M2 Staatsexamen: how Germany tests clinical reasoning
Table of contents
Three mornings in a row. Each one with five hours of a written exam. On the desk, a Heft with clinical vignettes, sometimes with an image or a lab table, and an Antwortbeleg for optical reading. The question does not ask for the isolated fact: it asks for the Bestantwort, the most suitable option among five, at the point where the case changes course. This is the written gateway to the second stage of the Ärztliche Prüfung in Germany: the Zweiter Abschnitt, the M2, the Staatsexamen that separates the clinical cycle from the Praktisches Jahr.
NEXO is a daily medical diagnosis game: one clinical case a day, six progressive clues, and ICD-10 proximity feedback across five levels.
The format looks close to a USMLE Step 2 CK or an Examen MIR, but the typical M2 item is the Key-Feature: a short sequence of critical decisions about the same case. This text explains what the M2 is, how the exam works, how the thresholds are set in the ÄApprO, what studies say about this format and how to train. At the end there is a table, frequently asked questions and a checklist.
Key points
- The Zweiter Abschnitt der Ärztlichen Prüfung (M2) is a written exam of 320 multiple-choice questions with five options and Bestantwort, over three consecutive days of five hours each (IMPP Medizin; ÄApprO § 28).
- The tasks come as Key-Features (three to five questions per case about critical decisions) and as Einzelaufgaben; the IMPP describes the Key-Feature as an exam of clinical decision-making, not only of memory of facts (Praktische Hinweise IMPP).
- The exam is passed if the candidate gets at least sixty percent of the questions right or does not fall below seventy-eight percent of the reference average (ÄApprO § 14 Abs. 6); with 320 valid tasks, the absolute threshold is 192 (IMPP Bestehensgrenzen).
- The Landesprüfungsämter admit candidates and communicate the result; the IMPP prepares the tasks, the Gegenstandskatalog (GK2) and the Blueprint (IMPP).
What the M2 Staatsexamen is
Medical education in Germany is federal: the Approbationsordnung für Ärzte (ÄApprO 2002) defines objectives, phases and the Ärztliche Prüfung in three Abschnitte. The Erste Abschnitt (M1) mixes written and oral-practical after about two years. The Zweite Abschnitt (M2) is written only, after three more years of the clinical cycle and before the Praktisches Jahr. The Dritte Abschnitt (M3) is oral-practical, with a patient, after the PJ (ÄApprO § 1 and § 13).
In student language, M2 and Staatsexamen often overlap. What this text deals with is the written part of the second stage: the exam that the ÄApprO calls the schriftliche Prüfung of the Zweiten Abschnitts and that the IMPP organizes on a national scale (IMPP Medizin).
Where the M2 fits in Miller's pyramid
George Miller described four steps: know, know how, show how and do (Miller, 1990). The M2 is at the second: application of knowledge in vignettes. The M3 rises to the third, with an exam and a patient report. Germany, like France with the EDN and ECOS, separates the written exam and clinical performance at different points along the path, instead of mixing them in a single session like the AMC Clinical Exam or the second stage of Revalida.
Who does what: IMPP and Landesprüfungsämter
The ÄApprO provides that the authorities of the Länder use an Einrichtung to prepare tasks and Gegenstandskataloge (ÄApprO § 14 Abs. 3). In practice, this Einrichtung is the Institut für medizinische und pharmazeutische Prüfungsfragen (IMPP), in Mainz. The IMPP publishes the GK2, the M2 Blueprint and the Praktische Hinweise brochure. The Landesprüfungsämter receive the registration (typical deadlines up to January 10 or June 10), admit the candidate, administer the exam and deliver the Ergebnismitteilung (ÄApprO § 10; IMPP).
How the written exam works
According to the IMPP and ÄApprO § 28, the Zweite Abschnitt brings together 320 tasks in the Antwort-Wahl-Verfahren. In each one, the candidate chooses the Bestantwort among five options. The exam runs over three consecutive days; each day lasts five hours (IMPP Medizin; ÄApprO § 28). The official dates are April and October (ÄApprO § 16).
