• Why doesn’t competition drive out inefficient health care technology?

    The following is a guest post by Nicholas Bagley, University of Michigan Assistant Professor of Law.

    So here’s a burning question. There’s a consensus that the primary driver of escalating health-care costs is the rapid adoption of new and expensive medical technology. Much of that technology is untested and of questionable medical value. Yet private insurance plans typically cover most any intervention that physicians say is medically necessary. There are exclusions here and there, to be sure. But it’s a decent rule of thumb that, if traditional Medicare covers a particular service or procedure, private plans will, too.

    Why? Why don’t private plans compete on price by refusing to cover costly, unproven therapies? (Recently, several have announced that they won’t cover proton-beam therapy for prostate cancer. But that’s the exception, not the norm.) One answer you sometimes hear is that the law gets in the way. Plans will point, for example, to a $77 million judgment entered against a California insurer that refused to cover an expensive (and, as it turned out, brutal and ineffective) breast-cancer therapy. See? We want to contain costs, but the courts won’t let us do it.

    The law can’t be the real story, however. As it stands, a federal statute—ERISA—gives employer-sponsored plans almost complete freedom to tailor their coverage packages as they like. If a plan excludes appendectomies from coverage, that’s legally unobjectionable. (Employers that don’t self-insure are subject to certain state coverage mandates. But ERISA preempts those mandates when it comes to self-funded plans, which today cover about three in five employees.) ERISA even shields a plan from liability if it negligently refuses to authorize coverage for care that it (wrongly) thinks is medically unnecessary. As safe legal harbors go, it doesn’t get any better than ERISA.

    Why, then, are private plans so cautious? I’m speculating a little here, but I’ve got some thoughts. For starters, it’s really hard to make good coverage decisions. The data for making them are usually quite poor, as Peter Neumann and his co-authors have demonstrated. And, absent convincing data, a plan that excludes a promising treatment risks alienating physicians and hospitals (not to mention patients). No individual plan has the right incentives to generate that kind of convincing data because, once it does, its competitors will jump on the bandwagon and take advantage of the leading plan’s research investments.

    What’s more, most coverage decisions aren’t crisp (“No proton-beam therapy, period.”). They’re qualified: if a patient has a certain risk-profile, or an identifiable need, then the intervention is covered. But once a plan has said that some patients are eligible for a particular treatment, it’s hard to stop those outside that group from getting the treatment, too. The data for denying the intervention in a borderline case won’t be great. Physicians may fudge the criteria. The plan will probably have to do some kind of utilization review, which is expensive and alienating. And a stingy plan can quickly scare off its customers, who won’t care when they’re sick and scared that their premiums are lower because their plan is aggressive about technology limitations.

    Against this backdrop, piggybacking on Medicare’s coverage determinations makes good sense. Not only does it allow plans to sidestep the collective-action problem that plagues efforts to develop good coverage data. It also helps plans avoid public backlash because they can be confident that their competitors will also follow Medicare’s lead. The government’s seal of approval lends legitimacy to a coverage exclusion that might otherwise appear hard-hearted.

    Employer-sponsored plans aren’t the only private plans around, of course. And, as it happens, the law has more bite outside of the employer setting. Medicare Advantage plans, for example, are legally obligated to cover everything that traditional Medicare covers. Similarly, starting January 1, 2014, the ACA will require private insurers outside the group market to cover essential health benefits.

    But for employer-sponsored plans, the law isn’t the problem. Far from restraining these plans, the law enables them to tackle the rising costs of technology. There’s just not a business case for it—at least not yet.

    • I do not have time to find it right now, but I think there are some detailed histories of what happened when some insurers tried to refuse payment for radical and expensive breast cancer treatments in the 1990;s.

      The insurers were sued and ridiculed, but after 15 years the medical research seemed to indicate that their refusals were justified.

      Medicare and the FDA are both quite passive about pricing, compared to other OECD nations. They use reference pricing more than we do — where the national insurance plan pays the cost of traditional treatments only,

    • Nick
      I would add another. The press. Plans get killed when they make the front pages. I dont have to give examples.

      I would also add, the same reason plans waffle on the coverage–for reasons you cite above–are the same ones which will prevent docs utilizing safe harbors for liability protection. Sounds great, and the strategy gets mentioned frequently. I just dont see it, except for very low hanging fruit.


      • Brad,

        Totally fair point. Publicity can be murder for these plans. And I agree that there’s some resonance with physician safe harbors. All too often, the law is a convenient scapegoat for insurers and physicians. It’s important to be skeptical of that kind of scapegoating.


    • The problem is that medicine is complicated.

      We need to keep in mind just how difficult it may be to determine whether a given treatment is worth doing. For the vast majority of medical decisions, even ignoring physician self interest, the treatments are selected by convention, not because they have been proven by RCT’s. Only a small fraction of what is done has even been studied at that level, let alone proven. So the great bulk of treatments for which insurers pay are justified because they are common practice, not because they are known to be valuable.

      Some treatments come along and years later prove to be valuable. If an insurance company has refused to cover them, and it turns out that it would have helped, then they will face “legitimate” accusations of having killed granny. The defense that “no one knew at the time it would have been lifesaving” does not go very far when the subsequent data supports the therapy.

      Even worse, you have to worry about the moving target effect. The initial application of a treatment may be discouraging, so the company may deny coverage. But the method evolves to something effective, and then the company is back in the position of having to justify its refusal to pay. Constantly reviewing every therapy that has not already been clearly proven useful in RCTs (which is nearly all of them) gets expensive, and as noted above is not necessarily protection from the risk of denying a therapy that in retrospect would have been useful. So the company has to worry about what will be proven about a therapy in the future.

