room assignment august 2023 mechanical engineering

Table IV—2—OIICS Codes (Version 3.0) for Heat-Related Health Effects †

Through a combination of survey staff and a specialized automated coding system, BLS applies OIICS codes to data collected through their worker safety and health surveillance systems, the Census of Fatal Occupational Injuries (CFOI) and the Survey of Occupational Injuries and Illnesses (SOII), to identify and document occupational heat-related deaths and occupational HRIs, respectively. Researchers have also relied on this system for identifying occupational HRIs ( e.g., Spector et al., 2016). However, BLS data does not currently specify discrete codes for all HRIs described in this health effects section. The CFOI is a cooperative program between the Federal Government and the States that relies on various administrative records, including death certificates, to accurately produce counts of fatal work injuries (BLS, 2012). The CFOI examines all cases marked “At work” on the death certificate, and the CFOI database relies on the death certificate (among other sources) to ascertain the cause(s) of death. Further details about BLS reporting using OIICS codes, as well as rates of HRIs, can be found in Section V., Risk Assessment.

A limitation to relying on these coding systems to identify heat-related fatalities and HRIs is underreporting. Numerous studies have found that HRIs are likely vastly underreported (see Section V., Risk Assessment). Reasons for the likely underreporting include underreporting of illness and injuries by workers to their employers (Kyung et al., 2023), underreporting of injuries and illnesses by employers to BLS and OSHA (Wuellner and Phipps, 2018; Fagan and Hodgson, 2017), underutilization of workers' compensation insurance (Fan et al., 2006; Bonauto et al., 2010), influence of structural factors and work culture on workers perceptions about seeking help (Wadsworth et al., 2019; Iglesias-Rios, 2023), and difficulties with determining heat-related causes of death ( e.g., Luber et al., 2006; Pradhan et al., 2019). As a result, there are likely many heat-related fatalities and cases of HRIs that are not ( print page 70711) captured in these coding systems. For a more detailed discussion of underreporting, see Section V., Risk Assessment.

As demonstrated by these coding systems, in which medical providers or other public health professionals assign one or more codes to identify a heat-related fatality or HRI, it is well accepted in the medical and scientific communities that heat exposure, including occupational heat exposure, can result in death and HRIs. Indeed, in its review of the best available scientific and medical literature on the health effects of occupational heat exposure, OSHA identified several studies that relied upon data from these coding systems to determine the incidence or prevalence of heat-related deaths and HRIs in workers. OSHA relies on these studies in both this Health Effects section and Section V., Risk Assessment, of this preamble to the proposed rule.

Heat is the deadliest weather phenomenon in the United States (NWS, 2022). Heat as a cause of death is widely recognized in the medical and scientific communities. Studies investigating relationships between heat and mortality have long demonstrated positive associations between heat exposure and increased all-cause mortality ( e.g., Weinberger et al., 2020; Basu and Samet, 2002; Whitman et al., 1997). As explained below, the connection between heat exposure, the body's physiological responses, and death ( i.e., heat-related death mechanisms) is clearly established. Exposure to occupational heat can be fatal. According to BLS's CFOI, occupational heat exposure has killed 1,042 U.S. workers between 1992-2022 (BLS, 2024c).

Death caused by exposure to heat can occur in occupational settings if the worker's body is not able to adequately cool in response to heat exposure or if treatment for symptoms of heat-related illness is not provided promptly. Nearly all body systems can be negatively affected by heat exposure. Mora et al. (2017) systematically reviewed mechanistic studies on heat-related deaths and identified five harmful physiological mechanisms triggered by heat exposure that can lead to death: ischemia (inadequate blood flow), heat cytotoxicity (damage to and breakdown of cells), inflammatory response (inflammation that disrupts cell and organ function), disseminated intravascular coagulation (widespread dysfunction of blood clotting mechanisms), and rhabdomyolysis (breakdown of muscle tissue). These mechanisms, with the exception of rhabdomyolysis, are associated with the development of heat stroke. Rhabdomyolysis, which is a potentially fatal illness resulting from the breakdown of muscle tissue, can also occur in conjunction with or in the absence of heat stroke. For a more detailed discussion on rhabdomyolysis, see Section IV.H., Rhabdomyolysis. Mora et al. (2017) also identified seven vital organs that can be critically impacted by heat exposure—the brain, heart, kidneys, lungs, pancreas, intestines, and liver. Across the five identified mechanisms and seven vital organs, Mora et al. (2017) found medical evidence for twenty-seven pathways whereby physiological mechanisms triggered by heat exposure could lead to organ failure and fatality.

The most common cause of heat-related occupational deaths is heat stroke. Heat stroke is a potentially fatal dysregulation of multiple physiological processes and organ systems resulting in widespread organ damage. Heat stroke is typically marked by significant elevation in core body temperature and cognitive impairment due to central nervous system damage. The physiological mechanisms involved in the development and progression of heat stroke are discussed in more detail in Section IV.E., Heat Stroke.

The identification of deaths caused by heat exposure can take place in a few different ways. Healthcare professionals may identify heat-related deaths in medical settings. For example, a heat-related death may be identified if an individual experiencing heat stroke presents to an emergency room and then later dies. The heat-related nature of the death should be documented by the healthcare professional in the chief complaint field during medical history taking and selection of relevant ICD diagnosis codes. The ICD system allows for identification of heat as either an underlying cause of death or a significant contributing condition. The ICD-10 instruction manual defines underlying cause as “(a) the disease or injury which initiated the train of morbid events leading directly to death, or (b) the circumstances of the accident or violence which produced the fatal injury” (WHO, 2016, p. 31). A significant contributing condition is defined as a condition that “contributed to the fatal outcome, but was not related to the disease or condition directly causing death” (WHO, 2004, p. 24).

Medical examiners or coroners can also identify heat as a cause of death or significant condition contributing to death during death investigations, which should be noted on the deceased individual's death certificate. The National Association of Medical Examiners (NAME), a professional organization for medical examiners, forensic pathologists, and medicolegal affiliates and administrators, defines “heat-related death” as “a death in which exposure to high ambient temperature either caused the death or significantly contributed to it” (Donoghue et al., 1997). This definition was developed in an effort to standardize the way in which heat-related deaths were identified and documented on death certificates. According to the NAME definition, cause is ascertained based on circumstances of the death, investigative reports of high environmental temperature ( e.g., a known heat wave), or a pre-death temperature ≥105 °F. Cause is also indicated in cases where the person may have a lower body temperature due to attempted cooling measures, but where the individual had a history of mental status changes and specific toxicological findings of elevated muscle and liver enzymes. Heat may be designated as a “significant contributing condition” if: (1) “antemortem body temperature cannot be established but the environmental temperature at the time of collapse was high”; and/or (2) heat stress exacerbated a pre-existing disease, in which case heat and the pre-existing disease would be listed as the cause and significant contributing condition, respectively, or vice versa. Importantly, Donoghue et al. note “The diagnosis of heat-related death is based principally on investigative information; autopsy findings are nonspecific.” (Donoghue et al., 1997). While this definition is the official definition of this professional organization, other definitions or processes for determining whether or not a death is heat-related may be used.

Additionally, there are processes in place to identify and document deaths that are work-related. Death certificates include a field that can be checked for “injury at work” (Russell and Conroy, 1991). Further, work-related fatalities due to heat are identified and documented through the CFOI (for more details, see Section IV.C., Overview of ICD and OIICS Codes for Heat-Related Health Effects). ( print page 70712)

Occupational heat exposure has led to worker fatalities in both indoor and outdoor work settings and across a variety of industries, occupations, and job tasks (Petitti et al., 2013; Arbury et al., 2014; Gubernot et al., 2015; NIOSH, 2016; Harduar Morano and Watkins, 2017). BLS's CFOI identified 1,042 U.S. worker deaths due to heat exposure between 1992 and 2022, with an average of 34 fatalities per year during that period (BLS, 2024c). Between 2011 and 2022, BLS reports 479 worker deaths (BLS, 2024c). During the latest three years for which BLS reports data (2020-2022), there was an average of 45 work-related deaths due to exposure to environmental heat per year (BLS, 2024c). However, for the reasons explained in Section V., Risk Assessment, these statistics likely do not capture the true magnitude and prevalence of heat-related fatalities because of underreporting.

There are numerous case studies documenting the circumstances under which occupational heat exposure led to death among workers. For example, in three NIOSH Fatality Assessment and Control Evaluations (FACE) investigations of worker fatalities, workers died of heat stroke after not receiving prompt treatment upon symptom onset (NIOSH, 2004; NIOSH, 2007; NIOSH, 2015). Another case report of a farmworker who died due to heat stroke indicates that confusion the worker experienced as a result of heat exposure may have played a role in his ability to seek help (Luginbuhl et al., 2008). Additional case reports show workers have collapsed and later died while working alone, such as in mail delivery (Shaikh, 2023), and that worker distress has been interpreted as drug use as opposed to symptoms of heat illness (Alsharif, 2023).

OSHA's review of the scientific and medical literature indicates that occupational heat exposure can and does cause death. The physiological mechanisms by which heat exposure can result in death are clearly established in the literature, and heat exposure being a cause of death is widely recognized in the medical and scientific communities. Indeed, occupational surveillance data demonstrates that numerous work-related deaths from occupational heat exposure occur every year.

Among HRIs, the most serious and deadly illness from occupational heat exposure is heat stroke. NIOSH (2016) defines heat stroke as “an acute medical emergency caused by exposure to heat from an excessive rise in body temperature [above 41.1 °C (106 °F)] and failure of the [body's] temperature-regulating mechanism.” When this happens, an individual's central nervous system is affected, which can result in a sudden and sustained loss of consciousness preceded by symptoms including vertigo, nausea, headache, cerebral dysfunction, bizarre behavior, and excessive body temperature (NIOSH 2016).

Because progression of symptoms varies and involves central nervous system function, it may be difficult for individuals, or those they are with, to know when they are experiencing serious heat illness or to understand that they need urgent medical care (Alsharif, 2023). If not treated promptly, early symptoms of heat stroke may progress to seizures, coma, and death (Bouchama et al., 2022). Thus, heat stroke is often referred to as a life-threatening form of hyperthermia ( i.e., elevated core body temperature) because it can cause damage to multiple organs such as the liver and kidneys. Of note, the term “stroke” in “heat stroke” is a misnomer in that it does not involve a blockage or hemorrhage of blood flow to the brain.

There are two types of heat stroke: classic heat stroke (CHS) and exertional heat stroke (EHS). CHS can occur without any activity or physical exertion, whereas EHS occurs as a result of physical activity. CHS typically occurs in environmental conditions where ambient temperature and humidity are high and is most often reported during heat waves (Bouchama et al., 2022). It is most likely to affect young children and the elderly (Laitano et al., 2019). Studies have found that EHS can occur with any amount of physical exertion, even within the first 60 minutes of exertion (Epstein and Yanovich, 2019; Garcia et al., 2022). Additionally, EHS can occur in healthy individuals who would otherwise be considered low risk performing physical activity, regardless of hot or cool environmental conditions (Periard et al., 2022; Epstein et al., 1999).

Cases of heat stroke can be identified in a few ways. Medical personnel who make a formal diagnosis of heat stroke record the corresponding ICD code in the patient's medical record. Medical examiners also identify heat stroke as a cause of death or significant condition contributing to death and note it on the deceased individual's death certificate.

Heat stroke happens when the body is under severe heat stress and is unable to dissipate excessive heat to keep the body temperature at 37 °C (98.6 °F), resulting in an elevated core body temperature (Epstein and Yanovich, 2019). The hallmark characteristics of heat stroke are: (1) central nervous system (CNS) dysfunction, including encephalopathy ( i.e., brain dysfunction manifesting as irrational behavior, confusion, coma, or convulsions); and (2) damage to multiple organs, including the kidneys, liver, heart, pancreas, gastrointestinal tract, as well as the circulatory system. There are three accepted mechanisms through which heat exposure can cause CNS dysfunction and/or multi-organ damage (Bouchama et al., 2022; Garcia et al., 2022; Iba et al., 2022). All three mechanisms share a common origin: heat exposure contributes to excessive heat stress, which results in hyperthermia.

One mechanism of heat stroke is reduced cerebral blood velocity (CBV) (an indicator of blood flow to the brain) that results in orthostatic intolerance ( i.e., the inability to remain upright without symptoms) (Wilson et al., 2006). As individuals experience whole body heating, CBV is reduced and cerebral vascular resistance (the ratio of carbon dioxide stimulus to cerebral blood flow) increases. These changes ultimately contribute to reduced cerebral perfusion (flow of blood from the circulatory system to cerebral tissue) and blood flow, as well as orthostatic intolerance (Wilson et al., 2006).

Another mechanism is damage to the vascular endothelium. Hyperthermia can damage or kill cells in the lining of blood vessels, known as the vascular endothelium. The body responds to vascular endothelium damage through a process called disseminated intravascular coagulation (DIC). DIC is characterized by two processes: (1) tiny clots form in the tissues of multiple organs, and (2) bleeding occurs at the sites of those tiny clots. DIC is extremely damaging and results in injury to organs (Bouchama and Knochel, 2002). Namely, DIC limits the delivery of oxygen and nutrients to several organs including the brain, heart, kidneys, and liver. Thus, DIC can result in both CNS dysfunction and multi-organ damage. Additionally, damage to the vascular endothelium makes it more permeable and creates an imbalance in the substances that control blood clotting, ( print page 70713) which promotes abnormal and increased blood clotting (Bouchama and Knochel, 2002; Wang et al., 2022).

A third mechanism is damage to the cells in the lining of the gut, known as the gut epithelium. Hyperthermia can alter the cell membranes' permeability (Roti Roti et al., 2008), or directly cause cells to die (Bynum et al., 1978). In either case, cells in the gut epithelium will leak endotoxins into the blood, a process known as endotoxemia. When these endotoxins circulate throughout the body, the immune system aggressively responds by activating cells to fight infection and inflammation, known as systemic inflammatory response syndrome (SIRS) (Leon and Helwig, 2010). The presence of endotoxins, as well as the body's aggressive immune response, can cause serious multi-organ damage (Epstein and Yanovich, 2019; Wang et al., 2022). In particular, the liver is usually one of the first organs to be damaged and is often what causes a heat stroke death (Wang et al., 2022).

Heat stroke is life-threatening and can severely impair workers' safety and health (Lucas et al., 2014). A study of work-related HRIs in Florida using hospital data reported that, during the warm seasons from May through October between 2005 through 2012, heat stroke was the primary diagnosis in 91% (21 of 23) of deaths. In total, they reported 160 cases of work-related heat stroke (Harduar Morano and Watkins, 2017). Analyses of heat stroke among military members indicate that roughly 73% of EHS patients require hospitalization for at least two days (Carter et al., 2007).

Heat stroke is a serious medical emergency that requires immediate rest, cooling, and usually hospitalization. Prognosis for heat stroke is highly dependent on how quickly heat stroke is recognized and how quickly an affected worker can be cooled. When an affected person can be diagnosed early and cooled rapidly, the prognosis is generally good. For example, rapid cooling within one hour of presentation of symptoms of CHS was found to reduce the mortality rate from 33% to 15% (Vicario et al., 1986). For EHS, cooling the body below 104 °F within 30 minutes of collapse is associated with very good outcomes (Casa et al., 2012; Casa et al., 2015). The authors also reported that they were unaware of any cases of fatalities among EHS victims where it was recorded that the body was cooled below 104 °F within 30 minutes of collapse (Casa et al., 2012).

Comparably, others have found that the risk of morbidity and mortality from heat stroke increases as treatment is delayed (Demartini et al., 2015; Schlader et al., 2022). Schlader et al. (2022) found that a delay in cooling can result in tissue damage, multi-organ dysfunction, and eventually death. Similarly, Zeller et al. (2011) found in their retrospective cohort study that patients who did not receive early or immediate cooling had worse outcomes, such as more severe forms of disease or death, although their study design does not allow for conclusions regarding causality (Zeller et al., 2011). Khogali and Weiner's (1980) case study report on 18 cases of heat stroke found that 72% of the patients took between 30-90 minutes to cool, whereas the other 28% were resistant to cooling, taking two to five hours to reach 38 °C (100.4 °F). This means that there is variation in how individuals respond to heat stroke treatment and that some individuals will respond quicker to treatment than others. Prompt treatment is likely even more critical for the individuals who take longer to cool.

Data from the general population also demonstrate the serious nature of heat stroke. One analysis of nationwide data estimated that nearly 55% of emergency department visits for heat stroke required hospitalization and roughly 3.5% of patients died in the emergency department or at the hospital (Wu et al., 2014). This study also found that heat stroke medical emergencies are more severe than other non-heat-related emergencies, with a 2.6-fold increase in admission rate and a 4.8-fold increase in case fatality compared to those other conditions (Wu et al., 2014).

Complete recovery for individuals who are affected by heat stroke may require time away from work. Some research suggests the length of recovery time and the need for time away from work is based on how long a person was at or above the critical core body temperature of 41 °C (105.8 °F), and how long it takes for biomarkers in blood to normalize (McDermott et al., 2007). Relevant biomarkers include those for acute liver dysfunction, myolysis (the breakdown of muscle tissue), and other organ system biomarkers (Ward et al., 2020; Schlader et al., 2022).

Guidelines for military personnel and athletes suggest that it may be weeks or months before a worker who has suffered heat stroke can safely return to work or perform the same level of work they did before suffering heat stroke. U.S. military members have clear return-to-work protocols post-heat stroke where members are assigned grades of functional capacity in six areas: physical capacity or stamina, upper extremities, lower extremities, hearing and ears, eyes, and psychiatric functioning (O'Connor et al., 2007). For example, when a soldier/airman experiences heat stroke, they automatically receive a reduced function capacity grade status in physical capacity. This also results in an automatic referral to a medical examination board. Soldiers and airmen are not cleared to return to duty until their laboratory results normalize, and even then, their status remains a trial of duty. If the individual has not exhibited any heat intolerance after three months, they are returned to a normal work schedule. However, maximal exertion and significant heat exposure remains prohibited for these individuals. If a military member experiences any heat intolerance during the period of restriction, or subsequent resumption to normal duty, a referral to the physical examination board for a hearing regarding their health status is required (O'Connor et al., 2007).

The U.S. Navy has its own set of guidelines, which does not distinguish between heat exhaustion and heat stroke, but uses laboratory tests, especially liver function tests, to determine when sailors are allowed to return to duty. For those who have suffered heat stroke, full return to duty is usually not granted until somewhere between two days to three weeks later (O'Connor et al., 2007).

In 2023, the American College of Sports Medicine (ACSM) published their consensus statement which provides evidence-based strategies to reduce and eliminate HRIs, including a return to activity protocol for athletes recovering from EHS (Roberts et al., 2023). Of note, ACSM names athletes (whether elite, recreational, or tactical) and occupational laborers as groups who are active and regularly perform exertional activities that could lead to EHS. Specifically, ACSM recommendations include refraining from exercise for at least seven days following release from the initial medical care for EHS treatment. Once all laboratory results and vital signs have normalized, ACSM recommends an individual can exercise in cool environments and gradually increase duration, intensity, and heat exposure over a two to four-week period to initiate environmental acclimatization (Roberts et al., 2023). If the affected athlete does not return to pre-EHS activity levels within four to six weeks, further medical evaluation is needed. ACSM recommends a full return to ( print page 70714) activity between two to four weeks after the individual has demonstrated exercise acclimatization and heat tolerance with no abnormal symptoms or test results during the re-acclimatization period (Roberts et al., 2023). Similarly, the National Athletic Trainer's Association proposes that individuals who experience EHS should complete a 7 to 21-day rest period, be asymptomatic, have normal blood-work values, and obtain a physician's clearance prior to beginning a gradual return to activity (Casa et al., 2015).

In the military setting it is accepted that returning to work too early and/or without adequate work restrictions can result in incomplete recovery from heat stroke, which may necessitate a prolonged restricted work status (McDermott et al., 2007). About 10-20% of people who have had heat stroke have been shown to experience heat intolerance roughly two months after having the heat stroke (Binkley et al., 2002). In some instances, this has lasted for five years and has increased the risk for another heat stroke (Binkley et al., 2002; McDermott et al., 2007). Similarly, a case study report of EHS cases amongst the U.S. Army found that in one of the ten cases examined, the person was heat intolerant for 11.5 months post-EHS (Armstrong et al., 1989).

Only a limited number of studies have focused on the long-term effects of heat stroke. This includes research by Wallace et al. (2007), whose retrospective review of military service members found that those who suffered an EHS event earlier in life were more likely to die due to cardiovascular disease and ischemic heart disease. Similarly, Wang et al. (2019) report that prior exertional heat illness was associated with a higher prevalence of acute ischemic stroke, acute myocardial infarction, and an almost three-fold higher prevalence of chronic kidney disease. Other research in mice support these claims and indicate that epigenetic effects post-EHS result in immunosuppression and an altered heat shock protein response as well as development of metabolic disorders that could negatively impact long-term cardiovascular health (Murray et al., 2020; Laitano et al., 2020).

OSHA's review of the scientific and medical literature indicates that occupational heat exposure can cause heat stroke, a medical emergency. The physiological mechanisms by which heat exposure can result in heat stroke are well-established in the literature, and heat exposure as a cause of heat stroke is well-recognized in the medical and scientific communities. The best available research demonstrates that heat stroke must be treated as soon as possible and that prolonged time between experiencing heat stroke and seeking treatment increases the likelihood of death and may result in long-term health effects.

NIOSH defines heat exhaustion as “[a] heat-related illness characterized by elevation of core body temperature above 38 °C (100.4 °F) and abnormal performance of one or more organ systems, without injury to the central nervous system” (NIOSH, 2016). Heat exhaustion can progress to heat stroke if not treated properly and promptly, and may require time away from work for a full recovery.

Signs and symptoms of heat exhaustion typically include profuse sweating, changes in mental status, dizziness, nausea, headache, irritability, weakness, decreased urine output and elevated core body temperature up to 40 °C (104 °F) (NIOSH, 2016; Kenny et al., 2018). Collapse may or may not occur. Significant injury to the central nervous system, and significant inflammatory response do not occur during heat exhaustion. However, there appears to be a fine line between heat exhaustion and heat stroke. Kenny et al. 2018 state that it can be difficult to clinically differentiate between heat exhaustion and early heat stroke. NIOSH also states that heat exhaustion “may signal impending heat stroke” (NIOSH, 2016). Armstrong et al. (2007) recommend that rectal temperature be taken to distinguish between heat exhaustion and heat stroke.

Heat exhaustion occurs when heat stress results in elevated body temperature between 98.6 °F and 104 °F (37 °C and 40 °C) and physiological changes occur (Kenny et al., 2018). Under these significant heat stress conditions, heavy sweating occurs, tissue perfusion is reduced, and inflammatory mediators are released. Electrolyte imbalances can occur due to fluid and electrolyte losses through sweating paired with inadequate replenishment. Voluntary and involuntary dehydration can exacerbate this process (Hendrie et al., 1997; Brake and Bates, 2003). “Voluntary dehydration,” as used by Brake and Bates, refers to the circumstance where a dehydrated worker does not adequately rehydrate, despite the availability of water. Upon review of several studies, Kenny et al. (2018) report that dehydration among workers is common, even when water is readily available. There is also evidence that even when water intake increases, as sweat rate and dehydration increase, intake may not be adequate to fully replace losses (Hendrie et al., 1997).

Brake and Bates (2003) summarized various hypothesized reasons for voluntary and involuntary dehydration. One hypothesized reason for voluntary dehydration is a delayed or decreased thirst response (Brake and Bates, 2003). Other reasons include mechanisms that affect fluid retention, such as the dependence of fluid retention on solutes such as sodium, which may be in imbalance under heat stress (Brake and Bates, 2003). Lack of adequate hydration could also be due to workplace pressures or concerns about sanitation (Rao, 2007; Iglesias-Rios, 2023).

The combination of heat stress, upright posture, and low vascular fluid volume (hypovolemia) can further dysregulate the circulatory system and affect clotting mechanisms (Kenny et al., 2018). Heat stress reduces blood flow to the abdominal organs, kidneys, muscles, and brain and increases blood flow to the skin to aid in cooling. These changes in the circulatory system and blood flow to the brain can potentially lead to dizziness or faintness upon standing (orthostatic intolerance), or collapse. Other factors that affect the development of heat exhaustion include individual health status, preparedness (such as acclimatization level), individual characteristics, knowledge, access to fluids, environmental factors, personal protective equipment use and work pacing and intensity (Kenny, 2018).

Heat exhaustion is one of the more common heat-related illnesses (Armstrong et al., 2007; Harduar Morano and Watkins, 2017; Lewandowski and Shaman, 2022). In their study of heat-illness hospitalizations in Florida during May to October from 2005-2012, Harduar Morano and Watkins (2017) reported that there were 2,659 cases of work-related heat exhaustion that resulted in emergency department visits or hospitalization, versus 181 cases of work-related heat stroke that resulted in emergency department visits, hospitalization, or death. Similar results have been reported in studies of heat-related illness among the United States Armed Forces and miners showing the frequency of heat exhaustion (Dickinson, 1994; Armed Forces Health Surveillance Division, 2022b; ( print page 70715) Lewandowski and Shaman, 2022; Donoghue et al., 2000; Donoghue, 2004). While in some studies heat exhaustion is not specifically diagnosed, several qualitative studies describe self-reported symptoms in workers that may be indicative of heat exhaustion ( e.g., Mirabelli et al., 2010; Fleischer et al., 2013; Kearney et al., 2016; Mutic et al., 2018). These symptoms included headache, nausea, vomiting, feeling faint, and heavy sweating.

Heat exhaustion may require treatment beyond basic first aid to prevent progression to heat stroke (Kenny et al., 2018). In cases where the degree of severity of heat illness is unclear, the individual should be treated as if they have heat stroke (Armstrong, 1989). For a worker experiencing heat exhaustion, NIOSH recommends the following steps to ensure the worker receives proper and adequate treatment: “Take worker to a clinic or emergency room for medical evaluation and treatment; If medical care is unavailable, call 911; Someone should stay with worker until help arrives; Remove worker from hot area and give liquids to drink; Remove unnecessary clothing, including shoes and socks; Cool the worker with cold compresses or have the worker wash head, face, and neck with cold water; Encourage frequent sips of cool water” (NIOSH, 2016).

