Category Archives: USMLE

The Paper Jungle and other Medical Questions from Real Readers

A reader writes: “Is there a solution to inefficient paperwork? With such great technological advancement, do you foresee any computerized forms of paperwork to make it less inefficient?”

Sure. We have a great electronic record system at one of our departments (the T-System) which makes documentation and order writing a lot easier than previously. The problem is the temptation to document more simply because we can. Not to mention that so long as there are bureaucrats, there will always be new ways to waste time and, unless it is aggresively cut back like creeping kudzu, paperwork will always spread into every available niche.

In other words, despite great technological advances in information technology, there has been no decrease in the amount of paperwork involved in medicine. The converse is true unfortunately as every year seems to find some new asinine JHACO compliance chore sucking up somebody’s otherwise valuable time. Now, to be fair, most of this stuff is what I call “automatic paperwork” meaning that they put a form in front of you, you sign it, and it vanishes into the mouldering realm of medical records where the evidence of your compliance with the hospital’s ass-wiping policy will be entombed forever. It’s not too bad I suppose but it is somewhat annoying and, just as a vigorous mongol warrior may easily ride down a few peasants here and there until you throw enough of them into his path to seriously impede his attack, so too is the modern doctor’s attempt to secure the medical equivalent of the goats and slave girls severely hampered by the reams of innocuous paperwork between him and his objective.

This is not even taking into account the tangled labyrinth of forms, regulations, and coding required to be compensated for your work in a system where nobody wants to pay anybody for anything and most of the bureaucracy is actively engaged in either deflecting invoices and delaying payments or in trying to get somebody else who isn’t involved, the so-called “third party,” to pay.

It all stems from a complete lack of trust at all levels of both medicine and society in general, a lack of trust engendered or at least encouraged by the legal profession against which most paperwork is directed. The hospital doesn’t trust its employees therefore they are required to complete endless forms foreswearing infractions of things that used to be common sense or to give fealty to their overlords in the diversity theocracy. Doctors don’t trust their patients not to sue so every discharge instruction includes the usual reams of boilerplate instructing the patient on differentiating their ass from a hole in the ground. Doctors don’t even trust other doctors and document in an attempt to drag as many people into the stew as possible under the theory that it’s a lot more fun to fry if you do it as a group. The hospital takes the opposite approach and, to minimize their liability, structure their paperwork to identify the one guy who can take the fall for everybody.

Insurance companies and the government don’t trust anybody on general principle.

The two things that have surprised me the most about medicine? Number one is the severity and number of illnesses people can collect and still grimly cling to their mortal coil. Number two is the absolutely astounding volume of paper generated in a hospital, the great bulk of which is completely useless except that the bureaucrats believe it to have talismanic powers against the legal vampires. That and it supplies work to countless people employed in tending it.

But as far as technology simplifying things, we have a long way to go. I still get a kick when patients from out of town or who frequent some other hospital airily dismiss my attempts to garner a past medical history or a medication list with a casual, “Oh, it’s all on the computer.”

Lady, at 3AM your primary care doctor’s electronic medical records might as well be on Neptune for all the good they can do you.

What do you think about the USMLE Step 3?

I took Step 3 today, or rather finished it, because it is a two-day test. I won’t elaborate on the questions so as not to subvert the exam but I will say that it is obviously slanted towards both primary care and, surprisingly enough, Emergency Medicine. It just seems to me that a surgery resident would have to study for it harder than a Family Medicine resident because I have a fairly good idea that most surgery residents rapidly forget all the primary care they ever knew. Sure, they know how to treat a lot of things tbut they may have forgotten how to handle some of the routine cases that are second nature to a Family Medicine intern who has the various preventative medicine guidlelines beaten into his head every day.

Confess. How many of you surgery residents know what to do with a pap smear?

An unusual feature for those of you working your way up the Steps are the interactive cases, nine of them, on the second day of the exam. Definitely do the practice cases provided by the USMLE before taking the real test because the interface, while easy to use, is not intuitive and you need to know how to handle the mechanics of the computer simulation. If you’re going to screw it up its best to do it honestly and not because you clicked the wrong button at the wrong time.

Since the practice cases are available publicly, I don’t think I’m giving anything away by describing how this section of the test, a section that I rather enjoyed, works. The first thing is to relax. You have been doing this since third year and the only difference now is that you don’t actually have a patient in front of you. After being presented with the history, you are free to write orders, ask for physical exam components as needed by system (which are given to you for the asking), and transfer the patient to any area of the simulated medical center in which you are working. A case might begin in the clinic, for example, and if by history and physical exam you realize that your patient is having a heart attack it’s time to transfer them to the Emergency Department where you continue your management. You manage the patient by writing orders asking for labs, studies, and consults. Some cases seemed (seemed) pretty simple and required simple management with some discharge instruction here and there to stop smoking and lose weight but on one I transferred the patient from the clinic to the Emergency Department to the ICU before my time was up.

