This study was conducted with male sugarcane cutters who were employed for either the 2021–2022 or 2022–2023 6-mo harvest season by the agribusiness Pantaleon in southwest Guatemala. These workers represent a subset of study participants whom were recruited as part of the “CKDu Heat and Air Pollution” (CHAP I) Study (Grant R01 ES31585) focused on occupational exposures and practices during the work shift. All participants were at least 18 yr old at the time of recruitment and were residing in local communities in the coastal lowland region surrounding the sugarcane fields. The sugarcane harvest lasts from November to May. During their work shift, sugarcane cutters manually cut and stack cane in the fields for approximately 8 h/day with 2 h of interspersed rest, roughly from 0700 to 1700, and each man cuts on average six tons of cane daily (31). Rest breaks include two 30-min breaks, taken in the shade, including one in the morning around 1000 and one in the afternoon around 1430, as well as a 1-h break for lunch at noon. Workers are provided with free access to purified water directly in the field and, in addition, receive at least 5 L of electrolyte solution daily. Other details of the work setting and practices have been published previously (5). When these workers have completed their work shift, they return home to their communities each evening on non-air-conditioned buses. This commute can take from 30 min up to 2 h, each way, to their worksite and back, depending on the daily field location. Workers return to homes in the communities that lack air conditioning or other cooling mechanisms, such as fans, with limited ventilation. As a pilot (n = 9), we assessed the conditions of the home environment that may contribute to heat exposure among a subset of workers involved in this study. Additional details will be described in DISCUSSION.

This cross-sectional descriptive study characterized individual heat exposures for this group of agricultural workers, following them individually for 24 h to include both their workday (CHAP I study) and off-work exposures (substudy). For the substudy, the period of observation and data collection was extended to include time off work (~1700 to 0600 the next morning). Across the 2 years of fieldwork, the study design was integrated within CHAP I at one of the three time points during the 6-mo harvest. Over a 6-day period in February 2022 (days 1–6) and a 4-day period in April 2023 (days 7–10), measures of heat exposure and T_c response were collected during the workday and the off-work period. February represents the approximate midpoint of the harvest season, whereas April is closer to the end of the season, allowing us to sample workers who are acclimatized to the climate and work (18). A total of 55 workers were evaluated from 10 work groups. Data were collected from one work group (2–9 workers) each study day.

Briefly, to begin the 24-h monitoring period, workers and the research team arrived at the sugarcane field at ~0600 where workers provided a urine sample, had body mass measured, were given a T_c monitoring pill to ingest, and were equipped with an ambient temperature logger; a wet bulb globe temperature (WBGT) meter was set up in the field. At the end of work around 1700, researchers returned to the field to meet the workers, collect another urine sample, and provide a second pill. To end the 24-h monitoring period, another urine sample and survey data were collected the following morning in the sugarcane field before the start of the next workday. Details about each measurement are provided below.

The participants for this study were recruited and provided written informed consent in November 2021 and November 2022. Workers who could not sign their name provided a thumbprint with a witness present. This research was approved by the Colorado Multiple Institutional Review Board (COMIRB No. 20-0509) of the University of Colorado and the Comité de Ética Independiente ZUGUEME in Guatemala.

Environmental data were collected on heat exposure by two methods. During the workday, wet bulb globe temperature (WBGT) data were measured continuously in the sugarcane fields within ~1 km of all workers using the Kestrel 5400 (Kestrel Instruments, Boothwyn, PA). These data were used to calculate the heat index (HI), which was derived from dry temperature and humidity using the National Weather Service formula through the weathermetrics package in R (32, 33).

Personal ambient temperature (T_pa) was collected for each individual worker in the fields during the workday and in those same individuals during off-work hours at home using a small, wireless temperature logger (Hygrochron iButtons, model DS1923, Maxim Integrated, California) set to record temperature once every 10 min. The temperature loggers were mounted to plastic key fobs and worn by workers on a running vest during the workday. At the end of the work shift, the temperature logger placement was moved from the vest to a bracelet around the wrist. The data were downloaded each day using the 1-Wire viewer software (Maxim Integrated, California).

Core body temperature (T_c) was measured from the start of the workday through the next morning using a pill-sized ingestible telemetric sensor (Bodycap e-Celsius Performance, Caen, France) that measures and saves data. Ingestible telemetric temperature capsules are used to measure gastrointestinal temperature, particularly in field and occupational settings, and are considered both a valid and reliable surrogate indicator of T_c (34-36).

