Recent years have seen an increase in the use of technology to measure behaviors in epidemiologic research. This has clear advantages because it reduces the need to rely on self-reporting and objective yet potentially biased measurements. This is particularly advantageous when the behaviors being measured are embedded in the complexity and unpredictability of the everyday world. Two recent examples of the success of this approach are studies on driving1 and hazard detection2 among persons with visual impairment. Pundlik et al2 take an in-the-world approach with their randomized clinical trial examining the at-home use of a chest-mounted collision warning device among persons with visual impairment and blindness. The device automatically detects potential collision hazards and relays their presence via a vibrotactile alert delivered to the person’s wrist. The rate (median [IQR]) of contacts with obstacles when the device was in the active intervention mode was 9.3 (8.4) vs 13.8 (17.0) when the device was in silent (non-alert) mode, reducing collisions by approximately 37%. While laboratory studies have demonstrated improvements in hazard avoidance with wearable devices,3 to our knowledge, this is the first study conducted in everyday settings; thus the results have the potential for greater generalizability.
McGwin G, Owsley C. The Complexity of Integrating Observations Into Observational Research. JAMA Ophthalmol. 2021;139(9):1005–1006. doi:10.1001/jamaophthalmol.2021.2633
Artificial Intelligence Resource Center
Customize your JAMA Network experience by selecting one or more topics from the list below.