Complete recovery from heat exhaustion may require a restricted work status (or limited work duties). Donoghue et al. (2000) reported that following heat exhaustion, 29% (22 of 77) of miners included in the study required a restricted work status for at least one shift. The military has specific protocols for return to duty following heat exhaustion. For example, the U.S. Army and Air Force follow the protocol outlines in AR 40-501 (O'Connor et al., 2007). Three instances of heat exhaustion in less than 24 months can result in referral to a Medical Evaluation Board before a full return to service. Some military units have additional or more specific guidelines. For example, one military unit, at Womack Army Medical Center in North Carolina, has guidelines that allow individuals who are considered to have mild illness, fully recovered in the emergency room, and have no abnormal laboratory findings to return to light duty the following day and limited duty the day after that. However, they also indicate that some effects of heat illness may be subtle or delayed and recommend individuals avoid strenuous exercise for several days and remain under observation (O'Connor et al., 2007).

The scientific and medical literature presented here clearly demonstrate that heat exhaustion is a recognized health effect of occupational heat exposure. The best available evidence on the symptoms, treatment, and recovery of heat exhaustion demonstrates that heat exhaustion can progress to heat stroke, a medical emergency, if not treated promptly and that heat exhaustion may require time away from work for a full recovery.

Occupational heat exposure can result in heat syncope. Syncope is the medical term for “fainting,” and heat syncope is defined as “fainting, dizziness, or light-headedness after standing or suddenly rising from a sitting/lying position” due to heat exposure (NIOSH, 2023a). Heat syncope may sometimes be referred to as “exercise-associated collapse” (EAC), but heat syncope can happen without significant levels of exertion (Asplund et al., 2011; Pearson et al., 2014). As explained below, heat syncope is an acknowledged and documented health effect of occupational heat exposure.

There are two mechanisms for how heat exposure can cause heat syncope (Schlader et al., 2016; Jimenez et al., 1999). One mechanism for heat syncope is reduced blood flow to the brain. Elevated core temperature induces vasodilation, sweating, and may result in blood pooling in certain areas of the body (see Section IV.B., General Mechanisms of Heat-Related Health Effects). Thus, there is a lower circulating blood volume, which can reduce blood flow to the brain and cause loss of consciousness (Wilson et al., 2006; Van Lieshout et al., 2003).

A second mechanism for heat syncope is reduced cerebral blood velocity (CBV) (indicative of reduced blood flow to the brain) that results in orthostatic intolerance (the inability to remain upright without symptoms) during a heat stress episode (Wilson et al., 2006). As individuals experience whole body heating, CBV is reduced and cerebral vascular resistance (the ratio of carbon dioxide stimulus to cerebral blood flow) increases. These changes ultimately contribute to reduced cerebral perfusion and blood flow, as well as orthostatic intolerance (Wilson et al., 2006). The orthostatic response to heat stress during “rest” ( i.e., standing/sitting) is essentially equivalent to the orthostatic response to heat stress after exercise if skin temperature is similarly elevated (Pearson et al., 2014). While core temperature is not always elevated in cases of heat syncope, skin temperature typically is (Department of the Army, 2022; Noakes et al., 2008).

Differentiating between heat syncope, heat exhaustion, and heat stroke is a critical step in proper diagnosis (Santelli et al., 2014; Coris et al., 2004). As stated above, heat syncope always involves loss of consciousness, but it does not require elevated core body temperature (Santelli et al., 2014; Holtzhausen et al., 1994). Conversely, heat exhaustion and stroke do not require loss of consciousness. Though central nervous system (CNS) disturbances are possible in heat stroke and heat stroke is always characterized by significantly elevated core temperature. Further, recovery of mental status is faster in heat syncope than in exhaustion and heat stroke, since cooling may not be required for treatment of heat syncope (Howe and Boden, 2007).

Workers have experienced heat syncope when exposed to heat. A survey-based study in southern Georgia found that 4% of 405 farmworkers experienced fainting within the previous week (Fleischer et al., 2013). Another survey-based study in North Carolina asked 281 farmworkers if they had ever experienced heat-related illness and found that 3% of workers had fainted (Mirabelli et al., 2010). While these cases were not formally diagnosed as heat syncope, Fleischer reported temperatures ranging from 34-40 °C (94-104 °F) and a heat index of 37-42 °C (100-108 °F) at the time workers fainted, and Mirabelli described the working conditions at the time of fainting as being in “extreme heat.”

NIOSH recommends treating heat syncope by having the worker sit down in a cool environment and hydrate with either water, juice, or a sports drink (NIOSH, 2016). The Department of the Army recommends that “victims of heat/parade syncope will recover rapidly once they sit or lay supine, though complete recovery of stable blood pressure and heart rate (resolution of orthostasis or ability to stand without fainting) in some individuals may take 1 to 2 hours” (Department of the Army, 2022). Treatment recommendations for athletes consist of moving the athlete to a cool area and laying them supine with elevated legs to assist in venous return, ( print page 70716) possibly with oral or intravenous rehydration (Peterkin et al., 2016; Howe and Boden, 2007; Seto et al., 2005; Lugo-Amador et al., 2004).

An episode of heat syncope may require time away from work for a thorough evaluation to ascertain one's risk for recurrent/future episodes of heat syncope. No studies have evaluated recurring episodes of syncope among workers specifically, but a study found that, for the general population, 1-year syncope recurrence (any type) was 14% in working-age people (18-65 years) (Barbic et al., 2019). The U.S. Army has a requirement to “obtain a complete history to rule out other causes of syncope, including an exertional heat illness or other medical diagnosis (for example, cardiac disorder)” (Department of the Army, 2022). Recommendations for athletes include thorough evaluation “for injury resulting from a fall, and all cardiac, neurologic, or other potentially serious causes for syncope” (Howe and Boden, 2007; Lugo-Amador et al., 2004; Binkley et al., 2002). Indeed, if an injury ( e.g., fall, collision) is sustained because of heat syncope, treatment beyond first aid (including hospitalization) may be necessary. Supporting this point, more general syncope has been linked to occupational accidents requiring hospitalizations (Nume et al., 2017).

The scientific and medical literature presented in this section demonstrate that heat syncope is a recognized health effect of occupational heat exposure. Studies suggest that heat syncope may require time away from work for further evaluation. Additionally, heat syncope can lead to injuries ( e.g., injury from a fall), some of which may require hospitalization.

Rhabdomyolysis is a life-threatening illness that can affect workers exposed to occupational heat. NIOSH defines rhabdomyolysis as “a medical condition associated with heat stress and prolonged physical exertion, resulting in the rapid breakdown of muscle and the rupture and necrosis of the affected muscles” (NIOSH, 2016). This definition is specific to exertional rhabdomyolysis. Another form of rhabdomyolysis, called traumatic rhabdomyolysis, is caused by direct muscle trauma ( e.g., from a fall or crush injury). Workers can experience such injuries, and consequently suffer from traumatic rhabdomyolysis, because of occupational heat exposure (see Section IV.P., Heat-Related Injuries). However, this section will focus only on exertional rhabdomyolysis. Unless otherwise specified, all references to rhabdomyolysis are shorthand for exertional rhabdomyolysis.

Signs and symptoms of rhabdomyolysis include myalgia (muscle pain), muscle weakness, muscle tenderness, muscle swelling, and/or dark-colored urine (Armed Forces Health Surveillance Division, 2023b; Dantas et al., 2022; O'Connor et al., 2008; Cervellin et al., 2010). Notably, the onset of these symptoms may be delayed by 24-72 hours (Kim et al., 2016). Rhabdomyolysis commonly affects individuals who are exposed to heat during physical exertion. For example, the Centers for Disease Control and Prevention (CDC) investigated an incident in which an entire cohort of 50 police trainees were diagnosed with rhabdomyolysis after the first 3 days of a 14-week training program; the trainees had engaged in heavy physical exertion outdoors with limited access to water. The CDC concluded that adequate hydration is particularly important when the HI approaches 80 °F (Goodman et al., 1990).

Rhabdomyolysis has long been recognized as a heat-related illness by NIOSH, the U.S. Armed Forces, and national athletic organizations such as the American College of Sports Medicine (Armstrong et al., 2007). Specifically, NIOSH lists rhabdomyolysis as an “acute heat disorder” in its Criteria for a Recommended Standard (2016) and provides detailed recommendations for recognition and treatment of rhabdomyolysis. NIOSH also conducted case studies and retrospective analyses to identify cases of rhabdomyolysis among workers exposed to heat, including firefighter cadets and instructors, as well as park rangers (Eisenberg et al., 2019; Eisenberg J et al., 2015; Eisenberg and Methner, 2014).

Similarly, the U.S. Armed Forces developed a case definition that specifies rhabdomyolysis can be heat-related (Armed Forces Health Surveillance Board, 2017), and this definition is applied in their annual surveillance reports of HRIs. From 2018 to 2022, most rhabdomyolysis cases (75.9%) occurred during warmer months ( i.e., May to October) (Armed Forces Health Surveillance Division, 2023b). In a retrospective study of hospital admissions for rhabdomyolysis in military members (2010-2013), 60.1% (193 out of 321) cases were deemed to be associated with exertion and exposure to heat (Oh et al., 2022).

Many studies have also found that rhabdomyolysis often coincides with exertional heat stroke and other HRIs such as heat exhaustion, heat cramps, hyponatremia, and dehydration. The frequent co-occurrence of rhabdomyolysis and other HRIs has been reported among workers, including police and firefighters (Eisenberg et al., 2019; Goodman et al., 1990), workers included in OSHA enforcement investigations (Tustin et al., 2018a), military members (Oh et al., 2022; Carter et al., 2005), athletes (Thompson et al., 2018), and in the general population (Thongprayoon et al., 2020).

Studies have identified two interrelated mechanisms through which heat exposure, combined with exertion, can cause rhabdomyolysis. Both mechanisms share a common origin: occupational heat exposure and exertion both contribute to excessive heat stress, which in turn causes an elevated core temperature. Both mechanisms also share a common outcome: the breakdown and death of muscle tissue, which is the hallmark characteristic of rhabdomyolysis. The first mechanism is thermal injury to muscle cells. When the body's core temperature is elevated, it creates a toxic environment that can directly injure or kill muscle cells. The temperature at which this occurs, known as the thermal maximum, is estimated to be about 107.6 °F (42 °C) (Bynum et al., 1978). At the thermal maximum, the structural components of the cells' membranes are liquified and the membrane breaks down. Proteins in the cells' mitochondria, which are key to energy production, change shape and no longer function properly. Calcium, which is normally maintained at a low level inside muscle cells, will rush into the cells and activate inflammatory processes that accelerate the death of those cells (Torres et al., 2015; Khan, 2009).

The second mechanism is lack of oxygen to muscle cells. When the body attempts to cool itself, it can lose high volumes of sweat. Sweat loss can deplete the body's stores of water and electrolytes, leading to low blood volume (see Section IV.B., General Mechanisms of Heat-Related Health Effects). Low blood volume, and low potassium in the blood (known as hypokalemia), can both contribute to muscle cell death. An adequate supply of blood is necessary to deliver oxygen to muscles, and an adequate supply of potassium is needed to support vasodilation (to support increased blood flow to the muscles during exertion). When neither blood volume nor ( print page 70717) potassium are sufficient, the muscle cells do not receive enough oxygen (known as ischemia). When this occurs, the muscle cells produce less energy and eventually will die if exertion continues (Knochel and Schlein, 1972).

While OSHA is not aware of surveillance data on the incidence of rhabdomyolysis in the worker population in the United States, there are surveillance data on the incidence of rhabdomyolysis among active military members in the Army, Navy, Air Force, and Marine Corps. These data have been reported for the U.S. Army from 2004 to 2006 (Hill et al., 2012) and for all military branches from 2008 through 2022 (Armed Forces Health Surveillance Division, 2023b; Armed Forces Health Surveillance Division, 2018; U.S. Armed Forces, 2013). These surveillance data and the studies described above by NIOSH and others indicate that workers performing strenuous tasks in the heat are at risk of developing rhabdomyolysis. The U.S. Armed Forces has successfully identified many cases of heat-related rhabdomyolysis by searching medical records for the presence of either the ICD-10 code for rhabdomyolysis and/or the ICD-10 code for myoglobinuria, along with any other heat-related codes (table IV-1) (Armed Forces Health Surveillance Division, 2023b; Oh et al., 2022).

Rhabdomyolysis is a serious heat-related illness that can cause life-threatening complications. Many cases of rhabdomyolysis may require hospitalization. For example, A CDC investigation into a police training program in Massachusetts found that 26% of police trainees (13 out of 50) were hospitalized for rhabdomyolysis only three days into their training (Goodman et al., 1990). The mean length of hospitalization was 6 days, with a range of 1 to 20 days (Goodman et al., 1990). Similarly, a military surveillance study identified 473 rhabdomyolysis cases among military members in 2022, with 35.3% of cases (167 out of 473) requiring hospitalization (Armed Forces Health Surveillance Division, 2023b). In a retrospective study of 193 military trainees hospitalized for rhabdomyolysis, the mean length of hospitalization was 2.6 days, with a range of 0 to 25 days (Oh et al., 2022).

The focus of treatment for rhabdomyolysis during hospitalization is to reduce levels of creatine kinase (CK) and myoglobin in the blood, as well as correct electrolyte imbalances, through aggressive administration of intravenous fluids (generally normal saline) (O'Connor et al., 2020; Luetmer et al., 2020; Manspeaker et al., 2016; Torres et al., 2015). Monitoring is used to repeatedly measure CK levels until a peak concentration is reached (often within 1-3 days), and then to ensure that CK levels are consistently trending downwards before discharge from the hospital (Kodadek et al., 2022; Oh et al., 2022).

Complications of rhabdomyolysis are also possible. When muscle cells die, they release several electrolytes and proteins into the bloodstream that can cause severe health complications. For example, the release of potassium from muscle cells can cause hyperkalemia (high level of potassium in the blood), which then leads to heart arrhythmias (abnormal heart rhythms) (Mora et al., 2017; Sauret et al., 2002). Also, the release of myoglobin into the bloodstream can be toxic for the kidneys. When blood is filtered by nephrons (functional units of the kidneys) to produce urine, the presence of even small amounts of myoglobin can obstruct and damage the nephrons (Mora et al., 2017; Sauret et al., 2002). In some cases, these complications from rhabdomyolysis can be life-threatening (Wesdock and Donoghue, 2019) and in fact fatalities have been reported (Gardner and Kark, 1994; Goodman et al., 1990). A more detailed discussion of how rhabdomyolysis can cause acute kidney injury or other kidney damage can be found in Section IV.M., Kidney Health Effects.

Guidelines for return to work among workers diagnosed with rhabdomyolysis are limited. In the U.S. military, soldiers deemed to be at low risk for recurrence of rhabdomyolysis are restricted to light, indoor duty and encouraged to rehydrate for at least 72 hours to allow for normalization of CK levels. If CK levels do not normalize, they must continue indoor, light duty; if CK levels do normalize, they can proceed to light, outdoor duty for at least 1 week and must show no return of clinical symptoms before they can gradually return to full duty. In contrast, soldiers deemed to be at high risk for recurrence of rhabdomyolysis must undergo additional diagnostic tests, with consultation from experts, and can be given an individualized, restricted exercise program while they await clearance for full return to duty (O'Connor et al., 2020; O'Connor et al., 2008). These guidelines have been adopted by the Armed Forces and restated in their surveillance reports of rhabdomyolysis (Armed Forces Health Surveillance Division, 2023b).

The available scientific literature indicates that rhabdomyolysis can result from physical exertion in the heat. Based on plausible mechanistic data, studies by NIOSH and others, and surveillance data indicating incidence of rhabdomyolysis among active military members, OSHA preliminarily determines that workers performing strenuous tasks in the heat are at risk of rhabdomyolysis.

Workers in hot environments may experience hyponatremia, a condition that occurs when the level of sodium in the blood falls below normal levels (<135 milliequivalents per liter (mEq/L)) (NIOSH, 2016). Hyponatremia is often caused by drinking too much water or hypotonic fluids, such as sports drinks, over a prolonged period of time. Without sodium replacement, the high water intake can result in losses of sodium in the blood as more sodium is lost due to increased sweating from heat exposure and urination (Korey Stringer Institute (KSI), n.d.). Mild forms of hyponatremia may not produce any signs or symptoms, or may present with symptoms including muscle weakness and/or twitching, dizziness, lightheadedness, headache, nausea and/or vomiting, weight gain, and swelling of the hands or feet (KSI, n.d.; NIOSH, 2016). In severe cases, hyponatremia may cause altered mental status, seizures, cerebral edema, pulmonary edema, and coma, which may be fatal (KSI, n.d.; NIOSH, 2016; Rosner and Kirven, 2007). NIOSH and the U.S. Army classify hyponatremia as a heat-related illness (NIOSH, 2016; Department of the Army, 2022).

When exposed to heat, the autonomic nervous system triggers the body's sweat response, in which sweat glands release water to wet the skin (Roddie et al., 1957; Grant and Holling, 1938). The purpose of the sweat response is to cool the body. However, in doing so, it can deplete the body's stores of water and electrolytes ( e.g., sodium, potassium, chloride, calcium, and magnesium) that are essential for normal bodily function (Shirreffs and Maughan, 1997). As the body's store of sodium is lessening and high quantities of water are consumed, hyponatremia may develop as sodium in the blood becomes diluted (<135 mEq/L). In some cases, this dilution may cause an osmotic disequilibrium—an imbalance in the amount of sodium inside and outside the cell resulting in ( print page 70718) cellular swelling—which can lead to the serious and fatal health outcomes discussed above.

Surveillance of hyponatremia among workers is limited. However, a recent case study demonstrates the potential severity and life-threatening nature of hyponatremia. After a seven-day planned absence from work, a 34-year-old male process control operator in an aluminum smelter pot room was hospitalized due to a variety of HRI symptoms including hyponatremia, with serum (the liquid portion of blood collected without clotting factors) sodium level of 114 millimoles per liter (mmol/L) (reference range: 136-145 mmol/L) (Wesdock and Donoghue, 2019). After 13 days in the hospital, the patient was discharged with a diagnosis of “severe hyponatremia likely triggered by heat exposure” (Wesdock and Donoghue, 2019). The patient was still out of work 32 weeks after the incident. While no temperature data for the pot room were available, an exposure assessment used outdoor temperatures that day and pot room temperatures from the literature to estimate that the WBGT could have been as high as 33 °C, which the authors state exceeds the ACGIH TLV for light work for acclimatized workers (Wesdock and Donoghue, 2019).

The relationship of heat exposure and hyponatremia was examined among male dockyard workers in Dubai, United Arab Emirates (Holmes et al., 2011). This population performed long periods of manual work in the heat and consumed a diet low in sodium. A first round of plasma ( i.e., the liquid part of blood collected that contains water, nutrients and clotting factors) samples were taken at the end of the summer (n=44), with a second round taken at the end of the winter among volunteers still willing to participate (n=38). In the summer, 55% of participants were found to be hyponatremic (<135 millimolar (mM)), whereas only 8% were hyponatremic in the winter. Although ambient temperature conditions were not reported, the authors indicate that hyponatremia was highest during the summer because of sodium losses through sweat and inadequate sodium replacement (Holmes et al., 2011).

Hyponatremia among the military population has been well documented by the Annual Armed Forces Health Surveillance Division, which releases annual reports on exertional hyponatremia among active duty component services members, each with surveillance data for the previous 15 years ( e.g., Armed Forces Health Surveillance Division, 2023a; Armed Forces Health Surveillance Division, 2022a; Armed Forces Health Surveillance Division, 2021; Armed Forces Health Surveillance Division, 2020). Cases come from the Defense Medical Surveillance System and include both ambulatory medical visits and hospitalizations in both military and civilian facilities. During the period of 2004 through 2022, the number of cases of hyponatremia among U.S. Armed Forces peaked in 2010 with 180 cases. The lowest number during that time period was 2013, when 72 cases were reported. During the last 15 years in which data were reported (2007-2022), 1,690 cases of hyponatremia occurred. Of these 1,690 cases, 86.8% (1,467) were diagnosed and treated during an ambulatory care visit (Armed Forces Health Surveillance Division, 2023a). As the diagnostic code for hyponatremia may include cases that are not heat-related, these data may be overestimates. However, such overestimation is reduced in this study as the authors controlled for many other related diagnoses ( e.g., kidney diseases, endocrine disorders, alcohol/illicit drug abuse), which can cause hyponatremia.

Treatment and recovery for hyponatremia can vary depending on severity and symptoms. Workers presenting with mild symptoms should increase salt intake by consuming salty foods or oral hypertonic saline and restrict fluid until symptoms resolve or sodium levels return to within normal limits (KSI, n.d.). Medical attention may be required in severe cases, which may be life-threating, and may be sought to address symptoms and personal risk factors ( e.g., history of heart conditions, on a low sodium diet) (NIOSH, 2016).

The available evidence in the scientific literature indicates that hyponatremia can result from occupational heat exposure. The evidence on treatment and recovery demonstrates that hyponatremia can require medical attention and, in some cases, may be life-threatening.

Workers exposed to environmental or radiant heat can experience sudden muscle cramps known as “heat cramps.” NIOSH defines heat cramps as “a heat-related illness characterized by spastic contractions of the voluntary muscles (mainly arms, hands, legs, and feet), usually associated with restricted salt intake and profuse sweating without significant body dehydration” (NIOSH, 2016). Someone can experience heat cramps even if they are frequently hydrating with water, but they are not replenishing electrolytes. Heat cramps are recognized as a “heat-related illness” by numerous organizations, including NIOSH, U.S. Army, U.S. Navy, National Athletic Trainers' Association (NATA), American College of Sports Medicine (ACSM), and World Medicine (formerly known as IAAF).

It is recognized in the medical and scientific communities that heat cramps result from heat exposure. However, the exact physiological mechanism is not known. In an early study of heat cramps, investigators included the following as the diagnostic criteria for heat cramps: exposure to high temperatures at work; painful muscle cramps; rapid loss of salt in the sweat that is not replaced (which may cause hyponatremia); diminished concentration of chloride in the blood and in the body tissues (also known as hypochloremia); and rapid amelioration of symptoms after appropriate treatment (Talbott and Michelsen, 1933).

The following mechanism has been proposed for the development of heat cramps: profuse sweating can deplete electrolyte stores ( e.g., sodium (Na), potassium (K), calcium (Ca)), which exacerbates muscle fatigue and can cause heat cramps (Bergeron, 2003; Horswill et al., 2009; Schallig et al., 2017; Derrick, 1934). The U.S. Army further posits that “intracellular calcium is increased via a reduction in the sodium concentration gradient across the cell membrane. The increased intracellular calcium accumulation then stimulates actin-myosin interactions (that is, filaments propelling muscle filaments) causing the muscle contractions” (Department of the Army, 2022). Heat cramps are sometimes referred to, more broadly, as exercise-associated muscle cramps (EAMCs) (Bergeron et al., 2008). However, heat cramps are distinct in that they only occur in hot conditions, which exacerbate electrolyte depletion, and may or may not be associated with exercise.

Surveillance data and survey study data demonstrate that workers exposed to environmental or radiant heat frequently experience heat cramps in the United States. In a study of heat-related illness hospitalizations and deaths for the U.S. Army from 1980- ( print page 70719) 2002, 8% of heat-related illness hospitalizations recorded were due to heat cramps (Carter et al., 2005). Similarly, in studies of self-reported heat-related illness, workers frequently cite heat cramps as a common symptom of heat exposure. Specifically, in several studies of self-reported heat-related symptoms among farmworkers in multiple States, participants reported experiencing sudden muscle cramps in the prior week in Georgia (33.7% of 405 respondents) (Fleischer et al., 2013), North Carolina (35.7% of 158 respondents) (Kearney et al., 2016), and Florida (30% of 198 respondents) (Mutic et al., 2018). In another study of self-reported symptoms among 60 migrant farmworkers in Georgia, heat-related muscle cramps were reported by 25% of participants, the second most frequently reported HRI symptom (Smith et al., 2021). In a study examining exertional heat illness and corresponding wet bulb globe temperatures in football players at five southeastern U.S. colleges from August to October 2003, the authors found that the highest incidences of exertional heat illness (EHI) occurred in August (88%, EHI rate= 8.95/1000 athlete-exposures (Aes)) and consisted of 70% heat cramps (6.13/1000 Aes) (Cooper et al., 2016).

Treatment for heat cramps includes electrolyte-containing fluid replacement (also known as isotonic fluid replacement), stretching, and massage (Gauer and Meyers, 2019; Peterkin et al., 2016). In some cases, sodium replacement may be a treatment for heat cramps (Talbott and Michelsen, 1933; Sandor, 1997; Jansen et al., 2002). In severe cases, it is recommended that magnesium levels of the patient are obtained and if necessary, magnesium replacement through IV therapy is provided (O'Brien et al., 2012). The ACSM recommends rest, prolonged stretching in targeted muscle groups, oral sodium chloride ingestion in fluids or foods, or intravenous normal saline fluids in severe cases (ACSM, 2007). NIOSH recommends that medical attention is needed if the worker has heart problems, is on a low sodium diet, or if cramps do not subside within 1 hour (NIOSH, 2016). If treated early and effectively, individuals may return to activity after heat cramps have subsided (Bergeron, 2007; Savioli et al., 2022; Gauer and Meyers, 2019). However, severe heat cramps may require an emergency department visit or hospitalization (Harduar Morano and Waller, 2017; Carter et al., 2005). While most cases of heat cramps do not require restricted work status or time away from work, guidelines for military personnel suggest some cases may require light workload the next day and limited workload the following day, with observation of the affected patient because some additional deficits may be delayed or subtle (O'Connor et al., 2007). In addition, guidelines for military personnel advise that strenuous exercise be avoided for several days in some cases of heat cramps (O'Connor et al., 2007). Severe heat cramps may also elicit soreness for several days which can lead to a longer recovery period (Casa et al., 2015).

