CCS Cases on USMLE Step 3 (2026): How Many You Actually Get
The short answer
Step 3 Day 2 still includes 13 to 14 computer-based case simulations, each allotted a maximum of 10 or 20 minutes of real time. That did notchange in March 2026. USMLE does not publish an exact CCS weighting — only an upper bound: CCS contributes no more than its share of the exam’s allotted time.
First, a correction: it is not 9 cases
You may have read that “on 10 March 2026 the USMLE reduced CCS cases from 13 to 9.” That is false, and it is now repeated across a surprising number of prep sites — ours included, until we checked the primary sources and corrected it. Two USMLE documents settle it.
The live Step 3 exam content page states verbatim: “13 to 14 case simulations, each of which is allotted a maximum of 10 or 20 minutes of real time.” That page is unambiguously current — it already describes Day 1 in the post-March-2026 format (232 multiple-choice items divided into 12 blocks of 18–20 items, 30 minutes per block).
And USMLE’s own announcement of the change includes a before/after table that lists case simulations explicitly:
| Component | Before 10 March 2026 | On / after 10 March 2026 |
|---|---|---|
| Day 1 MCQ blocks | 6 blocks, 38–39 items | 12 blocks, 18–20 items |
| Day 2 MCQ blocks | 6 blocks, 30 items | 9 blocks, 20 items |
| Day 2 case simulations | 13–14 cases | No change |
| Max minutes per case | 10 or 20 minutes | No change |
Where the myth came from is visible in that same table: on 10 March 2026, Day 2’s multiple-choice block count changed from 6 to 9. Somebody attached that 9 to the case simulations, and it propagated. If you see “9 CCS cases” anywhere, that is the error you are looking at.
Where CCS sits in the exam
Step 3 runs over two days. CCS appears only on Day 2, Advanced Clinical Medicine:
| Day 2 component | Detail |
|---|---|
| Multiple choice | 180 items in 9 blocks of 20, 30 minutes each |
| CCS tutorial | 6 minutes, optional — no practice cases at the centre |
| Case simulations | 13–14 cases, max 10 or 20 minutes real time each |
| Break | Minimum 45 minutes |
| Total | Approximately 9 hours |
How CCS is actually scored
The weighting is an upper bound, not a percentage
From the Bulletin of Information, Scoring and Score Reporting:
“For Step 3, your performance on the case simulations will affect your Step 3 score and could affect whether you pass or fail. The proportional contribution of the score on the case simulations is no greater than the proportional contribution of time allotted for CCS.”
That is the only weighting statement USMLE publishes. The widely quoted figure that CCS is “about 25% of your Step 3 score” appears in no official document — it is a reasonable inference from the time allotment, but treat it as an estimate rather than a fact. CCS is not separately scaled or separately reported; Step 3 comes back as one 3-digit score.
What earns and loses credit
USMLE lists the scored domains explicitly: diagnosis (including physical examination and appropriate diagnostic tests), therapy, monitoring, timing, sequencing, and location. Note that the last three are scored just as the first three are.
| Official wording | What it means in practice |
|---|---|
| “Indicated patient management actions are awarded credit” | Credit is positive and specific — do the right things. |
| “Actions that are not indicated and pose greater potential risk… decrease your score” | Unnecessary orders are penalised, keyed to risk rather than quantity. An extra CBC is not equivalent to an unindicated invasive procedure. |
| “Seemingly correct management decisions made in an incorrect sequence or after a delay… may receive no credit” | Right action, wrong order or too late, equals zero. Sequencing is scored. |
| “You will be scored lower if you take an aggressive approach when restraint and observation are the standard of care” | Over-treating is a scored error, not a safe default. |
The cost myth, corrected. USMLE states that cost is accounted for indirectly, based on the relative inappropriateness of patient-management actions. There is no direct cost penalty. You are not marked down for an expensive test that is indicated — you are marked down for ordering an inappropriate one, and expense correlates with that only loosely.
What the research says about CCS
CCS measures something the multiple-choice sections do not
Feinberg RA et al., Journal of General Internal Medicine 2012;27(1):65–70 (PMID 21879372), analysing 40,588 first-time Step 3 takers, found:
“Predictors of Step 1 and Step 2 CK explained 55% of overall Step 3 variability and only 9% of CCS score variability.”
Read that again. Your Step 1 and Step 2 CK performance predicts more than half of your overall Step 3 score — and almost none of your CCS performance. CCS tests a genuinely different skill, which is exactly why students who coast on question-bank ability get surprised by it. It is also why our Step 3 predictor treats CCS as a separate adjustment on top of the multiple-choice estimate rather than folding it into the same signal.
One in five candidates orders something dangerous
Harik P et al., Academic Medicine 2009;84(10 Suppl):S79–82 (PMID 19907393), analysing 25,283 first-time examinees, found that over 20% ordered at least one action with potential for significant patient harm, and that the propensity varied by case.
That is the practical argument for practising CCS rather than winging it. The failure mode is usually not missing the diagnosis — it is reaching for an intervention that is aggressive, mistimed, or simply not indicated.
What this means for your prep
- Do not skip CCS practice because your QBank average is high. The research says your MCQ ability predicts almost none of your CCS performance.
- Practise sequencing and timing, not just diagnosis. Both are explicitly scored, and both are where most credit is lost.
- Resist the urge to over-order. Restraint is the standard of care in some cases, and aggression is scored down.
- Use the 6-minute tutorial time wisely — there are no practice cases at the test centre, so interface fluency has to come from prep.
Factor CCS into your Step 3 estimate
Our predictor blends your Step 2 CK, UWorld %, UWSA, and NBME 6/7 results, then applies a CCS adjustment — because no multiple-choice form grades case simulations.
Predict my Step 3 score →References
- USMLE — Step 3 Exam Content (case count, block structure).
- USMLE — Test Delivery Software Updates Coming in 2026 (the before/after change table).
- USMLE — Bulletin of Information: Scoring and Score Reporting (CCS weighting statement).
- Feinberg RA et al. J Gen Intern Med 2012;27(1):65–70. PMID 21879372.
- Harik P et al. Acad Med 2009;84(10 Suppl):S79–82. PMID 19907393.
Educational content. USMLEPredictor is not affiliated with the NBME, USMLE program, or FSMB. Always confirm current exam format and policy on usmle.org.
Frequently asked questions
How many CCS cases are on USMLE Step 3 in 2026?
Step 3 Day 2 includes 13 to 14 computer-based case simulations, each allotted a maximum of 10 or 20 minutes of real time. This did not change on March 10, 2026 — USMLE’s own change table lists case simulations as “no change.” The claim that cases were reduced to 9 confuses the new Day 2 multiple-choice block count (which did go from 6 to 9) with the case count.
How much is CCS worth on Step 3?
USMLE does not publish an exact percentage. The Bulletin states only that the proportional contribution of the CCS score is no greater than the proportional contribution of time allotted for CCS. The widely quoted “about 25%” figure is a reasonable inference from the time allotment, but it is not an official number. Step 3 is reported as a single 3-digit score; CCS is not separately scaled or reported.
What is scored on a CCS case?
USMLE lists the scored domains explicitly: diagnosis (including physical examination and appropriate diagnostic tests), therapy, monitoring, timing, sequencing, and location. Timing, sequencing, and location are scored just as the clinical decisions are — a correct action taken in the wrong order or too late may receive no credit.
Are you penalised for ordering expensive tests on CCS?
Not directly. USMLE states that cost is accounted for indirectly based on the relative inappropriateness of patient management actions. You are not marked down for an expensive test that is indicated; you are marked down for ordering one that is not indicated, and expense correlates with that only loosely.