• Yes, it is amazing this is not widely understood

    When I don’t have time to blog I still share my thoughts, just in much more compressed form and mostly with Aaron. As we’ve come to learn, he and I basically share the same brain so I can communicate what would be a 500-word blog post to him in about 12 words. Why don’t I tweet it? Seriously, I don’t know. Sometimes I like things to be more private.

    Anyway, yesterday, as Paul Krugman was gearing up for his latest column, practicing his arguments on his blog, I kept emailing Aaron that what he was saying was old news, very old. Meanwhile, a few bloggers picked it up as if it was a new insight. I’m sure they know the history, but just didn’t want to go into it. What was Krugman talking about and why am I so sure it is an old idea? He just explained in a post, so now I don’t have to:

    Readers familiar with health care economics may have noticed that today’s column was largely a translation into English of part of Kenneth Arrow’s seminal article on the economics of medical care (pdf), especially part IIB — the article that explained, half a century ago, why health care can’t be treated as an ordinary market. It’s amazing that we’re still fighting to get people to understand the same lesson.

    What Krugman was focusing on was the fact that physicians (as well as other providers) have much more information relevant to health care than do consumers. That information asymmetry confers market and political power. (That power is there, whether exploited or not.) Much about our health system and costs follow. More here. See also the work of Paul Starr.

    Yes, it is amazing that people don’t understand this. I find it hard to believe that that is so. Is it? Or do some people just pretend not to understand it? Is this a hard concept or not?

    • It’s not a hard concept at all. It’s very easy to understand that it held 50 years ago, when Arrow’s article was published. It’s just no longer true today.

      Today, the information asymmetry to which you refer is largely negated by an inverse time asymmetry. Doctors spend so little time with patients that they are often unable to bring their vast store of knowledge to bear. The result is that patients–who do have the necessary time and motivation–do research on their own using the internet, self-diagnose, then shop for a doctor who will provide the treatment they’ve already determined they need. In other words, they are consumers.

      It is amazing that people don’t understand this. I find it hard to believe that that is so. Is it? Or do some people just pretend not to understand it? Is this a hard concept or not?

      • What proportion of patients do this and what proportion of the costs are associated with them? Even when those most able are motivated to do so with high deductible plans, the impact on health costs is trivial. See http://theincidentaleconomist.com/wordpress/the-health-spending-distribution-is-very-skewed/ .

      • As a practicing physician, I see what you describe in maybe 5% of our patients. Most patients just say “Whatever you think best doc”. Where people receive their care is determined mostly by their insurance, especially when there are limited networks, geography and referrals by their PCP. Of note, those who do come in with their newly found internet expertise often bring a lot of helpful information. I generally like having these patients. Every now and then they are way off base, like the lady who wanted to use meditation techniques in place of anesthesia for her heart surgery, but I like better informed patients so this is an ok tradeoff.


        • steve,

          Question for you: do better informed patients end up with better care? The only way you can answer this from your experience is to ignore the selection effect. I.e., you can only judge based on the patients you see. Next question: do better informed patients end up using less or more health care resources (costs)? Again, there is a selection problem, which we’ll have to ignore.

          My guess: better informed patients are not homogeneous. Some likely get better care, others probably just get more care (which we know can be a mix of better/helpful and unnecessary). Also, likely costs are higher, not lower, but if the health outcomes are better, that may be worth it. On the whole, I’d be shocked if patient knowledge made a huge difference, particularly on costs, though also on outcomes and quality. But I’d like to hear from you and other docs on this.

          • That is a tough question, and would be influenced by my personality. There are subgroups that definitely get better care. We are a referral center for peds imaging. I get a lot of kids with syndromes, some of which I have never heard of or only read about once or twice. Often, the mothers, and it is always the mothers, come equipped with pages of information. They have phone numbers for hotlines and experts to call. I think that their kids probably do get better care. Some of these Moms are real heroes I believe.

            For the run of the mill internet informed patient, I would guess that there is a long term trend towards better care, and maybe lower costs for those patients who are informed AND open minded. With those patients I can usually overcome any fears they might have about a specific treatment which will provide a better outcome, but may provide more short term pain or inconvenience, for example pain blocks after shoulder surgery. The internet experts who have their minds made up probably are at risk for worse outcomes or more care/less care. It may be a function of my specialty or just luck that I do not see many in this category. From talks with my PCP buddies, I think that this is more a problem for them.

            Overall, the number of patients who are uninformed or not interested details about their care, far outweighs the number who make an effort to search out and learn about their problems. I guess the world is not like the internet.


    • Are you encouraging patients to be ignorant of their care?

      I do not doubt the glaring fact that there are information flow problems. I believe that this can be at least be attempted to be solved, but it seems that your posts accept that fact and wishes to do nothing about it. I do not think it is sustainable for our system to constantly have disengaged patients. In fact, I think that is one of the reasons why healthcare is so expensive. Patients are held by their hand and are insulated by everything in the system, whether payment or information. With the coming shortage of doctors and shortage of facilities, an empowered patient is something that can be created and needs to be created. It is false to assume that it will always be the same environment.

      My suggestion… why not give it a try? Instead of rolling out trillions of new expenditures, why not empower patients? I guess that is what makes me lean to the right on these issues. I do not think the government can effectively and efficiently solve this problem.

      Always enjoy your posts. Every post makes me think!

    • How would we empower patients? I like the information provided by the Mayo Clinic. The problem is that most people don’t know it’s there.

      As to the quality of the care received: it seems that Paul Krugman is being almostly willfully naive. Kenneth Arrow did show that asymmetric information can lead to suboptimal results. He did NOT show that doctors’ ethics obviate that difficulty. He only hoped. Here too, patients might benefit from some metrics, don’t you think? IPABs can’t be the only answer.

      • I think the discussion is too narrow. A lot of knowledge is, almost by definition, beyond patients’ ability. Think what is done during surgery, in the ER, in the ICU. Think how imaging is read, tests interpreted, etc. Consider also the management of multiple and/or complex chronic conditions. Patients can learn a lot, but nowhere near enough to function as their own doctor. It may be possible to narrow the information gap, but it cannot be closed with current technology except in a small minority of cases.

        I’ve asked before: someday, will a significant amount of medicine be conducted by expert systems and robots? If not, why not?

        • will a significant amount of medicine be conducted by expert systems and robots

          There was an article in The New Yorker a few years ago about expert systems reading EKGs and mamograms. They did better than the typical doctor and radiologist.

      • Oh, about medical ethics: No, I do not think they address information asymmetry. If you read Paul Starr’s Social Transformation of American Medicine you’ll get a very different view from what Krugman implies.