How to Study for the USMLE Step 1

05.19.12 // Medicine

I’ve spent the majority of my “adult” life as a standardized test taker. As a resident physician, my skills are still developing. But as a student, as an aficionado of the multiple choice question, I’ve already reached (and probably lost) my peak abilities.

I am asked frequently about my thoughts on the Step exams, especially the USMLE Step 1. It is after all the single most important determining factor in what specialty a medical student can reasonably hope to enter and probably the easiest and most-used exclusion criteria used by program directors everywhere (with the likely addition of graduating year and US Allopathic schooling). I can hear your thoughts. They ask, how should I prepare? My answer is simple:

USMLE World.

Theoretically, any good question bank should do. In practice, UW is the best.

More so than First Aid, more so than Goljan, and absolutely more so than Doctors in Training, I believe the USMLE World question bank is far and away the most critical component of Step preparation (with a solid helping of Wikipedia). Here is why:

When you learn a fact in a book, you can congratulate yourself on adding a virtual index card into the Rolodex of your brain. However, you have not indexed this fact in a retrievable way. It isn’t necessary usable in a test-taking context. This is one of two reasons why many people who know things still do poorly on tests. They can talk about it, but they can’t apply it. You may know the right buzzwords but never seen them described in an exam question. When you do questions, you both learn facts and learn them in context. You learn them in a question format, instead of learning them in a paragraph and then struggling to integrate and apply it to a question. Your time is valuable, wouldn’t you rather kill two birds with one stone?

UW questions have extensive explanations and are essentially a textbook page unto themselves. You can learn what you need to know from UW. The only downside to UW is that is not “organized” like a book for those students without sufficient background in need of a stronger foundation (this is what a quick read-through of First Aid as fast as possible at the very beginning of your study process will give you: a reminder of what you’ve learned, a horrifying glimpse of how much you do not know). This ~~can be~~ is a strength for two reasons: 1) it’s [slightly] less boring 2) the USMLE is not divided into sections. You never know on what topic the next question will be. 3) Patchy random exposure to topics isn’t exactly “spaced repetition,” but it does work along those lines.

So read through First Aid in a week or less. Just get through it. Bring up all the hidden junk you memorized long ago, then go straight to questions. Questions, questions, questions. Use books only to memorize tables and diagrams or flesh out your knowledge when you find yourself stumped by concepts or totally out of your element. Many students spread themselves thin trying to get through multiple sources at the expense of not doing enough questions—this is a mistake. Focus and depth trump breadth, end stop. Go through UW and mark/flag all the questions you get wrong and all the questions you guess on. Then do the marked ones a second time. Unless you have a significant problem finishing exams on time, I believe in “tutor” mode: this mode allows you to learn the correct answer and read the explanation immediately after each question. The goal is here is to learn first (and simulate the test second). If you do NBME practice tests closer to test day, those will obviously be done in timed mode. Ultimately, this is a matter of preference.

UW also goes to the trouble of explaining not only why the right answer is right but also why the wrong answers are wrong. This is crucial. This means you learn how to answer similar questions correctly again as well as when the wrong choices would be correct. Several related facts in one caffeine-riddled swoop. These explanations are on the whole excellent. These exams no longer include the classic buzzwords found in review books; they describe those buzzwords and key phrases. UW employs this nuance well.

The second major reason people struggle with MCQ tests is the inability to “get in the head” of the question writer. There are individuals who seem to test better than they should. You could ask them to explain why they choose the correct answer and they typically cannot explain themselves. Or, if they know why, it often is not entirely based on their book-knowledge. They’re able to narrow it down to those same two choices as the next guy, but they pick the right one more regularly. These people are natural test-takers. Step 1 is a MCQ test, but its style is not identical to that of the SAT or MCAT. You need to do Step questions to know how to do Step questions. It’s like any other skill. Hone it.

