This audit is for emergency physicians relying on BoardVitals for the ABEM Qualifying Examination who want to know what its adaptive engine and analytics are actually optimising before they trust the dashboard. BoardVitals is a serviceable EM bank with adaptive testing, performance analytics and an exam-style simulator. Its principal limitation is one shared by every adaptive product: the engine optimises for your in-bank accuracy and engagement, and those numbers are not your exam readiness unless you audit coverage and measure on unseen, timed, mixed blocks.
What BoardVitals offers right now
BoardVitals is a multi-specialty board-review vendor whose emergency-medicine bank targets the ABEM Qualifying Examination and In-Training Exam. The figures below are vendor-reported as of 19 July 2026; confirm the live numbers and price on boardvitals.com.
| Feature | Vendor-reported position (19 July 2026) |
|---|---|
| Question volume | Over 900 EM board-review questions |
| Target exam | Follows the ABEM Emergency Medicine Qualifying Exam and In-Training Exam content outline |
| Adaptive behaviour | Adaptive testing that presents questions "tailored to your current competency level" |
| Analytics | Advanced performance tracking, comparison against a national average, and "AI-powered risk assessment" of at-risk subjects |
| Simulator | Exam simulator described as matching the ABEM interface |
| Access and price | Vendor-reported tiers — 1 month US$169, 3 months US$299, 6 months US$419; a "100% pass guarantee" applies subject to terms |
A note of proportion: at a vendor-reported 900-plus items, this is a more compact bank than some EM competitors. That is not disqualifying — a smaller, well-mapped bank can calibrate and cover — but it does mean the adaptive engine has less room to keep serving genuinely unseen items, which matters for the audit below.
The ABEM Qualifying Examination blueprint
The target is a computer-based, single-best-answer test at Pearson VUE of roughly 305 scored items, built on the EM Model: about twenty condition and organ-system categories over a set of physician tasks, with a published acuity frame of approximately 30% critical, 40% emergent and 21% lower-acuity presentations, minimums of at least 8% paediatric and at least 6% geriatric content, and image-based items. Passing leads to the Oral Certifying Examination. BoardVitals' "follows the ABEM content outline" is a vendor mapping claim to spot-check against ABEM's own specification.
Define every metric the platform shows
Before you act on a dashboard, define what each number is — and is not.
- First-attempt accuracy — your percentage on items the first time you see them. The only accuracy figure that approximates real readiness.
- Repeat accuracy — your percentage on items you have already attempted. Inflated by memory; near-useless as a readiness signal.
- Percentile / national average — where you sit among other BoardVitals users, not against ABEM's standard-setting.
- "Predicted" or risk-assessment output — a vendor model of your at-risk subjects; a study aid for prioritisation, not a pass forecast.
- Coverage — how much of the blueprint you have actually attempted. The metric most likely to be missing from the home screen.
- Difficulty — the bank's own difficulty label, not ABEM's.
- Time per item — the metric that most cheaply predicts trouble; guard your pace against the roughly one-minute-per-item exam tempo.
Explain the selection bias
An adaptive feed that serves questions "tailored to your current competency level" is, by design, over-sampling your weak areas and under-sampling your strong ones. That is useful for learning and corrosive for interpretation: your raw overall percentage on an adaptive feed is not comparable to your percentage on a mixed unseen block, because the two are drawn from different distributions. A percentage that dips when the engine feeds you hard items is not deterioration, and a percentage that climbs as you review familiar items is not readiness. Never compare an adaptive-feed average with a mixed-block average as if they measured the same thing.
Blueprint audit
Do not trust the home-screen average as evidence of coverage. Export or tally your attempted-question distribution and lay it against the official ABEM domain and acuity weighting. The questions to answer: has every EM Model category been attempted to a floor you set; does your paediatric attempt-share meet the blueprint's at-least-8%; does geriatric meet at-least-6%; and have you actually practised image items in proportion. With a compact bank, it is entirely possible to hit a high overall percentage while a whole category is thinly attempted. A blueprint-coverage matrix turns this from a guess into a check.
Readiness test
A dashboard number is only a credible readiness signal under specific conditions, and BoardVitals — like any single tool — cannot guarantee all of them from inside its own feed. For a signal you can trust, the block must be unseen (not items you have reviewed), timed (at exam pace), mixed (random across the blueprint, not topic-filtered), taken with no assistance (no notes, no pausing to look things up), and large enough to be stable (a handful of questions is noise). If any condition is missing, treat the number as practice feedback, not readiness.
Algorithm override rules
Because the engine optimises for accuracy and engagement, you must force it toward the material it will otherwise under-serve. Override the feed to guarantee minimum exposure to: low-volume EM Model domains that rarely surface in an adaptive queue; image-based items (ECGs, radiographs, clinical photographs); ethics, consent and communication content; and calculation-dependent items (anion gap, corrected values, weight-based paediatric dosing, drip rates). Set these as scheduled, deliberate blocks rather than hoping the algorithm surfaces them. The engine will not volunteer your blind spots; that is the one thing you must do by hand.
Worked dashboard example
Suppose your BoardVitals dashboard reads: overall 78%, first-attempt 71%, repeat 88%, "national average" 74%, and a risk flag on toxicology and paediatrics. Read it, do not celebrate it. The 88% repeat accuracy is memory and should be set aside. The 71% first-attempt is your real working figure. The 78% overall is a blend inflated by the repeats. "Above national average" tells you about other users, not about the ABEM standard. So next week's quotas write themselves: a toxicology block and a paediatric block to unique-item floors; an image and a calculation block, regardless of what the risk flag says, because those are the systematically under-surfaced categories; and one unseen, timed, mixed block — ideally on a second bank — to get a first-attempt figure the adaptive feed cannot inflate. No pass prediction is made or implied from these numbers; they set actions, not forecasts.
