Every OSCE variant keeps the three core ideas: the same task for everyone, wide sampling, and scoring by a trained observer. Each variant changes one thing. The OSPE changes what is tested. The TOSCE and GOSCE change who is tested. Linked stations change the shape of a station. Sequential testing changes how many stations a candidate sits. Virtual and hybrid delivery change where the exam happens. This page compares seven variants and says when each one fits.
Why the variants exist
Hodges called the OSCE's history variations on a theme. The original design solved one problem: the unreliable long case. Programs then bent it to solve others. Some wanted to test lab skills rather than clinical ones. Some wanted to test a team. Some could not afford to run every candidate through every station. Some, in 2020, could not put candidates in a room at all.
Each variant is a trade. It buys something the standard circuit does not give. It pays with something the circuit does give. A lead should know both sides before the blueprint is written. The variant changes what the blueprint can cover.
What is an OSPE?
An OSPE is an objective structured practical exam. It applies the circuit to lab and technical skills rather than to clinical encounters. A station presents a specimen, a slide, an instrument, a set of results, or a piece of kit. The candidate identifies, interprets, calculates or demonstrates. Many OSPE stations have no examiner in the room. The candidate writes an answer that is marked later.
The OSPE fits the basic sciences, pharmacy, lab medicine and any program where the skill is technical rather than interpersonal. It costs less to run than a clinical OSCE. It needs fewer examiners and no simulated patients. Its limit is the same as its strength. With no observer, it cannot score process, communication, or safe behavior. A program that needs those keeps observed stations for them.
What are TOSCE and GOSCE?
A TOSCE is a team OSCE. A team of candidates works through each station together. Each member is scored on their own actions, on the team's process, or on both. A GOSCE, or group OSCE, is usually the same idea in a formative setting. A small group works through stations with an examiner who both teaches and scores.
Both variants buy something the standard circuit cannot: evidence about teamwork, leadership and handover. Both pay for it in reliability. When several candidates share the same performance, each score depends on the others'. The scoring scheme has to separate one person's actions from the group's result. Examiners need training on that split. Programs use the TOSCE for cross-profession work and for formative exams. Few use it as the sole basis for a high-stakes decision about one person.
What is a linked station?
A linked station consists of two stations in a row in one scenario. The candidate takes a history in the first. In the second, they present the case, write it up, or interpret the results. The AMEE guide on OSCE organization lists it as one of four station types. The others are observed stations, unobserved stations such as data interpretation, and stations built around manikins or part-task trainers.
The link buys realism. It gives a longer task without a longer single station. It costs care in circuit design. No candidate can start the circuit on the second station of a pair. No candidate can end on the first. The rotation must place the pair in the order everyone meets it. The link also concentrates risk. A problem in the first station spoils the second for that candidate. The original 1975 design already used the idea. It paired a procedure station with an interpretation station.
How does a sequential OSCE work?
A sequential OSCE runs a short screening circuit for everyone. A second circuit runs only for candidates whose screening result is near or below the pass mark. Candidates who clearly pass the screen are done. Candidates near the line sit more stations. The decision about them then rests on more evidence.
The published Leeds design used 12 screening stations and 12 sequential stations. In one cohort of 228 students, 13.5 percent sat the second test. One student, 0.4 percent, would have passed under the previous system and failed on the extended testing. Nine students, 3.9 percent, would have needed a retest under the old system and instead passed the full sequence. The full sequence gave a reliability estimate of 0.79. The saving is examiner and simulated patient time for the majority who pass the screen. The cost is a two-stage timetable. It needs a second circuit that is ready and a standard-setting design that handles two sittings. The page on sequential OSCEs covers the screen length and where its cut sits. It suits large cohorts with a stable station bank.
What can a virtual OSCE assess?
A virtual OSCE moves the encounter onto a video platform. The candidate, the examiner, and the simulated patient join from separate places. A systematic review by Chan and colleagues found the format feasible across health professions. Saad and colleagues evaluated three Australian schools. They describe circuits of six to eight stations of eight to twelve minutes. Each station had a concierge to manage the video rooms. Recordings later supported appeals.
