User acceptance testing of the D-MOSS dengue predictive model as an early warning system in Malaysia

Abstract

Dengue remains a significant public health challenge in Malaysia with recurring outbreaks, highlighting the need for effective forecasting tools. This study aimed to evaluate user acceptance of the Dengue forecasting Model Satellite-based System (D-MOSS) in Malaysia, assessing its usability, strengths, challenges, and areas for improvement. A cross-sectional study was conducted among 32 stakeholders who were directly involved in using D-MOSS for dengue management. Data were collected during a national workshop using a structured survey adapted to the specific operational context of D-MOSS. Thematic analysis was employed to analyse open-ended responses using R for keyword-based categorisation and visualisation. Participants reported high comprehension of D-MOSS features, with a 100% understanding of the outbreak thresholds and forecast values. Common strengths included the system's forecasting capability (66%), impact on outbreak management (22%) and user-friendly interface (12%). Challenges highlighted included accuracy and credibility (54%), performance and technical issues (27%), and granularity and usability of data (19%). For improvements, 40% suggested greater customisation, such as smaller spatial resolution, 25% emphasised interface enhancements, 20% advocated for comparative analysis features, and 15% proposed integrating action planning tools. The system was perceived as a helpful tool for assisting proactive dengue management. D-MOSS demonstrated its user-friendliness and stakeholder adaptability, which support its application in dengue surveillance. Addressing identified challenges and enhancing system features could strengthen its utility and integration, contributing to improved dengue surveillance and control efforts.