http://orcid.org/0000-0003-1537-3355
Figures
Abstract
Background
Climate change has implications for human health and productivity. Models suggest that heat extremes affect worker health, reduce labor capacity, and commodity supply. Chronic health conditions are on the rise internationally. However there is a paucity of direct empirical evidence relating increasing temperatures to both agricultural worker health and productivity.
Methods and findings
We evaluated the relationship between temperature exposure, kidney function, and two measures of productivity—tons of commodity produced and job attrition, of 4,095 Guatemalan sugarcane cutters over a 6-month harvest. We used distributed lag non-linear models to evaluate associations between wet bulb globe temperature (WBGT) and productivity of workers with normal or impaired kidney function. The cumulative effect of exposure to a max WBGT of 34°C was 1.16 tons (95% CI: -2.87, 0.54) less sugarcane cut over the next five days by workers with impaired kidney function, compared to exposure to 29°C. Impaired kidney function was associated with premature workforce attrition. Workers starting the harvest season with impaired kidney function were more than twice as likely to leave employment (HR: 2.92, 95% CI: 1.88, 4.32).
Conclusions
Heat extremes may be associated with loss of agricultural worker productivity and employment, especially among those with impaired kidney function. Agricultural workers who develop health conditions, such as kidney disease, are particularly vulnerable in the face of climate change and increasing heat extremes. The resultant loss of employment and productivity has significant implications for global commodity supplies.
Citation: Dally M, Butler-Dawson J, Krisher L, Monaghan A, Weitzenkamp D, Sorensen C, et al. (2018) The impact of heat and impaired kidney function on productivity of Guatemalan sugarcane workers. PLoS ONE 13(10): e0205181. https://doi.org/10.1371/journal.pone.0205181
Editor: Peggy S. Lai, Massachusetts General Hospital, UNITED STATES
Received: June 7, 2018; Accepted: September 20, 2018; Published: October 5, 2018
Copyright: © 2018 Dally et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All files to reproduce the findings in this manuscript are housed on Open Science Framework (https://osf.io) DOI 10.17605/OSF.IO/R2CG7.
Funding: This evaluation was supported in part by Pantaleon; the Chancellor, University of Colorado, Denver – Anschutz; Centers for Disease Control and Prevention (https://www.cdc.gov) (U19 OH01127; LSN) and National Institutes of Health (https://www.nih.gov) (R21 ES028826; LSN). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the Centers for Disease Control and Prevention, the National Institutes of Health, or the Department of Health and Human Services. The National Center for Atmospheric Research is sponsored by the National Science Foundation. Pantaleon provided the University of Colorado investigators with de-identified productivity data and pre-employment medical screening data, all of which were collected by the company for business purposes prior to the analysis. The funder had no role in data analysis or data interpretation. A co-author employed by the company (CA) participated on the writing team. They provided details regarding work practices, but did not modify the results or conclusions of this report. The corresponding author had full access to all the raw data and had final responsibility for the decision to submit for publication.
Competing interests: This evaluation was partially supported by Pantaleon. This does not alter our adherence to PLOS ONE policies on sharing data and material.
Introduction
One of the most pressing challenges facing the world is the increasing impact of climate change on human health and productivity [1,2]. Rising temperatures will reduce labor capacity [3]. The decline in world economic productivity will disproportionately affect developing countries located in warm, tropical climates [4]. Decreased production in these major food producing regions will threaten food supplies worldwide [5] at a time when global demand for agricultural production in developing countries will rise by an estimated 77% [6].
There is emerging evidence that the adverse effects of increasing global temperatures may already be affecting worker productivity [7]. Multiple studies have demonstrated how worker productivity declines as heat increases [4,8–10]. However only a few studies have examined the meteorological effect on productivity in an agricultural setting, and those have focused on only healthy workers [7,11]. Less is known about how heat’s impact on agricultural worker health [12] contributes to productivity loss, although theoretical models predict negative effects of increasing temperatures on both worker health and performance [13].
For agricultural workers, occupational exposure to hot environments has been identified as a major climate related concern [12,14]. For example, the pandemic of Chronic Kidney Disease of unknown origin (CKDu) among sugarcane cutters and other workers in Central America [15,16], Sri Lanka [17], Egypt [18], and India [18] has been attributed, in part, to heat stress and resulting recurrent dehydration [2,12,19]. A small pilot study has suggested a relationship between agricultural worker productivity and reduced kidney function [20], but no study to date has examined how heat exposure and kidney function are associated with drop out from the workforce, despite the potential implications of attrition on agricultural production. There is an urgent need to understand the implications of heat exposure in agricultural workers internationally, given the global increase in disease and disability, including kidney disease [21].
