---
title: "Building an OSCE blueprint: a worked example for a 12-station final-year exam"
description: "A worked OSCE blueprint for a 12-station final-year exam: the competency by system grid, the weights, the coverage arithmetic, and the sign-off record."
canonical: https://studydrome.com/docs/osce/design/blueprint-worked-example/
updated: 2026-09-27
---

# Building an OSCE blueprint: a worked example for a 12-station final-year exam

A worked OSCE blueprint for a 12-station final-year exam: the competency by system grid, the weights, the coverage arithmetic, and the sign-off record.

An OSCE blueprint is a grid of competencies against clinical systems or problems. Each cell that the exam samples gets a station. For a 12-station final-year exam, the working group weights the competencies in the curriculum, converts the weights into station counts, places each station in a cell, and records which competencies the exam does not sample. The committee signs the grid before a station is written.

The general method of blueprinting is covered in [how to create an exam blueprint](/answers/exam-blueprinting/how-to-create-an-exam-blueprint/). This page works through one OSCE example to the signed document.

## Why the blueprint comes first

The blueprint is the exam's claim about what it measures. The AMEE guide on OSCE organization describes it as a two-dimensional matrix of competencies against problems or systems, and it calls content specificity "a major contributor to poor reliability". A candidate who is strong on one case is not reliably strong on the next. The only defense is sampling across the domain on purpose, and the blueprint is where that purpose is written down.

A blueprint built after the stations exist is a sorting exercise, not a design. It records what the writers happened to have. The example below starts from the curriculum and ends with commissions to writers.

## Step 1: fix the shape of the exam

The working group settles four numbers before it draws the grid.

| Decision | This exam | Reason |
| --- | --- | --- |
| Stations | 12 | The 2020 Ottawa consensus states that fewer than 12 stations will not yield reliable results; 12 rooms are available |
| Station length | 13 minutes | 12 stations at 10 minutes is 120 minutes of testing time, under the consensus floor of 150 minutes; 12 at 13 minutes is 156 |
| Candidate level | Final year, before graduation | Sets the standard every station is written to |
| Sittings | Two circuits, same day, same stations | Sets the security rule: the morning cohort is held until the afternoon cohort has arrived |

The station length is the decision most groups get wrong. They pick the number of rooms, pick a familiar station length, and only then discover the testing time. The AMEE guide's general recommendation is 14 to 18 stations, each lasting 5 to 10 minutes. A 2025 meta-analysis found the highest reliability in exams with 5 to 10 stations of under 10 minutes. Both point the same way: when rooms allow, add shorter stations rather than lengthening the ones you have. This group had 12 rooms, so it lengthened. The [reliability page](/docs/osce/foundations/validity-and-reliability/) sets out the evidence.

## Step 2: weight the competencies

The rows of the grid are the competencies the program says a graduate must have. The AMEE guide's example covers history-taking, physical examination, communication, procedures, and management. The working group weights them based on the curriculum's stated emphasis and on what the graduate will do in the first year of practice.

| Competency | Weight | Stations (of 12) |
| --- | --- | --- |
| History taking | 25% | 3 |
| Physical examination | 25% | 3 |
| Communication and explanation | 17% | 2 |
| Procedures and practical skills | 17% | 2 |
| Clinical reasoning and management | 17% | 2 |
| **Total** | **100%** | **12** |

The arithmetic is weight times 12, rounded to whole stations. Rounding is a decision, not an accident: the group records that reasoning and management got two stations rather than three because the written exams in the same year already sample them.

## Step 3: place the stations in the grid

The columns are the systems or presentations. The group lists 10 and places the 12 stations so that no system carries more than 2, and the systems most seen in the first year of practice carry at least 1.

| Competency | Cardio | Resp | GI | Neuro | MSK | Endocrine | Mental health | Pediatrics | Women's health | Acute care |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| History | 1 |  | 5 |  |  |  | 9 |  |  |  |
| Examination | 2 | 6 |  | 10 |  |  |  |  |  |  |
| Communication |  |  |  |  |  | 3 |  | 11 |  |  |
| Procedures |  |  |  |  |  |  |  |  | 7 | 4 |
| Reasoning and management |  |  |  |  |  |  |  | 12 |  | 8 |

Numbers are station positions in the circuit. Two things are visible at once. Each competency row holds its allocated count. No column holds more than two stations. The grid is the working group's evidence for both.

A third thing is visible only if you look for it. The grid has 50 cells, and the exam fills 12. That is 24 percent coverage of a single grid, before anyone argues about which cells matter most. Twelve stations cannot sample a curriculum. They can sample a defensible slice of it, and the blueprint's job is to make the slice deliberate.