Content and orientation of the case
§ 28 Abs. 1 calls for the knowledge and skills the physician needs to act autonomously. The exam is fallbezogen: requirements of practice, main clinical pictures, fächerübergreifend and problemorientiert questions (ÄApprO § 28). The official Stoffkatalog is in Anlage 15 of the ÄApprO; the IMPP translates this into the Gegenstandskatalog GK2 and a Blueprint that shows relative weights per domain, without expanding or reducing the Stoff of the Approbationsordnung (IMPP Gegenstandskataloge).
Materials on exam day
The Praktische Hinweise brochure describes the package for each morning: Aufgabenheft (with Bildbeilage when there are images), an optical-read Antwortbeleg and writing materials. There are two Auflagen (A and B) of the Heft. Calculators, phones and communication devices are prohibited. Tables of laboratory parameters with reference ranges come on the last pages of the Bildbeilage; values without a range in the table assume knowledge (Praktische Hinweise IMPP).
Key-Features and Einzelaufgaben
The IMPP divides the 320 tasks into two presentation formats: Fallstudien im Key-Feature-Format and Einzelaufgaben. Both mostly use the Aufgabentyp A (Einfachauswahl / Bestantwort) (Praktische Hinweise IMPP).
What a Key-Feature is in the M2
In the Praktische Hinweise, Key-Features are critical decisions needed to solve the clinical problem. The text describes a situation; then come three to five questions that test Schlüsselkompetenzen. A Key-Feature can also be a difficult node of identification or management (Weichenstellung). The stated goal is to measure clinical Entscheidungfindung and application of knowledge to the case, not just recall of facts (Praktische Hinweise IMPP).
This logic came from the Canadian tradition of key-feature problems, used in the MCCQE and studied by Page, Bordage and colleagues: the item focuses on the point where clinical error usually occurs, not the full script of the case (Page and Bordage, 1995; MCCQE).
Einzelaufgaben and image material
The Einzelaufgaben are standalone items, still fallbezogen. Some of them use Bildmaterial: histological sections, radiographs, clinical photos. The candidate consults the Bildbeilage and chooses the Bestantwort. There are also tasks that require reading laboratory results with the official reference tables (Praktische Hinweise IMPP).
Why this changes training
Training only with flashcards of isolated facts leaves the candidate vulnerable on the Key-Feature: the second and third questions of the same case depend on the first decision. An early error propagates. Training with short sequences (history → hypothesis → next test → management) mirrors better what the IMPP describes.
How scoring and thresholds work
ÄApprO § 14 Abs. 6 sets the Bestehen of the written exam: the candidate must answer at least sixty percent of the questions correctly, or the number of correct answers cannot be more than twenty-two percent below the reference average. The M2 reference is Erstteilnehmer with the Mindeststudienzeit of five years (ÄApprO § 14 Abs. 6; IMPP Bestehensgrenzen).
In IMPP practice, both numbers are calculated (absolute threshold and threshold relative to the average). The lower of them becomes the Bestehensgrenze for that Examen. With the full set of 320 tasks, the absolute threshold is 192. If the reference average is high, the relative threshold can fall below 192; the candidate passes at the lower of the two (IMPP Bestehensgrenzen).
Task elimination and Nachteilsausgleich
Before the result, the tasks go through review. Items considered fehlerhaft are removed from the count (§ 14 Abs. 4). The reduction cannot harm the candidate: those who answered the eliminated task correctly can have it reintroduced in the Vergleichsberechnung; those who answered it wrong are assessed on the reduced basis. This can produce slightly different Bestehensgrenzen between groups (IMPP Bestehensgrenzen).
Grades beyond passing
Anyone who reaches the threshold receives a grade from ausreichend to sehr gut according to the fraction of questions above the threshold that they still answered correctly: at least seventy-five percent of these extra questions for sehr gut; at least fifty and less than seventy-five for gut; at least twenty-five and less than fifty for befriedigend; less than twenty-five for ausreichend (ÄApprO § 14 Abs. 7).
What the candidate receives
The Ergebnismitteilung contains the absolute and relative score, the grade, the overall average, the reference average and the Bestehensgrenze used. The Landesprüfungsämter deliver this package; the IMPP publishes aggregated results on its homepage after the Auswertungsphase (IMPP Bestehensgrenzen).