To assess the relationship between worker health, heat exposure, and productivity, we evaluated daily, averaged agricultural worker productivity of 4,095 Guatemalan sugarcane cutters over a six-month harvest season. We identified those who met the definition of impaired kidney function at the start of the harvest season, and related these findings to local heat conditions. We hypothesized that there would be a direct inverse relationship between temperature exposure, measured as the wet bulb globe temperature (WBGT), and the amount of sugarcane produced. Furthermore, we hypothesized that the productivity of workers with impaired kidney function would be more highly affected by temperature exposure. Finally, we hypothesized that workers with impaired kidney function would be more likely to leave the workforce prior to the end of the six-month harvest season. To our knowledge, this is the first study to directly investigate the relationship between kidney function, heat, and agricultural productivity.
Methods
Population and study design
To explore the relationship between kidney insufficiency, increased heat exposure, and agricultural productivity, we examined a cohort of sugarcane cutters who were employed by a large agribusiness in southwest Guatemala. Guatemala is a recognized hotspot for chronic kidney disease [22], has a large agrarian workforce [23], and is at risk of experiencing more extreme heat days due to climate change [24].
After a one-week acclimatization period in November, during which employees work shorter days and cut fewer tons of sugarcane, workers typically work eight-hours during a ten-hour shift in six-day blocks before receiving one rest day. Cutting sugarcane is considered very heavy work involving swinging a machete to cut the stalk a few centimeters above ground level, followed by lifting, trimming and stacking the cane (Fig 1).
(Left) Sugarcane cutter using a machete to cut sugarcane. (Right) Cut sugarcane being collected and stacked.
Workers receive a base wage regardless of the amount of sugarcane harvested. Since 2009, the agribusiness has been promoting hydration, rest, and shade use for the cane cutters, aligned with U.S. Occupational Safety and Health Administration (OSHA) recommendations [25]. This includes instructing cane cutters to drink 16 L of water and 2.5 L of an electrolyte solution (composition per liter: 4.6 g NaCl, 34 g carbohydrates (26 g sucrose) and 2 g KCl) and take three 20-minute breaks and one 60-minute lunch break in the provided shade during the work shift. Water and electrolyte solution are provided daily by the agribusiness.
Our population was drawn from a cohort of workers employed by the agribusiness for the November 2015 to May 2016 harvest season. Per company protocol, the employer screened 5,138 sugarcane workers from August to November 2015, prior to the start of the six-month sugarcane harvest. This pre-employment screening included a 1) medical exam (blood pressure, heart rate, height, and weight), 2) survey that collected information about each individual’s demographics, lifestyle behaviors, and occupational and medical history, and 3) venipuncture to examine serum creatinine, a measure of kidney function. Company policy was to either not hire or place workers in lighter duty jobs if their pre-employment serum creatinine was 1.45 mg/dL or more. In 2015, 107 workers (2.5%) were excluded due to high creatinine. To be included in our analyses, workers must have worked the first week of the harvest, had pre-employment serum creatinine measurements, and had completed the pre-employment survey. Per company protocol, all workers must have been at least 18 years of age at the start of the harvest. Ethics review and approval for our evaluation of these data were received from the Colorado Multiple Institution Review Board (COMIRB). This was a secondary evaluation of de-identified data historically collected by the agribusiness and deemed non-human subjects research. Consent was not required.
Measures of productivity.
We considered two measures of productivity: average daily tons cut per work day and worker attrition. We calculated average daily tons cut per work day using the average of all workers for each day of the harvest. We calculated this measure for all workers combined, as well as stratified on kidney function. Worker attrition was defined if a worker did not work through all 28 weeks of the harvest. The company was able to provide de-identified productivity information for 4,873 cane cutters hired for the 2015–2016 harvest.
Wet bulb globe temperature.
