|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2011 flu ob 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0004891
Andersen, Arden Bruce.
Validation of the usf safe exposure time equation for heat stress
h [electronic resource] /
by Arden Bruce Andersen.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 32 pages.
(M.S.P.H.)--University of South Florida, 2011.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
ABSTRACT: Heat stress conditions are prevalent in the working environment around the world. Often they are not readily engineered out. Administrative controls and, in extreme/toxic environments, personal protective gear are the means available to protect workers. For every combination of metabolic work rate, clothing ensemble and environmental WBGT, there is a time of exposure threshold, beyond which the worker can no longer compensate for the heat stress, and signs and symptoms of heat strain appear. Increasingly, worker environments require specialty clothing either for worker protection or to maintain a clean/sanitary environment. Prior to the publication of the USF safe exposure time equation, no simple method was available for determining safe worker exposure time based on a clothing adjustment factor. To demonstrate the validity of the USF SET equation, both direct and indirect data from different environments, metabolic rates, and clothing ensembles were collected to compare observed tolerance times to the predicted safe exposure time. Statistical analysis was performed using the Kolmogorov-Smirnov test. The USF SET equation predicted an acceptable safe exposure time, 19 % of the trials. Based upon this data, the USF safe exposure time heat stress equation over estimates safe exposure time for workers in hot environments, in various clothing ensembles at various metabolic work rates.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Salihu, Hamisu .
x Occupational Health
t USF Electronic Theses and Dissertations.
Validation of the USF S afe E xposure T ime Equation for H eat S tress b y Arden B. Ande rsen A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Public Health Department of Environmental & Occupational Health College of Public Health University of South Florida Major P rofessor: Hamisu Salihu, M.D.,Ph.D. Thomas E. Bernard, Ph.D. Eve Hanna M D ., M.S.P.H. Dat e of Approval: December 2 1 2010 Key Words: Heat injury WBGT, clothing factor safe work environments Copyright 2011 Arden Bruce Andersen
Acknowledgments I will start by thanking the members of my thesis committee for providing guidance and support throughout th is project ; Dr Thomas Bernard who provided the leadership necessary to initiate this venture and along with Dr. Hamisu Salihu, guidance on how to Dr. Eve Hanna who contributed her expert advise time and effort as member of my thesis committee. Additional thanks are due to Don Doerr and Ken Cohen a t NASA for their assistance and suggestions. For assistance with the statistical analysis, thanks to Dr. Alfred Mbah. Partial support for my training c a me from the Suns hine Education and Research Center at USF, which is a CDC/NIOSH training program ( T42OH008438 )
i T able of Contents List of Tables i i List of Figures i ii Abstract iv Introduction 1 Literature Review 5 Methods 10 Results 21 Discussion 23 References 25
ii List of Tables Table 1 D irect Data 11 Table 2 Derived Literature Data 17
iii List of Figures Figure 1 Farm Worker Deaths 3 Figure 2 Direct Data Stratification 21 F igure 3 Indirect Data Stratification 22
iv Abstract Heat stress conditions are prevalent in the working environment around the world. Often t hey are not readily engineered out. Administrative controls and in extreme/toxic environments personal protective gear are the means available to protect workers. For every combination of metabolic work rate, clothing ensemble and environmental WBGT, ther e is a time of e xposure threshold, beyond which the worker can no longer compensate for the heat stress and signs and symptoms of heat strain appear Increasingly, worker environments require specialty clothing either for worker protection or to maintain a clean/sanitary environment. Prior to the publication of the USF safe exposure time equation, no simple method was available for determining safe worker exposure time based on a cloth ing adjustment factor. To demonstrate the validity of the USF SET equ ation, both direct and indirect data f rom different environments, metabolic rates, and clothing ensembles were collected to compare observed tolerance times to the predicted safe exposure time. Statistical analysis was performed using the Kolmogorov Smirnov test. The USF SET equation predicted an acceptable safe exposure time 1 9 % of the trials Based upon this data, the USF safe exposure time heat stress equation over estimates safe exposure time for workers in hot environments, in various clothi ng ensembles at various metabolic work rates.
