Uji Komparasi Eksperimental Smart-Anthro Berbasis IoT Dan Antropometri Manual: Analisis Akurasi Teknis Dan Persepsi Kemudahan Pengguna Non-Kesehatan

Abstract

The high prevalence of stunting at the community level necessitates early detection systems that are not only accurate but also adaptable to the capabilities of community health cadres. Conventional anthropometric measurements are often associated with measurement errors, time inefficiency, and difficulties in interpreting nutritional status manually. This study aimed to evaluate the effectiveness of the Internet of Things (IoT)-based Smart-Anthro system in improving measurement accuracy and facilitating stunting assessment among community health cadres. A quantitative study with a quasi-experimental comparative design was conducted involving 30 children aged 24–60 months in Ngleri Village, Gunungkidul, Indonesia. Anthropometric measurements were performed using three instruments: Smart-Anthro, conventional anthropometric tools, and a gold-standard reference instrument. The findings demonstrated that Smart-Anthro achieved an accuracy rate of 93.33%, significantly higher than that of conventional tools (56.67%) (p < 0.001). Regarding time efficiency, Smart-Anthro required an average measurement time of 31.23 seconds, whereas conventional measurements required up to 209 seconds (p = 0.001). Additionally, all participating cadres (100%) reported that the automated system simplified nutritional status assessment by providing immediate interpretation of measurement results (p = 0.008). The study concludes that the Smart-Anthro system effectively reduces measurement bias, enhances accuracy, and accelerates stunting screening at the community level. Future development should incorporate an offline-first data reporting feature to ensure functionality in areas with limited internet connectivity.