ACEM publishes a report after every Fellowship Written Examination. Each one contains the cohort size, the pass rate for each paper, the pass mark, the cohort mean and standard deviation, and the score achieved on every SAQ. Read one report and you get a snapshot. Read all eleven side by side and a much more useful picture appears: which paper decides the result, how close the average candidate sits to the line, and how the borderline review has quietly changed who passes.
Pass rates for every sitting, 2021.1 to 2026.1
Candidates must pass both papers to pass the exam, subject to the borderline review described below. The table gives the figures exactly as reported in the results section of each report.
| Sitting | Date | Candidates | SCQ pass | SAQ pass | Overall pass |
|---|---|---|---|---|---|
| 2021.1 | 21 May 2021 | 186 | 74.2% | 64.5% | 59.7% |
| 2021.2 | 19 Nov 2021 | 256 | 69.5% | 79.3% | 68.0% |
| 2022.1 | 20 May 2022 | 198 | 69.7% | 54.5% | 55.1% |
| 2022.2 | 18 Nov 2022 | 234 | 90.2% | 62.8% | 65.8% |
| 2023.1 | 19 May 2023 | 171 | 88.9% | 60.2% | 65.5% |
| 2023.2 | 17 Nov 2023 | 219 | 88.6% | 88.1% | 86.3% |
| 2024.1 | 17 May 2024 | 204 | 86.8% | 65.2% | 71.6% |
| 2024.2 | 15 Nov 2024 | 228 | 83.3% | 68.9% | 75.9% |
| 2025.1 | 23 May 2025 | 202 | 76.7% | 59.9% | 62.4% |
| 2025.2 | 21 Nov 2025 | 254 | 84.3% | 63.0% | 68.5% |
| 2026.1 | 22 May 2026 | 199 | 87.4% | 66.8% | 78.9% |
| All 11 sittings | 2021–2026 | 2,351 | 81.7% | 67.1% | 69.1% |
The SAQ paper decides the exam
Two things stand out in the chart. First, the SCQ pass rate jumped from around 70% in 2021 and early 2022 to between 83% and 90% from 2022.2 onwards, and has stayed there in every sitting except 2025.1 (77%). The reports do not explain the change, and we will not guess at a cause. What matters for a candidate is the consequence: for most people sitting today, the SCQ paper is not where the exam is won or lost.
Second, the SAQ pass rate has not moved. It sat between 60% and 69% in eight of the 11 sittings. The exceptions were 2021.2 (79%), 2022.1, when it fell to 54.5% and dragged the overall pass rate to its lowest point in the series (55.1%), and 2023.2, when it rose to 88% and the overall rate hit 86%. In every sitting the overall pass rate lands within a few points of the SAQ pass rate. If you want to predict the headline number for a sitting, the SAQ result is all you need.
That single fact should shape how you allocate preparation time. Most candidates who fail the written exam fail it on the SAQ paper, and the SAQ is the paper where answer technique, not just knowledge, is being marked. Our companion guide on the six examiner feedback factors covers what that technique looks like.
Where the pass mark sits, and how close the average candidate is to it
ACEM sets the pass mark for each paper after the exam using the Modified Angoff method, then adds one standard error of measurement. The reports publish the resulting pass mark alongside the cohort mean and standard deviation, which lets us ask a question the reports themselves never answer: how far above the line is the typical candidate?