The primary predictor of interest was wet bulb globe temperature (WBGT) which is a composite measure that combines ambient temperature, humidity, solar radiation, and wind speed and is commonly used in studies where workers are exposed to direct sunlight [7,8,11]. Meteorological observations from the Cengicaña weather station (14.33° N, 91.05° W, 300 meters above sea level) were used to compute the WBGT at 15-minute intervals. The observations required for the WBGT computations included ambient near-surface air temperature (Ta), relative humidity (RH), incident solar radiation at the surface (S), and wind speed (U). The WBGT is calculated for outdoor conditions with a solar load using the equation proposed by OSHA [26]:
(1)
Where Twb is the wet bulb temperature, Tg is the black globe temperature, as measured by a thermometer inserted in the center of a copper globe painted matte black [27], and Ta is the station-observed ambient air temperature as noted above (all temperature units are in °C). Twb is estimated from Ta and RH using the empirical equation of Stull, 2011 [28], which has a mean absolute error < 0.3°C for the range of temperatures encountered in this study. It is noteworthy that the method of Stull, 2011 [28] yields the psychrometric web bulb temperature (as opposed to the natural wet bulb temperature) which is required for the WBGT approach we employ that follows Dimiceli et al., 2011 [27], and Weatherly and Rosenbaum, 2017 [29], as described next. While Tg can be directly measured with proper instrumentation, it is not common to do so because of the expense, and therefore Tg is usually estimated using other weather variables. Tg was computed following the physically-based formulation of Dimiceli et al., 2011 [27], which requires observed Ta, RH, S, U, and the station coordinates as inputs. Several parameters are also required; we employed identical parameters to those used by Weatherly and Rosenbaum, 2017 [29], who also used the Dimiceli et al., 2011 [28] equations. Three additional modifications were made. First, we directly used the observed incident solar radiation (S), whereas Weatherly and Rosenbaum, 2017 [29] estimated S because they did not have measurements. Second, we found that the equation for Tg becomes unstable at very low solar zenith angles—i.e., at sunrise each day when the sun is low on the horizon and the solar zenith angle is between 85–90°–and therefore solar zenith angles > 85° were set to 85°. Third, we set a minimum wind speed, U, of 1 m s-1 following the approach of Lemke and Kjellstrom, 2012 [30], which accounts for the fact that airflow will rarely be zero relative to the skin of humans working outdoors.
For each day, we computed the distribution of the 15-minute WBGT measurements during the working hours of 07:00 to 17:00. We then selected the 95th percentile value as the WBGT95 measurement for that day. This approach allowed us to examine high values of daily WBGT exposure that may be the most important contributors to worker heat-related health effects [14] while protecting against values so high that they may be considered outliers. Additionally, we calculated the mean work-shift WBGT (WBGTmean) that workers were exposed to during typical working hours (07:00–17:00).
Kidney function.
We were interested in determining whether kidney function at the start of the harvest modified the relationship between WBGT exposure and worker productivity. Serum creatinine was used to calculate the estimated Glomerular Filtration Rate (eGFR) using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation [31] for each worker. All workers were male and race was considered “non-black” for all participants in the CKD-EPI equation. We dichotomized workers based on eGFR at or above 60 ml/min/1.73 m2, defined as having functioning kidneys, and below 60 ml/min/1.73 m2, defined in this analysis as having impaired kidney function, since this is a commonly agreed upon cut-off for defining CKD Stage 3 [32].
Statistical analysis
Building upon studies that have shown temperature and health do not follow a linear trend [33, 34], we chose to use distributed lag non-linear models as a flexible way to model the relationship between temperature exposure and productivity. Additionally, distributed lag non-linear models allowed us to also address the potentially important lagged effect of temperature exposure [35, 36] by estimating the effect on productivity in each future day following an exposure to increased WBGT.
When the effect of an exposure on the outcome may not be limited to the period when it is observed, lags can be used to determine the effect of the exposure at different times over the course of an event [33]. To model the initial exposure as well as the lag dimension, distributed lag non-linear models can be used [33]. Distributed lag non-linear models are based on the concept of a cross-basis, which is the joint modeling of two bases. In the framework of a distributed lag non-linear model, the two bases that the cross-basis is comprised of represent the exposure, as well as the lag. The cross-basis is then used as a predictor in a typical regression model to relate the exposure and lags simultaneously to the outcome. The complete theoretical framework for distributed lag non-linear models can be found in Gasparrini, et al 2010 [33].
For each of our outcomes, WBGT95 and WBGTmean, fit a linear regression model for the relationship between the cross-basis of temperature exposure and lagged temperature exposure, with average tons produced. Each model was then stratified on eGFR categories (<60 ml/min/1.73 m2 vs. ≥60 ml/min/1.73 m2).