1 Introduction Heat related injury occurs when the body is unable to maintain homeostasis between heat gain and heat loss from the body. The physiological response to environmental heat stress, which occurs bey ond homeostasis is termed heat strain. Heat strain manifests in progressive symptoms displayed by the worker beginning with fatigue, elevated heart rate, thirst, and headache progressing to mental status chang e and cessation of sweating and finally loss of consciousness and death. Unacclimatized workers will display heat strain signs and symptoms quicker than acclimatized workers. E ven mild heat strain can affect worker performance. NASA research data showed that telegraph key operators made five mist akes an hour, nineteen after three hours when the temperature was just 80 F (26.7 C) At 95 F (35 C), mistakes increased to 60 per hour, 138 after three hours. Dehydration further exacerbates heat strain. Wasterlund and Chaseling showed that dehydra tion equaling a 1 percent body weight loss of fluid correlated to a 12 percent decrease in productivity in forestry workers. Gopinthan et al demonstrated that a 2 percent body weight equivalent dehydration correlated to declines in vi sual motor tracking, short term memory, attention and arithmetic efficiency. A 4 percent equivalent body loss dehydration correlated to a 23 percent decline in reaction time. When the environmental temperature reaches 95 F ( 35 C ) or more, the only way the body can cool it self is via the transfer of core heat to the skin surfaces via blood and the subsequent evaporation of sweat. When the body is dehydrated, the efficiency of this cooling mechanism is impaired. (1 )
2 The NCAA completed a four year study finding that the risk of heat stress injury increases five fold when the WBGT rises above 82 F ( 27.8 C ) ( 2 ) The total number of hospitalizations with codes for heat illness among Army soldiers was 5 246 from 1980 through 2002. There were 37 deaths due to heat illness, and the mortality rate was 0.3 per 100,000 soldier s per year during the 22 y ea r period ( 3 ) Environmental Health and Safety Today reported that 2 554 workers missed work in 2000 due to heat related injury ( 4 ) Wallace found in researching marine training that cumulative heat stress from previous day s exposure is a factor in predicting exertional heat injury for that day beyond just the current WBGT suggesting that WBGT alone is inadequate for protecting workers/athletes from heat strain. ( 5 ) Military epidemio logical research suggests that exertional heat injury may increase long term mortality from organ failure including kidney, heart and liver. ( 6 ). Heat related death continues to threaten workplace safety, nationwide, especially in the agricultural industry The CDC reported 423 worker deaths from heat stress between 1992 and 2006 translating to 0.02 deaths/100,000 workers. Farm work ers had the highest death rates 68 deaths translating to 0.39 deaths/100,000 wor kers and heat related illness.( 7 ) As Figure 1 indicates, heat related deaths are actually increasing among American farm workers
3 Figure 1: Farm Worker Deaths ( 7 ) claim at $1 287 looking at claims between 1995 and 2004.( 8 ) In 1973, OSHA organized a Standards Advisory Committee for Heat Stress and heat related illness. Still by 2004, there was no specific OSHA standard for heat stress; therefore, safet y officials must rely general duty cl conditions in heat stress conditions.( 9 ) Having a viable method of predicting safe worker performance in a hot working environment gives management a means to determine if and when engineering, administrative or PPE corrections are needed for worker safety. The problem continues in answering how heat stress can be evaluated. Wet bulb globe temperature (WBGT ) is the traditional method used by the industry to set occupational s) that include NIOSH REL, ACGIH TLV and ISO 7243. The problem occurs where working conditions are extreme, e.g. tropical and desert environments, and working any time of the day would exceed the WBGT
4 recommendations. The problem also occurs in working e nvironments where workers must wear special clothing, such a s vapor barrier protective HAZMAT suits. These wor king conditions may not be tolerated for the 8 hours presumed by the usual WBGT based exposure assessment methods
5 Literature Review Many working conditions present with a combination of environmental conditions, work demands and clothing requirements that will significantly reduce the amount of time a person can work under those conditions. The recognition of this problem has led to the pu rsuit of alternative assessment schemes that can predict a safe exposure time. Navy because of significant incidences of heat related illness among Marine recruits in traini ng. WBGT was an improvement over looking at only temperature and humidity. It incorporated temperature and humidity along with wind and sun radiation. The military further identified thresholds for WBGT based upon epidemiological studies of casualty rec ords and additionally considered susceptible or vulnerable recruits and determined when training would be suspended or altered to reduce the inciden ce of heat related illness.(10 ) The limitations of WBGT become evident as environmental humidity increases, as compensation is needed for wear of special clothing, as individual acclimatization varies and as the environment becomes extreme and /or the work stress becomes extreme (11 ) Also, there is considerable room for variation in arriving at the WBGT depen ding upon equipment, conditions, operator error and interpretations of values gathered (10 ) One method based on WBGT is an approach used by most industries and formulated by ACGIH It is a table of WBGT readings corresponding to hourly work and recovery cycles for light, medium, heavy and very heavy work. WBGT has been criticized as