The same evaluation states the limit plainly. The virtual format could assess only about half of the skill range the in-person exam covered. History taking, communication, and data interpretation transferred. Physical examination and procedures did not. Security was weaker with candidates off campus. The authors' advice was to invest in staff training and planning rather than in technology.
A validity argument for a virtual OSCE has to address one step directly. An online encounter can slide from "shows how" toward "knows how". The candidate describes what they would do instead of doing it. Station writing for the format has to keep the candidate acting, not narrating.
What is a hybrid OSCE?
A hybrid OSCE combines multiple delivery modes in a single exam. Communication and history stations run by video. Examination and procedure stations run in person, often on a smaller circuit. Some programs also mix scoring modes. A remote examiner watches a live stream of an in-person station.
The hybrid keeps the skills the virtual format loses. It keeps some of the savings in rooms, travel, and simulated patient time. It costs two sets of logistics and two contingency plans. It also needs a cross-mode check. A video history station is not the same as the same case in person. Their scores should not be pooled without evidence that they behave alike.
Which variant fits which need?
Variant | What changes | Buys | Costs | Fits |
|---|---|---|---|---|
Standard OSCE | Nothing | The full set of design guarantees | Examiner and simulated patient time | Summative clinical assessment |
OSPE | What is tested | Cheap, wide sampling of technical skills | No process or communication scoring | Basic sciences, lab work, pharmacy |
TOSCE / GOSCE | Who is tested | Evidence on teamwork and handover | Reliability for the individual | Cross-profession and formative work |
Linked stations | Station shape | Realistic longer tasks without long stations | Rotation limits; concentrated risk | Consult-then-manage skills |
Sequential | Stations per candidate | Examiner time saved on clear passes; more evidence on borderline candidates | Two-stage timetable and standard setting | Large cohorts, mature banks |
Virtual | Where it happens | Access, recording, no travel | About half the skill range; weaker security | Communication-heavy or remote programs |
Hybrid | Where each station happens | Keeps hands-on skills, saves on the rest | Double logistics; cross-mode checks | Programs with a dispersed cohort |
Limits worth stating
The variants are not interchangeable. A program that swaps its summative OSCE for a virtual one has changed what the exam can certify. Its regulator may notice first. A program that adopts sequential testing without a stable station bank will find the second circuit weaker than the first. A TOSCE score is a poor basis for a pass-or-fail decision about an individual.
The safest path is to keep the standard circuit as the summative instrument. Use a variant where it clearly buys something the program needs. An OSPE for lab skills. A GOSCE for teaching. A sequential design for cohort size. A virtual station for a skill that transfers to video. Each choice should appear in the blueprint and in the validity argument.
Frequently asked questions
What is the difference between an OSCE and an OSPE?
An OSCE assesses clinical encounters. An examiner usually observes the candidate with a simulated or real patient. An OSPE uses the same circuit format for both practical and lab skills: specimens, instruments, results, and calculations. Many OSPE stations are unobserved and marked from written answers. They cannot score communication or process.
Is a virtual OSCE as valid as an in-person one?
Not for the whole domain. Evaluations report that the virtual format assessed about half of the skill range of the in-person exam. History taking, communication, and interpretation transferred well. Physical examination and procedures did not. A virtual OSCE is valid for what it can observe. Its blueprint should claim no more than that.
What is a sequential OSCE?
A design in which every candidate sits a screening circuit. Only those near or below the pass mark sit a second circuit. It saves examiners' time on candidates who clearly pass. It provides more evidence for borderline decisions. In the published Leeds design, 13.5 percent of one cohort sat the second circuit.
What is a linked station in an OSCE?
Two stations in a row in one scenario. A history in the first, then a presentation or a management plan in the second. The rotation must be set so that every candidate meets the pair in order. The idea dates from the original 1975 design, which paired a procedure with an interpretation station.
Can a team OSCE be used for a summative decision?
It can contribute, but rarely on its own. When several candidates share one performance, each score depends partly on the others. That lowers reliability for the individual. Programs use team formats for cross-professional and formative work. They keep summative decisions about one person on stations scored one person at a time.