• *Catastrophic Care*: Chapter 4

    Chapter 4 of Goldhill’s Catastrophic Care is a good read. From the incentives of insurance to their effects on the pharmaceutical industry to the overuse of screenings (and more), I found it all on target. Here are some of my highlights (all direct quotes):

    1. My fellow insured Americans would not have been willing to subsidize more fish and vegetables in my grocery cart, but insurance requires them to share the cost of my pharmaceutical alternative.
    2. The drug business may be the most straightforward example of how our refusal to use price to regulate supply and demand creates excess in both and discourages innovation.
    3. Since a drug company doesn’t face the risk of a price war, it overinvests in safer products with established markets and underinvests in riskier ones that may provide greater health benefit.
    4. Just because a test is sensible for some people doesn’t mean it should be performed on everyone.

    For more relevant to item 2 and 3, read Marcia Angell’s The Truth About the Drug Companies (related TIE post here and more in the FAQ). For more relevant to item 4, see this. Full and proper use of evidence in guiding health care decisions made by patients or surrogates — and the incentives to motivate just that — would reduce over-testing, avoiding its costs and harms.


    Comments closed
    • On the fourth point–according to some physicians I talk to, one of the main reasons for over-testing is that a lot of physicians take a “shotgun” approach rather than what I call a “bayesian-search” approach to prescribing tests. In the former, they order every test that might be relevant to the patient’s condition, all at once. In the latter, they only test for the most likely thing first, and go through the list of tests one by one in order of likelihood until they can make a confident diagnosis.

      Obviously, there are reasons to “shotgun” approaches some of the time, when urgency is required, but most of the time doctors can save money by following the Bayesian search pattern. It also turns out that that is the most pleasant to patients, who don’t like being poked and prodded anymore than they have to be.