Why is this an argument for osteopathic medicine, rather than an argument that the MD pipeline, cartelized by the AMA, should be expanded significantly?
I have nothing against DO’s, and I’d be happy to use one. But this article praises them primarily because they make up for the shortage of MDs, and because they’re roughly as effective as MDs. But the shortage of MDs is artificial, imposed by the cartel.
So, why not just expand the MD program so that the kids who currently get DOs get MDs instead? (MDs are also roughly as effective as MDs: this has been proven in extensive clinical trials). “We need more doctors, and DOs are more or less doctors,” is not a defense of osteopathy, just an indictment of the cartel.
A genuine defense of osteopathy on its own terms is a secondary theme in this article — maybe they do better with chronic pain and hypertension. But that defense is fairly weak: it’s not clear that the improvements are that large, that they are the result of anything distinctive about osteopathy, or that they cannot be swiftly imitated by the MD system.
So: hurray for more docs! Let’s train more docs! Let’s not worry too much about whether they are MDs or DOs!
All good points. But a reminder that the AMA intentionally produces fewer docs than our country needs.
I appreciate the piece. This is in the interest of combating some of the hostility in the comments but still providing feedback:
1. I’d be okay with more explainer type pieces on undercover topics, especially from writers other than Matt. On that goal, the piece really succeeded and I enjoyed it
2. The criticism of the studies you cite is substantive. It would be interesting to write a piece on the journalistic task of evaluating technical literature. What is that like and how can it be helped?
3. Recognizing that SB’s template is to mix an explainer with a policy recommendation, writers will be judged on the weaker of these. If you don’t have a cogent policy take, it’s okay to just say “I don’t know what this means going forward,” as Matt sometimes does. What I’d prefer is to turn the challenge into a policy statement. “The relative effectiveness of DOs is poorly studied and here are some hypotheses why. Here are the key questions I’d love to see answered in the future…”
I would have liked this article much more if it had simply presented osteopathy as an interesting topic, rather than recommending them. That part felt a bit like standard journalism based on shaky science, not up to the rigor of a typical slow boring piece.
But is it a growing medical field? Or is it just a label that is applied to certain medical schools based on an increasingly unimportant historical division.
I mean, if I started talking about and comparing academics educated at schools like Cambridge and Oxford that descend from an originally religious mission to schools that descend from a secular foundation am I actually discussing a academic split.
Or, as seems more likely, am I just seeing the lingering traces that the originally religious institutions are older and therefore tend to be higher prestige. Sure, Oxford and Cambridge have tradictions that UCL lacks just as DOs do but I doubt that it's more than peripheral/emotional (ppl grow attached to their educational traditions)?
Yeah, it’s good to be clear about what actually happened: osteopathy didn’t turn out to be good, its quack founder wasn’t vindicated, what happened is that osteopaths gradually shifted to mostly using real medicine and only teaching a bit of pseudoscience. Which is only beneficial in the context of a system where “just train more MDs” isn’t allowed.
Editing to acknowledge that, as people have pointed out, the restrictions on the doctor pipeline are more complicated than just the number of MDs trained. Does allowing DO schools actually increase the number of doctors that come out at the end?
I do think there is a perennial and difficult question here, though. Insofar as many people's health problems are psychosomatic, which is probably responsible for the vast bulk of the sincere experiences of pseudoscience improving people's subjective experience of their own health, there is some utility in a person who can give you real medicine for a real problem and also be willing to give you some made up stuff otherwise in case it helps resolve whatever psychological hangup is causing your symptoms. Dividing those two roles into pure allopathic medicine vs. pure quack practitioner has drawbacks -- former is unwilling to provide treatment that can relieve symptoms, which pushes patients toward the latter, who then might unscrupulously continue to prescribe fake medicine even when problems with physical causes emerge.
I thought the number of doctors was limited by the number of residency slots, which in turn was limited by Congress? Though I also thought there was a recent increase in that cap.
I am very curious about this. I don't know who limits the slots. I live a few blocks from a teaching hospital. What I have heard from them, especially since the pandemic and subsequent staffing shortages, is that they don't have the capacity to take on more residents, even if "they" (whoever that is) made more available
There are two pools, medical students and residents. There are currently more residency slots than US medical school graduates, so all graduates of US medical schools can get a residency. The remaining residency slots are filled by DO grads, American graduates of foreign medical schools, and foreign graduates of foreign schools. Foreign medical schools which include both US and foreign students supply 25% of US residents, and DO schools about 10%.
It is much less costly to fund a residency slot than a medical school slot in fact many hospitals need residents to get their work done, where one study at UVA calculated it cost $100,000 a year to educate a medical student, and that is at an established school, so likely to start a new school would be significantly more. Now the average tuition runs about $60,000 varying a wide degree with public vs private, so tuition likely pays a little more than half the cost.
I understood how the process works, but I am quite surprised to hear that there are more residency slots than US med school residents because of our local situation. I didn't think our hospital would be highly coveted. But your comment made me look at reviews, and now I see that it is popular for reasons I hadn't thought of - cheaper cost of living than big cities, but near enough to big cities that there is an overflow of talent; more supportive of residents in general, especially residents with families, than a lot of places; wide variety of patients since it is regional; better support than many places for urban, low-income patients.
The number of residency spots is determined by the Centers for Medicare & Medicaid Services at the federal level. They do limit it, but it is still much more than the number of medical school grads which I think is determined by the states or private universities. It can be hard to get into competitive residencies whether due to the specialty - orthopedics, surgery, etc. can be very competitive or whether the residency is affiliated with a prestigious school, but there are plenty of Family Practice and Internal Medicine programs that dont match their positions and then have to go outside the match to fill slots mostly with foreign grads.
Exactly. The AMA does not set the number of medical school positions. State legislatures and private entities determine the number of medical school slots and fund them. If NY or MT want to increase their medical school positions they can. The AMA is involved in accreditation. From my experience the pool of osteopathic students, just like the off shore Caribbean schools, is composed of people failing to get into medical schools.,
FWIW, if NY or MT increase their number of slots, the extra slots will be filled by people who otherwise would have failed to get into medical school.
Because the number of onshore MDs is artificially limited to less than people who could do the job, talking about people who failed to get into medical school is sort of like talking about people who "failed" to make the Olympic Swim Team or "failed" and less like talking about people who failed to get a drivers license or failed to finish college.
I have no problem with extending whatever standards you want lower to accompany more people going to medical schools in order to fill all the available residency positions. I don't think it would be a great hardship if the standard of gpa which now might be about 3.6 was lowered to 3.5, not a big deal, but among all those that fail to get into medical school, not everybody goes the DO route so that lowers the competitive level for DO schools.
I dont know what standards you suggest for your doctors, but given a choice between a higher and lower achiever, I would take the higher. I have nothing against NPs and PAs either, but at least that is more explicit and they are not called physicians.
"but given a choice between a higher and lower achiever"
It's very difficult to find a doctor, especially a primary care physician. My dad lives where he does now primarily because he can access doctors more easily than he could where he would otherwise like to live.
The problem is there's not enough choice, and it's frustrating to hear talk of about "standards" when they boil down primarily to GPAs. I'd almost agree that I'd prefer a doctor who had a 4.0 GPA to a doctor with a 3.5, but the truth is I'd evaluate on them o a dozen other different metrics first, and I've never heard of anyone asking a doctor how their college performance was.
So yes, give me a doctor nearby with a 3.0 who has time to treat me over a 4.0 GPA doctor who lives an hour away and is fully booked for the next 3 months. I know of no other profession that gets so hung up on this sort of gatekeeping.
Wigan, I sat on the admissions board of my medical school which had two students. We did consider things other than gpa and MCAT, but it can be a little fraught as prospective students all knew they had to volunteer with the red cross or migrant workers or what not. I actually likely benefitted from this because I worked overseas in relief work before going to Med School. But most resumes look similar and yes, grades hardly mean everything, but they are fairly objective say compared to an essay which students might have been coached on.
Ideally you want humane physicians, and if there was an easy way to measure that it would be great, but given two equal MDs, Ill take the one with the 4.0 over the 3.0 because at least things like Internal Medicine involve very challenging mental constructs for which I want the smarter doctor.
As to the shortages, it depends where you are. I have never had a problem finding a doctor, but in some rural areas, yes. I do think we should train more doctors and that doctors should spend more time with their patients, but a lot of that is determined by money. When I worked for a big employer we were graded on productivity, which you achieved by seeing the most amount of people in the least time, and charging the most. I think medicine should be divorced from the market place.
"I dont know what standards you suggest for your doctors, but given a choice between a higher and lower achiever, I would take the higher."
"The average medical school admission has a 3.72 gpa and osteopathic 3.54."
I don't have any strong opinions on osteopathy, but I think assessing whether someone will be a better doctor because they have 0.18 higher GPA in undergraduate work is a choice I decline to make. Give me some real evidence that MD's perform better than DO's and I might buy in, but as best I can tell there is no substantive difference in their patient results.
It would be very difficult to devise any study to determine which provide better care as that is so multifaceted. The MDs score higher on tests to get in so I would presume afterword, they would also. Tests mean something or you wouldn't have to pass your boards to practice medicine. They don't mean everything, and I was on the admissions board of a medical school, and we also look at other things like humanitarian concerns and hard work from people from challenged backgrounds, and as much as you can assess things like empathy and integrity, they play a role.
If gpa and MCATs means nothing and we should do a lottery and B- students would have as much chance as an A student. But medicine is very complex, and I would rather put my trust in an MD who was a higher achiever. There is a correlation between grades, test scores, and intelligence and hard work. We put our lives into the hands of MDs and I myself would want them to be very hard working and intelligent.
If you're asking me whether I want someone who had a 3.5 gpa in their undergraduate or a 2.5, then yes I'll take a 3.5. But I don't think the difference between a 3.72 and 3.54 in UNDERGRADUATE work provides meaningful information about whether someone is more hard working and intelligent as a doctor.
To be fair to your point, the AMA is one of the most powerful interest groups and the state congresses are certainly being driven by the AMA, not the other way around.
I spent several years in the anesthesiology device market and the American Society of Anesthesiologists two primary activities are limiting competition and containing malpractice risk of its members.
Agree, simply fund more medical school positions. Being that the DO schools are already there with faculty, doing the same thing, just incorporate them into the Medical system. I dont know other professions that have two tracks, so if you cant get into law school you would go to a lawgo school where you would learn the same thing and then do the same work. The bottleneck is that it costs a lot of money.
In the old days (30 years ago, when I was last paying attention), in California, there were three tracks to getting admitted to the bar. Three years in an accredited law school, or four years in an unaccredited law school, or five years of working in a lawyer's office learning the trade. Then you had to pass the CA bar exam, which is notoriously one of the more difficult ones. And then of course you had to get hired. Wasn't Kim Kardashian trying to become a lawyer via the third route? I think she went back to being ultra rich instead.
People still like to blame the AMA for their '90s positions, not their 2024 positions. It's annoying, but I don't really care about the AMA enough to defend them.
I think the median DO school reputation is still seen as generally below the median allopathic school. But now I think grads a mid-tier medical school and top quartile (or maybe half) good DO school are seen as approximately equivalent. Essentially, the bias is strongest at elite programs because they're picking between Harvard and Yale grads. It levels out the further down you go because elite candidates rarely trickle down that far so it becomes more individualized.
Caribbean medical schools are still seen as inferior and are heavily biased against. I'm not sure how they fare against other foreign medical schools since I can't recall any program I've been affiliated with matching any (aside from fellowship after completing a US residency).
I've heard an MD say that the MD and DO programs are overlapping bell curves. At the far ends, the best MDs are better than the best DOs, and the worst DOs are worse than the worst MDs. But the huge middle area was full of perfectly capable future physicians.