To calculate the cross-basis, the relationship between temperature exposure and average daily tons produced was modelled with natural splines with four degrees of freedom. The degrees of freedom were selected by plotting the smoothed relationship between WBGT and average daily tons while varying the degrees of freedom. To determine the number of lags to be included in the analysis, we examined the cross-correlation between both WBGT95 and WBGTmean with average daily tons. The cross-correlation suggested that five lags were appropriate. Five lags were also a theoretically appropriate number of lags to use, since a typical work week is six days long. We modeled the lagged relationship of temperature exposure and average daily tons produced with three strata. The effect on average tons produced was assumed constant within each stratum. The dlnm package [37] in R version 3.4.3 [38] was used to fit these models.
As a sensitivity analysis we re-ran the distributed lag non-linear models with the exclusion of the months November and May. This allowed us to account for effects that may be due in part to the seasonality of productivity, where production is naturally slower in the start of the season when they are in the acclimatization phase, as well as the end of the season when there is less sugarcane to harvest.
Cox proportional hazard models were run to assess differences in worker attrition between eGFR categories. Proportional hazard assumptions were visually checked using the log-negative log plot of the Kaplan Meier estimated survival curves. The dataset was stratified for those who dropped out prior to week 11 and those who dropped out during or after week 11 due to violations of the proportional hazards assumption. To account for the potential confounding effect of age, age was included as a continuous covariate in the Cox proportional hazard models. To perform demographic comparisons between eGFR categories (<60 ml/min/1.73 m2 vs. ≥60 ml/min/1.73 m2), t-Tests with Satterthwaite approximations were used for continuous variables and Chi-square tests were used for categorical variables. These analyses were done using SAS version 9.4 (Cary, NC).
Results
There were 4,095 (80%) hired cane cutters who met inclusion criteria. All cane cutters who met inclusion criteria were male. As shown in Table 1, 77 (2%) of the hired cohort had impaired kidney function at the time of pre-employment screening. The average worker was 30 years (SD: 9), had worked 7 previous harvests (SD: 7), and had a Body Mass Index (BMI) of 23 (SD: 3). The average amount of sugarcane cut per day was 5.9 tons (SD: 3.4). Workers with impaired kidneys were unable to cut as much sugarcane per day compared to those with higher kidney function; this amounted to an average of 5.62 (SD: 2.67) tons per day compared to 5.76 (SD: 2.72) tons per day, respectively. Workers with impaired kidney function were significantly less likely to complete the harvest, 42% (N: 32) of workers left before the end of the season compared to 25% (N: 1,019) (Table 1).
The median WBGT95 was 32.13°C (IQR: 1.84°C; Range: 29.11°C to 35.94°C). The median WBGTmean was 29.07°C (IQR: 1.42°C; Range: 26.84°C to 31.42°C). The maximum values WBGT95 occurred in November. Temperatures appear to decline from December through January and begin to rise again through March, at which time they appear to stabilize through the end of the harvest (Fig 2).
The use of a distributed lag non-linear model allowed us to assess the cumulative effect on productivity over future days following exposure to an increased WBGT [33]. In this case, lags up to five days were included in the model, thus providing us with the overall cumulative effect on the average productivity over the next five days given an increase in WBGT exposure (Fig 3, Tables 2 and 3). The cumulative effect on tons of sugarcane cut for workers with impaired kidney function who experienced exposure to a WBGT95 of 34°C is estimated to be a loss of 1.16 (95% confidence interval (CI): -2.87, 0.54) tons over the next five days compared to if they were exposed to a WBGT95 of 29°C. The estimated cumulative effect on tons of sugarcane cut by workers with functioning kidneys was 0.59 tons (95% CI: -2.05, 0.87) less under the same circumstances. The general trend for WBGT95 was that as the temperature of exposure increased, the cumulative reduction in the amount of sugarcane cut over the next five days was more pronounced. The cumulative effect on productivity in the WBGT95 model was greater for those with impaired kidney function, although not statistically different than those with functioning kidneys. There was no evidence to suggest that exposure to increased WBGTmean influenced cumulative productivity for either group. Lag specific estimates for a range of temperature exposures can be found in S1 Fig.
All estimates are calculated relative to a reference WBGT of 29°C.
All estimates are calculated relative to a reference WBGT of 27°C.
(Top) Temperature was defined using the 95th percentile of WBGT during the work-shift with a reference of 29°C (Bottom) Temperature was defined using the mean work-shift WBGT with a reference 27°C. (Left) Impaired kidney function: eGFR < 60 ml/min/1.73 m2. (Right) Functioning kidneys: eGFR ≥ 60 ml/min/1.73 m2.