## Step 4: record what the exam does not sample

The committee will be asked, usually at an appeal, why the exam had no station on a given topic. The blueprint answers in advance. The example carries a short table of the omissions and where else the program assesses them.

| Not sampled in this sitting | Assessed instead by |
| --- | --- |
| MSK examination | Year 4 OSCE, sitting 2 |
| Endocrine history | Written finals, paper 2 |
| Neurology management | Workplace-based assessment, medicine block |
| Respiratory procedures | Skills sign-off, simulation lab |

Over two years, the group rotates the columns so that each competency meets each major system at least once. That rotation plan is part of the blueprint and part of the [station bank](/docs/osce/design/station-bank-management/): a station written for the endocrine history cell this year is in the bank for the sitting that needs it.

## Step 5: sign off and commission

The blueprint becomes a document with five fields: version, date, the working group's names, the approving committee and its date, and a list of deviations from the previous version with reasons. The [governance page](/docs/osce/planning/governance-and-roles/) says who signs.

Then the commissions go out. Each writer receives a cell, not a topic: "History taking, gastrointestinal, final year, 13 minutes, station 5". The [station writing page](/docs/osce/design/writing-a-station/) starts from that line.

## What can go wrong?

- **Weights from the writers, not the curriculum.** The grid fills with the topics the writers like. The fix is to fix the weights before anyone volunteers a station.
- **A row that is not a competency.** "Cardiology" is a column. Putting it in a row doubles the axis, making the grid unreadable.
- **Coverage claimed for a station that tests something else.** A "history" station whose marking scheme is mostly examination items sits in the wrong cell. The [review workshop](/docs/osce/design/station-review-and-piloting/) checks the cell against the scheme.
- **No omissions table.** The exam cannot explain itself at appeal.
- **The blueprint changes on the day.** A station withdrawn for equipment failure changes the grid. The sign-off record should say what a withdrawal does to the pass decision, before it happens.

## Limits

A blueprint is a coverage argument, not a reliability guarantee. Twelve well-placed stations can still produce a low coefficient if the stations are badly written or the examiners untrained. And the Ottawa floor is a floor, stated for high-stakes exams. A formative OSCE can run fewer stations if the blueprint says so. The blueprint's honesty is the point: it states what the exam samples and what it does not.

## Frequently asked questions

### What is an OSCE blueprint?

A grid that maps the competencies an exam assesses against the clinical systems or problems it samples them in. Each station occupies a cell. The blueprint shows the committee what the exam covers, what it does not, and whether the sampling matches the curriculum's weights. It is signed before stations are written.

### How many stations does an OSCE blueprint need?

As many as the exam has. The 2020 Ottawa consensus states that fewer than 12 stations, or fewer than 150 minutes of testing time, will not yield reliable results for a high-stakes exam. The AMEE organization guide suggests 14 to 18 stations, each lasting 5 to 10 minutes. The blueprint then decides which cells those stations fill.

### Should the blueprint rows be competencies or topics?

Competencies in the rows, systems or presentations in the columns. History taking, examination, communication, procedures, and management are the usual rows. A topic like cardiology is a column. Mixing the two on one axis makes the grid impossible to read and the coverage impossible to check.

### Can a 12-station OSCE cover the whole curriculum?

No. A grid of five competencies and ten systems has 50 cells, and 12 stations fill 12 of them. The blueprint makes the sample deliberate, records what is not sampled and where else it is assessed, and rotates the cells across sittings so the program covers the grid over time.

### Who signs off an OSCE blueprint?

The working group that builds it proposes; the assessment committee approves. The signed record carries the version, the date, the names, the approving body, and the deviations from the previous version with reasons. It is filed with the exam as evidence of validity and consulted at appeal.

## Sources

**Sources**

1. [Khan KZ, Gaunt K, Ramachandran S, Pushkar P. The Objective Structured Clinical Examination (OSCE): AMEE Guide No. 81. Part II: organisation and administration](https://doi.org/10.3109/0142159X.2013.818635) (Medical Teacher, 2013)
2. [Boursicot K, Kemp S, Wilkinson T, et al. Performance assessment: consensus statement and recommendations from the 2020 Ottawa Conference](https://doi.org/10.1080/0142159X.2020.1830052) (Medical Teacher, 2021)
3. [Peng J, et al. Impact of station number and duration time per station on the reliability of OSCE scores: a systematic review and meta-analysis](https://doi.org/10.1186/s12909-025-06691-0) (BMC Medical Education, 2025)

> [!TIP]
> **In practice**
> The grid above is the document the committee signs. A coverage matrix built from the station bank shows the same cells with each station's past facility and how often it has run. To see how StudyDrome draws that matrix over a bank or one exam, read /docs/osce/design/in-studydrome-blueprint-and-stations/.