In many, or maybe most fields, a somewhat lower end performer is still useful. I'm sure the worst lawyer in town would not be my first option, but if everyone else is booked he / she may be a lot better than nothing.
And this would likely be especially true in a field with artificial supply constraints.
"Are the worst DOs not fully capable?" I mean...no? The worst MDs aren't fully capable either. Physicians do lose their licenses sometimes, and there are others who *should* lose their licenses but escape that fate. This is some deep Internet/Usenet dark lore, but I remember the Usenet diet groups being overrrun by one Andrew Chung, MD, who had 100% clearly lost his everloving mind and was, eventually, after many years, stripped of his license. Those guys do exist. Google "Dr. Andrew Chung" and "2 pound diet" if you're bored.
Haha yikes. I believe you. I've heard a few stories of off-their-rocker doctors, too. I think several of my ex-girlfriends being doctors also makes it easier for me to believe that doctors are not always hyper-rational, unbiased, 100%-knowledgable people, lol.
I was going to write a separate comment but yours hits the gist of it first, so I’ll add mine in support.
My understanding of DOs is that the school is easier to get into so people who really want to be doctors go there as a second choice. My initial reaction was to think that I should shy away from them, but if they take the same test and have the same outcomes, why have the difference without a distinction at all?
At one time they actually were different, more akin to chiropractic, and still incorporate some of the spinal manipulation, but people go their mostly just to become regular physicians for not getting into medical schools. Graduates with DOs and MDs go to the same residencies. As the article mentioned DOs tend to go more into primary care and I suspect that is just because they are the less competitive residencies, not that they are not needed, I am an FP myself.
You cite some evidence in favor of osteopathic medicine working. I have had the misfortune of reading them. I will be very blunt.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6134408/ — “simulated rct” on observational data. So an after the fact fishing expedition, with no attempt whatsoever for sorting on perceived severity. Junk, toss it.
I would bet you, at even odds, that this paper is fraudulent. Look at figure 2, page 7 — what is the theory here for why it should have a preposterously large effect on migraines? Does it not boggle imagination? You have found, not treatment but a panacea. So no, it’s probably horseshit.
They find no difference between dummy light touch treatment, and OMT as well. First decent set of methods so far.
Chiropractic medicine does not hold up — why does this?
You need to be more skeptical of the academic literature. It is filled with liars and fools — especially in medicine. If it’s not an RCT, it’s probably junk, and even if it is. You need to be more aware of sorting by unobservables. This article is far too credulous, and does not bring credit to the byline.
Hey listen OMT is a hard thing to conduct a randomized double blinded study on. But there are reputable organizations that clearly think it works and recommend it as a form of treatment. Below is a Cleveland Clinic explainer on it.
This also isn't just a complaint about blinding. You just absolutely cannot use a word like "proven" when linking to an observational study that made no attempt to correct for differing underlying patient characteristics.
Also, concerning a different citation, if I read a sentence like this in a paper it doesn't exactly suggest I should rely on it for treatment recommendations:
> A total of N = 31 out of N = 63 eligible subjects followed by a single cardiologist received osteopathic treatment in addition to routine care.
It's not "bias" to favor actual sciences in this instance.
You rely heavily on the performance of osteopaths vs medical doctors in the opioid crisis. The medical field obviously blew that one. But osteopaths didn't do better because they had better insights; they just are skeptical of all medical advancements and best practices. Consequently, the medical field went off the rails in this instance, they did not go with it.
Good for them - seriously; that's a lot of lives thankfully not ruined. But relying on this to elevate osteopaths is like praising the Catholic Church for its superior family planning approach because their adherents did not use thalidomide. It was right in that instance but not for the reasons its advocates (and this article) claim and ultimately does not indicate any particular insight that will sustain in the future.
It also does not indicate that residency programs are wrong or "biased" to favor medical doctors over osteopathy (or actual OB's over Catholic doctrine...)
That’s a generalization that doesn’t address the specific deficiencies of these particular studies. The hypertension study is biased because the DO is selecting which patients get OMT; that’s a huge no-no. The pneumonia study showed no difference on an Intentional To Treat (ITT) basis, which is the unbiased framework. Pointing to a per protocol (PP) advantage for OMT could, again, be a reflection of biased selection by only choosing patients who are likely to have better prognosis to actually get OMT.
But what I wrote is an accurate description of the short hand Nicholas leaned on. The biggest problem with medical studies is that people don't accurately describe what they are measuring.
Economists by now focus on "natural experiments" precicely because they realised that purely observational data is very bad at estimating effect sizes or even their direction.
Yeah, but ... natural experiments are purely observational data, just with some extra math to simulate a controlled experiment. They're vulnerable to all the same threats as a straight correlational study, plus the extra degrees of freedom given to the researchers. I don't think that the field is as enamored of natural experiments as you might suggest. A brief review of the 5 latest articles in QJE has three _controlled_ experiments, a correlational study, and an analytical model.
I didn't read the paper that Mr Decker linked, but "simulated RCT" is probably medical jargon for what social scientists call natural experiments.
That’s not true. Natural experiments are specific situations where there is an unintentional random variation that can be exploited as if were an intentional experiment. For example, when a charter school is oversubscribed, it would often randomly choose the kids to be admitted. You can then compare outcomes between those kids and kids who applied but were not admitted to estimate the impact of attending the charter school as if this were an RCT (whereas comparing the results of kids in the school to kids who did not apply to the school is misleading, because the parents sending their kids to charter schools are plausibly more committed to their education, so their kids would do better anyway)
I have found many laughable cases where economists declare something a “natural experiment” like school entry dates and timing of births.
It is particularly bad among Harvard grads who don’t have a handle on history or the data generating process but who have the egos to assert their source of variation is plausibly exogenous. (So many rainfall IVs or temperature “shock” claims.)
Wait, how is birth timing _not_ a natural experiment? Whenever you have an arbitrary cutoff, there will be people who are similar except for falling very slightly on one side vs. the other. If a school has a strict age cutoff, comparing students born one week prior to the cutoff vs. one week after is a plausible natural experiment, no?
Rich people strategically plan out when they are going to have children, usually during the summer. Poor people don't strategically plan fertility. There is a whole literature on debunking the Angrist and Krueger birth timing IV.
When you use birth timing, you are both picking up a selection effect and family income effect.
That's pretty interesting, I didn't know that but it makes sense.
In that case, couldn't you still have a natural experiment if you selected a small enough time window? Say, comparing children born two weeks before the cutoff vs. two weeks after? I wouldn't expect people to be able to have fine-grained enough control over birth timing where those populations would be meaningfully different. I do know some moms who deliberately scheduled c-sections, but I'd expect it would need to be close enough to the "natural" birth date for a doctor to go along with it.
(Not trying to pose a gotcha question, you just seem to know a lot about this topic and I'm genuinely curious)
I actually disagree with Mr. Quail, provided you can show two things. One, the discontinuity must be tight around the discontinuity. Two, you need to show that it genuinely is a hard cutoff, and people are not slipping from one side to another. We should be cautious, but it is not a priori a bad idea. You just need to work harder to show it.
Season of birth a la Angrist and Krueger is waaaaaaayyyy too big a time period.
Most things that aren't RCTs are pretty worthless. The level of rigor needed to draft a usable observational study is enormous and most academics fail to clear it, as the vast majority of effects hypothesized by observational studies fail RCTs with flying colors.
And yet many many many organizations use survey and observational data. You just stated that forecasting is worthless. I hope you realize how silly that sounds.
Not all questions are causal or need to identify causation.
Many organizations engage in many practices, some of which are good and some of which are not. A lot of those organizations include people invoking low-quality observational studies to advance their goals whether the results are true or not.
Questions like "do people prefer Coke or Pepsi" and "do people like eating dirt" are certainly accessible via survey, but for almost all actual policy proposals, the questions being investigated is casual: would implementing policy X have desired effect Y?
As another older economist, observational data can indeed sometimes be useful when an RCT is impossible, at least when you have many millions of clean observations. However, non-huge observational datasets are nearly always useless.
I can tell you this statement is also wrong. Small high quality datasets can help confirm or reject formal models.
Paul Rhode and Alan Olmsted disproved the whole ratcheting quota's model of cotton picking that historians loved by going back to a series of plantation records where they recorded daily production by worker. (Basically economic historians have shown that much of the 1619 Project is not substantiated in actual historical data or records.)
My favorite quasi conspiracy theory -- is that Linus Pauling ruined is Nobel winning reputation to push Vitamin C as a treatment *knowing* it was total bogus pharmacologically but that it activated the placebo effect and that's actually *real*.
EDIT: For those not familiar, Linus Pauling the only person to ever win two unshared Nobel prizes. Just makes no sense he didn't know exactly what was going on with Vitamin C -- which this Vox article completely misses as a possibility:
I agree that the literature on DOs is biased and probably not worth much, but I think you are focusing on the wrong thing. If there was a massive disparity between DOs and MDs, it would show up in the data "at a glance" and you wouldn't need complicated statistics to demonstrate DOs are bad - the MDs would have done that for you.
"Chiropractic medicine does not hold up — why does this?" Because MDs and DOs are similarly regulated, similarly treated by hospitals and patients, and held to approximately the same standards (based on their admission requirements). Therefore, I would expect the outcomes of MDs and DOs to be approximately similar, and that's what we see. Chiropractic is much less regulated and easier to get into, so I would expect there to be much higher variance in outcomes, and indeed that's what that field looks like (mostly useless, a small tail of terrible fraud and abuse, but lots of people believing that it does good for them). But I guess I'm coming at it from the idea that lots of allopathic medicine isn't terribly effective anyway; it's just that a few things like surgery and modern medicine have gotten so good that they make a massive improvement is patient outcomes.
The standards MDs are held to, in terms of hoops to jump through to get the credential, are far too high; the standards chiropractors are held to are far too low. Therefore, adding a marginal chiropractor will likely make the world slightly worse, but the marginal doctor (whether they be MD or DO) will make the world better.
That wasn't the only claim made though; these studies were brought up to support the idea that specific osteopathic practices have benefits, not just that DOs are similar to MDs overall (because, as you note, they mostly just do the same things).
No idea why the third paper included a sham group if they're not going to evaluate the treatment against it. But that one seems to have a pretty large effect size as well, placebo for the win.
Also -- have you noticed how generic placebos just don't work as well? I know, I know -- they're supposed to be bio-equivalent, made to the same standards, with the same inactive ingredients. Maybe they work the same, for some people.
But me? I only use Placebo™ brand placebos. Accept no substitutes.
The podcast "The Studies show" did an episode on the placebo effect. I think the issue is that a lot of studies don't include a "do-nothing" group and a lot of people just get better over time. A lot of the placebo effect can be attributed to regression to the mean.
Medical librarian here. I encourage people trying to evaluate this kind of evidence to look for systematic reviews and meta-analyses of individual studies, since it is a big time-saver especially when multiple trials have been conducted. In particular, look for Cochrane reviews. Here are a few:
I agree, though I am still skeptical of even meta-studies -- garbage in, garbage out. I'm reasonably confident the Italian study mentioned above is just actually fraudulent, but it would go right into a meta-analysis.
Oh garbage in garbage out for sure. If the meta-analysis was conducted properly (and for sure a lot of them aren't), the reviewers would have included a risk of bias assessment for each included study and incorporated that into the analysis, as well as checking for any retractions. You see that all the time in the discussion/conclusions of SRs - "most of the included trials were of low methodological quality so the evidence isn't strong" etc etc. The medical literature is just so flooded with garbage these days, not to mention how most researchers don't make their data available for peer review (although that's gradually starting to change).
