The Forerunners of EBM

The term “evidence-based medicine” first appeared in the medical literature in 1992. It quickly became popular and developed into a systematic enterprise. A book by Ulrich Tröhler To Improve the Evidence of Medicine: The 18th century British origins of a critical approach argues that its roots go back to the 1700s in Scotland and England. An e-mail correspondent recommended it to me. Can’t remember who, but I would like to thank him.

Francis Bacon (1561-1626) differentiated between “ordinary experience” (chance observations) and more objective “ordered experience” (methodological observations). Both of these involved empirical knowledge.  It’s hard to get back into the mindset of his time, when most physicians rejected empiricism as the sphere of quacks and surgeons.  Tröhler helps us understand why they did:

…since antiquity, the mark of distinction of a learned man had been the certainty of his knowledge. A doctor knew — he did not need to test his kind of knowledge empirically because this would imply acknowledgement of uncertainty.

Ancient wisdom, tradition, and dogmatism reigned supreme, in medicine as in religion. Rationalism based on accepted theory was sufficient: doctors could reason their way with no need for testing.

Surgeons were perhaps the first to empirically compare the effects of different treatments. In 1536, Paré tried a raw onion poultice on a burn case, on the advice of an old country woman, and found that the onion-treated areas of skin were free of the blisters seen on the conventionally treated areas of skin. He also found that treatment with boiling oil left wounded soldiers in agony, while an ointment containing turpentine did not. He wasn’t finding out what worked so much as finding out which remedies did the most harm.

The British Navy was an ideal place to experiment: the same doctor controlled the treatment of everyone on a ship and he had a captive group of subjects for months on end. James Lind’s famous scurvy experiment took place in 1747, but he didn’t invent his method out of the blue. He was influenced by other thinkers of the time.  It took 42 years for his discovery to be implemented, but Tröhler explains why it was logical that it took so long. No one had any idea of vitamins. Scurvy was envisioned to be a “putrid” disease. The theories of the time implicated the fermentative qualities of vegetables. Unfermented malt (“the wort”) was thought to have the same properties as citrus fruits, and some patients improved on wort, likely because it supplied missing B-vitamins that were the cause of some of the patients’ symptoms. Wort was cheaper and easier to supply than lemons. They continued to try other things, like sauerkraut, mustard, vinegar, sugar, and molasses. An early effort to supply ships with bottled lemon juice failed because it was boiled, destroying the vitamin C. When you try to re-create the thinking processes of the time, 42 years of resistance doesn’t look so shocking.

The story of fever is enlightening. They thought in terms of bad air; microbes were yet to be discovered. They lumped all fevers together into one diagnostic category. Cinchona bark is an effective treatment for malaria, but when they used it on other kinds of fevers, they naturally got varied results.

We laugh at bloodletting today, but the 4-humor theory was well established and generally believed, and early trials of bloodletting were equivocal: some showed benefit, others didn’t. It might have seemed a simple matter to test it and discard it, but the situation was much more complicated in practice. Those who favored it could actually point to convincing empirical evidence.  Battles raged for decades before it was abandoned.

Early controlled trials were flawed. They mostly compared two treatments, with no placebo or untreated group. Unrecognized factors affected results. The placebo-controlled, randomized trial wasn’t invented until well into the 20th century. But at least they were trying.

The idea of applying “arithmetic” to medicine was a new one. They were used to writing up a detailed case report of one successful treatment so that other doctors might read all the details and notice some common feature with their own patients. Reporting failures was not the custom: they didn’t see any point in reporting about something that didn’t work. Reporting numbers of successes and failures for a group of patients was a new idea that many found objectionable, because they favored individual evaluation and treatment.  Tröhler does a good job of describing their thought processes and the oh-so-gradual groping towards better ways of “improving the evidence of medicine.”

One of my personal heroes is William Withering, for his scientific rigor and sensible humility. He wrote his Account of the Foxglove in 1787. He had observed an apparent response to the folk remedy foxglove for dropsy (congestive heart failure) and he set out to test it systematically. He described every case in detail, including dosage, method of preparation and all the failures. With the striking results he got, all too many modern doctors would have contacted the media to boast that they had discovered an effective remedy. Withering limited himself to saying that if others confirmed his observations, the plant might become a useful part of the standard treatment of heart disease. Others did confirm, and the plant did become a useful part of standard treatment. The active ingredient was isolated and improved upon, and Digoxin is still widely used today. Interestingly, it has fallen out of favor for its original use in congestive heart failure: although it reduces hospitalizations, it has been shown not to reduce mortality and we have better drugs that do. It is still commonly used for certain heart arrhythmias, although it is not the drug of first choice. Medicine evolves in response to better evidence.

