Understanding the Heat-Work Injury Phenomenon in Australia: An Evaluation of Risk, Susceptibility and Attributable Burden

Abstract

Problem statement: Rising global temperatures along with increased frequency of hot days, warm nights, and more frequent and severe heatwaves, are of concern to the scientific community, policy makers, and the general public. The mortality and morbidity rates in the community often rise on hot days, and during heatwave events. The most vulnerable population subgroups include the elderly and the young, those with chronic conditions, and workers. Some workers spend a considerable amount of time in environments that are hot due to radiant heat or ambient heat. The generation of metabolic heat from physical exertion adds to the personal heat exposure. Heat-related illnesses (HRIs) arise when the thermoregulatory system is overwhelmed and the balance between heat gain and heat loss is impaired. While HRI is the well documented direct effect of heat on workers’ health, research has also shown that exposure to hot environments reduces worker productivity and increases the likelihood of work-related injuries (WRIs). The phenomenon of heat-related injuries may be attributable to heat-induced physiological and behavioural factors interacting with existing workplace hazards. As an indirect effect, the link between hot conditions and WRIs may not be fully appreciated. That said, workplace injury represents a major public health burden in Australia and worldwide, and new insights into injury reduction attract considerable interest. There exists a need to better understand the heat-work injury phenomenon and the underlying determinants, and the profile of workers at risk. This improved understanding of injuries associated with work in hot conditions will inform evidence-based adaptive strategies, practical guidelines, and tailored interventions to reduce the risk. Such research is also timely given that an increasing number of workers will be exposed to higher temperatures more often in the future, resulting in a greater burden on workers’ health and safety in a warmer climate. Research gap and opportunity: Following a systematised literature review, it was evident that hot conditions increase the risk of WRIs and HRIs in many industry sectors. However, despite Australia’s generally hot climate, evidence concerning the relationship between high temperatures and WRI is limited. Furthermore, little is also known about the underlying determinants of these injuries and the prevention practices adopted by workplaces in response to heat exposure. Therefore, the body of research in this thesis is designed to provide important new data and perspectives on injury prevention in hot environments, with potential implications for workers, supervisors, industry representatives, professionals and government. Purpose statement: The goal of this research is to obtain a better understanding of the heat-work injury phenomenon, thereby contributing to new knowledge that may be useful in reducing the frequency and incidence of workplace injuries. The project aims to achieve this goal by: (1) systematically examining the association between ambient heat and WRI in Australia; (2) exploring the underlying determinants of heat-related work injuries; and (3) exploring stakeholder perspectives regarding prevention and management. Research questions: 1. What is the epidemiology of heat-related injury and how does it vary across Australia? 2. What are the stakeholders’ perceptions and experiences of occupational injuries and adaptive strategies adopted to prevent WRI from occurring during hot weather? Methodology: There are two distinct parts to this research to address the abovementioned research questions. Part 1 consists of the analysis of workers’ compensation (WC) claims data and Part 2 consists of surveys of stakeholders. Part 1. Analysis of workers’ compensation claims data: A group of four studies was conducted in Part 1 of the research to investigate the effects of heat on the occurrence of workplace injuries in four Australian cities, namely, Adelaide, Brisbane, Melbourne, and Perth using daily ambient temperature and heatwaves as exposure metrics. Workplace injuries were identified from WC claims data which were obtained from SafeWork SA for data pertaining to Adelaide (2003–2013), and from Safe Work Australia for the other three cities (2005–2016). The WC data were transformed into a daily time-series format and merged with exposure data obtained from the Bureau of Meteorology. Climate data included daily maximum temperature (Tmax) and heatwave severity categories defined using the Excess Heat Factor (EHFsev: low-intensity, moderate/high-severity). Additionally, thermal composite indices of heat stress such as humidex and apparent temperature that combine relative humidity and temperature, and wet-bulb globe temperature and universal thermal comfort index incorporating relative humidity, wind speed, solar radiation and temperature, were also used in the estimation of injury risk. City-specific exposure-response curves summarising the relationship between ambient temperatures and WRI were generated utilising a time-stratified case-crossover design combined with a distributed lag-nonlinear model (DLNM) after adjusting for confounding factors such as day of the week and public holidays. Relative risks (RRs) of WRI at moderately hot (90th percentile) and extremely hot (99th percentile) temperatures, compared with a reference temperature were calculated. Finally, attributable fractions were derived to quantify the risk burden of WRI due to ambient temperatures. Similarly, city-specific associations between heatwaves of varying severity and WRI (restricted to the warm season) were estimated using time-stratified case-crossover design with generalised linear models. Comparisons were made between WRI occurring on heatwave days (defined using EHFsev heatwave categories) and non-heatwave days. Analyses were stratified by worker, work, work environment, and injury characteristics to identify at-risk subpopulations and types of injuries related to heat exposure (Tmax and EHFsev). Part 2. Surveys of stakeholders: A group of two studies was conducted in Part 2 of