COMPETENCIES TRAINING AND ASSESSMENT IN PATHOLOGY AND LABORATORY MEDICINE USING PATIENT SIMULATIONS

 

Fred Dee*, Clarence Kreiter, Christopher Goerdt, Timothy Leaven and Thomas Haugen.  University of Iowa, Iowa City IA, 52242 U.S.A.

 

PURPOSE 

A novel web-based computer-assisted patient simulation (LabCAPS) has been designed to foster evidence-based use of laboratory resources and reduction in unnecessary costs.  The simulation’s ability to generate valid and reliable scoring of performance is crucial to its educational and assessment role.

 

METHODS

LabCAPS is created in a MySQL database.  Perl-CGI scripting dynamically generates the simulations, and serves them over the web.  When a new case is entered, a web-based editor allows entry of history and physical findings.  Default normal results of 431 tests, procedures, and images can then be changed for each individual case to abnormal by the case editor.  When cases are engaged the LabCAPS web interface allows the trainee to sequential order and interpret tests and procedures through several encounters, and arrive at a diagnosis, all from check-box menus.  To develop a scoring key for comparison with trainee’s responses, experts engage the simulations.  The resultant electronically generated scoring key is based on the proportion of experts selecting responses, with disagreements adjudicated around evidence-based guidelines.

 

RESULTS

There are currently forty cases. Medical students in our Pathology Course rated three LabCAPS prototype units as the first, second, and fourth best units among 17 more traditional units.   Formative evaluations and reliability and validity studies of LabCAPS are underway in our Laboratory Medicine Course.  

 

CONCLUSION/FUTURE DIRECTIONS

Computer assisted simulations are expensive to develop and score.  This project, supported by a grant from the National Library of Medicine, investigates methodologies designed to significantly reduce the cost of both case development and scoring.