Abstract
Objective: This study evaluated a Web-based expert system for training junior residents in differential diagnosis using a Web site usability questionnaire scale (WUS) method.
Method: The Web-based expert system for training junior residents in differential diagnosis of the Japanese Organization for Quality Health Care was evaluated. The estimators were physicians with more than 3 years of clinical experience. Two evaluation figures were used to simulate eight symptoms and signs (chronic cough, nausea and vomiting, headache, sleep disturbance, vertigo, lymphadenopathy, abdominal pain, and hypertension): 1) to compare it with the results of other WUS; 2) to validate the degree to which it helped junior residents to learn differential diagnosis. We evaluated WUS and visual analogue scales (VAS) designed to evaluate the differential-diagnosis-specific learning support function. A mailed, self-administered anonymous questionnaire method was used. Statistical analysis was performed with SPSS (IBM, Armonk, NY) and R ver. 2.13.0.
Results: 1) The WUS scores of “Favorite” and “Useful” (p<0.01) and “Operability” and “Viewable” (p<0.05) were significantly lower than for Web sites reported in a paper on WUS development. 2) For the VAS of “chronic cough”, a significant difference between “attending physicians and senior residents” and “attending physicians and other physicians” was seen (p<0.05).
Discussion: The lower WUS might have been caused by an insufficient explanation of the use of Web applications with the mailed, self-administered, anonymous questionnaire method used.
Conclusion: WUS was useful for evaluating the Web-based expert system for training in differential diagnosis. Usability evaluation methods might also identify design challenges involving Web-based expert systems.
