Ideas that spread fast and slow

Atul Gawande (the thinking man’s Malcolm Gladwell) asks:

Why do some innovations spread so swiftly and others so slowly? Consider the very different trajectories of surgical anesthesia and antiseptics, both of which were discovered in the nineteenth century. The first public demonstration of anesthesia was in 1846. The Boston surgeon Henry Jacob Bigelow was approached by a local dentist named William Morton, who insisted that he had found a gas that could render patients insensible to the pain of surgery. That was a dramatic claim. In those days, even a minor tooth extraction was excruciating. Without effective pain control, surgeons learned to work with slashing speed. Attendants pinned patients down as they screamed and thrashed, until they fainted from the agony. Nothing ever tried had made much difference. Nonetheless, Bigelow agreed to let Morton demonstrate his claim.

On October 16, 1846, at Massachusetts General Hospital, Morton administered his gas through an inhaler in the mouth of a young man undergoing the excision of a tumor in his jaw. The patient only muttered to himself in a semi-conscious state during the procedure. The following day, the gas left a woman, undergoing surgery to cut a large tumor from her upper arm, completely silent and motionless. When she woke, she said she had experienced nothing at all.

Four weeks later, on November 18th, Bigelow published his report on the discovery of “insensibility produced by inhalation” in the Boston Medical and Surgical Journal. . . . The idea spread like a contagion, travelling through letters, meetings, and periodicals. By mid-December, surgeons were administering ether to patients in Paris and London. By February, anesthesia had been used in almost all the capitals of Europe, and by June in most regions of the world.

That was the happy story of the rapid acceptance of an innovation. Next comes the frustrating story:

Sepsis—infection—was the other great scourge of surgery. It was the single biggest killer of surgical patients . . . In the eighteen-sixties, the Edinburgh surgeon Joseph Lister read a paper by Louis Pasteur laying out his evidence that spoiling and fermentation were the consequence of microorganisms. Lister became convinced that the same process accounted for wound sepsis. . . . During the next few years, he perfected ways to use carbolic acid for cleansing hands and wounds and destroying any germs that might enter the operating field. The result was strikingly lower rates of sepsis and death. You would have thought that, when he published his observations in a groundbreaking series of reports in The Lancet, in 1867, his antiseptic method would have spread as rapidly as anesthesia.

Far from it. . . . It was a generation before Lister’s recommendations became routine and the next steps were taken toward the modern standard of asepsis—that is, entirely excluding germs from the surgical field, using heat-sterilized instruments and surgical teams clad in sterile gowns and gloves.

Pretty annoying, huh? Gawande asks why, and shoots down a couple of natural explanations:

Did the spread of anesthesia and antisepsis differ for economic reasons? Actually, the incentives for both ran in the right direction. If painless surgery attracted paying patients, so would a noticeably lower death rate. Besides, live patients were more likely to make good on their surgery bill. . . .

Maybe ideas that violate prior beliefs are harder to embrace. To nineteenth-century surgeons, germ theory seemed as illogical as, say, Darwin’s theory that human beings evolved from primates. Then again, so did the idea that you could inhale a gas and enter a pain-free state of suspended animation. . . .

The technical complexity might have been part of the difficulty. Giving Lister’s methods “a try” required painstaking attention to detail. . . . But anesthesia was no easier. Obtaining ether and constructing the inhaler could be difficult. You had to make sure that the device delivered an adequate dosage, and the mechanism required constant tinkering. Yet most surgeons stuck with it . . .

And then he gives his theory:

So what were the key differences? First, one combatted a visible and immediate problem (pain); the other combatted an invisible problem (germs) whose effects wouldn’t be manifest until well after the operation. Second, although both made life better for patients, only one made life better for doctors. Anesthesia changed surgery from a brutal, time-pressured assault on a shrieking patient to a quiet, considered procedure. Listerism, by contrast, required the operator to work in a shower of carbolic acid. Even low dilutions burned the surgeons’ hands. You can imagine why Lister’s crusade might have been a tough sell.

This has been the pattern of many important but stalled ideas. They attack problems that are big but, to most people, invisible; and making them work can be tedious, if not outright painful. . . .

Roaches

The above all makes sense, but I think there’s something else going on, something I find difficult to formulate but I think is real.