Retake
The Zweite Abschnitt may be retaken at most twice. A third attempt, even after further study, is not permitted. A passed section is not retaken (ÄApprO § 20).
| Item | Official rule | Source |
|---|---|---|
| Format | 320 MC, five options, Bestantwort | ÄApprO § 28; IMPP |
| Duration | Three consecutive days, five hours per day | ÄApprO § 28; IMPP |
| Item types | Key-Features (3–5 questions/case) and Einzelaufgaben | Praktische Hinweise IMPP |
| Passing | Sixty percent of the questions or ≥ seventy-eight percent of the reference average | ÄApprO § 14 Abs. 6 |
| Absolute threshold (320 valid) | 192 correct answers | IMPP Bestehensgrenzen |
| Admission and result | Landesprüfungsamt; content and scoring IMPP | ÄApprO § 9–10, § 14 |
| Next step | Praktisches Jahr, then the oral-practical M3 | ÄApprO § 3, § 30 |
How clinical reasoning appears
The M2 does not ask the candidate to demonstrate the physical examination in front of an examiner. It asks them to choose, under time pressure, the decision a physician at the start of autonomy would make. Three patterns recur.
A decision at the node, not a full script
In the Key-Feature, the useful question is: what changes the management now? Ordering one more useless test, starting therapy without criteria or delaying an emergency costs the item. Page and Bordage described this focus as a way to increase the validity of the written clinical decision test (Page e Bordage, 1995).
Call this moment the Weichenstellung: the first choice switches the case onto a different track. If you mark the wrong test in question 1, the following ones may seem coherent and still all be off the path. That is why useful practice is not just accumulating Einzelaufgaben: it is rehearsing the chain and, when you get the first node wrong, stopping and rewriting the priority before looking at the answer key.
The IMPP describes the Key-Feature format and the use of Bildmaterial in the practical documentation of the exam (IMPP). In the room, the discipline of reading the vignette to the end (age, allergies, instability) before the options reduces the error of a «complete» management plan for the wrong patient.
Integration across specialties
The ÄApprO requires Fragestellungen fächerübergreifend. A chest pain vignette may mix cardiology, pulmonology and emergency medicine. The Blueprint and the GK2 guide the coverage; they do not replace a cross-cutting reading of the case (IMPP).
Studying in silos (only Innere, only Chirurgie) leaves the candidate slow when the question crosses fields. One way to train the fächerübergreifend requirement is, after each block, to list which specialties entered the same case and which decision was the link between them. The GK2 remains the official map; the Blueprint shows the weights and emphases of the concrete application (IMPP).
Imaging and laboratory data as evidence
Some items only close with the Bildbeilage or with the reference table. Reading the finding and translating it into management is the same move that PLAB/UKMLA and Step 2 CK demand on a different scale: evidence, hypothesis, action.
What the studies say
The medical assessment literature treats key-feature problems as a written decision-making instrument with better sampling of critical decisions than long open cases (Page e Bordage, 1995; Bordage e Page, 2018). Miller places this type of exam at the knows how level (Miller, 1990).
Clinical reasoning studies remind us that errors come both from incomplete knowledge and from failure to select the right data under pressure (Eva, 2005; Norman e Eva, 2010). For the M2 candidate, the practical implication is obvious: reviewing only drug lists without practicing decision sequences leaves the Key-Feature opaque.
Compared with the USMLE Step 2 CK, the M2 sets a ceiling of 320 items and a known absolute threshold (192 when all are scored). Compared with the MCCQE, it shares the key-feature heritage, but applies five-option Bestantwort on a German federal scale. On the general Medicine page, the IMPP does not publish an open bank of corrected past papers the way the CNG does with the EDN. Practice depends on the Gegenstandskatalog, the Blueprint, the Beispielaufgaben in the brochure and material from universities and publishers aligned with the GK2 (IMPP Medizin).
How to train reasoning for the M2
Official base before the commercial bank
Start with the GK2 and the M2 Blueprint. They are not the new Stoff: they make explicit what the ÄApprO already set out. Use the Praktische Hinweise brochure to understand Key-Feature, Bildmaterial and room rules. Only then choose banks and books that state alignment with the GK2.
Practice in five-hour blocks
The real exam has three five-hour blocks. Simulate at least one full block per week in the final two months: same timing, no phone, breaks only at the pace you will use in the Examen. The goal is to calibrate pace (about one minute per item, with room for long Key-Features) and stamina.
