Starting With Curious

This may seem like a trick question: which is more valuable, resuscitating a newborn infant, or doing a hair transplant?

While I realize I have a pediatrician bias (though I am also follicularly challenged, so my biases probably balance), I think most people would place greater value on the newborn resuscitation. But if you ask the Center for Medicare and Medicaid Services and the American Medical Association, they would pick the hair transplant. In fact, hair transplant is assigned between 31% and 84% more value (depending on the number of plugs) than resuscitating a newborn. This is according to the Relative Value Unit (RVU) system, used in American medicine since the early 1990s. (Note that I am using the conventional method of calculating percent differences, not the alternative math used by the White House as described by RFK Jr.)

The word “value” has been thrown around gratuitously in healthcare for decades. To paraphrase the great philosopher Iñigo Montoya, people keep using that word, but I don’t think it means what they think it means. As I have written about before, traditionally value in healthcare has been defined as the ratio of quality to cost. People want good care – care that is effective and safe and patient-centered and timely and equitable – but they don’t want to be bankrupted by it. This makes sense, and it captures the idea of value as “the regard that something is held to deserve; the importance, worth, or usefulness of something” (Oxford English Dictionary).

Too often, though, in healthcare the denominator – cost – takes priority over quality, and indeed the term value becomes synonymous with money. There are many examples of this, but one of the most grating to physicians is the RVU. And the emphasis on RVUs can skew incentives away from true value for patients and society.

Before the 1990s, Medicare paid physicians based on “usual, customary, and reasonable” charges, which was highly variable and admittedly somewhat arbitrary. The Resource Based Relative Value Scale (RBRVS) payment system was created by legislation in 1989, and began implementation in 1992. (Since then, it has been widely adopted by private payers as well.) Under this system, physician payments would be based on a system of RVUs. The primary component, the work RVU, or wRVU, would be a measure of the work performed by a provider in delivering the service and relates to time, effort, and expertise required to perform the service. Each of thousands of procedure codes is assigned a wRVU weight. These weights are then multiplied by a factor to yield a payment amount. (There is also a component for the practice expense like supplies and office overhead, and malpractice.) In theory, this standardizes payment so that physicians doing similar work get similar payment, and greater effort is rewarded commensurately.

How are these wRVU weights determined? Initially they were assigned by a survey of physicians who were asked to rate the work of various procedures according to “time, mental effort and judgment, technical skill and physical effort, and psychological stress.” The original study, conducted by Harvard researchers and published in NEJM, concluded that “these ratings are highly reproducible, consistent, and therefore probably valid.” (Emphasis mine.) Subsequently, the American Medical Association created the Relative Value System Update Committee (RUC) which advises CMS on updates to the wRVU weights. The RUC consists of approximately 30 physicians, mostly specialists, the majority appointed by the relevant specialty society.

While the RBRVS has arguably brought standardization and rationalization to physician payment, there are important criticisms. One is that it tends to overvalue procedural over cognitive effort. A newborn circumcision, for example, has a wRVU of 2.47, compared with an initial office visit for a newborn with 1.5 wRVU. In my experience, each takes approximately the same amount of time (and that is only if the initial office visit is for a newborn with an uncomplicated history and the parents/caregiver have reasonably adequate resources). This gap has probably widened as technology has made many procedures faster and easier. At the same time, technology has increased the work required for many non-procedural tasks. Care coordination can require significant time reviewing the far greater volume of information contained in the electronic health record. And health counseling is complicated by the explosion of information on the Internet and from AI, of highly variable quality and validity, that primary care physicians must address.

And while the term “value” features prominently (note how many V’s are in these acronyms), do wRVUs really measure value? Do they measure what is important to patients and their health outcomes? Is performing a hair transplant or an eyelid lift (wRVU 4.5-6.7) really more than twice as valuable as an hour of chronic care coordination (wRVU 1.81)?

This is not only a moral or theoretical concern. Physician compensation relates directly to wRVUs. This is part of why specialties like plastic surgery (avg. annual salary $544,000) and gastroenterology ($520,000) pay so much more than public health ($269,000) and pediatrics (coming in at the bottom at $265,000). Is it any wonder why we are predicting a shortage of primary care and preventive medicine physicians?

If we really want to deliver value, we need to rethink how we define and measure it.