You have 20 minutes of “real time” per case plus five minutes at the end to finalize orders and provide a diagnosis. The diagnosis, as I understand from the tutorial, is not part of your grade but can serve to clarify what in the hell you were thinking. The case may span more than 20 minutes of “simulation time.” Within the limits of the “real time” (that is, the limit allowed by the test) you can advance the clock in whatever increment you desire. Sometimes you write your orders and then, with nothing to do, you need to advance the clock to a time when the next lab or study results are available. At other times you advance it to a follow-up appointment that may be the next day or later as you would do in a real clinic when presented with a non-emergent case. One of the practice cases, for example, is a guy with Giardia. I sent him home with lomotil pending the results of his giardia antigen assay and then on the follow-up visit, when it was positive, started him on Metronidazole. I guess this was the correct “play” because on advancing the clock the computer told me that he was feeling better and then abruptly ended the case.

That’s the disconcerting thing, however. You can be tooling along writing orders, managing like a big dog, kicking medical ass and taking health care names when the case, with time still on the clock, will suddenly and without warning come to an end. Naturally this can come as a shock because there may be no feedback. Did I cure him? Did I screw it up? Supposedly the case can end early if you handle it well but some hint would be appreciated. On one case I had no idea what was wrong with the patient and after shotgunning a whole bunch of labs and venturing some treatments just kept advancing the clock until I either killed the motherfucker or the computer decided I was an idiot and put me out of my miseries.

Without boring you with the mechanics, the program recognizes a couple thousand common orders for common studies, medications, and interventions. (I wrote an order to “intubate” and, mirabile dictu, a little window popped up saying that the patient was intubated and on the vent with the appropriate settings without any complications.) It is not necessary to know dosages of drugs but only the name (trade or generic) and the route of administration (PO, IM, IV, etc.). You don’t have to get that detailed with the orders so relax and don’t worry about the esoterica like vent settings and specifying everything you need for an Incision and Debridement. You do, however, need to order the stuff that most Emergency Medicine residents take for granted like intravenous access and cardiac monitors. In the computer simulation world, the nurses still dress like porn stars and don’t do a thing until explicitly told to do so by a doctor.

I guess the object is to show that you know how to manage efficiently and economically. Apparently some of the cases seem simple because they are simple and it is not necessary to admit every upper respiratory tract infection to the ICU. I imagine whoever or whatever grades the test takes points off for over-reacting, maybe putting Emergency Medicine residents at a disadvantage because I often found myself looking for the button labled “Indiscriminantly CT Everything.”

Other than that, all I can say is that like Step 1 and Step 2, knowing “what is the next step” is big. That is, knowing the diagnosis is not as important as knowing what to do about it and when. I don’t think anything came out of left field. Even the things I didn’t know I knew that I should have known. In other words, I knew what the question meant and what I was supposed to know even if I couldn’t exactly remember the details. Some of the answers seemed so obvious that I marked them for review and came back to them just to make sure they weren’t trick questions. You find yourself doubting if those massive ST elevations in the anterior leads are really pertinent to the answer and if the test is really asking you for some subtle, psychosocial management strategy.

Many of the questions are on ethics. Again, I will not subvert the test by giving you specific examples but I think we all know that “Taser the Patient” is probably an answer you can eliminate right off the bat. That and advising your Hispanic patient that “she needs to learn English.” I wasn’t so comfortable taking the test that I could afford to screw around so I voted the straight diversity party line like a good boy.

USMLE Step 2 Clinical Skills

Highway Robbery

Might as well come out and say it. The Step 2 Clinical Skills test is a swindle foisted on medical students by bureaucrats with too much time on their hands and not enough to keep them occupied. It had its origin in the the clinical skills test administered to foreign medical graduates to ascertain their level of English proficiency and their comfort level with Western clinical skills under the theory that advocating consulting the entrails of a lizard in broken English would preclude you from practicing in the United States.

Since money could be made by making American medical graduates take the test, beauracrats hired other beauracrats to prove that we were not being taught how to interact with patients during our four years of medical school. Despite strong resistance from the medical community and medical students, money was spread around and the result was an expensive solution to a non-existent problem.

Step 2 CS is a day-long standardized patient exercise which tests your ability to take a history, do a physical exam, and write a note with an assessment and a plan. Nothing to it, right?