The example I have in mind is roach extermination. When I worked with Ginger Chew and her colleagues in the school of public health at Columbia several years ago, I learned that the way to get rid of roaches in your apartment is to clean up your apartment, throw away all the open food, put boric acid in the cracks in your floor and walls, and seal up the cracks. It’s not easy but it does the job. But that’s not what they do in the Columbia-owned-and-operated building where I live. What they do is, every month they put a signup sheet up by the elevator and then an exterminator comes into the building and bombs the apartments of everyone on the list. The same people sign up every month, of course. Instead of thinking, “Hey, bombing doesn’t work,” they seem to think that it’s something they need to do monthly. Good business for the exterminators but not so effective at getting rid of roaches.

So why do they do it that way? One thing I’m definitely not going to do is talk with my neighbors and suggest they try a different approach. My impression is that people get very defensive about things like this. Also, I’m no roach expert; really I’d want to bring someone in from the school of public health to have this conversation.

Anyway, my impression is that people like any treatment that feels like “pushing a button” and they don’t like anything that feels like work. And if you tell people that pushing the button doesn’t really work, they get all bristly on you. Even though, in this case I think the effective treatment is ultimately less work than the bomb. Unfortunately, Columbia has it set up so they bomb for free, but they don’t provide a free cleanup and sealing service.

P-values

I feel like there’s something similar going on in scientific research, when statistically significant p-values are used to declare victory (for some recent examples, see here and here and here and here and here). I know that these methods have become popular—but, then again, my neighbors are getting their apartments bombed for roaches every month, they just keep on doing it. Pushing that button.

Please note: the previous paragraph is not an argument that there’s a big problem with p-values in scientific practice. Rather, conditional on you already agreeing with me that there’s a big problem with p-values in scientific practice, the previous paragraph is a speculation of one reason why this has happened.

P.S. Regarding the roach bombing, see this informative comment from Ryan Welch.

29 thoughts on “Ideas that spread fast and slow

  1. There may be parallels between this and the West’s apparent unwillingness to consider adaptations of Far East-style Rail + Property models for the funding of rapid transit infrastructure, especially where it aims to serve suburban car-dependent regions. We cling to the urban funding model (grants from governments) that has successfully employed passive value capture in older parts of New York, Toronto and Chicago, even though it has proven prohibitively expensive in largely sterile suburban areas. Out where the car is king, the land uses segregated and the landscape often hostile to the pedestrian, we need much more than the grafting-on of expensive subways or LRT lines to trigger the fertile urban forces that generate returns on investment. It would be wonderful if R+P could be merely adopted rather than adapted, and if R+P didn’t come with some moral-syndrome puzzles to work out. Alas, we’re going to need R+P to reverse our ongoing inertia toward unsustainable sprawl, so we’re going to have to get on with the difficult thinking and experimentation. Funny how such a great and important idea is so slow to catch on on the continent that needs it most.

  2. You seem awfully sure of your anti-roach strategy versus the exterminator-once-in-a-while approach. Perhaps both work? The exterminator “bomb” is likely a stronger version of your boric acid approach?

    • Rahul:

      I might be wrong; the recommendations I give are what I remember from my conversations with the experts. But, no, the bomb is not a stronger version. Key steps of the recommended approach are sealing the cracks. If you just drop the bomb without sealing the cracks, it will kill some roaches and then new ones will return. Hence the need for the bomb every month.

    • Although I am currently pursuing my PhD in political science, I’ve worked in pest control for 4 years and have a B.S. and M.S. in Entomology. I cannot help but weigh in a little here. Rahul seems skeptical of your embracing your “anti-roach” strategy. Let me just say that although the specifics are not exactly correct in my opinion, the main idea is — integrated pest management (IPM) is the most effective long-term strategy to controlling pests. IPM is basically tackling the pest management problem from all angles (sanitation, exclusion, as well as pesticides). Repeated “bombing” leads to frequent exposure to the active ingredient in the pesticide. Given the non-100% effectiveness of available active ingredients (whether due to the mode of action or exposure rate) and the incredibly quick generation times of roaches, resistance to the active ingredient is almost inevitable. But people continue to ask for “bombing” (or other wholesale chemical treatments), so pest control operators continue to supply it. Despite our repeated efforts to train pest control companies, many pest control operators also prefer wholesale chemical treatment. I believe this is probably due to a few reasons, but the most important one is that it is faster and results are good (in the short term) — both customers and operators like this. However, there is a collective action problem lurking (I told you I switched to social science). Roach populations that are repeatedly exposed develop resistance. Having just your apartment or restaurant treated wholesale with chemicals probably won’t cause the population of roaches to develop resistance, but a norm that subscribes to that management strategy will be bad for everyone because one day the “bombing” done every month won’t work anymore (this happened in the 90s with German cockroaches). And one may say, “eh, they’ll just develop another chemical.” But, this is unrealistic as chemical companies don’t have the required resources and incentives to always stay ahead of resistant pests. How does this relate to p-values? My guess is that p-values are quick and easy like “bombing.” However, one day we are going to turn around and say, “that’s not working anymore.” In fact, one can argue we are already there considering the many blog posts about how (ir)relevant political scientists are to policy makers.