Because you wouldn’t learn to play guitar by practicing the flute, would you? Sure, the flute is a musical instrument: your dexterity would improve, your knowledge of tempo, rhythm, and music itself would develop. But if you picked up a guitar, you would still suck. Don’t spend 10 hours a day for six weeks learning the flute, then pick up a guitar and hammer on a few chords for a week and call yourself a guitarist. Just play the damn guitar. Skip the middle man.

Here is my compilation of free USMLE questions, which can be helpful before you’re ready to shell out for a definitive resource. Here are some things to keep in mind while studying. And some more thoughts about how I read Step questions.

Studying for the NBME Pathology Shelf

05.24.10 // Medicine

Because Pathology is a cumulative all-encompassing subject, it makes sense that preparing for the Step 1—reading the First Aid—would be good preparation for Pathology Shelf. And while that would work, I don’t think that’s the best use of your time if you only have a few days to try to cram it all in. Pathology is cumulative, but the types of questions the National Board tends to ask demand a specific subset of knowledge: histology, gene mutations, responsible enzymes—these are the core of the pathology. Furthermore, reading a book (be it the First Aid, Goljan’s Rapid Review, or Robbins) is also a dangerous plan if you’re pressed for time. A) You probably can’t get through it. B) Knowing facts and applying them toward answering a question are separate steps. It’s not uncommon to need to see a question about a concept in order to the “bind” that knowledge appropriately.

A good Qbank (like USMLE World) works, but I think the very best way to review pathology is the Robbins and Cotran Review of Pathology, the question-book-companion of the big Robbins that many/most schools use. Benefits of this book:

Complete, system-organized review of pathology that covers all the important topics and factoids
Shelf-style questions and focus
Contains clear and concise but complete explanations—which is key. You’ll get a lot of questions wrong, but this review teaches you the salient distinctions quickly without being cryptic when you do.
Because it’s organized by system, you can tease apart related conditions and presentations. If you just do a blanket review, it’s hard to do this. Even if you use a Qbank instead, I would still recommend you do a run-through of the pathology questions by system first.

The link above is to the 3rd edition. I used a copy of the 2nd edition I bought at a local Half Price Books, and it certainly didn’t feel out of date. It’s also cheaper online. I think either one would work fine, though I’m sure there have been improvements made in the intervening five years. It’s a high quality resource; I only found two typos/mis-keyed answers in the entire book.

The point is this: there is so much material on Shelf and Step exams that literally anything you learn could be useful. Time and brain space are the limiting factors, so what you need is an efficient study aid. For the NBME Pathology Shelf, I had four days off to study. The Robbins question book is roughly 400 pages. I was able to do 100 pages a day and then follow it up with a few tables in the First Aid (important cytokines, for example), and that was 100% sufficient for Shelf purposes.

Sometimes when you do questions without having read a text first, the whole experience is just frustrating. Studying for the shelf is inherently painful, but this book really does it right.

The Mini Step 1

04.02.10 // Medicine

For around $35 a pop, your medical school can pay the NBME to let you and your classmates take the Mini Step 1 (“Comprehensive Basic Science Examination”), a 200 question multi-subject basic science test. The grading and its relationship to Step 1, according to the NBME:

The subject examination score is [was originally] scaled to a mean of 70 and a standard deviation of 8. A CBSE score of 70 is approximately equivalent to a score of 200 on the United States Medical Licensing Examination® (USMLE®) Step 1. The vast majority of scores range from 45 to 95, and although the scores have the “look and feel” of percent-correct scores, they are not…For this examination, the SEM [standard error of measurement] is approximately 3 points.

However, according the sample score report, the following is the scoring breakdown using the scores from 2008-11 (mean of 64; std 10 on the CBSE score scale):

CBSE Score	Percentile	Step 1 Equivalent
54 (or less)	18	160
62	50	180
70	77	200
78	92	220
86	98	240
94 (and above)	99	260

So it’s hard. Without doing any Step preparation (outside of attending to the usual coursework), I felt absolutely confident on only a handful of questions.

