This audit is for US emergency physicians and senior residents who assumed NEJM Knowledge+ runs an ABEM Qualifying Examination question bank. The honest finding first: it does not. NEJM Knowledge+ publishes adaptive board review for internal medicine, family medicine and paediatrics — not emergency medicine. Its content now sits inside AMBOSS, whose separately branded Emergency Medicine (ABEM) bank is the product that actually maps to your exam. Read your analytics with that in mind.
What NEJM Knowledge+ actually covers for ABEM right now
Last checked 19 July 2026. Before you commit a study budget on the strength of a familiar brand name, confirm what that brand covers, because the answer here is not what most candidates assume.
NEJM Knowledge+ built its reputation on adaptive, spaced board review, but its published catalogue is internal medicine, family medicine and paediatrics. There is no NEJM Knowledge+ emergency medicine product and nothing branded for the ABEM Qualifying Examination. The NEJM Knowledge+ line has since been folded into AMBOSS and is marketed as AMBOSS Knowledge+; the old product address now redirects into the AMBOSS and NEJM Group estate. That corporate move matters, because it is the reason an ABEM-shaped bank now exists under the same umbrella — just not under the NEJM Knowledge+ name you searched for.
| Product | Covers ABEM Qualifying Exam? | What it is |
|---|---|---|
| NEJM Knowledge+ Internal Medicine / Family Medicine / Paediatrics | No | Adaptive board review for ABIM, ABFM and ABP — not emergency medicine |
| AMBOSS Knowledge+ — Emergency Medicine (ABEM) | Yes (vendor-reported) | The EM board bank now sold under the same corporate umbrella |
The product that does map to your exam is AMBOSS's Emergency Medicine (ABEM) bank. AMBOSS reports it as "2,000+" single-best-answer questions built on the Emergency Medicine Model of Clinical Practice, with "smart analytics and adaptive study plans", an "AI Mode Learning" study copilot, a performance dashboard and peer comparison against other EM residents. Access is bundled as one year of library access plus a qbank window of one, six or twelve months, with list prices at the time of writing running from roughly $298 to $448. Every one of those figures is vendor-reported — verify the current count, features and price on the product page before you buy.
So if you arrived to read your "NEJM Knowledge+ ABEM analytics", the practical move is to read the analytics of whichever EM-specific bank you actually hold — the AMBOSS-branded successor, or a competitor such as Rosh Review, PEER or TrueLearn — and to treat any residual internal-medicine practice data as adjacent, never as an ABEM readiness signal. The rest of this article shows you how to do that without fooling yourself.
The exam you are actually preparing for
The ABEM Qualifying Examination is a computer-based, single-best-answer test of roughly 305 multiple-choice items delivered at Pearson VUE. It is built on the EM Model, which organises content across about 20 clinical domains — from abdominal and gastrointestinal disorders through to toxicological and traumatic disorders — plus the physician-task and procedural competencies that cut across them. Passing the Qualifying Examination is the gateway to the Oral Certifying Examination, a separate assessment of clinical care cases and communication and procedure scenarios that no multiple-choice bank reproduces.
The part of the blueprint that should drive how you read your analytics is the acuity and age weighting. The EM Model frames content by acuity: approximately 30% critical, 40% emergent and 21% lower-acuity presentations, alongside defined minimums of at least 8% paediatric and at least 6% geriatric content, plus image and pictorial items. That is the specification any bank claims to align with, and it is the yardstick your dashboard must be measured against — because a home-screen average of 78% tells you nothing about whether your 78% is concentrated in lower-acuity adult medicine or spread across the resuscitation and paediatric material that dominates the real paper.
Every metric on the dashboard, defined
Any credible EM board bank surfaces the same family of numbers. Define each before you trust any of them.
First-attempt accuracy is the percentage correct the first time you ever see an item. It is the only accuracy figure that approximates a fresh-eyes signal, and it is the one to watch.
Repeat accuracy is your percentage on items you have already answered. It rises mechanically as you re-encounter familiar stems, and it flatters. High repeat accuracy is memory of the bank, not readiness for the exam.
Percentile ranks you against other users of the same product. It is anchored to a self-selected cohort of paying candidates, not to the ABEM standard, so a strong percentile means you are ahead of other subscribers, not that you have cleared the pass line.
Predicted score or pass likelihood is a vendor model trained on prior users' behaviour and outcomes. Treat it as marketing-grade, not psychometric: it is a correlation on that platform's population, not an ABEM scaled score.
Coverage is the proportion of the bank you have attempted. Completion is not competence, and it is certainly not blueprint coverage — you can finish 100% of a bank and still have barely touched geriatric toxicology.
Difficulty is usually a bank-relative tag (easy/medium/hard by peer success rate), not an ABEM calibration. Time per item is your seconds-on-question; against a paper that gives you roughly 60 seconds per item, this is one of the few metrics that maps cleanly to exam conditions.
