• Medical necessity and the law

    One more thing about that paper by Baicker and Chandra I blogged about this morning, this bit caught my eye:

    U.S. corporate laws also make it difficult for individual insurers and hospitals to reduce the use of technologies with variable payments: insurers and hospitals are not permitted to interfere with the medical judgment of physicians.  State laws also require insurers to pay for any service deemed medically necessary by a physician.

    Since there was no citation at the end of this passage, I didn’t know where to look for the support. By email, Chandra pointed me to a 1997 paper by Einer Elhauge (ungated PDF).

    As a matter of state insurance law, courts have held that an insurance con-tract cannot define “medically necessary” to exclude beneficial care normally provided by medical providers. Courts have also refused to enforce contract provisions giving insurers the power to decide what procedures are “medically necessary.” […]

    Courts read any contractual ambiguity against the insurer. This has always been true as a matter of state insurance law. […]

    The real battleground has been over denials of coverage for “experimental” treatments, and even there insurers often lose even if a treatment has no proven health benefit at all. […]

    When an insurer responds by making the exclusion more specific and detailed, courts have still denied enforcement on the grounds that the detail makes the exclusion too complex and technical for insureds to understand. […]

    The hospital’s legal duty is to provide the services the physician orders or the support the physician needs to perform the services herself. No matter what cost pressures they might be under, hospitals have little legal power to interfere with the judgment of treating physicians. […]

    As a result studies find, not just legally but in actual practice, that “hospitals must cater to physicians’ desire for new technology.”

    Nor is there much difference when physicians are hired as employees. [Citations omitted.]

    This echos the 1993 paper by Ferguson, Dubinsky, and Kirsch. But both papers are relatively old. Is it still the case that US and state laws make it impossible for insurers and hospitals to control the use of health technologies? If so, this is a substantial problem for translation of comparative effectiveness research (CER) into practice. It seems that physicians, as well as hospitals and insurers, would need some legal safe harbor to put CER to use. Who is thinking or talking about this? Anyone?


    • How much this matters certainly depends on the context, but I wouldn’t overestimate its importance as a barrier. Go back to the implementation of the hospital DRG payment system (or any other PPS): the hospitals didn’t pay or employ the physicians, only physicians can admit or discharge, but length of stay dropped quickly and substantially.

    • Great law. 🙂

    • Peter Orszag?:

      “If you want me to restrict the care I give people in accordance with a particular set of guidelines, you have to constrain the liability I incur by doing so.”

      That was a familiar refrain during the debate preceding the passage of the ACA. AFAIK the final draft of the ACA included no such provisions. Holding doctors harmless if they follow EBM guidelines might be politically palatable, but extending this immunity to third parties such as insurance companies for denying care based on EBM guidelines would likely make the backlash against HMO’s in the 90’s seem mild by comparison. Should be interesting when and if EBM guidelines are linked to payment in a way that has real teeth.

    • Wow it is no wonder health insurance is so expensive. This might help to explain why there is such a huge difference in premium between high and low deductible insurance. With a high enough deductible the insurance company does not need to be able to refuse to pay for low benefit care because insured would . In my experiences Doctors will help in this if they know that you are paying. A doctor once told me he would let my wife out of the hospital a week earlier had he know then that I was paying and even then he was letter her out earlier than he would had she had insurance.

    • To put a point on Richard’s example above, while at the margins insurers may not be able to restrict a treatment they can make it so expensive and burdensome to obtain that it’s not used very much. Think of PET coverage for certain indications or the use of non-formulary drugs. While in theory people can get exceptions, most don’t.

    • Aside from outpatient (read: elective surgery) the US is not a big outlier in utilization. We are an outlier in costs. Incentives matter, and that is where insurers have more control- in theory. Don’t pay so richly for the experimental or poor value stuff and it won’t happen as much. Don’t pay so much for health care overall and we don’t even have a health care crisis to worry about.

      Insurers mostly follow Medicare’s lead, and Medicare’s fee weighting is set by RUC, and RUC is controlled by specialist physicians who know the value of scratching each others’ backs and the controls don’t exist to weight payments differently. That right there is your bigger problem for guiding the use of technology.

      And then there is the matter of why private insurers generally haven’t been able to negotiate fees at the level of Medicare or Medicaid: market power, politics, culture.