Sequences, not isolated cards
For each practice case, force three to five decisions: main problem, short differential diagnosis, next test, interpretation, management. This mirrors the Key-Feature. If the bank only offers Einzelaufgaben, group three items on the same topic and answer the chain out loud.
Early error
When you miss the first question of a Key-Feature, note why the Weichenstellung was wrong (ignored data, wrong priority, contraindication). The pattern repeats more than the name of the forgotten drug.
Three and a half to five hours per block call for energy management, not just content management. The ÄApprO and the IMPP practical Hinweise describe the structure in days and Heft; the candidate needs to rehearse the real pace (ÄApprO; IMPP). Anyone who only does sets of twenty items never finds out where attention drops at minute 180.
Steps on exam day (room discipline)
The M2 room has its own rules for materials and marking. Rehearsing the content without rehearsing the ritual creates avoidable errors on the Antwortbeleg.
- Confirm which Auflage (A/B) of the Heft matches the Antwortbeleg before the first mark.
- Read the full vignette, including lab notes and image captions, before the options.
- In Key-Features, decide the node (Weichenstellung) quietly and only then mark; if you hesitate, flag it and move on so you do not lose the block.
- Reserve the final minutes to review only the items marked as doubtful, not to start the booklet over.
- Do not carry the discussion of the previous case into the break between blocks: the next Heft restarts the mental clock.
- On the way out, note the types of error (priority, contraindication, image reading), not just the specialty topic.
These steps do not replace the GK2; they organize what the IMPP documentation and the ÄApprO already assume: long timing, standardized material and chained decisions (IMPP).
Quick checklist
- Read the GK2 and the Blueprint; mark weak domains.
- Do one five-hour block per week at the end of the cycle.
- Train Key-Features in sequences of three to five decisions.
- Include image and lab reading with reference ranges.
- Review errors by type of decision, not just by specialty.
- Confirm deadlines and documents with the Landesprüfungsamt (January 10 / June 10).
- On the day: check the Auflage A/B of the Heft and the Antwortbeleg before marking.
After the M2: Praktisches Jahr and M3
Those who pass the M2 enter the Praktisches Jahr (48 weeks in three blocks: Innere Medizin, Chirurgie and a third field, including Allgemeinmedizin) and only then face the Dritte Abschnitt, an oral-practical exam with Patientenvorstellung (ÄApprO § 3 and § 30). The M2 is therefore not the Approbation: it is the written gate that opens the practical year. Training reasoning for the M2 also prepares the case narrative that the M3 will ask for out loud, even if the format changes.
Admission documents (overview)
ÄApprO § 10 lists what accompanies the application to the Landesprüfungsamt for the Zweite Abschnitt: identity and civil status documents, proof of study, Leistungsnachweise for the Fächer and Querschnittsbereiche of § 27, Famulatur and the Zeugnis for the Erste Abschnitt. The application deadlines are January 10 and June 10. Form details vary by Land; the candidate confirms with the responsible Amt, not only in forums (ÄApprO § 10).
How NEXO helps you train
NEXO does not replace the GK2 or registration with the Landesprüfungsamt. It offers clinical cases in which the task is to decide the next step from a vignette, the same move the Key-Feature asks for. You can train in Portuguese or in the language of the exam, compare hypotheses and see where the decision diverged. Anyone already studying for MIR, Step 2 CK or EDN reuses the habit of the diagnostic chain. Also see the Economy page if you are putting together a study plan alongside the German cycle.
Frequently asked questions
Is the M2 oral or written?
Written only. The oral-practical part of the German pathway is the M3, after the Praktisches Jahr (ÄApprO § 13 and § 30).
How many questions and over how many days?
Three hundred and twenty tasks, three consecutive days, five hours per day (IMPP; ÄApprO § 28).
What is Bestantwort?
Among five options, the candidate marks the answer the official answer key considers correct (most appropriate). It is not all the true statements, nor free ranking (Praktische Hinweise IMPP).
How do you pass?
By getting at least sixty percent of the valid questions right, or by finishing no more than twenty-two percent below the reference average (equivalent to seventy-eight percent of that average). The lower of the two thresholds applies (ÄApprO § 14 Abs. 6).
