Examining Agricultural Workplace Micro and Macroclimate Data Using Decision Tree Analysis to Determine Heat Illness Risk.

OBJECTIVE: This study was designed to examine the associations between regional weather data and agricultural worksite temperatures in Florida.

METHODS: Florida farmworkers (n = 105) were each monitored using iButton technology paired with simultaneous data from regional weather stations. Conditional inference tree models were developed for (1) regional environmental temperatures and iButton (worksite) temperatures, and (2) regional heat index (HI) and iButton HI.

RESULTS: Worksite temperatures were partitioned by regional temperature at the primary node of 29.1°C. Worksite HI was partitioned at nodes of 33.0°C, 36.0°C, 37.0°C, and 40.0°C. The nodes at 33.0°C and 40.0°C mirror the National Weather Service's category entry points for "extreme caution" and "danger" regarding the risk of developing heat-related illness.

CONCLUSIONS:: Regional weather data have the potential to provide estimations of worksite environmental conditions allowing employers to quickly implement strategies to protect workers.