Why Doctors Are Not Scientists

Outgoing BMJ Editor Richard Smith explains why doctors are not scientists. A reader agrees, adding:

We doctors like people to think we know what we're talking about, and may be so convincing that we convince ourselves too. Because other people's lives depend on it, we have a big emotional need to be right and are uncomfortable with the thought that none of us really knows enough to be a good doctor….

Scientists, on the other hand, are very comfortable with the unknown; it is their bread and butter. When scientists disagree there is no more at stake than the scientists' amour propre, whereas medical disputes get rancorous because forever in the background is the thought that the other chap is damaging patients.

Science does not in itself make its practitioners haughty (the contrary, if done honestly), whereas medicine does. The main reason for that, I think, is because doctors get used to seeing other people undressed while they themselves are clothed. Once you have seen dukes and archbishops in their underpants they're never quite the same again.

[Thanks to Jeff Schaler for the links.]

Comments

[Brian]

I think the "purity" of one's science is determined by the methods used, not by the discipline. I know economists that develop falsifiable hypotheses and run controlled experiments and I know physicists who never set foot in anything that can properly be called a lab.

That said, neurobiology is the king of all science... 😉

On the arrogance of doctors...I once had occasion to be a patient in the emergency room of the hospital which also contains the lab in which I'm doing my PhD. The resident taking care of me asked what I did, and I told him. He asked me what my dissertation was about, and I told that I was working on the blood-brain barrier (a subject that usually merits part of an hour or so's lecture in the typical med school cirriculum.)

His response: "Wow. You might know more about that than I do."

I looked long and hard for a hint that he might be kidding, but I couldn't find it. He's lucky I was in a weakened condition.

[KentInDC]

Mo,
Are you aware that Mises once said something along the lines of it's a pity more economists don't know more about math or they would see how little math has to do with economics?

[Lucas Wiman]

The computer scientist I've had occasion to interact with (not an all-inclusive sample) were indistinguishable from mathematicians. However, it seems very clear to me that while mathematics is a useful tool in science and is very rigorous, it is not a science. As a mathematician, I perform no experiments, collect no data, etc. Sometimes I work through examples or argue by analogy to something else. But these are hypothesis-forming techniques. The method of certification of truth is radically different from experiment.

By the by, use of math also has little to do with whether your field is a science. You can apply math to any old crap, and the models will only be as correct as the assumptions.

[thoreau]

Mo-

I minored in economics as an undergrad. I liked it because I could take classes that combined math with current events. It was fun. Lately, while working on my Ph.D in physics, I've gotten interested in the theory of elections (a very rigorous area of mathematical social science, and that's not a joke, just look up "Arrow's Theorem" or my favorite "Gibbard-Satterthwaite Theorem"). On the side I'm working on a proof that basically says Approval Voting is more strategy-proof than any ranked voting method.

As to which science is the "purest", I used to get into those pissing contests as an undergrad. Now I see how pointless those pissing contests are. The simple fact is that you're either a careful investigator or you aren't, regardless of your field. Some fields may have more inherent limitations than others, but I'd rather hang out with a very careful psychologist rather than a sloppy physicist any day. As long as you appreciate and heed the limitations of your methodology and take care in drawing conclusions, you probably deserve the title of "scientist."

[Mo]

Kent,
That was kind of my point. Use of math in a field is worthless. I would hardly call accounting a science. I was the ex-science major at a science and engineering school trying to tweak the scientists in a nerd pissing contest. I had no illusions of economics as a science, though it is an interesting study of human behavior and choice theory. The problem a lot of people run into with economics is that they think it's a study of money when it is actually a study of people's choices.

I think it would do psychologists a lot of good to take some basic economics.

[joe]

On doctors' arrogance: consider a surgeon. He takes a knife, sticks it into people's vital organs, and screws around. If he cuts a little too much, or a little to lightly, or in slightly the wrong place, the patient dies. If he takes too long to finish, the patient dies. Freezes up: dead. He has to quickly, confidently slice a blade through another human being's body, and carry out a complicated procedure, on the clock.

