This audit is for the MCCQE Part I candidate weighing AiMedQs as a low-cost adaptive feed. Used well, it is a reasonable engine for drilling weak areas cheaply. Its principal limitation is that an adaptive feed optimises for your comfort with its own questions, not for a blueprint-fair readiness signal, and AiMedQs publishes no validated percentile or pass prediction. Treat it as a training feed, and keep your measurement somewhere unseen.
What AiMedQs offers for MCCQE Part I right now
The vendor styles itself AiMedQs. Figures below are vendor-reported and were checked on 19 July 2026; confirm them on aimedqs.com before you rely on them.
| Item | Vendor-reported (19 July 2026) | Audit note |
|---|---|---|
| MCCQE1 question volume | "Over 500 MCCQE1-specific practice questions" in the FAQ; a header figure of "1,300+ practice MCQs" spans all exams | The two numbers are not the same claim; verify the MCCQE1-specific count on the product page |
| AI / adaptive features | "Adaptive AI Question Selection", a "Spaced Repetition Engine", "Detailed Explanations" | Mechanism is not independently documented; treat as a weak-area prioritiser, not a validated algorithm |
| Access / pricing | Starter ~$29/mo (500 questions/month), Pro ~$49/mo (2,000/month), Premium ~$79/mo (5,000/month); 30 free questions, no card | Note the tiers cap monthly questions, not total bank size |
| Components covered | Multiple-choice MCCQE Part I content | See the exam-anchor caveat below |
One finding belongs at the top. On the day of checking, AiMedQs's own exam description referred to the MCCQE Part I as around 210 MCQs in a single session with the CDM component removed "in 2024". That is out of date. The MCC moved to the current two-section, 230-MCQ multiple-choice format in April 2025. When a vendor's own format summary lags the exam body, take its other claims as prompts to verify rather than as settled fact.
Exam anchor: what the adaptive feed has to cover
Since April 2025 the MCCQE Part I is multiple-choice only: 230 MCQs across two sections of 115 (including unscored pilots), a maximum of two hours 40 minutes per section, and three to five options per item (verify on mcc.ca). Content follows the MCC Objectives across Dimensions of Care (health promotion and illness prevention, acute, chronic, psychosocial) and Physician Activities (assessment and diagnosis, management, communication, and the legal, ethical and organisational, or CLEO, domain). The official MCC preparatory products, the Preparatory Examination (230 MCQs) and PE-Lite (115 MCQs across two forms), are the calibration standard; distinguish those official requirements from any third-party claim.
Define every metric before you trust it
An adaptive dashboard is only as honest as its definitions. Before you read an AiMedQs number, pin down what it means: first-attempt accuracy (your score on items you have never seen, the score that most cleanly tracks learning); repeat accuracy (score on re-served items, which inflates as recognition); percentile (position against other users, meaningful only if the cohort and denominator are stated); predicted score (a model output, unvalidated unless the vendor publishes its basis); coverage (proportion of the blueprint attempted, not proportion of the bank finished); difficulty (whose rating, and on what scale); and time per item (the pacing metric that most directly predicts a section-timed exam). If a metric's definition is not published, treat the metric as decorative.
What the adaptive engine is actually optimising
An adaptive feed that "prioritises your weak areas" is optimising to raise your score on the questions it keeps showing you. That is useful for drilling, but it has three side effects. It over-samples your weak domains, so the mix you see stops resembling the blueprint mix you will sit. It re-serves related items, so your rising accuracy is partly recognition, not new competence. And because everyone's feed is different, your percentage is not comparable to anyone else's, or to your own from last week. The consequence is simple: a number that climbs on an adaptive feed is not evidence of readiness on a blueprint-fair, unseen mock. It is evidence that the feed is working as designed.
Blueprint audit: compare your mix, not your average
Do not trust the home-screen average. Export or tally your attempted questions by dimension and compare that distribution against the MCC Objectives' weighting. Adaptive feeds routinely leave low-volume, high-stakes material thin: CLEO and ethics, population health, image and data interpretation, and calculation items. If your feed has served you 400 cardiology and respiratory items and forty CLEO items, your average tells you about the former and nothing reliable about the latter, which the exam will still test. The blueprint-coverage matrix method is the tool for this comparison.
Readiness test: the conditions for a credible signal
A score is only a readiness signal when five conditions hold at once: the items are unseen (not re-served by the feed), the block is timed to section pace, the mix is blueprint-representative rather than weak-area-weighted, you use no assistance or explanations mid-block, and the sample is large enough to be stable (a single 20-item block is noise). An adaptive feed, by construction, breaks the first and third conditions. That is not a flaw in AiMedQs; it is a reason to measure readiness elsewhere.
Override rules: force what the feed under-serves
Because the algorithm chases your weak areas as it defines them, you must manually override it to cover what it under-samples. Force blocks in CLEO, ethics and professionalism; population and preventive health with Canadian screening intervals; image, ECG and data interpretation; and calculation items. Also force periodic mixed, full-blueprint blocks even when the feed would rather keep drilling your worst domain, because the exam will not adapt to you.
Worked dashboard example
Suppose after two weeks your AiMedQs dashboard reads: first-attempt accuracy 68 per cent overall, chronic-disease management 74 per cent, acute care 71 per cent, psychosocial 58 per cent, CLEO 52 per cent, with 60 per cent of your attempts in acute and chronic domains and 8 per cent in CLEO. The wrong reading is "I'm at 68 per cent, nearly there." The right reading turns that into next week's quotas: cap acute and chronic blocks, force 40 per cent of volume into psychosocial and CLEO, add one image and one calculation block, and book an unseen mixed measurement block for the weekend. No pass prediction is made or implied; the dashboard sets quotas, not verdicts.
