Bias and Health Care

With only a few signficant exceptions, health insurance plans in the U.S. have quite broad provider panels. That is, there is almost no such thing as a plan that offers a full range of clinical services, but that restricts its members to a narrow panel of cheap providers. This is quite a remarkable fact in a country with tens of millions of people with no insurance at all, and many millions more who are terrified that they might lose theirs. Why wouldn’t employers that are on the bubble between continuing to offer broad insurance and cancelling insurance altogether offer some sort of narrow insurance plan instead? Why wouldn’t employees take such an offer?

The standard answer to this question is that people really dislike having to travel for health care, particularly for hospital care, because they want to be in their own community, treated by their own physician, and with their families and friends nearby and able to visit them easily. This strikes me as an incomplete answer, because there is a lot of price variation between providers, so a narrow panel insurance plan could be much cheaper, and it’s hard to imagine that people value being near home so much that they would be willing to pay (in expectation) thousands of additional dollars for a hospital admission just to avoid traveling to a hospital that was, say, an hour away from home.

It seems to me that a better answer is that when they actually get sick, a significant fraction of people adopt the attitude that care at their preferred provider is a fundamental right, and become genuinely indignant at the suggestion that they should have to travel just because their insurance company tells them to. This indignation leads them to take some kind of costly action against either the insurance company or against their employer (through which the narrow insurance plan was offered). That is, patients can’t credibly commit not to freak out and start kicking desks and peeing in water coolers when they are held to the terms of a narrow insurance plan. This makes such plans more expensive to offer, which may explain, at lest in part, why they are so scarce.

Whatever you think of this particular story, I find it odd that behavioral explanations for things are so common in finance, but so rare in health economics. I would think that health care would be an area where biases and other behavioral tics would be particularly common.

I have a paper about this, which you can find here. The contents of the paper and of this blog in no way reflect the opinions of my employer.

GD Star Rating
Tagged as:
Trackback URL:
  • How far are you talking about when you refer to “travel”? It can be very difficult to travel very far when one is sick, especially seriously ill. I personally find it outrageous that in the US medical system it is no longer the case that doctors do not visit people in their homes, except in the most unusual of circumstances. It is now accepted that this is perfectly OK, normal. But it is not the case in many, if not most, other countries, and it was not the case in the US in the not-too-distant past. So, I think talking about making people “travel” who are sick, is simply encouraging what has already been a very bad trend in medical care in the US that it is the duty of the sick to go moving all over the place, sitting endlessly in waiting rooms, and so forth. Maybe it saves money, but it is a dehumanizing degradation that certainly does nothing to improve the health of the population, and the thoroughly lousy stats of the US compared to other countries on infant mortality and life expectancy, continue to suggest that this is the case.

  • I agree that strong and puzzling emotions are very important to understanding health care. But since we are puzzled and do not understand much about health care, we must be cautious in claiming to see “bias.” Until you can identify the sign of the difference between what is and what should be, I’d say you haven’t yet identified a bias.

  • Curt Adams

    For examination purposes, long travel distances aren’t practical. If you’re having pain or fever, long travel isn’t practical, or even possible. You can’t drive an hour with cramping abdominal pains in heavy traffic and you can’t always get a ride. So primary care physicians, and any physician you see regularly for chronic conditions, need to be nearby. My bf’s allergist moved from 10 minutes away too 40 minutes away from his work. Since he can only see him in business hours, and as a manager frequently has to firefight, that’s proving a huge burden to him.

    Now for planned treatments that’s not the case – but in my experience, planned treatments often do require substantial travel.

  • David J. Balan wrote: “The standard answer to this question is that people really dislike having to travel for health care, particularly for hospital care, because they want to be in their own community, treated by their own physician, and with their families and friends nearby and able to visit them easily.”

    That’s news to me. For the most part, those I know take their family’s health very seriously, and most of them prefer to turn to those they trust when seeking health care.

    For most good health care is not about cost or convenience, but about quality, or at least your confidence in its quality.

    I find your post on “Overcoming Bias” ironic. You post strikes me as being so non-sensitcal that I’m left wondering as to whether it reflects a bias of your own 🙁

    However, I also acknowledge that my being puzzled by your opinion does not mean you are biased … no more so than does your puzzlement mean there exists a bias on the part of those you write of.

    p.s. Robin Hanson, very nice point. I hope it is ok that I took it a bit further.

  • David J. Balan

    Barkley and Curt, there is no doubt that travel costs are real, and that at least in some instances travling is out of the question. Presumably even narrow plans would not require travel in those instances. But that does not change the fact that there is a great deal of heterogeneity in prices across providers (particularly hospitals) that are separated by distances no greater than those covered by many people in *daily* commutes. It seems very reasonable to me to suppose that if there were no insurance and people could get a surgical admission for $3000 less at a hospital an hour away, a great many would do so. By the same token, many people (assuming they rationally anticipate future illness) would be willing to buy an insurance product that requires such travel and is correspondingly cheaper if they were free of the indignation that I describe.

    You are right that the “bias” here is not a biased estimate of a parameter, but it is an irrational mental attribute that, conditional on you having it, makes you unable to make a contractual arrangement that otherwise would be in your interest. I suppose that strictly speaking that doesn’t count as bias, but it seems like it’s in the same neighborhood.

  • David, I meant that when consumer behavior is systematically odd, that suggests you don’t understand what exactly the consumers are demanding, which suggests that you are not well positioned to know whether any particular product package is a good deal for them or not.