| Sitting | SAQ pass mark | SAQ cohort mean | Mean minus pass mark | SCQ pass mark | SCQ cohort mean | Mean minus pass mark |
|---|---|---|---|---|---|---|
| 2021.1 | 228/348 (66%) | 231.9 (67%) | +3.9 marks (0.17 SD) | 78/118 (66%) | 82.3 (70%) | +4.3 marks (0.54 SD) |
| 2021.2 | 206/348 (59%) | 222.5 (64%) | +16.5 marks (0.65 SD) | 78/116 (67%) | 81.6 (70%) | +3.6 marks (0.44 SD) |
| 2022.1 | 232/360 (64%) | 232.3 (65%) | +0.3 marks (0.01 SD) | 77/113 (68%) | 79.7 (71%) | +2.7 marks (0.39 SD) |
| 2022.2 | 229/355 (65%) | 234.6 (66%) | +5.6 marks (0.25 SD) | 79/116 (68%) | 87.3 (75%) | +8.3 marks (1.22 SD) |
| 2023.1 | 248/360 (69%) | 251.7 (70%) | +3.7 marks (0.15 SD) | 79/114 (69%) | 87.1 (76%) | +8.1 marks (1.19 SD) |
| 2023.2 | 229/359 (64%) | 256.0 (71%) | +27.0 marks (1.13 SD) | 75/112 (67%) | 81.9 (73%) | +6.9 marks (1.11 SD) |
| 2024.1 | 239/360 (66%) | 246.7 (69%) | +7.7 marks (0.27 SD) | 81/116 (70%) | 86.5 (75%) | +5.5 marks (0.77 SD) |
| 2024.2 | 232/348 (67%) | 241.1 (69%) | +9.1 marks (0.43 SD) | 78/117 (67%) | 85.5 (73%) | +7.5 marks (0.93 SD) |
| 2025.1 | 241/360 (67%) | 244.4 (68%) | +3.4 marks (0.13 SD) | 79/114 (69%) | 83.9 (74%) | +4.9 marks (0.64 SD) |
| 2025.2 | 235/360 (65%) | 242.4 (67%) | +7.4 marks (0.27 SD) | 78/115 (68%) | 87.1 (76%) | +9.1 marks (1.02 SD) |
| 2026.1 | 214/318 (67%) | 221.3 (70%) | +7.3 marks (0.32 SD) | 74/113 (65%) | 81.7 (72%) | +7.7 marks (1.12 SD) |
For the SAQ, the answer is: barely. Across the 11 sittings the cohort mean sat between 0.01 and 0.65 standard deviations above the pass mark in every sitting except 2023.2. In marks, that is a gap of 0.3 to 16.5 marks out of 360. In 2022.1 the mean score was 232.3 and the pass mark was 232. The average candidate that year was, to within a fraction of a mark, exactly on the line. In eight of the 11 sittings the gap was less than eight marks, which is less than a single 12-mark question.
The SCQ tells a different story. Since 2022.2 the cohort mean has been 0.6 to 1.2 standard deviations above the SCQ pass mark. A candidate of average ability passes the SCQ with room to spare. The same candidate has roughly even odds on the SAQ.
There is a practical reading of this. The SAQ pass mark averaged 65% of the available marks (range 59% to 69%). Cohort averages ran 64% to 71%. A candidate who can reliably score 70% on practice SAQs marked to the standard of the model answers is above the line in every sitting in this series. One who scores 65% is a coin toss.
How spread out the scores are
The standard deviation on both papers has been remarkably stable at about 7% of the marks available (22 to 28 marks on the SAQ, 6 to 9 marks on the SCQ). The lowest and highest scores show the full range.
| Sitting | SAQ lowest | SAQ highest | SAQ SD | SCQ lowest | SCQ highest | SCQ SD |
|---|---|---|---|---|---|---|
| 2021.1 | 138/348 (40%) | 280/348 (80%) | 23.3 marks (6.7%) | 61/118 (52%) | 101/118 (86%) | 8.0 marks (6.8%) |
| 2021.2 | 131/348 (38%) | 281/348 (81%) | 25.4 marks (7.3%) | 56/116 (48%) | 99/116 (85%) | 8.2 marks (7.1%) |
| 2022.1 | 138/360 (38%) | 311/360 (86%) | 27.8 marks (7.7%) | 61/113 (54%) | 95/113 (84%) | 6.9 marks (6.1%) |
| 2022.2 | 159/355 (45%) | 285/355 (80%) | 22.7 marks (6.4%) | 65/116 (56%) | 104/116 (90%) | 6.8 marks (5.9%) |
| 2023.1 | 177/360 (49%) | 320/360 (89%) | 24.3 marks (6.8%) | 61/114 (54%) | 102/114 (89%) | 6.8 marks (6.0%) |
| 2023.2 | 152/359 (42%) | 316/359 (88%) | 23.9 marks (6.7%) | 65/112 (58%) | 98/112 (88%) | 6.2 marks (5.5%) |
| 2024.1 | 119/360 (33%) | 340/360 (94%) | 28.3 marks (7.9%) | 51/116 (44%) | 103/116 (89%) | 7.1 marks (6.1%) |
| 2024.2 | 154/348 (44%) | 286/348 (82%) | 21.1 marks (6.1%) | 61/117 (52%) | 102/117 (87%) | 8.1 marks (6.9%) |
| 2025.1 | 152/360 (42%) | 294/360 (82%) | 27.0 marks (7.5%) | 61/114 (54%) | 100/114 (88%) | 7.7 marks (6.8%) |
| 2025.2 | 138/360 (38%) | 299/360 (83%) | 27.6 marks (7.7%) | 60/115 (52%) | 106/115 (92%) | 8.9 marks (7.7%) |
| 2026.1 | 139/318 (44%) | 268/318 (84%) | 22.6 marks (7.1%) | 60/113 (53%) | 100/113 (88%) | 6.9 marks (6.1%) |
Two observations. The best SAQ score in each sitting was between 80% and 94% of the marks (340 of 360 in 2024.1 was the highest in the series), so nobody is getting near full marks and there is no need to. And the SCQ floor is between 44% and 58%, meaning even the weakest candidates get roughly half the MCQs right; the paper does not separate candidates the way the SAQ does.