OSHA's review of the scientific and medical literature indicates that heat cramps are a recognized health effect of occupational heat exposure. Indeed, several studies of self-reported symptoms of HRI among farmworkers in multiple States have indicated that heat cramps are quite common. The best available evidence on treatment and recovery indicates that heat cramps can, in some cases, require medical attention and may require time away from work or an adjusted workload.

Workers in hot environments may experience heat rash. Heat rash is defined by NIOSH as “a skin irritation caused by excessive sweating during hot, humid weather” (NIOSH, 2022). NIOSH, the U.S. Army, and the U.S. Navy classify heat rash as a heat-related illness (NIOSH, 2016; Department of the Army, 2022; Department of the Navy, 2023). Also known as miliaria rubra or prickly heat, workers with heat rash develop red clusters of pimples or small blisters, which can produce itchy or prickly sensations that become more irritating as sweating persists in the affected area. Heat rash can last for several days and tends to form in areas where clothing is restrictive and rubs against the skin, most commonly on the neck, upper chest, groin, under the breasts, and in elbow creases (OSHA, 2011; NIOSH, 2022; OSHA, 2024a). If left untreated, heat rash can become infected, and more severe cases can lead to high fevers and heat exhaustion (Wenzel and Horn, 1998). In some cases, heat rash can lead to hypohidrosis ( i.e., the reduced ability to sweat) in the affected area, even weeks after the heat rash is no longer visible, which impairs thermoregulation and can cause predisposition for heat stress (Sulzberger and Griffin, 1969; Pandolf et al., 1980; DiBeneditto and Worobec, 1985). This can impair an employee's ability to work and prevent resumption of normal work activities in hot environments to allow for the area to heal, which in some cases can take 3-4 weeks for heat intolerance to subside (Pandolf et al., 1980).

The development of heat rash has been studied for centuries (Renbourn, 1958). While working in hot environments with a high relative humidity, the body's ability to cool itself is greatly reduced, as sweat is less likely to evaporate from the skin (Sulzberger and Griffin, 1969; DiBeneditto and Worobec, 1985). Heat rash occurs when sweat remains on the skin and causes a blockage of sweat (eccrine) glands and ducts (Wenzel and Horn, 1998). Since the sweat ducts are blocked, sweat secretions can leak and accumulate beneath the skin, causing an inflammatory response and resulting in clusters of red bumps or pimples (Dibeneditto and Worobec, 1985). If left untreated, heat rash may become infected (Holzle and Kligman, 1978). Depending on the level of blockage, this can manifest as various types of miliaria, with miliaria rubra being the most common form of heat rash (Wenzel and Horn, 1998).

Surveillance of heat rash in worker populations is limited. However, farmworkers have reported cases of skin rash or skin bumps while working in summer months (Bethel and Harger, 2014; Kearney et al., 2016; Luque et al., 2020). From these studies, the percentage of participants surveyed or interviewed that report experiencing skin rash or skin bumps in the previous week were 10% (n=100, Beth and Harger, 2014), 12.1% (n=158, Kearney et al., 2016) and 5% (n=101, Luque et al., 2020). Although these studies do not purport a diagnosis, presentation of skin rash or skin bumps while working in hot environments with reported average high temperatures ranging to the mid-90s °F indicates respondents may have developed heat rash.

Similar findings with diagnosis of heat rash or related symptoms have been recorded outside of the U.S. among workers in the following professions: 17% of indoor electronics store employees in air-conditioned (4%) and non-air-conditioned (13%) areas in Singapore (n=52, Koh, 1995); 2% of underground miners at a site in Australia (n=1,252, Donoghue and Sinclair, 2000); 34% of maize farmers in Nigeria (n=396, Sadiq et al., 2019); 68% of sugarcane cutters and 23% of ( print page 70720) sugarcane factory workers in Thailand (n=183, Boonruksa et al., 2020); 41% of sugarcane farmers in Thailand (n=200, Kiatkitroj et al., 2021); 17% of autorickshaw drivers (n=78), 23% of outdoor street vendors (n=75), 16% of street sweepers (n=75) in India (n=228, Barthwal et al., 2022); and 13% of underground and open pit miners across Australia (n=515, Taggart et al., 2024). Although these studies illustrate the prevalence of heat rash in various worker populations, OSHA notes that differences in study methodologies and the populations studied mean that the results of these studies are not necessarily directly comparable to each other or to similar industries or worker populations in the United States.

The type of clothing worn may also contribute to formation of heat rash while working in higher temperatures. Heat rash was formally diagnosed among U.S. military personnel wearing flame resistant army combat uniforms in hot and arid environments (102.2 °F to 122 °F (39 °C to 50 °C), 5% to 25% relative humidity) (Carter et al., 2011). In this case series, 18 patients with heat rash presented with moderate to severe skin irritation, which was worsened by reactions to chemical additives not removed from the laundering process and increased heat retention from sweat-soaked clothing, as well as the friction from the fabric and the occlusive effect of the clothing, which allowed sweat to accumulate on the skin despite the lower humidity (Carter et al., 2011). This study calls attention to the effect of clothing on the development of heat rash and factors that may influence its severity.

Although most cases of heat rash can be self-treated without seeking medical attention, symptoms typically last for several days (Wenzel and Horn, 1998). It is important that heat rash is kept dry and cool to avoid possible infection. Workers experiencing heat rash should move to a cooler and less humid work environment and avoid tight-fitting clothing, when possible (NIOSH, 2022). The affected area should be kept dry, and ointments and creams, especially if oil-based, should not be used (NIOSH, 2022). However, powder may be used for relief.

The available evidence in the scientific literature indicates that heat rash can result from occupational heat exposure. Although heat rash usually resolves on its own without medical attention, symptoms often persist for several days and more severe cases can impair an employee's ability to work and lead to infection if left untreated.

Workers in hot environments may experience heat edema. Heat edema is the swelling of soft tissues, typically in the lower extremities (feet, ankles, and legs) and hands, and may be accompanied by facial flushing (Gauer and Meyers, 2019). Surveillance systems and the U.S. Army classify heat edema as a heat-related illness (Department of the Army, 2022). Workers who are sitting or standing for prolonged periods may be at higher risk for heat edema (Barrow and Clark, 1998). Workers who are not fully acclimatized to the work site may be more prone to developing heat edema as the body adjusts to hotter temperatures (Howe and Boden, 2007).

When exposed to heat, the body increases blood flow and induces vasodilation to cool itself and thermoregulate. This means, as blood is shunted towards the skin and vasodilation begins, the blood vessels near the skin's surface become wider (Hough and Ballantyne, 1899; Kamijo et al., 2005). However, blood can pool in areas of the body that are most subject to gravity ( e.g., legs), and fluid can seep from blood vessels causing noticeable swelling under the skin—this is known as heat edema (Gauer and Meyers, 2019).

Surveillance of heat edema is limited. Many studies include heat edema as one of many HRIs that contributed to an aggregate measure of HRI in worker, military, or general populations, but very few were found to quantify heat edema alone.

Multiple studies outside of the U.S. have examined HRIs among farm and factory workers in the sugarcane industry through surveys and interviews (Crowe et al., 2015; Boonruksa et al., 2020; Kiatkitroj et al., 2021; Debela et al., 2023). Respondents in the studies were asked if they experienced swelling of the feet or hands (with varying degrees of frequency) during periods of heat exposure, which could indicate presentation of heat edema. In different samples of sugarcane workers in two provinces of Thailand, two studies found incidence of swelling of the hands and feet. Among sugarcane cutters, 16.7% self-reported ever experiencing swelling of the hands or feet and 5.6% self-reported experiencing these symptoms (mean 30.6 °C WBGT) (n=90, Boonruksa et al., 2020). In another province, 10.5% self-reported swelling of the hands/feet while working one summer (n=200, Kiatkitroj et al., 2021).

While comparing HRI symptoms among sugarcane harvesters and non-harvesters in Costa Rica, 15.1% of harvesters (n=106) and 7.9% of non-harvesters (n=63) self-reported having ever experienced swelling of hands/feet (p=0.173) (n=169, Crowe et al., 2015). While 7.5% of harvesters, who worked outdoors in the field, self-reported experiencing this symptom at least once per week, no non-harvesters self-reported swelling with this level of frequency (p=0.026) (Crowe et al., 2015). The sample of non-harvesters included both workers that were intermediately exposed to heat ( e.g., in the processing plant or machinery shop) and workers not exposed to heat ( e.g., in offices).

In a sample of sugarcane factory workers (n=1,524) in Ethiopia, 72.4% (1,104) were considered exposed to heat defined as conditions exceeding the ACGIH's TLV (Debela et al., 2023). Of the total sample (including workers considered exposed to heat and not), 78% (1,189) self-reported having experienced swelling of hands and feet at least once per week, which was the most commonly reported HRI symptom (Debela et al., 2023). Although these studies do not purport a diagnosis, presentation of swelling of the hands and feet while working in hot environments suggests respondents may have developed heat edema.

Although most cases of heat edema can be self-treated without seeking medical attention, symptoms can last for days and reoccurrence is less likely if individuals are properly acclimatized (Howe and Boden, 2007; Department of the Army, 2023). It is important that the affected individual moves out of the heat and elevates the swollen area. Diuretics are not typically recommended for treatment (Howe and Boden, 2007; Gauer and Meyers, 2019; CDC, 2024a).

The available evidence in the scientific literature indicates that heat edema can result from occupational heat exposure, causing swelling of the lower extremities (feet, ankles, and legs) and hands. It may be difficult to move swollen body parts, thereby impeding an employee's ability to perform their job. The need for medical attention can typically be avoided if the condition is properly treated. ( print page 70721)

The kidneys perform many functions in the body, including filtering toxins out of the blood and balancing the body's water and electrolyte levels (NIDDK, 2018). Working in the heat places a lot of demand on the kidneys to conserve water and regulate electrolytes, like sodium, lost through sweat. A growing body of experimental and observational literature suggests that intense heat strain can cause damage to the kidneys in the form of acute kidney injury (AKI), even independent of conditions like heat stroke and rhabdomyolysis. An epidemic of chronic kidney disease in Central America and other regions around the world has placed additional attention on the potential of recurrent heat stress-related AKI to cause chronic kidney disease (CKD) over time (Johnson et al., 2019; Schlader et al., 2019). Working in the heat has also been associated with the development of kidney stones among workers outside the U.S., likely a result of decreased urine volume leading to increased concentration of minerals in the urine that crystallize into stones.

Each kidney is comprised of hundreds of thousands of functional units called nephrons. Each nephron has multiple parts, including the glomerulus (a cluster of blood vessels that conduct the initial filtering of large molecules) and the tubules (tubes that reabsorb needed water and minerals and secrete waste products). The fluid that remains after traveling through the glomeruli and tubules becomes urine and is eliminated from the body (NIDDK, 2018).

This section will discuss three kidney-related health effects associated with heat exposure: kidney stones, AKI, and CKD.

Kidney stones are hard objects that form in the kidney from the accumulation of minerals. They range in size from a grain of sand to a pea (NIDDK, 2017a). Symptoms include sharp pain in the back, side, lower abdomen, or groin; pink, red, or brown blood in the urine; a constant need to urinate; pain while urinating; inability to urinate or only able to urinate a small amount; and cloudy or foul-smelling urine (NIDDK, 2017b). Nausea, vomiting, fever, and chills are also possible, and symptoms may be brief, prolonged, or come in waves (NIDDK, 2017b). In rare cases or when medical care is delayed, kidney stones can lead to complications including severe pain, urinary tract infections (UTI), and loss of kidney function (NIDDK, 2017a). Risk factors for kidney stones include being male, a family history of kidney stones, having previously had kidney stones, not drinking enough liquids, other medical conditions ( e.g., chronic inflammation of the bowel, digestive problems, hyperparathyroidism, recurrent UTIs), drinking sugary beverages, and working in the heat, especially if unacclimatized (NIDDK, 2017a; Maline and Goldfarb, 2024). NIOSH has also cautioned workers that experiencing chronic dehydration can increase the risk of developing kidney stones (NIOSH, 2017a).

Kidney stones form when concentrations of minerals are high enough to the point of forming crystals, which then aggregate into a stone in either the renal tubular or interstitial fluid (Ratkalkar and Kleinman, 2011). Reduced urine volume, altered urine pH, diet, genetics, or many other factors may cause this concentration of minerals (Ratkalker and Kleinman, 2011). Heat exposure has the potential to cause kidney stones through heat-induced sweating and dehydration. Loss of extracellular fluid increases osmolality ( i.e., increased concentration of solutes, like sodium and glucose) which leads to increased secretion of vasopressin, an antidiuretic hormone. Vasopressin signals to the kidneys to conserve water by reducing urine volume, leading to increased concentration of relatively insoluble salts, like calcium oxalate, in the urine. These salts can eventually form crystals which can develop into stones (Fakheri and Goldfarb, 2011).

Epidemiological studies conducted outside the U.S. have documented the association between working in heat and developing kidney stones. One of the earliest publications on occupational heat and kidney stones was a small study of beach lifeguards in Israel (Better et al., 1980). Eleven of 45 randomly selected lifeguards (24%) were found to have had kidney stones, which Better et al. noted was approximately 20 times the incidence rate of the general Israeli population at the time. The authors attributed this finding to low urine output due to dehydration, hyperuricemia (elevated levels of uric acid in the blood), and absorptive hypercalciuria (elevated levels of calcium in the urine), among other factors. In 1992, Pin et al. compared outdoor workers exposed to hot environmental conditions to indoor workers exposed to cooler conditions (Pin et al., 1992). This study of 406 men in Taiwan included quarry, postal, and hospital engineering support workers. The prevalence of kidney stones was found to be significantly higher in the outdoor workers than the indoor workers (5.2% versus 0.85%, p<0.05). The authors posited that chronic dehydration from working outdoors in a tropical environment might explain the higher prevalence of kidney stones among outdoor workers (Pin et al., 1992).

Several studies have also considered occupational exposure to indoor heat sources. Borghi et al. studied machinists who had been working in the blast furnaces of a glass plant in Parma, Italy for five or more years, excluding those who had kidney stones before working at the plant (Borghi et al., 1993). The prevalence of kidney stones was significantly higher among machinists exposed to heat (n=236) than among those working in cooler temperatures (n=165) (8.5% vs. 2.4%, p=0.03) (Borghi et al., 1993). An analysis of risk factors revealed that workers in the heat lost substantially more water to sweat and that their urine had higher concentrations of uric acid, higher specific gravity, and lower pH than workers in normal temperatures (Borghi et al., 1993).

In a large study in Brazil, the prevalence of at least one episode of kidney stones was 8.0% among the 1,289 workers in hot areas, which was significantly higher than the 1.75% prevalence found among the 9,037 people working in room temperature conditions (p<0.001) (Atan et al., 2005). An analysis of a subset of workers demonstrated that workers in hot temperatures had significantly less citrate in their urine (p=0.03) and lower urinary volume (p=0.01) compared to room-temperature workers.

Venugopal et al. studied 340 steel workers in southern India engaged in moderate to heavy labor with three or more years of heat exposure (Venugopal et al., 2020). Of the 340 participants, 91 workers without other risk factors for kidney disease, but who had reported a symptom of kidney or urethral issues, underwent renal ultrasounds, which revealed that 27% had kidney stones. 84% of the participants with kidney stones were occupationally exposed to heat, as defined as working in conditions above the ACGIH TLV. Having five or more years of heat exposure was significantly associated with risk of kidney stones, while ( print page 70722) controlling for smoking (OR: 3.6, 95% CI: 1.2, 10.7).

Most recently, Lu et al. studied 1,681 steel workers in Taiwan, 12% of whom had kidney stones, compared to the age-adjusted prevalence among men in Taiwan of 9% (Lu et al., 2022). Heat exposure was found to be positively associated with prevalence of stones, particularly among workers ≤35 years old (OR: 2.7, 95% CI: 1.2, 6.0) (Lu et al., 2022).

Overall, the peer-reviewed literature supports occupational heat exposure as a risk factor for kidney stones, in both indoor and outdoor environments, across multiple countries, and in several industries.

Treatment of kidney stones depends on their size, location, and type. Someone with a small kidney stone may be able to pass it by drinking plenty of water and taking pain medications as prescribed by a doctor (NIDDK, 2017c). Larger kidney stones can block the urinary tract, cause intense pain, and may require medical intervention such as shock wave lithotripsy, cystoscopy, ureteroscopy, or percutaneous nephrolithotomy to remove or break up the stone (NIDDK, 2017c). Percutaneous nephrolithotomy, whereby kidney stones are removed through a surgical incision in the skin, requires several days of hospitalization, but the other interventions typically do not require an overnight hospital stay (NIDDK, 2017c). One study found that among working aged adults, approximately one third of people treated for kidney stones miss work and that they miss, on average, 19 hours of work per person (Saigal et al., 2005). With monitoring or treatment, people typically recover from kidney stones. However, over the long term, individuals who develop kidney stones are at increased risk of chronic kidney disease and end-stage renal disease, particularly if kidney stones are recurrent (Uribarri, 2020).

The available peer-reviewed scientific literature demonstrates occupational heat exposure as a risk factor for kidney stones, in both indoor and outdoor environments. Kidney stones may require medical treatment and in some cases hospitalization. Finally, individuals who develop kidney stones are at increased risk of other kidney diseases.

Acute kidney injury (AKI) can affect workers exposed to occupational heat. AKI is an abrupt decline in kidney function in a short period ( e.g., a few days). As normally functioning kidneys filter blood and maintain fluid balance in the body, AKI events can disrupt this fluid balance, which can impact major organs like the heart. AKI can also have metabolic consequences, like a build-up of too much potassium in the blood (hyperkalemia) (Goyal et al., 2023). AKI is not always accompanied by symptoms and is typically diagnosed with blood and/or urine tests ( e.g., increase in serum creatinine). While damage to the kidneys is one potential consequence of heat stroke (such as in the context of multi-organ failure, as mentioned in Section IV.E., Heat Stroke), this section is focused on AKI that is not necessarily preceded by clinical heat stroke.

There are three categories of AKI used to distinguish the location of the cause(s) of AKI—prerenal, intrarenal, and postrenal (Goyal et al., 2023). Prerenal AKI represents a reduction in blood volume being delivered to the kidneys ( i.e., renal hypoperfusion). This can be the result of heat-induced sweating that leads to reduced circulating blood volume. Prerenal AKI that is reversed ( e.g., dehydration is quickly reversed) is typically not associated with impairment to the kidney glomeruli or tubules, however prolonged exposure can lead to direct injury to renal cells through ischemia (inadequate blood and oxygen supply to cells). Intrarenal AKI is when the function of the glomeruli, tubules, or interstitium are affected, such as in the case of nephrotoxic exposures ( e.g., heavy metals) or prolonged ischemia. Rhabdomyolysis, which was previously discussed in Section IV.H., Rhabdomyolysis, is one potential cause of necrosis of tubular cells resulting from myoglobin precipitation and direct iron toxicity (Sauret et al., 2002, Patel et al., 2009). Postrenal AKI is when there is an obstruction to the flow of urine, such as kidney stones, pelvic masses, or prostate enlargement. Postrenal AKI is less relevant to a discussion of heat-related health effects, apart from kidney stones, which is discussed in Section IV.M.II., Kidney Stones.

Researchers have written specifically about potential mechanisms leading from occupational heat exposure to AKI (Roncal-Jiménez et al., 2015; Johnson et al., 2019; Schlader et al., 2019; Hansson et al., 2020), often in the context of chronic kidney disease. As previously discussed in Section IV.B., General Mechanisms of Heat-Related Health Effects, working in the heat can lead to increases in core temperature and reductions in circulating blood volume. Researchers hypothesize that elevated core temperature could directly injure renal tissue or that injury could be mediated through subclinical (mild and asymptomatic) rhabdomyolysis or increases in intestinal permeability that can cause inflammation. Reductions in blood volume could inflame or injure the kidneys through reduced renal blood flow that leads to ischemia and/or local reductions in adenosine triphosphate (ATP) availability. Reduced blood flow and increased blood osmolality also trigger physiologic pathways ( e.g., renin-angiotensin-aldosterone system, polyol-fructokinase pathway) which are energy-intensive and may lead to oxidative stress and inflammation. Other mechanistic pathways under investigation include urate crystal-induced injury (Roncal-Jiménez et al., 2015) and increased reabsorption of nephrotoxicants (Johnson et al., 2019).

Serum creatinine levels are used in clinical settings to estimate kidney function (glomerular filtration rate, or GFR), as it is typically produced in the body at a relatively stable rate and is removed from circulation by the kidneys. Multiple criteria exist for defining AKI based on increases in serum creatinine over hours or days, such as the KDIGO criteria published by a non-profit organization that produces recommendations on kidney disease (KDIGO, 2012). There are multiple factors that could affect the reliability of using serum creatinine to estimate GFR, including the increased production of creatinine during exercise. As a result of the limitations of serum creatinine, there is growing use of alternative biomarkers to identify cases of AKI, which may be more reliable and specific to AKI, such as neutrophil gelatinase-associated lipocalin, or NGAL.

Researchers have documented an association between heat strain and biomarkers of AKI in controlled experimental conditions. In 2013, Junglee et al. documented elevations in urine and plasma NGAL and reductions in urine flow rate in participants after a heat stress trial that induced elevations in core temperature and reductions in body mass (an indication of hydration status) (Junglee et al., 2013). These increases in NGAL were higher in an experimental group that underwent a muscle damaging, downhill (−10% gradient) run (compared to a non- ( print page 70723) muscle damaging run on a 1% gradient) prior to the heat stress trial, providing support for the argument that subclinical rhabdomyolysis may be a pathway from heat stress to kidney injury. Schlader et al. conducted a trial in which participants wearing firefighting gear completed two separate exercise trials in hot conditions of different durations. The longer duration trial was intended to induce higher levels of heat strain, while the shorter duration was intended to induce lower levels (Schlader et al., 2017). The researchers found that the longer trial was associated with elevated core temperature and reduced blood volume, as well as increases in serum creatinine and plasma NGAL, suggesting the magnitude of kidney injury may be proportional to the magnitude of heat strain. McDermott et al. tested longer durations of exercise in the heat (5.7 ± 1.2 hours) and similarly found elevations in serum creatinine and serum NGAL from before the trial to after (McDermott et al., 2018). To determine whether it is elevated core temperature or reduced blood volume that primarily drives heat-induced AKI, Chapman et al. conducted four trials in which subjects exercised for two hours in the same conditions, but received different interventions (water, cooling, water plus cooling, and no intervention) (Chapman et al., 2020). The group with no intervention had the highest levels of urinary AKI biomarkers in the recovery period, whereas the water and cooling groups each experienced reductions in AKI biomarker levels relative to the control group. The researchers concluded that limiting hyperthermia and/or dehydration reduces the risk of AKI.

The relationship between AKI and hyperthermia and/or dehydration has also been demonstrated in animal models (Hope and Tyssebotn 1983; Miyamoto 1994; Roncal-Jiménez et al., 2014; Sato et al., 2019).

In addition to experimental evidence, heat-related AKI has also been observed in “real world” conditions going back to the 1960s. In 1967, Schrier et al. documented evidence of military recruits developing AKI (referred to as “acute renal failure”) following training exercises in the heat (Schrier et al., 1967). It was soon after reported that AKI cases linked to exercise in the heat represented a sizeable portion (approximately 10%) of all AKI cases treated at Walter Reed General Hospital in the early 1960s (Schrier et al., 1970).

More recently, serum creatinine-defined AKI has been observed in agricultural workers in both Florida and California. Among a cohort of field workers from the Central Valley of California, Moyce et al. report a post-work shift incidence of AKI of 12.3% (35 of 283 workers) (Moyce et al., 2017). Workers with heat strain, characterized by increased core temperature and heart rate, were significantly more likely to have AKI (OR: 1.34, 95% CI: 1.04, 1.74). Among a cohort of agricultural workers in Florida, Mix et al. found that heat index (based on nearest weather monitor) was positively associated with the risk of AKI—47% increase in the odds of AKI for every 5 °F increase in heat index. The authors reported an incidence of AKI of 33% ( i.e., 33% of workers had AKI on at least one day of monitoring) in this study (Mix et al., 2018).

OSHA researchers have also identified cases of heat-related AKI among workers in the agency's own databases: the Severe Injury Reports (SIR) database and case files from consultations by the Office of Occupational Medicine and Nursing (OOMN) (Shi et al., 2022). Shi et al. identified 22 cases of heat-related AKI between 2010 and 2020 in the OOMN consultation records (based on serum creatine elevations meeting the KDIGO requirements) after excluding cases related to severe hyperthermia, multi-organ failure, or death. Using inclusion criteria of a heat-related OIICS code (172*) and a mention of AKI in the narrative, they also identified 57 cases of probable heat-related AKI between 2015 and 2020 in the SIR database.

Studies conducted among workers outside the U.S. have also reported a relationship between working in the heat and acute elevations in serum creatinine or increased risk of AKI (García-Trabanino et al., 2015; Wegman et al., 2018; Nerbass et al., 2019; Sorensen et al., 2019).

There are a few limitations to these observational studies, such as the use of serum creatinine to characterize AKI, as described above. An additional limitation is the inability to determine from these studies whether the AKI observed is due to prerenal or intrarenal causes. As discussed in Physiological Mechanisms, prerenal AKI may be due to reductions in renal blood flow (which would be expected in cases of dehydration) and is not necessarily indicative of clinically significant structural injury. Another limitation may be the use of serum creatinine measures taken over relatively short spans of time, which may be too short to see true reductions in GFR (Waikar and Bonventre, 2009). However, there are a growing number of studies that find a relationship between short-term fluctuations in serum creatinine and longer-term declines in kidney function among outdoor workers (see discussion in Section IV.M.IV., Chronic Kidney Disease).

