Objective structured clinical examinations

Abstract

The introduction of the objective structured clinical examination (OSCE) format to the assessment of clinical skills in 1975 Harden initiated a groundswell of innovation and research that continues to grow. Briefly, an OSCE is a format for performance assessment based on a circuit of stations. At each station examinees complete assigned clinical tasks within the context of a simulation. The most common form of simulation is a standardized patient, an individual trained to present a specific patient problem consistently and realistically. Examinees are scored at each station according to predetermined criteria, either by the standardized patient or by an observing rater. Today the OSCE is widely used in the United Kingdom, North America and many other countries to assess examinees at all levels of medical training. There are quite literally thousands of articles published on the OSCE and this chapter highlights the most important developments of the past three decades, including the literature establishing that standardized patients can provide realistic and consistent simulations of patient problems. Other research of note includes studies comparing checklists and rating scales that demonstrate checklists as being more appropriate for assessing structured tasks and junior trainees, and rating scales as better for distinguishing novices from experts. Further highlights include the literature regarding multiple types of raters, such as standardized patients, trained paramedical personnel, and physicians. Standards setting poses unique challenges and suggestions are provided. There has been accumulating evidence regarding the valid interpretation of OSCE scores and this work is linked to the practical information provided in the second half of the chapter. Basic topics including blueprinting, length of the examination, number of stations, score reliability, generalizability coefficients, item-total correlations, other correlation analyses, and issues of test security are reviewed. The initial step in developing an OSCE requires answers to six questions that cover the purpose of the assessment, performance standards, the domains to be assessed, choice of raters, and the number and length of stations. Other topics covered are cost factors, choice of scoring instruments, standardized patient training, tips for developing test materials, and a comprehensive tasks and timeline checklist to guide OSCE developers.