Periods of lower productivity occurred at the start of the season during the acclimatization phase, as well as the end of the season when production is wrapping up. These natural periods of lower productivity corresponded with increased temperatures (Fig 2). To assess the stability of our estimates, we performed the analysis with data from November and May removed. Interestingly, the effect on productivity of increased WBGT exposure became more pronounced in this analysis (Fig 4, Tables 4 and 5). For those with impaired kidney function, the cumulative effect on productivity was estimated to be 1.41 tons (95% CI: -3.04, 0.22) less over five days for an exposure to WBGT95 of 34°C compared to an exposure of 29°C. For those with functioning kidneys, the cumulative effect on productivity was estimated to be 0.80 tons (95% CI: -2.12, 0.52) less under the same circumstances. The most noticeable difference was in the estimates of the effect of WBGTmean on cumulative productivity. For both the impaired and functioning groups, there was a statistically significant effect of exposure to WBGTmean of 31°C compared to 27°C on cumulative productivity. The impaired group was estimated to produce 1.28 tons (95% CI: -2.38, -0.18) less over five days, while the functioning group was estimated to produce 1.13 tons (95% CI: -2.03, -0.23) less.
All estimates are calculated relative to a reference WBGT of 29°C. November and May removed.
All estimates are calculated relative to a reference WBGT of 27°C. November and May removed.
(Top) Temperature was defined using the 95th percentile of WBGT during the work-shift with a reference of 29°C (Bottom) Temperature was defined using the mean work-shift WBGT with a reference 27°C. (Left) Impaired kidney function: eGFR < 60 ml/min/1.73 m2. (Right) Functioning kidneys: eGFR ≥ 60 ml/min/1.73 m2.
Productivity can be adversely affected in two ways: reduced daily tons produced and premature workforce attrition. We observed a notable difference in job attrition between those with impaired kidney function and those with functioning kidneys (Fig 5). Table 6 summarizes the estimated hazard ratios for those who left the workforce before week 11 of the 28-week harvest compared to those who left during week 11 or later. Workers who started the season with kidney impairment were more likely to leave the workforce before the harvest season was over, even after controlling for age. There was no effect of kidney function on attrition prior to week 11. However, kidney function was related to attrition on or after week 11. The likelihood of leaving the workforce during week 11 or later for those with impaired kidney function was 2.92 (95% CI: 1.88, 4.32) the likelihood for those with functioning kidneys after adjusting for age.
(Impaired kidney function) eGFR < 60 ml/min/1.73 m2 represented by the dashed lined. (Functioning kidneys) eGFR ≥ 60 ml/min/1.73 m2 represented by the solid line.
Discussion
In this study, we evaluated the health and daily productivity of 4,095 sugarcane workers in Guatemala to address the hypothesis that there would be a direct inverse relationship between temperature exposure and the amount of sugarcane produced with the greatest impact being on those with impaired kidney function at the start of the harvest. While not definitive, our analyses suggest that increases in daily maximum WBGT negatively affected productivity over the next five days. This was more apparent in the impaired kidney function group, although not statistically significant, when compared to those with an eGFR > 60 ml/min/1.73 m2 at the start of the harvest. This study provides initial field evidence for the link between impaired kidney function, increased heat exposure, and agricultural worker productivity.
Our data suggest that as exposure to heat increases, agricultural production may decrease. This supports findings from other studies on heat and productivity [4,10]. Other investigators have estimated that there is a 0.57% reduction in direct work time [8] and 5% reduction in work output of rice harvesters [7] with increases in WBGT. Conversely, one study found only limited evidence of an effect of WBGT on declined tree fruit harvest [11] although the authors point out that it was conducted in a more moderate climate than that found in Central America. Estimates show that in parts of Central and South America, up to 10–15% of annual daylight hours are so hot that productivity is lost [14]. Under heavy work conditions, regular rest is recommended when the WBGT exceeds 26°C [14]. There is evidence that workers with impaired kidney function cut less sugarcane per day than do those with functioning kidneys. Our data suggest that the productivity of workers with impaired kidney function may be more greatly impacted by exposure to increase heat extremes. It is important to consider the role of acute and chronic health conditions on worker productivity in future projections of climate’s effect on worker productivity.