I am in the medical training system. We have DO's in prominent places at my institutions and I have hired DO's. But there is nothing in the article specific to DO's that is relevant about implications for health care today. Insofar as they have same training as MD's they are fine for the things that MD's do. Insofar as they do things that some MD's should or could do, that's fine too but also not specific to DO's. The more typical Slow Boring take would be to emphasize role of mid-level practitioners (PA's, APRN's) both in rural and primary care and as physician extenders in specialty practices. Those are much cheaper and easier to produce, support the "abundance mindset" and challenge the status quo in more fundamental ways.
Think about why. Mid levels are encroaching on physicians’ authority. Doctors hate competition.
I’m not saying that there aren’t legitimate concerns about this. I observed conscious sedation procedures in Utah where RNs are allowed to deliver propofol and it was terrifying. But the root of MDs perspective is to shield their prerogatives and incomes
I should have been more precise. In Utah, nurses manage propofol — dose and administer without any real supervision by an anesthesiologist or CRNA. It’s technically done under the prescription authority of an MD, but the RN is making all dosing decisions. It’s equivalent to how CRNAs are empowered in many other states. That’s different from a nurse’s standard task of administering medications.
I’ve been in many ICUs, but have never observed propofol delivery, so educate me if it’s happening there also. That said, the amount of monitoring and availability of personnel to intervene is much higher in an ICU than in a GI suite where most of Utah’s RN-managed sedation occurs. I had nurses tell me that they were terrified doing it. If a surgeon/gastroenterologist pushed the nurse to over sedate the patient, the nurse would have a harder time pushing back than an anesthesiologist would and much less able to perform airway rescue if the patient stopped spontaneous respiration.
There is usually no issue if the mid-level is working for an MD. Midwives do very well on their own but that is due at least in part because they are trained on how to refer complex cases quickly and safely. The issue becomes problematic when it comes to independent practices. If you worked hard (years, no sleep, less pay than others in your cohort) to get a license that you were told was needed to practice safely and then others with much less training got the same privileges, you would be pretty mad too. Not saying there isn't potentially a public interest in ignoring that, but it is the real lived experience of physicians. Of course, there is more to the story and, if done correctly, could be a good SB post.
From my experience almost all DOs roll their eyes at the OMT stuff and never think about it after medical school, nor do they encounter it in their residency programs unless it’s one of the small OMT/Family Medicine programs. The DOs I know call it osteopathic massage therapy.
If you attend the OMT specific conferences, you’ll find a lot of quackery represented in the vendor and sponsor booths. Homeopathy, herbal medicine, crystals… The presentations aren’t so bad but they need to clean house and that they have not tells me that the OMT component of DO education is vestigial woo woo.
That said, one thing that I worry about in medicine today is NDs - Naturopathic Doctors. N and M easily sound the same and they claim to have gone to medical school but they are not following the same curriculum as MDs and DOs. They also don’t attend the same residency programs. Despite these differences, some states license NDs as medical providers. I feel like their patients could potentially be misled into thinking they have the same training and capabilities as doctors. Perhaps they will follow the same path as DOs and one day integrate themselves into the medical field at large.
There's always someone to pick up the woo woo for the woo-woo-craving masses.
I think the OMT convention issue is a similar one to activist movements. Once the normies move mainstream, it's mostly weirdos left behind. I'm sympathetic to complimentary medicine (in that some actually works and gets adopted) but naturopathy is not even in the same quality ballpark as DO or MDs.
My family saw DO family med docs back in the '90s (just because they took our HMO and were nearby) and they never did OMT.
I’m always wondering about massage and yoga and things like that. They really do seem to have lasting benefits for my body and how I feel , but no one is ever able to do controlled studies that show there’s something real. I suspect that a lot of it is that bodies are really hard to measure systematically in any but crude ways, though many traditions have found things that do work to some extent, even if we haven’t figured out how to quantify the endpoints. (There’s a lot of crudeness that gets tolerated by medical practice just to make it work statistically - Moderna is probably giving us huge overdoses of their mRNA because they used a big dose in their tests to ensure that it worked, and there’s no incentive to cut the dose to get the side effect size down to Pfizer, or even a regular flu shot.)
I mean ... but this is my problem with "controlled studies". We don't need a controlled study to know building a stronger, leaner, more flexible, higher VO2 max, etc. body is better. It's better by default or by first principles if you want to frame it that way.
I didn’t mean no one is able to do controlled studies to show anything with human bodies (obviously, vaccine studies are very straightforward as are many others). But I mean that whatever the benefits of yoga and massage and stretching and posture might be, they haven’t been able to operationalize it in a way that shows up statistically.
I'm totally with you. I just expanded it to the strength training and aerobic training because the actual studies there too are super light and the benefits should be easier to quantify. The best example of this non sense is BMI. Ok, sure population aggregates that are easy to calculate are fine to comp. but I don't need my GP assessing my BMI when they can measure my actual lean mass and body fat %.
- things like yoga have health benefits through well-known, boring mechanisms like muscle building and flexibility
- the woo woo associated with these practices describes some subjective experience associated with the practice, and/or acts via placebo effect to benefit people who believe in it even though these things have no empirical basis
I've heard yoga teachers reference chakras and energy centers, for example. Are those things "real"? No. But I can squint and see how they might be appropriate metaphors to focus the practitioner's attention on the right thing.
It’s plausible that those benefits exist but belief in them comes from motivated reasoning anchored in placebo effects, selection bias and wishful thinking. It’s a lot like Moliere’s observation that if God did not exist, people would invent Him.
Yes, this is very plausible too! There are problems with being completely credulous that every traditionally-used system that has a traditional practice with a long history of incremental and progressive developments is effective, but there are also problems with being completely skeptical of everything that we haven't been able to operationalize into statistically-detectable benefits.
I'll admit I didn't know the difference between osteopathy and chiropractic (which is, I think, a quack system).
But I can't help feeling that maybe DOs and MDs differ in two ways. DOs focus more on a holistic approach to the patient, which is probably good. And they're also taught that most illnesses originate in the musculoskeletal system, which is probably wrong.
If the two professions get similar patient outcomes, isn't it possible that you could do better just by making MDs think more holistically, without adding the crank etiology?
Osteopathy started out more akin to chiropractic but has largely morphed into an equivalent of medical schools, and since osteopathic graduates can apply for the same residencies, it has largely just become an alternate route to become a physician if you fail to get into medical school, just like Caribbean medical schools, because, for some reason, the US has more residency spots than medical school spots. My guess is that is because of the cost of medical education where residents are a source of lower paid MDs in hospitals.
I was definitely surprised by the number “1 in 5 Americans don’t know about osteopathy”. I consider myself fairly knowledgeable, and yet if you gave me a list of terms like this, I’m not sure that I would have picked out “osteopathy” as the actual name for a type of doctor!
Also, I’m interested in the naming convention - I’ve seen the contrast of allopathy vs homeopathy for historical medical practices that treat a symptom by giving you something that does the opposite of the symptom vs medical practices that treat a symptom by giving you a tiny amount of something that causes the same symptom. I’m wondering how treating the bone fits into this same schema.
It’s based upon the quack presumption that most ailments originate from an imbalance in the musculoskeletal system - how your bones are arranged. That’s largely abandoned by contemporary osteopathy
The idea that DO's focus more on a holistical approach to the patient than MD's is DO propaganda. Is there actually evidence of this?
My (MD) doctor has given me plenty of holistic/non-drug advice:
- Change my diet (weight loss, blood pressure)
- Exercise more (general health, weight loss, blood pressure)
- For a foot pain issue, try different insoles
- For an old injury, a set of exercises that should help and a referral to a PT
- Stop taking allergy meds to sleep and use a better sleep routine (or at least switch to melatonin which has fewer side effects)
They did eventually prescribe blood pressure lowering medications, but they did it after trying the non-drug solutions (didn't work) and their purpose is to prevent problems in the future, so we can avoid "treating symptoms" later.
I'm my experience as a physician, I notice no effective difference in practice along equivalent physicians.
Most practice differences ("quality") seemed to be linked to quality of training (community program vs university program). DOs seem to be more likely to have trained at community programs.
I've never even heard of someone even discuss OMT in the hospital setting (where I work).
Agree with all. I've never heard a resident who attended a DO school even mention OMT/OMM or OPP. And now that DO-only residency programs are basically gone, it'll become even more rare (unless there's some specific demand for it that gets paid).
It's far more common for old-timers to say residents these days are lazy and useless because they're young and have work hour protections, not because they went to a less esteemed medical school. Which is...progress?
It is true that if you work fewer hours per year, you get less experience. I.e., a resident who works 60 or 80 hours a week gets the same amount of experience under their belt in one year as a resident who only works 40 hours a week for a year and a half or two years.
I think we'll eventually see a push to extend residencies but make hours shorter. But it would have to come with a significant increase in the number of residency slots. We're heading toward shift work instead of call.
I’m genuinely curious whether there is any data determining how long a particular residency should be. My guess is no, it’s all just habit and history at this point, and everyone assumes it’s fine as long as their residents are passing the boards in sufficiently high numbers.
That's my impression. Residents do have case logs with minimum numbers of various types they need to graduate. Some of the required patients are somewhat rare so you go way over on regular patients trying to at least see a few of the rarities. And there isn't any simple way to make that easier given how much luck is involved.
This is the general clinical benefit to a larger, which tends to be university/academic, program. You will likely see a wider range of cases, and more rare and complex things. This matters for practice, I think.
"I can teach them how to do surgery in 3 years, but it takes more than 5 to determine when to do surgery"
("when" meaning which patients really would benefit from surgical treatment and which would not... this wisdom is the art and mastery for sure)
I actually suspect training duration has been well studied, since the effects of duty hours is well studied, but I'm not sure how that had affected anything.
I have doubts that assumption would hold up to real scrutiny. The quality of the hours of sleep deprived experience that I saw my friends in residency experience was probably fairly low compared to what they could learn when they were awake, alert, and well rested. Not to mention the quality of care, or lack there of, said sleep deprived residents were providing to their patients.
I think in general the research on duty hours shows positive benefit to residents (well being, burnout, etc), and has not shown any significant change in patient outcomes (good or bad).
Many studies and review articles are available on pubmed, most with this theme.
The duty hour era coincides with the modern EMR era so that may be a factor affecting (helping) outcomes across the increased shift handoffs as well.
Yes I'm sure at some point there're diminishing educational returns from cramming more hours in the same number of days; there's a physical limit--whether it's 40 or 60 or 80 hours a week or something else depends a lot on the individual's physical stamina. But subject of course to that the basic point holds, for any craft or endeavor where competency requires accumulating hours of hands-on experience.
This is the US vs UK difference. A us 3 year residency is often 5 years or so to get equivalent credential in UK. Though their system is a little more varied whereas US is more standardized (as far as I understand things).
US: high cost medical education, shorter but intense residencies (60-80 hr weeks, though honestly 60 is probably more true more often than 80), higher eventually salary.
UK: less expensive education, longer more "relaxed" training (40-50 hr weeks), lower eventual salary.
My wife and I have a DO as our primary care doctor. He's been a fine doctor, and I have no complaints, but osteopathy has been totally irrelevant to either of our care. He's never recommended OMT, he's prescribed drugs when anyone else would, and basically he's been identical to an MD.
Arguing from academic prestige is a bit of a pet peeve of mine. A bad article can almost always be criticized substantively. Eg, this one uses a non-random sample - are most medical studies better than that? Idk but maybe you do
Having worked in devices and Biopharma in many therapeutic areas, I’d recommend having more skepticism of literature from less prestigious journals and reserve judgment until finding are confirmed by multiple studies. It’s very difficult to discern which reason a study appeared in a less prestigious journal:
1. it’s a weak study that couldn’t get published anywhere else
2. The researcher lacks the credibility or network to get considered by better journals
3. It cuts against the orthodoxy and is being boxed out by the establishment.
My sense is that these are in order of frequency with #1 being most common
I agree that one should be more cautious approaching articles in lower-tiered journals. But I think that should lead to greater diligence reading the article, which can then inform a substantive critique - as opposed to just using the journal as the basis for the critique alone. If you don’t have sufficient background knowledge to assess the study’s quality, then I’m not sure you should be citing individual medical studies in the first place.
