Every ACEM written examination report contains a short section headed "Candidate Feedback". It explains that, for each SAQ, examiners record which of six standard factors explain why an unsuccessful candidate lost marks. The six factors have not changed in eleven sittings. They are the closest thing that exists to an official statement of how the SAQ paper is failed, and most candidates have never read them.
The six examiner feedback factors (paraphrased from the reports):
1. Parts of the SAQ had no answer.
2. A significant number of incorrect or irrelevant answers were provided.
3. The answer used vague or non-consultant-level terminology.
4. Some of the mandatory answers were missing, so the answer lacked prioritisation.
5. More than one answer on one line, or answers in excess of the number required, were not marked.
6. A response was provided or omitted that would cause significant harm to the patient, so the zero-score rule was applied to a sub-question.
Why technique decides the SAQ paper
Before the six factors, the arithmetic that makes them matter. The reports publish the SAQ pass mark, the cohort mean and the standard deviation for every sitting. In eight of the 11 sittings from 2021 to 2026 the average candidate finished within eight marks of the pass mark, out of 360. In 2022.1 the gap was 0.3 marks. One standard deviation on the SAQ is about 25 marks, or 7% of the paper.
Put those together. A candidate who writes three answers on one line in two questions, leaves one sub-question blank because they ran out of time, and gives "IV fluids" where the model answer wanted a dose can lose 10 to 15 marks without a single gap in their knowledge. That is more than the distance between the average candidate and a fail. Our pass-rate analysis has the full numbers; the short version is that the SAQ paper is the one that decides the exam, and technique is a large part of what it measures.
Factor 1: parts of the question had no answer
What it means. A sub-question, or several, left blank. A blank scores zero; a wrong answer also scores zero, so a blank is never better than an attempt (with the single exception of factor 6, below).
Why it happens. Almost always time. The paper is 26 to 28 questions in 180 minutes, two marks a minute. Candidates who spend ten minutes on an early 12-mark question they know well are borrowing from a question at the end of the booklet that they will never see. The second cause is a question on an unfamiliar topic, where the candidate freezes rather than writing what the stem allows them to infer.
What to do. Budget six minutes for a 12-mark question and nine for an 18-mark question, and move on when the time is up even if you have more to say. When a topic is unfamiliar, answer the sub-questions you can from first principles: a structured assessment, the investigations you would order for any patient with that presentation, the disposition. Partial marks on a hard question are worth exactly as much as partial marks on an easy one. Never hand in a blank sub-question.
Factor 2: a significant number of incorrect or irrelevant answers
What it means. The candidate wrote plenty, but much of it was wrong, or was a correct answer to a different question. The examiner cannot award marks for content that is not in the model answer for that sub-question.
Why it happens. Not reading the stem. A question that asks for "clinical features that would support the diagnosis" is answered with investigations. A question that specifies a 4-year-old is answered with adult doses. A question that says "list the complications of the procedure" gets a description of the procedure. Under pressure, candidates answer the question they prepared for instead of the one on the page. The other common cause is padding: writing more items than you are confident of, in the hope that some will land. On the ACEM paper this backfires (see factor 5).
What to do. Underline the operative words in every sub-question before writing: the patient's age, the setting, whether it asks for features, investigations, management or complications, and how many items it wants. Write to that question and only that question. If the stem gives you a rural hospital, a pregnant patient or a child, every answer should reflect it.
Example: a 3-year-old with a suspected foreign body aspiration. "List four features on history that increase your concern."
Weak: "Chest X-ray, bronchoscopy, oxygen saturation, stridor."
Better: "Witnessed choking episode; sudden onset cough or wheeze while eating or playing with small objects; persistent unilateral wheeze or cough after an initial event; history of previous aspiration or developmental delay."
The weak answer contains two investigations and a sign. Only "stridor" is arguably a feature, and it is an examination finding, not history. It would score at most one mark of four.