Additionally, we have shown that workers with impaired kidney function have more than twice the risk of not finishing a harvest, resulting in fewer days worked. This is consistent with the estimate that global labor capacity has decreased 5.3% from 2000 to 2016 due to temperature change [1]. Productivity, economic output, pay, and family income are all reduced as a result of an injured worker’s natural inclination to slow down work or limit working hours [14]. In low- and middle- income countries that are dependent on manual labor, the health and welfare of workers are of paramount importance for sustained industrial growth [10]. Thus, our data illustrate the inherent conflict between preserving health and maintaining productivity that workers and employers must address [14]. This challenge is particularly urgent in the case of CKDu, in light of evidence implicating high energy expenditure during temperature extremes as a potential contributor to kidney insufficiency [12,19,39].
Our data demonstrate that lost productivity in agricultural workers with impaired kidney function is reflected by the likelihood that workers will drop out of the workforce as well as how much of a commodity is produced. These findings, taken in context with the body of evidence suggesting a relationship between work in hot climates and the epidemic of CKDu [2,12,39] leads us to propose a ‘two-hit’ hypothesis for how heat exposures impact worker productivity and health. Multiple studies have implicated high intensity work in hot climates as contributors to the global epidemic of CKDu [2]. In turn, we have provided evidence suggesting that sugarcane workers with kidney insufficiency are more vulnerable, suffering greater impact on their productivity and employment as temperature exposure increases. Our data provide initial empirical support that climate can directly impact commodity production [40]. Taken together, published models and our data suggest that the extent of food insecurity is being underestimated when climate models of our world’s food supply neglect heat’s effect on worker health and workforce sustainability. Our findings reinforce the need to better understand the role of heat in the development of both acute and chronic health conditions, such as chronic kidney disease, especially as the global burden of kidney disease and disability continues to rise [21].
This study carries a few limitations. We evaluated the effect of heat on productivity of only hired workers during a single season in a single large agribusiness, thus results are not necessarily generalizable to all agricultural workers. Measurements of WBGT were calculated using data from a weather station in proximity to the fields where the sugarcane cutters were working. The station is at a higher altitude than most of the fields, thus providing us with relatively lower temperature readings. While this provided us with a conservative estimate of WBGT, we acknowledge that the estimated impact of heat on productivity is likely an underestimate. Additionally, we related average daily productivity to aggregate measures of WBGT during work hours. By doing so we were limited in our ability to determine how heat progression throughout the work day impacted productivity as well as our ability to determine the effect of temperature exposure outside of work hours. As shown in Fig 2, week 11 is around the time that the seasonal temperatures begin to rise. Consequently, we have fewer observations on the productivity of impaired kidney function workers at higher values of WBGT, making estimates for those workers at increased temperatures less stable and making relationships between temperature exposure and productivity harder to detect for those workers. In this analysis we did not temporally decompose long-term trends in productivity, which may explain some of the observed association between increased temperature exposure and decreased productivity.
In conclusion, this study provides support for connecting heat extremes to productivity of agricultural workers, which is more apparent in those experiencing impaired kidney function. In light of the rise in the global burden of non-vector borne, chronic illness [21,41], the epidemic of work-related chronic illnesses that have been related to heat, including CKDu [2], and evidence of chronic kidney disease in Central American agricultural communities [15,16], we call for further research and interventions that help address this public health crisis. There is an urgent need to identify how these relationships impact productivity, labor loss, income, food accessibility, and most importantly, health and economic outcome disparities among vulnerable workers and their families.
Supporting information
S1 Fig. Lag specific effects on average daily tons produced estimated at three different WBGT exposures.
(Top) Temperature was defined using the 95th percentile of WBGT during the work-shift with a reference of 29°C (Bottom) Temperature was defined using the mean work-shift WBGT with a reference 27°C. (Left) Impaired kidney function: eGFR < 60 ml/min/1.73 m2. (Right) Functioning kidneys: eGFR ≥ 60 ml/min/1.73 m2.
https://doi.org/10.1371/journal.pone.0205181.s001
(PDF)
References
- 1. Watts N, Amann M, Ayeb-Karlsson S, Belesova K, Bouley T, Boykoff M, et al. The Lancet Countdown on health and climate change: from 25 years of inaction to a global transformation for public health. Lancet. 2017. Epub 2017/10/30. pmid:29096948.
- 2. Glaser J, Lemery J, Rajagopalan B, Diaz HF, García-Trabanino R, Taduri G, et al. Climate Change and the Emergent Epidemic of CKD from Heat Stress in Rural Communities: The Case for Heat Stress Nephropathy. Clin J Am Soc Nephrol. 2016;11(8):1472–83. Epub 2016/05/05. pmid:27151892.
