Statistics from Altmetric.com
In the environmental epidemiology setting, dozens of time series studies have been used to link high ambient temperatures to daily counts of mortality or hospital admissions.1 ,2 In the occupational setting, studies looking at the effect of short-term exposure to high temperatures have focused on, for example, the occurrence of heatstroke cases in different occupational groups3 or the effects of high temperature on productivity.4 ,5 Xiang et al6 used data from compensation claims to build a time series of daily work-related injuries, and then used time series techniques to examine how fluctuations in ambient temperature are associated with the number of daily injuries.
In their study, ambient temperature was indeed associated with work injuries, but the shape of the association was not what one would expect a priori. We have learnt from mortality and morbidity studies that, as soon as temperatures exceed a certain comfort temperature range, mortality and morbidity risks start to increase, first slowly and then with steep increases when temperatures become extremely hot.7 Interestingly, Xiang et al6 found that as temperatures rise, the number of daily injuries keep increasing but only up to a certain temperature, from which point on the number of injuries starts to decrease. This apparently contradictory finding can be probably explained by the fact that some work activities may be stopped in situations where heat warnings are issued. Thus, at very high temperatures, there is a decrease in the population at risk, and as the time series analysis method used does not take denominators into account, estimated work-related injury risk goes down as well. This is not the case in population-based studies, where all the population remains at risk (eg, of death) when temperatures become extremely hot.
This non-linear dose–response pattern which was consistently observed in almost all industrial sectors—except for critical sectors like electricity, gas and water supply, where production cannot be stopped—has important practical consequences. It indicates that increases in temperature in a range that is not considered to be extreme increase the risk of work-related injuries. Heat prevention plans tend to focus on extreme temperatures,8 and the results from Xiang et al6 seem to suggest that these plans may indeed be effective at that temperature end, since a decrease in work-related injuries was observed. However, hot days that are not extreme, which occur more often and are not usually targeted by prevention plans, may importantly contribute to the burden of work-related injuries. This needs to be taken into account to design new policies and education plans to reduce the harmful effects of heat in the workplace.
Xiang et al6 also identified the industrial sectors that were at higher risk. Not surprisingly, an increased risk was found in sectors that mostly work outdoors, such as agriculture, construction and transport. Indoor industries may tend to control indoor temperature, especially in high-income countries affected by extreme temperatures such as Australia, and in that case ambient (outdoor) temperature is not a good marker to detect an association. Indoor temperature control may be less frequent in medium and small enterprises, where the temperature-related injury risk was observed to be higher.
Often, time series studies on the relationship between temperature and mortality or morbidity exclude from the analyses deaths or hospitalisations from external causes,9 ,10 most of the times because these studies follow from previous ones examining the health effects of air pollution, and air pollution is not believed to contribute to health events falling in the external causes group. However, high temperatures are known to diminish our ability to carry out mental and physical tasks,3 and this may be the mechanism leading to more injuries. Indeed, some studies have found that on hot days there are increased risks of falls, injuries or traffic accidents.11 ,12 The present article by Xiang et al6 further supports the inclusion of injuries in studies aiming to quantify the total effects of heat exposure on health.
Competing interests None.
Patient consent Obtained.
Provenance and peer review Commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.