the research which comprised two national online cross-sectional surveys investigating perceptions of key stakeholders (such as health and safety professionals (HSPs) and health and safety representatives (HSRs)) on heat-associated injury risks, determinants, management and prevention. Collected data included perspectives on injury experiences, current preventive measures, workplace training, policies and guidelines, barriers and suggestions for prevention, and productivity loss. Descriptive analysis and log-Poisson regression models were conducted to identify risk factors associated with the reported frequency of injury experience. Main findings: Part 1. Ambient temperature and WRI: In Adelaide, as the daily Tmax rose above 25 °C, the risk of WRI also increased. Compared with 25 °C, there was an 8% (RR 1.08; 95% CI: 1.05–1.12) and 30% (RR 1.30; 95% CI: 1.18–1.44) increase in WRI associated with moderately hot temperatures and extremely high temperatures (defined above), respectively. The proportion of WRI attributable to hot temperatures was 2.1% (95% CI: 1.21–2.98%) with moderately hot temperatures responsible for a higher fraction than extremely hot temperatures (1.5% vs 0.6%). Findings varied for the other three cities. In Melbourne, there was a 5% (RR 1.05; 95% CI: 0.99–1.10) and 14% increase in WRI (RR 1.14; 95% CI: 1.03–1.25) associated with moderately and extremely hot temperatures compared with the median Tmax of 20 °C, respectively. On the other hand, there were no observed effects in Brisbane or Perth, with the exception of traumatic injuries that increased by 17% (RR 1.17; 95% CI: 1.03–1.35) during extreme heat in Perth. Nevertheless, in all three cities there was a decreased injury risk at cooler temperatures which was greater in Brisbane resulting in a higher attributable fraction of WRI due to temperatures in Brisbane (26.5%) than in Perth (5.7%) and Melbourne (1.9%). Who is affected?: Associations between extreme heat and WRI were observed among several groups of workers in Adelaide. These included males and females, and both young workers (15–24 years), and experienced workers. Specific occupations affected included food service and warehouse workers, workers in medium-strength (somewhat physically demanding) occupations, workers in regulated indoor climates, and industries such as ‘electricity, gas and water’ and ‘transport and storage’. In Melbourne, young workers, female workers, workers in regulated indoor climates, and workers ‘in a vehicle/cab’, workers in medium-strength occupations, and those in indoor industries were identified to be at-risk of heat-related injuries. What types of injuries occur in hot conditions?: The types of injuries that significantly increased in Adelaide during moderate and extreme heat were: burns, wounds, lacerations and amputations, along with injuries resulting from vehicle incidents, ‘heat, electricity and other environmental factors’ and ‘chemical and other substances’. In Melbourne, the types of injuries associated with heat exposure were: traumatic injuries and injuries from ‘being hit by moving objects’. Injuries attributed to ‘heat, electricity and other environmental’ factors and ‘mental stresses’ also increased at extremely high temperatures in Melbourne. Part 1. Heatwaves and WRI: Uniform impacts of heatwaves were observed in Brisbane (subtropical climate) and Melbourne, Perth and Adelaide (temperate climates). For three cities (Brisbane, Melbourne and Perth), there was either a small-reduction in risk or a null effect during heatwaves of low-intensity. However, injury risk increased consistently across the four cities during moderate/high-severity heatwaves with the highest effect estimate evident in Brisbane (RR 1.45; 95% CI: 1.42–1.48), followed by Adelaide (RR 1.31; 95% CI: 1.28–1.34), Perth (RR 1.26; 95% CI: 1.24–1.29) and Melbourne (RR 1.25; 95% CI: 1.22–1.28). Who is affected during heatwaves?: Stratified analysis by worker characteristics identified highest associations between heatwaves of moderate/high-severity and the risk of WRI among male workers, workers aged up to 24 years, apprentice/trainee workers, new workers, and workers working through labour hire firms. When physical demands of the occupation were taken into account, workers employed in the medium-strength and heavy-strength (physically demanding) occupations were at increased risk of WRI during moderate/high-severity heatwaves. Workers in regulated indoor climates and ‘in a vehicle or cab’ environments were also at risk of WRI. Stratified analyses according to industrial sectors identified positive associations in both outdoor and indoor industries. What types of injuries occur during heatwaves?: Injury claims that significantly increased across the four cities during moderate/high-severity heatwaves included those for traumatic injuries, ‘mental health’ and injuries from ‘body stressing’, ‘chemicals and other substances’ and ‘falls, trips, and slips of a person’. Part 2. Stakeholder perceptions: In total there were 307 HSPs who completed the first survey, the majority (74%) of whom acknowledged the potential for increased risk of occupational injuries in hot weather. A variety of injury types and mechanisms were reported, including manual handling injuries, hand injuries, wounds or lacerations, loss of control of power tools, fatigue, and dehydration. Work factors significantly associated with reported injuries included problems with heat-retentive personal protective equipment (PPE); lack of shade for workers; inadequate hydration and rest breaks, and problems with supervision. Whereas ceasing outdoor work when temperatures are extreme can be a preventive measure, only 54% of HSPs reported this occurring in workplaces they visited or managed. Furthermore, less than half (42%) stated the availability of adequate heat training for staff. Reported barriers for prevention included: lack of awareness by workers and supervisors of injury risks, and management concerns about productivity loss and/or deadlines. A second survey was conducted among workplace HSRs. In total, 222 HSRs completed the survey. Overall, 43% of respondent reported that injuries or incidents caused by hot/very humid weather occur sometimes/often in their