The borderline review: who it rescues
The 2021.1 report says simply that both papers must be passed. From 2021.2 the reports state that the overall result is decided “after the borderline review process” without describing it, and from 2022.2 they spell it out. Under that original version, a candidate who passed one paper by at least one standard error of measurement (that is, cleared the raw cut score by two SEM) and failed the other by no more than half an SEM was deemed to have passed.
From 2024 the Council of Education revised the rule. A candidate is now awarded an overall pass when both of the following hold:
- they passed one paper, and their score on the other is not more than one SEM below its pass mark; and
- the amount by which they exceeded the pass mark in the passed paper is equal to or greater than the amount by which they fell short in the other, in percentage terms.
The reports state pass rates for each paper and overall, but not how many candidates passed both papers outright. That means we cannot count borderline passes exactly. We can, however, put a floor under the number: the overall pass rate can only exceed the lower of the two paper pass rates because of the borderline review, so the difference is the minimum number of candidates it rescued.
| Sitting | Rule in force | SCQ pass | SAQ pass | Overall pass | Passed on borderline review (at least) |
|---|---|---|---|---|---|
| 2021.1 | Both papers; no review mentioned | 74.2% | 64.5% | 59.7% | 0 of 186 (0.0%) |
| 2021.2 | Borderline review (not described) | 69.5% | 79.3% | 68.0% | 0 of 256 (0.0%) |
| 2022.1 | Borderline review (not described) | 69.7% | 54.5% | 55.1% | 1 of 198 (0.5%) |
| 2022.2 | Original borderline review | 90.2% | 62.8% | 65.8% | 7 of 234 (3.0%) |
| 2023.1 | Original borderline review | 88.9% | 60.2% | 65.5% | 9 of 171 (5.3%) |
| 2023.2 | Original borderline review | 88.6% | 88.1% | 86.3% | 0 of 219 (0.0%) |
| 2024.1 | Revised borderline review | 86.8% | 65.2% | 71.6% | 13 of 204 (6.4%) |
| 2024.2 | Revised borderline review | 83.3% | 68.9% | 75.9% | 16 of 228 (7.0%) |
| 2025.1 | Revised borderline review | 76.7% | 59.9% | 62.4% | 5 of 202 (2.5%) |
| 2025.2 | Revised borderline review | 84.3% | 63.0% | 68.5% | 14 of 254 (5.5%) |
| 2026.1 | Revised borderline review | 87.4% | 66.8% | 78.9% | 24 of 199 (12.1%) |
Under the revised rule the effect is substantial. In 2026.1 at least 24 of 199 candidates (12%) passed the exam despite failing the SAQ paper. In 2024.2 it was at least 16 of 228 (7%). Because these are lower bounds, the true figures are higher.
The strategic implication is not subtle. A strong SCQ score is insurance against a narrow SAQ miss, and under the revised rule the insurance pays out in proportion to the surplus. A candidate who scrapes the SCQ pass mark has no cover at all. Candidates who treat the SCQ as the easy paper and stop preparing for it once they are passing practice papers are giving up the one thing that can rescue a bad SAQ morning.