6 being too conservative a method for determin ing allowable work for workers who must work in extreme environments or heavy work environments such as mining in the tropics. It may indicate they should not be working in these conditions, but the fact of the matter is that they are going to work in these environments (Australia and UAE) so another method of predicting safe exposure time is need ed to protect these workers. ( 12 ) Chin Lee reported problems with ISO WBGT approach as it requires extensive calculations and interpretation not practical for real world working. He contended that there was l ed him to contend that heat strain indices correlate poorly with heat stress indices at high work levels.( 13 ) Generally methods of assessment of high heat stress may fall into two categories. Rational methods like the Belding and Hatch Heat Stress Index and the ISO Predicted Heat Strain have been used to set a prescribed work time. Alternatively, empirically based methods have also received attention. Rational Methods The most current rational model of heat stress analysis is ISO7933: Ergonomics of the thermal environment -Analytical determination and interpretation of heat stress using calculation of the predicted heat strain (PHS). ( 14 ) Under circumstances in which the body cannot maintain sufficient levels of heat loss, a safe exposure time is b ased on the time to reach a predicted body core temperature of 38 C. For longer exposures where there is a risk of dehydration, the ISO standard also provides a time limit. This model has been validated and most body core temperatures will be below 38.5 C .( 1 5 )
7 An alternative method is TWL, thermal work limit, developed in Australia and elaborated upon by Graham Bates at Curtin University, Perth, Australia in his Ph.D. dissertation in 2002. (16 ). This equ ation evolved out of the need for the industry to develop a environments of the Australian mining industry and the UAE construction industry. Brake and colleagues recognized that under high heat stress con ditions, people reduce their work demands to maintain th ermal equilibrium. Brake monitored heat stress responses in mine workers in Australia and subsequently developed the Thermal Work Limit (TWL) as an index of thermal stress which is fundamentally a r ational model but interesting in the work limiting thought process. (16 ) The result of the analysis is an average rate of work that can be sustained. It has been used in the extreme conditions encountered by miners in Australia and construction workers in the United Arab WBGT, which was found t ( 1 7 ) As evaluated, workers wore standard cotton work clothing. Empirical Methods The occupational exposure limits based on WBGT are empirical limits to heat stress that assume that the exposures will be repeated throughout an 8 hour day. Developing w ork/ r est cycles that limit the overall time weighted average to the exposure limit is a me thod of setting an exposure time limit These methods are straightforward but
8 potentially over protective. For this reason, alternative empirical methods have been explored. evere hea in the mid rectangular hyperbolic curve with y=b+c/x a with x=p(db) + (1 p)(wb); y is tolerance time; a, b, and c are constants and x the weighted sum of the wet and dry bulb te mperatures. It was developed by studying test subjects in the lab under heat conditions and work until an observer determined that heat strain collapse was imminent and the trial was ended and data recorded. A safety factor was taken using the 95 th perce ntile confidence limit.(18 ) The corresponding hyperbolic curves predict maximum tolerance time under various wet and dry bulb conditions for workers with primary activities of sitting, standing or were based on standard work clo thes. The maximum worker tolerance was defined as near complete collapse. Bell claimed his equation allowed for a margin of safety based on individual differences. This claim is belied by a table in his paper that sho w ed the predicted time was often gre ater than the mean observed tolerance time. Further, collapse as a criterion was too aggressive to be used for occupational exposures. The small range of activities and aggressive criterion eliminate d consideration of B e limits. The US Navy has many locations for which the environmental conditions would limit the amount of work time. For this reason, Da sler developed empirical limits called physiological heat exposure limits (PHEL s ). ( 19 ) The environmental conditions were represented by WBGT and there were six separate curves for light to moderate metabolic rates. The clothing was the u tilities uniform The time limits were based on a body core
9 temperature reaching 39 C. Other than a higher body core temperature than is usually considered as a target for occupational exposures, the PHEL charts have been useful fo r setting acceptable work times for up to six hours. It is interesting to note that the six hour limits were lower than the commonly accepted WBGT limits for eight hours at the same metabolic rates. Bernard and Ashley (2009) developed the USF safe exposure time equation with the intent of using WBGT as the environmental index and the well established Clothing Adj ustment Factors (CAFs). This equation includes WBGT a djusted for metabolic work rate and CAF. It does assume short sustained ex posures of 120 minutes or less. ( 20 ) The following is their formulation of a safe exposure time (SET). SET [min] = 26000/(AdjWB GT[ o C WBGT] TLV) 3 + 10 = 26000/(WBGT measured + CAF 0.02(365 M[W]) (56.7 11.5log 10 M)) 3 + 10 The purpose of this effort is to validate the USF safe exposure time (SET) model with independent data.
