• The job-killing medical device tax, ctd.

    When I posted earlier, I had forgotten about the recent paper by Bryan Schmutz and Rex Santerre on the device tax and R&D. I still have not read it in full. Nevertheless, here’s the abstract:

    Unlike the pharmaceutical industry, no empirical research has focused on the factors influencing research and development (R&D) spending in the medical device industry. To fill that gap, this study examines how R&D spending is influenced by prior year cash flow and corporate market value using multiple regression analysis and a panel data set of medical device companies over the period 1962–2008. The empirical findings suggest that the elasticities of R&D spending with respect to cash flow and corporate market value equal 0.58 and 0.31, respectively. Moreover, based upon these estimates, simulations show that the recently enacted excise tax on medical devices, taken alone, will reduce R&D spending by approximately $4 billion and thereby lead to a minimum loss of $20 billion worth of human life years over the first 10 years of its enactment.

    I took a quick enough look to find out how the authors estimated the loss of life years. The key bits are these:

    To translate that lost amount of medical device R&D spending in to the value of human lives lost, we use Neumann et al. (2000) median estimate of $40 000 cost per quality-adjusted life year (QALY) gained from medical devices to calculate a reasonable estimate of the amount of R&D spending needed to gain one QALY. In other words, prior research has determined that every $40 000 medical device sales gained one QALY. Using the 2007 median estimate for R&D intensity of approximately 0.06 (which is R&D divided by Sales) along with the 2007 adjusted median estimate of $51 444 cost per QALY (adjusting the $40 000 to year 2007 dollars), we calculate an estimate of $3087 of R&D spending needed to gain one QALY. […]

    [W]e use a range of $200 000 to $300 000 to calculate the dollar value of life years lost to depressed medical device R&D spending.

    By email, I asked Santerre if his work incorporates any effect on the industry of coverage expansion. His answer:

    No we didn’t but the demand expansion effect may be slight for medical devices. Hospitals are the primary buyers rather than consumers and they have always strived to have the newest bells and whistles when it comes to medical devices. Also, they typically feel obligated to treat most patients when they come through the door even without insurance (legally they just have to stabilize the patient as Aaron reminds us). I don’t see much changing in that regard after the insurance mandate. However, if medical technology is the culprit behind the growth of health care spending and many (and I) believe, the tax on medical devices may be one way to slow down its growth. In other words, to slow the growth of health care spending, maybe people should die sooner than they would have otherwise.

    I think we’d all prefer to slow the use of medical devices we know to be too broadly and harmfully applied. If we did so, instead of killing people, we might actually save a few.


    • A naive comment/question, but won’t the elasticity of demand for these devices be some small negative number close to 0? and if so, won’t the entire tax be passed through ultimately to consumers? if you need a new set of knees or a stent, are you going to say no thanks because your co-pay or some other amount you pay went up 2.3%? Sure, there will be some reduction in demand on the margin, because there is demand curve, but I am just not sure who will make that calculation?

    • This makes me laugh – as though the vast US healthcare market suddenly becomes unattractive because you have to pay a bit more to fund its expansion. And the other factor of course is that many devices will never be assessed in RCTs, so to make a hyperbolic projection about people dying sooner is absurd.

    • The unspoken assumption in the QALY calculations is that extra years of life will add to the nation’s wealth.

      This is a confusion. Extra years of life of course feel good to the person who is living. I want all the extra years I can get.

      But longetivity costs society at least as much as it gains.

      If a 90 year old with Alzheimers lives an extra year, then his family loses financially, Medicare and Social Security lose financially, etc etc.

      That is an extreme example, but only a difference in scale.

      When a child survives with a very serious illness, once again the family is often bankrupted.

      • Do GDP calculations “care” about personal bankruptcy? As a measure of economic activity the more you spend on your child’s treatment the more you increase GDP. Of course if you cut back on all spending afterwards because you are broke that lowers GDP in the second period. But I think the outcome of bankruptcy alone may not be measured in GDP.
        Of course as humans we care about bankruptcy and emotional distress and other forms of harm not captured by GDP.

    • The medical device tax, is partly about self-referral and over-utilization. It’s become common-place for medical device manufacturers/vendors to compensate physicians who use their products, either from consulting, speaking, and research fees, or joint ventures in which the physician is an equity owner in the chain of distribution. What’s likely to be a greater deterrent, however, are the new disclosure rules, which require the medical device vendors to disclose these arrangements with physicians. No, the disclosure won’t discourage these arrangement by putting the physicians at greater risk of government sanctions (such as being barred from being a Medicare/Medicaid provider), but by increasing the likelihood of qui tam (whistleblower) lawsuits against the physicians.