What else the reports tell you about the paper
| Sitting | SAQs | Marks | 18-mark questions | Paediatric questions | Paediatric share of marks | SCQ items removed after review |
|---|---|---|---|---|---|---|
| 2021.1 | 26 | 360 | 8 | 5 | 20% | 2 |
| 2021.2 | 26 | 360 | 8 | 8 | 32% | 4 |
| 2022.1 | 27 | 360 | 6 | 6 | 22% | 7 |
| 2022.2 | 28 | 360 | 4 | 5 | 18% | 4 |
| 2023.1 | 28 | 360 | 4 | 7 | 27% | 6 |
| 2023.2 | 28 | 359 | 4 | 7 | 25% | 8 |
| 2024.1 | 28 | 360 | 4 | 9 | 32% | 4 |
| 2024.2 | 28 | 360 | 4 | 7 | 25% | 3 |
| 2025.1 | 27 | 360 | 6 | 6 | 27% | 6 |
| 2025.2 | 28 | 360 | 4 | 9 | 33% | 5 |
| 2026.1 | 26 | 330 | 3 | 5 | 21% | 7 |
A few things worth knowing that are easy to miss:
- The paper structure changed in 2026.1. Every sitting from 2021 to 2025 had 26 to 28 SAQs worth 360 marks. The 2026.1 paper had 26 questions worth 330 marks, with only three 18-mark questions. Watch the next report to see whether that holds.
- Questions get removed after the exam. Two to eight SCQ items were excluded after psychometric review in every sitting (median five), and a whole SAQ was removed in four of the 11 sittings. If a question in the exam seems ambiguous or unanswerable, it may well not count. Do not let it cost you time or composure on the questions that do.
- Paediatrics is a quarter of the SAQ paper. Between 18% and 33% of the marks in every sitting, averaging 26%. Our topic analysis breaks down the rest.
- Cohorts are larger in November. The November sittings averaged 238 candidates against 193 in May. Overall pass rates were on average higher in November too (73% against 66%), but that gap is driven almost entirely by the 2023.2 outlier and should not be read as a seasonal effect.
- The timeline is fixed. Every sitting in the series was held on a Friday in the second half of May or November. The SAQ marking day is about 11 or 12 days later, standard setting about a week after that, and the report follows in the subsequent weeks.
What to do with this
- Prepare for the SAQ as the deciding paper. The reports make it clear the average candidate is on the line. Practise under timed conditions and mark against model answers with mandatory items, not against your own impression of what you would have said.
- Aim for 70% on the SAQ, not 60%. Sixty percent is below the pass mark in ten of the 11 sittings. Seventy is above it in all of them.
- Do not neglect the SCQ once you are passing it. Every mark above the SCQ pass line is borderline-review currency.
- Learn how marks are lost. Examiners record the reasons, and the same six appear in every report. Read how candidates lose marks in the SAQ before your next practice paper.
Sources and method
Figures come from the ACEM Fellowship Written Examination Reports for 2021.1, 2021.2, 2022.1, 2022.2, 2023.1, 2023.2, 2024.1, 2024.2, 2025.1, 2025.2 and 2026.1, published by the Australasian College for Emergency Medicine. Where a report's introduction and results section disagree (the 2024.2 report gives 84.2%, 65.4% and 73.2% in its introduction but 83.3%, 68.9% and 75.9% in its results section), we use the results section. Weighted averages are weighted by cohort size. "Mean minus pass mark in SD" divides the difference between the published cohort mean and the published pass mark by the published standard deviation. Borderline-review counts are lower bounds calculated as the overall pass count minus the lower of the two paper pass counts. ACEMCQ is independent of ACEM and this page is not endorsed by the College.
Frequently asked questions
What is the pass rate for the ACEM fellowship written exam?
Across the 11 sittings from 2021.1 to 2026.1, 69% of candidates passed overall (1,625 of 2,351). Individual sittings ranged from 55% (2022.1) to 86% (2023.2). The SCQ paper pass rate averaged 82% and the SAQ paper 67%.
Is the SAQ or the SCQ paper harder to pass?
The SAQ. Since 2022.2 the SCQ pass rate has been between 77% and 90%, while the SAQ pass rate has stayed between 60% and 69% in every sitting except 2023.2. The overall result tracks the SAQ result closely.
What score do you need to pass the ACEM written exam?
The pass mark is set each sitting by the Modified Angoff method plus one standard error of measurement. Between 2021 and 2026 the SAQ pass mark ranged from 59% to 69% of available marks (average 65%) and the SCQ pass mark from 66% to 70% (average 68%).
What is the ACEM borderline review?
Since 2022.2 a candidate who passes one paper and narrowly fails the other can be awarded an overall pass. Under the revised rule (applied from 2024), the failed paper must be no more than one standard error of measurement below its pass mark, and the surplus in the passed paper must be at least as large as the shortfall, in percentage terms. In the sittings since the revision, at least 2% to 12% of each cohort passed this way.