Factor 3: vague or non-consultant-level terminology
What it means. The examiner is marking against a standard: would a newly qualified FACEM answer this way? "Give fluids", "treat the pain", "refer to surgeons", "start antibiotics" and "manage airway" are not consultant-level answers. They are headings. The model answer has the specifics under each heading, and the marks are attached to the specifics.
Why it happens. Partly time, partly habit. Registrars spend their working lives giving verbal handovers where "fluids and antibiotics" is enough because everyone knows what is meant. On paper, the examiner is instructed not to fill in the gaps for you.
What to do. For any treatment, write the drug, the dose (weight-based for children), the route and the endpoint or reassessment. For any referral, write the specialty, the urgency and what you are asking them to do. For any investigation, write what you are looking for. This costs seconds and it is where a large fraction of the marks live.
Example: a hypotensive adult with suspected septic shock. "Outline your initial management."
Weak: "IV fluids, antibiotics, oxygen, monitor, ICU referral."
Better: "30 mL/kg crystalloid bolus (e.g. 2 L Hartmann's for a 70 kg adult) with reassessment of BP, lactate and perfusion after each litre; blood cultures then IV piperacillin-tazobactam 4.5 g (or per local guideline) within the hour; titrated oxygen to SpO₂ 94–98%; noradrenaline via central or large peripheral access if MAP under 65 after fluid; early ICU referral for vasopressor support and source control discussion."
Both answers are "correct". The first would score perhaps two of eight marks; the second would score most of them. The difference is not knowledge.
A useful test when practising: read your answer back and ask whether a junior doctor could act on it without asking you a question. If they would need to ask "how much?", "which one?" or "how urgently?", the examiner would too.
Factor 4: mandatory answers missing, so the answer lacked prioritisation
What it means. The model answer for most sub-questions contains mandatory items: the things a consultant must say. A sub-question worth four marks might have two mandatory items and a list of acceptable others. Candidates who list four true but minor points, and omit the mandatory ones, score poorly even though nothing they wrote was wrong. The word "prioritisation" in the factor is the key: the examiner is looking for the most important answers first.
Why it happens. Writing in the order things come to mind. A candidate asked for the key steps in managing a patient with a tension pneumothorax who begins with "IV access" and "analgesia" and reaches decompression fifth has demonstrated that they do not know what matters most, whatever else they wrote.
What to do. Spend the first 20 seconds of every sub-question deciding what the mandatory items are, and write those first. Order everything else by clinical priority: life threats, then time-critical treatment, then the rest. When a question asks for "the three most important", it is telling you that the model answer has three mandatory items and that items four onwards will not be read.
Factor 5: more than one answer on a line, or more answers than asked for
What it means. This is the most mechanical of the six and probably the most avoidable. Answer booklets provide a set number of lines. If a sub-question asks for three items and gives three lines, the examiner marks three lines. A fourth item is not marked. Two items on one line are treated as one answer, and if one of them is wrong the line may score zero.
Why it happens. Hedging. A candidate who is unsure which of two differentials the examiner wants writes both on the same line, or lists five investigations where three were asked for, reasoning that the examiner will pick the right ones. The rule exists precisely to stop that: it forces candidates to commit.
What to do. One answer per line, exactly the number asked for, the most important first. If you are torn between two answers, choose the one a consultant would say first. Practise on printed answer templates with the real line counts, so the constraint is familiar before exam day.
Example: "List three investigations you would perform and what you are looking for."
Weak: line 1: "ECG / troponin – ischaemia"; line 2: "CXR, CT chest – dissection, PE, pneumothorax"; line 3: "bloods"; line 4: "bedside echo – effusion".
Better: line 1: "12-lead ECG – ST elevation, dynamic ischaemic change, arrhythmia"; line 2: "Bedside echocardiogram – pericardial effusion, RV dilatation, regional wall motion abnormality"; line 3: "CT aortogram – aortic dissection flap or intramural haematoma".
The weak answer puts two investigations on line 1 and three on line 2, gives no target for "bloods", and puts a good answer on a fourth line that will not be marked.
