Feasibility and Utility of Wearable Bluetooth and RFID Sensors to Measure Care Interactions

Abstract

Abstract Clinical care interactions influence important health outcomes in older adults such as nosocomial infections, falls, and care satisfaction. However, traditional measurements of interactions are often time-consuming, costly, biased, or they interfere with normal clinical care. Wearable sensors measure interpersonal contact with a high degree of spatial and temporal detail and accuracy when applied to examining the spread of infection. This study aims to test the feasibility of implementing two types of commercially available wearable sensors to collect spatial and temporal data for measuring care interactions in health care settings. EMBC02 (Bluetooth) and OpenBeacon (RFID) sensors were tested in a simulation laboratory. Five nursing students wore the sensors along with name tags and enacted a scenario of a fallen nursing home resident. Sensor data were plotted in Python and compared with video recordings of the simulated care interaction to determine sensor usability, accuracy, and precision. EMBC02 and OpenBeacon detected multiple wearers and provided spatial and temporal data. OpenBeacon showed better usability and validity than EMBC02 for using proximity data collected by sensors to infer care interactions. Both OpenBeacon and EMCB02 showed some limitations in accuracy and precision, such as increased data missingness due to idling function and high data noise. Bluetooth and RFID sensors measure different aspects of proximity. OpenBeacon outperformed EMBC02 for measuring care interactions. Combining Bluetooth with RFID may provide richer information for measuring and understanding care interactions. Simulation laboratories can be leveraged to test health technology before use in clinical research.