10 M ethods In order to validate the USF SET equation independent data are needed. The validation process can be done in two ways: collecting direct data from relevant laboratory or field trials and collecting indirect data inferred from the literature. Both types o f data were collected and used in this validation process. Direct Data For the direct data source, individual tolerance times were taken from trials performed at USF for DuPont under DOD contract and for MSA and Scott Paper Company In addition, there were two trials performed at NASA. These data were not published but were available in the study records The direct data provided assessment opportunities using different clothing ensembles in different heat stress environments These data are summarized by study and clothing ensemble in Table 1. The summary data were then used to calculate the predicted safe exposure time using the USF SET equation and also reported in Table 1.
11 Table 1. Summary of direct d ata by study, clothi ng ensemble and environment. Trial Information WBGT (C WBGT) CAF (C WBGT) Metabolic Rate (W) Mean Obs.Time (min) Predicted SET (min) DuPont 1 Control, DI N=25 28.8 3 344 6212 > 120 Control, DII N=8 23.9 3 338 105.916 > 120 Control, J N=8 28.1 3 374 112.812 > 120 BaseSPM, DI N=8 28.8 11 353 31.24.9 25 BaseSPM, DII N=8 23.9 11 357 55.810.5 75 BaseSPM, J N=8 28.1 11 350 96.228 28 PropSPM, DI N=8 28.8 3.5 379 334 > 120 PropSPM, DII N=8 23.9 3.5 343.7 75.426.6 > 120 PropSPM, J N=8 28.1 3.5 360 10216.5 > 120 DuPont II Control, DI N=8 28.8 3 334 619 >120 SPMA, DI N=8 28.8 11 333 42.759.9 27 SPMB, DI N=8 28.8 3.5 331.5 46.754.3 > 120 SPMC, DI N=8 28.8 3.5 338 36 >120 DuPont III Control, DI N=8 28.8 3 367 6315 >120 AB A, DI N=8 28.8 3.5 351 47 6 >120 AB A, J N=4 28.8 3.5 323 78 24 >120 A B B, DI N=8 28.8 3.5 350 45 8 >120 MSA Phase I Vapor Barrier N=5 32 11 27171 318 24 MSA Phase 2 Vapor Barrier, N=5 32 11 27662 3112 23 Scott Study ComfortGard I N=5 32 1 25439 8028 >120 ComfortGard II N=5 32 4 26256 7217 >120 ComfortGard III N=2 32 4 26225 725 >120 Tyvek 1422A N=5 32 2 27570 5012 >120 NASA NASA Vapor Barrier N=2 37.05 11 300 44 14
12 General Description of USF Studies The combinations of clothing and heat stress level were assigned to participants in random order ; the schedule for each participant was random as well Participants were monitored for rectal temperature and heart rate while walking on a Burdick T500 treadmill between 2.5 and 3.0 mph with no grade at about 190 W/ m 2 Metabolic rate was calculated from oxygen consumption, sampled at approximately 30 minute intervals. Trial termination criteria were when 1) rectal temperature reached 38.5 C, 2) maximum age predicted heart rate (0.85*[220 Age]) was reached, 3) participant asked to stop, or 4) participant accomplished 120 minutes on the treadmill without reac hing any of the other termination criteria. DuPont Studies All t he DuPont/DOD studies were similar in that they examined potential chemical protective ensembles (sometimes known as MOPP gear). The studies used similar environments and metabolic rates bu t differed in the clothing. Three environments, distinguished by the relative humidity with no radiant heat source, were selected for the study. The environmental conditions were: Jungle: 35 C at 50% relative humidity (Vp = 2.81 kPa) Desert I: 49 C at 20% relative humidity (Vp = 2.35 kPa) Desert II: 40 C at 30% relative humidity (Vp = 2.21 kPa)
13 DuPont 1 In the first study the following three clothing ensembles and three environments (Jungle, Desert I and Desert II) were examined in a full fac torial design with eight participants The primary ensembles based on fabric were: Control: The current standard (Saratoga Hammer) ensemble Base SPM Fabric Proprietary SPM Fabric The summary information including tolerance times is reported in Table 1 DuPont 2 In the second DuPont study the following four clothing ensembles and one environment (Desert I) were examined in a full factorial design with eight participants. The prima ry ensembles based on fabric were: Control: The current standard (Saratoga Hammer) ensemble SPM A SPM B SPM C
