Greetings, this is John Miller from the Heart Rhythm Society Core Concepts in Electrophysiology. I want to spend a few moments for those who are interested in taking board exams for initial certification or recertification, as to ways that you can improve your chances of a successful outcome of your venture. Here are my disclosures. I want to talk a little bit about how the exam is constructed. Several of us that are on the Core Concepts faculty have served with the American Board of Internal Medicine in the construction of either the maintenance of certification examinations or the secure initial certification examination. We know a thing or two about how to make questions. The way this goes is the exam committee, at least in past years, I was on the committee for many years and the chairman for a few years. The immediate prior exam is reviewed. Each question is reviewed very carefully. Questions that performed well will come to what that means subsequently are accepted for repeat use on another exam because it discriminated well, it worked well as a means of telling who should pass the examination. Those questions may have to be updated, new techniques, new terminology may be necessary, they may be modified, or if it's just a good question, but for instance, something about signal average ECGs, pretty much nobody uses those anymore, so it's not maybe a great question about signal averaging ECGs, but it's just not relevant and so should be discarded. It was relevant at one time. The ideal question is one in which somewhere between 55 and 85 percent of the candidates answered it correctly and it had a good correlation with overall exam performance, meaning if the people who did well on the overall exam got this question correctly and the people who did poorly on the overall exam got this question incorrectly, that's a good correlation. If on the other hand, the people who scored well on the exam overall got had trouble with this question, got it wrong, whereas the people who didn't do so well got it right, there's a problem with the question, it's not discriminating very well, so it needs to be answered. It can't be too easy, otherwise everybody gets it and it doesn't discriminate. It can't be too hard, otherwise nobody gets it or very few get it and it doesn't discriminate well. We wanted to discriminate who has the goods to pass the exam, the written exam, and who doesn't have all the cognitive stuff that's necessary. Questions that are constructed by the exam committee will have a trial run on its first use, so it's presented on the secure exam a bunch of questions that don't count. They're merely testing the question for how good a question it is, how well it performs. If it performs well, it goes into the pool for an actual test where it actually counts for the score. Each question is then assigned a degree of difficulty that's been accepted for going on the examination, and an absolute passing standard is calculated prior to the exam. Roughly 65 to 70 percent of questions are correctly answered. That's a passing score. Very few times are 90 percent or more. Very few candidates get 90 percent or more correct. It's a tough exam. Performance outlier questions are reviewed before the final results are presented, so the chairman of the board for those years gets to review questions that really didn't perform very well at all and gets the option of saying, okay, everybody gets this one right. We throw out the question. It's important because you have to accumulate an absolute number of correctly answered questions. That's the passing score. X number of questions you pass, X minus one you don't pass. So throwing out a question is a big deal. Giving the correct answer, giving everybody a correct response is a big deal also. The development of the exam hopefully reflects practice in electrophysiology such that invasive diagnosis and therapy, about 40 percent of the exam, intracardiac electrograms, ablation, how do you do this, how do you do that, technology techniques. Basic science and pharmacology, about not quite 20 percent. Device management, about 20 percent or so. This number fluctuates a little bit according to the weightings given by the exam committee at the time. Noninvasive diagnosis and therapy, there's cardioversions, tilt table tests, monitor strips, ultra monitors, loop recorders, things like that. Clinical scenarios and syndromes, this is a patient with syncope, a patient with an inherited disorder, how do you manage this, how do you evaluate this. Test questions are supposed to test knowledge, which of the following drugs does this, knowledge. Synthesis, given this situation, the most likely correct answer is this. Giving a couple of facts, you have to come up with a hypothesis. And finally, judgment. Judgment, as we'll see, are the hardest questions. Now, the mix of the questions varies a little bit depending on whether it's initial certification or maintenance of certification, mainly in the distribution of basic physiology, anatomy, and pharmacology, and genetics. Not too much of these are on the maintenance of certification, you can't throw those away. Every question counts, but they're weighted a little bit differently, and a little bit more in devices amongst those who have been around for a while. The initial certification exam is offered once a year in the fall, and maintenance of certification twice a year in fall and spring. How is the pass rate set? Who in the world is going to pass this exam? There is not a passing percent that's pre-established. It's a seemingly funny way of doing things. There's a modified Angoff method, and each question, each item on the exam called an item, is evaluated by the test committee. And the members are supposed to think of a candidate who is just barely going to get this question correct. 