• It’s been a great century – ctd

    Imagine:

    Until 1961, patients with acute myocardial infarction — if fortunate enough to survive until they reached a hospital — were placed in beds located throughout the hospital and far enough away from nurses’ stations that their rest would not be disturbed. Patients were commonly found dead in their beds, presumably from a fatal tachyarrhythmia. Indeed, the risk of death occurring in the hospital was approximately 30%.

    That’s from “A Tale of Coronary Artery Disease and Myocardial Infarction,” by Elizabeth Nabel and Eugene Braunwald (NEJM). We have come a long way since 1961. Improvement in care for patients with cardiovascular disease is one of the great medical success stories of the 20th century, complementing great strides in public health. We owe this triumph of medicine to the interdisciplinary efforts at enhancement to the technology of care (from devices to drugs to surgical technique), evidence from clinical trials, and dissemination of best practice and lifestyle improvements via education programs.

    For these efforts, we have been rewarded with dramatic reduction in cardiovascular disease specific mortality. This chart, from the Nabel and Braunwald paper, illustrates:

    A more detailed timeline of milestones in cardiovascular disease is found in a supplementary appendix (pdf). (The chart does not prove causality. Other factors may have contributed to mortality decline.)

    For all that, heart disease remains one of the top causes of death in the US and “cardiovascular disease, including heart disease and stroke, is the leading cause of death worldwide, including low-income and middle-income countries.”

    The experience with cardiovascular disease illustrates evidence-based medicine at its best or, at least, closer to “best” than is achieved in much of medical science. Unfortunately, not all of medicine has undergone similar, decades long scrutiny and improvement with sound scientific and public health methods. It’s well known that there is far too much eminence-based medicine,* driven by the anecdote or intuition of powerful physicians and medical groups operating under perverse incentives and within a fractured health system. We can do so much better. Our experience with cardiovascular disease proves it.

    * My source for the brilliant “eminence-based medicine” is Ashish Jha.

    AF

    Share
    Comments closed
     
    • Did all of those healthcare advances put together have half as much impact as the decline in smoking after the partial ban on cigarette advertising?

      • It’s a good question and not one I can answer. It’s for this reason that I noted in the post that one cannot infer that the indicated studies and interventions are the cause of (all) the mortality decline.

    • I have the same question as Anderson above. This reminds me of charts that show the decline in infectious disease mortality and how much of that came before antibiotics and other fancy advances, and was probably caused by (mostly) improvements in hygiene and (to a somewhat lesser extent but still very important!) vaccines.

    • It would be interesting to plot the percentage of smokers over the same years.

      • @Richard A
        don’t have plot handy but smoking prevalence was ~50% adults in 1950 and ~25% in 1990….down to ~21-22% today.

    • I guess I’m a little puzzled by the graph. It’s true that since a peak around 1968, deaths due to cardiac disease have been declining. Of course that’s a good thing. But what’s strange is that at which those deaths decline hasn’t really changed at all — that is, the downward slope of the graph is pretty much constant for the whole duration, regardless of the innovations. That suggests that something else might very well be responsible for the main effect. I’m very skeptical of the last two “innovations”, the genome-wide association and the deep sequencing of 2009 — have those actually led to improvements in real-world clinical practice, or are they currently of theoretical interest only?

    • Wonder if it is just a coincidence that this happened when Medicare came along? How much of this happens w/o some way to pay for the advances that were made?

      Steve

      • @steve
        hard to imagine that payment for revascularization by Medicare didn’t have an effect….both on developing more techniques and on outcomes. Would like to see the graph for Cancer over same period and expect the average would wash out tumor/types differences.

    • It’s worth asking how many of the scientific and clinical advances shown in the above timeline had their genesis in the countries that have embraced the kinds of health care systems that the authors of this blog have championed.

      There seems to be a tendency to take both scientific and clinical progress as givens that will occur with essentially the same frequency no matter what resource constraints or incentives are in play. It’s far from clear that the empirical record supports such a claim, whether we’re talking about drugs, devices, or clinical regimes.

