Impact of rising temperatures on occupational accidents in Brazil in the period 2006 to 2019: A multiple correspondence analysis

•We combined occupational accidents data and climatic variables into a new database.•We verified a positive association between heat stress and occupational accidents.•There is an association between high levels heat stress and death of the worker.•Workers of color are associated to higher levels of heat stress.•Unprecedented results and database on Brazilian research. Work accidents result in consequences to the employment of the population and increasing public spendings. Caused by workplace and work activity characteristics, occupational accidents may also derive from ergonomics and comfort issues. Heat stress is a discomfort factor that affects workers when exposed to temperatures above the body limits, resulting in exhaustion, dizziness, reduced cognitive performance and, eventually, injuries and accidents. Under the current climate change scenario characterized by increase of temperature projections all around the world, the heat stress issue becomes even more significant. However, in Brazil, this topic is yet little explored, especially regarding the investigation of historical data on occupational accidents considering the climatic variables. This paper aims at filling a part of this gap by presenting a new database that unifies a work accident database - recording from 2006 to 2019 - with meteorological data of the place and time of the accident. We investigate the relationship between these two datasets through the application of Multiple Correspondence Analysis (MCA) in the R Software. Our results show some association between accident variables and heat stress variables. We identify some of the more critical workers' characteristics in this context and the most exposed regions of Brazil. Our database allows the continuity and expansion of this type of research in Brazil, and the MCA results point to a positive association between the occurrence of accidents with climatic variables. It may pave a new path for research that can detail and deepen the discussion on the behavior of these variables.