  3. Even conditional on agreeing with Andrew that there’s a big problem with p-values, the reason may well be that there’s no good, universal alternative.

  4. Another example: the British Navy was bluntly informed that citrus juice stopped scurvy. Scurvy was the biggest cost the Navy had in terms of crew: unless there was a major battle, they’d lose many times more men each year from scurvy than from any form of violence. In the thousands. And scurvy especially ravaged British ships at the limits of the Empire and weakened their ability to fight in the Far East, in the Southern oceans, etc.

    Nothing happened. It took a few decades before lime juice became part of rations and, as you might guess, limes were the worst choice of the fruits that worked.

    Why? People blame bureaucracy but I think it’s the structure of the enterprise. The Navy had a recruiting system which was designed to produce x number of men. (Remember, it sometime included press gangs.) People were invested in this structure. Lots of changes would flow from lowering the number of replacements. You’d think that would be good but think about the people invested in the process of recruitment, from admirals on down, and all the money that entails.

    Apply to sepsis: you have a system. It kills people but people have always died. And you have a lot of people invested in the system as it is. Think of all the jobs that have to change.

    In American business, I often saw a similar mindset in which companies chase revenues without much if any regard for the marginal cost of that revenue – only when things don’t work do they realize they were losing money. But they put a high bar on cost savings initiatives. I’ve seen surveys about what executives expect as rates of return to approve cost savings measures and these matched my observations. The requirements are often absurd: all money back in 5 or even 3 years (and they don’t recognize how that carries forward). It’s often said that US business is a sales-oriented culture. That bespeaks the same issue: a structure invested in sales which can’t see the value in cost-savings measures. We saw what happened when that culture met culture that valued cost-savings, that worked to reduce WIP inventory through JIT processes, etc. (That is, we got our asses kicked.) We see today that many companies react to harder times by reducing headcount rather than explore cost savings measures; they remain wedded to that other model.

    In terms of statistical thinking, I see this kind of thing as a persistent normal distribution which necessarily rejects tail information. It means any system has future blindness; it is unable to recognize which tail information signals a change in direction that incorporates the past tail into the future central interval. We all do this to some extent. We can’t always be open-minded. We need to commit to ideas and courses. Systems become more rigidity with size and certain kinds of organizational structure.

    And they certainly become more rigid as one reinforces one’s conclusions by seeking out confirmation and reducing or eliminating exposure to alternatives. See FoxNews.

    • I wonder, how many times previously (if at all) had the British Navy been told of other alleged solutions to scurvy? I don’t know.

  5. I think you may be hitting on something important that also points in the direction of how change is likely to happen, and why it hasn’t already. P values are what people know, they’re (relatively) easy, and they pass the threshold for working well enough — just like the roach bombs. The alternative seems to most people — and I’m speculating here — to be difficult and uncertain and certainly a departure from the path of least resistance. If you want people to move from roach bombs to something that is harder in the short run, it’s rarely enough to try to convince them that they’ll be better off in the long run. You need to make it easier for people to change, you need to give them a road map to change, you need to show them that everyone else is doing it, and even then it’s an uphill battle. Thanks for the food for thought!

  6. This is economics. If UC offers bombs for free, then that is what people choose.

    If instead they offered a subsidy of $x per year, enought to cover one Gelman Roach Exterminator treatment a year, but not enough to cover a bomb every month, then you might witness a different outcome.

    But there is a possibility that bombing is cheaper overall.

    • Anon:

      Yes, of course people are doing it because it’s free. My point is that the university thought it was a good idea to offer this particular treatment for free, rather than instead offering something more effective that would not feel like “pushing a button.” Also, I don’t think the recommended treatment would be needed as often as once per year. But I’m not sure on that.

      • The “free” part is a red herring I think. Exterminators are often hired by people paying from their own pockets too. So maybe bombing is indeed cheaper overall, after discounting the value of lost time & inconvenience?

        • Rahul:

          I suspect that bombing is cheaper only after including the psychic cost of “not pushing a button.” That is, I think bombing is less effective and more expensive (as indicated by the perceived need to do it every month) than the alternative, but that people do it because it feels more like pushing a button. Hence my connection to to Gawande’s stories of people using a worse treatment despite their being a clearly better alternative. People can get really upset at the idea that the button they are pushing isn’t really working.