Worked example: a seven-day IMG plan
For an international graduate balancing content review with Canadian conventions, give AiMedQs one job (weak-area drilling) and iatroX another (unseen transfer measurement). Days 1 to 2: AiMedQs blocks forced onto psychosocial care and CLEO, reviewing every rationale against Canadian norms. Days 3 to 4: AiMedQs image, data and calculation blocks. Day 5: consolidate logged errors, no new questions. Day 6: an unseen, timed, mixed iatroX block, no assistance, to see whether the drilling transferred. Day 7: review the iatroX errors and reset next week's quotas. No claim is made about either platform's internal algorithm; the design simply keeps training and measurement on separate surfaces. If you run a second bank, follow the two-Q-bank rule to avoid duplicating items.
Decision checklist: continue, supplement, switch or stop
Continue with AiMedQs if your unseen scores are rising and the low cost suits you. Supplement it if a blueprint audit shows persistent thin domains that the feed will not fill, adding official MCC material and one unseen measurement layer. Switch if the feed keeps re-serving recognisable items and your unseen scores stall despite a climbing dashboard. Stop adding new questions when unseen, timed blocks have plateaued across dimensions and your errors are careless rather than conceptual. Base each move on a measured gap, not on novelty or the sunk cost of a subscription.
Three mistakes this audit is designed to stop
The first is reading the adaptive average as a readiness score. A number that climbs on a feed built to drill your weak areas measures your comfort with that feed, not your standing on a blueprint-fair, unseen section, and the two can diverge widely. The second is trusting the platform's own exam description: on the day of checking it still framed the MCCQE Part I as a single-session, roughly 210-MCQ exam, a format the MCC retired in April 2025, so a candidate who calibrates to the in-app description will rehearse the wrong structure. The third is letting the algorithm set your blueprint mix. Because it chases the weaknesses it can see, it will happily leave CLEO, ethics, population health and modality items thin while your organ-system scores rise, producing a confident average over an incomplete map. Each mistake has the same antidote: force the blueprint yourself, verify the format against the MCC, and measure readiness on unseen material.
Bottom line
AiMedQs is a serviceable, inexpensive adaptive drilling feed for MCCQE Part I, with the honest caveats that its question count is ambiguously reported, its own format description has lagged the exam, and its adaptive number is a training metric rather than a readiness signal. Use it to raise weak-area competence, override it to cover the blueprint, and keep your true readiness measurement on unseen, blueprint-fair blocks.
Frequently asked questions
Is AiMedQs enough for MCCQE Part I on its own? For most candidates it is not sufficient alone, because its MCCQE1-specific volume is modestly and ambiguously reported (over 500 questions in the FAQ against a cross-exam header figure), and an adaptive feed cannot supply the blueprint-fair, unseen measurement you need to judge readiness. It works best as one component: a cheap weak-area drill, paired with official MCC calibration and an unseen measurement layer.
Which MCCQE Part I component does AiMedQs not reproduce well? The applied CLEO domain, legal, ethical and organisational items, together with population and preventive health framed to Canadian norms, is where an adaptive feed is weakest, because those low-volume categories are exactly what the algorithm under-samples while it chases your organ-system weaknesses. Image, data and calculation items are the second common gap. Force these manually; do not wait for the feed to surface them.
How many AiMedQs questions should I complete per day for MCCQE Part I? There is no universal number, and the tiers cap monthly rather than daily volume, so anchor to your calendar: enough that every blueprint dimension gets a meaningful weekly sample and you can review every rationale, rather than a daily maximum you cannot consolidate. Quality of review and blueprint spread beat raw counts; a smaller number of fully understood, well-distributed items is worth more than a large undigested tally.
When should I stop using AiMedQs and move to mixed mocks? Move to unseen mixed mocks once your forced-blueprint blocks are stable across dimensions and the feed starts re-serving items you recognise, because at that point the adaptive engine is measuring memory, not learning. In the final two to three weeks the higher-value activity is timed, blueprint-representative mocks and official MCC forms, with AiMedQs relegated to spot-drilling any single domain that remains soft.
How should I combine AiMedQs with iatroX without duplicating practice? Give each platform a distinct job: AiMedQs for weak-area drilling, iatroX for unseen, blueprint-mapped measurement, and never run the same item on both. Practically, drill on AiMedQs during the week, then sit a fresh iatroX block at the weekend as your read-out; if you notice overlapping content, rotate topics so the measurement block stays genuinely unseen, following the two-Q-bank rule to protect the comparability of your scores.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 19 July 2026; AiMedQs question counts, prices and feature names are vendor-reported as at that date and should be reconfirmed on aimedqs.com, which on the day of checking still described an outdated MCCQE Part I format. Disclosure: iatroX operates a competing MCCQE Part I question bank; to avoid conflict, this audit confines iatroX to the unseen-measurement job AiMedQs does not claim to provide, and makes no claim about any platform's proprietary algorithm. Corrections are welcome via the feedback route on iatrox.com.
References: Medical Council of Canada, MCCQE Part I multiple-choice and 2025 change pages, and preparatory products (mcc.ca); AiMedQs MCCQE1 product and pricing pages (aimedqs.com), vendor-reported; iatroX internal resources including Your Q-Bank Percentage Is Not Your Exam Score and the iatroX comparison hub.