I don't think this would be possible without superhuman arrogance.

[J. Goard]

Almost every linguist has some version of the anecdote where a different kind of scientist, professional, or just rather literate person wants to discuss language on an equal footing. Most people (and I don't exclude myself!) seem to readily conflate exposure or interest with expertise. In a society where a large percentage of survivors become self-anointed experts on their respective diseases, I'm hardly surprised that many doctors have trouble distinguishing between levels of expertise vis a vis human bodies as systems.

[Lucas Wiman]

Ooops. Should have said "computer Scientists" instead of "computer scientist."

[george]

"When scientists disagree there is no more at stake than the scientists' amour propre"
Not always. There's a story I read about the first A-bomb test: there were three theories about the result; one, that there would be a small explosion that would scatter radioactive matter across the desert; two, that there would be a big explosion (as, in fact, there was); three, that there would be an enormous explosion that would destroy all life on earth. # Three was given a low probability but it was considered possible.
Scientists and technicians often work with things that, wrongly applied, could kill or injure one, ten, or a million people. I was "only" a technician in my working days, maintaining an air navigation system, which, if miscalibrated, could have put a plane into the ocean. Doctors need not flatter themselves that they're the only ones with someone's life in their hands.

[Sandy]

Well a proof is not a double blind study--and I've been a subject of an HCA study for a dissertation where I chatted with the researcher while attempting the tasks (so not randomized or double-blind--possibly controlled, but I doubt it). So I might put you on equal footing with political scientists, but that's it. But then I'm son of physicist and don't believe mathematicians are scientists. Smart guys who are fiends with logic, but not scientists--it's a different job.

[thoreau]

I had no illusions of economics as a science, though it is an interesting study of human behavior and choice theory.

As a physicist who took a lot of economics classes, I more or less agree with this. Economics itself isn't so much a science as a body of knowledge that's very useful to keep in mind when making decisions. However, economic questions are often best studied with a scientific approach, and obviously math can be a useful tool for shedding light on economic questions.

So I'd say that a good economist can be considered a scientist if he uses a systematic and quantitative approach to gathering knowledge and testing theories, even if the body of knowledge isn't a science in the same sense as physics or chemistry.

[KentInDC]

Mo,
Sorry I missed your point. When I mention to others my interest in economics, their eyes usually glaze over. I tell them that real economics is more like psychology rather than meaningless formulae that often pass for economics. Mises felt that economists should have a broad knowledge of of science, math, geography, etc., but it was partly so that they would know when the bodies of knowledge were being misapplied to economics.

[J. Goard]

I frequently use (micro)economics as a point of comparison in my thinking about conversation dynamics. Aside from the possibly underappreciated connection between the phenomena, my primary reason is that both are best investigated with an analytic backbone, tweaked to account for features of psychological implementation. In either case, if we were to attempt a foundation strictly in terms of the "scientific method", we would a) probably never get past arguments about what counts as "observation", and b) never address these social dynamics in a way that remains useful where individual physiology/psychology changes or varies, or be able to model them productively in AI.

[Douglas Fletcher]

Personally I prefer a doctor who's willing to perform experiments.

'Say Doc, what happens when you stick a needle like that in someone's eye?'

[Sandy]

Doctors at least are trained in science and aspects of the scientific method (double-blind controlled and replicable studies), which is more than, say, computer "scientists" can say--which is why lots of "scientists" in the creation science/intelligent design movement are computer scientists rather than, say biologists.

Of course, training does not equal understanding: most political "scientists" receive some (brief) training in the scientific method, but ape it much like cargo cults in WWII: the forms and rhetoric are there, but the fundamental substance is missing.

[Jos Bleau]

Old, old joke but true none-the-less -

Q: What the difference between God and doctor?

A: God doesn't think He's a doctor.

[Jennifer]

Scientists spend most of their careers searching for new information, whereas doctors spend most of their careers making do with the information already available.

[ron hard]

Friend,

An English scientist seems to have discovered a cure for breast and prostrate cancer.

All it requires is that a person give up all dairy products.

If you would like the complete story e-mail me [e-mail].