- 3. Dunne JP, Stouffer RJ, John JG. Reductions in labour capacity from heat stress under climate warming. 2013;3:563.
- 4. Lundgren K, Kuklane K, Gao C, Holmér I. Effects of heat stress on working populations when facing climate change. Ind Health. 2013;51(1):3–15. pmid:23411752.
- 5.
Janetos A, Justice C, Jahn M, Obersteiner M, Glauber J, Mulhern W. The Risks of Multiple Breadbasket Failures in the 21st Century: A Science Research Agenda. Boston, MA: Boston University, The Frederick S. Pardee Center for the Study of the Longer-Range Future, 2017.
- 6.
Alexandratos N, Bruinsma J. World agriculture towards 2030/2050: the 2012 revision. Rome: FAO, 2012 Contract No.: ESA Working paper No. 12–03.
- 7. Sahu S, Sett M, Kjellstrom T. Heat exposure, cardiovascular stress and work productivity in rice harvesters in India: implications for a climate change future. Ind Health. 2013;51(4):424–31. Epub 2013/05/20. pmid:23685851.
- 8. Li X, Chow KH, Zhu Y, Lin Y. Evaluating the impacts of high-temperature outdoor working environments on construction labor productivity in China: A case study of rebar workers. Building and Environment. 2016;95(Supplement C):42–52. https://doi.org/10.1016/j.buildenv.2015.09.005.
- 9. Zander KK, Botzen WJW, Oppermann E, Kjellstrom T, Garnett ST. Heat stress causes substantial labour productivity loss in Australia. Nature Climate Change. 2015;5:647.
- 10. Krishnamurthy M, Ramalingam P, Perumal K, Kamalakannan LP, Chinnadurai J, Shanmugam R, et al. Occupational Heat Stress Impacts on Health and Productivity in a Steel Industry in Southern India. Saf Health Work. 2017;8(1):99–104. Epub 2016/11/03. pmid:28344848.
- 11. Quiller G, Krenz J, Ebi K, Hess JJ, Fenske RA, Sampson PD, et al. Heat exposure and productivity in orchards: Implications for climate change research. Arch Environ Occup Health. 2017;72(6):313–6. Epub 2017/01/31. pmid:28139172.
- 12. Kiefer M, Rodríguez-Guzmán J, Watson J, van Wendel de Joode B, Mergler D, da Silva AS. Worker health and safety and climate change in the Americas: issues and research needs. Rev Panam Salud Publica. 2016;40(3):192–7. pmid:27991978.
- 13. Kjellstrom T, Kovats RS, Lloyd SJ, Holt T, Tol RS. The direct impact of climate change on regional labor productivity. Arch Environ Occup Health. 2009;64(4):217–27. pmid:20007118.
- 14.
International Labor Organization (ILO). Climate Change and Labour: Impacts of Heat in the Workplace: United Nations Development Programme, 2016.
- 15. Correa-Rotter R, Wesseling C, Johnson RJ. CKD of unknown origin in Central America: the case for a Mesoamerican nephropathy. Am J Kidney Dis. 2014;63(3):506–20. Epub 2014/01/10. pmid:24412050.
- 16. Cohen J. Mesoamerica’s mystery killer. Science. 2014;344(6180):143–7. pmid:24723592.
- 17. Wijetunge S, Ratnatunga NV, Abeysekera DT, Wazil AW, Selvarajah M, Ratnatunga CN. Retrospective analysis of renal histology in asymptomatic patients with probable chronic kidney disease of unknown aetiology in Sri Lanka. Ceylon Med J. 2013;58(4):142–7. pmid:24385054.
- 18. Weaver VM, Fadrowski JJ, Jaar BG. Global dimensions of chronic kidney disease of unknown etiology (CKDu): a modern era environmental and/or occupational nephropathy? BMC Nephrol. 2015;16:145. Epub 2015/08/19. pmid:26282933.145.
- 19. Laws RL, Brooks DR, Amador JJ, Weiner DE, Kaufman JS, Ramírez-Rubio O, et al. Changes in kidney function among Nicaraguan sugarcane workers. Int J Occup Environ Health. 2015;21(3):241–50. Epub 2015/01/28. pmid:25631575.
- 20. Wegman DH, Apelqvist J, Bottai M, Ekström U, García-Trabanino R, Glaser J, et al. Intervention to diminish dehydration and kidney damage among sugarcane workers. Scand J Work Environ Health. 2017. Epub 2017/07/07. pmid:28691728.