      • No it is not worth asking that because your question is not well formed. What do you mean by “the kinds of health care systems that the authors of this blog have championed”? What kind of backward, passive-aggressive phrasing is that? Come out and say what you mean instead of trying to turn it into a barb.

    • It wasn’t meant as a barb. I find it odd that you took it as such, given that there’s is a clear policy bent on display in your posts that makes it possible to click back and forth between your posts, and say, John Goodman’s or the text of a speech delivered by Richard Epstein and determine that they favor less government intervention in health care and you favor more.

      That’s neither objectively good nor bad, but it is an obvious difference and that difference does have important implications related to the chart which is the subject of your blog post.

      The overriding theme of your posts seems to be that the best way to deliver health care is via a centrally administered public system where bureaucracy, price-controls, and rationing supplant the functions that private actors, market-prices, and competition serve in the rest of the economy. All-payer rate setting, etc. All things being equal, the closer we approximate the system in Canada, the better, no?

      My contention is that the more price controls and bureaucratic rationing you have, the less innovation you get. Maybe I’m wrong and these systems produce more new drugs, new technologies, and more clinical innovation of the kind shown in your chart. I don’t think that’s the case.

      I happen to think that the vast majority of the improvement in health care has come about as a result of innovation, and the welfare gains driven by reshuffling the same pieces that existed in 1950-60-70-80-etc have been incredibly small if not entirely non-existent. Hence the incentives that influence the rate of innovation matter a great deal, and It *is* worth asking how much of the improvement in health care has been driven by innovation, and how much has been driven by better management of resources that existed as of some arbitrary date.

      If I’m wrong, I’m wrong and the case for imposing Canada-like administrative controls only gets stronger. If I’m right, all that follows is that reduced innovation – and all of the consequences that arise from that – is one of the potential trade-offs that honest, serious advocates of such systems will have to incorporate into their analysis of the pros and cons of moving health care in that direction.

      • “You favor more” :: What I write has nothing to do with what I favor unless I’ve explicitly indicated that. If I seem to have overlooked some consequence of some policy, I welcome that being pointed out, but not with arm-chair accusations and conjecture, but with actual citations of studies. I really don’t care what people — at least those with little power to implement change — believe apart from what can be demonstrated with evidence.

        “important implications related to the chart which is the subject of your blog post.” :: Many of the studies indicated in the chart were precisely of the form I favor. We should have more of them, not less. I think you’ll find that I’ve been very supportive of science, research, and evidence-based medicine. That’s not what you seemed to be suggesting.

        “The overriding theme of your posts seems to be that the best way to deliver health care is via a centrally administered public system where bureaucracy, price-controls, and rationing supplant the functions that private actors, market-prices, and competition serve in the rest of the economy. All-payer rate setting, etc. All things being equal, the closer we approximate the system in Canada, the better, no?” :: Wow, no. Are you confusing me with someone else? All-payer, in particular, can and should include market signals. I’ve written explicitly about that. The context in which I’ve suggested it as an option is in the instance ACOs exacerbate provider market power. I’ve written elsewhere about the value of market signals (competitive bidding). It is hard for me to understand those who would not want to take advantage of them, where well applied and within an appropriately regulated market.

        “My contention is that the more price controls and bureaucratic rationing you have, the less innovation you get.” :: A perfectly fine contention and a concern I share. At the same time, along with all the wonderful technologies developed in the US, we also excel at medical innovation that is not very valuable, of unproven value, or misapplied coinciding with underprovision of proven, effective care. There’s more than one problem to address.

        “honest, serious advocates of such systems will have to incorporate into their analysis of the pros and cons of moving health care in that direction.” :: Agreed. I do hope you’re not claiming anyone here is not honest or serious. I think you will find that nearly every time I suggest an avenue for reform I point out the limitations, if not in the same post than in another. It really frustrates me when people like you seem to overlook that and presume to believe I have an agenda that I don’t. Moreover, even if I did, I think it is irrelevant. I’ve proven time and again to being receptive to evidence and scholarship. Show me some of that and I will have much more respect for your contribution to the conversation.

