Affordable measurement of core body temperature (Tc) in a continuous, real-time fashion is now possible. With this advance comes a new data analysis paradigm for occupational epidemiology. We characterize issues arising after obtaining Tc data over 188 workdays for 83 participating farmworkers, a population vulnerable to effects of rising temperatures due to climate change. We describe a novel approach to these data using smoothing and functional data analysis. This approach highlights different data aspects compared with describing Tc at a single time point or summaries of the time course into an indicator function (e.g., did Tc ever exceed 38 °C, the threshold limit value for occupational heat exposure). Participants working in ferneries had significantly higher Tc at some point during the workday compared with those working in nurseries, despite a shorter workday for fernery participants. Our results typify the challenges and opportunities in analyzing Big Data streams from real-time physiologic monitoring.
Keywords: LOESS; core body temperature; farmworker; functional data analysis; heat stress; heat-related illness; occupational epidemiology; smoothing.