That said, and perhaps it’s just an extra year of medical school talking, the questions seem more doable and slightly less minutia-dependent than those found on the NBME Shelf subject exams. On this run, for example, the demanded anatomy is fairly basic—reserved for the highest yield topics like major artery and nerve distributions & common injuries and syndromes—especially compared to the anatomy shelf I “took” last year. While I assuredly failed this exam with soaring colors, it seems slightly less intimidating then before (edit: I did not fail. Goes to show that taking a lot of board-style questions in a row feels worse than it actually is). Still frightening, quiver-in-your-boots hard, but potentially doable. For most topics, it’s breadth, not depth. Only for key topics (basic metabolism, common bacterial and viral pathogens, big-name diseases like CF, CAD, MI, DM, Crohn’s, Addison’s, etc) is minute detail demanded.

For the question style itself, I was surprised overall with the frequency of first-order questions and the amount of useless writing. If you read Kaplan style questions too much, you see a lot of long vignettes with this scenario:

Long-winded clinical presentation of Strep throat (without identification). The question might ask, what should the patient’s physician ask before administering the therapy of choice? As we use Penicillin for Strep, we need to ask about a Penicillin allergy.

The ID of the bug is a first-order question. The drug of choice is a second-order question. The common adverse side effect of the drug of choice is yet a third-order question. On the Mini Step, most questions were actually first-order questions. Third order was much much rarer. Most frequently—and annoyingly—the long vignettes end with a diagnosis or ID, thereby negating the need to read the vignette at all! My advice: if you’re the type to run short on time, read the last sentence or two before reading the whole vignette. My other piece of advice is that you shouldn’t let Kaplan or other sample tests scare the crap out of you. They pick the most ridiculous questions they can find in order to frighten you into buying their product.

The Role of Ritual in Medical Training

09.27.09 // Medicine, Reading

While Final Exam, a memoir by transplant surgeon Dr. Pauline Chen, deals primarily with doctors’ troubled relationship with death and dying, I was struck most by an essay that deals directly with medical training’s preoccupation with protocol, algorithm, routine, and ritual. For Chen, rituals during her medical training were the foundation on which she built her persona and expertise as a doctor. Medicine is challenging, and ritual is the mechanism by which students—and later, physicians—break down complicated or otherwise difficult tasks in order to approach situations calmly, competently, and treat patients effectively. The harder the situation, the more essential it is to have a ritual to fall back on, as Chen describes how her routines helped steady her during an emotionally challenging pediatric transplant by allowing her to mindlessly do a procedure she had long since mastered.

My favorite ritual example in Final Exam, pre-surgical hand-washing, illustrates both its positive and negative effects. At first, the routine of scrubbing helped Chen ensure that she observed proper sterile technique; by following the ritual, she achieved technical competence and kept her patients safe. Furthermore, the mindless routine of the ritual was a form of calming meditation, a quiet break that helped separated her—emotionally and temporally—from both her clinical and surgical duties.

Years later (and after years of physical discomfort from an aggressive, skin-damaging style), Chen discovered that she was behind the times: she could achieve the same results by scrubbing for five minutes instead of ten and using a soft sponge instead of hard irritating bristles. The danger of ritual is that it leads doctors to routines that may reinforce bad habits, make it challenging to adapt to advances in patient care, or shield us from responding emotionally to our patients. Chen writes:

After nine years of clinical training, I found it hard to conceive of doing these clinical tasks any differently. In, I fact, I believed there was no other way, because these rituals were what assured the quality of my practice. They were what made me a good doctor.

This devotion to ritual is what helps training doctors learn the way of doing things correctly, even when the way is perhaps not the best way. While rituals may be a necessary first step in the learning process, the art of medicine lies not just in following the ritual effectively—but rather in when knowing to deviate. As Chen argues, a good surgeon doesn’t just know how to perform the right maneuvers; she knows how to fix the surprises that invariably pop up in the moment. It is when we fail to leave room for change in our devotion to ritual that our development as physicians stagnates, because while “they protect us from doing the wrong thing, their protective logic can shield us from fully shouldering responsibility.” (94) If we do everything correctly, the logic goes, then the negative consequences must be beyond our control.

