A Short History of Evidence-Based Medicine

Evidence-based medicine is the practice of grounding clinical decisions in the best available research evidence, weighed together with clinical judgment and the patient's own values. The term itself is young, popularized in the early 1990s, but the habit of mind behind it grew slowly over centuries. Medicine moved, unevenly and with plenty of resistance, from trusting authority and memorable stories toward testing claims in ways designed to reveal when we are fooling ourselves. This article is general education about the history of an idea, not medical advice; decisions about your own care belong with a clinician who knows your history.

I spend my working life inside this tradition. As a physician-scientist, I read studies as an appraiser and produce them as an author, including a systematic review and meta-analysis in Diabetes Care and genetic work on type 2 diabetes at the Lund University Diabetes Centre. The story below is one I find myself returning to, because knowing where these tools came from makes it easier to use them honestly.

Before evidence: authority and the single case

For most of medical history, the strongest argument was the oldest one. A treatment was right because a respected teacher said so, or because a physician remembered a patient who recovered after using it. Neither reason is worthless. Experience matters, and a striking case can point somewhere important. The trouble is that memory is a biased recorder. We remember the patient who got better and forget the ones who did not, and we credit the remedy rather than the illness that would have resolved on its own.

Bloodletting is the classic cautionary tale. It was standard for centuries, defended by the finest minds of each era, and it persisted precisely because there was no fair way to see that it was not helping. Authority and anecdote could sustain a harmful practice indefinitely, because nothing in the method was built to catch the error.

Early attempts at a fair comparison

The first move toward evidence was the comparison group. If you want to know whether a treatment works, you have to ask what would have happened without it, and that means comparing similar people who did and did not receive it. A famous eighteenth-century shipboard experiment on scurvy did roughly this, giving different sailors different remedies and noticing that citrus stood out. It was small and imperfect, but the logic was modern: divide, treat differently, and watch.

Through the nineteenth century, a few physicians began counting instead of merely recalling. Applying arithmetic to clinical outcomes was controversial, because it seemed to reduce the art of medicine to bookkeeping. Yet counting is what exposed treatments that felt effective but were not. The quarrel between the artful individual case and the cold aggregate has never fully ended, and both sides are partly right.

The randomized trial and clinical epidemiology

The decisive twentieth-century idea was randomization. Assigning treatment by chance, rather than by physician preference or patient characteristics, balances the groups on average across everything that might distort the result, including factors nobody thought to measure. A landmark mid-century trial of an antibiotic for tuberculosis is often cited as the model, not because it was the first comparison ever run, but because it made randomization a deliberate, describable method others could copy.

Around the same time, a discipline took shape that treated the study itself as an object worth analyzing: clinical epidemiology. Its contribution was a vocabulary for the ways a study can mislead, through chance, bias, and confounding, and a set of designs built to disarm each threat. This is the intellectual soil from which the phrase evidence-based medicine later grew. The movement that named itself in the 1990s did not invent these tools. It gave them a banner and argued, forcefully, that they should sit at the center of teaching and practice.

Synthesis: from single studies to systematic reviews

One study, however clean, is rarely the last word. By the late twentieth century a further insight had taken hold: the useful unit of evidence is often not a single trial but a careful summary of all the trials on a question. A systematic review is a structured, repeatable search for every relevant study, and a meta-analysis is the optional statistics layer that pools their results. Done honestly, this averages out the noise of individual studies and reveals a signal none of them could show alone.

An international collaboration devoted to producing such reviews became one of the movement's most visible institutions, and the practice of synthesizing evidence rather than cherry-picking it is now expected rather than exceptional. The catch, which I return to often in my own appraisal work, is that a synthesis inherits every flaw of the studies it pools. A precise summary of biased trials is a confident wrong answer, so the honesty of the search matters as much as the arithmetic.

What the movement got right, and still debates

The core achievement is hard to overstate. Evidence-based medicine gave clinicians a shared standard for asking whether a claim survives scrutiny, and a way to challenge a practice without having to out-rank the person defending it. That is a genuine democratization of judgment, and it has retired more than one comfortable but useless treatment.

The debates are real and worth keeping open. One is about balance: the original definition always paired research evidence with clinical expertise and patient values, yet the middle term can get squeezed until guidelines feel like recipes rather than tools for thought. Another is about generalizability, because a tidy trial in selected patients does not always describe the person in front of you. A third is about who sets the questions and funds the studies, since the evidence base can only answer what someone chose to investigate. None of these is a reason to return to authority and anecdote. They are reasons to practice the method more carefully, remembering that the point was never to replace judgment but to discipline it.