Development and Evaluation of a Mobile-Based Asthma Clinical Decision Support System to Enhance Evidence-Based Patient Management in Primary Care Publisher



Abtahi H¹ ; Amini S² ; Gholamzadeh M^{1, 3} ; Gharabaghi MA^{1, 4} Show Affiliations
Source: Informatics in Medicine Unlocked Published:2023

Abstract

Background: Despite tremendous efforts, satisfactory adherence to asthma care guidelines has remained an unmet need worldwide. There is a growing body of evidence that applying informatics-based solutions such as clinical decision support systems (CDSS) could enhance guideline adherence. The main objectives of our research include proposing a framework for translating evidence into a guideline-based CDSS and developing an asthma-specific CDSS based on the proposed framework. Methods: The stepwise framework for those who want to implement guideline-based CDSS in form of a simple-to-use application was described in this research. Nested decision tree modeling was performed in an iterative process to model patient-related decisions under expert consultation. Six expert panel members were asked to evaluate the application based on actual patient data, once based on paper-based guidelines, and once based on the expert system to calculate the accuracy of our CDSS. In the usability evaluation phase, evaluators completed tasks based on the think-aloud protocol to find usability problems. Then, to determine the level of usability a standard SUS questionnaire was applied. Results: In total, 28 clinical decision trees (CDTs) and more than 220 rules were necessary to cover the whole guideline. Moreover, 336 knowledge statements were extracted from the GINA guideline. After modification with feedback from seven experts, our application was developed on the Android Platform. For asthma diagnosis, CDSS had a sensitivity, specificity, and accuracy of 100%, 86.71%, and 93.33%, respectively. The think-aloud usability evaluation revealed that the total problems (n = 32) were classified deductively into layout problems (43.8%), terminology problems (9.4%), data entry problems (18.8%), and comprehensiveness problems (28.1%). The overall usability score level was 90.22(±9.487), while learnability and usability scores were 4.43 and 4.52 out of 5, respectively. Conclusion: The ginasthma mobile-based CDSS can provide healthcare providers with an automated individualized action plan by combining patient data with embedded knowledge from the GINA guidelines with high accuracy. © 2023