I was surprised to see how large a role military medicine played in testing treatments and accumulating numerical evidence for everything from amputations to syphilis. (No need for informed consent, of course.) Outstanding work was done by thoughtful doctors outside the mainstream establishment of prestigious upper-class physicians. They worked on their own initiative, without monetary reward, to put the received wisdom of medical authority to the test. We owe those pioneers a great debt of gratitude.

One thing that struck me was the recurrent theme of “inflammation” as the alleged cause of fevers and practically everything else. It was a simplistic explanation that was wrong, and it reminded me of today’s simplistic claims for anti-inflammatory diets and antioxidant supplements.

Tröhler’s book is no literary masterpiece and its recital of details becomes frankly boring at times, but it’s full of tidbits of interesting information for history buffs and of lucid explanations for those who want to understand the evolution in thinking processes that led, slowly but surely, to modern scientific medicine. The book is not easy to find. It was published in 2000 by the Royal College of Physicians of Edinburgh. It is out of print and not available in used bookstores. The only way to obtain it is through interlibrary loan. My local public library located a copy at a university library in Ohio and shipped it all the way to Washington State for me to read. Didn’t take long, and didn’t cost me a penny. I found the book enlightening, and I feel more charitable towards our predecessors than I once did. 

Posted in: Book & movie reviews, History

Leave a Comment (14) ↓

14 thoughts on “The Forerunners of EBM

  1. kathy says:

    A favourite author of mine called it “chronological snobbery”. He wasn’t talking about medicine or science … but the term applies pretty widely I guess. He meant the assumption that, because we know more and have better methods, we are cleverer than those who lived and worked in other centuries.

  2. windriven says:

    “Reporting numbers of successes and failures for a group of patients was a new idea that many found objectionable, ,because they favored individual evaluation and treatment.” Emphasis mine.

    The vehement resistance of some present day physicians to best practices schemes suggests this sentiment outlived the 18th century.

  3. Janet Camp says:

    Thanks, Dr. Hall, for the historical insights and also for mentioning the tireless efforts of public libraries to make information truly accessible to all.


    Anthropologists call it “ethnocentrism” and advocate for “cultural relativity”. It doesn’t only apply to the past, but can be seen whenever people are exposed to something “different”. Sadly, as reported here and in comments, some colleges are twisting these concepts to advocate for some very woo-ish concepts that are an embarrassment to the field. I am hopeful that this is limited to the sub-field of Medical Anthropology–a field that would definitely attract those prone to magical thinking. They have forgotten that the point is to respect other cultures and work with them, not to “go native”. The book Dr. Hall discusses here seems to discuss the difference between observation and testing, as well as the evolution of these ideas. It should be required reading in Medical Anthropology.

  4. Scott says:

    @ windriven:

    I was very powerfully struck by that same parallel myself. Many of the quotes in the recent “How do you feel about EBM” post are very much along that same theme, where one physician’s personal experience trumps robust data.

    Dr. Jay Gordon is of course a particularly prominent example of this.

  5. cervantes says:

    Oh ferchrissake:

    Vitamin drip infusions gain favor in Hollywood:

    “Taking vitamins orally might soon fall out of favor as more Americans opt to drip them straight into their veins.

    More exhausted people are heading to private clinics and getting hooked up to a vitamin I.V. for their energy-boosting cocktail infusion.

    The technique was fueled, in part, by a photo that pop star Rihanna tweeted last month with her arm hooked up to an intravenous drip. Turns out she was enjoying the so-called “party-girl drip.” . . .

    I’ll retire to bedlam.

  6. tuck says:

    “One thing that struck me was the recurrent theme of “inflammation” as the alleged cause of fevers and practically everything else. ”

    They’re still at it. Now that the cholesterol hypothesis for heart disease has fallen apart, we’re told that statins are effective because they’re anti-inflammatory.

    Inflammation is an effect, not a cause…

    “Are statins anti-inflammatory?”

  7. “An early effort to supply ships with bottled lemon juice failed because it was boiled, destroying the vitamin C.”

    Fun fact: it was James Lind himself who tried to bottle lemon juice! And it was a sustained effort, for the rest of his career. After his famous clinical trial, Lind never conducted another one. Instead, having “proved” that lemons worked in his original simple test, he became fixated on producing a mass produceable scurvy cure based on processed lemons (that he could sell to the Royal Navy) — and it never occurred to him to test it (or anything else) ever again. Fascinating.

    Highly recommended book: Scurvy: How a surgeon, a mariner, and a gentleman solved the greatest medical mystery of the age of sail, by Stephen Brown.

  8. Jan Willem Nienhuys says:

    One of the first, maybe THE first, randomized double blind trials dates from 1835. These are the so-called Nuremberg salt trials. The aim was to test the claim that kitchen salt diluted in a 1 : 1, 000, 000, 000, 000, 000, 000. 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000 ratio would produce remarkable ‘symptoms’ in healthy volunteers.

    I wonder whether this is mentioned in the book. Unfortunately this trial had a negligible impact on homeopathic belief.