- 21. Collaborators GDaH. Global, regional, and national disability-adjusted life-years (DALYs) for 333 diseases and injuries and healthy life expectancy (HALE) for 195 countries and territories, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet. 2017;390(10100):1260–344. pmid:28919118.
- 22. Laux TS, Barnoya J, Cipriano E, Herrera E, Lopez N, Polo VS, et al. Prevalence of chronic kidney disease of non-traditional causes in patients on hemodialysis in southwest Guatemala. Rev Panam Salud Publica. 2016;39(4):186–93. pmid:27657183.
- 23.
The World Bank: World Bank Open Data [updated Dec 20, 2017]. http://databank.worldbank.org/data/reports.aspx?source=2&series=SL.AGR.EMPL.MA.ZS.
- 24.
Magrin GO, Marengo JA, Boulanger J-P, Buckeridge MS, Castellanos E, Poveda G, et al. Climate Change 2014: Impacts, Adaptation, and Vulnerability Part B: Regional Aspects. Cambridge, United Kingdom and New York, NY, USA: Cambridge University Press. p. 1499–566.
- 25.
National Institute of Safety and Health (NIOSH). Working in Hot Environments. NIOSH Pub. No. 86–112. Cincinnati, OH: NIOSH, 1986.
- 26.
Occupational Safety and Health Administration (OSHA). OSHA Technical Manual, TED 01-00-015 [TED 1–0.15A]. Washington, D.C.: OSHA, 2008.
- 27.
Dimiceli, VE, Piltz SF, Amburn SA. Estimation of black globe temperature for calculation of the wet bulb temperature index. Proc. World Congress on Engineering and Computer Science. San Francisco, CA: International Association of Engineers, 2011, 591–599.
- 28. Stull R. Wet-Bulb Temperature from Relative Humidity and Air Temperature. Journal of Applied Meteorology and Climatology. 2011;50(11):2267–9.
- 29. Weatherly J, Rosenbaum M. Future Projections of Heat and Fire-Risk Indices for the Contiguous United States. Journal of Applied Meteorology and Climatology. 2017;56(4):863–76.
- 30. Lemke B, Kjellstrom T. Calculating Workplace WBGT from Meteorological Data: A Tool for Climate Change Assessment. Industrial Health. 2012;50(4):267–78. pmid:22673363
- 31. Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF, Feldman HI, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604–12. pmid:19414839.
- 32.
The Renal Association [Internet]. [cited Dec 20, 2017]. https://renal.org/information-resources/the-uk-eckd-guide/ckd-stages/.
- 33. Gasparrini A, Armstrong B, Kenward MG. Distributed lag non-linear models. Stat Med 2010; 29(21): 2224–2234. pmid:20812303
- 34. Curriero FC, Heiner KS, Samet JM, Zeger SL, Strug L, Patz JA. Temperature and mortality in 11 cities of the eastern United States. Am J Epidemiol. 2002;155(1):80–7. pmid:11772788.
- 35. Bobb JF, Peng RD, Bell ML, Dominici F. Heat-related mortality and adaptation to heat in the United States. Environ Health Perspect. 2014;122(8):811–6. Epub 2014/04/29. pmid:24780880.
- 36. Chen K, Zhou L, Chen X, Ma Z, Liu Y, Huang L, et al. Urbanization Level and Vulnerability to Heat-Related Mortality in Jiangsu Province, China. Environ Health Perspect. 2016;124(12):1863–9. Epub 2016/05/06. pmid:27152420.
- 37. Gasparrini A, Distributed lag linear and non-linear models in R: the package dlnm. Journal of Statistical Software 2011; 43(8):1–20. pmid:22003319
- 38.
R Core Team (2017). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.
- 39. Ordunez P, Saenz C, Martinez R, Chapman E, Reveiz L, Becerra F. The epidemic of chronic kidney disease in Central America. Lancet Glob Health. 2014;2(8):e440–1. Epub 2014/06/24. pmid:25103508.
- 40. Kjellstrom T, Freyberg C, Lemke B, Otto M, Briggs D. Estimating population heat exposure and impacts on working people in conjunction with climate change. Int J Biometeorol. 2017. Epub 2017/08/01. pmid:28766042.
- 41. Levy BS, Patz JA. Climate Change, Human Rights, and Social Justice. Ann Glob Health. 2015;81(3):310–22. pmid:26615065.