Hand-washing is a relatively benign example because Chen was only hurting herself, but ritual pervades every aspect of medical training and practice, from memorizing the steps of the physical exam to sharing difficult news with a terminally-ill patient. The negative consequences of these rituals are only complicated by the role of the “informal curriculum” in medical training, the instruction that indoctrinates young doctors with the habits of their superiors. What happens when the rituals themselves are faulty? What happens when the carefully rehearsed patterns are themselves a source of doctor error?

In our first year training we learned physical exam techniques from both fourth year students and faculty preceptors. Both groups stressed the importance of learning the rituals of different exams, the routines on which to build our future competence, and so we robotically went through the motions, verbalizing our steps and performing the exam with techniques that only appeared analogous to the real thing. The emphasis was on “pretend” competency: the ability to look like a doctor on camera. This is not a shortcoming of any one school but rather an unfortunate result of the nation’s century-old curriculum design, one that places inordinate importance on some topics to the exclusion of others (oblivious of clinical importance). Soon, undoubtedly and embarrassingly, our class will have to relearn how to perform exam techniques in order to actually evaluate our patients. Right now, the sham ritual is all we have.

As Chen says, the clinical aphorism is “see one, do one,” which means that as doctors we train to master the mistakes of our mentors. Our early success will depend directly on how well we copy our teachers (because it is our teachers, with their idiosyncrasies, that evaluate us). And while rituals may be a useful crutch in the short term, it’s not hard to imagine the future consequences. When our patient interactions become ritualized—each sentence just another item on a mental checklist—our patients will be reduced to a given number of steps. The more times we use our algorithms, the easier it will be to categorize our patients as cases, people as diseases, and conversation as a technical skill—instead of an intrinsic part of what makes us human. This reduction is the process of dehumanization that comes with the epidemic of physician burn-out, naked cynicism, and is a chief component of patient dissatisfaction. It is a mainstay of a generation of medicine we should hope to overcome.

Free USMLE Step 1 Questions

08.30.09 // Medicine

No matter how much money you spend on books, every medical student needs to do a ton of practice questions for the USMLE Step 1. Questions are an excellent way to learn the useful tip-offs and keywords, and—depending on the source—get a better feel for the board format. They’re also a form of active learning, unlike trying to self-induce a coma with the universally-utilized First Aid for the USMLE Step 1. I believe USMLEWorld is the best question bank out there—despite its draconian efforts to prevent IP theft—and there is no free source out there that matches it (especially for the final marathon push before the big day). That said, there are other ways to study, especially during the basic science years.

For question books, post-Step MS3s and your local used book store are always good resources to buy study materials on the cheap. But free is better, and the internet is undeniably convenient and portable. I scoured the web to find free question banks online (updated June 2019):