Why practice percentages are not comparable to the real paper
The reason a practice average overstates readiness is selection. Adaptive study plans preferentially re-serve your weak areas, which is pedagogically sound but statistically distorting: the questions you see are conditioned on your past errors, so your accuracy on that stream cannot be compared with a fresh, mixed, blueprint-proportioned block. Add tutor mode, where you answer with explanations visible or immediate feedback after each item, and you are measuring assisted recognition, not unassisted retrieval. None of this is a flaw in the product; it is simply why the number on the home screen answers a different question from the one your exam asks.
Blueprint audit: hold your attempts against the EM Model, not the average
This is the step most candidates skip, and it is the one that matters most. Export your attempted-question distribution and lay it against the EM Model acuity and age weighting rather than trusting the composite percentage.
| EM Model dimension | Target weighting | A candidate's attempts | Read |
|---|---|---|---|
| Critical acuity | ~30% | 18% | Under-sampled — the highest-stakes third of the paper is thinly practised |
| Emergent acuity | ~40% | 47% | Over-sampled — comfortable, familiar territory |
| Lower acuity | ~21% | 27% | Over-sampled |
| Paediatric | ≥8% | 4% | Below the exam minimum — a structural gap |
| Geriatric | ≥6% | 5% | Marginal — needs a deliberate top-up |
| Image / pictorial items | Present | rarely filtered | Unknown — untracked is untested |
A composite of 79% hides all of this. The audit shows a candidate who is strong where the paper is easiest and thin exactly where acuity is highest and where the paediatric and geriatric minimums bite. That is the difference between practice data and readiness, and no dashboard average will surface it for you — you have to build the matrix. Our completion-is-not-coverage blueprint matrix walks through the mechanics for any exam.
The readiness test: what a credible signal actually requires
A number is only a readiness signal if it was produced under exam-like conditions. Require all five: the items were unseen (first attempt, not review); the block was timed at roughly a minute an item; the content was mixed and blueprint-proportioned rather than a single weak domain; you worked with no assistance (no visible explanations, no search, no tutor mode); and the sample was large enough — a couple of hundred items across a full blueprint sweep, not a lucky 20. Miss any one of those and you have a practice figure, not a readiness estimate.
Override rules: force what the algorithm under-serves
Adaptive and self-directed study both leave predictable blind spots. Override them deliberately. Force a minimum weekly quota of critical-acuity resuscitation items even when the feed judges you competent. Force paediatric and geriatric blocks to clear the ≥8% and ≥6% floors, because these are the first areas an adult-medicine-heavy stream neglects. Force image and pictorial items — ECGs, radiographs, rashes, ultrasound stills — because visual diagnosis rarely surfaces on its own. And schedule the low-volume, high-yield strands the feed treats as satisfied after a handful of items: toxicology, environmental emergencies, and the ethics, safety and medico-legal material bundled into the physician-task competencies.
Worked example: turning a dashboard into next week's quotas
Suppose your EM bank shows: first-attempt accuracy 74%, repeat accuracy 91%, coverage 62%, paediatric attempts 4%, critical-acuity attempts 18%, and a vendor pass likelihood of "high". Ignore the pass likelihood — it is a spurious comfort. Read the structure instead: the 17-point gap between repeat and first-attempt accuracy tells you the composite is inflated by review; the coverage figure is irrelevant next to the fact that critical acuity and paediatrics are both under the blueprint.
Convert that directly into quotas for the week, not a prediction:
- 40 critical-acuity items, unseen, timed, no assistance — closing the biggest structural gap.
- 25 paediatric items to push attempts from 4% toward the ≥8% floor.
- 15 geriatric items to hold above ≥6%.
- 20 image-based items filtered specifically for pictorial content.
- One 100-item mixed, blueprint-proportioned, timed block at week's end, scored on first-attempt accuracy only.
That final block is your readiness instrument. If unassisted first-attempt accuracy on a proportioned mix holds up across two or three such blocks, you have a signal. A single number on the home screen is not that.
A seven-day pattern: one job per tool
Give each resource one job. Use your EM bank — the AMBOSS-branded Knowledge+ EM product, Rosh Review, or whichever you hold — as the learning and spaced-review engine, where you read explanations, build category quizzes and close knowledge gaps. Use iatroX as the unseen-measurement layer, where you run fresh, timed, blueprint-proportioned blocks that no explanation has touched, to check whether learning has transferred. iatroX makes no claim on any competitor's proprietary algorithm; it simply provides the clean, unseen sample your own bank cannot, because you have already learned from those items.
- Days 1–2: learn in your EM bank on the two weakest EM Model domains; explanations on.
- Day 3: 25 paediatric items in your bank; force the age quota.
- Day 4: 50 unseen, timed items in iatroX, mixed; score first-attempt only.
- Day 5: image and toxicology override blocks in your bank.