Study workers ingested the telemetric pill the morning of their monitored workday, shortly before starting their shift. The external wireless receiver (e-Viewer, BodyCap, Caen, France) for the telemetric pill was worn in a fanny pack around the waist by the workers to receive the data from the pill in real time (at work and off-work) before it was excreted (36). Pilot testing with the telemetric pills in this population revealed two main takeaways. First, if workers were provided with the pill the night before their workshift, many workers excreted the pill before work began the next morning, as other researchers have observed (37); therefore, the pill was given to workers shortly before the workshift. Second, although the pill records data, it was often excreted before meeting the research team at the end of the workday. Therefore, to limit missing data, workers were equipped with wireless receivers to collect the data before excretion.

Because of the rapid excretion of the pill observed in pilot testing, workers were asked to take a second pill at the end of their work shift before returning home, a practice that had been cleared as safe by the pill manufacturer. The number of workers who could have their T_c monitored overnight on each of the 10 study days varied based on how many receivers were available (Table 2).

Workers were weighed in work clothing only (pants and long-sleeved shirt) with boots, shin guards, and other equipment removed. The morning weight was measured with a digital scale that was calibrated before each data collection session (Seca 874 DR, Seca Corporation, Chino, CA) and was used to calculate body mass index (BMI) with height, which had been collected at baseline in November (beginning of the harvest) of each study year. Hydration was assessed by urine specific gravity (USG) immediately in the field using a digital refractometer (5) (ATAGO PAL-10S, Tokyo, Japan). Urine samples were collected at the end of the workday to evaluate workday hydration and the next morning preshift to estimate off-work hydration. A recovery survey was administered verbally in Spanish at the end of the 24-h monitoring period to ask workers about their recovery practices after work, that is, the evening and overnight prior. The survey included questions related to hydration, sleep and rest quantity and quality (38-40), and nutrition, behaviors including off-work medication, tobacco, and alcohol use, how and when they arrived home the previous evening, activities they did after arriving home, the time they went to bed, and the time they woke up that morning.

Table 2 shows the average workday heat index and T_pa across the 10 study days, with the frequency (%) of heat strain (exceeding T_c of 38.0°C, 38.3°C, and 38.5°C) on each day. Based on the T_pa data, study days 7 and 10 had the highest mean temperatures. The highest average HI was on day 9 (39.4°C). Study days 8 and 10 saw the highest frequency of individuals exceeding a T_c of 38.0°C (80%), followed by day 4 and then day 3. Day 4 was the coolest day based on HI, and yet 75% of workers exceeded a T_c of 38.0°C, whereas three workers also exceeded 38.3°C, and two workers exceeded 38.5°C. At least 50% of study participants exceeded a T_c of 38.0°C on four of the 10 days.

Overall, 44% of workers experienced workday heat strain (exceeded 38.0°C), 16% exceeded 38.3°C, and 6% exceeded 38.5°C (Table 2). Figure 1 shows a time series with a LOESS estimate of the central tendency of T_c of the workers across the workday period, and on average, T_c increased across the workday. A total of five individuals (9%) were dehydrated (USG > 1.020) postshift. Two of the five workers (one on study day 4 and the other on study day 8) also exceeded T_c 38.0°C on the day they were dehydrated.

During the workday, T_pa increased until ~1300 then decreased slightly until the end of the workday (Fig. 2). As shown in Table 2, the average T_c across all workers during the workday was 37.3°C ± 0.5°C and reached an average maximum of 37.9 ± 0.4°C in the early afternoon. The average and maximum T_c tended to increase across the day (Fig. 1). Workers who exceeded T_c of 38.0°C did so starting from 1000 until the end of the workday.

The T_pa temperatures were lower during the early evening hours and started to increase around 2100–2200 (Fig. 2) until ~0300, when they started to decrease again, on average. T_c measures were highest from 1900 to 2200. The hourly off-work T_pa and T_c data can be viewed in APPENDIX.

Figure 3 displays the off-work trajectory of T_c following the workday. The average off-work max was 37.6°C and was recorded early in the off-work period, between 1800 and 2000 before starting to decrease after 2000. As described earlier, after 2000 when T_c started to decrease, T_c gradually increased sometime after 0400, which is also when many workers are beginning to wake up and get ready for the commute to work the next day, as described further below. On average, T_c decreased by one degree during the off-work/overnight period. Seven workers (41%) exceeded 37.5°C while off work. No one exceeded a T_c of 38.0°C during the off-work period. Five individuals (29%) experienced a higher or equal off-work max T_c compared with their max workday T_c. These data are shown in the Appendix Table A2 and Appendix Fig. A1, which contains T_c and T_pa data for each of the workers individually across both the workday and off-work period, as well as the workers’ reported time returning home from work and time they reported going to bed, as a preliminary assessment of off-work practices. Notably, in the same survey, which is detailed further below, most workers, including those individuals with higher (>37.5°C) off-work T_c, reported not doing any physical activity like playing sports after work, and most arrived home after dark (Appendix Table A3).