Completely agree. Disregarding a study a priori based on where it is published is lazy and sloppy.
I also agree that people shouldn’t cite a study if they have no ability to gauge its quality. The consequence is that most journalists shouldn’t cite studies at all. I’m not sure that’s optimal. There are too many issues deserving of coverage and needing journalists to bridge the expert-public divide and covering them in a non-technical way is unsatisfying. Ideally a journalist would consult with experts who explain the issue and literature. That has flaws though. It tends toward “both-sidesism” given that most important topics are subject to debate. And it’s not feasible to interview multiple experts for every story. Given that, I can’t think of a better rubric than to see journal prestige as a proxy for something being worth relying on.
I’m a little gobsmacked by some of the hostility in these comments (canceling your subscription seems a bit much) because I found this an interesting explanation of something I was curious about and while it makes the case that DOs have value it isn’t exactly a strong rallying cry for them. As the parent of someone who aspires to attend medical school I found this post informative and useful. Thanks Ben.
Canceling your subscription over this would be ridiculous, but the article really misleadingly friendly to DO claims. The mainstream establishment treated them like quacks because they were quacks. Modern DO's are fine because they're basically MD's who are also taught some random quackery that most of them are smart enough to ignore.
Journalism needs places for younger writers to hone their skills that aren’t centered on advocacy, as James Bennnet described about his time at NYT. Standards should be the same, but I try to be more constructive in my comments.
Speaking as an MD and practicing oncologist, I'm not surprised to hear future MD's higher performance on the pre-requisite coursework and testing required to enter medical training doesn't lead to eventual higher performance on patient outcomes like mortality and hospital re-admission. There is a huge space between knowing the best medical approach for a given problem and actually working with an individual patient to execute that approach successfully. Patients need to feel like they trust the recommendations they're receiving. They need to feel heard when they describe side effects from a treatment or the practical barriers they might face to the medically optimal approach. Allopathic training has been certainly been placing more emphasis on this aspect of care in the past 10-20 years. But an innate talent for making those connections with patients is under measured by GPA or MCAT scores. I haven't gone through osteopathic training, but I suspect that the foundational emphasis on holistic care probably leaves DOs somewhat better suited to that aspect of practicing medicine. The success with more granular outcomes like treating chronic pain seems consistent with that conclusion too. Approaching patients as individual people with unique circumstances, experiences, and needs is the bridge between knowing the best medical/pharmacologic strategy and actually achieve the best possible outcome. Even if MD's are slightly more likely to have mastered guidelines, they are not necessarily better suited for the day to day work of caring for patients.
Of course we want MDs to be humanistic and caring etc. and that is not measured with GPA or MCATs, but then it isn't really measured by anything else at the time of admissions. I was on the admissions board at UNC School of Medicine when I was a medical student, and there were some efforts to find people with strong backgrounds suggesting more humanity, compassion, but it was not easy, being that all the prospective students knew they had to do all those volunteer things, work with handicapped children, or in some lab. I think that matters, but it is hard to tell who is the next Jonas Salk from the person just doing those things because they know it will help them get into school.
And GPA and MCATs do correlate with intelligence and/or hard work, and they are hard to game, though I support giving a boost to people from more challenged backgrounds (not race but challenges) than those likely to have had Kaplan and the best college preparations.
We got a lot of humanistic type training in med school before we actually hit the wards, and I feel it was a bit wasted, and would have been more useful later, because when we got to residency it disappeared and that is when the real need was to encourage a caring approach, but I found we got little in residency, where you had to keep your head down and work the work, and sleep through lectures.
I would be curious to know the number of researcher-weeks that goes into a story like this. It feels a bit of a weighty topic to handle with the usual rigor of "we should build more houses because people need to live somewhere".
With the benefit of five minutes' thought, I'm not sure I'm super comfortable with this.
The analysis and supporting evidence seemed incredibly weak- I was very confused halfway through the article until I scrolled back up and confirmed it wasn't a MY piece. The footnote opened my eyes though- once Ben said his interest was triggered by the fact that his sister was in a DO program, the level of analysis and particular perspective espoused by the article made more sense to me.
When there is a big news story, there should be a thread to discuss it even if the article is just a stub. I suspect more of us want to discuss the presidential race than osteopathic medicine, but Ben wrote a good piece that deserves focused discussion and comment
I would rather hear about osteopathic medicine. What more is there to say about the race at this point, and what can any of us do about it? But from yesterday's article, I know that my demographics aren't typical of the SB readers beyond the education thing.
No, I'm right there with you. Some stuff will happen (the race will move, or it won't, or Biden will bail, or he won't), but right now all of the chattering seems well and sufficiently chattered.
I've been checking out the Ezraklein subreddit a lot lately (I don't think Ezra is involved in it at all. Also, there doesn't seem to be a Matt subreddit):
Interesting article. The point about the lower rates of opioid prescription by DOs highlights the role played by physicians in the opioid epidemic. While the Sacklers are great villains and bear a great deal of blame, it’s also true that doctors can’t pretend to be unwitting pawns. In fact, doctors had the agency, training, and intelligence to make better choices for their patients but in many cases failed to do so.
It'd be interesting to really get into the data there too. I work in the health care field, specifically focusing on fraud/waste/abuse, and what we see with doctors is a huge discrepancy between the regular levels of opioid prescribing and the outlier levels- there are some incredibly bad actors that are, essentially, just drug dealers. They're known in the community, they have connections with literal street level drug dealers, and they just issue enormous numbers of prescriptions. In my experience, they're exclusively MDs. So I'd be curious to see how much of the "rate of opioid prescribing by practice type" is driven by the extreme outliers and what the comparative rates between MDs and DOs would be if you eliminated those clearly criminal actors from the analysis.
Thanks for this, Ben. I have been going to a DO for many years, but knew almost nothing about the differences and similarities with MDs. If I actually see the DO at an upcoming appointment, I will ask her about it. She is the most popular doc in the practice, though, so it is hard to get her for appointments that aren't scheduled well in advance.
The fact that the average grades and scores are slightly lower for DOs does not bother me a bit. A doctor who really listens to what I am saying is what I need. I also want a doctor who isn't over confident and is willing to look things up. Sometimes the doctors with the highest scores are unwilling to do either, especially at the beginning of their career.
Whatever philosophical basis osteopathy has, most students in the current are just aspiring physicians who fail to get into medical school. Like the Caribbean schools, it is an end run around the system. The average medical school admission has a 3.72 gpa and osteopathic 3.54. I could not find the data on MCATS, but presume similar.
Why the US, one of the richest countries in the world, doesn't not train enough MDs to fill all the available residency slots is unclear, but these spots are always filled preferentially with graduates of American medical schools, and the remaining slots by osteopathic graduates and foreign medical school graduates. It is not some philosophical difference that draws people to schools of osteopathy, as there might have been in the past, and they should just call the osteopathic schools, medical schools, and they can continue their traditions, but not pretend that people choose them because of that, as opposed to just failing to get into medical school.
I am not claiming that every student who cannot make the average grade point and MCAT score needed to get into a medical schools is incompetent and I have known osteopaths in my career as a physician who were great doctors, but we should be honest about why people choose that route, and we should train as many physicians as we need.
"Why the US, one of the richest countries in the world, doesn't not train enough MDs to fill all the available residency slots is unclear, but these spots are always filled preferentially with graduates of American medical schools, and the remaining slots by osteopathic graduates and foreign medical school graduates."
& "we should train as many physicians as we need."
If we moved the people going to DO schools because they couldn't get into MD schools into MD schools, what difference would it make?
Not a lot, but it would end the charade that there is really some difference between the two. Also half of the medical school slots are filled by foreign grads, either Americans going to lower quality less competitive schools or foreign grads, usually who are very high quality, but they are needed more in their own countries and come here b/c they make a lot more money and it is part of the brain drain from abroad.
“ and they should just call the osteopathic schools, medical schools,”
Are they not already? Like, if someone said they were going to medical school, one would not correct them and say “actually you are attending a college of osteopathic medicine” or something.
You do not get an MD but an OD from osteopathic schools, even though residencies accept both and most if not all states treat them the same. They have different accrediting and governing bodies, but if they are essentially producing the same practitioners, what is the purpose of two tracks. At one time osteopathy was more alternative like chiropractic but not anymore.
I’m aware of all that. I read the post. I’m just saying, at least colloquially if not technically, “med school” encompasses both. You would be an asshole to tell a DO student that they aren’t attending medical school.
Twenty years ago when I graduated college, plenty of students chose the Caribbean medical schools over DO schools largely because they wanted the MD on their ID badge. Now my understanding is that almost no one would choose Caribbean over DO.
At one point the DO school in Fort Worth was (for in-state students) by far the cheapest tuition in the nation. A lot easier to choose primary care when your debit is < $100k.
Why is this an argument for osteopathic medicine, rather than an argument that the MD pipeline, cartelized by the AMA, should be expanded significantly?
I have nothing against DO’s, and I’d be happy to use one. But this article praises them primarily because they make up for the shortage of MDs, and because they’re roughly as effective as MDs. But the shortage of MDs is artificial, imposed by the cartel.
So, why not just expand the MD program so that the kids who currently get DOs get MDs instead? (MDs are also roughly as effective as MDs: this has been proven in extensive clinical trials). “We need more doctors, and DOs are more or less doctors,” is not a defense of osteopathy, just an indictment of the cartel.
A genuine defense of osteopathy on its own terms is a secondary theme in this article — maybe they do better with chronic pain and hypertension. But that defense is fairly weak: it’s not clear that the improvements are that large, that they are the result of anything distinctive about osteopathy, or that they cannot be swiftly imitated by the MD system.
So: hurray for more docs! Let’s train more docs! Let’s not worry too much about whether they are MDs or DOs!
All good points. But a reminder that the AMA intentionally produces fewer docs than our country needs.
That’s definitely a part of it! This article was merely an attempt to discuss a growing medical field!
I appreciate the piece. This is in the interest of combating some of the hostility in the comments but still providing feedback:
1. I’d be okay with more explainer type pieces on undercover topics, especially from writers other than Matt. On that goal, the piece really succeeded and I enjoyed it
2. The criticism of the studies you cite is substantive. It would be interesting to write a piece on the journalistic task of evaluating technical literature. What is that like and how can it be helped?
3. Recognizing that SB’s template is to mix an explainer with a policy recommendation, writers will be judged on the weaker of these. If you don’t have a cogent policy take, it’s okay to just say “I don’t know what this means going forward,” as Matt sometimes does. What I’d prefer is to turn the challenge into a policy statement. “The relative effectiveness of DOs is poorly studied and here are some hypotheses why. Here are the key questions I’d love to see answered in the future…”
I would have liked this article much more if it had simply presented osteopathy as an interesting topic, rather than recommending them. That part felt a bit like standard journalism based on shaky science, not up to the rigor of a typical slow boring piece.
“…an attempt to discuss a growing medical field….”
And it succeeded as such! Thanks, Ben.
But is it a growing medical field? Or is it just a label that is applied to certain medical schools based on an increasingly unimportant historical division.
I mean, if I started talking about and comparing academics educated at schools like Cambridge and Oxford that descend from an originally religious mission to schools that descend from a secular foundation am I actually discussing a academic split.
Or, as seems more likely, am I just seeing the lingering traces that the originally religious institutions are older and therefore tend to be higher prestige. Sure, Oxford and Cambridge have tradictions that UCL lacks just as DOs do but I doubt that it's more than peripheral/emotional (ppl grow attached to their educational traditions)?
