objection to what I have just said is that we are surrounded in our everyday lives with quality scales for consumer goods and services. No one seems to complain against Michelin stars or Yelp ratings, so why couldn’t we have the same thing in healthcare?
In my judgment, the use of quality measures in medical care is problematic for several reasons.
First, consumer guides are simply testimonials reflecting other people’s experiences and opinions. They are widely recognized as being subjective. Furthermore, they are produced for the sake of consumers. In healthcare, on the other hand, quality ratings are primarily for the sake of insurers, and particularly for the Medicare program. The interest of these entities may not be the same as the interests of patients.
Second, because healthcare is generally paid for by insurance, patients lose an important independent indicator of quality: market prices. A freely set market price is one of the most reliable signs of quality, because such a price is the end result of voluntary exchanges between large numbers of consumers and producers. In a system dominated by third-party payment, however, prices do not reflect the needs and satisfactions of patients. Instead, they mainly convey the outcomes of negotiations between providers and payers.
Third, healthcare is highly regulated. As a result, quality measures are essentially arcane rules articulated in bureaucratic lingo. Unlike rating systems for consumer goods, where one can directly verify the veracity of the report, healthcare quality measures invite ambiguities, errors, and obfuscations.
Finally, medical care consists of choices and decisions made in the face of inherent uncertainty. Unlike a hamburger or a hotel room, every human being is unique, with constituent ingredients that are beyond the complete control of healthcare workers. Therefore, the quality of a coronary bypass or that of a stay in the intensive care unit must be gauged in real time, in its specific context, and in light of all possible alternatives. Government regulators, third-party payers, and auditing agencies are far removed from these contextual details. Their quality determinations are unlikely to be more accurate than those spread by word-of-mouth from patients who have experienced the care.
In conclusion, when it comes to promoting quality in healthcare, the words of sociologist William Bruce Cameron come to mind: “not everything that can be counted counts, and not everything that counts can be counted.”
Note: This post was significantly edited for clarity on July 27, 2016