The NBME offers its own small set of free practice materials for the Steps 1, 2, and 3. You see the most recent set here, which includes a browser-based software version that mirrors the actual program Step uses (Fred V2), a tutorial, and 100+ question practice test. A must do. A pdf file is also available from the above link, which contains the same questions for your offline viewing pleasure. I’ve written answers/explanations to the past several sets, which are linked here.
Lecturio has made their 2200 question USMLE question bank completely free after registering for a free account. If you’re interested in buying their full-featured video lecture/qbank product, you can get a 25% discount with code hpG6C.
Pastest is a 2300-question commercial qbank that has a 48-hour free trial.
WikiDoc has a 696 question board-style USMLE Step 1 qbank. Robust, very nice. Qbank appearance approximates the USMLE Fred software. Totally free but requires a login.
MedBullets has a 1000+ question Step 1 qbank in clinical vignette style. Registration required, pretty robust software (tutor mode, tracks prior questions, etc).
USMLE Sapphire is a free online qbank (registration required), currently with 520 questions. Style is more concise/abbreviated/clinical-flashcardy than the real clinical vignettes and the software handles the explanations in an annoying way, but the site keeps your test history, lets you review prior answers (no tutor mode), and pick questions based on subject and body system. Some of the bits I saw were a bit obscure, particularly given its size.
Osmosis is a completely free big (>5000 question) qbank and video product. Many questions are more on the Step 2 side of things, but an impressive collection nonetheless.
~~USMLEQuickPrep~~ is a large (~4500 questions) and entirely free qbank. It’s the largest and most exhaustive free source out there. The questions are a mixed bag, and not all are in Step-style, but most have explanations, the site isn’t too clunky, and it certainly stands out for its sheer volume. [site is down again]
~~Lippincott’s 350-Question Practice Test for USMLE Step 1 is solid, but you must register (for free) before using it.~~ [now defunct]
MedMaster (makers of the “made ridiculously simple” series) has a USMLE Step 1 qbank (among others). The questions are not step-style but rather content review. It’s a good foundational accompaniment to book learning, as it clearly highlights key facts and distinctions that are crucial for Step 1, but it does not prepare you for the exam proper. There are also no puns or goofy diagrams like the book series.
Test Prep Review has a USMLE practice self-assessment section. There are 20 modules of 20 questions for 400 questions. They’re mainly fact-recall and not vignette-based, but it’s easy to use and accessible.
Wiki Test Prep [now defunct, but with questions available as a pdf for download] is was a student-written qbank with over 900 questions with explanations. The site is great, and you can browse questions by keyword, flag questions, and create your own tests. It also lets you know what percentage of students answer the question correctly, which is interesting. The questions are in clinical-vignette board format.
MDLexicon has a bunch of vignette questions organized by category, it’s hard to tell exactly how many. The site design is bit odd, but it works.
4tests.com hosts an old 60-question Kaplan diagnostic exam. Answers can be exposed during the test if desired and do contain explanations. (Mom MD also has the identical sampler, only organized in six 10-question pages with answers directly below questions)
ValueMD has a large question bank divided up by subject. The site also requires a free registration. The questions are straightforward fact-recall type and the site itself is clunky and hideous, but it’s still decent review.
Kaplan lets you try one 48-question section for free after signing up.
Learntheheart.com has 50 cardiology USMLE Step 1 questions, with plans to add more.

Enjoy. I can’t vouch for the quality of these resources, but WikiDoc, Lecturio, MedBullets, Sapphire, and Wiki Test Prep together are about 3500 questions, bigger than UW (though assuredly with lots of overlap between sources). Osmosis, a new free player, adds in a lot as well. Add in the past few years of official practice questions (the “Free 150”) and you’ve got even more.

There are also several free questions sources for the MRCP (The UK’s version of the Step), for which there is considerable overlap:

Confidence (3500 questions)
ReviseMRCP
MRCP Study

(For more information on how I personally would recommend studying, feel free to peruse my post: How to Approach the USMLE Step 1. You can also find my compilation of free study resources for the basic sciences here.)

Studying for the Basic Sciences and (preclinical) NBME Shelf Exams

05.22.09 // Medicine

Some thoughts for intrepid and probably antsy first-years as they stumble upon this page on their search for the truth. Here are some tidbits about the Anatomy, Physiology, Biochemistry, and Microbiology shelves.

Anatomy:

Anatomy, if learned well over the year, is doable. It is crammable if (and only if) you’ve actually paid attention and learned everything once. This is not the time to learn anything for the first time. Even with a good background, the anatomy shelf tests a ton of material. Pay special attention to high yield parts—things that are often injured or easily framed in a clinical vignette. While everything is there, it’s not there equally often.

For example, collateral circulation is a big favorite for several systems. You’ll see questions that involve arterial transections from stab wounds as well as a occluded arteries from cholesterol plaques. You need to know what other arteries can supply these areas–they’ll be there in spades. Focus not just on knowing body parts but also on 3D relationships between body parts, because they’ll often ask for one part in relation to another (the structure immediately lateral to the X is….).