50, 60, 40, he or she is going to get it correct. And so what percentage of minimally competent examinees would get this correct? If you say 90% will get this correct, then it's going to be a harder exam. If you say only 15% are going to get this correct, it's going to be an easier exam. And so it goes around the table and people make their estimates, and the sum of the item ratings represents the passing cut score. This is the cut score here. The passing score was only 376. It goes from 200 to 800 scale like the old SATs. 376, really? Well, that's the number you needed to do to get this correct. This is one of my prior fellows. He got a 757. He's up here. I'm not sure who got better than he did, but okay. Somebody maybe. We got lucky. The mean score was over here. The passing score is here. So everybody who didn't get those number of questions correct does not pass the exam. Everybody who does passes the exam. This is the number of candidates who have been certified and maintenance certifications on there. You see it kind of peaked in 2016. And then we went to two years. And so that group is coming through, and I suspect it's going to come up again here a little bit. You see there was an initial influx when the first exam was given in 1992. It was only every two years then until enough questions became available to do it every year. The passing likelihood pass rate for first-time takers and retakers is pretty good. In some years, it's been as high as 95, 96 percent. Way over here, it was in the early days, it was in the high 70s, low 80 percent. It really meant something. Not quite like the interventional boards where you just have to kind of be able to sign your name twice and seem to pass those, but this is not terribly hard. The question types are, as I said earlier, knowledge, just straightforward facts. There's no controversy here. The primary effect of a Demzine on a cell is which is following. Synthesis, this is a little bit harder. It is using pieces of information to reach a straightforward conclusion. Applies to a lot of EP tracings. The following tracing shows what. I'm looking at these electrograms and it shows this. Judgment questions are hard. They're hard to make a question like this. First, everybody on the committee has to agree that the correct answer is the correct answer, and not just your opinion has to be validated by objective data. But it's decision-making based on the information provided. We're given a clinical scenario, a couple of lines about the patient situation, and then these are the choices, maybe based on some recordings, device recordings, electrograms from the EP lab. The answers have to be at least plausible. The temperature outside today was 60, 80, 100 degrees, 40,000 degrees. That doesn't make any sense. That wouldn't be a plausible distractor. Judgment questions ideally comprise about 40 percent of the examination, but are difficult to write it, as I said. Now, the recertification modules are different. It's a different committee that deals with these questions are circulated to select board-certified individuals to mull over and critique. When on the maintenance of certification exam is with practice tests is to make you look up relevant data and make you keep on continuing to learn. The content of the examination of the maintenance examination is more recent information. The stuff that's on the secure exam for initial certification has to be information that is well-validated in literature. There's no controversy about it whatsoever. It has to go through these multiple cycles of evaluation of the question and its validity. That information has to be two, three, five years old, probably, whereas the maintenance certification can be more recent information. The certification and recertification guys do not take the same examination. For instance, the basic science has decreased substantially. There are a lot of different sources from which the exam committee can take questions. There are only a limited number of arrhythmias that people have had. Most of these will be covered on the examination. The amount of information that they make up of the questions is not proportional to the likelihood of seeing a case of whatever in real life. A lot of people joke that they saw their first atriofascicular case ever on the exam. It's got a lot of physiology to it. It tests a lot of things. You might have one or two or even more questions about atriofascicular physiology on the exam, not a reflection of how often that occurs in real life, but it's important stuff and testable. Questions aim to be relatively generic. For instance, you won't be asked a question about some specific algorithm that is embedded in a mapping system or an ablation characteristic or some unique pacemaker mode of operation. If it's something about minimizing ventricular pacing, most of the manufacturers have something like that and it should be recognizable on an exam, but you won't be tested as to the finer points of that shouldn't be. Controversial topics are off limits. There has to be an agreed upon right answer. In recent years, for instance, should adenosine be used to test the durability of pulmonary vein isolation? The papers that have been for and against it showing that it works, it doesn't help. You can test physiology of adenosine, but you can't test that you have to use it to get the result of durability of pulmonary vein isolation. Older questions may persist, especially with devices, a lot of updates of features and some recognizably very old intracardiac recordings or device download recordings probably should have been purged by now. All systems exam facilities now have onscreen calipers that have worked out reasonably well. Use them where you need to. Sometimes you don't need to, but be very careful if you think you don't need to. It has to be very, very obvious that you don't need it. Don't overthink questions. A lot of really smart people have not done well on exams because they think, well, did they mean this or did they mean this? I know the straightforward answer, but maybe they meant this instead. Don't overthink things. You should be one best answer here. All exam questions are multiple choice, single best answer. There's no true false. There's no which of the following except that that doesn't occur. It's all positive. Which of the following is the correct answer? There are two parts to a question. There's the stem and the distractors. The distractors are the choices you have to make A, B, C, D, or E. About half of the question, and the stem is the introduction of 47 year old person with such and such as this. How do you interpret this figure? About half will have figures. Figures take up a lot of time on the examination. We'll come back