Objective Structured Clinical Examination (OSCE)
An Objective Structured Clinical Examination is a performance-based assessment in which learners rotate through a series of timed stations, each presenting a standardized clinical task scored against a predefined rubric. Because every learner encounters the same scenarios and is scored on the same criteria, the OSCE is a structured way to assess clinical and communication skills more objectively than unstructured observation.
Source: Harden RM, Stevenson M, Downie WW, and Wilson GM, British Medical Journal, 1975
The format comes from a 1975 paper in the British Medical Journal by Harden, Stevenson, Downie, and Wilson, who had students rotate round a series of stations in a hospital ward while examiners scored them against check lists agreed in advance. The acronym as programs use it now, with the word clinical in it, arrived with Harden and Gleeson's 1979 paper in Medical Education, which set out station design and scoring in more detail.
Objective and structured describe the design, not a claim that judgment has been removed. The task, the time limit, the patient portrayal, and the scoring criteria are all fixed before anyone walks in, so a difference in scores is more likely to reflect the learner than which examiner or which patient they happened to draw. Faculty still make judgments at every station, and the quality of an OSCE rests almost entirely on how well the cases and the rubrics were written.
In assessment terms the OSCE sits at the third level of Miller's Pyramid, shows how, where a learner demonstrates a skill under observation. That is a real step up from a written exam, and it is also the ceiling: an OSCE tells you what someone did once in an exam room, not what they do at two in the morning with a frightened family. Programs that treat the OSCE as the whole picture end up making progression decisions from a single annual data point per learner.
Communication stations are where the constraint bites hardest. Recruiting, training, and paying standardized patients holds most programs to one or two encounters per learner per year, which is enough to score a skill and nowhere near enough to build one. Learners can get repetitions between exams by practicing with AI patients, so the OSCE measures something they have actually rehearsed. Extend your SP program, don't replace it.
What this looks like in a program
- Write the checklist and the case together, because a rubric drafted after the case tends to score what is easy to observe rather than what matters.
- Train more than one examiner per station and compare their scores on the same recorded encounter before the exam counts for anything.
- Report station-level results to the competency committee rather than one composite score, since a learner who fails only the disclosure station needs something different from a learner who fails the physical exam stations.
Common questions
What is an OSCE examination?
An OSCE is a clinical exam built from a series of short timed stations, each with one set task and a rubric written before anyone sits the exam. Harden, Stevenson, Downie, and Wilson described the format in the British Medical Journal in 1975 as an alternative to the traditional long case, where a learner's grade depended heavily on which patient and which examiner they happened to draw. Programs use it to make progression decisions about history taking, physical examination, clinical reasoning, and communication.
What does OSCE stand for in medicine?
OSCE stands for objective structured clinical examination. Objective means scoring against criteria fixed in advance rather than against an examiner's overall impression, and structured means every learner rotates through the same stations under the same time limit. Harden and Gleeson set the acronym in its current form in Medical Education in 1979, which is also where station design and scoring were first laid out in detail. It gets shortened to objective clinical examination often enough that the two phrases are used interchangeably, though the structure is the part doing the work.
What happens in a medical OSCE exam?
A learner reads a short task instruction outside the station, spends a fixed number of minutes inside with a standardized patient or a manikin, then moves on when the timer goes. Common tasks are taking a focused history, performing a physical examination, explaining a diagnosis or a procedure, and handling a conversation with a family. An examiner scores the station while it happens, against a checklist of required actions, a rating scale, or both. Stations are scored independently of each other, so a bad first station does not sink the whole exam.
How many stations does an OSCE have?
There is no standard number, and station count is a sampling decision rather than a convention. Reliability comes from sampling widely across stations, cases, and examiners rather than from making any one station longer, so a program running three or four stations should read any single station result as a weak signal about the learner. For scale, the USMLE Step 2 Clinical Skills examination used 12 standardized patient encounters of 15 minutes each, 10 of which counted toward the score, before the NBME and the Federation of State Medical Boards discontinued it on January 26, 2021. AMEE Guide 49 sets out the metrics programs use to check whether their own station set is holding up (Pell, Fuller, Homer, and Roberts, Medical Teacher, 2010).
How are OSCE stations scored?
Stations are scored with a checklist of required actions, a global rating scale, or both, and the choice changes what the score can support. Regehr, MacRae, Reznick, and Szalay compared the two on an OSCE format examination of surgical skills in Academic Medicine in 1998 and found that global rating scales completed by expert examiners had higher inter-station reliability, a Cronbach's alpha of 0.85 against 0.79 for checklists, and better construct validity, since a checklist rewards thoroughness rather than judgment. The usual compromise is a checklist for the actions that have to happen and a rating scale for communication quality or clinical reasoning.
Read next
Last updated August 2026