What is the reference?
In the M2, Erstteilnehmer with a Mindeststudienzeit of five years (ÄApprO § 14 Abs. 6).
How many times can I retake it?
At most two retakes of the Zweite Abschnitt. After that, there is no new chance even with further study (ÄApprO § 20).
Does the GK2 increase the exam content?
No. The IMPP states that the Gegenstandskatalog makes explicit the Stoff already set out in the Approbationsordnungen; it neither expands nor reduces it (IMPP Medizin).
How does it differ from MIR, EDN or Step 2 CK?
All are written exams of clinical application. The M2 emphasizes Key-Features in series and an absolute threshold or one relative to the national reference. The MIR ranks candidates for residency in Spain; the EDN mix digital formats and require ECOS afterwards; the Step 2 CK uses the NBME scale. The decision move is similar; the passing rule and the role in the curriculum change.
Play the case of the day on NEXO and practice the Weichenstellung before the next five-hour block.
References
- Institut für medizinische und pharmazeutische Prüfungsfragen. Prüfungen Medizin. Available at: [https://www.impp.de/pruefungen/medizin](https://www.impp.de/pruefungen/medizin)
- IMPP. Bestehens- und Notengrenzen. Available at: [https://www.impp.de/pruefungen/allgemein/bestehens-und-notengrenzen](https://www.impp.de/pruefungen/allgemein/bestehens-und-notengrenzen)
- Bundesrepublik Deutschland. Approbationsordnung für Ärzte (ÄApprO 2002), in particular §§ 13, 14, 16, 20, 28. Available at: [https://www.gesetze-im-internet.de/_appro_2002/BJNR240500002.html](https://www.gesetze-im-internet.de/_appro_2002/BJNR240500002.html)
- IMPP. Praktische Hinweise zur Durchführung der schriftlichen Prüfungen nach der Approbationsordnung für Ärzte (current brochure). Accessed via [IMPP Medizin](https://www.impp.de/pruefungen/medizin)
- IMPP. Blueprint M2-Examen and Gegenstandskatalog GK2. Available via [IMPP Gegenstandskataloge](https://www.impp.de/pruefungen/medizin)
- Miller GE. The assessment of clinical skills/competence/performance. Acad Med. 1990. PMID: [2400509](https://pubmed.ncbi.nlm.nih.gov/2400509/)
- Page G, Bordage G. The Medical Council of Canada's key features project: a more valid written examination of clinical decision-making skills. Acad Med. 1995. PMID: [7748379](https://pubmed.ncbi.nlm.nih.gov/7748379/)
- Bordage G, Page G. An alternative approach to scoring and interpreting clinical decision-making skills: the key features approach. Adv Health Sci Educ Theory Pract. 2018. PMID: [29777464](https://pubmed.ncbi.nlm.nih.gov/29777464/)
- Eva KW. What every teacher needs to know about clinical reasoning. Med Educ. 2005. PMID: [15813700](https://pubmed.ncbi.nlm.nih.gov/15813700/)
- Norman GR, Eva KW. Diagnostic error and clinical reasoning. Med Educ. 2010. PMID: [20051831](https://pubmed.ncbi.nlm.nih.gov/20051831/)
<!-- revisao_clinica_thiago:
- Confirm that the cited Praktische Hinweise brochure is still the current version at the time of publication; replace the direct PDF link if the IMPP rotates the file.
- Confirm that Blueprint_M2-Examen.pdf at the path /files/PDF/Blueprint/ remains stable; the post links to the IMPP Medizin page, not to the loose PDF.
- Threshold: the text uses sixty percent / seventy-eight percent / 192 according to IMPP Bestehensgrenzen and ÄApprO §14 Abs.6; check whether there is a recent legislative change not reflected in gesetze-im-internet.
- Key-Feature: definition aligned with Praktische Hinweise (3–5 questions; Entscheidungfindung); do not invent a fixed number of Key-Features per day (IMPP does not publish a fixed quota on the general page).
- Do not mix M1 rules (2 days × 4 h) with M2 (3 × 5 h).
- Cover Needle_holder2.JPG CC0 with no visible brand text in the 1200×630 crop.
-->