In an era where people marry AI chatbots, online markets allow people to place bets on when the US will bomb another country, and wealthy individuals and corporations openly purchase political power,  it is perhaps time to reconsider the old adage “money can’t buy everything.” As I wrote in my previous blog, Michael Sandel describes in his book What Money Can’t Buy: The Moral Limits of Markets our transition from a market economy to a market society.  While free market capitalism has demonstrated its ability to efficiently organize productive activity and allocate goods and services, this efficiency comes at a cost. One concern, discussed previously, is how markets can produce inequity. Inequitable distribution may be tolerable for some things (video games, Broadway shows) but not for others (insulin, cancer screening.) A more subtle, but perhaps more concerning, concern about marketization is its corrupting effect. Even if we can make a market for healthcare less unfair, it may still be undesirable.

                Market theory assumes that the value of a thing is fully reflected in people’s willingness to pay for it. But it is easy to think of examples of things that have an intrinsic value that cannot be readily expressed in monetary terms. I have a hat that belonged to my grandfather. I am sure that I could find a similar hat in a vintage store, but it wouldn’t be the same hat. The hat that I have can never be replaced. It is, effectively, priceless. Indeed, there are things that are literally priceless. Sandel uses the example of the Nobel Prize. It has value exactly because it cannot be purchased; it can only be earned. His point is that if we try to commoditize something, we can reduce its intrinsic value. Once there is a market for Nobel Prizes, they cease to have any real value. (Keep in mind, he was writing in 2012, long before a US President would openly campaign to buy one. That craziness notwithstanding, Sandel’s point remains valid.)

                Even if commercialization doesn’t devalue something, it can change the nature of it. Blood donation in the US is an act of altruism. If we begin to pay for it, then it becomes a commercial transaction. The motivation for doing it shifts from an intrinsic one (desire to help others) to an extrinsic one (monetary reward). And studies have shown that extrinsic motivations can crowd out intrinsic ones. If the expectation is that you get paid to donate blood, willingness to do so without pay decreases. At the same time, extrinsic motivations are generally weaker and more fickle. During economic boom times, the blood supply becomes vulnerable as fewer people need the money they can get from giving blood.

                If you want to get under a clinician’s skin, ask them about RVUs (relative value units). This is a system of quantifying the “value” of a physician or other provider’s work. They are used to measure clinician productivity, and compensation is often tied to worked RVUs. The intrinsic motivation of taking care of people is supplanted by the extrinsic motivation of generating RVUs. Studies have linked RVU-based compensation with decreased physician satisfaction and burnout.

                Monetary incentives may be effective in delivering a result, but not necessarily the desired one. Paying a child to read will likely lead them to read more books, but it does not necessarily lead to an increased love of reading. It may even lead to gaming the system by the child reading books below their reading level – quantity over quality. Paying a clinician to generate more RVUs will lead to more patients being seen, but not necessarily to better or more compassionate care.

                Indeed, by supplanting intrinsic motivation, monetary incentives can backfire. A study in Israel found that after instituting a payment for people who picked up their children late from a day care, late pickups actually increased. Rather than being motivated by shame for inconveniencing the staff, many parents simply felt that coming late was a preference they could pay for if they wanted.

                One subtle form of corruption produced by marketizing things is it can send an unintended moral signal. According to market theory, motivations are preferences, these preferences are equally valid, and they are rationally expressed in the form of a willingness to pay for them. It does not draw a moral distinction between, say, choosing to buy a TV versus a gun, or one gun versus 100. Making healthcare a market commodity, it simply becomes one among many equally morally valid options. It signals that if someone chooses to spend their money gambling in Vegas instead of purchasing health insurance, that is morally OK. Libertarians would undoubtedly agree with that; who are we to judge someone for choosing craps over coverage? Even if this did not unfairly create a societal burden (the gambler can rest easy knowing that they can’t be turned away from an ER even if they can’t afford to pay), and they really would suffer the consequences of their decision, I know for me – and I think for many of us – the idea that healthcare is no different from anything else you can obtain for money doesn’t sit quite right.

                There will, of course, always be economic considerations. After all, nurses and doctors, manufacturers of drugs and medical devices, discoverers of new cures, and others all expect and deserve to be appropriately paid for their contributions. And as with education, to an extent the healthcare system needs to accommodate differences in individual needs and preferences, often most efficiently expressed in market terms. But healthcare, like education, is not a commodity. These are social goods. When people are healthy and well-educated, not only do they benefit themselves, but all of society benefits. As more and more of the healthcare system becomes subject entirely to the forces of the market, we risk unacceptable inequities in access, and corruption of the underlying values that underpin it.  I don’t think it’s being overly sentimental to hope that the noble intrinsic motivations of the millions of dedicated people working in that system are not supplanted and degraded by more materialistic external ones. It’s called healthcare for a reason.