There is a spectrum of severity for AKI. For example, some individuals may not know they are experiencing AKI without a serum or urine test. There is also a spectrum of time and medical treatment needed for recovery, dependent on whether the AKI is quickly reversed or sustained for longer periods of time. In Schlader et al. 2017, researchers noted that the biomarkers of AKI for participants in their trial returned to baseline the following day. However, intrarenal causes of AKI may require longer periods of time for recovery and may potentially require the need for medication or dialysis (Goyal et al., 2023). AKI can be severe, which can be the case when resulting from heat stroke, where it may represent irreversible damage to the kidneys and can be fatal (Roberts et al., 2008; King et al., 2015; Wu et al., 2021). Recurrent AKI may also lead to chronic kidney disease (as discussed in Section IV.M.IV., Chronic Kidney Disease).

The available peer-reviewed scientific literature, both experimental and observational studies, suggests that occupational heat exposure causes AKI among workers. However, there are limitations in the case definitions used to define AKI in observational settings.

Chronic kidney disease (CKD) is a progressive disease characterized by a gradual decline in kidney function over months to years. It is typically asymptomatic or mildly symptomatic until later stages of the disease, when symptoms such as edema, weight loss, nausea, and vomiting can occur (NIDDK 2017d). People with CKD can be at a greater risk for other health conditions, like AKI, heart attacks, hypertension, and stroke. The diagnosis typically requires multiple blood and urine tests taken over time (NIDDK 2016). Typical risk factors for CKD include hypertension and diabetes.

Epidemics of CKD in Central America and other pockets of the world, such as India and Sri Lanka, that appear to be afflicting mostly young, outdoor workers with no history of hypertension or diabetes have raised questions about ( print page 70724) whether working in hot conditions can cause the development of CKD (Johnson et al., 2019). Researchers have been investigating this question and the cause of the epidemic over the past 20 years, including other potential exposures, such as heavy metals, agrichemicals, silica, and infectious agents (Crowe et al., 2020).

Researchers have proposed that working in the heat could lead to the development of CKD through repetitive AKI events (see discussion of heat-related mechanisms in Section IV.M.III., Acute Kidney Injury). However, some researchers acknowledge the possibility that the unexplained CKD cases observed in Central America and elsewhere may instead represent a chronic disease process that begins earlier in life which places workers at increased risk of AKI (Johnson et al., 2019; Schlader et al., 2019). Additionally, as discussed above in Section IV.M.III., Acute Kidney Injury, some occupational cases of AKI could be transient, the result of prerenal causes, and possibly unrelated to the development of CKD.

Independent of the epidemic of unexplained CKD, frequent and/or severe AKI has been identified as a risk factor for developing CKD (Ishani et al., 2009; Coca et al., 2012; Chawla et al., 2014; Hsu and Hsu 2016; Heung et al., 2016). The relationship between heat-related AKI and risk of developing CKD is untested in the experimental literature because of the ethical implications (Schlader et al., 2019; Hansson et al., 2020).

As discussed in Section IV.E., Heat Stroke, there is also evidence that experiencing heat stroke may increase an individual's risk of developing CKD (Wang et al., 2019; Tseng et al., 2020).

As discussed previously in the context of AKI, serum creatinine is commonly used to estimate glomerular filtration rate (GFR), the indicator of kidney function. When measures of serum creatinine (and therefore estimates of GFR) are taken over periods of months to years, medical professionals can determine if an individual's kidney function is declining. CKD is typically diagnosed when the estimated GFR is below a rate of 60 mL/min/1.73m 2 for at least 3 months, although there are other indicators, like a high albumin-to-creatinine ratio. There are various stages of CKD; the final stage is called end-stage renal disease (ESRD) and represents a point at which the kidneys can no longer function on their own and require dialysis or transplant.

There is a growing body of evidence that suggests that heat-exposed workers who experience AKI (or short-term fluctuations in serum creatinine) are at greater risk of experiencing declines in kidney function over a period of months to years. For instance, sugarcane workers in Nicaragua who experienced cross-shift increases ( i.e., increase from pre-shift to post-shift) in serum creatinine at the beginning of the harvest season were more likely to experience declines in estimate GFR nine weeks later (Wesseling et al., 2016). Another study conducted among Nicaraguan sugarcane workers found that approximately one third of workers who experienced AKI during the harvest season had newly decreased kidney function (greater than 30% decline) and a measure of estimated GFR of less than 60 mL/min/1.73m2 one year later (Kupferman et al., 2018). In an analysis among Guatemalan sugarcane workers, Dally et al. found that workers with severe fluctuations in serum creatinine over a period of 6 workdays had greater declines in estimated GFR (−20% on average) (Dally et al., 2020). In a separate study conducted in Northwest Mexico, researchers observed declines in estimated GFR among migrant and seasonal farm workers from March to July that were not observed in a reference group of office workers in the same region (López-Gálvez et al., 2021).

Further support for the hypothesis that working in the heat may lead to declines in GFR and increased risk of CKD comes from intervention studies in Central America, in which workers were given water-rest-shade interventions and observed longitudinally for kidney outcomes. In these studies, implementation of the heat stress controls was associated with reductions in the declines in kidney function and reduced rates of kidney injury (Glaser et al., 2020; Wegman et al., 2018).

While much of the literature is focused on Central American workers, OSHA did identify one paper conducted among a cohort of U.S. firefighters. Pinkerton et al. (2022) found lower than expected rates of ESRD in the cohort (relative to the general U.S. population) despite high levels of occupational exposure to heat. However, as the authors point out, this may be due to the healthy worker effect ( i.e., a phenomenon in occupational epidemiology by which workers appear to be healthier than the general population due to individuals with health conditions leaving the workforce) (Pinkerton et al., 2022). The authors also examined associations between proxies for heat exposure and risk of developing ESRD and found non-significant associations between the number of exposed days and all-cause ESRD, systemic ESRD, and hypertensive ESRD. Very few of the ESRD cases identified in this cohort were due to interstitial nephritis (which would be most consistent with the CKD cases observed in Central America), limiting the authors' ability to examine associations between those cases and exposure.

There may be differences between the heat-exposed worker populations in Central America and the U.S. that could limit the ability to extrapolate findings from that region, such as differences in other potentially nephrotoxic exposures ( e.g., agrichemicals, infectious agents). There is also evidence that children in regions with epidemics of unexplained CKD have signs of kidney injury (Leibler et al., 2021). Unfortunately, surveillance of CKD in the U.S. (namely the U.S. Renal Data System) may be missing cases among susceptible workers, such as migrant agricultural workers, limiting the ability to detect a potential epidemic of heat-related CKD in this country.

In addition to the general lack of studies conducted among U.S. workers, there may be other limitations with these observational studies, such as limited data on longer-term follow-up ( i.e., years instead of months) and the potential for reverse causality ( i.e., undetected CKD is causing AKI).

Often kidney disease gets worse over time and function continues to decline as scarring occurs (NIDDK 2017d). As discussed above, late-stage CKD (or ESRD) requires dialysis or a kidney transplant for an individual to survive. Kidney failure is permanent. Having even early-stage CKD may impair workers' urine concentrating ability, which could increase their heat strain and risk of HRIs while working (Petropoulos et al., 2023).

There is growing evidence suggesting that heat stress and dehydration may be contributing to an epidemic of CKD among workers in Central America and other parts of the world, although the cause is still being investigated by researchers. There is currently limited information as to whether this type of CKD is affecting U.S. workers and if so, to what extent. Experiencing heat stroke has been identified in the literature as a risk factor for developing CKD. ( print page 70725)

In addition to the health effects discussed in the previous sub-sections, heat exposures have also been linked to reproductive health effects. Additionally, health effects have been associated with prior episodes of heat illness.

There is mixed evidence that heat affects reproductive and developmental health outcomes. NIOSH reported two mechanisms by which heat may affect reproductive and developmental health: infertility ( e.g., such as through damaged sperm) and teratogenicity (harm to the developing fetus, e.g., spontaneous abortion or birth defects) (NIOSH, 2016). NIOSH concluded that while human data about reproductive risks at exposure limits (see NIOSH, 2016, table 5-1, p. 70) were limited, results of research and animal experiments support the conclusion heat-related infertility and teratogenicity are possible (NIOSH, 2016, p. 91).

More recent evidence, although also limited, continues to provide support of a reproductive risk to people who are pregnant and developmental risk to their children. Numerous epidemiological studies have reported that heat exposure during pregnancy is associated with poor outcomes, such as pre-term labor and birth and low-birth weight babies ( e.g., Kuehn and McCormick, 2017; Basu et al., 2018; Chersich et al., 2020; Rekha et al., 2023). While most studies assess this relationship in the general population of pregnant women and do not specifically address occupational exposures, Rekha et al. show that occupational exposures to heat were associated with adverse pregnancy and fetal outcomes, as well as adverse outcomes during birth in a cohort of pregnant women in Tamil Nadu, India (Rekha et al., 2023). Although the mechanisms for these outcomes are unclear, a study of pregnant women conducting agricultural work or similar activities for their homes in The Gambia reported an association between heat exposure and fetal strain (through measures of fetal heart rate and umbilical artery resistance) (Bonell et al., 2022). Further, a recent longitudinal prospective cohort study in Germany found that heat exposure was associated with vascular changes in the uterine artery. This study reports that changes of increased placental perfusion and decreased peripheral resistance in the uterine artery indicate blood redistribution to the fetus during the body's response to heat stress. They also report increased maternal cardiovascular strain. This data may support a mechanistic role for uterine and placental blood flow changes during heat exposures in resultant birth outcomes, such as pre-term birth (Yuzen et al., 2023; Bonell et al., 2022).

There is evidence that occupational heat exposures can affect male reproductive health ( e.g., Mieusset and Bujan, 1995). Some research studies report associations between occupational heat exposure and time to conceive ( e.g., Rachootin and Olsen, 1983; Thonneau et al., 1997), sperm velocity (Figa-Talamanca et al., 1992), and measures of semen quality such as sperm abnormalities (Rachootin and Olsen, 1983; Bonde, 1992; Figa-Talamanca et al., 1992; De Fleurian et al., 2009). Effects of heat on sperm have also been demonstrated in experiments in animal models (Waites, 1991). Cao et al. report that in their study of heat stress in mice, heat stress reduced sperm count and motility (Cao et al., 2023). In this study, the heat exposed mice were exposed to 38°C (100.4 °F) temperatures for 2 hours per day for two weeks. When the mice were not being exposed to heat, they were kept at 25°C (77 °F). Control mice were kept at 25°C for the duration of the study. Their study results indicate that reduced sperm quality may be a result of disrupted testicular microbial environment and disruption in retinol metabolism that occurs during heat stress. Although, the authors note that the heat exposure does not accurately mimic real world heat exposures in humans.

While it is accepted that heat impairs spermatogenesis, or development of sperm ( e.g., MacLeod and Hotchkiss, 1941; Mieusset et al., 1987; Thonneau et al., 1997), some studies of occupational heat exposure find no relationship between heat and semen quality (Eisenberg ML et al., 2015). Another study found observable but not statistically significant associations between heat and semen quality (Jurewicz et al., 2014). Many studies of the effects of occupational heat exposure on reproductive outcomes are cross-sectional in nature and measure exposures through occupation categories or self-report answers on questionnaires ( e.g., Figa-Talamanca et al., 1992; Thonneau et al., 1997; Jurewicz et al., 2014). These methods can be susceptible to recall bias and misclassification errors, which can reduce accuracy in characterizing the association between occupational heat exposures and reproductive health outcomes, and they are also unable to determine causality on their own. Additional research that quantifies occupational heat exposures directly ( e.g., through measures of heat strain or on-site temperatures) would help to clarify the impacts of occupational heat exposures on male reproductive outcomes.

A limited number of studies have focused on a variety of long-term effects following a prior episode of heat illness. This includes research by Wallace et al., also reviewed by NIOSH in the 2016 Criteria for a Recommended Standard Occupational Exposure to Heat and Hot Environments, whose retrospective case control study of military members found that those who experienced an exertional heat illness event earlier in life were more likely to die due to cardiovascular or ischemic heart disease (Wallace et al., 2007). Similarly, Wang et al. reports that, in their retrospective cohort study in Taiwan, prior heat stroke was associated with a higher incidence of acute ischemic stroke, acute myocardial infarction, and an almost three-fold higher incidence of chronic kidney disease compared to patients who had other forms of heat illness or compared to the control group that had no prior heat illness, over the study's 14 year follow-up period (Wang et al., 2019). They also found significantly higher incidence of cardiovascular events, cardiovascular disease, and chronic kidney disease among individuals in the study who had other forms of heat illness (heat syncope, heat cramps, heat exhaustion, heat fatigue, heat edema and other unspecified effects) compared to the control group that had no prior heat illness. In a long-term follow-up study of military personnel who had experienced exertional heat illness, Phinney et al. reported a transient and small but observable increase in the rate of subsequent hospitalizations and decreased retention in the military (Phinney et al., 2001). While these studies suggest a relationship between episodes of serious heat illness and subsequent health effects, this body of research is small and subject to some limitations. The cross-sectional nature of some of these studies does not allow for determination of causality on their own. Additionally, given the retrospective nature of some of these studies it is possible that important confounding variables were not adjusted for in analyses, including occupation in some cases. ( print page 70726)

The description of evidence presented here demonstrates that there is some evidence to support a link between occupational heat exposures and adverse reproductive health outcomes. There is also limited evidence that prior episodes of heat illness may affect health outcomes later in life such as increased risk of cardiovascular disease and kidney diseases. This evidence of reproductive and developmental health effects and health effects associated with prior episodes of heat illness, while suggestive, is still nascent and requires further investigation.

This section discusses individual risk factors for heat-related injury and illness. The purpose of this discussion is to summarize the factors that may exacerbate the risk of workplace heat-related hazards and to provide information to better inform workers and employers about those hazards. However, exposure to workplace heat contributes to heat stress for all workers and can be detrimental to workers' health and safety regardless of individual risk factors. OSHA is not suggesting that application of the proposed standard would depend on an employer's knowledge or analysis of these factors for their individual workers. Nor do these individual risk factors detract from the causal link between occupational exposure to heat and adverse safety and health outcomes or an employer's obligation to address that occupational risk (see Reich v. Arcadian Corp., 110 F.3d 1192, 1198 (5th Cir. 1997) (Congress intended the Act to protect all employees, “regardless of their individual susceptibilities”); Pepperidge Farm, Inc., 17 O.S.H. Cas. (BNA) ¶ 1993 (O.S.H.R.C. Apr. 26, 1997) (that non-workplace factors may render some workers more susceptible to causal factors does not preclude finding the existence of an occupational hazard); see also Bldg. & Const. Trades Dep't, AFL-CIO v. Brock, 838 F.2d 1258, 1265 (D.C. Cir. 1988) (holding that OSHA did not err in including smokers in its analysis of the significant risk posed by occupational exposure to asbestos, despite the “synergistic effects” of smoking and asbestos)). Many factors can influence an individual's risk of developing heat-related health effects. These factors include variation in genetics and physiology, demographic factors, certain co-occurring health conditions or illnesses, acclimatization status, certain medications and substances, and structural factors ( e.g., economic, environmental, political and institutional factors) that lead to disproportionate exposures and outcomes. Although there is a lack of evidence that explores the full extent to which these factors interact to affect heat-related health effects, or how various risk factors compare in their impacts, there is evidence that each of these factors can affect risk of heat-related health effects. This section focuses on factors that relate to an individual's health status. For an in-depth discussion on acclimatization as a risk factor, see Section V., Risk Assessment, and for an in-depth discussion on demographic factors and structural factors that affect risk of heat-related illness, see Section VIII.I., Distributional Analysis.

There are a number of factors that can impact an individual's response to heat stress and lead to variation in heat stress response between individuals. These include variation in genotype (Heled et al., 2004), gene expression (Murray et al., 2022), body mass and differences in thermoregulation between the biological sexes (Notley et al., 2017), differences in thermoregulation as people age ( e.g., Pandolf 1997, Kenny et al., 2010; Kenny et al., 2017), and pregnancy (Wells, 2002; NIOSH, 2016). Normal variation across individuals in genetics, physiology, and body mass results in variation in how individuals respond to heat stress. There is some evidence that, at least in some specific populations, variation in genotype ( i.e., genetic makeup) can affect heat storage and heat strain (Heled et al., 2004; Gardner et al., 2020). Normal variation in body mass can also correspond to variation in thermoregulation between individuals ( e.g., Havenith et al., 1998). Results from Havenith et al.'s experimental study of heat stress under different climate and exercise types indicates that one reason for this effect may be due to the relationship between size and surface area of the skin which plays an important role in cooling capacity (Havenith et al., 1998). A more detailed discussion of the relationship between obesity and heat stress response can be found below.

There is some evidence that biological sex could be considered a risk factor for heat-related illness, although the evidence is mixed. Some studies find differences in heat stress response between males and females ( e.g., Gagnon et al., 2008; Gagnon and Kenny, 2011; Gagnon and Kenny, 2012). These differences may be due to differences in body mass (Notley et al., 2017), lower sweat output in females or differences in metabolic heat production (Gagnon et al., 2008; Gagnon and Kenny, 2012). However, recent experimental data assessing differences in thermoeffector responses (autonomic responses that affect thermoregulation, such as skin blood flow and sweat rate) between males and females exposed to exercise show that differences between the sexes in heat stress response are mostly explained by differences in morphology (body shape and size and the resultant mass-surface ratios) (Notley et al., 2017). Although, Notley et al.'s (2017) experiment only involved heat environments where enough heat could be lost so that the body does not continue to gain heat (compensable heat stress), so it is unclear if an increased effect due to biological sex would occur in conditions where heat gain is expected, such as in occupational settings where environmental heat or environmental heat and exertion exceed the body's ability to cool.

Healthy aging processes can also make individuals more susceptible to heat-related illness. Aging may impact thermoregulation through reduced cardiovascular capacity (Minson et al., 1998; Lucas et al., 2015), reduced cutaneous vasodilation (the widening of blood vessels at the skin to aid heat loss), sweat rate, altered sensory function (Dufour and Candas, 2007; Wong and Hollowed, 2017), and changes in fluid balance and thirst sensation (Pandolf, 1997). Observational evidence tends to show that elderly individuals, particularly those with co-existing chronic or acute diseases, are at highest risk for morbidity or mortality related to heat exposures, and that risk increases with age ( e.g., Semenza et al., 1999; Fouillet et al., 2006; Knowlton et al., 2008). However, experimental evidence shows that, under certain conditions, when individuals are matched for fitness level and body build and composition, middle-aged individuals can compensate for heat exposures similarly to younger adults (Lind et al., 1970; Pandolf, 1997, Kenny et al., 2017). Conversely, observational studies of occupational populations often find that younger workers experience greater rates of heat-related illness than do older workers ( e.g., Harduar Morano et al., 2015; Hesketh et al., 2020; Heinzerling et al., 2020). While it is unclear why younger workers appear to have greater rates of heat-related illness in epidemiological data, Heinzerling et al. (2020) suggest that this could be a result of a greater number of younger workers being ( print page 70727) employed in high-risk occupations. Further, younger workers have less work experience, meaning that younger workers are less familiar with the heat risks associated with their jobs, how their body responds to heat, and/or how to respond if they experience symptoms of heat-related illness.

Health status is another factor that plays a role in how someone responds to heat stress ( e.g., Semenza et al., 1999; Knowlton et al., 2008; NIOSH, 2016; Vaidyanathan et al., 2019, 2020). Conditions such as cardiovascular disease and diabetes can affect risk of heat-related illness ( e.g., Kenny et al., 2016; Kenny et al., 2018). The cardiovascular system plays an integral role in thermoregulation and heat stress response (Costrini et al., 1979; Lucas et al., 2015; Wong and Hollowed, 2017; Kenny et al., 2018). Cardiovascular diseases can affect the heart and blood vessels, increasing cardiovascular strain and decreasing cardiovascular function and thermoregulatory capacity (Kenny et al., 2010) and, as a result, increase risk of heat-related illness during heat stress (Kenny et al., 2010; Semenza et al., 1999). For example, people with hypertension ( i.e., high blood pressure) may be at increased risk of heat-related illness due to changes in skin blood flow that can impair heat dissipation during heat stress (Kenny et al., 2010). Further, many individuals with hypertension and cardiovascular diseases may take prescription medications that reduce thermoregulatory functions, through mechanisms like reduced blood flow to the skin, which can increase sensitivity to heat (Wee et al., 2023). Studies estimate that a substantial percentage of the population, and therefore the population of workers, have the type of health status ( i.e., having a chronic condition such as cardiovascular diseases) (Boersma et al., 2020; Watson et al., 2022) that could affect their response to heat stress. For example, Watson et al. (2022) estimate that of the 46,781 surveyed adults between the ages of 18 and 34 who reported being employed, 26.1% have obesity, 11% have high blood pressure, and 9.7% have high cholesterol. Additionally, 19.4% were estimated to have depression, which is sometimes treated with medications that can affect thermoregulation.

Diabetes and obesity are other factors that may affect risk of developing heat-related illness (Kenny et al., 2016). Both diabetes and obesity may affect thermoregulation by reducing a person's ability to dissipate heat through changes in skin blood flow and sweat response (Kenny et al., 2016). While some evidence shows that individuals with well-controlled diabetes may be able to maintain normal thermoregulatory capacity (Kenny et al., 2016), some evidence indicates that individuals with poorly controlled diabetes (Kenny et al., 2016) or older individuals with Type 2 diabetes (Notley et al., 2021) may experience decreased heat tolerance. Obesity has also been identified as a risk factor for exertional heat illness in the military ( e.g., Bedno et al., 2014; Nelson et al., 2018b; Alele et al., 2020). Gardner et al. (1996) reported increasing risk of exertional heat illness among male Marine Corps recruits as BMI increased. Additionally, a smaller body mass to surface area ratio can reduce capacity for heat loss since surface area is relatively smaller in relationship to mass (Bar-Or et al., 1969; Kenny et al., 2016). Differences in tissue properties between adipose (fat) tissue and other body tissues may indicate that a higher body fat mass can lead to greater rises in core temperature for a given amount of heat storage in the body (Kenny et al., 2016).

Beyond chronic health conditions, prior episodes of significant heat-related illness and recent or concurrent acute illness or infection may also affect an individual's response to heat stress and increase the risk of heat-related illness ( e.g., Carter et al., 2007; Nelson et al., 2018a; Nelson et al., 2018b; Alele et al., 2020). Reviews of research and case studies of heat-related illness indicate that acute illnesses that may affect risk of heat-related illness include upper respiratory infections and gastrointestinal infections (Casa et al., 2012; Alele et al., 2020). However, statistical evidence is limited (Alele et al., 2020). Leon and Kenefick (2012) discuss results from a study of four marine recruits who presented with exertional heat illness and who also had an acute illness separate from heat-related illness. The recruits' blood tests showed elevated levels of immune-related substances which Leon and Kenefick identify as being substances that are both mediators of viral infection symptoms and substances associated with exertional heat illness. Leon and Kenefick interpret this observation, along with evidence from a study on rats that showed that bacteria exposure exacerbated inflammation and organ dysfunction due to heat stress, to suggest that pre-existing inflammatory states, such as those that occur with acute viral illness, compromise the ability to thermoregulate appropriately (Carter et al., 2007; Leon and Kenefick, 2012) (see also Bouchama and Knochel, 2002). Several studies in military populations also show that a prior heat illness may increase risk of a future episode of heat illness (Nelson et al., 2018b; Alele et al., 2020). Assessments of heat and epigenetics (the study of how the environment and behavior affects genes) suggest that the complex physiological responses to heat impact genetic mechanisms that could play a role in increasing susceptibility to future heat illness following an episode of heat illness (Sonna et al., 2004; Murray et al., 2022).

Certain medications can also affect thermoregulation and risk of heat-related illness. Medications that may decrease thermoregulatory capability include medications that treat cardiovascular diseases, diabetes, neuropsychiatric diseases, neurological diseases, and cancer (Wee et al., 2023). Some of these medications affect thermoregulation by directly affecting the region of the brain that controls thermoregulation or through other central nervous system effects ( e.g., antipsychotics, dopaminergics, opioids, amphetamines) (Cuddy, 2004; Stollberger et al., 2009; Musselman and Saely, 2013; Gessel and Lin, 2020; Wee et al., 2023). Other medications affect thermoregulation through effects on heat dissipation that occur due to changes in sweat response and/or blood flow to the skin ( e.g., anticholinergics, antihypertensives, antiplatelets, some antidepressants and antihistamines, aspirin) (see, e.g., Freund et al., 1987; Cuddy, 2004; Stollberger et al., 2009; Wee et al., 2023; CDC, 2024b). There are also medications that may affect ability to perceive heat and exertion ( e.g., dopaminergics) (Wee et al., 2023). Some medications can affect electrolyte balances ( e.g., diuretics, beta-blockers, calcium channel blockers, and antacids) (CDC, 2024b). When accompanied by dehydration, some medications also pose a toxicity risk ( e.g., apixaban, lithium, carbamazepine) (CDC, 2024b). Finally, some medications can affect fluid volume, kidney function, hydration status, thirst perception, or cardiac output ( e.g., diuretics, ACE inhibitors, some anti-diabetics, beta-blockers, non-steroidal anti-inflammatories (NSAIDs), tricyclic antidepressants, laxatives, and antihistamines) (Stollberger et al., 2009; Wee et al., 2023; CDC, 2024b). The NIOSH Criteria for a Recommended Standard for Occupational Exposure to Heat and Hot Environments (table 4-2), the Department of the Army's Technical Bulletin 507 (table 4-2), and CDC's Heat and Medications—Guidance for Clinicians contain additional information about classes of ( print page 70728) medications and the proposed mechanisms for how they affect thermoregulation (NIOSH, 2016; Department of the Army, 2022; CDC, 2024b).