  9. Harriet Hall says:

    The Nuremberg salt trials were not mentioned in the book, but then it was about real medicine, not about homeopathy or other superstitions. As you say, the salt trials had no impact on the development of the scientific method.

  10. pmoran says:

    Good Heavens! I find it astonishing that this Nuremberg group was able to see the logic of that kind of study in 1835 yet it was over a hundred years before placebo controls began to be used at all regularly in mainstream clinical studies and the first ever randomized study was not performed until 1898.

    Why so slow? I suppose its the same old problem. When misled by the powerful illusions of everyday medical practice, it is difficult for doctors to conceive that they might be wrong, let alone to take the risk of proving that in such a test.

    I think this would have delayed progress even when science had largely dragged itself out its earlier reliance upon navel-gazing.

  11. Lytrigian says:


    – Relying on the certainty of received wisdom instead of empirical evidence.
    – “Individual evaluation and treatment” as opposed to understanding of harm vs. benefit based on large-scale trials.
    – No use of adequate controls, even when trials are conducted.
    – Overly-broad diagnostic categories, each with its own “one true cause”.
    – No attempt to isolate active agents in apparently effective treatments.

    This all sounds very familiar. Perhaps it suggests a different tack to take vs. CAM. Highlight the horrors of pre-modern medicine, and then point out how CAM practitioners are doing the same thing almost word-for-word. I bet it’d even be possible to find modern CAMish statements from some of the more alarming practitioners of the time, who may have employed very scary-sounding treatments.

    The sort of emotional hooks sCAMmers freely employ, and which their public can find so convincing, are hard to come by for honest practitioners of EBM and SBM who want to counter them. Perhaps a few can be found in this context.

  12. kathy says:

    Janet: I don’t think chronological snobbery is the same as ethnocentrism. It refers specifically to the assumption (presumption?) of the people of any century that they are automatically better/more intelligent/whatever, than the people of earlier days. That is the opposite of what Dr Hall was describing, the excessive veneration of those who were authorities in some long-ago ‘golden age’. Both are an error in any scientific discipline, and indeed in any field that needs objectivity. (Just btw, the author I spoke of was referring to literary criticism).

    Lytrigian: would it be a good idea to use the methods of CAM to scare people back into the SBM fold? Wouldn’t that destroy the only real way of progressing, which is to motivate them to think for themselves? I speak as someone convinced of the superiority of SBM. But it seems to me that, above and beyong SBM, we stand for certain values in society: the right – and responsibility – of every individual capable of doing so, to think carefully before deciding what to believe, whether in medicine, science, politics or religion.

  13. Lytrigian says:

    @Kathy — I believe that people “thinking for themselves” is part of the problem. CAM users think they already are thinking for themselves and making decisions based on the best information available.

    You cannot simply make the facts available and expect people to make use of them. That’s just not realistic; it’s not how people think. It would be one thing if SCIENCE had the power to filter the information people receive, so that we can be assured that they only hear the truth, but the sad fact is that woo-vendors of every subspecies can say pretty much whatever they damn well please. I’d bet you that close to 100% of acupuncture users think there’s solid evidence behind claims of that treatment’s effectiveness, for instance. At the same time they have mental filters in place, cognitive biases probably erected via the sort of emotional appeal that almost always trumps cold fact, which shut out the genuine scientific message.

    Cognitive biases are very real, and to get your message across you have to take them into account.

    I would disapprove of this kind of tactic in general, but when it’s actually the truth — when it can be shown that the modalities and techniques of CAM really are throwbacks to the bad old days of bloodletting and witchcraft — and if that truth, presented in such a way to carry emotional impact, can cause people to alter their biases in favor of a clearer view of the facts, then I’d have to think that a good thing.

    As to your reply to Janet, I think ethnocentrism is exactly right. It has NOT been true of every century, but is governed by the culture of the time. Medicine is the perfect example: for centuries, professional medicine in the West was governed by Galen, who was deferred to because 1) to know him was to be learned, and to be learned was to be certain; and 2) his antiquity lent him authority; and 3) he was approved by the Church. Not just in medicine, but in every branch of knowledge, for medieval thinkers a question was decided not by the latest experiments but by appeal to ancient authorities regardless of the subject. It was the techniques of theology applied to the physical word. Only since the Renaissance has there been even a hint that progress was a good thing in and of itself, that recent empirical information is more likely to be useful than ancient opinion; that reason had to be controlled by experiment. It took centuries *after* the Renaissance for this mode of thought to become the norm. (Galen was considered a useful authority until well into the 19th century, hundreds of years after his anatomical studies had been shown to be wrong.)

  14. Scott says:

    It took centuries *after* the Renaissance for this mode of thought to become the norm.

    It can be credibly argued that it still hasn’t truly become the norm. Far more people pay lip service to it, than actually adhere to it.

Comments are closed.