Good evening, this is John Miller, one of the co-directors of the Heart Rhythm Society Core Concepts in EP and Board Preparation course. I'm here to welcome you and we'll start off right off with an introductory presentation on strategies for success, how the examination is constructed, how to take it, how to do well on it. These are my disclosures, you can dwell on those for as long as you want, I'm not going to. Now, the ABM exam is a very highly structured, very carefully prepared presentation. And it goes as this, once an examination has been administered, that is a person has taken the examination, all the scores come back, all the scores are collated, and immediately afterwards there is a meeting of the members of the exam committee, the ones who had written the questions. Old questions that have done well on the exam are accepted for further use on the exam. A question may appear on several different administrations of the exam over several years if it's a good basic question that works well. I'll get to that in a second here. If the question hasn't worked well, it is updated, assigned for editing, maybe distractor B was not a good one, maybe choice C got too many people. So that's modified some along the way, and sometimes it needs to be updated. Sometimes it's just hopelessly broken, it won't work for a variety of reasons that we'll talk about, and it has to be discarded, even if it's a beautiful tracing and so forth. The ideal question on the board exam is one that a little over half, but not quite everybody gets correct. That way it is not too hard that nobody can get it right, and it also is not too easy that it doesn't discriminate amongst individuals who are really good electrophysiologists taking the exam, and the ones who are not very good. And it also should have a pretty good correlation with the overall score. So for instance, the people who ended up doing very well on the exam should have gotten that question right. People who did poorly on the exam should have gotten it incorrect, and predicting that way. Each question that is invented by a member of the test committee that is edited a couple of times and accepted for use on the exam gets a test of itself. So it is included in the 50 questions that are trial questions that are not counted into the score. We're actually evaluating those questions during the examination administration. If it works well, discriminates well, and is not too terribly hard, it goes into the pool of questions that may form an actual test in the future. Each question is assigned a degree of difficulty, we'll come to that shortly as well, by the test committee members, and the absolute passing score is calculated prior to the administration of the exam. Usually it's about 65 to 70% correct is passing. That's not very good. When you think of it, getting a 65 on an exam when you pass, it's not very good. Very few will get 90% or more correct. We'll see a histogram of this. The performance of outlier questions is reviewed before the final results, and we'll come to that as well. The committee members decide on the weighting of what types of questions are on the examination. This varies from year to year. Sometimes it shifts very little, but you see here the medical content, invasive diagnosis and therapy, that's endocardiac electrograms and so forth, ablation, gets the lion's share. Basic science and pharmacology, they used to be separate, now they're all rolled in together. 18%, device management, also 18%, non-invasive diagnosis and therapy, rhythm strips, telemetry, event recorders, cardioversion, things like that, 15%, and then clinical scenarios and syndromes, 7% of Bergada syndrome, syncope, things like that. Questions test in three different bundles, knowledge, synthesis, and judgment. We'll come back to that in a second as well. Now the exams are, that's the old style of exam, the newer administration is slightly different distribution of materials, as you can see here. Note also that individuals who are taking the maintenance of certification examination have a slightly different distribution, a little bit less in basic science, anatomy, pharmacology, and genetics. The recertification exam is given twice a year, whereas the initial certification is only in the fall. Who passes? Who decides who passes? Well, the passing percent is not established a priori. It is a method that people on the exam committee have a difficult time understanding, even I was on the exam committee for a total of 18 years or something like that, and I never did quite understand this Engolf method, but what happens is each question is evaluated by all the members sitting around the table at the test committee meeting, and each individual is asked to think of one of their trainees who's just barely going to make it, and what percent of the time would that individual get this question correct? Each question is then tallied in this way, and it goes through some mathematical transformations and ends up devising a cut score, so if, for instance, everybody thinks, well, this would be a really hard question for this person, then the cut score is actually lower, and so more people might pass the exam, and vice versa. This is a distribution for, I think, 2017, 2018, maybe, and you see there were a couple of people who did this. This is on the old SAT, I'm sorry, the SAT type of format here, and you see there's some people who didn't do very well at all and some people who did moderately well, some people who did very, very well here. This is the median, the mean score here was 