                I recently spent 40+ hours over 4 months getting my wife enrolled in Medicare. The craziness of how challenging this was for someone with, arguably, a lot more specialized knowledge of the healthcare system than the average person (and even I needed to invoke the help of my Congressperson’s staff to get across the finish line) could itself be the topic for a column. There are many reasons for this situation, but one I want to explore is the role of the market. Although Medicare is ostensibly a government program, much of it is privatized. Consumerism has been proposed as the answer to the American healthcare woes of high costs with lagging results in terms of actual health outcomes. Give people more choices, with transparency about costs and quality, and the invisible hand of the market will make it all better. I have long been suspicious that this can actually work with healthcare (it certainly hasn’t so far). But now I find myself wondering whether even if it could deliver better value, is it the right thing to do?

                To someone like me who went into healthcare out of a sense of mission and calling, consumerism has always felt a little icky, though I couldn’t well articulate why. I recently read Michael Sandel’s 2012 book What Money Can’t Buy: The Moral Limits of Markets. In it Sandel, a Harvard political philosopher who has also written on justice and “meritocracy,” talks about the enormous expansion in recent decades, particularly in the US, of the role of markets in society. He describes the shift from a market economy, which is a tool for efficiently organizing productive activity, to a market society where market values dictate nearly every aspect of human activity, not just economic transactions but social relations. As the title of the book suggests, he believes that some things should not be left to the market, not for practical reasons but for moral ones. He raises two overall causes for concern: equity and corruption. While the book is not primarily about healthcare, I couldn’t help but conclude that it is one of the things where more marketization should raise alarm bells. I’ll tackle the latter concern, corruption, in the next installment of this blog. For now, let’s consider how marketizing healthcare threatens equity.

                The basic principle behind market transactions is that people, acting in their self-interest, pursue those goods they most want. Prices are determined by the balance of supply and demand, and goods are then allocated to the people who most value them based on their willingness to pay. If tickets to the Vikings and a dinner at the Four Seasons each cost $300, then the person who prefers football to fine dining will get the tickets because they are willing to pay for them, leaving the dinner reservation to the gourmand who really wants them. Each one, pursuing their own interest, gets what they want. Everyone is happy.

                This assumes that those individuals have the means to purchase one of these goods. I might really want those Vikings tickets, and be willing to pay $300 for them, but if I don’t have that $300 I can’t have them. Markets are based on allocation according to willingness to pay. But willingness to pay is not the same as ability to pay. Money is unevenly distributed in society – increasingly so in the US – so the ability of the market to allocate goods to those who most value them is restricted. (Indeed, when someone is constrained enough, choices may be not only limited, but coercive. I may not want to sell my plasma, but if I’m desperate enough I will.)

                With greatly uneven distribution of economic resources – of ability to pay – some people will be unable to obtain something they value a great deal. “I’d give anything for…” is a common sentiment, but very few people are in a position to literally give anything.  At the same time others with excess resources may obtain things they don’t have a great desire (or need) for simply because they can. This may not matter for stereo equipment or luxury cars, but can lead to troubling inequities if we are talking about the distribution of medicines and healthcare services.

                Fair markets also assume equal access to information about the transaction. In reality, not only is money distributed unevenly, so is information. As my Medicare enrollment experience illustrated, modern healthcare is incredibly complex. And transparency is at best a tiny step toward equalizing information. There are so many variables, with so many degrees of uncertainty, from so many different sources, it is unrealistic to think that the ability to process it can be equal. (Even if there weren’t active disinformation going on, including from our own government.) If people have different levels of knowledge about the benefits and drawbacks of a good, different abilities to know what they are getting for their money, then the transaction cannot be truly equitable.

                If a healthcare market is prone to inequity by allocating the goods and services based on willingness (or more accurately ability) to pay, is there a more equitable alternative? In many systems, willingness to pay is replaced by willingness to wait, or queuing. This is a criticism often raised by market advocates about single-payer, non-market-based systems, that they lead to waits for care. This is often true. But it can be argued that time is distributed more evenly in society than money is, though not completely so, and therefore willingness to wait is a fairer system. And to be clear, we also have waits, though in our system, it’s a bimodal distribution: some people don’t wait very long, while others wait literally forever.

                And there are ways to make a market-based healthcare system more equitable. Health insurance does that by spreading risk across the population, and income-based subsidies for insurance can reduce disparities in access. Standardizing benefits and information, like the ACA marketplace or the letter system for supplemental Medicare (“Medigap”) plans, can reduce information disparities, though trust me, it remains complex.

                But inequity isn’t the only concern about markets in healthcare. Markets are about allocating value, but are agnostic about values. It is the corrupting effect of markets that is even more troubling. We’ll look at that next time.