Lastly, pay attention to common injuries and conditions. If someone falls on an outstretched hand, what bone is probably broken? Scaphoid. What bone is most likely dislocated? Lunate. So on and so forth.

Favorite books: Moore for the textbook (though your school syllabus may be enough). Rohen for the atlas, end stop. (I prefer real photos to Netter drawings, though the combination is synergistic; Netter shows you the ideal relationships; Rohen shows you how to actually identify structures on your practicals).

If you want a combination of Netter-style illustrations mixed with a healthier amount of explanatory text and some nice tables, consider Thieme’s Atlas of Anatomy (it’s excellent; the downside is that your school probably isn’t using it as the official text).

Physiology:

Memorize lab values, especially if you’re the type who doesn’t finish tests quickly (this is also a time-saver on the Step 1). More so than other shelves, Physiology clinical vignettes often include lab values that are actually key to understanding what’s wrong (not the case for Microbiology, for example, where lab values are almost always superfluous). Because the test demands more problem solving/reasoning over rote memorization, the physio test is also one of the more tiring, in my humble opinion.

Almost all physio questions require knowledge of the various trends and formulas; they usually demand that you know why something happened or what happens next. For example, you need to know the body’s compensatory mechanisms to keep fluid volume stable, and how these are used to correct for hemorrhage, dehydration, diabetes insipidus, etc etc etc. If something going up makes something else go up or down, then there will be a question about it.

Also, know your renal.

Favorite book: Costanzo, easily. She also wrote the BRS Physiology Review, which I’ve heard good things about.

Biochemistry:

Buy a set of notecards and memorize them. Know all of the incredibly rare diseases and what enzyme is deficient in the pathway. Not much to say here: it’s a hard test. The things that are most often covered in detail in your biochemistry class are the things you must know in greater detail. So, you have to know metabolism incredibly well. Intimate knowledge of pathways like glycolysis, glycogenolysis, and gluconeogenesis is very high yield. For less important topics, like microtubules, basic gist-level knowledge is sufficient.

Favorite book: I used the course syllabi supplemented with the wordy but quite good Lange flashcards. Lippincott is the common favorite. Goljan’s Rapid Review covers the topic nicely as well and many people end up buying it anyway. I’ve also heard pretty good things about First Aid for the Basic Sciences.

Microbiology:

The most crammable exam. First, know basic virology and bacteriology. How they work, change, evolve etc. You definitely must learn about the multiple ways bacteria become virulent and gain new genetic material. A lot of the test comes directly from the sort of stuff everyone overlooks when they start memorizing pathogens.

Second, for clinical vignettes, a good notecard set is always helpful. For most pathogens, there are a handful of key words that will set you off to the question—they usually ask for either an identification or a secondary fact about it. Everything else in the question is just time-wasting fluff. Memorize the key relationship, and you will be fine on the majority of the vignettes. Will you really understand what’s going on? Probably not, but these questions are often designed to test your knowledge of these stereotypical cases, not whether or not you’ll actually be a competent physician.

For example, H. pylori, which causes gastric ulcers produces urease. If you see a patient with an ulcer, the answer is probably either the ID or the enzyme. Aspergillus makes a “fungus ball.” Coccioides forms a spherule in the body. People who get a non-healing lesion after being pricked by a rose thorn have an infection from Sporothrix. If a bacteria is coagulase positive, it’s Staph aureus. It really is that superficial. Key word -> answer. The difficulty comes from information overload. The more rare the disease, the more likely the question will be an easy form of memory recall (e.g. fever that goes up and down (undulates) is Brucellosis). For more common bugs like Staph and Strep and the very well-known classical diseases (Malaria, TB, etc)—do a better job nailing down greater details.

Favorite book: Clinical Microbiology Made Ridiculously Simple, hands down. For rapid bug review, try the BRS flash cards. Fast and painless and better than making your own.

And that’s it. You’ll be fine.

(Keep in mind: Questions are good. Do questions. If you have cash to burn, USMLE World has plenty of them. You can also find my compilation of free study resources for the basic sciences here. You can find my views on the pathology shelf here.)