A seven-day worked example
One week using BoardVitals for adaptive drilling and iatroX for unseen transfer measurement. No proprietary-algorithm claims about either tool.
- Monday — 40 timed BoardVitals items in the two categories flagged at-risk; tag every miss by error type.
- Tuesday — Source reads for Monday's knowledge gaps; write three one-line rules.
- Wednesday — 30 unseen, timed iatroX items; record first-attempt accuracy and compare error types with Monday.
- Thursday — Override block: images plus calculations, the categories the adaptive feed under-serves.
- Friday — 40 BoardVitals items in the next weakest category; check paediatric and geriatric attempt-shares against the blueprint.
- Saturday — One timed random mixed block at exam length; log pacing and first-attempt accuracy.
- Sunday — Reconcile: adaptive-feed average versus unseen first-attempt average, and set next week's floors from the gap.
BoardVitals drives the adaptive learning; iatroX supplies fresh, unseen items so the figure you trust for readiness is one the adaptive engine cannot inflate. That separation of roles is the two-Q-bank rule.
Decision checklist: continue, supplement, switch or stop
- Continue BoardVitals while first-attempt accuracy on unseen blocks is still improving and coverage floors are unmet.
- Supplement with an unseen second bank once the compact library starts recycling familiar items — with a 900-plus-item bank, that point can arrive sooner than with larger banks.
- Switch your main volume source only for a measurable coverage gap against the blueprint, not for a discount or the pass guarantee.
- Stop active drilling at the exit criteria — coverage floors met, stable unseen first-attempt accuracy, comfortable pace, calibration against official material — not when the bank is "finished".
Three mistakes this workflow is designed to stop
First, quoting repeat accuracy or "above national average" as readiness; the first is memory and the second is a peer ranking, and neither maps to the ABEM standard. Second, trusting the adaptive feed to cover the blueprint for you; it optimises for accuracy, not coverage, and low-volume domains, images and calculations are exactly what it under-serves. Third, treating the pass guarantee as a readiness signal; it is a commercial term with conditions, not a measure of whether you are prepared.
Bottom line
BoardVitals gives you a workable adaptive EM bank and a dashboard full of numbers — the skill is knowing which numbers mean anything. First-attempt accuracy on unseen, timed, mixed blocks is the readiness signal; repeat accuracy and peer percentiles are not. Override the engine toward images, calculations and low-volume domains it will otherwise skip, audit your attempted distribution against the blueprint, and use iatroX to measure transfer on fresh items. Do that and the adaptive engine works for you; trust the home-screen average and it optimises you into a comfortable, misleading number.
Frequently asked questions
Is BoardVitals enough for ABEM on its own? BoardVitals can serve as a primary or supplementary bank, but at a vendor-reported 900-plus items it is more compact than some EM banks, which shortens the runway of genuinely unseen questions. Whether it is "enough" depends on your coverage against the blueprint and, crucially, on measuring readiness with unseen mixed blocks rather than the adaptive-feed average. Many candidates will want a second bank for volume and calibration.
Which ABEM component does BoardVitals not reproduce well? As a multiple-choice product it does not reproduce the ABEM Oral Certifying Examination, which assesses structured verbal case management rather than option selection. It targets the written Qualifying Examination and In-Training Exam; prepare for the oral separately with case-based verbal practice.
How many BoardVitals questions should I complete per day for ABEM? A sustainable target is roughly 40 to 60 timed items per day in the core months, weighted toward flagged-weak categories, with error-type tagging and pace mattering more than raw volume. With a compact bank, resist the urge to burn through it quickly — spacing your first attempts preserves more unseen material for genuine measurement.
When should I stop using BoardVitals and move to mixed mocks? Move to timed mixed blocks once every EM Model category has cleared its coverage floor and your first-attempt accuracy has been stable for two to three weeks. Because a smaller bank starts recycling sooner, you may reach the point of diminishing unseen items — and therefore the case for mixed mocks on a second bank — earlier than with a larger library.
How should I combine BoardVitals with iatroX without duplicating practice? Keep them in separate roles. BoardVitals is your adaptive learning and analytics engine; iatroX is your unseen-measurement layer, used in short timed blocks to get a first-attempt figure the adaptive feed cannot inflate. Do not re-test the same items across both — the value of the second bank is fresh material, so you compare error types and unseen accuracy rather than re-answering questions you have already met.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 19 July 2026. Question counts, adaptive and analytics features, access periods, prices and any pass guarantee are vendor-reported and change without notice; confirm the current details on boardvitals.com before relying on them. Disclosure: iatroX operates a competing question bank, so this audit confines iatroX's role to a job BoardVitals' adaptive feed does not do for you — supplying unseen, blueprint-stratified ABEM items for transfer measurement — and takes no position on the pass-guarantee terms beyond noting they are commercial, not a readiness measure. Corrections are welcome via the feedback route on iatrox.com. References: ABEM Qualifying Examination content specifications and the EM Model (abem.org); BoardVitals Emergency Medicine Board Review (boardvitals.com/emergency-medicine-board-review); iatroX ABEM bank (iatrox.com/abem-emergency-medicine) and comparison hub (iatrox.com/compare); "Your Q-Bank Percentage Is Not Your Exam Score" (iatrox.com/blog/qbank-percentage-not-your-exam-score); and the blueprint-coverage matrix guide (iatrox.com/blog/question-bank-completion-is-not-coverage-how-to-build-a-blueprint-coverage-matrix-for-any-medical-exam).