      Way safer to be relatively generous on what you will cover, make up the difference with higher premiums, and let the federal government pay the substantial costs of evaluating the therapies.

    • Americans do excess better than others. I’m not sure why. Some say it’s because Americans are exceptional. Others say its because Americans are optimistic. Still others say it’s because Americans are gullible. Whatever the reason, Americans do excess. And by excess, I don’t mean quality, I mean quantity, whether it’s food, housing, cars, clothes, or health care.

    • What Brad said, plus market share. Talk to execs in the insurance industry and you find that they closely monitor market share. (Hospitals too, but that is another story.) A few stories about not covering a cute young kid or any sympathetic patient, and they risk losing market share. Margins are not especially high in the insurance industry, so executive compensation is often linked to size. Also, if you lose too much market share, you lose the ability to dictate fee rates to most providers.


    • “There’s just not a business case for it— ..”

      Bingo! For the insurance industry this is not about healthcare and what is appropriate for it, it is about business, first. last, and always – and as long as we leave healthcare to the tender mercies of “the market” this will be true … For too long now we have witnessed the dire results of that policy ….

      • There are many ways in which competition, which is potentially beneficial, is blocked by the profit motive. I think it was Frontline that recently did a piece on the rise of so-called “superbugs’, drug-resistant bacteria. There is no collection and dissemination of data about the rise of these pathogens because no hospital wants to give its competitors an advantage by openly disclosing their findings. Competition, or the profit motive, is actually killing people because collection and publication of data could do serious, bottom-line damage to these for-profit institutions. Research by drug companies is stymied by this lack of data. But the drug industry also doesn’t see meds that are used only occasionally and for brief periods as having blockbuster potential. So again, an emerging threat to the general health and welfare of the country is hurt by the concentration on profit above all else.

        • Agree – there are other areas too – so many institutions feel the need to have the latest multimillion dollar technology to “compete for market share” with neighboring institutions – flooding “the market” with redundancies. And then of course there is all the money spent on advertising ,,,,

          One could make a good argument that all this “competition” is raising, not lowering, the cost of health care …

    • A lot of the problem on costs is that the new fancy expensive technology may not be better than older technology (proton beam, for example) but it reimbursed at a much higher rate. Even when the new technology comes down in cost, the reimbursement stays at a very high rate (ie. MRI vs CT).
      Perhaps they should just say they will reimburse the new at the same rate as the old technology unless it can be shown to be better.

    • Almost forgot. I suspect that the real drivers of health costs when it comes to unproven tech is actually stuff we have been doing for a long time. Knee arthroscopies and Back surgeries come to mind. Why do private insurers pay for that stuff? Well studied. We know it is mostly useless. Austin has published a number of studies along this line.



    • Thank you for addressing a topic that is incredibly important but poorly understood! I’d like to add my speculation to yours.

      You say there’s not a business case for more coverage exclusions, and I believe the reason is search frictions (see Cebul et al. in AER, August 2011). Because consumers (before insurance exchanges, at least) generally can’t make apples-to-apples comparisons of insurance plans’ coverage, they use reputation as a quality surrogate (see Hibbard et al. in Health Affairs, March 2012), which means risking not covering something is also risking losing a lot of market share if it gets bad publicity.

      Another result of search frictions is that even if an attempt to avoid covering low-value services succeeds and they lower their premiums accordingly, most consumers wouldn’t be able to tell that the value of that plan has just risen higher than competitors’ plans, so market share/additional profit probably wouldn’t flow to an insurer taking all these risks.

      Insurers know these things, and I would guess these are the same problems keeping insurers from jumping all over other premium-lowering innovations like reference pricing and tiered provider networks.

    • Most people do not even know what they pay for health insurance.

    • The insurrance company in California that lost a $77 million judgement was for just cause. Payment for ineffective treatment was not the issue. The company created a benefit for bone marrow transplantation as a treatment for advanced breast cancer when it granted coverage to the wife of one of the Company’s executives. Even though this strategy was still experimental, now proven ineffective, paying for the treatment created a benefit for all the other insured persons. The lawsuit was filed in behalf of an insured with advanced breast cancer who was refused coverage of a bone marrow transplant after the company had already granted coverage for this therapy eventually given to the executive employee’s wife. Obviously, the judgement was granted for illegal governance.

    • Insurers get to keep a percentage of whatever is spent. Why would they want to decrease the amount that is spent. They can always blame providers and expensive technology for perpetual premium increases.

    • “There’s a consensus that the primary driver of escalating health-care costs is the rapid adoption of new and expensive medical technology.”

      There may be a consensus, but that consensus is WRONG.

      I don’t get this (except I do). We know what causes inflation: Demand exceeds supply. Except when it comes to medical care apparently. Suddenly with medical care, the rules have to be different? Why?

      The fact is that people want more healthcare than is available. You solve that either by making people want less healthcare (the GOP plan), or by increasing the amount of healthcare for sale (no one’s plan). Why is this so hard to understand, and why does this difficulty only happen when it comes to medical care?

      The reason of course is obvious. The vested interests have a massive cash cow on their hands, and the last thing they want is for the upward pressure on prices to be contained. And so these interests pay a lot of money to throw subterfuge out there to mute the real problem, which can be solved, by replacing it with artificial problems that can’t be solved.

    • I don’t know if this is a 1% or 25% problem, but lets talk about a baby elephant in the room:
      Doctor’s Greed
      My 85 year old, dying grandmother was taken to the dentist by her son. The dentist said, you need 35,000 dollars worth of bridge work.
      Her son nearly through the dentist out the window.

      Yeah , the majority of MDs are hardworking, etc etc, but there is a large minority that are not honest.