NBME Shelf Exam scores, with a grain of salt

04.21.09 // Medicine

The NBME Shelf exams are enjoyable standardized tests that every first year looks forward to with almost unbearable glee. Each tests a single subject (“Anatomy”) and is (for the preclinical years) made up from the old or junior varsity questions from the USMLE Step 1, a test that makes the MCAT look like the GRE and the SAT look like building with Lincoln logs.

Some schools force their students to take a variety of Shelf exams (spending/wasting $30 a pop) to help measure how well their students have mastered the material (AKA how they are doing compared to their national counterparts). What is a bit amusing and misleading about the whole ordeal is that the national norms are probably a big crock.

Different schools use the “shelves” differently. Some use them as a just-for-fun intellectual exercise, others as extra-credit, and still others as a true final exam. Don’t get me wrong, it’s not a bad thing to get some USMLE Step 1 experience, but it’s highly dependent on the environment: if you take five shelf exams in a single week, you are clearly not going to be prepared or even particularly focused. If it’s your final exam, you are going to do your best to rock it.

So if the national average is computed from all of these groups together, then it’s going to have a huge unseen left tail: if people are taking the exam who don’t care how they perform, they’re going to be dragging the average down from where it would otherwise be. So while the test is technically normalized, it’s not the same normal as a regular standardized test: Unlike the MCAT, not every student has something riding on the exam. I personally knew people who filled out all C’s on an exam that was for extra-credit only.

While your school receives the group’s average and your grade relative to your test group (classmates), the theoretically more interesting numbers a student receives are the grade based on the national average and corresponding percentile. I’m curious as to how far off the scores really are. If all those people who weren’t making a good faith effort actually tried (as they do on the USMLE Steps 1, 2, 3), then I’d wager it’d be a different ball game. It’s essentially an unstandardized standardized test.

Further reading: How NBME Shelf Scores Work

Lip service surveys

03.17.09 // Medicine, Miscellany

The world is full of surveys: surveys for free meals at TGI Friday’s, surveys for news polls, and at school, surveys for curricular reform:

“In order to improve this course for next year, we would appreciate it very much if you would take a few minutes and fill out this evaluation form.”

And the idea behind a survey is a good (nay, excellent) one: to gather feedback and ostensibly make changes and corrections based on it. The issue is in survey construction and follow-through. The usual survey has a variety of broadly worded statements with answer choices 1-5, 1 being “strongly agree” and 5 being “strongly disagree.” There will usually be a text-box for general comments at the end. You take this survey and your answers disappear into the depths of the internet never to be heard from again.

But from the beginning, the idea that you can sum up whether something works effectively or not based on a numerical average is a kludge. Furthermore, even if an average of 4 does approximate satisfaction, that doesn’t mean there aren’t better ways to do things. It’s an understandable shorthand, but anyone hoping that it’s sufficient to understand reception is fooling themselves. If people’s responses show that weekly quizzes are on the whole useful, that doesn’t mean they wouldn’t prefer or think it better if they were biweekly, on Mondays, on Fridays, longer, shorter, or anything else. If people say dividing the year into 4 chunks is no good, it doesn’t mean 7 would be better. A number is all well and good, but at the end of the day, how someone feels isn’t the crux: it’s why they feel the way they do.

In order for a survey to be effective, it has to take time. Each question needs to have its own comment box. Then, someone needs to go through those comment boxes and compile all of the suggestions and problems. Take the suggestions and complaints, then formulate new courses of action. Then, before implementing them, offer them anew in a survey: What do you think about these choices? Do they sound good? How good? Better then before? If not, why not? If that takes too much time to do, have students volunteer to do the grunt work. They’ll put in on their CVs, the administrators can continue doing whatever it is that administrators do, and everyone is happy. This is also how you make changes quickly. It doesn’t need to take years.

