Development of Test Toolkit of Hard Review to Evaluate a Random Clinical Decision Support System for the Management of Chronic Adult Diseases

Abstract

Over 30 % of Korean population suffers from chronic diseases, and the management cost of this portion of the population is estimated to be more than 7 % of the gross domestic product and is expected to continue to increase with the aging of society. The u-healthcare (ubiquitous healthcare)-based home healthcare service has been emerging as a solution to this issue. Clinical decision support system (CDSS) is software that provides professional medical knowledge to help people make medical decisions, and is required for the efficient operation of u-healthcare as a core technology. U-healthcare-based CDSS is meant to improve accessibility of medical information for the public, but it at the same time carries a substantial risk of misuse. Hence, a certain level of standards and specifications is required to guarantee the quality of u-healthcare-based CDSS to protect public health. In this context, this project aimed to develop technologies to evaluate reliabilities of u-healthcare-based CDSS dealing with diabetes mellitus and hypertension. This study created a Test Toolkit to evaluate random CDSS with a function to confirm results obtained from CDSS performance evaluation technologies. Results of CDSS were obtained based on screening test items. The design descriptions are defined as (1) creation of test scenario to test CDSS, (2) grading function using returned answers to CDSS, and (3) confirmation of test result. The functions of the Test Toolkit are divided into two types. (1) Test scenarios and the answers to the scenarios are created by reading and fabricating standard scenarios from the DB based on test items input by the evaluator. (2) Appraisal CDSS creates a file of result values using a given test scenario file, and then a CDSS evaluation report, classified by test item, is formed by comparing answers to those found in (1). It should be noted that functions (1) and (2) are completely independent. Conclusively, researchers developed a toolkit to evaluate clinical reliabilities and software stabilities of CDSS for healthcare free from misjudgments caused by human factors and to speed up overall evaluation processes tremendously and efficiently. This toolkit will ensure proper usage of CDSS by the public for health benefits.