Medications that can affect how individuals respond to heat are used by a significant portion of the U.S. population. Survey data from the National Health and Nutrition Examination Survey from 2015-2016 showed that 60% of adults aged 40-79 used a prescription medication within the last thirty days and approximately 22% of adults in that same age range took five or more prescription medications (Hales et al., 2019). Many of the medications reported by survey respondents are medications that can affect an individual's response to heat ( e.g., commonly used blood pressure and diabetes medications).

Amphetamines (whether prescription or illicit), methamphetamines, and cocaine can also affect thermoregulation and increase risk of heat-related illness (NIOSH, 2016; Department of the Army, 2022). These substances can affect the central nervous system's thermoregulatory functions, stimulate heat generation, and reduce heat dissipation through vasoconstriction (Cuddy, 2004). The synergy between the hyperthermia induced by these substances, physical activity, and heat exposure can increase risk of heat-related illness (Kiyatkin and Sharma, 2009). Analyses of occupational heat-related fatalities find amphetamines and methamphetamines to be an important risk factor (Tustin et al., 2018a, Karasick et al., 2020; Lin et al., 2023). In Lin et al.'s 2023 review of heat-related hospitalizations and fatalities documented through NIOSH Fatalities in Oil and Gas Database (2014-2019) and OSHA's Severe Injury Report Database (2015-2021), 50% of identified fatalities occurred in workers that had tested positive for amphetamines or methamphetamines after they died. However, small sample sizes, sampling strategies, and incomplete data have so far limited the ability of studies to fully characterize the association between these substances and risk of heat-related illness or fatality. Poor data quality or limited data has also limited current studies from concluding if and when amphetamine-like substances are from prescription or non-prescription use.

Alcohol and caffeine use may also affect risk of heat-related illness through effects on hydration status and heat tolerance (NIOSH, 2016; Tustin, 2018; Department of the Army, 2022). There have been cases of fatalities due to occupational heat exposure in individuals with a history of “alcohol abuse or high-risk drinking” (Tustin et al., 2018a, p. e385). Both alcohol and caffeine may affect how someone responds to heat stress due to their ability to cause loss of fluids and subsequently dehydration, and alcohol also affects central nervous system function (NIOSH, 2016). In the case of caffeine, it appears that moderate consumption associated with normally caffeinated beverages ( e.g., one cup of coffee, tea, soda) may not interfere with thermoregulation in a way that negatively affects response to heat stress (NIOSH, 2016; Kazman et al., 2020; Department of the Army, 2022). However, heavily caffeinated beverages, such as energy drinks, have been linked to negative health outcomes (Costantino et al., 2023) and could potentially exacerbate heat stress through diuretic (salt and water loss) mechanisms and cardiovascular strain (NIOSH, 2016). Overall, there is a lack of robust data that quantify the specific amounts of alcohol or caffeine that are problematic for heat stress response. However, experts generally advise against drinking alcohol or caffeinated beverages before or during work or exercise in the heat (NIOSH, 2016; Department of the Army, 2022; CDC, 2022).

The evidence presented in this section demonstrates that there are numerous factors that can affect risk of heat-related illness ( e.g., genetics, age, body mass, some chronic conditions, prescription medications and drugs). Because prevalence data show that a majority of working-age adults live with or experience at least one risk factor, these factors should be considered an important component of understanding how individuals can be at increased risk for heat-related illness. OSHA acknowledges, however, that for most of the described risk factors, the evidence is not robust enough to determine the full picture of how the factor impacts risk of heat-related illness or to establish the degree to which the risk factor contributes to overall risk of developing heat-related illness. There is also a lack of evidence evaluating the way in which multiple risk factors combine to affect risk of heat-related health outcomes.

In addition to heat-related illnesses, heat exposure can lead to a range of occupational heat-related injuries. A heat-related injury means an injury, such as a fall or cut, that is linked to heat exposure. A heat-related injury may occur as a result of a heat-related illness, such as a fracture following heat syncope. The association between heat exposure and heat-related injury among workers has been well documented over the last decade (Tawatsupa et al., 2013; Xiang et al., 2014b; Adam-Poupart et al., 2015; Spector et al., 2016; McInnes et al., 2017; Calkins et al., 2019; Dillender, 2021; Dally et al., 2020; Park et al., 2021; Negrusa et al., 2024). In particular, analyses of workers' compensation claim data has demonstrated the increased risk of occupational traumatic injury with increasing heat exposure (Xiang et al., 2014b; Adam-Poupart et al., 2015; Spector et al., 2016; McInnes et al., 2017; Calkins et al., 2019; Dillender, 2021; Park et al., 2021; Negrusa et al., 2024). These types of heat-related injuries can cause hospitalizations, extended time out of work, and reduced productivity. In some instances, a heat-related injury may be fatal, like in the event of accidents such as a slip, trip, or fall. In 1972, NIOSH identified occupational heat exposure as contributing to workplace injuries, and discussed how accidents and injuries were outcomes that could be prevented by a heat stress standard (NIOSH, 1972). Specifically, NIOSH highlighted how reduced physical and psychological performance, fatigue, accuracy of response, psychomotor performance, sweaty palms, and impaired vision may result in a workplace heat-related injury.

Since multiple types of injuries can be heat-related ( e.g., strain, fracture, crushing) and the mechanisms underlying those injuries vary ( e.g., impaired speed and reaction time, impaired vision, impaired dexterity), the identification and classification of heat-related injuries varies on a case-by-case basis. Although there are no ICD or OIICS codes specific to diagnosing heat-related injuries, medical professionals and occupational health professionals can combine a heat-related illness code with other injury related codes to indicate an injury is heat-related. An injury specifically attributed to heat would be expected to be assigned both a heat-related OIICS or ICD code and an injury OIICS or ICD code. Numerous researchers have used ICD and OIICS code to conduct studies on heat-related injuries (Dillender, 2021; Garzon-Villalba et al., 2016; Morabito et al., 2006; Spector et al., 2016).

This section first presents the epidemiological evidence of increasing occupational injuries during periods of hotter temperatures, followed by a discussion of mechanisms that can lead to heat-related injuries. ( print page 70729)

A multitude of studies have identified an association between heat exposure and occupational injury in the U.S. (Knapik et al., 2002; Fogleman et al., 2005; Garzon-Villalba et al., 2016; Spector et al., 2016; Calkins et al., 2019; Dillender, 2021; Park et al., 2021; Negrusa et al., 2024). These analyses primarily rely on workers' compensation claim data and meteorological data and are often case-crossover or observational time-series in design.

In two studies of outdoor agricultural workers (Spector et al., 2016) and outdoor construction workers (Calkins et al., 2019) in Washington State, traumatic injury claims were significantly associated with heat exposure. Among outdoor agricultural workers (n=12,213 claims), Spector et al. (2016) found a statistically significant increased risk of traumatic injuries at a daily maximum humidex (the apparent, or “feels like,” temperature calculated from air temperature and dew point, similar to heat index) above 25 °C (77 °F). Among outdoor construction workers (n=63,720 claims), Calkins et al. (2019) found an almost linear statistically significant association between traumatic injury risk and humidex. Both studies reported that injuries most commonly resulted from falls or bodily reaction and exertion, which may include sudden occurrences of strains, sprains, fractures, or loss of balance, among others (Spector et al., 2016; Calkins et al., 2019).

Using workers' compensation claim data from Texas, Dillender (2021) found that hotter temperatures resulted in larger percent increases in traumatic injuries among two similar sets of injury types, “open wounds, crushing injuries, and factures” and “sprains, strains, bruises, and muscle issues.” Park et al. (2021) examined over 11 million workers' compensation records in California and estimated that approximately 20,000 additional injuries per year between 2001 and 2018 were related to hotter temperatures. In comparison to a day with temperatures in the 60s °F, the risk of occupational heat-related injury increased by 5-7% (p<0.05) and 10-15% (p<0.05) on days with high temperatures between 85-90 °F and above 100 °F, respectively (Park et al., 2021).

In these case-crossover studies, cases serve as their own controls, allowing for variables such as age, sex, race, and ethnicity, as well as other known and unknown time-invariant confounders to be controlled. However, there are still some limitations to these studies, such as the potential for time-varying confounders ( e.g., air pollutants like ozone and sleep duration influenced by nighttime temperatures).

Studies conducted among workers outside the U.S. have also reported a relationship between working in the heat and increased risk of injuries (Morabito et al., 2006; Tawatsupa et al., 2013; Adam-Poupart et al., 2015; McInnes et al., 2017; Martinez-Solanas et al., 2018). Analyses from Dally et al. (2020), found an increase in injury risk with increasing average daily mean WBGT above 30 °C (86 °F) among sugarcane harvesters in Guatemala; although this result was not statistically significant, this may have been due to small sample and event size.

Heat exposure can impair workers' psychomotor and mental performance, which can interfere with routine occupational tasks. Consequently, the risk of work-related injuries, including slips, trips, and falls, as well as cuts and other traumatic injuries, is exacerbated when job tasks are performed in hot environments. As summarized in the prior health effects sections of this preamble, heat can impair a variety of physiological systems and produce a range of symptoms. Changes in the cardiorespiratory, locomotor, and nervous systems due to heat exposure can induce various bodily responses such as fatigue, which may lead to injury (Ross et al., 2016). Changes from elevated skin and core body temperatures, which may result in increased sweating and dehydration, can cause decrements in physical, visuomotor, psychomotor, and cognitive performance (Grandjean and Grandjean, 2007; Lieberman, 2007). Even experiencing a high level of heat sensation may contribute to discomfort and distress, causing distraction and other behavioral changes that can result in accidents and injuries (Simmons et al., 2008). An explanation of how heat exposure can impair psychomotor and mental performance, and consequently lead to occupational heat-related injuries is provided below.

Heat exposure can impair psychomotor function ( i.e., the connection between mental and muscle functions) which may cause heat-related injuries. Impaired psychomotor function from heat exposure can take multiple forms, including impaired movement, strength, or coordination (fatigue); impaired postural stability and balance; and impaired accuracy, speed, and reaction time. Each of these impairments to psychomotor performance are discussed in turn below.

Heat exposure can hamper psychomotor performance by impairing workers' movement, strength, or coordination and causing fatigue. Fatigue has been described as having a lack of energy or a feeling of weariness or tiredness (NIOSH, 2023b). Effects from heat strain on the cardiorespiratory and locomotor systems can cause both central and peripheral fatigue due to increased heat storage at the brain and muscle levels, along with other physiological mechanisms (Ross et al., 2016). As an individual's metabolic rate increases in hot environments, blood pH level may become more acidic and cause muscle fatigue from increased muscle glycogen degradation, lactate accumulation, and elevated carbohydrate metabolism (Varghese et al., 2018). These changes have been shown to compromise performance.

Numerous studies demonstrate the relationship between heat exposure and fatigue. In a cross-sectional survey of 256 occupational health and safety professionals in Australia, fatigue was the most reported incident in workers during higher temperatures (Varghese et al., 2020). Among two groups of 55 steel plant workers who completed a questionnaire assessing fatigue, the group of workers exposed to hotter environments (30-33.2 °C (80-91.76 °F) WBGT) were significantly more likely to report symptoms of fatigue in comparison to workers in cooler environments (25.4-28.7 °C (77.7-83.6 °F) WBGT) (Chen et al., 2003). This study highlights how fatigue symptoms increase with rising heat exposure levels (Chen et al., 2003).

Moreover, in a review of 55 studies on workplace heat exposure, core temperature elevation and dehydration have been shown to have numerous negative behavioral effects including fatigue, lethargy, and impaired coordination, which may lead to injury (Xiang et al., 2014a). These 55 articles included ecological (22%), cross-sectional (64%), and cohort (5%) studies, as well as epidemiological experiments (9%). From one study included in the review, 42% of construction workers surveyed reported it was “easy to get fatigued” while working in the summer (Inaba and Mirbod, 2007). In another review of heat stress risks in the construction industry, Rowlinson et al. (2014) also discussed the association of high temperatures and ( print page 70730) level of fatigue, which has been considered one of the critical factors leading to construction accidents (Garrett and Teizer, 2009; Chan, 2011). In a case study of 15 workers who experienced fatigue-related accidents, fatigue was shown to trigger other safety risks, such as not following proper safety procedures or becoming distracted, which can induce injury (Chan, 2011).

Heat exposure has also been shown to impair postural stability and balance as increases in metabolic heat can impact workers' gross motor capacity ( i.e., the ability to move the body with appropriate sequencing and timing to perform bodily movements with refined control), including postural balance. As individuals become dehydrated, they may experience negative neuromuscular effects. Distefano et al. (2013) demonstrated the detrimental impact of dehydration during task performance in hot conditions, where subjects experienced decreased neuromuscular control as characterized by poorer postural stability. The authors found that neuromuscular control was impaired while participants were hypohydrated (defined as uncompensated loss of body water) and hyperthermic. Additionally, when an individual is experiencing high-intensity exertion in hot environments and is already dehydrated, this can result in further dilution of blood sodium. When blood sodium is diluted, water may be forced from the extracellular compartment into the intracellular compartment, which could lead to pulmonary congestion, brain swelling, and heat stroke (Distefano et al., 2013). At this stage, neurons begin degenerating in the cerebellum and cerebral cortex, and this process coupled with the rise in body temperature, impairs central nervous system functionality (Sawka et al., 2011; Nybo, 2007; Distefano et al., 2013).

Research also indicates that performing exertional activities in a hot environment may impair balance. To better understand lower extremity biomechanics, Distefano et al. (2013) used an assessment tool to measure gross movement errors, such as medial knee displacement, hip or knee rotation, and limited sagittal plane (front to back) motion. The authors found that after performing the exercise protocol, participants demonstrated poorer movement technique when they were hypohydrated in a hot environment compared with when they were hypohydrated in a temperate environment or in a hot environment but euhydrated (state of optimal total body water content) (Distefano et al., 2013). These findings suggest that working in hot temperatures while dehydrated may increase risk for injury due to impaired balance (Distefano et al., 2013).

The compromising effects of heat strain on psychomotor function have long been established, but the level of performance deterioration is dependent on the severity of heat strain and the complexity of the task (Taylor et al., 2016; Hancock, 1986; Ramsey, 1995; Pilcher et al., 2002; Hancock and Vasmatzidis, 2003). Some research has found that when high skin and core temperatures increase cardiovascular strain, heat exposure results in faster reaction times where individuals respond more quickly, but less accurately when in the heat (Simmons et al., 2008). Other research, such as Mazloumi et al. (2014), found that heat stress conditions impair selective attention (the ability to select and focus on a particular task while simultaneously ignoring other stimuli) and reaction time. In their study of 70 workers in Iran, where half of the workers experienced heat stress and half worked in air-conditioning, the authors found impaired psychomotor function among the exposed workers indicated through an increase in the duration of a task and response time as well as an increase in the number of errors (Mazloumi et al., 2014).

Additional studies examine the impacts of high skin and core temperatures on psychomotor function contributing to more mistakes (Allan and Gibson, 1979; Gibson and Allan, 1979; Gibson et al., 1980). In one study of foundry workers, response time, reaction time, and number of errors were reported to be adversely affected when workers were exposed to WBGTs of 31-35 °C (87.8-95 °F) compared to unexposed workers in a WBGT of 17 °C (62.6 °F) (Mazlomi et al., 2017). A meta-analysis review of 23 studies supports these conclusions, finding that under hot conditions, performance on mathematical-related tasks and reaction time tasks can be negatively impacted at 32.2 °C (89.9 °F) with a roughly 15% average decrement in performance (Pilcher et al., 2002).

Pyschomotor performance is an important factor when considering job tasks that require precision and concentration to prevent injuries. In a study observing steel plant workers, it was found that electrical arc melting workers who were exposed to hotter environments (30-33.2 °C WBGT) experienced a significant decrease in their attention span and slower response time compared to the continuous cast workers, who worked in cooler environments (25.4-28.7 °C WBGT) (Chen et al., 2003). A decline in psychomotor function could also negatively affect speed of response, reasoning ability, associative learning, mental alertness, and visual perception, which has been reported as a key cause of fatal accidents (Rowlinson et al., 2014).

The effects of heat exposure on mental performance can also play a significant role in increasing workplace accidents and injuries and compromise workplace safety. Heat exposure can result in impaired cognition or cognitive performance; impaired visual motor tracking; and impaired decision-making or judgment, which can lead to unsafe behaviors (like the removal of required PPE). Each of these are discussed in turn below.

Declines in cognitive function from heat are correlated with an elevated risk of injury. Evidence indicates a statistically significant increase in unsafe behaviors above 23 °C WBGT and an increased risk of accidents (Ramsey et al., 1983). When an individual experiences hyperthermia, even if it is mild and only occurring for a short period, the central nervous system is vulnerable to damage (Hancock and Vasmatzidis, 2003). This can acutely affect memory, attention, and ability to process information (Walter and Carraretto, 2016). When hyperthermia triggers cerebral damage, these cerebral injuries can be characterized into three broad areas. The first area includes cellular effects (where cells are damaged as temperatures continue to rise and normal cell function is disrupted and cell replication is no longer possible). The second area includes local effects (like inflammatory changes and vascular damage), and the third area includes systemic changes (like changes in cerebral blood flow (Walter and Carraretto, 2016). These negative effects are typically seen when core body temperatures reach 40 °C (104 °F), although some changes can begin at temperatures of 38 °C (100.4 °F) (Walter and Carraretto, 2016). These physiological changes also negatively impact cognitive performance.

Heat exposure has been shown to affect cognitive performance ( print page 70731) differentially, based on type of cognitive task (Yeoman et al., 2022). The more complex a task, especially if it requires motor accuracy, the more likely an individual's cognitive ability to perform the task will decline because of heat stress (Hancock and Vasmatzidis, 2003). Some research indicates a decrease in cognitive performance for tasks requiring more perceptual motor skills will be observed in the 30-33 °C (80-91.4 °F) range, well before the physiological system reaches its tolerance limit (Ramsey and Kwon, 1992; Hancock and Vasmatzidis, 2003; Piil et al., 2017). Ramsey and Kwon (1992) have summarized over 150 studies looking at task exposure time and task type and found statistically significant performance decrements at the 30-33 °C (80-91.4 °F) range. The decrements at this range occurred regardless of duration of exposure (from short exposures under 30 minutes and longer exposures up to 8 hours) (Ramsey and Kwon, 1992). Furthermore, in a case study of nine male volunteers, results indicate that highly motivated subjects were strongly affected by heat load within the first two hours of exposure, and that these subjects' performance was significantly impaired when assigned complex tasks requiring a significant amount of reasoning and judgment (Epstein et al., 1980). The authors found that performance began to decrease when workers were exposed to temperatures above 27 °C (80.6 °F).

Moreover, in a review of fifteen laboratory experiments assessing the effects of high ambient temperature on mental performance, one study found that mental performance declines were statistically significant at exposure durations of four consecutive hours in 87 °F (30.55 °C) temperatures (Wing, 1965). Similarly, in a study of the effects of hot-humid and hot-dry environments on mental functioning, 25 participants were exposed to a variety of temperatures in humid and dry conditions, while performing physical exercises with bouts of rest, to assess mental alertness, associative learning, reasoning ability and dual-performance efficiency (Sharma et al., 1983). The authors found that all the psychological functions tested were adversely affected under heat stress, and that a significant drop in various psychological functions was seen at temperatures of 32.2 °C (89.9 °F) and 33.3 °C (91.9 °F) in hot-humid and hot-dry conditions, respectively. Moreover, the authors suggest that, for heat-acclimatized subjects who continuously work for four hours, that the temperature should not exceed 31.1 °C (87.9 °F) in hot and humid conditions, and 32.2 °C (89.9 °F) for workers in hot desert conditions (Sharma et al., 1983).

Hyperthermia and dehydration, a common symptom of heat exposure, have been found to impair visual-motor tracking ( i.e., the eyes' ability to focus on and follow an object), increasing the risk of workplace injury. In a review of studies on hydration and cognition, the authors indicate that a 2% or more loss of body weight due to dehydration from heat and exercise can result in significant reduction in visual-motor tracking (Lieberman, 2007). In an experimental study assessing performance in complex motor tasks in hyperthermic humans (Piil et al., 2017), the authors found that visual-motor tracking performance was reduced following exercise-induced hyperthermia. Participants were exposed to hot (40 °C (104 °F)) and control (20 °C (68 °F)) conditions. At baseline, and after exercise, participants completed simple and complex motor tasks, which included visual tracking assessment. The authors concluded that visual-motor tracking is impaired by hyperthermia, and especially so when multiple tasks are combined (Piil et al., 2017).

Heat exposure has been found to affect decision-making or judgment amongst workers, increasing the risk of injury. In a review of ecological, cross-sectional, and cohort studies, as well as epidemiological experiments, Xiang, et al. indicate that core temperature elevation and dehydration impair judgment and concentration (Xiang, et al., 2014a). In a study analyzing over 17,000 observations of unsafe behavioral acts ( e.g. mishandling tools, equipment, or materials) in two industrial facilities with varying temperature conditions, authors found that unsafe behavioral acts decreased within the zone of preferred temperature (approximately 17 °C (62.6 °F) to 23 °C (73.4 °F), WBGT) and increased outside of this zone (when the temperature was equal to or less than 17 °C WBGT or equal to or greater than 23 °C WBGT) (Ramsey et al., 1983). This study indicates that the risk of unsafe behavioral acts may increase when the temperature increases.

In addition to psychomotor and mental impairments that can result from heat exposure, other mechanisms may also contribute to heat-related injuries. The purpose of this section is to summarize some additional factors that may exacerbate the risk of workplace heat-related injuries and to provide information to better inform workers and employers about those hazards.

PPE is another factor that plays a role in increasing susceptibility to a heat-related injury given that some PPE insolates the body and reduces evaporative cooling capacity. For instance, research among firefighters finds that a self-contained breathing apparatus can lead to heat buildup and can impact postural stability and balance (Hur et al., 2015; Hur et al., 2013; Games et al., 2020; Mani et al., 2013; Ross, 2016). Other examples of PPE that may result in heat stress, and therefore increase the risk of heat-related injuries, include reflective vests that are made of water impermeable material that block effective heat dissipation and safety helmets with no ventilation that can raise the temperature inside the helmet. In one case, the air temperature inside a worker's helmet (57 °C (134.6 °F)) was measured to be over 20 °C hotter than the environmental temperature (33 °C (91.4 °F)) they were working in (Rowlinson et al., 2014). The authors found that workers will often remove helmets in these situations to alleviate heat stress, exposing them to other workplace hazards ( e.g., falling objects) (Rowlinson et al., 2014). Other research by Karthick et al. (2023) found that in hot weather conditions, physical health challenges, specifically major accidents at the job site, minor injuries, physical fatigue, excessive sweating, and dermatological problems were found to be significant based on a workers' clothing comfort. The authors highlighted how PPE can make workers feel uncomfortable, and when combined with extremely hot weather, it creates fatigue which may increase the number of workplace injuries and accidents (Karthick et al., 2023).

There is also evidence indicating heat exposure can contribute to impaired vision, which may lead to workplace injuries. For example, fogged safety glasses or sweat in eyes due to heat exposure can reduce workers' visibility, creating additional hazards and increasing risk of injury (NIOSH, 2016). Individual case studies also report issues with protective eyewear in hot temperatures, noting the uncomfortable feeling of the eyewear under heat and in sunlight as well as difficulty seeing through the glasses (Choudhry and Fang, 2008). In a survey conducted among occupational health and safety professionals in Australia, one of the most frequently cited causes of heat- ( print page 70732) related injuries was from “impaired vision due to fogged safety glasses (39%)” (Varghese et al., 2020). Injuries resulting from impaired vision may include manual handling (musculoskeletal injuries), joint/ligament injuries, hand injuries, wounds or lacerations, burns, head or neck injuries, motor vehicle accidents, eye injuries, or fractures (Varghese et al., 2020).

When exposed to heat, workers may also experience impaired dexterity (or fine motor skills) leading to workplace injuries. For example, sweaty palms and hands due to heat exposure can reduce workers' ability to handle tools or other work-related materials, increasing the risk of injury. Occupational health and safety professionals have reported losing control of tools as one of the most common causes for heat-related injuries (Varghese et al., 2020). Researchers have also found sweaty palms to increase the risk of workplace injuries (Shulte et al., 2016).

The scientific and mechanistic data and association studies on heat-related injuries summarized in this section demonstrate that heat-related injuries are a recognized health effect of occupational heat exposure. While the types of heat-related injuries can be broad, the scientific community recognizes that heat exposure can diminish the body's senses through various mechanisms like impaired psychomotor performance ( e.g., fatigue, impaired balance, or impaired dexterity), and impaired mental performance ( e.g., impaired cognition or vision) which can result in various types of injuries. The best available evidence demonstrates that heat-related injuries can have serious adverse effects on worker safety and health.

OSHA requests information and comments on the following question and requests that stakeholders provide any relevant data, information, or additional studies (or citations) supporting their view, and explain the reasoning for including such studies:

• Has OSHA adequately identified and documented the studies and other information relevant to its conclusions regarding heat-related health effects, and are there additional studies OSHA should consider?

In this risk assessment, OSHA relied on surveillance data of occupational heat-related fatalities and non-fatal injuries and illnesses reported by the Bureau of Labor Statistics (BLS). Additionally, OSHA relied on annual incidence estimates derived from State workers' compensation systems and hospital discharge datasets. These estimates were calculated and reported in a variety of sources, such as reports from State health departments, as well as the peer-reviewed scientific literature. OSHA has preliminarily concluded that inclusion criteria for HRIs in these data sources (days away from work, workers' compensation claim, emergency department visit, or inpatient hospitalization) demonstrate that the HRIs are a material impairment of health, thus making these data sources relevant to OSHA's determination of significant risk.

OSHA has previously relied on such injury, illness, and death data to demonstrate the extent of risk (see, e.g., Fall Protection, 81 FR 82494 (2016); Working Conditions in Shipyards, 76 FR 24576 (2011); Permit-Required Confined Spaces, 58 FR 4462 , 4465 (1993) (finding significant risk based on available accident data showing that confined space hazards had caused deaths and injuries); Hazard Communication, 48 FR 53280 , 53284-85 , 53321 (1983) (finding significant risk of harm from inadequate chemical hazard communication based on BLS chemical source injury and illness data)).