Yeah, it’s good to be clear about what actually happened: osteopathy didn’t turn out to be good, its quack founder wasn’t vindicated, what happened is that osteopaths gradually shifted to mostly using real medicine and only teaching a bit of pseudoscience. Which is only beneficial in the context of a system where “just train more MDs” isn’t allowed.
Editing to acknowledge that, as people have pointed out, the restrictions on the doctor pipeline are more complicated than just the number of MDs trained. Does allowing DO schools actually increase the number of doctors that come out at the end?
I had the same thought.
I do think there is a perennial and difficult question here, though. Insofar as many people's health problems are psychosomatic, which is probably responsible for the vast bulk of the sincere experiences of pseudoscience improving people's subjective experience of their own health, there is some utility in a person who can give you real medicine for a real problem and also be willing to give you some made up stuff otherwise in case it helps resolve whatever psychological hangup is causing your symptoms. Dividing those two roles into pure allopathic medicine vs. pure quack practitioner has drawbacks -- former is unwilling to provide treatment that can relieve symptoms, which pushes patients toward the latter, who then might unscrupulously continue to prescribe fake medicine even when problems with physical causes emerge.
I thought the number of doctors was limited by the number of residency slots, which in turn was limited by Congress? Though I also thought there was a recent increase in that cap.
I am very curious about this. I don't know who limits the slots. I live a few blocks from a teaching hospital. What I have heard from them, especially since the pandemic and subsequent staffing shortages, is that they don't have the capacity to take on more residents, even if "they" (whoever that is) made more available
There are two pools, medical students and residents. There are currently more residency slots than US medical school graduates, so all graduates of US medical schools can get a residency. The remaining residency slots are filled by DO grads, American graduates of foreign medical schools, and foreign graduates of foreign schools. Foreign medical schools which include both US and foreign students supply 25% of US residents, and DO schools about 10%.
It is much less costly to fund a residency slot than a medical school slot in fact many hospitals need residents to get their work done, where one study at UVA calculated it cost $100,000 a year to educate a medical student, and that is at an established school, so likely to start a new school would be significantly more. Now the average tuition runs about $60,000 varying a wide degree with public vs private, so tuition likely pays a little more than half the cost.
I understood how the process works, but I am quite surprised to hear that there are more residency slots than US med school residents because of our local situation. I didn't think our hospital would be highly coveted. But your comment made me look at reviews, and now I see that it is popular for reasons I hadn't thought of - cheaper cost of living than big cities, but near enough to big cities that there is an overflow of talent; more supportive of residents in general, especially residents with families, than a lot of places; wide variety of patients since it is regional; better support than many places for urban, low-income patients.
The number of residency spots is determined by the Centers for Medicare & Medicaid Services at the federal level. They do limit it, but it is still much more than the number of medical school grads which I think is determined by the states or private universities. It can be hard to get into competitive residencies whether due to the specialty - orthopedics, surgery, etc. can be very competitive or whether the residency is affiliated with a prestigious school, but there are plenty of Family Practice and Internal Medicine programs that dont match their positions and then have to go outside the match to fill slots mostly with foreign grads.
Exactly. The AMA does not set the number of medical school positions. State legislatures and private entities determine the number of medical school slots and fund them. If NY or MT want to increase their medical school positions they can. The AMA is involved in accreditation. From my experience the pool of osteopathic students, just like the off shore Caribbean schools, is composed of people failing to get into medical schools.,
FWIW, if NY or MT increase their number of slots, the extra slots will be filled by people who otherwise would have failed to get into medical school.
Because the number of onshore MDs is artificially limited to less than people who could do the job, talking about people who failed to get into medical school is sort of like talking about people who "failed" to make the Olympic Swim Team or "failed" and less like talking about people who failed to get a drivers license or failed to finish college.
I have no problem with extending whatever standards you want lower to accompany more people going to medical schools in order to fill all the available residency positions. I don't think it would be a great hardship if the standard of gpa which now might be about 3.6 was lowered to 3.5, not a big deal, but among all those that fail to get into medical school, not everybody goes the DO route so that lowers the competitive level for DO schools.
I dont know what standards you suggest for your doctors, but given a choice between a higher and lower achiever, I would take the higher. I have nothing against NPs and PAs either, but at least that is more explicit and they are not called physicians.
"but given a choice between a higher and lower achiever"
It's very difficult to find a doctor, especially a primary care physician. My dad lives where he does now primarily because he can access doctors more easily than he could where he would otherwise like to live.
The problem is there's not enough choice, and it's frustrating to hear talk of about "standards" when they boil down primarily to GPAs. I'd almost agree that I'd prefer a doctor who had a 4.0 GPA to a doctor with a 3.5, but the truth is I'd evaluate on them o a dozen other different metrics first, and I've never heard of anyone asking a doctor how their college performance was.
So yes, give me a doctor nearby with a 3.0 who has time to treat me over a 4.0 GPA doctor who lives an hour away and is fully booked for the next 3 months. I know of no other profession that gets so hung up on this sort of gatekeeping.
Wigan, I sat on the admissions board of my medical school which had two students. We did consider things other than gpa and MCAT, but it can be a little fraught as prospective students all knew they had to volunteer with the red cross or migrant workers or what not. I actually likely benefitted from this because I worked overseas in relief work before going to Med School. But most resumes look similar and yes, grades hardly mean everything, but they are fairly objective say compared to an essay which students might have been coached on.
Ideally you want humane physicians, and if there was an easy way to measure that it would be great, but given two equal MDs, Ill take the one with the 4.0 over the 3.0 because at least things like Internal Medicine involve very challenging mental constructs for which I want the smarter doctor.
As to the shortages, it depends where you are. I have never had a problem finding a doctor, but in some rural areas, yes. I do think we should train more doctors and that doctors should spend more time with their patients, but a lot of that is determined by money. When I worked for a big employer we were graded on productivity, which you achieved by seeing the most amount of people in the least time, and charging the most. I think medicine should be divorced from the market place.
This all makes sense and thanks for speaking to your experience.
"I dont know what standards you suggest for your doctors, but given a choice between a higher and lower achiever, I would take the higher."
"The average medical school admission has a 3.72 gpa and osteopathic 3.54."
I don't have any strong opinions on osteopathy, but I think assessing whether someone will be a better doctor because they have 0.18 higher GPA in undergraduate work is a choice I decline to make. Give me some real evidence that MD's perform better than DO's and I might buy in, but as best I can tell there is no substantive difference in their patient results.
It would be very difficult to devise any study to determine which provide better care as that is so multifaceted. The MDs score higher on tests to get in so I would presume afterword, they would also. Tests mean something or you wouldn't have to pass your boards to practice medicine. They don't mean everything, and I was on the admissions board of a medical school, and we also look at other things like humanitarian concerns and hard work from people from challenged backgrounds, and as much as you can assess things like empathy and integrity, they play a role.
If gpa and MCATs means nothing and we should do a lottery and B- students would have as much chance as an A student. But medicine is very complex, and I would rather put my trust in an MD who was a higher achiever. There is a correlation between grades, test scores, and intelligence and hard work. We put our lives into the hands of MDs and I myself would want them to be very hard working and intelligent.
If you're asking me whether I want someone who had a 3.5 gpa in their undergraduate or a 2.5, then yes I'll take a 3.5. But I don't think the difference between a 3.72 and 3.54 in UNDERGRADUATE work provides meaningful information about whether someone is more hard working and intelligent as a doctor.
Looks like I’m just wrong about who creates the bottleneck in producing MDs. If it’s not the AMA, then I apologize to them for the false accusation.
Nonetheless, the bottleneck is artificial, whoever is creating it. If legislators are responsible, then they should stop.
To be fair to your point, the AMA is one of the most powerful interest groups and the state congresses are certainly being driven by the AMA, not the other way around.
I spent several years in the anesthesiology device market and the American Society of Anesthesiologists two primary activities are limiting competition and containing malpractice risk of its members.
Agree, simply fund more medical school positions. Being that the DO schools are already there with faculty, doing the same thing, just incorporate them into the Medical system. I dont know other professions that have two tracks, so if you cant get into law school you would go to a lawgo school where you would learn the same thing and then do the same work. The bottleneck is that it costs a lot of money.
In the old days (30 years ago, when I was last paying attention), in California, there were three tracks to getting admitted to the bar. Three years in an accredited law school, or four years in an unaccredited law school, or five years of working in a lawyer's office learning the trade. Then you had to pass the CA bar exam, which is notoriously one of the more difficult ones. And then of course you had to get hired. Wasn't Kim Kardashian trying to become a lawyer via the third route? I think she went back to being ultra rich instead.
Yes, Kim K. was purportedly "reading the law," but that's now reportedly "on pause," in part because she needs to focus on playing a lawyer in an upcoming TV series: https://abovethelaw.com/2024/02/is-kim-kardashian-giving-up-on-her-dream-of-becoming-a-lawyer/
People still like to blame the AMA for their '90s positions, not their 2024 positions. It's annoying, but I don't really care about the AMA enough to defend them.
I think the median DO school reputation is still seen as generally below the median allopathic school. But now I think grads a mid-tier medical school and top quartile (or maybe half) good DO school are seen as approximately equivalent. Essentially, the bias is strongest at elite programs because they're picking between Harvard and Yale grads. It levels out the further down you go because elite candidates rarely trickle down that far so it becomes more individualized.
Caribbean medical schools are still seen as inferior and are heavily biased against. I'm not sure how they fare against other foreign medical schools since I can't recall any program I've been affiliated with matching any (aside from fellowship after completing a US residency).
I've heard an MD say that the MD and DO programs are overlapping bell curves. At the far ends, the best MDs are better than the best DOs, and the worst DOs are worse than the worst MDs. But the huge middle area was full of perfectly capable future physicians.
Are the worst DOs not fully capable?
In many, or maybe most fields, a somewhat lower end performer is still useful. I'm sure the worst lawyer in town would not be my first option, but if everyone else is booked he / she may be a lot better than nothing.
And this would likely be especially true in a field with artificial supply constraints.
"Are the worst DOs not fully capable?" I mean...no? The worst MDs aren't fully capable either. Physicians do lose their licenses sometimes, and there are others who *should* lose their licenses but escape that fate. This is some deep Internet/Usenet dark lore, but I remember the Usenet diet groups being overrrun by one Andrew Chung, MD, who had 100% clearly lost his everloving mind and was, eventually, after many years, stripped of his license. Those guys do exist. Google "Dr. Andrew Chung" and "2 pound diet" if you're bored.
Haha yikes. I believe you. I've heard a few stories of off-their-rocker doctors, too. I think several of my ex-girlfriends being doctors also makes it easier for me to believe that doctors are not always hyper-rational, unbiased, 100%-knowledgable people, lol.
I was going to write a separate comment but yours hits the gist of it first, so I’ll add mine in support.
My understanding of DOs is that the school is easier to get into so people who really want to be doctors go there as a second choice. My initial reaction was to think that I should shy away from them, but if they take the same test and have the same outcomes, why have the difference without a distinction at all?
At one time they actually were different, more akin to chiropractic, and still incorporate some of the spinal manipulation, but people go their mostly just to become regular physicians for not getting into medical schools. Graduates with DOs and MDs go to the same residencies. As the article mentioned DOs tend to go more into primary care and I suspect that is just because they are the less competitive residencies, not that they are not needed, I am an FP myself.