Right. The first time pass rate is in the mid 90 percent for American medical school graduates and very people study for it at all. So you see, and try to follow me here, if there really is a problem then people would be failing this thing left and right leading to the kind of studying we normally only see for the clinical knowledge portion of Step 2. Which does not happen. Medical schools do a good job of teaching you clinical skills. There is no problem.

Not to mention that Step 2 CS is pass/fail so it can’t even be used as a measure of anything other than your having at least the same skills as some guy from Bolivia who’s father paid the Minister of Health to get him into medical school.

But I digress. Step 2 CS is a done deal so you are going to have to take it. It was about a thousand bucks in 2005 not counting transportation and accommodation at one of the only twelve testing centers in the United States. (I took the test in Houston.)

As I mentioned, Step 2 CS is a day-long standardized patient exercise. For those of you who are not familiar with them, “standardized patients” are actors trained to pretend that they have various clinical conditions. Many medical schools use them to introduce students to the history and physical exam.

The Step 2 CS site is set up ostensibly like a real outpatient clinic. During the day you will see approximatley 12 standardized patients in a round-robin fashion moving from room to room rotating patients with your fellow examinees. Each clinical encounter consists of up to 15 minutes for a history and physical examination and then up to ten minutes to write your note. You can leave the patient’s room before your history and physical exam time is up and use this extra time for writing your note. Once you leave the room however you may not reenter.

Posted on the door of each room is a board with the chief complaint (as would be elicited by a nurse) the patient’s demographic information, and his vital signs. At a signal from the proctors, you knock on the door, enter, and begin the song and dance.

I will explain the history and physical exam in a later post (for those of you who are not in medical school yet, of course) but suffice to say that one enters the room, exchanges the usual pleasantries, elicits the Chief Complaint (CC), gets the History of Present Illness (HPI), and performs a physical exam.

To assist you the actor playing your patient may have moulage (makeup simulating an injury) as well as cards telling you the result of invasive exams like the digital rectal exam which you will not do on standardized patients.

(Some view the Step 2 CS as a digital rectal exam done on medical students.)

Now, here is the key to this portion of the test. Since this is a simulated real world clinic, you should do a focused HPI and physical exam. The patients will all have classic presentations of common clinical problems. I don’t think I will be violating the NBME’s non-disclosure agreement if I tell you that you might get a patient who’s chief complaint is chest pain and shortness of breath with exertion. In this case a complete neurological exam is not necessary, will gain you no extra points, and will eat into your time. There are no zebras on Step 2 CS.

Once you are done you exit the room and start writing your note which is essentially a SOAP note. You can write this on a form that is provided or enter it in a computer. The form has a space for the history, the results of the physical exam, your assessment and your plan. In the case of Step 2 CS your assessment is a differential diagnosis ( a list of the most likely causes for the patient’s symptoms) and your plan is the next step in the diagnostic work up.

Note that unlike on a real SOAP note your plan will not specify a treatment but only your next proposed steps in the diagnostic work-up.

In the case of chest pain with shortness of breath, your differential diagnosis might include Acute MI, Pulmonary embolus, GERD (heartburn), or PUD (peptic ulcer disease). Your plan for diagnostic work-up could be to draw cardiac enzymes, get an EKG, obtain a spiral CT scan, do a heart cath, or anything else you feel would be appropriate. You will not get any credit for suggesting treatment so save yourself the effort.

How long should your note be? Well, go look here:

Does that look like a long note? Of course not. It is not necessary to write the great American novel. I usually only needed ten minutes with the patient and five minutes for the note leaving me with ten minutes of thumb-twiddling time. Yet I saw most people leaving the room on the fifteen minute mark and scribbling furiously for the entire documentation time. They were obviously over-thinking it.

Focused history and physical. Concise note. That’s all there is too it.

The Step 2 CS exam is scored in three separate components each of which his pass/fail and all three of which must be passed. The first part is called the ICE or integrated Clinical Encounter which includes your history and physical exam skills as well as your documentation and assessment.

Communication and Interpersonal Skills (CIS) assesses your demeanor, your bearing, your use of empathy, your sensitivity, and even your appearance. Wear conservative clothes and a clean white coat.

The third tested area is English Proficiency.

All you will need for the test is your white coat and your Stethoscope. Every other piece of diagnostic equipment you need wll be in the room. PDAs and reference books are not allowed.

Do you need to study for the Clinical Skills test? I say no. I know there is a mini-industry of test prep material but your four years of medical school should be enough. Just act natural, do what you have been doing for your entire third and fourth year and don’t worry about it.