14 The order of the ensembles was balanced to minimize the effects of order. A cotton tee shirt and gym shorts were worn under the clothing as the base ensemble. The summary information including tolerance times is reported in Table 1. DuPont 3 In the third DuPont study the following five clothing ensembles and one environment (Desert I) were examined in a full factorial design with eight pa rticipants. In addition a sub study with one ensemble ( AB A) in the Jungle environment was examined with four participants The primary ensembles based on fabric were: Control: The current standard (Saratoga Hammer) ensemble AB A (Internal Control us ing standard SPM) AB B (No porosity) AB C ( Porosity 1 ) AB D ( Porosity 2) The order of the ensembles was balanced to minimize the effects of order. A cotton tee shirt and gym shorts were worn under the clothing as the base ensemble. The summary information including tolerance times is reported in Table 1.
15 Mine Safety Appliances Company (MSA) The MSA studies were performed in two phases to test different cooling systems against a no cooling control The no cooling control was a vapor barrier ens emble with hood and respirator The environment for both phases was representative of the Gulf Coast in the summer time with T db = 35 C and rh = 55% (WBGT = 32 C WBGT) The summary information including tolerance times is reported in Table 1. Scott Pape r Company The Scott Paper Company studies, conducted with three clothing ensembles in one environment (Gulf Coast) were examined in a full factorial design with five participants with a sub study for one clothing ensemble (Tyvek). The primary ensembles bas ed upon fabric were: Comfort Gard I Comfort Gard II Comfort Gard III The order of the ensembles was balanced to minimize the effects of order. A cotton tee shirt and gym shorts were worn under the clothing as the base ensemble. The summary information including tolerance times is reported in Table 1.
16 NASA Trials Two individual trials at NASA were observed. Termination criteria for the NASA trial was 38.9 C so work times were standar dized to 38. 5 C core body tempe rature using a factor of 0.8.( 21 ) Th e correction factor of 0.8 was calculated by taking 38.5 minus 36.8 divided by 39.0 minus 36.8 yielding 0.773 rounded to 0.8 The summary information for the two NASA trials including tolerance times is reported in Table 1. Indirect Data The indirect data was taken from data reported in studies providing clothing description, metabolic work rate, WBGT, observed working times and trial termination criteria tha t could be standardized to 38.5 C core body temperature. Mean values were either given or calcula ted for each study and a 5 th percentile time was calculated that was 95 percent protective. The descriptive data from the reported studies are provided in Table 2. The assigned data were then used to calculate the predicted safe exposure time using the U SF SET equation and reported in Table 2
17 Table 2: Derived literature Data Study WBGT CAF Work(W) Obs. Work time Ave. Time Std Protective Work Time Predicted SET Zhang N=10 30 0 465 7426 50 24 132 Bishop N=14 26 11 430 13511.9 95 55 233 Bell 1 37.4 0 325 55.19.8 45 31 48 Bell 2 40.9 0 325 31.45.3 22 15.4 24 Bell 3 43.9 0 325 19.62.1 14 12 17 Bell 4 46.9 0 325 15.22.1 9 7 14 Bell 5 51.8 0 325 9.92.2 9 5.4 12 Muir 1 18 11 450 109.219.6 98 70 293 Muir 2 23 11 450 62.37.7 56 51 40 Muir 3 28 11 450 42.56.1 39 34 19 Hostler 18.45 11 355 45 41 17 121 Zhang et al Zhang, et al conducted a trial with 10 individuals to test the effectiveness of a carbon dioxide cooling device imbedded in the work clothing. ( 22 ) He conduc ted a non cooling control trial which was the trial of interest. Extracting rectal temperature and exposure time, 38.5 C correspond ed to 50 minutes exposure time at 465 W of metabolic work in cotton shirts and pants The CAF was taken to be 0. Adjusting the exposure time plot by 1.65 Z score correlating to the 95% safety factor equated to 24 minutes as the safe exposure time. Bishop, Nunneley and Constable Bishop, Nunneley and C onstable conducted a trial with 14 individuals to test the effectiveness of intermittent cooling in emergency response workers wearing chemical