You cite some evidence in favor of osteopathic medicine working. I have had the misfortune of reading them. I will be very blunt.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6134408/ — “simulated rct” on observational data. So an after the fact fishing expedition, with no attempt whatsoever for sorting on perceived severity. Junk, toss it.
https://pubmed.ncbi.nlm.nih.gov/24481801/
Another observational study. Don’t let doctors do statistics! They’re so bad at it! Junk, toss it.
https://dacemirror.sci-hub.ru/journal-article/e956b3f56e92cb0c2a48ba030511a3cc/cerritelli2015.pdf?download=true
I would bet you, at even odds, that this paper is fraudulent. Look at figure 2, page 7 — what is the theory here for why it should have a preposterously large effect on migraines? Does it not boggle imagination? You have found, not treatment but a panacea. So no, it’s probably horseshit.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2848182/
They find no difference between dummy light touch treatment, and OMT as well. First decent set of methods so far.
Chiropractic medicine does not hold up — why does this?
You need to be more skeptical of the academic literature. It is filled with liars and fools — especially in medicine. If it’s not an RCT, it’s probably junk, and even if it is. You need to be more aware of sorting by unobservables. This article is far too credulous, and does not bring credit to the byline.
Hey listen OMT is a hard thing to conduct a randomized double blinded study on. But there are reputable organizations that clearly think it works and recommend it as a form of treatment. Below is a Cleveland Clinic explainer on it.
https://my.clevelandclinic.org/health/treatments/9095-omt-osteopathic-manipulation-treatment
The Cleveland Clinic will also let you pay for Acupuncture if you really want to.
https://my.clevelandclinic.org/health/treatments/4767-acupuncture
That explainer is no more than marketing. Without question, the idea of a doctor being ‘holistic’ and ‘hands-on’ is appealing to patients.
This also isn't just a complaint about blinding. You just absolutely cannot use a word like "proven" when linking to an observational study that made no attempt to correct for differing underlying patient characteristics.
Also, concerning a different citation, if I read a sentence like this in a paper it doesn't exactly suggest I should rely on it for treatment recommendations:
> A total of N = 31 out of N = 63 eligible subjects followed by a single cardiologist received osteopathic treatment in addition to routine care.
This article is irresponsible and misleading.
It's not "bias" to favor actual sciences in this instance.
You rely heavily on the performance of osteopaths vs medical doctors in the opioid crisis. The medical field obviously blew that one. But osteopaths didn't do better because they had better insights; they just are skeptical of all medical advancements and best practices. Consequently, the medical field went off the rails in this instance, they did not go with it.
Good for them - seriously; that's a lot of lives thankfully not ruined. But relying on this to elevate osteopaths is like praising the Catholic Church for its superior family planning approach because their adherents did not use thalidomide. It was right in that instance but not for the reasons its advocates (and this article) claim and ultimately does not indicate any particular insight that will sustain in the future.
It also does not indicate that residency programs are wrong or "biased" to favor medical doctors over osteopathy (or actual OB's over Catholic doctrine...)
As an older economist don't fall into the "observational data is not informative because it's not an RCT trap."
Also everything is endogenous when it comes to referees and they demand that you prove a negative.
That’s a generalization that doesn’t address the specific deficiencies of these particular studies. The hypertension study is biased because the DO is selecting which patients get OMT; that’s a huge no-no. The pneumonia study showed no difference on an Intentional To Treat (ITT) basis, which is the unbiased framework. Pointing to a per protocol (PP) advantage for OMT could, again, be a reflection of biased selection by only choosing patients who are likely to have better prognosis to actually get OMT.
But what I wrote is an accurate description of the short hand Nicholas leaned on. The biggest problem with medical studies is that people don't accurately describe what they are measuring.
Economists by now focus on "natural experiments" precicely because they realised that purely observational data is very bad at estimating effect sizes or even their direction.
Yeah, but ... natural experiments are purely observational data, just with some extra math to simulate a controlled experiment. They're vulnerable to all the same threats as a straight correlational study, plus the extra degrees of freedom given to the researchers. I don't think that the field is as enamored of natural experiments as you might suggest. A brief review of the 5 latest articles in QJE has three _controlled_ experiments, a correlational study, and an analytical model.
I didn't read the paper that Mr Decker linked, but "simulated RCT" is probably medical jargon for what social scientists call natural experiments.
That’s not true. Natural experiments are specific situations where there is an unintentional random variation that can be exploited as if were an intentional experiment. For example, when a charter school is oversubscribed, it would often randomly choose the kids to be admitted. You can then compare outcomes between those kids and kids who applied but were not admitted to estimate the impact of attending the charter school as if this were an RCT (whereas comparing the results of kids in the school to kids who did not apply to the school is misleading, because the parents sending their kids to charter schools are plausibly more committed to their education, so their kids would do better anyway)
It isn’t! It’s controlling on unobservables again. Nothing to do with an RCT.
QJE is just a publication mill so Harvard grads can get tenure. (Sort of jest but sort of true.)
True. 😞 And the name of St. Peter is “Katz”
Larry has taken a back step and has been replaced with arbitrary identification police
I have found many laughable cases where economists declare something a “natural experiment” like school entry dates and timing of births.
It is particularly bad among Harvard grads who don’t have a handle on history or the data generating process but who have the egos to assert their source of variation is plausibly exogenous. (So many rainfall IVs or temperature “shock” claims.)
Wait, how is birth timing _not_ a natural experiment? Whenever you have an arbitrary cutoff, there will be people who are similar except for falling very slightly on one side vs. the other. If a school has a strict age cutoff, comparing students born one week prior to the cutoff vs. one week after is a plausible natural experiment, no?
Rich people strategically plan out when they are going to have children, usually during the summer. Poor people don't strategically plan fertility. There is a whole literature on debunking the Angrist and Krueger birth timing IV.
When you use birth timing, you are both picking up a selection effect and family income effect.
That's pretty interesting, I didn't know that but it makes sense.
In that case, couldn't you still have a natural experiment if you selected a small enough time window? Say, comparing children born two weeks before the cutoff vs. two weeks after? I wouldn't expect people to be able to have fine-grained enough control over birth timing where those populations would be meaningfully different. I do know some moms who deliberately scheduled c-sections, but I'd expect it would need to be close enough to the "natural" birth date for a doctor to go along with it.
(Not trying to pose a gotcha question, you just seem to know a lot about this topic and I'm genuinely curious)
I actually disagree with Mr. Quail, provided you can show two things. One, the discontinuity must be tight around the discontinuity. Two, you need to show that it genuinely is a hard cutoff, and people are not slipping from one side to another. We should be cautious, but it is not a priori a bad idea. You just need to work harder to show it.
Season of birth a la Angrist and Krueger is waaaaaaayyyy too big a time period.
Obviously there's better and worse versions of that, but looking at raw correlations is still a bad idea.
Most things that aren't RCTs are pretty worthless. The level of rigor needed to draft a usable observational study is enormous and most academics fail to clear it, as the vast majority of effects hypothesized by observational studies fail RCTs with flying colors.
And yet many many many organizations use survey and observational data. You just stated that forecasting is worthless. I hope you realize how silly that sounds.
Not all questions are causal or need to identify causation.
Many organizations engage in many practices, some of which are good and some of which are not. A lot of those organizations include people invoking low-quality observational studies to advance their goals whether the results are true or not.
Questions like "do people prefer Coke or Pepsi" and "do people like eating dirt" are certainly accessible via survey, but for almost all actual policy proposals, the questions being investigated is casual: would implementing policy X have desired effect Y?
As another older economist, observational data can indeed sometimes be useful when an RCT is impossible, at least when you have many millions of clean observations. However, non-huge observational datasets are nearly always useless.
I can tell you this statement is also wrong. Small high quality datasets can help confirm or reject formal models.
Paul Rhode and Alan Olmsted disproved the whole ratcheting quota's model of cotton picking that historians loved by going back to a series of plantation records where they recorded daily production by worker. (Basically economic historians have shown that much of the 1619 Project is not substantiated in actual historical data or records.)
Quail is correct, sample size cannot undo bias.
Maybe doctors could try thinking seriously about causality. I’ll consider observational studies then.
Alternatively, the placebo effect is real! People who think they are being treated do get better than people who don’t.
My favorite quasi conspiracy theory -- is that Linus Pauling ruined is Nobel winning reputation to push Vitamin C as a treatment *knowing* it was total bogus pharmacologically but that it activated the placebo effect and that's actually *real*.
EDIT: For those not familiar, Linus Pauling the only person to ever win two unshared Nobel prizes. Just makes no sense he didn't know exactly what was going on with Vitamin C -- which this Vox article completely misses as a possibility:
https://www.vox.com/2015/1/15/7547741/vitamin-c-myth-pauling
I agree that the literature on DOs is biased and probably not worth much, but I think you are focusing on the wrong thing. If there was a massive disparity between DOs and MDs, it would show up in the data "at a glance" and you wouldn't need complicated statistics to demonstrate DOs are bad - the MDs would have done that for you.
"Chiropractic medicine does not hold up — why does this?" Because MDs and DOs are similarly regulated, similarly treated by hospitals and patients, and held to approximately the same standards (based on their admission requirements). Therefore, I would expect the outcomes of MDs and DOs to be approximately similar, and that's what we see. Chiropractic is much less regulated and easier to get into, so I would expect there to be much higher variance in outcomes, and indeed that's what that field looks like (mostly useless, a small tail of terrible fraud and abuse, but lots of people believing that it does good for them). But I guess I'm coming at it from the idea that lots of allopathic medicine isn't terribly effective anyway; it's just that a few things like surgery and modern medicine have gotten so good that they make a massive improvement is patient outcomes.
The standards MDs are held to, in terms of hoops to jump through to get the credential, are far too high; the standards chiropractors are held to are far too low. Therefore, adding a marginal chiropractor will likely make the world slightly worse, but the marginal doctor (whether they be MD or DO) will make the world better.
That wasn't the only claim made though; these studies were brought up to support the idea that specific osteopathic practices have benefits, not just that DOs are similar to MDs overall (because, as you note, they mostly just do the same things).
No idea why the third paper included a sham group if they're not going to evaluate the treatment against it. But that one seems to have a pretty large effect size as well, placebo for the win.
Placebo is an example of a treatment that's done an amazing job and is being recognized more and more, I notice.
Unfortunately, that's led to the market being flooded with fake and low-quality placebos imported from abroad.
"...flooded with fake and low-quality placebos imported from abroad."
Why can't the FDA do its job here, to ensure that Americans receive the high-quality, genuine American placebos we deserve?
Also -- have you noticed how generic placebos just don't work as well? I know, I know -- they're supposed to be bio-equivalent, made to the same standards, with the same inactive ingredients. Maybe they work the same, for some people.
But me? I only use Placebo™ brand placebos. Accept no substitutes.
The podcast "The Studies show" did an episode on the placebo effect. I think the issue is that a lot of studies don't include a "do-nothing" group and a lot of people just get better over time. A lot of the placebo effect can be attributed to regression to the mean.
Medical librarian here. I encourage people trying to evaluate this kind of evidence to look for systematic reviews and meta-analyses of individual studies, since it is a big time-saver especially when multiple trials have been conducted. In particular, look for Cochrane reviews. Here are a few:
- https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD008880.pub2/full
- https://pubmed.ncbi.nlm.nih.gov/30973196/
And a recent review of SRs: https://bmjopen.bmj.com/content/12/4/e053468?fbclid=IwAR0z3L3G6TpwvWMjmf3DZX8kqVHC2Dwmr8OxTUAT5JXBSmKRQuHbTyqLN9c
I agree, though I am still skeptical of even meta-studies -- garbage in, garbage out. I'm reasonably confident the Italian study mentioned above is just actually fraudulent, but it would go right into a meta-analysis.
Oh garbage in garbage out for sure. If the meta-analysis was conducted properly (and for sure a lot of them aren't), the reviewers would have included a risk of bias assessment for each included study and incorporated that into the analysis, as well as checking for any retractions. You see that all the time in the discussion/conclusions of SRs - "most of the included trials were of low methodological quality so the evidence isn't strong" etc etc. The medical literature is just so flooded with garbage these days, not to mention how most researchers don't make their data available for peer review (although that's gradually starting to change).