560. The passing score is 431, so there were people who got 430, and that doesn't work. This was one of my trainees who did pretty well out here at 693. There were a few people that did better. That's the standard distribution of things. Over the years, this is the number of candidates who have taken the examination and become certified, you see, in the first administration in 1992 was a whole bunch of folks who were legacy trained, decided to take the exam on his first pass through. It was taken every two years for a while, and then every year, and you see it's kind of oscillated a little bit, but the numbers have gradually climbed a little bit. 2018 was a little bit more than this year. Now, it's not a terribly hard test nowadays. Back in my day on the exam committee, we were down in the 70s and so forth percent passing. We see now in 2018, 96% of first-time takers pass the exam, 88% of recertification. This is a pretty good score here. We don't ordinarily like it to be quite that high, but if that's what the exam committee has come up with in their Angoff methodology, then that's what it is. These are the three different bundles of types of questions that are asked. Under our knowledge question, this is straightforward facts. The ion that passes through the sodium channel is which of the following? Sodium channel. That's not too hard. That's no controversy. Everyone can agree on the correct answer there. Synthesis is a question where you have to know something, you have to make inferences about something, for instance, interpreting a tracing, and then based on the tracing, what would you do next? Judgment is a harder type. It's making a decision based on the data that you have, and which choice of therapies would be the best one here, is B really better than C, or how does that work? These judgment questions should be about 40% of the examination. They're really tough to write well. It's nearly impossible to, even though you've got a very good topic to test the individual on, it's difficult to construct a question that's neither too obvious nor too difficult. Now, for the recertification module, there's a separate subcommittee. It has really nothing to do with the certifying exam. The question material is different from which the questions are drawn. The candidates can use OpenBook for this, and it tends to be more geared towards maintenance of certification, keeping current with the literature and new ideas, whereas on the certifying exam, as we'll see, material that's on the examination has to be what we would call settled law. It's something that is indisputable. It's not just one paper in the literature. It's been replicated by several groups. The diagnostic test has been shown to be perfectly valid and so forth. So that could be on the certifying exam. Something that is relatively new and somewhat controversial might be on the MOC exam. This bothers some people who take the MOC exam because they want a right answer. Well, it's meant to test individuals' currency with what is happening in the literature. Something such as the role of adenosine in testing at the end of a pulmonary vein isolation procedure. You can test the physiology of that. That's settled. But as far as what to do with that information, that's still somewhat controversial. The performance evaluation is more difficult since this is a test where it basically is OpenBook. There's a time limit on it, but you can use any kind of resources, and it's not clear exactly how the pass rate is configured with that except that most people pass. It's important to know that the recertification people and the first-time test takers don't take the same exam. You see that the basic science for the research people, as I said before, has decreased to only about 10% of the examination as opposed to 20% for the initial certification. There are only a limited number of arrhythmias that people can have. Most of these will be covered, some out of proportion to their prevalence in real life. For instance, atrial fascicular pathways. Some people have never seen them in their career, and yet they might see one or two questions on the certifying exam concerning atrial fascicular pathways because they're so rich in testing the ability of people to think correctly and the physiology. Questions try to be generic, not having a product-specific feature. When one manufacturer was the first one out with minimizing the amount of ventricular pacing, that was not a feature that could be widely used on the examination. Now, everybody has that feature, so it's something that can be tested. Rate response is the same sort of thing. There will be generic questions about how to configure rate response or whether to use rate response in a given individual, but there won't be anything specific about one manufacturer's algorithm versus another. There may be questions that have been used on many examinations, and some of them are frankly – they were great questions in their day, but their day has passed, and so they have to be retired. There will be on-screen calipers for use during the examination. We have our own emulation for the prep course here. They tend to work reasonably well. Initially, a few years ago, there were problems, but they tend to work reasonably well. We haven't heard too many complaints about them recently. Now, one problem that really good candidates, really smart people have is sometimes they overthink a question. So they think, well, I would really like to pick A because that seems the best, but I wonder if the guys and gals who wrote the