The basic format
The AMC MCQ Examination Specifications describe a computer-administered examination of 150 multiple-choice questions delivered over 3.5 hours in one session. Every question has five options, and there is only one best answer. The AMC's instruction to candidates is to read each question carefully to understand what is being asked, then select the option they consider to be the most appropriate.
Note the word 'most'. Several options may be reasonable in real life. The exam asks which one is best for this patient, at this moment, in Australia. This is a single-best-answer exam, not a true-or-false test of isolated facts.
The exam is fully integrated. Questions from all six patient groups are mixed, so you may move from a newborn with jaundice to an ethics dilemma to an older man with hip pain in three consecutive screens. Practise that switching. Studying one subject at a time is useful early, but your final weeks should use mixed blocks.
- 150 questions in a single 3.5-hour session.
- Five options, one best answer.
- Mixed disciplines throughout.
- Some questions include images; the AMC suggests reading the question first, then studying the image, so you know what to look for.
Scored questions and new questions
You may read online that a fixed number of questions are unscored pilot items. Go by the AMC's current wording instead. The specifications (version 8, September 2025) say at least half of the questions are taken from a pool of previously calibrated questions, and that new questions and candidate responses are inspected and calibrated before being used for scoring.
The practical message is simple. You have no way of knowing whether a question is a new one. A strange or very hard question may be new, or it may be a calibrated item at the top of your ability range. Either way, spending five minutes on it hurts you. Give every question your best reasoned answer within your time budget and move on.
The AMC also states that candidates should complete all 150 questions in the 3.5 hours. Not completing all items may result in a 'Fail – insufficient data to obtain result'. Finishing is part of passing.
How the adaptive test chooses your questions
The AMC explains the core rule in its specifications. The first question is selected at random from the pool. If you answer correctly, the next question is more difficult. If you answer incorrectly, the next one is less difficult. Over 150 questions the test settles on the difficulty level you can answer reliably.
Because of this, every candidate sits a different exam. Your friend's exam may have felt easier because it was easier for them at that moment, not because they had a lucky paper.
Three rules follow from the adaptive design, and all three are stated in the AMC specifications. You must answer each question before the next is presented. Skipping is not permitted, and you cannot return to unanswered questions. Answers cannot be reviewed or changed once submitted. The test needs your answer to decide what to show you next, so there is no going back.
- Correct answer: the next question gets harder.
- Incorrect answer: the next question gets easier.
- No skipping and no returning.
- No changing answers after you submit.
The 0–500 scale and the pass standard
Results are reported on a scale from 0 to 500, and the AMC says a passing score is represented on that scale at 250. This is a scaled score. It is not a percentage and it is not a count of correct answers.
In December 2025 the AMC announced a slight increase to the MCQ pass standard for exams from 2026 onwards. Results are still reported on the 0–500 scale with the pass described as 250. The new, slightly higher cut score is built into the scale. The AMC's FAQ addresses the obvious worry directly: candidates will not have to answer a greater percentage of questions correctly, because the adaptive test selects questions targeted to each candidate's ability.
Why set the standard where it is? The AMC says the pass standard is set at the level expected of graduating medical students in Australia, and that it reviews the standard periodically, including through benchmarking with Australian medical schools. If you think like a safe new Australian intern, you are aiming at the right target.
Marks are not deducted for incorrect answers. A blank answer cannot happen anyway, because you must answer to move on, but this rule means a reasoned guess is always better than wasting time hoping for certainty.
- Scale: 0–500.
- Pass: 250, with the underlying standard raised slightly from 2026.
- Each result is judged against the standard in place when that exam was sat.
- No negative marking.
What adaptive scoring means for your strategy
A run of hard questions is usually a good sign. It means you have been answering correctly and the test has moved you up. Candidates who panic at this point start second-guessing and slowing down, which is the worst response. Stay with your method.
A run of easy questions does not mean you are failing. Difficulty moves both ways all the time. Do not try to read the test. Read the patient.
Accuracy on each question matters more than speed on any single one, but the clock is fixed. Set a budget. With 210 minutes for 150 questions you have an average of about 84 seconds per question. Long vignettes will take longer and short knowledge questions less. Check the on-screen timer at fixed points rather than after every question.
Do not 'save' hard questions for later. There is no later. If you are stuck between two options, use the exam's own logic: which option is safest, which addresses the most immediate threat, and which matches what an Australian guideline would advise first? Choose, submit, and reset for the next screen.
- Budget about 84 seconds per question on average.
- Check the timer at set points, for example after every 25 questions.
- If stuck after about two minutes, eliminate, choose, and move on.
- Do not interpret difficulty as a sign of passing or failing.
What a typical question looks like
Most questions open with a short clinical vignette: the patient's age and sex, the setting (general practice, emergency department, ward, rural hospital), the presenting problem, and key findings. A single lead-in then asks one thing, such as the most likely diagnosis, the most appropriate next step, the most appropriate investigation, or the most appropriate management.
The setting is often the most important detail. A patient with a suspected ST-elevation myocardial infarction in a rural town hours from a catheter lab needs a different answer from the same patient at a metropolitan hospital. A 15-year-old asking for contraception raises questions of capacity and confidentiality, not only pharmacology.
The AMC clinician tasks map onto lead-ins. Data gathering questions ask what to ask, examine or test. Data interpretation questions ask what a result or picture means. Management questions ask what to do. Our syllabus page lists these tasks and patient groups in detail.
- Read the lead-in first, then the stem, so you know what you are looking for.
- Note the setting and any time pressure.
- Look for red flags that change the 'next step'.
- Answer the question asked, not the question you expected.
On-screen tools and the testing room
The AMC specifications state that a timer clock is displayed on screen and that keystrokes are saved every three seconds, so a technical fault does not wipe your work. If there is a problem, raise your hand to alert the invigilator.
You cannot bring paper, pens, pencils, rulers or tissues into the exam room. A whiteboard and pen are provided for notes and collected at the end. Practise with a small whiteboard or a single sheet at home so jotting down a quick 2×2 table or a fluid calculation feels normal.
If a technical problem such as a power outage or computer failure disrupts your exam, the AMC states it will consider extenuating circumstances. Its MCQ Results Panel may analyse partial responses and, if no conclusion can be reached, may arrange a new exam at no additional fee.
| AMC rule | What it means | Your action |
|---|---|---|
| 150 questions in 3.5 hours | About 84 seconds per question on average | Practise timed blocks at this pace |
| Adaptive difficulty | Correct answers lead to harder questions | Ignore how hard it feels; focus on each patient |
| No skipping or returning | Every answer is final | Decide within your time budget, then move on |
| No negative marking | Wrong answers cost nothing extra | Always choose your best remaining option |
| Complete all 150 | An unfinished exam can fail for insufficient data | Watch the clock at fixed checkpoints |
| Single best answer | More than one option may look reasonable | Pick what an Australian intern would do first |
| Whiteboard provided | No paper or pens of your own | Practise brief notes on a small whiteboard |