    • Last post on this topic:

      You’re nothing if not honest and serious. Really. Should have made that clear. As to having an agenda/perspective, there’s *really* nothing wrong with that. Maybe perspective is the better word. It’s not what you believe, but why, and I’m far from the only person who believes that it’s important to consider the potential impacts of broad legistlative reforms on medical innovation since the potential implications for future welfare are huge. It’s true that I have zero power to implement change, but my goodness(!), that’s a high bar for anyone to clear and didn’t seem to be listed as a precondition to comment here.

      This impact of innovation on growth/welfare isn’t a new topic in economics nor in the sub-discipline of econ applied to health care, and as a full-time professional working in the field I presume that you have access to all of the pertinent literature that addresses this issue. It’s true that there aren’t many, but they’re out there. Maybe Whitman and Raad’s 2009 piece, and Abe Dunn’s 2011 pieces are useful places to start. Here’s the key sentence from the Dunn piece: ”

      “Thus, accounting for changes in quality appears to be very important for properly measuring prices in the market for anti-cholesterol drugs. This

      result highlights the potential importance of accounting for quality changes when measuring prices and output in the health sector where technology is a primary driver of expenditure growth.”

      As far as formal literature citations dealing with the structural requirements for innovation in healthcare are concerned, there’s a couple reasons why there aren’t thousands of papers on the topic. One is that it it’s an exceptionally complex topic, and very poorly understood. There are too many variables to hold any particular factor or subset of factors constant and run a regression, or develop a formal model with much predictive value. In cases like this where the magnitude of the measurement-errors and unknowns is large, and often the best you can do is apply principles/heuristics to retrospective data and draw tentative conclusions such as – if something that you don’t understand very well seems to have worked well in the past, be very careful when you change it, and if you must change it – an incremental approach is much less likely to have large and destructive unintended consequences.

      We also know things like that demand curves tend to slope downwards, that capital tends to migrate to the areas in which it can earn the highest marginal return, etc. Consequently limiting the conversation to citations of previous studies and excluding informal principles-based arguments like “If we constrain the rate of return on medical innovation by imposing price controls we may well get less of it,” seems like a puzzling pre-condition for a dialogue related to this topic, and others of its kind.

      • “It’s true that I have zero power to implement change, but my goodness(!), that’s a high bar for anyone to clear and didn’t seem to be listed as a precondition to comment here.”

        No, it’s not a precondition at all. This is my philosophy, that beliefs of most people should not matter when one is seeking truth. I don’t even care much what I believe, in the sense that I don’t think it is important. I don’t think anyone should care what I believe. I think the evidence and argument based on it are paramount, but this is not widely held. Most are comfortable substituting the beliefs of those they revere or respect for their own thinking, somewhat independent of the virtues of their argument.

        “Consequently limiting the conversation to citations of previous studies and excluding informal principles-based arguments like “If we constrain the rate of return on medical innovation by imposing price controls we may well get less of it,” seems like a puzzling pre-condition for a dialogue related to this topic, and others of its kind.”

        But this is a dodge. The issue is not what type of market regulation maximizes innovation. It’s far, far more complex than that. Innovation is not the only criterion, or I should hope not. As I wrote previously, we also have underutilization of valuable health services, a problem “freer” markets have not solved. Additionally, it is completely reasonable that we can have too much innovation — and particularly of certain types — as well as too little. It is not enough to say, “Be careful not to stifle innovation,” one has to strive to indicate just how much we need and to foster the right types. For this, some study, with actual data, may be of value. It is not too much to ask.

        Granted, it may not exist, but then we cannot make quantitative headway. We can bicker back and forth — if we choose — about principles, but it does no good. I can just as easily justify we have too much innovation now as you can justify we have too little. Who is right? We will never know.

        That’s not a useful conversation. There’s no point having it. And, to my original point, the way you raised it was not helpful to the conversation at all, implying I have made up my mind on the question. I most certainly have not! That would be foolish in the absence of data. I would do no such thing.