workplace. Factors found to be associated with these injuries included ‘the wearing of PPE’, ‘inadequate resources and facilities’ and being a new worker. For outdoor workers, the most frequently adopted preventive measures were provision of PPE, sunscreen and access to cool drinking water. HSRs reported more injuries if certain preventive measures (i.e. rescheduling work to cooler times and shaded rest/work areas) were adopted never/rarely/sometimes. Access to cool drinking water and provision of PPE were the most frequently adopted preventive measures for indoor workers. HSRs reported that more injuries occurred if self-pacing, shielding of heat sources and adequate ventilation were adopted never/rarely/sometimes. Implications of findings: Part 1. Ambient temperature and WRI: The findings from this research show that WRI do not only occur in extreme temperatures but also during moderate temperatures and that the risks apply to both outdoor and some indoor workers. Interestingly, in Adelaide which had the greatest risks, the burden of WRI was substantially higher during moderate hot temperatures than extremes. This finding suggests that while extremely hot temperatures that rarely occur are dangerous for workers, the risks during the more common hot days should not be ignored. On the other hand, a higher burden of WRI was seen in Brisbane than in Melbourne and Perth. This finding has implications for the future as extreme hot days become more frequent. This study clearly indicates that increasing exposure to higher temperatures poses a risk for workers’ health and safety. Part 1. Heatwaves and WRI: The consistency in the impacts of moderate/high-severity heatwaves on WRI across study sites suggest that it is not only severe heatwaves (during which work may cease) that are cause for concern. Forecasts for moderate-severity heatwaves, based on EHF, should also signal the need for heightened heat awareness and preventive measures, to minimise the risks to workers in Australian workplaces. Part 2. Stakeholder perceptions: The stakeholder evidence suggest that the burden of heat-related work injuries could be reduced by wider adoption of prevention measures such as work rescheduling, self-pacing, provision of shade and adequate ventilation. There should also be an increased awareness of heat as an occupational hazard, particularly with regard to injury occurrence and modifiable risk factors. The integration of multiple lines of evidence (Parts 1 and 2) suggests that underlying mechanisms of injuries in hot weather are complex and multi-factorial. Strengths and weaknesses: There are a number of strengths of this research. Firstly, the research has evaluated the risk and susceptibility of workers and quantified the associated attributable burden of heat and WRI across four Australian cities with diverse climate and worker profiles. This is the first time such an extensive multi-city study has been conducted in Australia on heat-related work injuries and their determinants. A unique multiple data source approach was used to address the research questions, combining both complex statistical models incorporating administrative data, and surveys of key stakeholders. A novel standard heatwave definition across the study sites, and classifications of indoor/outdoor workers by occupation, were incorporated in the estimation of injury risk, adding to the rigour of the study methods. The multiple data source approach yielded a triangulation of findings with the outcomes of surveys supporting and supplementing the major findings of the analysis of the WC data. This research also has a number of limitations. The WC data have inherent limitations in the identification of WRI, as not all workers who are injured lodge a claim. Therefore these data can underestimate the true burden of WRI due to under-reporting. Weather data were assessed at an aggregate level using data from only one monitoring station per study site. Additionally, being an ecological study, personal exposures could not be ascertained. These factors can therefore introduce exposure misclassification. Similarly, surveys may produce biased results if only certain sections of stakeholders (such as those particularly interested in the topic) respond. Furthermore, self-reported responses may introduce recall or reporting bias. Nevertheless, this study has added considerably to the body of knowledge of heat and WRI and may provide valuable evidence for policymakers and workplaces moving forward as heat exposure becomes a greater hazard for exposed workers. Novelty: This is the first Australian investigation to utilise a multiple data source approach to systematically examine the epidemiology and specific determinants of heat-related work-injury occurrence. Conclusions: The findings showed while the highest extreme temperatures have the greatest risk, WRI can also occur on moderate-hot days which are more common and therefore have the greatest burden. The findings also identified a range of modifiable risk factors for injuries. These can be associated with the work (e.g. PPE, no shade), the worker (e.g. lack of awareness) and organisational issues (e.g. poor supervision and lack of training). Recommendations: Based on the findings from this body of research, it is evident that a holistic approach is needed to address the risks of injuries associated with work in hot conditions. The following recommendations are made for jurisdictional safety regulators and policy makers, employers, and manufacturers of PPE. ● It is recommended that there be awareness raising at both the employer and the worker levels, that the health risks of working in hot environments includes injuries as well as HRIs. ● It is recommended that workplaces be aware of forecasted heatwaves (moderate or severe) so that plans, policies and preventive measures can be enacted in advance to minimise risks to the health and safety of workers. ● It is recommended that workplaces examine the potential interaction between existing hazards and heat induced physiological and behavioural effects. ● As some PPE can impair heat loss, and alter behaviour, it is recommended that that research is done on improving PPE comfort and suitability for hot conditions.