Among at least four individuals (24%), their T_c off-work increased again after having already reported going to bed before they woke up the next morning. Indeed, in Fig. 3, the average T_c increased just before 0200 and decreased again just before 0400. As shown in the Appendix Table A1 and Fig. 2, the average T_pa was also elevated at this time, during 0100–0159.

Workers’ reported practices from the recovery survey and off-work hydration are displayed in Table 3. The survey was administered just before workers began their next workday, at the end of the 24-h monitoring period; the average amount of time elapsed since waking up and the morning survey time was 2.8 h (range: 1.5–4.4 h). Few workers were dehydrated in the morning (5.5%), although approximately three-quarters reported drinking fewer than five cups of water after leaving work the previous day, before bed, and 35% had only had one cup of water since waking up that morning. Workers consumed a range of other beverages during the off-work period, shown in Appendix Table A3, including atol, which is a warm, traditional corn-based beverage, coffee, juice or soda, and electrolyte solution.

In terms of sleep and rest, 15% slept fewer than 7 h overnight; approximately the same proportion reported this to be their usual amount of sleep (18%). The average amount of time spent at home between shifts was 10 h (range: 8–13 h). Several questions were asked related to sleep issues, shown in Table 3. Six workers (11%) reported feeling exhausted when they woke up that morning. One of the six workers who felt exhausted was among those who reported sleeping fewer than 7 h overnight. As shown in Appendix Table A3, five workers (9%) reported feeling neither alert nor sleepy at the time of the survey, and one worker felt sleepy. Most workers were sharing a bed with at least one other person (partner or child), with 15% sharing with two other people and 4% with three others. Just two workers reported doing any physical activity after work—either playing sports or doing work around their home/land.

No one reported drinking alcohol during off-work hours, and one individual reported taking an unnamed medication to help with sleeping. Four workers (7%) reported feeling a little to moderately hungry before going to bed. Of those four, one of them was among the six who reported feeling exhausted. Four percent (two workers) reported that their family did not have enough food to eat during the previous month. The most common foods that workers reported eating for dinner were beans (44%), chicken or meat (33%), rice or pasta (27%), and eggs (18%). Other examples of foods consumed by a minority of workers (<10%) included soup, cheese, and potatoes. Corn tortillas are commonly consumed at most meals by the workers in this population, based on anecdotal evidence and our prior experience with and knowledge of workers’ diets (50); however, workers did not specifically report tortillas in this survey.

This study carries several limitations. We were limited by a small sample size during the off-work period, restricting our ability to draw comparisons. Nonetheless, this was the first study to examine off-work T_c and one of the very few to date to have measured T_c in the field during the workday in a relatively large agricultural worker population. We did not analyze the potential lag between environmental heat exposure and T_c. This may be an important consideration in understanding the relationship between T_c and the ambient environment that may have implications for proposed mitigation strategies. In this analysis, we did not account for the potential variability in workload during the workday, which will influence workday T_c. We speculate from the recovery survey data that off-work activity levels are more similar across the cohort, though this will require further study. In addition, we conducted this study across 6 days in February 2022 and 4 days in April 2023, and therefore, the results may not be indicative of other times in the harvest season or generalizable to agricultural workers in other regions. Nonetheless, we observed variability in the HI and T_pa data even across those 10 days. Ultimately, future studies will be needed to understand the generalizability of our findings to other conditions of work and to other home environments of agricultural workers.

The data presented in this study lay the groundwork for further investigation into the physiological recovery process of workers exposed to daily heat and physical exertion. In this study, male agricultural workers in Guatemala experienced high rates of heat strain during their workday, as measured internally through T_c. T_c continued to increase across the workday, including into the off-work hours after completing the work shift. Questions remain around how to define “recovery” of T_c in this population, including following workday rest breaks, right after work, and during overnight sleep. Longitudinal studies are needed to understand whether workers are experiencing heat strain daily and whether such cumulative stress, without adequate recovery, leads to adverse health outcomes over the longer term. Future research in this cohort will further explore the relationship between T_c and ambient temperature and humidity, as well as other factors that can impact heat-related illness, including physical exertion. Our results point to the need to better understand the conditions and practices relevant to the home environment and off-work hours that may influence recovery and the subsequent response to heat and exertion at work. By also capturing round-the-clock environmental heat exposure, this work will help inform future intervention research to identify strategies that better protect workers from heat-related illness and related effects.