        • I’m not sure. Perhaps you underrate the “pleasure” people derive from having the freedom of letting food lie around (and dirty dishes) and yet have to face no roaches.

          Just ask a typical grad student or the stereotypical absent minded scientist. :)

          Sometimes the IPM approaches are fine but just way too much hassle. Maybe people just prefer the bombing once a month to the discipline needed to have no food / dirty-dishes / trash accessible.

  7. Not just the British navy, and earlier than you may have supposed. I was reading a book about Magellan’s voyage (title and author escape my aging memory) and it claimed that the survivors arrived in Zanzibar, all with advanced cases of scurvy. The local Moslem doctor sent over bags of oranges, and they all recovered within days. The doc explained that sailors in the Arabian sea and Indian Ocean always carried oranges or lemons with them, and this prevented scurvy. when the survivors got to Portugal they told this to the authorities, and sailors went to work for other naval powers. It took several hundred years before citrus became an onboard staple for Europeans on long voyages.

  8. I agree with Dave that you are on the right track with your ‘button’ theory. i’d suggest, however, that while it seems different from what Atul is saying it strikes me that you could think if his theory as an explanation for *why* some innovations will be perceived as easy button pushes and others won’t. in other words, i think you’re both describing different aspects of the same psychological process. Dave’s point about the importance if familiarity is also noteworthy. perhaps there is a ‘familiar = easy’ heuristic that contributes to making people resistant to adopting new methods. the continued use of misleading and inefficient statistical routines would seem to support this idea. one thing that has always frustrated me about statistical methods papers is the lack of inclusion of statistical code for commonly used software packages in an appendix. people aren’t going to adopt a recommendation unless they have some concrete tools, like starter code. you obviously understand this but i continue to be amazed at how few methods scholars do.

  9. The economist in me wonders how often people sign up for bombing every other month. You know, because you bomb your apartment, and then everything is sorta-OK/not-great-but-tolerable for a month because most of your roaches move next door. So the next month your neighbor signs up, and the deluge returns to your apartment. This basically happened to us, so we switched to the boric acid/gel-in-cabinets method, and things have been much, much better. Of course, we live in a semi-slum (grad school!) so there is no way to actually deal with all the cracks, and the roaches move apt to apt through the walls anyway, but even then, the boric acid along the floor boards works pretty well (this gel stuff our landlord puts in the cabinets is pretty good too).

    I wonder, though, if you might find a stable, roach-free equilibrium by bombing the whole building, and then following up with some “boric acid encouragement” in people’s apartments and with choice public/hallway areas to get the stragglers and keep the place mostly roach free.

    Now…as for the actual statistical content of this post – that was a really enjoyable read, and I think that the only thing I would add to the discussion is the idea that changing ones behavior in these kinds of things (i know, intentional roach v. unintentional human killing are so, so similar!) is a tacit admission that you’ve been doing it wrong for a while. So if you switch to boric acid, you’ve been complaining for a long time for no reason, and that makes you look stupid. If you’ve accidentally been killing hundreds upon hundreds of your human patients because you refused to acknowledge the value of some finding, that kinda makes you look guilty of reckless manslaughter. I think Andrew was getting at that when he mentioned people getting “defensive.” But I think the interesting part is why they get defensive, and I think it has something to do with the difficulty of admitting we have been consistently hurting someone (ourselves or others) for no good reason.

    • > a tacit admission that you’ve been doing it wrong for a while
      That was often my sense of why folks often get very upset when statistical methods are really being explained to them, they can’t help realizing that how they had been making sense of the world is/was seriously flawed.

  10. Very interesting. It also seems important that (roughly) a single dramatic case suffices to show that anesthesia works, but many cases and some rudimentary statistical reasoning are needed to show that antisepsis works.

