Is statistics taught in med school?

[Bernoull]

This is directed at med students or beyond.

I'm interested in some level of statistics training and public health. I can take a summer course dealing with one of the above but not both so I'm wondering which of the two will be taught in med school so that I'll forego that course for this summer. I start med school this fall.

For statistics, I'm basically interested in being able to better interpret research articles (i.e. are the conclusions supported by the data? is the statistical analysis crappy? are the results statistically significant? etc etc).

For public health, I'm interested in a course that introduces population level health/medicine.... macro-medicine if u will...

I'm also deciding b/t two graduate courses:

1. Intro to Public Health Epidemiology

2. Statistical Methods in Research: Obvious choice for my statistics interest and also include training in SPSS.

Also this should not be misconstrued as prestudying or gunning... I HAVE to take at least one course this summer to remain an active grad student and finish my research project. Therefore I want to take the course in something i'm truly interested in (since I'm done with all grad courses).


Thanks for any insights..

---

Members don't see this ad.

 

[Law2Doc]

This is directed at med students or beyond.

I'm interested in some level of statistics training and public health. I can take a summer course dealing with one of the above but not both so I'm wondering which of the two will be taught in med school so that I'll forego that course for this summer. I start med school this fall.

For statistics, I'm basically interested in being able to better interpret research articles (i.e. are the conclusions supported by the data? is the statistical analysis crappy? are the results statistically significant? etc etc).

For public health, I'm interested in a course that introduces population level health/medicine.... macro-medicine if u will...

I'm also deciding b/t two graduate courses:

1. Intro to Public Health Epidemiology

2. Statistical Methods in Research: Obvious choice for my statistics interest and also include training in SPSS.

Also this should not be misconstrued as prestudying or gunning... I HAVE to take at least one course this summer to remain an active grad student and finish my research project. Therefore I want to take the course in something i'm truly interested in (since I'm done with all grad courses).


Thanks for any insights..

Beyond a basic working knowledge (using a 4x4 table) of concepts like sensitivity, specificity, PPV, NPV, p values and the like (basically to the extent they appear on boards), you aren't going to learn statistics in med school. It's rarely a full blown course -- frequently they squeeze it into the brief coverage of epidemiology. If you plan to do more research I highly recommend taking stats before med school.

 

[Bernoull]

Beyond a basic working knowledge (using a 4x4 table) of concepts like sensitivity, specificity, PPV, NPV, p values and the like (basically to the extent they appear on boards), you aren't going to learn statistics in med school. It's rarely a full blown course -- frequently they squeeze it into the brief coverage of epidemiology. If you plan to do more research I highly recommend taking stats before med school.

Thanks for the response, I'm definitely inclined to do the stat course and yes I have research interests.

To clarify, epidemiology (or some other public health course) is taught somewhere during med school. Basically is one exposed to public health courses at some point?

 

[armybound]

It's my understanding that they are now required to teach you a certain amount of epidemiology as a medical student. We had a short crash-course in it that just finished.

Law2Doc is right on the money with the extent of what they taught us. Those topics were basically everything we learned.

it is a 2x2 table, though 🙂

 

[Bernoull]

Awesome, thanks to both of you for your response!!

 

[Law2Doc]

...
it is a 2x2 table, though 🙂

You are right -- 2x2 table, 4 boxes. My bad.

 

[krs881]

I haven't heard of any/many med schools that have more than the cursory review of both of these topics, but particularly statistics, like the other posters mentioned. I think that taking statistics will be more worthwhile as being in a classroom will be particularly helpful for learning, and in the long run it will be a very useful tool and background as a physician. The Epi intro is also very useful, particularly if you are interested in public health, but is something that you can sort "pick up" on your own more (not so much epi study design, but if it is really more an intro to issues in public health, you can read up on your own, attend talks and keep on top of news and literature). Not to downplay what a great course it could be, and very enriching, but if you can only choose one... Having good statistics instruction is something very useful that your med school most likely won't provide.

 

[Jihad]

Or how bout taking it easy for a while during potentially your last free summer? (With the knowledge that you can always learn what you want in the future from an MPH or independently in a lab)

 

[Rayhaan]

My perception of statistics is forever jaded. I donno if it is just my program or what, but they made me take 2 statistics course that were LEGIT the SAME THING. Is there anything beyond statistics but different formulas and tables for analyzing data?

 

[loveoforganic]

Knowing which analysis to use and why?