People tend to make incremental changes to the status quo. It’s hard to make drastic changes, especially if those changes reverse your hard work or go against your own inclinations; it’s even harder to come up with these changes yourself when necessary. This difficulty then breeds the stagnation that allows bad systems to continue even when their obsolescence is practically taken for granted. And yet, this is how you get curricular form with a stethoscope on the heartbeat of a student body.

Sometimes things don’t work—but if a goal is truly to teach a subject effectively, then no one can tell you better what does and does not work than students. This is how you don’t spin your wheels around a problem, making arbitrary changes. You need to ask for feedback, but more importantly, you need to be willing to listen to it.

Anatomy of an NBME Shelf Exam

03.04.09 // Medicine

The NBME offers comprehensive subject exams, ostensibly to torture students and devour whatever scraps of self-worth they have left. The typical question format:

A X-year-old [type of person] reports to the doctor/ER with a X-hour/day/week/month/year history of not-feeling-so-hot. Upon examination, distracting details. Irrelevant information. Single key relationship. More words to make the test take longer. Talkie-talkie. What is the likely cause of this you-should-have-learned-in-your-class-and-probably-did-but-maybe-you-didn’t-who-knows-there-are-so-many-questions-on-this-test-it’s-all-a-blur person’s problem?

Rinse and repeat for three hours. Then do it for it every class you take. The joys of biochemistry could never be fully appreciated until they were compiled in such a form.

In-ear, shmin-ear

02.22.09 // Medicine

Every few months, I see some news report on the revelation that listening to loud music can cause hearing loss. Yesterday, Time online posted “iPod Safety: Preventing Hearing Loss in Teens,” the latest in this series of mediocre ear-science. I’ll admit that in this case the facts are accurate, it reminds me of a lot of stories that are less so.

The point, which is true, is that listening to loud noises of any kind tends to be bad for the ears. The louder the noise, the less time it takes to cause damage. Prolonged exposure to loud noise leads to both tinnitus (ear-ringing, which sucks) and sensorineural hearing loss (which sucks and is irreversible). This is actually a serious problem, and it’s caused the kind of hearing loss in teens that used to be reserved for old-time factory workers. My beef is that there is one finding several years ago that has given rise to a huge misconception:

In-ear head phones generate more sound pressure at a given volume setting than over-the-ear counterparts [source].

This is logical, given that in-ear headphones are actually in your ear, literally closer to your tympanic membrane, which transmits the physical pressure waves to your middle ear. Because iPods are generally used with in-ear headphones, some news outlets and people came away with the idea that in-ear headphones are automagically more dangerous—which is hogwash.

This is silly because pressure and volume are essentially the same thing. When the in-ear headphones produce more “volume” at a given setting, the user actually hears the music louder. If I were to switch from over-the-ear to in-ear phones, chances are I’d adjust the volume accordingly. The fact that earbuds can pump out more decibels in and of itself is meaningless. Admittedly, there has been some work that has shown that some earbuds don’t cancel external sound all that well and therefore might lead to higher volumes when used, but this varies wildly between brands. All that means is that the government should subsidize some new Bose headphones for people who work in loud places, because good sound-canceling headphones are the only ones that eliminate this problem effectively. Being closer to the ear is not an inherent problem unless the volume isn’t adjusted accordingly. This is not an unnoticeable danger increase.

What studies have shown is that individuals have a preferred ambient listening volume. Some very angry teenagers who like thrash metal tend to like to blow a hole out of their eardrum, but the rest of us tend to fall somewhere on a decent curve. What matters is what relative volume we prefer, not what method we use to get there. When people taken off the street were tested for average listening volume, the data reflect this reality: the biggest problem is background noise. We tend to like our music somewhere around 60dB. If the ambient noise is 20dB, many people will turn up the volume to 80dB. If you correct for background noise, preferred volume is nearly constant. So when people listen to their iPod somewhere loud (on an airplane or the subway), they’re probably doing a lot more damage than if they’re at home. It really is that simple.

The idea that in-ear headphones are actually worse for you is based on this distortion. They’re not; your preferences and habitat may be.