    • The funny thing is that the hand-washing before assisting with childbirth idea was also pushed a few years earlier by Dr. Oliver Wendell Holmes Sr. (the father of the famous Supreme Court Justice). Unlike poor Semmelweiz, who had a difficult personality, the elder Holmes, who was also a beloved poet and humorist, was wildly popular. Wikipedia says:

      “Surrounded by Boston’s literary elite—which included friends such as Ralph Waldo Emerson, Henry Wadsworth Longfellow, and James Russell Lowell—Holmes made an indelible imprint on the literary world of the 19th century. ”

      But even Holmes didn’t get much traction with his call for doctors to wash their hands.

      http://en.wikipedia.org/wiki/Oliver_Wendell_Holmes,_Sr.#Medical_reformer.2C_marriage_and_family

      • Reading further, I find that Holmes invented the word “anesthesia:”

        “In 1846, Holmes coined the word “anesthesia”. In a letter to dentist William T. G. Morton, the first practitioner to publicly demonstrate the use of ether during surgery, he wrote: “Everybody wants to have a hand in a great discovery. All I will do is to give a hint or two as to names—or the name—to be applied to the state produced and the agent. The state should, I think, be called ‘Anaesthesia.’ This signifies insensibility—more particularly … to objects of touch.”[66] Holmes predicted his new term “will be repeated by the tongues of every civilized race of mankind.”[67]”

        So, Holmes’ most important invention — doctors should was their hands — didn’t accomplish much of anything, but his invention of the name for somebody else’s invention was an instant hit worldwide.

        That’s bizarre, but not atypical, either.

  11. Thanks to peer review, the same people using inferior statistical methods are the ones reviewing your manuscript, and so the perception is that such methods are the ones to use in order for the manuscript to be published. In my field, such people are not statisticians, but heavy users of statistics. This is a recipe for inertia. Not only do inappropriate statistical methods keep being used (e.g. automated stepwise regression), but inferior plots (e.g. “dynamite” plots).

  12. The idea that exterminators have an economic motive for deliberate inefficiency is old enough to have been used metaphorically in by Duff Cooper, in 1932:

    “Fouche, as Minister of Police, dealt with the Jacobins as professional rat destroyers are said to deal with their quarry, sacrificing a certain number, from time to time in order to satisfy their clients, but leaving a sufficient stock to render their further services necessary.”

    • It’s tough to track good ideas that have probabilistic effects. However, spectacularly *bad* ideas really do have a way of lingering. The following from Robert Edgerton* describes how the Mae Enga treated arrow wounds:

      > Acting on the Mae Enga belief that the wounded man’s blood was contaminated, a specialist used a bamboo knife to make an incision under the victim’s armpit and then broke a rib with a wedge so that he could insert two fingers into the thorax, collapsing a lung in the process. He next poured water into the cavity and shook the patient vigorously to mix the blood and water. Finally, he rolled the patient over to drain out the mixture. (p. 106)

      Not sure how accurate that report is, but there’s no shortage of clearly appalling medical atrocities. It’s a good point that people suffer from recalcitrant intellectual habits, and continue to suffer from unwarranted confidence in oversold statistical models. It’s not enough to point out the flaws. You need to change a culture.

      *
      [1] R. Edgerton. Sick Societies: Challenging The Myth of Primitive Harmony. Free Press, New York, 1992.
      cited in
      [2] K. Sterelny. Snafus : an evolutionary perspective. Biological Theory, 2(3):317–328, 2007.

  13. Is it not rather plausible that the reason Lister & Semmelweis & co. had such a hard time getting people to pay attention to them was that doctors were extremely hostile to the suggestion that people were dying because of pathogens that the doctors themselves were carrying. In the case of anesthesia, the doctors were being offered a way to control a harmful mechanism, internal to the patient, i.e. not really caused by the doctor. With hand washing and sterilization, people were trying to tell the doctors, ‘you are actually killing people because you are dirty,’ but perhaps the doctors of the time cared more about the unpleasantness of this implication than about whether or not it was true.

  14. I can’t help but think the anaesthesia/anestetic problem might find its roots in information assymetry. Patients, even more than doctors of the time, lack the knowledge of medicine and germs to demand that their doctor use sterile procedures, while many would have heard of the amazing painless surgeries and demanded it. I have no proof this is so, but it’s the first idea I would investigate.

  15. Hi Andrew,

    Great points! I think your theory is already right there in the mathematics of the diffusion of innovations, though. If we take Everett Roger’s old findings as a guide, and import some economics like you’re doing (looking at the relative benefits and costs to adopters), it would seem that the particular shape of the standard logistic function that describes tipping points will change relative to the costs adopters face — the amount of work they have to put in to adopt.

    I have to insist, as a sociology student, that the cultural context of belief *does* matter as well. You’re assuming that these doctors were thinking “I don’t have to know how the microwave works to heat up my burrito, and don’t care.” I doubt that’s true: however wrong the ideas about the origins of disease were, they were widespread and had their own locked-in system of (incorrect) logic. These beliefs, too, raise the cost of adoption of a belief, potentially dramatically, and compound the effects of “I have to work to adopt” vis a vis “I have to push a button to adopt.”

Comments are closed.