Maybe you should see a DO for your hypertension.
My blood pressure is within healthy bounds.
Thanks for doing the work here. Alarm bells were going off as I read this.
I am in the medical training system. We have DO's in prominent places at my institutions and I have hired DO's. But there is nothing in the article specific to DO's that is relevant about implications for health care today. Insofar as they have same training as MD's they are fine for the things that MD's do. Insofar as they do things that some MD's should or could do, that's fine too but also not specific to DO's. The more typical Slow Boring take would be to emphasize role of mid-level practitioners (PA's, APRN's) both in rural and primary care and as physician extenders in specialty practices. Those are much cheaper and easier to produce, support the "abundance mindset" and challenge the status quo in more fundamental ways.
If you want to see some vitriol, go over to Reddit and look at various doctor groups. The hatred of "mid-levels" is intense.
Think about why. Mid levels are encroaching on physicians’ authority. Doctors hate competition.
I’m not saying that there aren’t legitimate concerns about this. I observed conscious sedation procedures in Utah where RNs are allowed to deliver propofol and it was terrifying. But the root of MDs perspective is to shield their prerogatives and incomes
I should have been more precise. In Utah, nurses manage propofol — dose and administer without any real supervision by an anesthesiologist or CRNA. It’s technically done under the prescription authority of an MD, but the RN is making all dosing decisions. It’s equivalent to how CRNAs are empowered in many other states. That’s different from a nurse’s standard task of administering medications.
I’ve been in many ICUs, but have never observed propofol delivery, so educate me if it’s happening there also. That said, the amount of monitoring and availability of personnel to intervene is much higher in an ICU than in a GI suite where most of Utah’s RN-managed sedation occurs. I had nurses tell me that they were terrified doing it. If a surgeon/gastroenterologist pushed the nurse to over sedate the patient, the nurse would have a harder time pushing back than an anesthesiologist would and much less able to perform airway rescue if the patient stopped spontaneous respiration.
Thank you for clarifying.
There is usually no issue if the mid-level is working for an MD. Midwives do very well on their own but that is due at least in part because they are trained on how to refer complex cases quickly and safely. The issue becomes problematic when it comes to independent practices. If you worked hard (years, no sleep, less pay than others in your cohort) to get a license that you were told was needed to practice safely and then others with much less training got the same privileges, you would be pretty mad too. Not saying there isn't potentially a public interest in ignoring that, but it is the real lived experience of physicians. Of course, there is more to the story and, if done correctly, could be a good SB post.
Ah yes, the RC Cola of physicians.
From my experience almost all DOs roll their eyes at the OMT stuff and never think about it after medical school, nor do they encounter it in their residency programs unless it’s one of the small OMT/Family Medicine programs. The DOs I know call it osteopathic massage therapy.
If you attend the OMT specific conferences, you’ll find a lot of quackery represented in the vendor and sponsor booths. Homeopathy, herbal medicine, crystals… The presentations aren’t so bad but they need to clean house and that they have not tells me that the OMT component of DO education is vestigial woo woo.
That said, one thing that I worry about in medicine today is NDs - Naturopathic Doctors. N and M easily sound the same and they claim to have gone to medical school but they are not following the same curriculum as MDs and DOs. They also don’t attend the same residency programs. Despite these differences, some states license NDs as medical providers. I feel like their patients could potentially be misled into thinking they have the same training and capabilities as doctors. Perhaps they will follow the same path as DOs and one day integrate themselves into the medical field at large.
Osteopathy that focuses on OMT can definitely fall in the woo woo territory. Some forms of OMT are more mainstream, others are not.
There's always someone to pick up the woo woo for the woo-woo-craving masses.
I think the OMT convention issue is a similar one to activist movements. Once the normies move mainstream, it's mostly weirdos left behind. I'm sympathetic to complimentary medicine (in that some actually works and gets adopted) but naturopathy is not even in the same quality ballpark as DO or MDs.
My family saw DO family med docs back in the '90s (just because they took our HMO and were nearby) and they never did OMT.
I’m always wondering about massage and yoga and things like that. They really do seem to have lasting benefits for my body and how I feel , but no one is ever able to do controlled studies that show there’s something real. I suspect that a lot of it is that bodies are really hard to measure systematically in any but crude ways, though many traditions have found things that do work to some extent, even if we haven’t figured out how to quantify the endpoints. (There’s a lot of crudeness that gets tolerated by medical practice just to make it work statistically - Moderna is probably giving us huge overdoses of their mRNA because they used a big dose in their tests to ensure that it worked, and there’s no incentive to cut the dose to get the side effect size down to Pfizer, or even a regular flu shot.)
I mean ... but this is my problem with "controlled studies". We don't need a controlled study to know building a stronger, leaner, more flexible, higher VO2 max, etc. body is better. It's better by default or by first principles if you want to frame it that way.
Ok, but we do actually have those studies!
That's awesome. Definitely let Kenny know (i.e., "no one is ever able to do controlled studies").
I didn’t mean no one is able to do controlled studies to show anything with human bodies (obviously, vaccine studies are very straightforward as are many others). But I mean that whatever the benefits of yoga and massage and stretching and posture might be, they haven’t been able to operationalize it in a way that shows up statistically.
I'm totally with you. I just expanded it to the strength training and aerobic training because the actual studies there too are super light and the benefits should be easier to quantify. The best example of this non sense is BMI. Ok, sure population aggregates that are easy to calculate are fine to comp. but I don't need my GP assessing my BMI when they can measure my actual lean mass and body fat %.
It seems plausible to me both that:
- things like yoga have health benefits through well-known, boring mechanisms like muscle building and flexibility
- the woo woo associated with these practices describes some subjective experience associated with the practice, and/or acts via placebo effect to benefit people who believe in it even though these things have no empirical basis
I've heard yoga teachers reference chakras and energy centers, for example. Are those things "real"? No. But I can squint and see how they might be appropriate metaphors to focus the practitioner's attention on the right thing.
It’s plausible that those benefits exist but belief in them comes from motivated reasoning anchored in placebo effects, selection bias and wishful thinking. It’s a lot like Moliere’s observation that if God did not exist, people would invent Him.
"...like Moliere’s observation that if God did not exist, people would invent Him...."
Moliere might have invented that quote, if Voltaire had not existed.
I'm here for the debate as to whether Moliere's plays were actually written by Voltaire!
I think that might require time travel? I.e. Voltaire later than Moliere?
Twist: Voltaire's philosophic works were written by Moliere but hidden and then published years later under a pseudonym!
Yes, this is very plausible too! There are problems with being completely credulous that every traditionally-used system that has a traditional practice with a long history of incremental and progressive developments is effective, but there are also problems with being completely skeptical of everything that we haven't been able to operationalize into statistically-detectable benefits.
*Norman Rockwell Guy Meme*
RC Cola is Good.
I'll admit I didn't know the difference between osteopathy and chiropractic (which is, I think, a quack system).
But I can't help feeling that maybe DOs and MDs differ in two ways. DOs focus more on a holistic approach to the patient, which is probably good. And they're also taught that most illnesses originate in the musculoskeletal system, which is probably wrong.
If the two professions get similar patient outcomes, isn't it possible that you could do better just by making MDs think more holistically, without adding the crank etiology?
Osteopathy started out more akin to chiropractic but has largely morphed into an equivalent of medical schools, and since osteopathic graduates can apply for the same residencies, it has largely just become an alternate route to become a physician if you fail to get into medical school, just like Caribbean medical schools, because, for some reason, the US has more residency spots than medical school spots. My guess is that is because of the cost of medical education where residents are a source of lower paid MDs in hospitals.
I was definitely surprised by the number “1 in 5 Americans don’t know about osteopathy”. I consider myself fairly knowledgeable, and yet if you gave me a list of terms like this, I’m not sure that I would have picked out “osteopathy” as the actual name for a type of doctor!
Also, I’m interested in the naming convention - I’ve seen the contrast of allopathy vs homeopathy for historical medical practices that treat a symptom by giving you something that does the opposite of the symptom vs medical practices that treat a symptom by giving you a tiny amount of something that causes the same symptom. I’m wondering how treating the bone fits into this same schema.
It’s based upon the quack presumption that most ailments originate from an imbalance in the musculoskeletal system - how your bones are arranged. That’s largely abandoned by contemporary osteopathy
Embarrassingly, I would have had a hard time picking out the term "allopathy" --- and I'm married to one, an MD.
I’m the opposite, I would have recognized “osteopath” but thought it was a specialty, like neurologist.
Unfair to chiropractors: they make a lot of high quality ASMR videos!
The idea that DO's focus more on a holistical approach to the patient than MD's is DO propaganda. Is there actually evidence of this?
My (MD) doctor has given me plenty of holistic/non-drug advice:
- Change my diet (weight loss, blood pressure)
- Exercise more (general health, weight loss, blood pressure)
- For a foot pain issue, try different insoles
- For an old injury, a set of exercises that should help and a referral to a PT
- Stop taking allergy meds to sleep and use a better sleep routine (or at least switch to melatonin which has fewer side effects)
They did eventually prescribe blood pressure lowering medications, but they did it after trying the non-drug solutions (didn't work) and their purpose is to prevent problems in the future, so we can avoid "treating symptoms" later.
I'm my experience as a physician, I notice no effective difference in practice along equivalent physicians.
Most practice differences ("quality") seemed to be linked to quality of training (community program vs university program). DOs seem to be more likely to have trained at community programs.
I've never even heard of someone even discuss OMT in the hospital setting (where I work).
Agree with all. I've never heard a resident who attended a DO school even mention OMT/OMM or OPP. And now that DO-only residency programs are basically gone, it'll become even more rare (unless there's some specific demand for it that gets paid).
It's far more common for old-timers to say residents these days are lazy and useless because they're young and have work hour protections, not because they went to a less esteemed medical school. Which is...progress?
I think medical school has become fairly standardized, at least for the "facts" part of it.
It is true that if you work fewer hours per year, you get less experience. I.e., a resident who works 60 or 80 hours a week gets the same amount of experience under their belt in one year as a resident who only works 40 hours a week for a year and a half or two years.
I think we'll eventually see a push to extend residencies but make hours shorter. But it would have to come with a significant increase in the number of residency slots. We're heading toward shift work instead of call.
I’m genuinely curious whether there is any data determining how long a particular residency should be. My guess is no, it’s all just habit and history at this point, and everyone assumes it’s fine as long as their residents are passing the boards in sufficiently high numbers.
That's my impression. Residents do have case logs with minimum numbers of various types they need to graduate. Some of the required patients are somewhat rare so you go way over on regular patients trying to at least see a few of the rarities. And there isn't any simple way to make that easier given how much luck is involved.
This is the general clinical benefit to a larger, which tends to be university/academic, program. You will likely see a wider range of cases, and more rare and complex things. This matters for practice, I think.
A surgeon once told me:
"I can teach them how to do surgery in 3 years, but it takes more than 5 to determine when to do surgery"
("when" meaning which patients really would benefit from surgical treatment and which would not... this wisdom is the art and mastery for sure)
I actually suspect training duration has been well studied, since the effects of duty hours is well studied, but I'm not sure how that had affected anything.
I have doubts that assumption would hold up to real scrutiny. The quality of the hours of sleep deprived experience that I saw my friends in residency experience was probably fairly low compared to what they could learn when they were awake, alert, and well rested. Not to mention the quality of care, or lack there of, said sleep deprived residents were providing to their patients.
I think in general the research on duty hours shows positive benefit to residents (well being, burnout, etc), and has not shown any significant change in patient outcomes (good or bad).
Many studies and review articles are available on pubmed, most with this theme.