to that in a second. Exam figures could be used more than once in an exam, typically or not. There's enough material. The stem is a description of the case followed by the question. Again, there's no use of the least likely or except which of the following, except all are trying to be positive. Sometimes it results in some funny wording, awkward wording, but it's trying to turn it into a positive. Quite often there's some demographics, 47 year old woman, a 53 year old Asian man. Sometimes that's important. Sometimes that's not important, but it gives just some face validity it's called. There will not be run on cases where, you know, question number 32 is on this figure is also be referred to question number 33, number 34. None of that, each question stands on its own. There are only U.S. FDA approved drugs and devices that will be source material. Nothing that's investigational at this point. And for basic science questions, and it goes, there are not that many of them, but I think it goes. Figures on AP tracings are sometimes lightly edited, but they're not retouched. Actually, as they come out, the team works very hard to make these as clear as possible. Formats are usually uniform. When I was on the committee, I made sure that we had service leads one through whatever at the top, and then hybrid atrium, his proximal distal, CS proximal distal, and so on. If there was some variation in that, it was designated as much. Our feeling was that the candidates should not have to spend time thinking, what am I looking at? The CSs are upside down. I got to do this in my head. And so everything should be a uniform presentation. For coronary sinus or tricuspid annular catheters, sometimes, unless it says the proximal electrodes are the CSOs, there'll be a diagram of exactly where the electrodes were. There might be one or two tracings per question, ideally only one. Artifacts are a legitimate testing point. I can spot an artifact on my own. Maybe you can't, maybe you can't. They occur in real life. You have to get these right in real life. You have to get them right on the examination as well. The distractors, the possible answers, almost always there are four or five, sometimes as few as three, rarely as many as six. There is one best answer, always one best answer. Another may be really close to right, but it's one best answer. And there will not be all of the above or none of the above that doesn't occur. So here's an example of a question. 13-year-old Caucasian girl is brought from a local school complaining about hour of palpitations, blood pressure as such. She's aware of the rapid heart rate, but has no other complaints. ECG is shown in figure whatever. The rhythm is most likely shown on the ECG. This is the stem, this is the question line, and these are the distractors here. And I'm not gonna show you the actual question. Several aspects of the question development deserve some note in how you approach a question. Think patient safety and doing conservative measures first. So for instance, you might have somebody with T-wave over-sensing, and they got a shock. You'll probably do reprogramming first before you put in a new lead or reposition an old lead. So think conservative first. That may not be the ultimate solution for that patient. In your heart of hearts, no, this is not gonna work. We're gonna have to put in a new lead. And maybe that's true, but ordinarily you'd try something non-invasive first. When the question is, what is the most likely diagnosis? That doesn't mean the only possible diagnosis. It means the most likely diagnosis. So you winnow it out and say, yeah, it could be these other things, but the most likely one based on the information is this. What you would do next, if it asks, what is your next step? What is the appropriate next step? What's the ultimate solution for this patient? You may need a lead extraction, or you may just need some antibiotics for a minor cellulitis instead of a full extraction. So again, think conservative and non-invasive first. Percentages are hard to test. 47% of patients will have this response, or what is the anticipated rate of heart block after even a re-entry? That's a tough number to get. Sometimes it's not ridiculous, like 10%, 5%, 2.5%, or less than 0.5%. Nobody would probably get that wrong. Questions are rarely based on a single trial. I don't think I can remember if there were questions that are like this. Recertification may ask about a specific trial and the important points of that trial, but most have to be validated and made certain of by multiple trials in the literature before it appears on the secure exam. Technical issues or problems are fair game. Even if you're an electrophysiologist who doesn't do any devices, you're still expected to know a thing or two about devices and maybe about device implantation techniques. If you don't do epicardial mapping in a place, you're still responsible for knowing a thing or two about indications for it, potential complications of it, because you're covering for partners who may be doing it. It must be stuff that's well-known and represented in the literature, nonetheless. Questions may, on the secure exam, have to be the so-called settled law where there's no controversy. Everybody around the table can agree that the right answer is the one that's represented as the correct answer on the exam. These have to be at least a couple of years old because they have to go through these cycles of making sure the question is valid and all rarely brand or device-specific stuff will appear on the exam. You shouldn't, if you only ever deal with products from manufacturer A and B, why should you know stuff that's specific to manufacturer C or D? Please do yourself a favor and look at the whole tracing when you have intracardiac or device recordings. Look specifically for changes in activation sequences that may be pivotal for getting the correct answer, changes in the VA intervals that are not easily discerned from B to B, but if you look at the beginning of the tracing, at the end of the tracing, compare those that may be much more clear. Are pacing stimuli consistently capturing? This is a favorite of those people that write these questions nowadays. Chronic problem areas are electrocardiograms. You wouldn't