Estimating annual incidence among heat-exposed workers ( i.e., the number of annual work-related HRIs divided by the number of heat-exposed workers) requires being able to accurately estimate the number of exposed workers and using that number in the denominator. Unfortunately, there is no published estimate for the number of U.S. workers exposed to hazardous heat on the job and the majority of the incidence estimates that OSHA identified used a denominator that would include both exposed and unexposed workers. This use of a larger denominator has the effect of diluting the resulting annual incidence estimates. For instance, BLS estimates and reports annual incidence of injuries and illnesses involving days away from work that were the result of “exposure to environmental heat,” but in their calculation, BLS captures the broader U.S. workforce in the denominator, which includes a large number of unexposed workers ( e.g., office workers in climate-controlled buildings).

Some of the annual incidence estimates that OSHA identified, such as those based on workers' compensation claims in California and Washington State, were stratified by sector, industry, or occupation. OSHA considers these incidence estimates to be helpful in getting to a more accurate estimate of risk among heat-exposed workers, specifically the sectors, industries, and occupations where exposure to hazardous heat on the job is more common. Furthermore, OSHA identified incidence estimates from cohort data in which the entire cohort was presumed to be exposed to hazardous heat on the job. These estimates are much higher than the estimates based on surveillance data. One potential reason for this difference is that the denominator used in the cohort studies contains much less unexposed worker-time.

In the following sections (V.A.II., and V.A.III.), OSHA has summarized the best available incidence data that the agency identified. Given the limitations with these data, OSHA relied on this incidence data as a range of possible incidence estimates with the assumption that many of these estimates represent a lower bound and that the true incidence is likely higher.

The BLS Survey of Occupational Injuries and Illnesses (SOII) is the primary nationwide source of surveillance data for nonfatal occupational injuries and illnesses. The scope includes both private and public (State and local government) sector employees, but excludes the self-employed, workers on farms with 10 or fewer employees, private household workers, volunteers, and Federal Government employees. The data are derived from a two-stage sampling process, during which a sample of employers are surveyed and report to BLS the number of injuries and illnesses occurring at their workplace. To reduce the reporting burden on employers, BLS only requires detailed case information on a sample of the injuries and illnesses that occurred at each establishment. BLS uses these survey responses to estimate the counts and incidence for nonfatal injuries and illnesses across all workplaces. In estimating annual incidence, BLS uses a denominator of full-time equivalent (FTE) workers, ( print page 70733) which is based on 2,000 hours worked per year ( i.e., 40 hours per week over 50 weeks). Relevant Occupational Injury and Illness Classification System (OIICS) v2.01 event and nature codes for this proposed standard include “Exposure to environmental heat” (event code-531) and “Effects of heat and light” (nature codes beginning in 172-). Codes beginning with 172- include heat stroke and heat exhaustion (among other outcomes) but exclude sunburn and loss of consciousness without reference to heat. For more information about OIICS codes generally, see Section IV., Health Effects.

Between 2011 and 2020, there were an estimated 33,890 work-related injuries and illnesses that involved days away from work that were coded with event code 531, for an annual average of 3,389 such injuries and illnesses during this period (BLS 2023b). In 2023, BLS reported biennial rather than annual estimates for work-related injuries and illnesses that involved days away from work (as well as for the first time reporting an estimate of injuries and illnesses involving job restriction or job transfer). The biennial estimate for 2021-2022 for heat-related cases meeting either of these criteria was 6,550 (5,560 cases involved days away from work; 990 cases involved job transfer or restriction) (BLS 2023g). The estimated annual heat-related injury and illness incidence (for cases involving days away from work) calculated by BLS for all workers covered by SOII from 2011-2020 varied by year but ranged from 2.0/100,000 workers to 4.0/100,000 workers. The average estimated annual incidence for the entire time period was 3.0/100,000 workers. However, as stated above, OSHA considers these incidence estimates to be underestimated for heat-exposed workers because BLS calculates the incidence rate for the entire U.S. workforce covered by SOII. Therefore, they are including workers who are not exposed to hazardous heat. In subsectors and industries where OSHA expects a greater proportion of workers to be exposed to hazardous heat, the incidence rate estimates are much higher. For instance, according to unpublished data from BLS SOII for the period 2011-2020, the crop production subsector (NAICS code 111) had an annual average incidence of 14.2/100,000 workers, and the specialty trade contractors subsector (NAICS code 238) had an annual average of 9.3/100,000 workers. This was also true of subsectors with primarily indoor workers where OSHA expects a greater proportion of those workers to be exposed to hazardous heat, including the primary metal manufacturing subsector (NAICS code 331), which had an annual average incidence of 13.1/100,000 workers for the period 2011-2020.

Workers' compensation claims are an alternative way to quantify occupational injuries and illnesses, particularly those that involve outpatient medical treatment, inpatient hospitalization, intensive care, and/or lost workdays. OSHA identified five papers and a report from Wisconsin that have evaluated State workers' compensation data and calculated statewide incidence for heat-related injuries and illnesses.

The earliest of these, a paper by Bonauto et al., in 2007, evaluated workers' compensation claims submitted to and accepted by the Washington State Fund between 1995 and 2005 (Bonauto et al., 2007). The State Fund is the sole provider of workers' compensation insurance to Washington employers unless they are self-insured or fall under an alternative system ( e.g., Federal employees) and it covers approximately two-thirds of the State's workers. Certain workers are exempt from mandatory coverage, such as self-employed and household workers. The authors identified heat-related cases using the American National Standards Institute (ANSI) Z16.2 codes  [ 2 ] submitted in the claims by workers or their physicians, the ICD-9 codes submitted on bills from healthcare providers and hospitals, and a physician review of cases that included relevant Z16.2 or ICD-9 codes. The researchers used all ICD-9 codes beginning in 992 (“Effects of heat and light,” specifically 992.0-992.9) and the ANSI Z16.2 type code 151 (“Contact with general heat—atmosphere or environment”). ICD-9 codes were not available for claims from the self-insured, so the authors restricted the analysis to State Fund claims only. They also excluded claims in which the employer's physical location was outside of Washington (n=12).

Over the 11-year study period, 480 accepted claims met the authors' inclusion criteria after physician review, in which they identified and removed cases where the recorded illness had been miscoded, contained incorrect data, or represented a burn. Most of the 480 claims (n=442; 92.1%) were medical-only claims, meaning the State Fund only paid for the medical bills and did not compensate the worker otherwise ( e.g., wage replacement, disability benefits). The claims included the employer's NAICS code, which the authors used to stratify cases by industry sectors and industries. Employers covered under the Washington State Fund are required to report hours worked by their employees every quarter ( i.e., three-month increments), which the authors used to estimate denominators for rates assuming 2,000 work hours is 1 FTE. This means the authors could calculate rates for certain portions of the year rather than the whole year without needing to divide by the total number of annual workers ( i.e., they could adjust for hours worked only during the specified portion). The employment reporting by quarter also allowed for the authors to estimate claim rates for the third quarter only (July, August, and September), which corresponded to the time of year with the “greatest level of exposure to elevated environmental temperatures” (Bonauto et al., 2007, p. 5).

The authors reported an average annual claim rate (which can be thought of similarly to an injury or illness incidence rate) of 3.1 claims/100,000 FTE for the overall workforce covered by the State Fund during the study period, with annual rates ranging from 1.9 to 5.1/100,000 FTE. They reported a corresponding average third-quarter claim rate of 8.6 claims/100,000 FTE for the overall workforce covered by the State Fund during the study period. In their paper, Bonauto et al. report annual and third-quarter rates for all sectors and industries that had more than five claims during the study period. The sectors (2-digit NAICS) with the highest annual average claim rates were:

1. Construction (12.1/100,000 FTE),

2. Public administration (12.0/100,000 FTE),

3. Agriculture, forestry, fishing, and hunting (5.2/100,000 FTE),

4. Administrative and support and waste management and remediation services (3.9/100,000 FTE), and

5. Transportation and warehousing (3.5/100,000 FTE).

The corresponding average third-quarter claim rates for these sectors were more than double the annual averages: 33.8/100,000 FTE, 31.2/100,000 FTE, 12.6/100,000 FTE, 9.9/100,000 FTE, and 10.6/100,000 FTE, respectively. This pattern was also true ( print page 70734) for some sectors with a majority of indoor claims. For example, Manufacturing (3.0/100,000 FTE vs. 7.6/100,000 FTE) and Accommodation and food services (1.7/100,000 FTE vs. 5.1/100,000 FTE).

The industries (6-digit NAICS) with the highest annual average claim rates were:

1. Fire protection (80.8/100,000 FTE),

2. Roofing construction (59.0/100,000 FTE),

3. Highway, street and bridge construction (44.8/100,000 FTE),

4. Site preparation construction (35.9/100,000 FTE) (tie), and

5. Poured concrete foundation and structural construction (35.9/100,000 FTE) (tie).

Similar to the pattern observed among sectors, the corresponding third-quarter claim rates for the top 5 industries were more than double the annual averages, except for fire protection—158.8/100,000 FTE, 161.2/100,000, 105.6/100,000 FTE, 106.5/100,000 FTE, and 102.6/100,000 FTE, respectively. This was also true for restaurants: limited service restaurants (2.4/100,000 FTE vs. 6.0/100,000 FTE) and full service restaurants (1.6/100,000 FTE vs. 5.3/100,000 FTE). These industries have few to no outdoor claims, indicating that even some industries that involve primarily indoor work are at higher risk in the summer months.

A follow-up paper to Bonauto et al., 2007, published in 2014, examined heat-related illnesses among workers in Washington State in certain agriculture and forestry subsectors between 1995 and 2009 (Spector et al., 2014). The State changed their injury and illness codes from ANSI to OIICS in July 2005, so for this paper, the researchers used a combination of ANSI (prior to July 2005), OIICS (beginning in July 2005), and ICD-9 codes to identify potential heat-related claims and then reviewed each claim to ensure it was heat-related. These authors used additional ICD-9 codes that were not included in the 2007 paper, specifically: prickly heat (705.1), hyperosmolality and/or hypernatremia (276.0), volume depletion (276.5 and 276.50), dehydration (276.51), hypovolemia (276.52), and acute renal failure (584 and 584.9). The authors identified 84 accepted claims meeting their eligibility criteria, the majority of which (n=76; 90%) were medical only claims. Of the 84 claims, 61 (73%) met the diagnostic code criteria used in the 2007 paper (ICD-9 codes beginning in 992). The average annual claim rate for the agriculture and forestry subsectors the authors examined over the 15-year period was 7.0/100,000 FTE and the average third-quarter (July-September) claim rate was 15.7/100,000 FTE. The majority of claims (61%) were among crop production and support workers (NAICS 111 or 1151).

A second follow-up paper to Bonauto et al., 2007, was published in 2020 and included all Washington State Fund-covered workers over a more recent 12-year period, 2006 to 2017 (Hesketh et al., 2020). The authors used similar methods, except for different screening criteria for ascertaining cases prior to investigators reviewing each case. To identify potential heat-related claims, they used OIICS v1.01 event/exposure code 321, OIICS nature code 072*, OIICS source codes 9362 and 9392 (Sun), and the ICD-9 codes used in Spector et al., 2014. (Note that these OIICS codes are v1.01 OIICS, which was the coding scheme used from 1992-2010. BLS updated the coding scheme in 2010, which first applied to 2011 data.) The State adopted ICD-10 coding in October 2015, so the following ICD-10 codes were used for claims after that date: E86* (Volume depletion), T67* (Effects of heat and light), T73.2* (Exhaustion due to exposure), W92* (Exposure to excessive heat of man-made origin), X30* (Exposure to excessive natural heat), and Z57.6 (Occupational exposure to extreme temperature). The researchers excluded claims in which service date for treatment of dehydration or kidney failure was not within one day of the illness date or claims in which dehydration or kidney failure were the only identifiers flagged, as they noted that these cases often did not represent heat-related illnesses.

The authors reported a total of 918 confirmed heat-related claims, of which 654 (71%) were accepted claims. Of the accepted claims, 595 (91%) were medical-only claims. Using only accepted claims, they estimated an average annual claim rate of 3.2 claims/100,000 FTE for the overall workforce covered by the State Fund during the study period (Communication with David Bonauto and June Spector, June 2024). Similar to Bonauto et al., 2007, the authors reported claim rates for all sectors and industries with more than 11 claims. The sectors (2-digit NAICS) with the highest annual average accepted claim rates were:

1. Agriculture, forestry, fishing, and hunting (13.0/100,000 FTE),

2. Construction (10.8/100,000 FTE),

3. Public administration (10.3/100,000 FTE),

4. Administrative and support and waste management and remediation services (4.6/100,000 FTE), and

5. Transportation and Warehousing (3.8/100,000 FTE).

The average third-quarter (July-September) claim rates for some sectors were more than 10 times greater than the average annual rates. These third-quarter claim rates were also much higher than those calculated for 1995-2005 in Bonauto et al., 2007. The sectors with the highest average third-quarter accepted claim rates were:

1. Public administration (131.3/100,000 FTE),

2. Agriculture, forestry, fishing, and hunting (102.6/100,000 FTE),

3. Construction (70.0/100,000 FTE),

4. Administrative and support and waste management and remediation services (61.5/100,000 FTE), and

5. Wholesale trade (44.9/100,000 FTE).

The industries (6-digit NAICS) with the highest annual average accepted claims rates were:

1. Farm labor contractors and crew leaders (77.3/100,000 FTE),

2. Fire protection (60.0/100,000 FTE),

3. Structural steel and precast concrete contractors (54.2/100,000 FTE),

4. Poured concrete foundation and structure contractors (31.6/100,000 FTE), and

5. Roofing contractors (29.0/100,000 FTE).

The ratio between third-quarter rates and annual rates for all industries reported in table 3 of the paper ranged from 2.5-13.7, with the highest average third-quarter accepted claim rates in the following industries:

1. Farm labor contractors and crew leaders (600.9/100,000 FTE),

2. Fire protection (394.6/100,000 FTE),

3. Administration of conservation programs (282.7/100,000 FTE),

4. Site preparation contractors (232.1/100,000 FTE), and

5. Poured concrete foundation and structure contractors (172.3/100,000 FTE).

A group of researchers conducted a similar analysis for the State of California, using data from the California Workers' Compensation Information System (WCIS) between 2000 and 2017 (Heinzerling et al., 2020). Virtually all California employees are required to be covered by workers' compensation; voluntary, non-compensated workers, owners, and workers covered under separate programs are excluded. The WCIS contains all accepted and rejected workers' compensation claims in the State since 2000 that required medical treatment beyond first aid or more than ( print page 70735) one day of lost work time. The investigators identified heat-related claims in the system using WCIS-specific nature of injury and cause of injury codes ( e.g., “temperature extremes”), heat-related illness keywords ( e.g., “heat stroke”), and certain ICD-9 (992.0-992.9 and E900.0-E900.9) and ICD-10 (T67.0-T67.9, X30, and W92) codes. They also manually reviewed all claims that met only the ICD code identification criteria to ensure the claims were heat-related, as some of the codes they used to identify claims were not specific to heat-related illness or injury. In WCIS, the employer's industry is coded using NAICS codes classified by the claims adjusters. The authors converted the NAICS codes into the appropriate 2002 census industry codes using the NIOSH Industry and Occupation Computerized Coding System (NIOCCS). This was necessary to obtain the corresponding employment denominator estimates from the NIOSH Employed Labor Force Tool, which relies on data from the Current Population Survey (CPS), a Census Bureau survey conducted for BLS. The CPS data provide estimates of all employed and non-institutionalized civilian workers over the age of 15. To account for changes in coding schemes implemented in 2002, the investigators extrapolated 2002-2017 data to estimate denominators for 2000 and 2001.

The authors excluded claims for workers below 16 years of age (n=104 claims) and institutionalized workers (n=455 claims), as these workers are excluded from CPS data. They reported a final estimate of 15,996 claims meeting their inclusion criteria, corresponding to an overall annual claims rate of 6.0/100,000 workers. Industry and occupation codes were available for 86% and 74% of the included claims, respectively. The authors reported claim rates for all sectors, but the sectors with the highest annual claim rates were:

1. Agriculture, forestry, fishing, and hunting (38.6/100,000 workers; 95% CI: 26.9, 40.4),

2. Public administration (35.3/100,000 workers; 95% CI: 34.3, 36.3),

3. Mining (21.3/100,000 workers; 95% CI: 17.6, 25.7),

4. Utilities (11.4/100,000 workers; 95% CI: 10.1, 12.8), and

5. Administrative and support and waste management (8.8/100,000 workers; 95% CI: 8.3, 9.3).

The major occupational groups with the highest annual claim rates were:

1. Protective services (56.7/100,000 workers; 95% CI: 54.9, 58.7),

2. Farming, fishing, and forestry (35.9/100,000 workers; 95% CI: 34.1, 37.9),

3. Material moving (12.3/100,000 workers; 95% CI: 11.5, 13.1),

4. Construction and extraction (8.9/100,000 workers; 95% CI: 8.4, 9.4), and

5. Building and grounds cleaning and maintenance (6.0/100,000 workers; 95% CI: 5.6, 6.5).

Another study examined workers' compensation claims in an unnamed, mid-sized Texas city before and after an intervention among a cohort of 604 municipal workers and calculated the incidence of HRI claims from 2009 to 2017 (McCarthy et al., 2019). The municipal departments included in the study were picked because the job descriptions for workers within each included work in hot environments with moderate and heavy physical activity. These departments were Streets and Traffic, Parks and Recreation, Utilities, and Solid Waste. After removing worker-time contributed by administrative personnel who were not exposed to heat on the job, the remaining worker-time represented 329 FTEs per year. Prior to the intervention in 2011, the heat-exposed workers experienced 17 total HRIs between 2009 and 2010. The authors reported an average annual rate of HRIs among the heat-exposed workers during this time of 25.5/1,000 FTEs (McCarthy et al., 2019, Figure 2). These estimates are much higher than other incidence estimates reported in this section, possibly because the denominator is solely comprised of heat-exposed workers. This explanation is supported by evidence of higher incidences reported in other cohort studies ( e.g., approximately 3 HRIs/1,000 National Guard troops involved in flood relief activities between July 5 and August 18, 1993, calculated from data in Dellinger et al., 1996). The results of the voluntary intervention are discussed in Section V.C., Risk Reduction.

Finally, a report issued by the Wisconsin Occupational Health and Safety Surveillance Program in 2024 summarized an analysis of heat-related workers' compensation claims in the State from 2010-2022 (Fall et al., 2024). The authors analyzed lost work time claims (under Wisconsin workers' compensation, there must be more than three days of lost work time to be compensable) reported by both insurance carriers and self-insured employers and reported rates by industry sector and industry subsector (rather than overall workforce rates). These do not include medical-only claims, which were the majority of HRI claims reported in the Washington State Fund database. The authors reported cumulative claim rates only. To convert cumulative rates to annual average rates, OSHA divided the reported rates by 13 (the number of years' worth of data reported). The sectors with the highest annual average claim rates were:

1. Administrative and Support and Waste Management and Remediation Services (2.9/100,000 FTE),

2. Public Administration (2.8/100,000 FTE),

3. Wholesale Trade (1.9/100,000 FTE),

4. Construction (1.4/100,000 FTE), and

5. Transportation and Warehousing (1.1/100,000 FTE).

The major occupational groups with the highest annual average claims rates were:

1. Protective Service (4.1/100,000 FTE),

2. Transportation and Material Moving (2.6/100,000 FTE),

3. Production (1.6/100,000 FTE),

4. Construction and Extraction (1.5/100,000 FTE), and

5. Building and Grounds Cleaning and Maintenance (1.5/100,000 FTE).

Similarly, the minor occupational groups with the highest annual average claims rates were:

1. Fire Fighting and Prevention (14.7/100,000 FTE),

2. Material Moving Workers (3.3/100,000 FTE),

3. Metal and Plastic Workers (2.8/100,000 FTE),

4. Motor Vehicle Operations (2.2/100,000 FTE), and

5. Assemblers and Fabricators (2.2/100,000 FTE).

Another way to quantify occupational injury and illnesses requiring medical treatment is to use data reported directly by hospitals to public health departments or national databases, such as the National Electronic Injury Surveillance System (NEISS). Data in NEISS are estimated from a nationally representative probability sample of hospitals across the country, which report data for every injury-related ED visit. A paper from 2010 analyzed NEISS data for heat-related emergency department visits from 2001-2004 (Sanchez et al., 2010). The authors reported an annual average of 8,376 work-related ED visits for nonfatal heat injuries and illnesses. OSHA used annual average employment estimates from NIOSH's Employed Labor Force query system for 2001-2004 (both total workers and FTEs) to estimate a nationwide annual average rate of 6.1 ( print page 70736) visits/100,000 workers and 6.3 visits/100,000 FTEs from this study. More recent studies estimating the incidence of work-related ED visits and/or hospitalizations for HRIs within individual or multiple States are discussed below.

A group of public health researchers from nine States in the Southeast (Florida, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, South Carolina, Tennessee, and Virginia) used hospital discharge data reported directly to State health departments to characterize rates of heat-related inpatient hospitalization and ED visits among workers from 2007—2011 (Harduar Morano et al., 2015). The researchers used ICD-9 codes to identify heat-related cases, specifically 992.0-992.9, E900.0, E900.1, and E900.9. To assess work-relatedness, they determined whether the expected payer was workers' compensation or if a work-related external cause of injury code (sometimes referred to as E-codes) was noted by the physician ( e.g., E000.0 Civilian activity done for income). They restricted cases only to those where the patient was at least 16 years old but included both State residents and non-residents in reported case counts. To calculate rates, the investigators used CPS data for estimating denominators, which were age-adjusted using direct standardization and population weights for the entire U.S. Non-residents were not included in the rate calculations. The authors noted that hospital discharge data weren't available for every year in every State and that the missing data were primarily for discharges following ED visits.

Across the five-year study period, the authors identified 8,315 occupational heat-related ED visits (7,664 of these among residents, or 92%), which corresponded to an overall age-adjusted rate of 6.5 visits/100,000 workers (95% confidence interval, CI = 6.4, 6.7). While they reported rates for each State ( e.g., 4.8 visits/100,000 workers in Florida and 17.3 visits/100,000 workers in Louisiana), they cautioned against directly comparing between States given differences in the data collection methods, data availability, and use of work-related variables. They identified 1,051 occupational heat-related inpatient hospitalizations (930 among residents, or 88%), which corresponded to an overall age-adjusted rate of 0.61 hospitalizations/100,000 workers (95% CI = 0.58, 0.66). The average length of stay for State residents was 2.7 days, which was comparable to non-residents (2.4 days).

The Florida Department of Health published a similar analysis in 2011 using the same methods for the State of Florida for the years 2005—2009 (Florida DOH, 2011). They identified 2,198 occupational heat-related hospitalizations and ED visits, which corresponded to an average overall age-adjusted annual rate of 3.7 cases/100,000 workers (95% CI = 1.9, 5.5) and a crude rate (no age adjustment) of 5.1/100,000 workers (Communication with Laurel Harduar Morano, October 2023). The majority of these (89.4%) were ED visits. They identified 3 fatalities in this subset, which they noted corresponds to a case fatality rate of 1.4 fatalities/1,000 cases. They reported a third-quarter (July, August, and September) rate of 3.2 cases/100,000 workers using a denominator of total number of workers, whereas using a denominator of FTEs instead produced a third-quarter rate of 13.0 cases/100,000 FTE (Communication with Laurel Harduar Morano, October 2023). A 2016 study conducted a more in-depth analysis of the statewide Florida hospitalization data and included data for three additional years (2010, 2011, and 2012) (Harduar Morano et al., 2016). The authors restricted the data to cases occurring in May-October of each year and identified a total of 2,979 work-related ED visits and 415 work-related hospitalizations between 2005-2012. Using total number of workers in the denominator (calculated from monthly CPS data), these corresponded to average annual age-adjusted rates of 8.5 ED visits/100,000 workers and 1.1 hospitalizations/100,000 workers.

In March 2023, the Louisiana Department of Health published a report on heat-related illnesses in the State using ED and hospitalization data from 2010-2020 (Louisiana DOH 2023). The authors used workers' compensation as payer and work-related ICD codes to determine which cases were among workers. They reported an annual average of 320 work-related ED visits and 20 work-related hospitalizations for heat-related illness during this period. Using State employment data from CPS, the authors calculated an overall age-adjusted rate of 15.1 work-related ED visits/100,000 workers and 0.9 work-related hospitalizations/100,000 workers. In 2024, the Department of Health released a syndromic surveillance report on ED visits for HRIs between April 1 and October 31, 2023 (Louisiana DOH 2024). They identified 1,412 ED visits for HRIs among workers during this time period.

Since 2013 over 20 States have reported rates of heat-related ED visits among workers to the Council of State and Territorial Epidemiologists (CSTE), comprising the organization's Occupational Health Indicator #24 (see www.cste.org/​page/​ohindicatorstable ). These data are compiled by the State health departments using workers' compensation as primary payer and external cause of injury codes to determine work-relatedness. Rates are calculated using CPS estimates of total employed persons by State. While multiple States report their annual rates to CSTE, the organization cautions against directly comparing these rates between States because “workers' compensation eligibility criteria and availability of data from workers' compensation programs varies among states, prohibiting state-level data from being directly compared to other states or with national estimates.”

Additionally, given that these data are not available for every State, they cannot be combined to produce an accurate national rate. The State-reported rates are currently available for 2013-2019. During this period, the annual rates for heat-related ED visits ranged from 0.1 to 18.7 ED visits per 100,000 workers.