question really had something else in mind. Don't overthink. There is one best answer. It will fit all the other information on the exam. All questions are multiple choice. There's no essay. There's no true-false. There's no A and B or all of the above or none of the above. A question is constructed of two different parts, a stem and the distractors. The stem presents the background of the question and asks the actual question. The distractors are the actual choices, A, B, C, D, and E. Over half will have figures associated with them. Our discipline is very rich in figures, intracardiac recordings, device, electrograms, and so on. Some figures might appear more than once on the examination for different purposes. It won't be the same question used twice. The same question asked twice, but it may be the same very good ECG or intracardiac recording twice. That's happened a couple of times. The stem is that part of the question, of the item where the background is given, and then the question is posed. Such language as what is the least likely or everything except will not be on the examination, only positive things. All of the following are true, except that this can't occur. Usually, the stem includes information about the patient age and gender, and that's called for face validity, just instead of a person, it's a 59-year-old man who had a myocardial infarction, something such as that. Occasionally, that's very important information, such as a Southeast Asian man who has a myocardial night terror or things like that with Brugada syndrome. There won't be any questions that are question number 37 uses the same figure as question number 36. It doesn't work that way. Each question stands alone by itself. The questions that have to do with specific medications or devices will only be concerned with agents that are FDA approved and in clinical market use. Basic science questions, however, may cover drugs that are in development, technologies that are not available in the United States, but are prevalent in the literature. Figures are the mainstream of this examination. They are maybe edited, but they're not retouched. They are as they are. People try to get their best figures, very nicely cleaned up original figures. For intracardiac recordings, many of us strived over the years to get things as uniform as possible. So you have the surface ECT leads, and you have high red atrium, proximal to distal head, proximal to distal coronary sinus, and so forth, and right ventricle down at the bottom. The reason we did that is because different laboratories may have different setups, and is the way they record during a procedure. During a procedure, we thought it was most appropriate to have a uniform format that most could agree on. And we didn't want the candidate who's taking the examination to be staring at a question, saying, Kelly, is a CS proximal to distal or distal to proximal? You don't need to spend your time trying to figure out what you're looking at. You need to, we need to test knowledge, not how can you recombinate things in your head. For coronary sinus or tricuspid annulus catheters that can be in variable locations, if it's important to know where that is, it will say that the proximal electrodes are at the coronary sinus ostium, or the distal electrodes are at the lateral margin of the heart. There may be a diagram showing where a tricuspid annular catheter is located, if that's important. There may be one, maybe up to two questions per tracing. You won't be having to flip back and forth over three or four different tracings, as you might in real life. But this is not acceptable on examination. Single tracing, as I said, may be used more than once on different questions. Unusual, but it might happen. Artifacts, noise are part of life. And we have to make real judgments on these things in real life. So we have to make judgments on the examinations. While you're supposed to be an expert, you ought to be able to recognize artifact and know to not be misled by it. It'll be pretty obvious artifact when it is, so it's not gonna be terribly subtle. The distractors are your choices, A, B, C, D, and E. Minimum of three, maximum of six, usually four or five. There is only one best answer. There may be another one that's pretty close, and you might be thinking about that. But when you think long and hard about it, one answer is the best, and that's what you should put down. All the above, none of the above is for both. Here's an example of a question here. A 13-year-old Caucasian girl is brought from a local school complaining of one hour of palpitations. Her blood pressure is 118 over 80 millimeters of mercury. She is aware of a rapid heartbeat but has no other complaints. Her ECG is shown in figure 44, which I'm not gonna show. What rhythm is most likely shown on the ECG? It's not the only one that it could possibly be, but based on the ECG that I'm not even gonna show, the most likely one is, you see the choices there. So you shouldn't have too much trouble distinguishing ventricular tachycardia from junctional tachycardia and so on. But these are the types of questions and the types of wording you would have. Rarely will you be asked to give a definitive diagnosis. It's exclusive of all others. It's what is the most likely, not the least likely. When you're trying to evaluate among the different distractors and you're given a choice to reoperate on a patient, to reposition a lead or do reprogramming, pick the least invasive as the default. Think patient safety, what's the easiest for the patient. But