think, but it does turn out that candidates are not real good at reading ECGs, localizing arrhythmias from them, diagnosing arrhythmias from them, picking out ST elevation. This was a real eye-opener to be on the exam committee with this, and retrograde conduction. This was not specifically part of... It wasn't the point of questions, but people had a great deal of difficulty discerning whether retrograde conduction was present or not, over what pathway it was, and that was the reason why the committee put in an entire talk on retrograde conduction, because it's an important thing in real life and an important thing on the exam. The process of preparation should have already begun if you're going to be taking an exam and you're seeing this clip here. You really ought to be reviewing things pretty critically. Reference critical review articles that are represented in core concepts, lectures, and workshops. Some are replete with references for you. Take a general approach to all tracings, a consistent methodical approach. Look at the tracing, what is your first impression, what is excluded by this, and so on. You do 50 questions in each of four two-hour blocks. That is roughly 2.4 minutes per question. You can't perseverate. Pace yourself. Don't get tied up with one or two questions that take way too much time. Scratch down some thoughts or remember where you were with this and come back to it. You can always, you don't get a penalty for incorrect answer. Might as well mark something for everybody. So don't leave anything blank. Read the questions carefully. All the information is presented for a reason. Sometimes it's a distraction, but there's only a certain amount of space that the exam committee has to write a question. They fill it with not extraneous stuff. It's not war and peace. It's a concise stem. So laboratory studies, device settings, ejection fractions, electrograms are all there for a reason. And everything you need to get the correct answer will be there. Maybe some extra stuff too. Just to, you know, you wouldn't want to say the device settings, you just say what the rate is. You probably want the outputs and the thresholds and things like that. Just what ordinarily people would ordinarily get. And you make your conclusions from it. Develop a differential diagnosis for what you think is going on in the question. Read the answers, read them carefully. There's one best answer. Always be one best answer. Pretty soon you'll say, yeah, that's the best answer. Look at the whole tracing. Don't focus on only one part that looks interesting. Look at the whole tracing. Remember, there might be stuff at the very end of the tracing, at the very beginning of the tracing that is interesting, very important. And if you don't look at the whole thing, measure intervals don't just estimate. You got calipers, use them. However, the correct answer is not going to be based on a difference of five or 10, five milliseconds. It'll be much more than that. Focus on zones of transition where changes occur during arrhythmia going from sinus rhythm to arrhythmia, from arrhythmia to sinus rhythm, from pacing to tachycardia, tachycardia to pacing, resumption of tachycardia after pacing, bundle branch block, presence or absence of information, all these zones of changes. Test each distractor if you're not certain about what's going on for goodness of fit. Try a what-if scenario. Well, if C were correct, if this really were an atrial tachycardia, would this fit with everything else that I know about this question? If it doesn't, then that's probably not the correct answer. So here's the question. 45-year-old man presents with YQRS tachycardia. At EP study, the effect of an extra stimulus is shown in figure one. The best interpretation is PT is present. Pre-excited adenoviral reentry is present. Antedromic tachycardia is present. Orthodromic tachycardia is present, or no diagnosis can be made. Here's the tracing. And you see it's a reasonably regular YQRS tachycardia. There's a single stimulus from the high-rate atrium. And I'll leave the rest to you. Turn me off and come back when you're ready. Okay. The correct answer is antedromic tachycardia is present. Why is this? All right, well, here's the onset of the pre-excited QRS here. And you see that the HISS potential is after it. Therefore, this is not any type of SVT with aberration or bundle branch reentry. It could be VT, but it is either pre-excited or it's VT. You see that the VV intervals are regular, except for where they're not. This is where calipers come in handy. This is a shorter interval. This is a longer one. Furthermore, where you see the HISS is here, they have a consistent relationship to the onset of the QRS. They're after it. And the HISS-HISS intervals follow the, are correlated with the VB intervals, but because the V, the HISS is after the V onset, the HISS is retrograde after the VB. All right, and the AA intervals also follow here. All right, so what do you make of that here? Well, this is a PAC that's delivered in a white QRS tachycardia while the HISS is refractory and doesn't affect septal atrial activation. The AAs here are, right here, are right on time. This is 3.30, so this extra stimulus did not affect the timing of this A. Therefore, it could not affect anything having to do with the AP node. And yet it brings in the next QRS complex. You can't premature, can't bring in the next QRS complex of a VT with an atrial stimulus. You've never heard of that. This is, therefore, a pre-excited tachycardia, okay? It's not AV nodal reentry because we didn't affect the perinodal atrial activity over here. Couldn't get it into the AV node. The HISS after the QRS complex excludes orthodromic SVT. And the atrial activation sequence, atrial activation that is after the advanced QRS is advanced to the same degree and is identical to all the others. So another, coming up another accessory pathway, unlikely, it's concentric, it looks like, AV nodal activation. And thus, the most likely diagnosis is antedromic tachycardia, not just a pre-excited tachycardia, but antedromic tachycardia. And I thank you for your attention. Do well on the exam. ♪♪

John Miller

Heart Rhythm Society

board exams

electrophysiology

exam construction

question selection

question types