Your First Big Hurdle in Medical School

I got a fairly decent but not spectacular score on both Step 1 and Step 2 of the USMLE (United State Medical Licensing Exam) so I am not claiming any revealed wisdom or special insight into the tests. You must take what follows as my opinion alone and I caution you to study for the test in a manner which feels right for you.


For the uninitiated, the USMLE is a three part test which you will take at various times in your medical education. Step 1 is typically taken at the end of your second year of medical school and covers what you should have learned during that time. Step 2 is typically taken during fourth year and emphasizes the clinical aspect of medicine. Step 2 includes the “Clinical Skills” test which is a day-long standardized patient exercise that will cost you a thousand bucks. The only difference between paying a thousand bucks for the clinical skills test and being mugged is that you can’t file a police report after the test. (But more on that in a future post.)

Step 3 is usually taken early in your second year of residency and is your last hurdle to becoming a licensed physician.

Let’s talk about Step 1. Although you have spent the previous two years in the intense study of medical knowledge you will have to study for this test. I am sure that the average medical student can take the exam “cold” and probably pass it three times out of four with a grade close to the minimum passing score. A pass is a pass of course but a low Step 1 score will adversely impact your ability to match into a competitive specialty or a generally non-competitive specialty at a competitive program. In fact, many residency programs screen applicants by Step 1 scores so a low score will automatically close many doors for you.

Additionally, at all American medical schools Step 1 is a “must pass” test and you cannot advance through third and fourth year until you do so. At most schools you will get three chances to pass it. Upon failing the first time you will probably be pulled out of the third year rotation schedule and during the block you sit out you will be expected to study for and pass the test (this time comes out of your vacation). If you fail it for a second time you will be dropped back a year. Fail it a third time and they stick the fork in you as you are done.

At most schools you must attempt Step 2 before graduation but passing or even having the scores back is not required to graduate.

The question then becomes how long and what should study?

The answer depends on your school. My medical school had an eight-week break between the end of second year and the first day of third year. Most of my classmates elected to take Step 1 during this time either earlier or later depending on their comfort level and study habits. I believe that eight week is too long as you will invariably both lose your edge and also start to forget what you studied ealier. Three weeks is not enough time as you do have a lot of information to cover. Five to six weeks seemed to be the average in my class.

If your school doesn’t give you this kind of time then you need to make the best of it.

To study effectively you need to understand the format of the test. Step 1 and Step 2 are day long, computer-based, multiple choice tests. They are broken into hour long blocks of fifty questions selected more or less randomly from various subject areas of medicine. No two tests are exactly alike but an effort is made to keep the relative level of difficulty and mix of subjects more-or-less constant.

The tests are multiple choice but are not like the multiple choice tests which are probably the norm at your medical school. First of all, many of the questions have choices “a” through “k” rather then the standard “a,b,c,d.” You are consequently going to work a lot harder at eliminating wrong answers.

Additionally, the tests are concept-based, not fact-based, and feature many of what I like to call “double pump” questions. Rather than presenting you with a set of information and asking you to name the disease, for example, the typical Step question presents you with a brief case summary after which rather than asking you to name the disease the question might be, “What is the next step in the management of this patient?” Your excitement over knowing the disease was premature as this is not the answer they are looking for.

Heck, in many questions they tell you what is wrong with the patient. Consequently it pays to read the question at the end of a long paragraph first as the presentation is irrelevant once you know the diagnosis.

Another popular question style is to present you with a case followed by a selection of different lab results, your task being to pick the one which fits the presentation. You can usually eliminate most of the possible answers as obviously wrong (high pH in a set of lab values that you know should describe acidosis, for example) but invariably you will be left with a handful of reasonable looking results.

A variation on this is to present you with various simplified graphs and ask you to select the one which correctly represents the case.

Then there are the deceptively simple Behavioral Science questions which present you with a scenario and give you a list of choices only a few of which you can obviously rule out. The rule of thumb here is when in doubt go with the politically correct answer. Spanking is always wrong and it is never correct to advise a patient take responsibility for their actions.

I would say that of all the questions on Step 1 and Step 2, only a handful had a discrete, definite answer like “Sickle Cell” or “Guillan Barre Syndrome.”

So how should you study?

First, do not attempt to go back and review two years worth of class notes. The typical lecture curriculum probably has close to 4000 lectures in two years and you simply do not have the time. Not to mention that since the Step tests are standardized and your curriculum is not (despite the best efforts of your faculty) you might waste your time studying something taught at your school which is not emphasized at another and not likely to be included on the test.

Second, do not read text books because if your lecture notes are full of trivia the text is even more so. Again, you do not have time. This also goes for review books and I confess that I had a change of heart on this between Step 1 and Step 2. You simply do not have the time to review two years worth of material. The four weeks or so you will devote to study will dwindle quickly and your studying will bog down in a quagmire of detail.