if you really need to reposition a lead, if there's just no other way around it, then that's okay. Most likely as, which is the most likely diagnosis is not the same as I said, as the only possible diagnosis. So don't get too twisted around about that. Next is not the same as ever in a management question. So for instance, you might decide to reprogram a device because of T-wave over-sensing as opposed to replacing a lead. Although patients had many different episodes of T-wave over-sensing, many shots from this. And you can see that the T-wave is always larger than the ventricular electrogram. That's probably not gonna be a solution. You probably are gonna have to intervene. So what is your best next step versus what would you really need to do eventually to make that discrimination? Percentages are hard to test. For instance, the likely success rate of a procedure or the likely complication rate of a procedure should not ordinarily appear as a 17% or 43%, something like that, because that's not the case. Most individuals could probably find a study that gives their answer and the answer that they put down on the examination. We wouldn't want to count that wrong. Questions are rarely based on one specific trial unless it's a really landmark trial and there will be no reason to replicate it. Everyone can agree that the answer is correct. Recertification questions may have information about a recent trial, but there's no reason to replicate it. We have information about a recent specific trial that's come out, especially if it seems very solid in its information. Technical issues or problems, complications of the procedure are fair game. Individuals have access to the literature and even if for instance, one doesn't do transeptals or epicardial mapping, you ought to know something about what can go wrong with these procedures and what to avoid. The exam, as I said earlier, covers material that pretty much everybody can agree on. There won't be any concern around the exam committee table that, well, I don't know about this. I kind of reserve judgment. Everybody has to agree that that is the right answer. We're testing whether an individual gets certified or not. It's not, we can't be equivocal about the answer on these questions. Material taken for the examination has to be a couple of years old, just because even if it's really new and hot information, the exam, the question has to be written. It has to go through a couple of edits. Those are, that's a year long process. It has to be pre-tested on our examination where it won't count. And if it performs well, it can then be on the exam. But if it doesn't perform well, needs a little tweaking, it goes back in the bin. So it's a couple of years before information that's current gets on an examination, no matter how important it is. Rarely will a brand or a device-specific algorithm be on. I don't think that has happened that I'm aware. Be sure you look at the whole tracing. Look for changes in atrial or ventricular activation sequence. Look for changes in VA intervals that may not be appreciated from beat to beat. But if you look at the first beats of the tracing and look at the last beats on the tracing, you can say, oh yeah, those are different. So this is an atrial tachycardia. It's not a fixed VA interval. Are pacing stimuli capturing consistently? Quite often in an examination, we'll have a question about attempted entrainment and what looks like a good match with pacing, but it was never capturing. So of course, it looks like a good match. Be careful to make sure that before you start analyzing post-pacing intervals and so on, that you have consistent capture. There's some chronic problem areas on the exam as people who've been on the board have scratched their heads about. Electrocardiograms. Electrophysiologists ought to be able to diagnose some disorders on the ECG. Maybe not everything, but the ECGs that are on examinations are not that hard and should be gotten correctly by most individuals, and yet a lot of examinees fall on this. Likewise, retrograde conduction. I'm going to be talking about that a little bit later. It's an important concept, and it really isn't that difficult, but again, many individuals are just not that well-steeped, not that versed in it, and it causes a lot of problems on the examination, so we'll go over that as well. This is the makeup of the current test committee. You see some real good people there that have been around for a good long time, and we have some good diversity of gender as well as legacy of institutions from which they derive. Very good people. You can direct your complaints to them. Some general principles for taking the exam. If you haven't already started preparing for it, there's a lot of material out there with our course as well as others to branch out onto, and you should really be underway with your preparation if you haven't gotten so already. Look at critical review articles during some of these lectures and workshops. They'll be referenced. There should be a pretty standard approach to all the tracings. Look at the whole tracing. Don't rush to judgment at the first thing you see. There are 50 questions over two hours and four modules that you take. That averages out to only about 2.4 minutes per question. If it's a text question like a pharmacology question, you breeze through that. If it's a complex figure, it's tough. Characteristically, examinees take almost the entire time for all of their modules, so be careful to not get behind. If you get stuck on one question, just