Arizona is not one of the States to share their ED visit data to CSTE, but the most populated county in the State—Maricopa County—has published a Heat Morbidity Report in which they provide case counts for heat-related hospitalization discharges, including a breakdown of the “preceding activity type” (determined by ICD activity E-codes) (Maricopa County Public Health Department, n.d.). Using the case counts reported under “occupational” activity type and yearly estimates of the average annual employment for Maricopa County provided by the BLS Quarterly Census of Employment and Wages, there was an average annual hospitalization rate among workers of 4.1 cases/100,000 workers (range: 3.1-6.4/100,000) between 2010-2017. Primary payer of workers' compensation was not used to determine work-relatedness, which means some occupational cases not involving E-codes may have been missed. Given that for the majority of cases (77%-83% per year), the preceding activity was marked as “unknown”, it's likely that some number of these were occupational in nature and just not listed as such. This ( print page 70737) is supported by the fact that an “Industrial Site” was the place of injury for, on average, 8% of cases, which may also be an underestimate. It should be noted that the authors only used the following ICD-9/ICD-10 activity E-codes to determine work-relatedness: E011/Y93.C Activities involving computer technology and electronic devices; E012/Y93.D Activities involving arts and handcrafts; and E016/Y93.H Activities involving exterior property and land maintenance, building and construction. To OSHA's knowledge, the authors did not use any other external cause of injury codes, such as E000.0 Civilian activity done for income, but it is not clear from the report if these E-codes were not available or were just not used.

As discussed in Section IV.P., Heat Related Injuries, one area of research has used the natural fluctuations in temperatures to conduct quasi-experimental studies examining the relationship between heat and workers' compensation claims for traumatic injuries ( e.g., Spector et al., 2016; Calkins et al., 2019; Dillender 2021; Park et al., 2021). The findings of these papers suggest that there may be many workers' compensation claims that are heat-related but not coded as such. For instance, Park, Pankratz, and Behrer (2021) estimated that approximately 20,000 injuries per year in California between 2001-2018 resulted from hotter temperatures (relative to “optimal” temperature). For comparison, for a similar time period (2000-2017), Heinzerling et al. (2020) only identified an average of 889 HRI workers' compensation claims per year in California (a 22-fold difference), suggesting that relying on workers' compensation claims coded as HRIs alone does not capture the higher incidence of injuries of other kinds where heat may have played a role. A research report from the Workers Compensation Research Institute expanded this type of analysis to 24 States, using a convenience sample of workers' compensation claims from May-October 2016-2021 (Negrusa et al., 2024). They found that the number of injuries increased 3.2-6.1% when the daily maximum temperature was 75 °F or higher relative to a day with a daily maximum temperature of 65-70 °F. This relationship was even more pronounced for the construction industry.

Another source of incidence data is surveys of workers exposed to heat. Multiple papers describe the results of surveys of outdoor workers, typically agricultural workers, who are asked about heat-related symptoms experienced over a week-long period while working in the summer months (Fleischer et al., 2013; Kearney et al., 2016; Mutic et al., 2018). Commonly reported symptoms in these studies include heavy sweating (38-66% of surveyed workers), headache (44-58%), muscle cramps (30-36%), dizziness (14-32%), weakness or fatigue (18%), and nausea or vomiting (9-17%). Notably, in two of these studies, multiple workers reported fainting on the job. A study in southern Georgia found that 4% of 405 farmworkers experienced fainting within the previous week, during which the heat index ranged from 100-108 °F (Fleischer et al., 2013). Another study involved asking 281 farmworkers in North Carolina if they had ever worked in “extreme heat.” Of those answering “yes”, 3% reported having ever fainted on the job (Mirabelli et al., 2010). When asked about symptoms over a single workday, a separate study found that 25% of workers reported cramps, 22% headache, 10% dizziness, and 3% nausea (Smith et al., 2021).

OSHA identified multiple sources that have reported annual incidence estimates for nonfatal HRIs among workers. These studies and reports generally reported heat-related incidence across an entire workforce (either National or State), using the total workforce as the denominator. This would understate the risk to workers who are actually exposed to heat on the job since the denominator includes a large percentage of workers who are not exposed to heat ( e.g., office workers). Evidence in support of this claim comes from studies showing higher incidence of HRI when populations are stratified by sector, industry, or occupation, as well as those reporting incidence that occurred only during the third quarter (July, August, and September). For instance, in Heinzerling et al., 2020, the authors report an overall annual incidence of 6.0/100,000 workers whereas they report an annual incidence of 38.6/100,000 workers for workers in the agriculture, forestry, fishing, and hunting sector (a greater than 6-fold difference). OSHA considers these stratified estimates to be more accurate estimates of the “true” incidence of HRIs among heat-exposed workers.

A summary of the annual incidence estimates for nonfatal occupational HRIs discussed above can be found in table V-1. In the same table, OSHA calculated the number of non-fatal HRIs that would be expected over a working lifetime (assuming a working lifetime is 45 years long) based on those annual incidence estimates ( i.e., the annual incidence multiplied by 45). These estimates represent the total number of HRIs that may be expected to occur in a cohort of 100,000 workers all of whom enter the workforce at the same time and all of whom work for 45 years. Estimates of HRI risk over a working lifetime based on annual incidence among entire working populations (National or State) range from 90-180/100,000 for HRIs requiring days away from work, 140-270/100,000 for HRIs leading to a workers' compensation claim, and 4.5-842/100,000 for HRIs leading to emergency department visits or inpatient hospitalizations. Like incidence estimates, these values understate the risk to workers who are actually exposed to heat on the job since the denominator includes a large percentage of workers who are not exposed to heat ( e.g., office workers). However, when using incidence estimates specific to individual sectors, industries, or occupations, the HRI estimates over a working lifetime are much higher, ranging from 49.5-114,750/100,000 for HRIs leading to a workers' compensation claim.

The BLS Census of Fatal Occupational Injuries (CFOI), established in 1992, is the primary source of surveillance data on work-related fatalities, including fatalities due to environmental heat exposure, for the United States. The fatality data in CFOI come from diverse data sources to identify, verify, and describe work-related fatalities. In each case, at least two sources ( e.g., death certificates, workers' compensation reports, media reports, and government agency administrative reports) and an average of four are used to validate that the fatality was work-related and to verify the event or exposure leading to death and the nature of injury or illness in each case, which are then classified with OIICS codes. Heat-related fatalities can be identified with an event code (“Exposure to environmental heat”) and/or a nature code (“Effects of heat and light”).

According to BLS's CFOI, occupational heat exposure killed 1,042 U.S. workers between 1992 and 2022 (BLS, 2024c). Between 2011 and 2022, BLS reports 479 worker deaths, an average of 40 fatalities per year during that time. During the latest three years ( print page 70738) for which BLS reports data (2020-2022), there was an average of 45 work-related deaths due to exposure to environmental heat per year. Multiple sources have relied on BLS surveillance data to estimate annual incidence rates of occupational heat-related fatalities.

Gubernot et al. (2015) calculated overall fatality rates and fatality rates by industry sector using BLS CFOI data from 2000-2010 (Gubernot et al., 2015). The authors focused on the three industry sectors with the highest rates in preliminary analyses: Agriculture, Forestry, Fishing and Hunting (NAICS code 11); Construction (NAICS code 23); and Administrative and Support and Waste Management and Remediation Services (NAICS code 56). All other industry sectors were combined for comparison as a referent group. The authors used nationwide worker population data from the CPS to estimate fatality rates. The CPS data provide estimates of all employed and non-institutionalized civilian workers over the age of 15.

The authors identified 339 occupational heat-related deaths from 2000-2010, after excluding volunteers and military personnel. They reported an average annual heat-related fatality rate of 0.022 fatalities per 100,000 workers for the overall workforce.

For the three industry sectors preliminarily identified as having the highest rates, the authors reported the following average annual fatality rates:

1. Agriculture, forestry, fishing and hunting (0.306 fatalities per 100,000 workers),

2. Construction (0.113 fatalities per 100,000 workers), and

3. Administrative and Support and Waste Management and Remediation Services (0.056 fatalities per 100,000 workers).

For all other industry sectors combined, the average annual fatality rate was substantially smaller (0.009 fatalities per 100,000 workers). The agriculture and construction sectors combined accounted for 58% of the fatalities during the study period (n=207).

A CDC Morbidity and Mortality Weekly Report (MMWR) from 2008 reported by Luginbuhl et al. investigated heat-related fatalities among all workers—and agriculture workers in particular—using BLS CFOI data from 1992-2006 (Luginbuhl et al., 2008). During the study period, the authors identified 423 deaths related to environmental heat in CFOI using the OIICS v1.01 event/exposure code 321 (Exposure to environmental heat) and nature code 072* (Effects of heat and light). Similar to the approach taken by Gubernot et al., the authors calculated rates using CPS estimates of the average annual worker population for denominators.

For the overall workforce, the authors calculated an average annual incidence of 0.02 fatalities/100,000 workers, which is similar to the estimate reported by Gubernot et al. for 2000-2010 (0.022/100,000). Of the 423 fatalities identified, 102 (24%) occurred in the agriculture, forestry, fishing, and hunting sector (average annual fatality rate of 0.16/100,000 workers) and 68 occurred among workers in crop production or support activities for crop production (annual fatality rate of 0.39/100,000 workers). The rates for crop workers in North Carolina, Florida, and California were 2.36/100,000 workers, 0.74/100,000 workers, and 0.49/100,000 workers, respectively. These findings were later included in a peer-reviewed article (Jackson and Rosenberg 2010).

The editorial note accompanying this MMWR report mentioned, among other limitations, that CPS estimates used for denominators likely underestimate the number of crop workers—because of the potential lack of stable residences among these workers and the seasonal trends in employment—which would lead to an overestimate of risk for these workers. This limitation would presumably apply to any rate estimates calculated with CPS data for this specific population. To OSHA's knowledge, this is the only reported limitation in the included articles that would suggest a potential overestimation of incidence.

A third paper analyzed BLS CFOI heat-related fatality data for the construction sector, estimating fatality rates for various occupations within the sector using Standard Occupational Classification codes (Dong et al., 2019). Using the OIICS v2.01 nature code 172* (Effects of heat and light) to determine heat-relatedness and CPS estimates for sector-wide and occupation-specific denominators, the authors identified 82 heat-related construction deaths between 2011-2016 and estimated an average annual fatality rate for the entire sector (0.15 fatalities/100,000 workers) as well as for specific occupations. The occupations with the highest fatality rates included cement masons (1.62/100,000); roofers (1.04/100,000); helpers (1.03/100,000); brick masons (0.50/100,000); and laborers (0.29/100,000).

Finally, a paper from 2005 by Mirabelli and Richardson identified heat-related fatalities using medical examiner records from North Carolina for the period from 1977 to 2001, including 15 years of data before the creation of CFOI (Mirabelli and Richardson 2005). They determined that heat was a primary or underlying cause of death based on ICD-9 codes. The researchers used the decedents' location and activities reported in the records to determine work-relatedness, and they excluded cases in which the decedent was <10 years old or those which involved manufactured sources of heat.

The authors identified 40 occupational heat-related deaths. They classified 18 of these as farm workers and reported an annual fatality rate among these farm workers of 1.52 fatalities/100,000 workers. They reported 10 cases having occurred at a construction site but did not report a fatality rate for this group of workers. The average annual fatality rate for the entire State working population was 0.05 fatalities/100,000 workers.

As none of the identified papers reported fatality rates for the overall workforce for years beyond 2010, OSHA used the heat-related fatality counts reported by BLS for 2011-2022 (479 worker deaths) and employment estimates for the same years from CPS to calculate fatality rates for these years. For the denominator, OSHA used the total number of workers and average hours worked to estimate total FTEs per year. The average annual fatality rate during this period was 0.029 deaths/100,000 FTEs.

OSHA identified multiple studies that calculated and reported annual incidence estimates for heat-related fatalities among workers using data from BLS CFOI or medical examiner records. These studies reported heat-related fatality rates across an entire workforce (either National or State), using the total workforce as the denominator. As mentioned above, this would understate the risk to workers who are actually exposed to heat on the job since the denominator includes a large percentage of workers who are not exposed to heat ( e.g., office workers). Evidence in support of this claim comes from studies showing higher fatality rates when populations are stratified by sector, industry, or occupation. For instance, in Gubernot et al., 2015, the authors report an overall annual fatality rate of 0.022/100,000 workers whereas they report an annual fatality rate of 0.306/100,000 workers for workers in the agriculture, forestry, fishing, and hunting sector (a 14-fold difference). OSHA considers these stratified estimates to be more accurate estimates of the “true” incidence of heat-related fatalities among heat-exposed workers. ( print page 70739)

Evidence suggests that existing surveillance data undercount the total number of heat-related injuries, illnesses, and fatalities, among workers. The incident rates presented in the previous section are likely vast underestimates both because they use this surveillance data as the numerator when calculating incidence rates and because they overestimate the number of workers exposed to hot work environments ( i.e., the denominator for incidence rates). These sources of uncertainty are described below.

Incidence estimates based on BLS data are likely to underestimate the true risk to workers who are exposed to specific hazards, like heat, in part because of difficulties in estimating the population of exposed workers. The current approach for BLS SOII rate estimates is to use the population of all workers in the U.S. for the denominator, not just those exposed to the hazard of interest. For instance, the denominators used for the risk estimates presented above would include most office workers who work in climate-controlled buildings and would therefore not have occupational exposure to the levels of heat stress that have been associated with adverse outcomes. For 2022, BLS reported 116,435,925 full-time workers in the U.S. However, OSHA estimates the proposed standard would cover approximately 36 million workers, approximately one-third of the total full-time workers in the U.S. Therefore, BLS's use of a larger denominator likely underestimates risk because it includes workers not exposed to hazardous heat and therefore less likely to experience an HRI.

The denominators for the annual incidence estimates presented above also include worker-time for the entire year, even though for many workers, exposure to potentially harmful levels of heat only occurs during the hottest months of the year. Including unexposed worker-time in the denominator has the effect of diluting the incidence estimates, meaning annual incidence estimates do not accurately represent the risk to workers when they are actually exposed to hazardous heat. The risk to workers whose jobs do expose them to harmful levels of heat, on the days on which those exposures occur, would therefore be expected to be higher than the estimates published by BLS. In addition, using total worker populations as a basis for estimating incidence likely will underestimate the risk to particularly susceptible workers, such as older workers, workers with pre-existing conditions, and workers not acclimatized to the heat.

OSHA believes that studies that reported illness rates by sector or occupation provide evidence showing that the annual average illness rates reported across the entire workforce underestimate risk for exposed workers. For example, the Washington State and California workers' compensation studies found that heat-related illness rates for sector- or occupation-specific populations were substantially higher than the rates for the general working population in the State (Heinzerling et al., 2020; Bonauto et al., 2007; Hesketh et al., 2020). The sectors and occupations examined included those where exposure to hot environments was more likely than for the population as a whole ( e.g., Construction and Agriculture, Forestry, Fishing, and Hunting). Additionally, many of the surveillance papers described above also reported the month in which the injury, illness, or fatality occurred and found that most cases were clustered in the hotter, summer months ( e.g., June, July, and August). When researchers in Washington and Florida restricted their rate estimates to include data only for the third quarter (July, August, and September), they found rates that were several-fold higher than annual average illness rates over the whole population, which include many unexposed worker-days.

The general underreporting and undercounting of occupational injuries and illnesses has been a topic of multiple government reports ( e.g., Ruser, 2008; Miller, 2008; GAO, 2009; Wiatrowski, 2014). The authors of the peer-reviewed papers described in sections V.A.II., and V.A.III., above list underreporting or misclassification of cases as a limitation in their analyses that would have the effect of underestimating risk.

Two papers from the early 2000s that linked workers' compensation records to BLS SOII data found evidence that SOII missed a substantial amount of workers' compensation claims, depending on the State analyzed and the assumptions and methodology used (Rosenman et al., 2006; Boden and Ozonoff, 2008). In response to increased attention around this topic at the time, BLS funded additional research to examine the extent of underestimation in SOII and potential reasons (Wiatrowski, 2014). One of these studies involved linking multiple data sources ( i.e., not just SOII and workers' compensation) for cases of amputation and carpal tunnel syndrome (Joe et al., 2014). The authors found that the State-based surveillance systems included 5 times and 10 times more cases than BLS SOII, respectively.

Another study conducted as part of this broader effort estimated that approximately 30% of all workers' compensation claims in Washington between 2003-2011 were not captured in BLS SOII (Wuellner et al., 2016). This included sectors with higher rates of heat-related injuries and illnesses, such as Agriculture, Forestry, Fishing, and Hunting (28% of cases uncaptured) and Construction (28% uncaptured) (Wuellner et al., 2016, Table III). The rate of underreporting was particularly high for large construction firms (Wuellner et al., 2016, Table IV).

In response to the studies on SOII undercount, BLS authors have argued that differences in the inclusion criteria, scope, and purpose between BLS SOII and workers' compensation explain some of differences in the estimates and complicate the interpretations of the linkage-based studies (Ruser, 2008; Wiatrowski, 2014). SOII estimates OSHA-recordable injuries and illnesses each year and provides detailed case and demographic information ( e.g., nature of injury) for a specific subset of the more severe cases ( e.g., those involving days away from work). This scope (OSHA-recordable injuries and illnesses) inherently limits the ability for SOII to be used to estimate all occupational injuries and illnesses. Additionally, injuries and illnesses involving days away from work represent a limited percentage of the total injuries and illnesses reported to BLS. In 2022, these cases were 42% of total recordable cases, suggesting the case counts for HRIs in SOII could be missing up to 58% of all OSHA-recordable HRIs ( i.e., those not involving days away from work) ( https://www.bls.gov/​iif/​latest-numbers.htm ).

The injury and illness data that employers report to BLS come from the employer's OSHA Form 300 Log of Work-Related Injuries and Illnesses and OSHA Form 301 Injury and Illness Incident Report, so information on the quality of the data in these forms is relevant for understanding limitations of SOII. Through the Recordkeeping National Emphasis Program (NEP) from 2009-2012, OSHA found that almost half (47%) of establishments inspected by the agency had unrecorded and/or under-recorded cases, which were more common at establishments that ( print page 70741) originally reported low rates (Fagan and Hodgson, 2017). Several factors contributed to the under-recording and unrecording cases. First, in conducting thousands of interviews, the authors found that workers do not always report injuries to their employers because of fear of retaliation or disciplinary action. Second, some employers used on-site medical units, which the authors explained could contribute to underreporting ( e.g., if these units were used to provide first aid when additional medical care, which would have warranted reporting on OSHA forms, should have been provided).

Employers rely on workers to report injuries and illnesses that may otherwise be unobserved, but workers have multiple reasons to not do so. In addition to Fagan and Hodgson 2017, multiple studies have interviewed or surveyed workers on this topic. A recent systematic review of 20 studies found that 20-74% of workers—which included cleaning staff, carpenters, construction workers, and healthcare workers—did not report injuries or illnesses to management (Kyung et al., 2023). Some of the researchers asked workers about the barriers to reporting, which included fear, a lack of knowledge on the reporting process, and considering the injury to be a part of the job or not serious.

Finally, employers are disincentivized from reporting injuries and illnesses on their OSHA logs. Disincentives for reporting include workers' compensation premiums being tied to injury and illness rates, competition for contracts involving safety records, and a perception that reporting will increase the probability of being inspected by OSHA (GAO, 2009).

In interviews with employers selected to respond to SOII, researchers found that 42% of them were not maintaining a log (Wuellner and Phipps, 2018). In the same study, researchers found evidence to suggest that misunderstandings about the reporting requirements would likely lead to employers underreporting cases involving days away from work. A similar study conducted among SOII respondents in Washington State found that 12% weren't maintaining a log and 90% weren't complying with some aspect of OSHA's recordkeeping requirements (Wuellner and Bonauto, 2014).

While the general underreporting articles described here are not specific to heat, Heinzerling et al. 2020 examined rates of heat-related injuries and illnesses among workers in California and found that California's workers' compensation database, WCIS, had 3-6 times the number of heat-related cases between 2009-2017 than the official BLS SOII estimates for California for each year in that period (Heinzerling et al., 2020). Part of the reason for this discrepancy could be the difference in inclusion criteria between the two datasets, however, it is still a useful estimate for contextualizing the potential magnitude of underreporting of heat-related cases when using only SOII. While outside the U.S., a recent survey of 51 Canadian health and safety professionals in the mining industry found that 71% of respondents believed HRIs were underreported (Tetzlaff et al., 2024).

While workers' compensation data may capture injury and illness cases not included in BLS SOII, the data are not available for the entire U.S., as insurance coverage and reporting requirements vary across States, and most States do not have single-payer systems. Therefore, the majority of claims data are compiled by various insurers and not within a single database. Even when the data are available for an entire State, it is generally presumed that not all worker injuries and illnesses are captured in these data, in part because of eligibility criteria and in part because of underutilization of workers' compensation for reimbursement of work-related medical expenses.

Multiple papers have examined the extent to which and reasons why workers don't always use workers' compensation insurance to pay for work-related medical expenses and other reimbursable expenses. Some reasons workers have reported for not filing workers' compensation claims include fear, a lack of knowledge, “too much trouble” or effort, and considering the injury to be a part of the job or not serious (Kyung et al., 2023; Scherzer et al., 2005). Using the Washington State Behavioral Risk Factor Surveillance System (BRFSS), a telephone survey, Fan et al. (2006) found that 52% of the respondents in 2002 reporting a work-related injury or illness filed a workers' compensation claim. Using similar methodology across 10 States, Bonauto et al. (2010) found that among respondents who reported a work-related injury, there was a wide range in the proportion who reported having their treatment paid for by workers' compensation by State—47% in Texas to 77% in Kentucky (with a median of 61%). A study from 2013 estimated that 40% of work-related ED visits were paid for by a source other than workers' compensation (Groenewold and Baron, 2013). Worker race, geography, and having an illness rather than an injury were all predictors of whether workers' compensation was the expected payer.

There are a few papers that suggest this phenomenon is occurring for heat-related outcomes. Harduar Morano et al. 2015 (described above in Section V.A.II., Reported Annual Incidence of Nonfatal Occupational Heat-Related Injuries and Illnesses) found that across several southeastern States, workers' compensation as expected primary payer alone captured 60% of all emergency department visits and inpatient hospitalizations, which varied by State (50-80% for emergency department visits and 38-84% for inpatient hospitalizations) (Harduar Morano et al., 2015). Similarly, in the 2011 report by the Florida Department of Health (described above in Section V.A.II., Reported Annual Incidence of Nonfatal Occupational Heat-Related Injuries and Illnesses), 83% of claims identified were captured by workers' compensation as primary payer (Florida DOH, 2011). It should be noted that these percentages are influenced by the total number of captured cases and in both sources the authors presume that they did not capture all relevant cases.

Hospital discharge data are the only surveillance data presented in this risk assessment for which work-relatedness is not an inclusion criterion; therefore, researchers relying on this data need to take an additional step to assess work-relatedness for each case that introduces the possibility that work-related cases are not recognized as such and are thus excluded. Researchers identifying work-related cases typically use a combination of workers' compensation as the primary payer or ICD codes for external cause of injury. As discussed in the previous section, workers' compensation is not always used by workers, so relying on this variable will lead to undercounting. For external cause of injury codes ( e.g., E900.9 Excessive heat of unspecified origin), researchers have found that these are not always present or accurate for work-related injury cases (Hunt et al., 2007), which isn't unexpected given that they aren't required for reimbursement. For instance, codes indicating the location of occurrence were present in 43% of probable work-related injury cases the authors reviewed (Hunt et al., 2007). Harduar Morano and Watkins (2017) used external cause of injury codes to identify work-related emergency department visits and hospitalizations for heat-related illnesses in Florida. They found that 2.8% of emergency ( print page 70742) department visits, 1.2% of hospitalizations, and 0% of deaths were identified solely by an external cause of injury code for work.

Both workers' compensation claims and hospitalization data are also affected by the accuracy of diagnostic codes for identifying heat-related cases. While the use of ICD codes for surveillance of heat-related deaths, illnesses, and injuries is widely accepted, it is not infallible, as these codes are designed for billing rather than surveillance. The use of specific codes is up to the discretion of healthcare providers, so practices may vary by provider and facility. Healthcare providers may not always recognize that a patient's symptoms are heat-related and thus, they may not record a heat-specific ICD code. For example, a patient who presents to the emergency room after fainting would likely be diagnosed with “syncope” (the medical term for fainting). If the provider is aware that the patient fainted due to heat exposure, they should record a heat-specific ICD-10 code, T67.1 Heat syncope. However, if the provider is unaware that the patient fainted due to heat exposure (or otherwise fails to recognize the connection between the two), they may record a non-heat-specific ICD-10 code, R55 Syncope and collapse. Researchers suspect underreporting when ICD codes are used for surveillance of HRIs (Harduar Morano and Watkins, 2017) and recommend researchers use all possible fields available ( e.g., primary diagnosis, secondary diagnosis, underlying cause of death, contributing cause of death).

Researchers examining trends in heat-related illnesses using electronic health records for the Veterans Health Administration identified a dramatic increase in cases when ICD-10 was adopted, suggesting that the coding scheme in ICD-9 may have led to systematic underreporting of heat-related cases, at least for this population (Osborne et al., 2023). The authors also note that 8.4% of the HRI cases they identified were captured using unstructured fields ( e.g., chief complaint, reason for admission) and not ICD codes.

Not all sick and injured workers go to an emergency department or hospital and those that do are likely to be more severe cases. Unfortunately, estimating the proportion of injured and sick workers who do go to the hospital or emergency room is difficult, given a lack of data on this topic. In a 1998 CDC Morbidity and Mortality Weekly Report written by NIOSH safety researchers, the authors reported an analysis of unpublished data from the 1988 National Health Interview Survey (NHIS) Occupational Health Supplement which found that 34% of all occupational injuries were first treated in hospital emergency departments, 34% in doctors' offices/clinics, 14% in work site health clinics, and 9% in walk-in clinics (NIOSH DSR 1998). 1988 was the last year that NIOSH asked that question in the NHIS.