 

[Transmogrifier]

I've had the same amount of stats as the others have mentioned. A little bit on different study designs and odds ratio/relative risk and that's about it. Nothing that would make me feel remotely qualified to design and run my own research. There are lots of options for doing an MPH or other combined degree if you're into that. But med school itself is only designed to make you an informed consumer of the literature, not a researcher.

 

[Bernoull]

Or how bout taking it easy for a while during potentially your last free summer? (With the knowledge that you can always learn what you want in the future from an MPH or independently in a lab)

Trust me, chillin, travellin and doing nothin academic was my plan for this summer. However, I had to push my research project into May/June bcos of a crazy spring semester. To remain an active student and finish my research, I MUST enroll in at least one summer course. Initially, I wanted to go with Spanish or French but then I learned that the course has to be grad level which ruled out any foreign language so i'll take something else that I think will be useful for med school n beyond.

 

[Bernoull]

I've had the same amount of stats as the others have mentioned. A little bit on different study designs and odds ratio/relative risk and that's about it. Nothing that would make me feel remotely qualified to design and run my own research. There are lots of options for doing an MPH or other combined degree if you're into that. But med school itself is only designed to make you an informed consumer of the literature, not a researcher.

I was very close to doing an MPH over BME but I went with the latter. At this point I'm too old for another Masters...

For the most part, I just want to be able to critically read research articles and be able to evaluate whether sample size is adequate, conclusions are supported by the data, data interpretation is sound, experimental design is rigorous etc etc

 

[IH8ColdWeath3r]

Bernoull, why did you remove your We Be Steady Mobbin quote. Terrific song...who says you can't go to med school and listen to rap w/ vulgar lyrics (AKA keep a hard **** for your girlfriend to wobble on"


Keep it dirty like im lying in a Ditch,
Like Eli Im from New Orleans, Im a GIANT in the this Bitch,
Move the G and and an S, and put the I before the N,
Put the A in front of that and thats what I am till the end,
thats a SAINT mother****A simplify it for them,
where your funeral comes with a second line at the end,
But you wont be second linin with them because you'll be in the herse thats behind all of them,
I see your homies dressed in black like the finna ride,
But that fashion statement is tellin a lie,
Im Pelican Fly, MAC 11 Inside,
My bullets come with wings and your Cerebellum can Fly!!!!


WEEZY F-BABY.

 

[sunset823]

We had a 'full-blown' semester long class in statistics at my med school, but the material is very basic. If you have any stats background at all it will be a total blow-off. My masters degree was basically applied statistics, and I TAed a stats class. In the med school class I studied probably a grand total of 2 hours and got the highest grade in the class (unfortunately, that's the only class I can say that for).

I do like stats, and want to pursue health outcomes research so it is obviously important for me, and I do appreciate the fact that everyone is forced to learn it to interpret clinical studies and such, but if you plan to use it in your career, I would say you should take higher level courses, as part of an MPH degree or such.

 

[fahimaz7]

We had 3, 2 hour lectures in stats. Then we were tested.. It sucked.

 

[armybound]

I was very close to doing an MPH over BME but I went with the latter. At this point I'm too old for another Masters...

For the most part, I just want to be able to critically read research articles and be able to evaluate whether sample size is adequate, conclusions are supported by the data, data interpretation is sound, experimental design is rigorous etc etc

I would take the statistical methods course, then. I've heard of other courses in MPH programs that are about study design and such that would probably be useful.

The info we learned in med school definitely was not enough to help us understand the importance of analyzing a study before believing it all. I did learn a lot of that in undergrad/grad school in statistics courses, though.

 

[Bernoull]

Bernoull, why did you remove your We Be Steady Mobbin quote. Terrific song...who says you can't go to med school and listen to rap w/ vulgar lyrics (AKA keep a hard **** for your girlfriend to wobble on"


Keep it dirty like im lying in a Ditch,
Like Eli Im from New Orleans, Im a GIANT in the this Bitch,
Move the G and and an S, and put the I before the N,
Put the A in front of that and thats what I am till the end,
thats a SAINT mother****A simplify it for them,
where your funeral comes with a second line at the end,
But you wont be second linin with them because you'll be in the herse thats behind all of them,
I see your homies dressed in black like the finna ride,
But that fashion statement is tellin a lie,
Im Pelican Fly, MAC 11 Inside,
My bullets come with wings and your Cerebellum can Fly!!!!


WEEZY F-BABY.

I've been wondering the same, don't wanna blame mods 4 censorship but it somehow went up in to da ether even though I scrubbed it clean... could be a glitch also so I'll put it on again. Song is str8 up ill...

