$2 trillion. Forty-four million uninsured. Sixteen percent of gross domestic product. Respectively, that's how much the U.S. spends on health care each year, how many people are uninsured and the portion of the GDP that health care consumes. To those of us who follow health care, these numbers are more than just familiar -- they are macroeconomic symbols of our woeful health care system. As health care reform continues to burn near the top of political issues in the U.S. presidential race, these numbers increasingly are being used as weapons. Weapons against insurers, pharmaceutical companies and providers.
These numbers can be somewhat misleading. For example, just a few weeks ago, the New York Times published a controversial article -- "Sending Back the Doctor's Bill" by Alex Berenson -- about how many health care economists believe that both political parties were missing the real source of cost savings: physician incomes. The article argued that physician incomes were more than double those of their European counterparts and that health spending could be reduced significantly if doctors were paid less and were salaried rather than being paid per procedure.
Not surprisingly, the article drew plenty of criticism, including some from noted economist Uwe Reinhardt. Reinhardt counterargued that physician take-home incomes represent only 10% of health care spending and that even a 20% reduction in incomes would reduce national health care expenditures only by 2%.
Regardless of how you think physicians should be compensated, health care reform inevitably will be determined by the politics and economics of change. Health care is an enormous industry with large, entrenched, multibillion dollar constituents. Numbers increasingly will be wielded as fact, argument and weapons of ideology. But is it possible that, by focusing on the numbers, we're missing something equally, if not more, important? When it comes to health care reform, is there more than just the numbers?
Double-Edged Specificity
There's no question that numbers help us better grasp the nature and magnitude of issues. It makes a big difference to know that in the U.S. we spend $1 out of every $6 on health care, rather than just "a lot of money." Or to know that Wellpoint earned about $3.1 billion in net income in 2006 -- a 25% increase from the previous year. Numbers provide us with a reliable metric against which we can compare dissimilar ideas, issues and concepts. They are perceived to be less subject to misinterpretation and nuance. For that reason, numbers make great weapons for attack.
However, for the same reasons, numbers can be used to mislead and distort reality, whether intentional or not. They can overrepresent ideas or concepts that may be taken out of context. Most recently, oil companies received a significant amount of criticism for unprecedented profit levels. Exxon Mobile, for example, reported a profit of $39.5 billion in 2006. But, when taken in context, one can see that on a relative basis that number may not be as large as it seems. As a percent of total sales, oil and gas companies showed about the same level of total profit as a percent of sales compared with all other industries. Another number, profit margin, shows that oil companies may actually be half as profitable as the 21% benchmark, which, of course, happens to have been set by the pharmaceutical industry.
In addition to being taken out of context, numbers also can represent facts that go beyond any normal range of awareness or understanding. For example, if you were told that the sun is about 150 million kilometers from earth, would that provide you with any more clarity than what you already know: that the sun is very far away? To most people, 150 million kilometers is a meaningless metric. Unless put in another form (say, for instance, the sun is about 18,000 roundtrip flights between San Francisco and New York City), very little is gained by the added specificity. Numbers can be powerful instruments, but they do have their limitations.
The Importance of Context
More importantly, numbers can be taken out of context. Sources can be misinterpreted. Denominators can be distortionary. And the real issues determining the outcomes can be lost.
Today, anywhere between 10% and 25% of physicians in the U.S. are using an electronic health record. This statistic has been used to variably attack the electronic literacy of physicians, the design expertise of EHR vendors and the administrative sophistication of health care organizations. It has provided politicians with noncontroversial policy platforms that mandate and/or fund greater adoption of EHRs, but, as with many key statistics, the danger of numbers is that they may overly focus efforts on the wrong issue.
Should we be focused on the adoption of technology per se? Are we convinced that EHR adoption will lead to greater outcomes at lower cost? Is the low adoption rate simply a reflection of something else more contextual?
The danger we have in focusing too much on the numbers, particularly on the few surrogate metrics we have in health care, is that the underlying economics and incentives are highly distorted. Unlike many free-market industries, health care is trapped by volume-based, highly opaque third-party reimbursement. It creates outcomes and incentives that most would find nonintuitive. By attempting to control metrics that are merely surrogates of the system, we actually may be causing more harm than good.
Traditional analogies from other industries simply fall apart when we talk about health care, and we end up directing energy on the wrong goals. Something as benign as "ensuring adoption of EHRs by 2010" ignores the context in which those numbers exist. Perhaps, the low adoption rates are not reflective of the will to adopt EHRs but, rather, the lack of time and money for providers to invest in new systems? Perhaps the real source of focus should be on financing how providers get paid, rather than on how many are using EHRs? By disproportionately encouraging doctors to spend more money adopting systems that decrease their incomes, we might be increasing their pressures to see more patients per day -- which we know is not good for quality, cost-effective care.
The upcoming presidential election will provide us with many opportunities to take a look at what's wrong with our health care system and begin discussing ways to fix it. During the next several months, the candidates will be sure to demonstrate their expertise, citing statistics and setting new metrics for a change in health care policy. But it might be best for us to be wary of dictums that focus on improving surrogate statistics and metrics that don't look at fixing the critical root causes of a system gone wrong -- such as health care reimbursement reform. Individuals who advocate changing the system through forced adjustments of intermediate metrics, whether they be EHR adoption rates, physician incomes or pharmaceutical profits, might be making the problem worse, not better.