Care-seeking for workers experiencing heat-related symptoms specifically may be low. In a study evaluating post-deployment survey response data among a subset of the Deepwater Horizon oil spill responders (U.S. Coast Guard), Erickson et al. found that less than 1% of respondents reported seeking medical treatment for heat-related illness, yet 12% reported experiencing any heat-related symptoms (Erickson et al., 2019).

CFOI is well-regarded as the most complete and authoritative source on fatal workplace injuries. However, the approach used to classify the event and nature codes by BLS is not immune to misclassification of heat-related deaths. BLS relies on death certificates, OSHA fatality reports, news articles, and coroner reports (among other sources) to determine the primary or contributing causes of death. The criteria for defining a heat-related death or illness can vary by State, and among physicians, medical examiners, and coroners. Additionally, individuals who fill out death certificates are not necessarily equipped to make these distinctions or confident in their accuracy (Wexelman, 2013). Depending on State policies, individuals performing this role may be a medical professional or an elected official with limited or no medically relevant experience (National Research Council, 2009; CDC, 2023).

Researchers estimating fatality rates attributable to heat in the overall U.S. population using historical temperature records have produced much higher counts than approaches solely using death certificates (Weinberger et al., 2020). While outside the U.S., a recent study examining causes of death among migrant Nepali workers in Qatar from 2009-2017 demonstrated that deaths coded as cardiovascular-related ( e.g., “cardiac arrest”) among these mostly young workers were unexpectedly common and correlated with higher wet bulb globe temperatures, suggesting that these deaths may have been heat-related but not coded as such (Pradhan et al., 2019). Heat-related deaths are uniquely hard to identify if the medical professional didn't witness the events preceding the death, particularly because heat can exacerbate an existing medical condition, acting as a contributing factor (Luber et al., 2006).

In conclusion, the available evidence indicates that the existing surveillance data vastly undercount cases of heat-related injuries and illnesses among workers. OSHA additionally believes that the inclusion of unexposed worker-time in the denominator for incidence estimates underestimates the true risk among heat-exposed workers.

OSHA requests information and comments on the following questions and requests that stakeholders provide any relevant data, information, or additional studies (or citations) supporting their view, and explain the reasoning for including such studies:

• Are there additional data or studies OSHA should consider regarding the annual incidence of HRIs and heat-related fatalities among workers?
• OSHA has identified data from cohort-based and time series studies that would suggest higher incidence rates than data from surveillance datasets ( e.g., BLS SOII, workers' compensation claims). Are there other data from cohort-based or time series studies that OSHA should rely on for determining risk of HRIs to heat-exposed workers?
• Are employers aware of occupational HRIs that are not reported through BLS SOII, workers' compensation claims, or hospital discharge data? How commonly do HRIs occur that are not recorded on OSHA 300 logs?
• Are there additional data or studies that OSHA should consider regarding the extent of underreporting and underestimating of HRIs or heat-related fatalities?

In this section, OSHA presents the evidence that forms the basis of the heat triggers contained in the proposed standard. These triggers are based on the heat index and wet bulb globe temperature (WBGT). The WBGT triggers are based on NIOSH exposure limits ( i.e., the REL and RAL), which are supported by empirical evidence dating back to the 1960s and have been found to be highly sensitive in capturing unsustainable heat exposures.

Although there are no consensus-based heat index exposure limits for workers, the question of which heat ( print page 70743) index values represent a highly sensitive and appropriate screening threshold for heat stress controls in the workplace has been evaluated in the peer-reviewed scientific literature. The evidence described below provides information on the sensitivity of alternative heat index values, that is, the degree to which a particular heat index value can be used to screen for potential risk of heat-related injuries and illnesses (HRIs) and fatalities. OSHA looked at both experimental and observational evidence, including efforts to derive more accessible and easily understood heat index-based triggers from WBGT-based exposure limits, to preliminarily determine appropriate heat index values for triggering heat stress control measures. Each of these evidence streams has strengths and limitations in informing this question.

Relevant experimental evidence in the physiology literature is often conducted in controlled laboratory settings among healthy, young volunteers, but the conditions may not always mimic conditions experienced by workers ( e.g., workers often experience multiple days in a row of working in high temperatures). Observational evidence does not have this limitation because the data are collected among actual workers in real-world settings. However, observational evidence is potentially affected by exposure misclassification since exposure metrics are often derived from local weather stations and rely on maximum daily values. Experimental data does not have this limitation, since the laboratory conditions are highly controlled, including the exposure levels.

OSHA used both streams of evidence to support proposing an initial heat trigger of 80 °F (heat index) and a high heat trigger of 90 °F (heat index). The observational evidence that OSHA identified suggests that the vast majority of known occupational heat-related fatalities occur above the initial heat index trigger, making it a sensitive trigger for heat-related fatalities. The vast majority of nonfatal occupational HRIs also occur above this trigger. The experimental evidence (specifically the WBGT-based exposure limits) also suggests that when there is high radiant heat, a heat index of 90 °F would be an appropriate time to institute additional controls ( e.g., mandatory rest breaks). This is supported by observational evidence that shows a rapidly declining sensitivity above a heat index of 90 °F. OSHA has preliminarily concluded that the experimental evidence also supports the selection of these triggers as highly sensitive and therefore protective.

To determine an appropriate initial heat trigger, OSHA sought to identify a highly sensitive screening level above which the majority of fatal and nonfatal HRIs occur. This could presumably be used to identify the environmental conditions for which engineering and administrative controls would be most important to prevent HRIs from occurring. One challenge for determining this trigger level is that many factors influence an individual's risk of developing an HRI. In addition to workload, PPE, and acclimatization status, the risk of developing an HRI is also influenced by workers' abilities to self-pace at their jobs as well as whether there had been exposure to hot conditions on the prior day(s). There are also medications and comorbidities that may increase workers' risk of HRIs (see discussion in Section IV.O., Factors that Affect Risk for Heat-Related Health Effects).

The observational studies reviewed by OSHA used retrospective temperature and humidity data matched to the locations where HRIs and fatalities occurred over a period of time. Although these studies did not account specifically for workload, PPE use, acclimatization status, or other relevant factors, the HRI cases studied included worker populations where these factors were likely present to varying degrees. Therefore, OSHA has preliminarily determined that retrospective observational data collected among workers who have experienced fatal or nonfatal HRIs on the job is valuable to informing a screening level that reflects the presence of these multiple risk factors among worker populations. These studies are summarized in the following sections.

In a doctoral dissertation from 2015, Gubernot matched historic weather data to the heat-related fatalities reported in BLS CFOI (fatality data described in Section V.A., Risk Assessment) between 2000-2010 (Gubernot, 2015). Gubernot used historic, weather monitor-based temperature and dew point measurements from the National Climatic Data Center to recreate the heat index (using daily maximum temperature and daily average dew point) on the day of each fatality. If there was not already a monitor in the county where a fatality occurred, then the next closest weather monitor to that county was used. Of the 327 fatalities identified as being related to ambient heat exposure ( i.e., cases with secondary heat sources, like ovens, were excluded), 96.3% occurred on a day with a calculated heat index above 80 °F and 86.9% occurred on a day above 90 °F. Using a higher threshold such as a heat index of 95 °F would have only captured approximately 71% of fatalities (estimated from Figure 4-2 of the study). The author also evaluated how many cases occurred on a day when a National Weather Service (NWS)-defined excessive heat event (EHE) was declared. In a directive to field offices, the NWS outlines when offices should issue excessive heat warnings—when there will be 2 or more days that meet or exceed a heat index of 105 °F for the Northern U.S. and 110 °F for the Southern U.S., with temperatures not falling below 75 °F (although local offices are allowed to use their own criteria) (NWS, 2024a). Gubernot appears to have used a simpler criterion to evaluate the sensitivity of these EHEs—whether the heat index on the day of the fatality was at or above 105 °F for northern States and at or above 110 °F for southern States. Only 42 fatalities (12.8%) occurred on days meeting the EHE definitions, suggesting EHEs are not a sensitive trigger for occupational heat-related fatalities. During the SBREFA process, small entity representatives suggested that OSHA consider the NWS EHE definitions as options for the initial and/or high heat triggers, but based on these findings (and those reported in other studies summarized in this section), OSHA has preliminarily determined that these criteria are not sensitive enough and would not adequately protect workers.

Some limitations of this analysis include the use of nearest-monitor exposure assignment, as well as the use of maximum temperature with average dew point to calculate heat index, both of which may introduce exposure misclassification. Although the author did not refer to the latter as a daily maximum heat index, this estimate would most closely approximate that value, which would suggest that workers were likely exposed to heat index values below that level during the work shift leading up to the fatality.

In a meta-analysis published in 2020, Maung and Tustin (both affiliated with OSHA at the time) conducted a systematic review of studies, such as the one described above by Gubernot, where researchers retrospectively assigned heat exposure estimates to occupational heat-related fatalities (Maung and Tustin, 2020). The purpose of their meta-analysis was to identify a heat index threshold below which occupational heat-related fatalities do not occur ( i.e., a highly sensitive ( print page 70744) threshold). Maung and Tustin identified 418 heat-related fatalities among civilian workers across 8 studies. Approximately three quarters of these civilian fatalities (n=327; 78%) came from Gubernot 2015. The authors found a heat index threshold of 80 °F to be highly sensitive for civilian workers—96% of fatalities (402 of 418) occurred on days with a heat index estimate at or above this level. A heat index threshold of 90 °F had slightly lower sensitivity—approximately 86% (estimated from table 1 and figure 3 of their study). Similar to the findings reported in Gubernot 2015, one of the NWS thresholds for issuing heat advisories (heat index of 105 °F) did not appear to be a sensitive trigger, missing 68% of civilian worker fatalities.

The limitations for Gubernot 2015 apply to this analysis as well. These analyses (including the data from Gubernot, 2015) were limited to outdoor workers, potentially limiting the generalizability of the findings. This analysis also relied on single values ( e.g., daily maximum heat index) to capture exposure across a work shift. As pointed out by Maung and Tustin, it is important to consider that exposure characterizations using daily maximum heat index likely over-estimates the exposures that workers experience throughout the shift leading to the fatality. For example, a fatality occurring on a day with a daily maximum heat index of 90 °F likely involved prolonged exposure to heat index values in the 80s °F.

In 2019, a group of OSHA researchers published a similar analysis for both fatal and nonfatal HRIs reported to OSHA in 2016 among outdoor workers (Morris CE et al., 2019). They identified 17 fatalities in this subset and used nearest weather station data to estimate daily maximum heat index on the day of the fatality. All 17 fatalities occurred on a day with a daily maximum heat index of at least 80 °F (the lowest was at 88 °F). A daily maximum heat index of 90 °F had a sensitivity of approximately 94%, while 100 °F had a sensitivity of approximately 35%. A major limitation with this analysis is its small sample size (n=17 fatalities).

Morris et al., identified 217 nonfatal HRIs among outdoor workers reported to OSHA in 2016 (Morris CE et al., 2019). They found that 99% of these cases happened on a day with a daily maximum heat index of at least 80 °F. There is a steep decline in sensitivity for daily maximum heat index values in the 90s °F—89% for 90 °F but approximately 58% for 100 °F (estimated from Figure 5 of the study which combines fatal and nonfatal cases)—suggesting that many nonfatal HRIs occur on days when the heat index does not reach 100 °F. One limitation of this dataset is potential selection bias, because the dataset only included cases that were reported to OSHA. This study therefore did not include cases in State Plan States.

A much larger analysis conducted among emergency department (ED) visits in the Southeastern U.S. was published by Shire et al. (Shire et al., 2020). The authors identified 5,017 hyperthermia-related ED visits among workers in 5 southeastern States (Florida, Georgia, Kentucky, Louisiana, and Tennessee) between May and September in 2010-2012. While the previously described studies used nearest monitor data, Shire et al. used data from the North American Land Data Assimilation System (NLDAS), which incorporates both observation and modeled data to fill in gaps between locations of monitors, providing data at a higher geographic resolution (0.125° grid). Since the authors only had ED visit data at the county level, they used the NLDAS data to compute population-weighted, county-level estimates of daily maximum heat index using all the grids within each county. They found that approximately 99% of ED visits occurred on days with a daily maximum heat index of at least 80 °F and about 95% of cases on days with a maximum heat index of at least 90 °F. Approximately 54% of cases occurred on days with a daily maximum heat index of 103 °F or higher. This further supports the finding from Morris et al. (2019) that sensitivity declines steeply above a heat index of 90 °F. One limitation of this analysis is the use of the emergency department location as the basis for the exposure assignment, which has the potential to introduce exposure misclassification if workers were working far away from the ED facility.

In a 2016 doctoral dissertation, Harduar Morano conducted a retrospective analysis of 3,394 heat-related hospitalizations and ED visits among Florida workers in May-October between 2005-2012, using data from the weather monitor nearest to the zip codes where the hospitalizations and ED visits occurred to characterize heat exposure (Harduar Morano, 2016). The vast majority of cases occurred on a day with a daily maximum heat index of at least 80 °F, with approximately 91% of cases occurring on a day with a maximum heat index of at least 90 °F (estimated from Figure 6-4). There was also a 13% increase in the HRI hospitalization and ED visit rate for every 1 °F increase in heat index at values below 99 °F (Figure 6-4, Lag 0 plot of the study), suggesting that potential triggers in the mid-to-high 90's would increasingly miss many cases. One limitation of this analysis and that conducted by Shire et al. is that hospitalization and ED visit data did not include enough information to distinguish between indoor vs outdoor workers; it is possible that indoor workers could have been exposed to conditions not captured by the weather data (such as working near hot industrial processes).

In addition, four studies of workers' compensation data in Washington State—three of which were reported in Section V.A., Risk Assessment—have examined maximum temperature or heat index on the days of reported HRIs (Bonauto et al., 2007; Spector et al., 2014; Hesketh et al., 2020; Spector et al., 2023). Hesketh et al., 2020 (an update on Bonauto et al., 2007) matched weather data to addresses for the HRI claims in the State's workers' compensation database between 2006 and 2017 (Hesketh et al., 2020). They found that, of the 905 claims for which they had temperature data, over 75% of HRIs occurred on days with a maximum temperature of at least 80 °F and approximately 50% of claims occurred on days with a maximum temperature of at least 90 °F (estimated from Figure 2). They also reported that approximately 75% of claim cases occurred when the hourly maximum temperature was at least approximately 79 °F. This paper is part of the rationale for Washington State lowering the trigger level in its heat-specific standard from 89 °F to 80 °F—the old trigger of 89 °F had missed 45% of cases in this dataset (Washington Dept. of Labor & Industries, 2023). A similar study published in 2023 expanded the dataset used by Hesketh et al. to include HRI claims from 2006 to 2021 (n=1,241) (Spector et al., 2023). The authors used gridded meteorological data from the PRISM Climate Group at Oregon State University and geocoded accident location (or business location or provider location if accident location was unable to be used) to determine the maximum temperature on the day of the event. They found that 76% of HRI claims occurred on a day with a maximum temperature of at least 80 °F (this increased to 79% when restricted to cases that were “definitely” or “probably” outdoors). A major limitation of these studies is the use of ambient temperature, limiting the ability to compare findings to other papers that relied on the heat index. In ( print page 70745) Spector et al. 2014, the authors calculated the daily maximum heat index for each county with an HRI in their dataset on the date of injury (Spector et al., 2014). They obtained the county of injury and, when not available, imputed the location of the injury rather than using the employer address, which is assumed to be more accurate for characterizing exposure. In their analysis of 45 agriculture and forestry worker HRI claims between 1995-2009 that had corresponding weather data, Spector et al. found that 75% of HRI claims occurred on days when the maximum heat index was at least 90 °F, whereas only 50% occurred on days when it was at least 99 °F and 25% for 106 °F.

In summary, researchers have identified a heat index of 80 °F as a highly sensitive trigger for heat-related fatalities (capturing 96-100% of fatalities) and nonfatalities (99-100%) among workers (excluding results from Washington State). When looking at ambient temperature, researchers in Washington found that 75-76% of HRI claims occurred on a day with a maximum ambient temperature of 80 °F or greater. Multiple studies additionally identified a rapidly declining sensitivity above a heat index of 90 °F, suggesting that additional protective measures ( e.g., observation for signs and symptoms of HRIs) are needed once the heat index reaches approximately 90 °F.

One of the common limitations of the analyses presented in this section is the use of a single reading ( e.g., daily maximum heat index) to capture each affected worker's exposure on the day of the event. In reality, conditions fluctuate throughout the day, so relying on maximum measures would likely overestimate heat exposure across the workday. The use of nearest monitor weather data is also likely to lead to exposure misclassification. The inclusion of indoor workers in some of the studies is also a limitation, since the exposure for those workers could be very different ( e.g., if there is process heat). In Spector et al. 2023, the authors noted an increase in the percent of cases occurring on days with a maximum temperature of 80 °F when restricting to cases that definitely or probably occurred outdoors. In all these studies, researchers can only examine conditions for the cases that were captured in the surveillance systems. There could be a bias such that cases occurring on hotter days were more likely to have been coded as heat-related and included in these databases. Failure to ascertain HRI cases occurring at lower heat indices could have skewed the findings upwards, making it appear that hotter thresholds were more sensitive than they actually were. Finally, the use of heat index (or ambient temperature) ignores the impacts of air movement as well as radiant heat, which can substantially increase the heat stress a worker is exposed to and increase the risk of an HRI.

NIOSH has published exposure limits based on WBGT in its Criteria for a Recommended Standard going back multiple decades. [ 3 ] These exposure limits—the REL and RAL—account for the contributions of wind velocity and solar irradiance, in addition to ambient temperature and humidity. (ACGIH has published similar exposure limits—the TLV and AL.) In addition to WBGT, NIOSH and ACGIH heat stress guidelines require the user to account for metabolic heat production (through the estimation of workload) and the contributions of PPE and clothing. The user adds an adjustment factor to the measured WBGT to account for the specific clothing or PPE worn (specifically those ensembles that impair heat loss) and uses a formula based on workload to estimate the exposure limit. They then compare the measured (or adjusted, if using a clothing adjustment factor) WBGT to the calculated exposure limit to determine if the limit is exceeded. Work-rest schedules with increasing time spent on break can further increase the exposure limit.

These exposure limits and guidelines are based in empirical evidence, such as laboratory-based trials conducted in the 1960s and 1970s. This basis for WBGT exposure limits is described in detail by both NIOSH and ACGIH (NIOSH, 2016; ACGIH, 2017). These exposure limits have been tested and found to be highly sensitive (100%) in modern laboratory conditions in capturing unsustainable heat exposures ( i.e., when a steady increase in core temperature is observed) (Garzon-Villalba et al., 2017). Among workers in real-world settings, these WBGT-based exposure limits have been found to be highly sensitive for fatal outcomes (100% in one study; 92-100% in another) and, although slightly less so, still sensitive for nonfatal outcomes (73% in one study; 88-97% in another); however, these studies are limited by their small sample size and retrospective characterization of workload, acclimatization status, and clothing/PPE use (which are required for accurately estimating WBGT-based exposure limits) (Tustin et al., 2018b; Morris CE et al., 2019).

Two papers have attempted to apply the concepts of the WBGT-based exposure limits to the more easily accessible and understood heat index metric. Based on the relationship between WBGT and heat index, Bernard and Iheanacho developed a screening tool that reflects heat stress risk based on heat index and workload category—light (180 W), moderate (300 W), and heavy (415 W)—using assumptions about radiant heat but ignoring the contributions of wind and clothing (Bernard and Iheanacho, 2015). To do this, they created a model predicting WBGT from the heat index. From this model, WBGT estimates were produced within a 1 °C range for heat index values of 100 °F or more but the model was less accurate at heat index values below 100 °F. Using their reported screening table, which allows the user to adjust for low vs high radiant heat, an acclimatized worker performing a heavy (415 W) workload in high radiant heat outdoors would be above the WBGT-based exposure limit and in need of a break at a heat index of 90 °F. The same worker, if unacclimatized, would be above the exposure limit at a heat index of 80 °F. These findings support the provision of 15-minute breaks at a heat index of 90 °F in OSHA's proposed standard, as well as the provision requiring these breaks for unacclimatized workers at a heat index of 80 °F (unless the employer is following the gradual acclimatization schedule and providing breaks if needed). The authors noted that high radiant heat indoors could require even greater adjustments to the heat index. As further evidence for the need to adjust these values for radiant heat exposure, Morris et al. (2019) reported that for the days on which HRIs occurred in their dataset, cloud cover was often minimal suggesting there was exposure to high radiant heat when the HRIs occurred.

More recently, Garzón-Villalba et al. used an experimental approach to derive workload-based HI heat stress thresholds (Garzón-Villalba et al., 2019). The researchers used data from two progressive heat stress studies of 29 acclimatized individuals. Participants were assigned different work rates and wore different clothing throughout the trials, serving as their own controls. Once thermal equilibrium was established, the ambient temperature was increased in five-minute intervals while holding relative humidity ( print page 70746) constant. The critical condition defined for each subject was the condition at which there was a transition from a stable core body temperature to an increasing core body temperature ( i.e., the point at which heat exposure became unsustainable). Using the results from these trials, the authors established an equation deriving a heat index exposure limit (equivalent to the TLV or REL) at different metabolic rates for a worker wearing woven clothing:

HI benchmark (°C) = 49−0.026 M

Where M is workload in Watts.

Garzón-Villalba et al. assessed the effectiveness of the proposed heat index thresholds for predicting unsustainable heat stress by using receiver operating characteristic curves and area-under-the-curve (AUC) values to determine predictive power (this technique is commonly used to evaluate the predictive power of diagnostic tests). The AUC value for the proposed heat index thresholds with subjects wearing woven clothing was 0.86, which is similar to that of the WBGT-based thresholds, based on the authors' prior analysis (Garzón-Villalba et al., 2017). This result showed that the heat index thresholds derived by Garzón-Villalba et al. (2019) would reasonably identify unsustainable heat exposure conditions.

Compared to the heat index thresholds proposed by Bernard and Iheanacho (2015), the heat index thresholds proposed by Garzón-Villalba et al. are the same at low metabolic rates (111 °F for 180 W) but higher at higher metabolic rates: 105.8 °F versus 100 °F at 300 W and 100.4 °F versus 95 °F at 415 W ( Note: these values are unadjusted for radiant heat). This is likely because the ACGIH WBGT-based exposure limits, upon which Bernard and Iheanacho based their heat index thresholds, are intentionally more conservative at higher metabolic rates, whereas Garzón-Villalba used a less conservative linear model to derive their heat index thresholds (Garzón-Villalba et al., 2019). When adding an adjustment for full sunshine provided by the authors, the proposed heat index-based exposure limit derived from the Garzón-Villalba et al. (2019) equation for a worker performing a very heavy workload (450 W) is 92.8 °F.

Thus, laboratory-derived heat index thresholds for unsustainable heat exposure are higher than heat index thresholds shown in observational studies to be sensitive for predicting the occurrence of HRIs. There are several reasons that may explain why values determined to be sensitive in laboratory settings are higher than those reported among workers in real-world settings. For one, volunteers in laboratory studies are often young, healthy, and euhydrated ( i.e., beginning the trial adequately hydrated). They are also not exposed to consecutive days of heat exposure for eight-hour or longer work shifts. Working in hot conditions on the prior day has been demonstrated in the literature to be a risk factor for HRIs, even among acclimatized individuals (Garzón-Villalba et al., 2016; Wallace et al., 2005). Therefore, the use of volunteers and exposure conditions in laboratory-based trials may not always provide good proxies for workers and the environments in which they work. There is also significant inter-individual variability in heat stress tolerance, which may mean trial studies with few participants might not capture the full range of heat susceptibilities faced by workers.

In summary, long-established and empirically validated occupational exposure limits exist for WBGT. In observational studies, WBGT exposure limits have been found to be highly sensitive for detecting fatal HRIs among workers and, although slightly less so, still sensitive for nonfatal outcomes (although these studies are limited by small sample size and retrospective work characterization). Research efforts to crosswalk the WBGT-based exposure limits to the more accessible heat index metric have demonstrated that a heat index of 90-92.8 °F would represent an appropriate trigger for controls such as mandatory rest breaks for acclimatized workers performing heavy or very heavy workloads in high radiant heat conditions (Bernard and Iheanacho, 2015; Garzón-Villalba et al., 2019). For unacclimatized workers performing heavy workloads in high radiant heat conditions, a heat index trigger of 80 °F would be in line with the WBGT-based exposure limits (Bernard and Iheanacho, 2015). Although these two studies suggest that higher triggers could reasonably be applied to workers performing lighter workloads, the assumptions used may not always apply to workers ( e.g., no exposure to working in the heat the prior day, healthy, euhydrated). This may explain, at least in part, the discrepancy in findings between the observational and experimental studies discussed in this section.

In their heat-specific standards, summarized in the table below, States use various initial and high heat triggers, some of which depend on the clothing or gear worn by workers. OSHA's proposed triggers are generally in line with those used by these States.

OSHA is proposing using the same initial heat trigger (heat index of 80 °F) as Oregon's existing standard and Maryland's proposed standard (Or. Admin. R. 437-002-0156 (2022); Or. Admin. R. 437-004-1131 (2022); Code of Maryland Regulations 09.12.32: Heat Stress Standards (2024)). California and Colorado use an ambient temperature trigger of 80 °F for outdoor work sites and agricultural sites, respectively, as does the Washington standard for workers wearing breathable clothing (Cal. Code of Regulations (CCR), tit. 8, section 3395 (2015); 7 Colo. Code Regs. section 1103-15 (2022); Wash. Admin. Code sections 296-62-095 through 296-62-09560; 296-307-097 through 296-307-09760 (2023)). California's proposed indoor standard uses an ambient temperature trigger of 82 °F (CCR, tit. 8, section 3396 (2023)).

The high heat trigger that OSHA is proposing (heat index of 90 °F) is the same as Oregon's existing standard and Maryland's proposed standard. California and Colorado use an ambient temperature high heat trigger of 95 °F, while the Washington standard uses 90 °F. The California indoor proposal uses an ambient temperature or heat index trigger of 87 °F to impose additional requirements.