Factor 6: the zero-score rule
What it means. If a candidate writes something, or omits something, that would cause significant harm to the patient, the examiner can award zero for that whole sub-question, regardless of the other content. It is the only factor that can wipe out correct answers you have already written.
Why it happens. Usually a reflexive answer that ignores the stem. Suxamethonium for a patient with a spinal cord injury from three weeks ago. A beta-blocker in cocaine-associated chest pain. Discharging a child in whom the stem has planted an NAI concern. Thrombolysis in a patient with a described contraindication. Omitting the airway from a resuscitation sequence. The stem often contains the detail that makes the standard answer dangerous, and candidates writing from memory miss it.
What to do. Read every stem twice, the second time looking specifically for the detail that changes the standard management: the age, the pregnancy, the allergy, the medication list, the time since injury, the renal function. Before you write any drug, ask whether anything in the stem contraindicates it. And in any resuscitation question, write the life-threat steps first even if they feel too obvious to mention. This is the one situation where a cautious blank is better than a confident wrong answer: if you are unsure whether a treatment is safe in the scenario described, leave it out.
Putting the six factors to work
The factors are more useful as a marking tool than as a list to memorise. After every practice SAQ, mark your own answer against a model answer and code each lost mark to a factor. Most candidates find that two or three of the six account for almost all their losses, and they are rarely the ones they expected. Common patterns:
- Candidates with strong knowledge who fail on factors 3 and 5: vague answers and hedged lines. They know the medicine and are not writing it down at consultant level.
- Candidates who read widely and fail on factor 4: they list true things without prioritising, and miss the mandatory items under a pile of acceptable ones.
- Candidates who fail on factor 1 in the last third of the paper: a time-management problem that no amount of extra reading will fix.
A practical routine for each question:
- Read the stem twice. Underline the age, setting, and any detail that changes standard management (factor 6).
- For each sub-question, underline what is asked for and how many (factors 2 and 5).
- Decide the mandatory items before writing, and write them first (factor 4).
- For every treatment, write drug, dose, route, endpoint; for every referral, who, how urgently and for what (factor 3).
- One answer per line, exactly the number asked for (factor 5).
- Watch the clock: six minutes for 12 marks, nine for 18. Move on with a partial answer rather than leaving a later question blank (factor 1).
The exam reports show the average candidate sitting on the SAQ pass line. Recovering ten marks of technique is the difference between the two sides of it. For the topics those marks are attached to, see what the SAQ paper actually tests.
Sources
The six factors are listed, in the same form, in the "Candidate Feedback" section of every ACEM Fellowship Written Examination Report from 2021.1 to 2026.1. Pass marks, cohort means and standard deviations are from the results sections and appendices of the same reports. The worked examples are illustrative and written by ACEMCQ; they are not model answers from the College. ACEMCQ is independent of ACEM and this guide is not endorsed by the College.
Frequently asked questions
How are ACEM SAQs marked?
Each question has a model answer with mandatory items. Examiners mark all SAQs on a single marking day, and for unsuccessful candidates they record which of six standard factors explain the lost marks. Those factors are listed in every ACEM written examination report from 2021 to 2026.
What is the zero-score rule in the ACEM SAQ?
If an answer includes, or omits, something that would cause significant harm to the patient, the examiner can award zero for that sub-question regardless of what else was written. It is one of the six feedback factors.
Why do candidates fail the ACEM SAQ paper?
The six examiner-recorded reasons are: parts of the question left blank; a significant number of incorrect or irrelevant answers; vague or non-consultant-level terminology; missing mandatory answers and poor prioritisation; more than one answer on a line or more answers than asked for; and a response that would cause significant harm, triggering the zero-score rule.
How much does SAQ technique matter?
A great deal. In eight of the 11 sittings from 2021 to 2026 the average candidate finished within eight marks of the SAQ pass mark, which is less than one 12-mark question. Marks lost to technique rather than knowledge decide the paper for a large share of candidates.