Yours,

Ron Hard

[thoreau]

An old joke goes:

Biologists think they're biochemists.
Biochemists think they're chemists.
Chemists think they're physical chemists.
Physical chemists think they're physicists.
Physicists think they're God.
God thinks He's a mathematician.

Also, I swear to God I'm telling the truth when I say that one night in college I was actually in a situation where a physicist, an engineer, and a mathematician were trying to fix a vacuum cleaner. Strangely enough, no great joke emerged from the situation, however.

[J]

thoreau,

I've heard the joke slightly differently, but the gist is the same (and unfortunate, for biologists such as myself). But you forgot the one that usually goes at the top: social scientists (or psychologists) think they're biologists.

And how about this for a physicist joke: A farmer can't understand why his chicken has quit laying eggs, and after consulting numerous other farmers, veternarians, etc., he decides to talk to a physicist (it is the next logical choice, you know...). Without even looking at the chicken, the physicist locks himself away in his office for a few days, then comes out and tells the farmer he has the answer. He starts, "First, assume a spherical chicken...."

[pigwiggle]

Ron Hard-
Keep your pseudo-science holistic crap to yourself. When the study has been peer reviewed and published post a link to the pdf.
Thoreau-
I'm a Mathematician and a Physical Chemist, where does this put me in your scheme.

[Mark]

Of course, there's also the problem that shrinks(MD and PhD) think they are both doctors and scientists.

[Alison]

From Sandy--

Doctors at least are trained in science and aspects of the scientific method (double-blind controlled and replicable studies), which is more than, say, computer "scientists" can say--which is why lots of "scientists" in the creation science/intelligent design movement are computer scientists rather than, say biologists.

As computer scientist I take issue with that remark. Anyone with a Ph. D. in any science or engineering field MUST be trained in the scientific method and know how to conduct experiments. In the field of computer science, and especially within the subfield of Human-Computer Interaction, double-blind user studies are essential tools in R&D. A Turing test would be a famous example. Due to the mathematical nature of CS, a proof might be more appropriate than a double blind study, but please don't malign my field because of a few disreputable individuals.

[Mo]

I remember when I was in college, there was a huge debate over the "purest" science. It went in ascending order: bio, chem, physics and math. It was quite humorous to see the physicists and (especially) computer scientists calling biology a "psuedoscience." The compsci guys tried to claim they were real scientists because of the amount of math used. I said, if amount of math used is a measure of scientific purity, then I, the econ guy, am up with the comp sci guys.

I promptly got dirty looks for the rest of the night.

thoreau-
You'd appreciate an experience I had back in college. A bunch of my fraternity brothers and I had to change a windsheild wiper. I didn't participate and just watched as we played, "How many MIT students does it take to change a windshield wiper blade?" The answer is 7, one stabs his hand and 6 continue trying to fix it as the lady in the gas station fixes it in 10 seconds and proceeds to laugh at the group.

[jon b.]

Doctors are not scientists and you don?t want them to be. Doctors are tradesmen who take the latest information provided by scientists and apply it to a specific task. You don?t want a doctor who is testing the bounds of knowledge, and one that did so would not long be allowed to practice. You want a doctor who is using the best proven solution for your problem.

As far as using the scientific method, all medical professionals use it daily. When I taught Nursing in college we called it the Nursing Process: Gather data on the patient; formulate a hypothesis as to what is happening; administer the proper treatment; reassess the patient to see if the intervention is working; do something different based on the new knowledge. The scientific method is simply the most basic problem solving methodology.

If the idea that we don't really know what is going on bothers you then you fall into a group which includes most patients. They don't want to hear that their doctor is figuring out what is going on, they WANT to see him as omnipotent.

[Lucas Wiman]

"So I might put [mathematicians] on equal footing with political scientists, but that's it."

I don't understand this at all. How are mathematics and political science comparable? Political science is based entirrely on historical analysis, which mathematics doesn't use at all. Similarly, political scientists don't use proofs, and the kinds of things they theorize about are large groups of people, not abstract entities. Mathematics also has a whole lot more provably useful applications.