Guest author: Yves Smith, author of the best-selling book “Econned – How unenlightened self interest undermined democracy and corrupted capitalism”. Yves writes at the popular blog Naked Capitalism.
When analysts and commentators go through their laundry list of what ails the hopelessly overpriced US health care system, one of the most commonly cited reasons is the cost of “defensive medicine” meaning both the cost of medical malpractice policies to doctors, plus the costs of extra tests and procedures to provide doctors with cover in case a malpractice claim arises.
A new study suggests that these issues are greatly overstated as a culprit in America’s burgeoning medical costs. Despite the considerable (and very visible to doctors) cost of malpractice insurance, it appears that the savings resulting from the threat of litigation is underestimated. For instance, some of those “defensive” tests actually turn up real problems; the threat of lawsuits actually keeps doctors and hospitals on their toes. Put more simply, lawsuits appear to be a check on a system that otherwise runs the risk of having inadequate protections of patients.
This study pegs the cost of defensive medicine at 2.4% of total health care expenditures. By contrast, administrative costs for US healthcare are much higher in the US than in other advanced economies. From the New York Times:
One thing Americans do buy with this extra spending is an administrative overhead load that is huge by international standards. The McKinsey Global Institute estimated that excess spending on “health administration and insurance” accounted for as much as 21 percent of the estimated total excess spending ($477 billion in 2003). Brought forward, that 21 percent of excess spending on administration would amount to about $120 billion in 2006 and about $150 billion in 2008. It would have been more than enough to finance universal health insurance this year.
The McKinsey team estimated that about 85 percent of this excess administrative overhead can be attributed to the highly complex private health insurance system in the United States. Product design, underwriting and marketing account for about two-thirds of that total.
That puts the additional costs at a considerably higher level than what this study found for medical malpractice. Some analyses contend the administrative cost differentials are even greater).
This write up comes via Joe Paduda (hat tip reader Francois T):
A new study [abstract only] reported in this morning’s Health Affairs makes a compelling case for the latter view, and adds valuable insight into what is a politically-charged issue, one rife with misinformation and sloppy math.
The study found “Overall annual medical liability system costs, including defensive medicine, are estimated to be $55.6 billion in 2008 dollars, or 2.4 percent of total health care spending.” [emphasis added]
Recall total system costs are in excess of $2.2 trillion. While $55 billion is a lot of money, compared to total system costs of $2.3 trillion, it, well, isn’t much.
In fact, costs would be much higher if the real toll of medical malpractice – lousy care, incompetent providers, poorly managed facilities, was adequately accounted for. Solid research indicates the vast majority of medical malpractice problems are never litigated. One study indicated that the cost of ‘adverse events approached 5% of total health care costs; over a hundred billion dollars in today’s world.
The med mal reform issue has been raised by opponents of health reform, who contend the failure to include med mal reform in the Accountable Care Act was a missed opportunity to significantly reduce costs Of note, the study estimated the most significant cost associated with medical malpractice was defensive medicine, which accounted for $45.6 billion of the total, most of which was spent on hospital services.
In an email conversation, I asked the study’s principal author, Michelle M. Mello, PhD, to clarify the study’s findings re the impact of med mal on defensive medicine – the theory that physicians change the way they practice to protect themselves against medical malpractice by prescribing more tests and studies.
Here’s Dr Mello’s response.
There are two ways to measure defensive medicine. One is to ask physicians, using surveys, how often they order extra tests, procedures, and referrals primarily because of liability pressure. We didn’t use this method because it has two major shortcomings: (1) physicians may consciously or unconsciously overreport defensive practices because they want to help build the case for taking action to solve what they perceive as a problem with the liability environment; and (2) they may not be able to separate out different motivations they have for ordering services. In many cases, they may feel that ordering an extra test is a good idea both because it’s in the patient’s best interest and because it helps them reduce their liability risk.
The other method — the one we used — is to compare rates of health services that we think are indicative of defensive medicine in areas of high and low liability risk. If rates are higher in high-liability areas, and we can rule out other explanations for the differences, we can conclude that there is an association between liability and physician practices. The main challenge associated with this method is adequately controlling for other factors that could explain the differences. Researchers have extensively documented that physicians in different geographic areas have different practice styles, and it is believed that this is due to many factors, of which liability concern may be one.
We based our defensive medicine estimates for hospital services on previous analyses by Dan Kessler & Mark McClellan. Having reviewed the literature extensively as it has evolved over the past decade, our firm belief is that the Kessler & McClellan analyses provide the best available figures. Their statistical design enabled the researchers to control for other sources of variation in physician practices.
The main weakness of the Kessler & McClellan analysis, as we discuss in the paper, is that it was based on a narrow range of health services (cardiac care services) provided to a specific type of patient (Medicare beneficiaries). Is it appropriate to generalize from these data to all services provided to all patients? We have some concern about that, and consequently characterize the quality of the evidence supporting our defensive medicine estimate as low. Other kinds of health services may be less subject to physician discretion over treatment intensity than the cardiac services that Kessler & McClellan studied, so it’s possible that extrapolating to all services yields an estimate of defensive medicine costs that is too high. Nevertheless, we believe Kessler & McClellan’s analysis of the strongest one available.
The paper provides additional background on the methodology used, and the challenges with that methodology. While it isn’t perfect, one has to compare it to the methods used by others who contend the tort system is a major driver of health care costs. Those ‘methods’ are rather less rigorous.