The duty hour era coincides with the modern EMR era so that may be a factor affecting (helping) outcomes across the increased shift handoffs as well.
Yes I'm sure at some point there're diminishing educational returns from cramming more hours in the same number of days; there's a physical limit--whether it's 40 or 60 or 80 hours a week or something else depends a lot on the individual's physical stamina. But subject of course to that the basic point holds, for any craft or endeavor where competency requires accumulating hours of hands-on experience.
There can be _negative_ returns too, where working more hours isn't just diminishing, but worse than not working those hours.
The curve may be different in training vs. learning(classroom) vs. working a job.
This is the US vs UK difference. A us 3 year residency is often 5 years or so to get equivalent credential in UK. Though their system is a little more varied whereas US is more standardized (as far as I understand things).
US: high cost medical education, shorter but intense residencies (60-80 hr weeks, though honestly 60 is probably more true more often than 80), higher eventually salary.
UK: less expensive education, longer more "relaxed" training (40-50 hr weeks), lower eventual salary.
Note that this comment is neither meant to defend nor disparge DOs vs MDs.
My wife and I have a DO as our primary care doctor. He's been a fine doctor, and I have no complaints, but osteopathy has been totally irrelevant to either of our care. He's never recommended OMT, he's prescribed drugs when anyone else would, and basically he's been identical to an MD.
Citing articles published in the Journal of Bodywork and Movement Therapies probably isn’t the best way to make an argument.
Arguing from academic prestige is a bit of a pet peeve of mine. A bad article can almost always be criticized substantively. Eg, this one uses a non-random sample - are most medical studies better than that? Idk but maybe you do
I read them all. I think the Italian one is fraudulent, and the rest are bad. Comment is above (or perhaps below).
Having worked in devices and Biopharma in many therapeutic areas, I’d recommend having more skepticism of literature from less prestigious journals and reserve judgment until finding are confirmed by multiple studies. It’s very difficult to discern which reason a study appeared in a less prestigious journal:
1. it’s a weak study that couldn’t get published anywhere else
2. The researcher lacks the credibility or network to get considered by better journals
3. It cuts against the orthodoxy and is being boxed out by the establishment.
My sense is that these are in order of frequency with #1 being most common
I agree that one should be more cautious approaching articles in lower-tiered journals. But I think that should lead to greater diligence reading the article, which can then inform a substantive critique - as opposed to just using the journal as the basis for the critique alone. If you don’t have sufficient background knowledge to assess the study’s quality, then I’m not sure you should be citing individual medical studies in the first place.
Completely agree. Disregarding a study a priori based on where it is published is lazy and sloppy.
I also agree that people shouldn’t cite a study if they have no ability to gauge its quality. The consequence is that most journalists shouldn’t cite studies at all. I’m not sure that’s optimal. There are too many issues deserving of coverage and needing journalists to bridge the expert-public divide and covering them in a non-technical way is unsatisfying. Ideally a journalist would consult with experts who explain the issue and literature. That has flaws though. It tends toward “both-sidesism” given that most important topics are subject to debate. And it’s not feasible to interview multiple experts for every story. Given that, I can’t think of a better rubric than to see journal prestige as a proxy for something being worth relying on.
That’s fair. I’d probably want some combination - ie, journalists with enough background knowledge to assess articles, then report on them
I’m a little gobsmacked by some of the hostility in these comments (canceling your subscription seems a bit much) because I found this an interesting explanation of something I was curious about and while it makes the case that DOs have value it isn’t exactly a strong rallying cry for them. As the parent of someone who aspires to attend medical school I found this post informative and useful. Thanks Ben.
Canceling your subscription over this would be ridiculous, but the article really misleadingly friendly to DO claims. The mainstream establishment treated them like quacks because they were quacks. Modern DO's are fine because they're basically MD's who are also taught some random quackery that most of them are smart enough to ignore.
I agree.
Journalism needs places for younger writers to hone their skills that aren’t centered on advocacy, as James Bennnet described about his time at NYT. Standards should be the same, but I try to be more constructive in my comments.
Speaking as an MD and practicing oncologist, I'm not surprised to hear future MD's higher performance on the pre-requisite coursework and testing required to enter medical training doesn't lead to eventual higher performance on patient outcomes like mortality and hospital re-admission. There is a huge space between knowing the best medical approach for a given problem and actually working with an individual patient to execute that approach successfully. Patients need to feel like they trust the recommendations they're receiving. They need to feel heard when they describe side effects from a treatment or the practical barriers they might face to the medically optimal approach. Allopathic training has been certainly been placing more emphasis on this aspect of care in the past 10-20 years. But an innate talent for making those connections with patients is under measured by GPA or MCAT scores. I haven't gone through osteopathic training, but I suspect that the foundational emphasis on holistic care probably leaves DOs somewhat better suited to that aspect of practicing medicine. The success with more granular outcomes like treating chronic pain seems consistent with that conclusion too. Approaching patients as individual people with unique circumstances, experiences, and needs is the bridge between knowing the best medical/pharmacologic strategy and actually achieve the best possible outcome. Even if MD's are slightly more likely to have mastered guidelines, they are not necessarily better suited for the day to day work of caring for patients.
I am skeptical Doc.
Of course we want MDs to be humanistic and caring etc. and that is not measured with GPA or MCATs, but then it isn't really measured by anything else at the time of admissions. I was on the admissions board at UNC School of Medicine when I was a medical student, and there were some efforts to find people with strong backgrounds suggesting more humanity, compassion, but it was not easy, being that all the prospective students knew they had to do all those volunteer things, work with handicapped children, or in some lab. I think that matters, but it is hard to tell who is the next Jonas Salk from the person just doing those things because they know it will help them get into school.
And GPA and MCATs do correlate with intelligence and/or hard work, and they are hard to game, though I support giving a boost to people from more challenged backgrounds (not race but challenges) than those likely to have had Kaplan and the best college preparations.
We got a lot of humanistic type training in med school before we actually hit the wards, and I feel it was a bit wasted, and would have been more useful later, because when we got to residency it disappeared and that is when the real need was to encourage a caring approach, but I found we got little in residency, where you had to keep your head down and work the work, and sleep through lectures.
I would be curious to know the number of researcher-weeks that goes into a story like this. It feels a bit of a weighty topic to handle with the usual rigor of "we should build more houses because people need to live somewhere".
With the benefit of five minutes' thought, I'm not sure I'm super comfortable with this.
The analysis and supporting evidence seemed incredibly weak- I was very confused halfway through the article until I scrolled back up and confirmed it wasn't a MY piece. The footnote opened my eyes though- once Ben said his interest was triggered by the fact that his sister was in a DO program, the level of analysis and particular perspective espoused by the article made more sense to me.
I didn't reach the footnote :/
When there is a big news story, there should be a thread to discuss it even if the article is just a stub. I suspect more of us want to discuss the presidential race than osteopathic medicine, but Ben wrote a good piece that deserves focused discussion and comment
I would rather hear about osteopathic medicine. What more is there to say about the race at this point, and what can any of us do about it? But from yesterday's article, I know that my demographics aren't typical of the SB readers beyond the education thing.
No, I'm right there with you. Some stuff will happen (the race will move, or it won't, or Biden will bail, or he won't), but right now all of the chattering seems well and sufficiently chattered.
I've been checking out the Ezraklein subreddit a lot lately (I don't think Ezra is involved in it at all. Also, there doesn't seem to be a Matt subreddit):
https://www.reddit.com/r/ezraklein/new/
I don't post there, but it's been a decent source for information and smart commentary, especially for reddit.
What a bad week for SB to go on vacation!
Interesting article. The point about the lower rates of opioid prescription by DOs highlights the role played by physicians in the opioid epidemic. While the Sacklers are great villains and bear a great deal of blame, it’s also true that doctors can’t pretend to be unwitting pawns. In fact, doctors had the agency, training, and intelligence to make better choices for their patients but in many cases failed to do so.
It'd be interesting to really get into the data there too. I work in the health care field, specifically focusing on fraud/waste/abuse, and what we see with doctors is a huge discrepancy between the regular levels of opioid prescribing and the outlier levels- there are some incredibly bad actors that are, essentially, just drug dealers. They're known in the community, they have connections with literal street level drug dealers, and they just issue enormous numbers of prescriptions. In my experience, they're exclusively MDs. So I'd be curious to see how much of the "rate of opioid prescribing by practice type" is driven by the extreme outliers and what the comparative rates between MDs and DOs would be if you eliminated those clearly criminal actors from the analysis.
I believe the Sacklers pushed the idea that their Oxy wasn’t addictive and for a while that was the consensus view.
There was a huge gap in opioid addiction research back then. "Thanks" to the Sacklers, that gap has since been significantly closed.
Thanks for this, Ben. I have been going to a DO for many years, but knew almost nothing about the differences and similarities with MDs. If I actually see the DO at an upcoming appointment, I will ask her about it. She is the most popular doc in the practice, though, so it is hard to get her for appointments that aren't scheduled well in advance.
The fact that the average grades and scores are slightly lower for DOs does not bother me a bit. A doctor who really listens to what I am saying is what I need. I also want a doctor who isn't over confident and is willing to look things up. Sometimes the doctors with the highest scores are unwilling to do either, especially at the beginning of their career.
Whatever philosophical basis osteopathy has, most students in the current are just aspiring physicians who fail to get into medical school. Like the Caribbean schools, it is an end run around the system. The average medical school admission has a 3.72 gpa and osteopathic 3.54. I could not find the data on MCATS, but presume similar.
Why the US, one of the richest countries in the world, doesn't not train enough MDs to fill all the available residency slots is unclear, but these spots are always filled preferentially with graduates of American medical schools, and the remaining slots by osteopathic graduates and foreign medical school graduates. It is not some philosophical difference that draws people to schools of osteopathy, as there might have been in the past, and they should just call the osteopathic schools, medical schools, and they can continue their traditions, but not pretend that people choose them because of that, as opposed to just failing to get into medical school.
I am not claiming that every student who cannot make the average grade point and MCAT score needed to get into a medical schools is incompetent and I have known osteopaths in my career as a physician who were great doctors, but we should be honest about why people choose that route, and we should train as many physicians as we need.
"Why the US, one of the richest countries in the world, doesn't not train enough MDs to fill all the available residency slots is unclear, but these spots are always filled preferentially with graduates of American medical schools, and the remaining slots by osteopathic graduates and foreign medical school graduates."
& "we should train as many physicians as we need."
If we moved the people going to DO schools because they couldn't get into MD schools into MD schools, what difference would it make?
Not a lot, but it would end the charade that there is really some difference between the two. Also half of the medical school slots are filled by foreign grads, either Americans going to lower quality less competitive schools or foreign grads, usually who are very high quality, but they are needed more in their own countries and come here b/c they make a lot more money and it is part of the brain drain from abroad.
“ and they should just call the osteopathic schools, medical schools,”
Are they not already? Like, if someone said they were going to medical school, one would not correct them and say “actually you are attending a college of osteopathic medicine” or something.
You do not get an MD but an OD from osteopathic schools, even though residencies accept both and most if not all states treat them the same. They have different accrediting and governing bodies, but if they are essentially producing the same practitioners, what is the purpose of two tracks. At one time osteopathy was more alternative like chiropractic but not anymore.
I’m aware of all that. I read the post. I’m just saying, at least colloquially if not technically, “med school” encompasses both. You would be an asshole to tell a DO student that they aren’t attending medical school.
Twenty years ago when I graduated college, plenty of students chose the Caribbean medical schools over DO schools largely because they wanted the MD on their ID badge. Now my understanding is that almost no one would choose Caribbean over DO.
At one point the DO school in Fort Worth was (for in-state students) by far the cheapest tuition in the nation. A lot easier to choose primary care when your debit is < $100k.