say, well, I think it's A or B and move on and come back to that. Please answer every question. There's no penalty for guessing or for wrong answers. If you come in the last minute to have 10 questions left, answer them all and get some of them right here. Now, George Klein, who was on the examination committee for several years, came up with some pretty pithy things here that helped the examinee. Read all the questions carefully. It may seem that some of the information is extraneous. Sometimes it is, but the exam question writers are pretty conservative in the use of language and they will include all the information that's necessary to solve, to get the correct answer. And then some distractions, but all the information you need is there. Once you've looked at the question, think, okay, what is this technical, it could be three or four things, but it can't be 15 things. So you've already excluded some of the things. And then you have to decide, okay, which of the things, just like a differential diagnosis, which can I eliminate, which can I not eliminate? There is one best answer, as I said, look at the whole tracing, don't focus on just one part. Measure intervals, don't just estimate or guesstimate behavior calipers. There's no reason to discuss things unless you're really time pressed. Focus on zones of transition during a tracing. So going from pacing to tachycardia, tachycardia to pacing, bundle branch block to no bundle branch block, whenever there's a change somewhere along the way, there's information, George used to say, still says, and make use of that. These are very important areas for the electrophysiologist and a lot of information there. Test each of the distractors if you're uncertain for goodness of fit and play the game of what if. What if C is correct? Does that satisfy everything else about the question? If it doesn't, then it's probably not the correct answer and you keep looking. Let's do a question here. 45-year-old man presented with wide caress tachycardia. In an EP study, the effect of an extra stimulus is shown in figure one, the best interpretation, not the only possible, but the best interpretation is ventricular tachycardia is present, pre-excited avian ovary entry is present, anterodromic tachycardia is present, orthodromic tachycardia is present, or no diagnosis can be made. And that may be correct in some questions. You don't want to force a conclusion where not enough information is present. All right, here's the question. This is a wide complex tachycardia with five surface ECG leads, high-rate atrium, his proximal distal CS, proximal distal and right ventricular recordings, and a single extra stimulus is introduced from the high-rate atrium. Now, this is how it appears. No numbers on there, no intervals. You have to decide what's important to measure, if anything's important to measure, what to do with this information. So we'll go on to the answer here. And when making the appropriate measurements, it's important to find out that anterodromic tachycardia is the most likely diagnosis here. Why is that? Well, here is the onset of the QRS here, I guess. And we're looking at the VV intervals on the top here, 330, 330. The one that is in response to the atrial extra stimulus is short. And then the one after that is longer. There's information there has been a change, zone of transition. The HYS is shown here, and the HYS-HYS intervals are as shown, 330, 330, the same as the VV intervals. And the HYS-HYS is the same as the VVs throughout, okay? So since the QRS starts before the HYS, the HYS is beholden to the ventricular recording here. And then finally, the AA intervals. Again, at the proximal HYS here, 330, 330, they also track along. So the rationale here is that a PAC during this wide complex tachycardia occurs when the HYS has already occurred here. Here's the anterograde HYS. And it does not affect the septal atrial activation. This AA is right on time here, okay? So if this is our earliest A, and for instance, if it were AV nodal reentry, we should not have any effect on the timing of these next A's because if it's pre-excited AV nodal reentry, or VT with AV nodal reentry, we haven't done anything to this A here, so we can't affect that next H. Because we did affect the next HYS, that's not the correct diagnosis. So it doesn't affect the septal atrial activation, but does advance the next QRS and the next HYS, and resets the tachycardia because this next interval is different than expected. That excludes ventricular tachycardia. You can't do anything about VT with an atrial extra stimulus unless you have an accessory pathway to get to the ventricles, and we have an accessory pathway here. The HYS after the QRS onset, as you see here, there's a QRS onset along here somewhere, HYS is afterwards, excludes orthodromic tachycardia because this has to be for the QRS under those circumstances. The atrial activation after this QRS, the one that is advanced, is identical to the others, and the VA or HA interval is slightly longer than on the other complexes, if you measure it. This makes retrograde conduction over another pathway unlikely because the A would be dissociated from the HYS under those circumstances. And thus, it seems most reasonable that the diagnosis is pre-excited tachycardia going down a pathway up the AV node to antedropic tachycardia. Thank you for your attention, and we'll move on to another topic. Thank you.

Heart Rhythm Society

exam preparation

question evaluation

passing score

types of questions

ECG interpretation

retrograde conduction