- Day 6: 100-item unseen, timed, proportioned iatroX block; no assistance.
- Day 7: audit both dashboards against the EM Model; set next week's quotas.
This is the two-Q-bank rule applied to ABEM: a learning bank and a measurement bank, never duplicating the same items in both roles.
Decision checklist: continue, supplement, switch or stop
Continue with your current EM bank if first-attempt accuracy on unseen, timed, proportioned blocks is trending up and your attempts now match the acuity and age weighting. Supplement with a dedicated unseen-measurement layer if your only figures come from reviewed or assisted items — you cannot self-certify readiness from data you learned on. Switch only for a measurable coverage gap, not novelty: if your bank genuinely under-serves paediatrics, geriatrics or imaging and you cannot force enough volume, a different bank is justified. Stop a resource when it has become pure repeat-accuracy inflation — high familiar-item scores, flat first-attempt accuracy, no new coverage. Sunk cost is not a reason to keep answering questions you have memorised. Anchor every one of these calls to a measurable gap, and remember that your Q-bank percentage is not your exam score.
Bottom line
For ABEM specifically, the headline is a coverage finding, not an analytics tip: NEJM Knowledge+ does not publish an emergency medicine bank, so there is no NEJM Knowledge+ ABEM dashboard to read. Its EM-shaped successor lives inside AMBOSS Knowledge+, and it is a reasonable option to include in a revision stack — but whichever EM bank you use, the discipline is identical. Separate first-attempt from repeat accuracy, audit your attempts against the EM Model's acuity, paediatric and geriatric weighting, and reserve your readiness judgement for unseen, timed, unassisted, proportioned blocks of adequate size. Do that, and you will never mistake a comfortable practice average for the exam it does not predict.
FAQ
Is NEJM Knowledge+ enough for ABEM on its own? No, and not for the reason candidates expect — NEJM Knowledge+ does not publish an emergency medicine or ABEM Qualifying Examination product at all, so it cannot be enough on its own for this exam. Its board-review catalogue covers internal medicine, family medicine and paediatrics; for ABEM you need an EM-specific bank, such as the AMBOSS-branded Emergency Medicine (ABEM) product that now sits under the same corporate umbrella, verified on its own product page.
Which ABEM component does NEJM Knowledge+ not reproduce well? All of the ABEM components, because it does not cover the specialty; and even the EM-specific successor bank, like every multiple-choice resource, does not reproduce the Oral Certifying Examination. The oral exam assesses clinical care cases and communication and procedure scenarios interactively, and no single-best-answer qbank — NEJM Knowledge+, AMBOSS or otherwise — simulates that format.
How many NEJM Knowledge+ questions should I complete per day for ABEM? There is no meaningful per-day count for NEJM Knowledge+ against ABEM because it holds no ABEM questions. Working in whichever EM bank you actually own, a sustainable target for most candidates is 40 to 60 items a day, weighted toward first-attempt, unseen practice and forced quotas in critical-acuity, paediatric and geriatric content rather than raw volume in comfortable domains.
When should I stop using NEJM Knowledge+ and move to mixed mocks? If you are studying ABEM, you are not using NEJM Knowledge+ in the first place; the real question is when to move from single-domain learning to mixed mocks in your EM bank. The trigger is when your attempted distribution matches the EM Model and your first-attempt accuracy on proportioned blocks has stabilised — at that point, switch your emphasis to full-length, timed, unseen mixed blocks.
How should I combine NEJM Knowledge+ with iatroX without duplicating practice? Since NEJM Knowledge+ is not part of an ABEM stack, the practical version is to combine your EM learning bank with iatroX by role: learn and review in the bank, then measure transfer on fresh, unseen, timed iatroX blocks you have not learned from. Keeping the two in separate roles is exactly how you avoid duplicating items and destroying your own calibration.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 19 July 2026. Question counts, features and prices attributed to NEJM Knowledge+, AMBOSS and other vendors are vendor-reported and change without notice; confirm the current figures on each product page before purchase. Disclosure: iatroX operates its own question bank and clinical-knowledge platform and therefore competes with the products discussed here; this audit confines iatroX to the unseen-measurement role that the audited products do not claim, and leads with the coverage finding rather than a recommendation. Corrections are welcome via the feedback route on iatrox.com.
References: ABEM Qualifying Examination (abem.org/get-certified/qualifying-exam/); ABEM EM Model (abem.org/resources/em-model/); NEJM Knowledge+ (knowledgeplus.nejm.org); AMBOSS Knowledge+ Emergency Medicine (ABEM) board review (amboss.com/us/board-review/emergency-medicine); iatroX ABEM guide (iatrox.com/abem-emergency-medicine); "Your Q-Bank Percentage Is Not Your Exam Score" (iatrox.com/blog/qbank-percentage-not-your-exam-score); iatroX comparison hub (iatrox.com/compare).