 

[45408]

Beyond a basic working knowledge (using a 4x4 table) of concepts like sensitivity, specificity, PPV, NPV, p values and the like (basically to the extent they appear on boards), you aren't going to learn statistics in med school. It's rarely a full blown course -- frequently they squeeze it into the brief coverage of epidemiology. If you plan to do more research I highly recommend taking stats before med school.

This is about how much we learned. They might have taught us a little more, but this is all anyone took away from it. That's all that's on Step 1 anyways.

 

[Charles_Carmichael]

From what I understand, based on talking to med students, if you're interested in doing research, etc, it might be better to take a strong stats course in undergrad since the med school one doesn't really go into depth.

I've taken an applied stats course that taught SAS programming in addition to going into detail about experimental design, etc, and it's easily been one of my favorite and most helpful classes that I've taken.

PS. I prefer SAS over SPSS. I mean, each comes with its own set of problems but I feel like SAS is a bit more powerful and easier to program. I don't know what the majority of labs use these days though. My own lab uses StatViewer pretty frequently, IIRC.

 

[Food]

I'm pretty sure MD/PhDs take it as part of their PhD coursework

 

[Bernoull]

I've dropped the Epi course and signed up for the STAT and I appreciate all the comments/advice.

 

[Narmerguy]

Even at some of the more research heavy schools these aren't offered more frequently?

 

[Bernoull]

Even at some of the more research heavy schools these aren't offered more frequently?

Also this whole discussion makes me wonder how well newly minted physicians have been prepared to be become critical consumers of literature. I'm assuming that new doctors will stay abreast with developments in their fields primarily through research/medical articles if so, clearly not all published material are created equal. Some are definitely crappier than others.

So how is one to know how much salt to take with such literature (a pinch, a boatload...) or is the expectation that the "quality control" done by Journal gatekeepers is sufficient.

 

[armybound]

The criticism of research should come during the publication process. If the research or analysis wasn't done well, it should be screened out.

Besides, most clinicians probably keep up with the highest impact-factor publications, which should be more strenuous in their screening.

 

[45408]

Also this whole discussion makes me wonder how well newly minted physicians have been prepared to be become critical consumers of literature. I'm assuming that new doctors will stay abreast with developments in their fields primarily through research/medical articles if so, clearly not all published material are created equal. Some are definitely crappier than others.

So how is one to know how much salt to take with such literature (a pinch, a boatload...) or is the expectation that the "quality control" done by Journal gatekeepers is sufficient.

That mostly happens in your residency, because it's incredibly difficult to evaluate medical literature from other specialties. Just one is usually enough. The issue is usually not the nitty gritty statistics, but rather the sample size, the methods, the authors, the location, etc. It's pretty interesting to listen to a department chairman or someone of equivalent standing evaluate an article, because they know all kinds of inside details about it that escape a lot of people at lower levels.

 

[TheFlashMD]

Song is str8 up ill...

Iiiiiill! Not Sick!

 

[ThaliaNox]

Like others above, our statistics was two lectures during epidemiology. Epidemiology was basically "how researches can mess with statistics to say whatever they want to." The whole exam was stupid trick questions. Better than 20% of the med school class didn't pass the class. Over half didn't pass the midterm. Overall, it is the most hated class taken during first year.

 

[canada1311]

I was very close to doing an MPH over BME but I went with the latter. At this point I'm too old for another Masters...

For the most part, I just want to be able to critically read research articles and be able to evaluate whether sample size is adequate, conclusions are supported by the data, data interpretation is sound, experimental design is rigorous etc etc

I just finished a Master's in Epidemiology, which involved taking a bunch of biostatistics courses too - I may be able to provide a bit of help. Although biostatistics courses are very useful, if you want to be able to critically appraise clinical research (and accomplish some of the things I quoted you writing above), I would advise taking an epidemiology course. Basic calculations like ORs, RRs, etc. are usually taught in epi courses too, while a statistics course may not give you a solid grounding in study design and controlling bias.

Furthermore, in order to get the grounding in biostatistics to truly understand analysis in clinical research, you would need to be quite familiar with regression and other multivariable methods of analysis, which usually takes multiple courses in order to understand and carry out. Basic statistical calculations, like ANOVA tables or sample size calculations (which are useless to calculate post-hoc anyway), will not do as much for you as a well-taught epi course focusing on epidemiological methods and critical appraisal.