Frankly, the best way to study for Step 1 is to do well and study hard during first and second year. You will retain more than you imagine and this is the best base for the most effective Step test study method which is to do practice questions and little or nothing else for your alloted preparation time.

The best practice questions, in my opinion, are those sold on-line by Kaplan or USMLEworld qbank. These are subscriptions to about 2000 questions each which are formatted exactly like the real questions, cover the same mix of subjects, and can be taken in Step-like blocks of questions exactly like the real test. The best part about the service is that you can read the explanations to the correct answer as well as to the wrong answers which in my book is just as important.

As I did the questions I made a point of reading or at least skimming all of the explanations to the right and wrong answers on all of the questions about which I had any doubts at all.

The advantage of these (and similar) on-line question banks are many. First, they are the exact same format and feel of the real test and will get you used to the “physics” of the computer test. That way there will be no nasty suprises on test day when you suddenly run out of time on several blocks or wrestle with the interface.

Second the questions focus on subjects which are actually tested. This means that you are going to get the typical standard presentations and basic medical knowledge with nothing coming out of left field. It was my sense that the questions are not written to trick you and it is only your lack of knowledge which can confuse you. The hoofbeats are always horses, never zebras.

Another advantage is that the questions are phrased and presented in a similar manner to the real test with the important distinction that the practice questions from Kaplan and USMLEworld seemed harder than the real questions. In fact, I scored considerably lower on the practice tests than on the real thing. The practice questions distributed by the USMLE are almost exactly like the real questions in difficulty.

Finally, doing practice questions will put you in the test-taking mode. If all you do is read review books you will have a store of rapidly disintegrating facts jumbled in your head. Doing practice questions narrows your focus to answering questions based on patterns which you have seen before.

Each of these services costs about 100 bucks for a one month subscription . A couple or three thousand quesitons should keep you busy for a month.

A word about “First Aid,” a popular USMLE review book which has an almost cult-like following.

Save your money. First Aid is a collection of “buzzwords” and supposedly high yield facts which are touted by some as all you need to study for the Step tests. I understand that in the old days the Step tests were twice as long but had shorter questions with discrete answers. First Aid might have been useful for a test like that but the nature of the test has changed making First Aid less useful.

Like I said, some people swear by it and claim to to make fantastic scores by studying nothing else. I am skeptical. I hit First Aid pretty hard for Step 1 (before I got smart and started doing practice questions) but I could remember only a few questions on the test where I thought First Aid had even remotely helped me.

Just a few random thoughts:

1. Avoid the temptation to cram the night before a Step test. How much are you really going to retain and more importantly, of the two years of knowledge required to take the test, how much of it can you cram into one sleepless night? Relax. I stopped studying a week before I took Step 1 because I literally could not stand doing another question or reading another page of review material. Remember what I said about peaking early. You just have to know when to say “no mas.” Let your conscience be your guide but it is better to go into the test relaxed (but alert and ready) than to panic and lose sleep over it.

2. It will be a long day. Bring a lunch, of course, and wear comfortable clothes. Paradoxically the day will seem to fly by once you get into the test. You are given plenty of break time but most people just plow through most of their breaks to just get the test over with.

3. I thought I had failed Step 1. I mean, looking back it seemed that every question was renal physiology and that there were only a handful of the 350 question on the test of which I was sure of the answer. It seemed that I could get most of the questions down to two or three choices but I was never really sure on most of my answers.

So I dreaded getting my score and anticipated failing even though the first time pass rate for American medical students is something like 93 percent, I am not stupid, and I studied pretty hard. I asked around and this seems to be a common perception after finishing the test. Almost everybody thinks they bomb it but most people don’t. Move on. Enjoy the rest of your vacation. Unfortunately it will take six weeks to get your results so you have a long wait.

4. Step one is divided into seven one hour blocks of fifty questions. I am a very fast reader so my strategy was to read the question, quickly select an answer, and if I wasn’t sure mark it for later and move on. (The software allows you to select questions within the block for review.) I found that I was able to get through the block in about half and hour leaving me with half an hour for review. Typically I had about half of the questions marked and spent the time wrestling with them. There were always a handful of questions for which I didn’t have a clue and these I marked “C” without wasting any more time trying to figure them out.

The key is to have a system that works for you and not to get hung up on one question. If you don’t know the answer mark it for review and move on. Don’t win the skirmish but lose the war.

I repeat: Take my advice as free advice and thus worth what you paid for it. I welcome your comments but please don’t flame me because we disagree. I am perfectly willing to be corrected, proven wrong, or convinced of your point of view.