18 protective clothing. ( 23 ) The non cooling control trial data was used in this study. Extrapolating from non cooling gra ph data by determining the standard deviation of the data times the 1.65 Z score corresponding to the 9 th percentile protective factor gives us 55 minutes of exposure time corresponding to 38.5 C core temperature termination criteria. Trial conditions in cluded WBGT of 26 C, metabolic work rate of 430 W and a clothing factor of 4. Bell et al Bell et al. conducted a trial with eight men in various different environments and metabolic work rates from sitting, standing and working. (18 ) Data was taken only from the working trials of 37.4 to 51.8 C at a working metabolic rate of 325 W and a clothing adjustment factor of 0 for cotton boiler suites. Since Bell use d oral temperatures, 0.5 C was added to approximate the body core te mperatures. For each trial, a time adjustment factor was done to standardize the calculated core body temperatures to 38.5 o C as the termination point for the trials. This was needed as Bell continued the trials almost to the point of participant collapse Standardization was done by taking 38.5 36.5 divided by the calculated core body temperature for the trial minus 36.8. For the 37.4 WBGT trial this equated to 38.5 36.5 di vided by 38.9 36.5 which equaled 5 minutes which ; multiplied by 0. 8 equaled 45 minutes, average. Taking the standard deviation of 9.8 for this t rial multiplied time s 1.65 equaled 16; subtracted from 55 gives 39, multiplied by the temperature sta ndardization factor of 0.8 gave 31 minutes protected time. Each trial was
19 adjusted/standardized accordingly to arrive at both the average times worked until core body temperatures reached 38.5 C and t he safe work times to the 95 th percentile Muir, Bishop, and Ray Muir, B ishop and R ay conducted a trial with 6 male participants to test a new ice cooling system for impermeable protective clothing having a clothing adjustment factor of 11. (2 4 ) Data from the control trial was extracted for 18 23 and 28 C WBGT at a metabolic work rate of 450 W. Rectal temperatures were recorded and were adjusted Hos tler et al. Hostler et al. conducted a study with ten male participants varyin g participant hydration status using a control and a test with hyperhydration via intravenous fluid replacemen t. (2 5 ) The control data was used from this study. The clothing adjustment factor was 11 for the chemical resistant enclosed clothing ensemble and air purifying respirator. Exposure times were adjusted to reflect standardized maximum heart rates giving an adjustment factor of 0.9; multiplied time s the average exposure time of 45 minutes g ave 41 minutes average exposure time for the trial. Takin g the standard deviation and adjusting it to the 95 th percentile safety factor g ave 17 minutes of protected exposure time.
20 Combined Data Analysis T ables 1 and 2 summarized t he representative data points for WBGT, clothing adjustment factor, and metabolic work rate These values were then used for the USF SET equation to predict a safe exposure time If the computed time was greater than 120 minutes, the predicted time was set to 121 minutes In comparing the predicted to observed time, if the observed t ime was less than or equal to the predicted time the outcome of the decision was For an observed value less than the predicted, the decision was Data were analyzed using the Kolmogorov Smirnov test.
21 Results The purpose of this study was to validate the USF Safe Exposure Time (SET) equation using direct data from DuPont, MSA, Scott Paper and NASA studies and indirect data from the literature. Figure 2 displays the relationship between the predicted and observed time for the direct data. A n identity line was provided in the figure for a reference point such that data to Of all the observations, 30 of 160 or 19% of direct data points f e ll below the line in contrast to 120 or 81 % of direct data points above the line. Figure 2. Comparison of observed to predicted time for the direct data. Observations to the right of the identity line are protective predictions. 0 30 60 90 120 0 30 60 90 120 Predicted Safe Time [min] Observed Time [min]
22 In a similar fashion, Figure 3 represents the comparison be tween predicted and observed for the indirect data. Though several points are near or below the line, adding the 5 th percentile safety factor places 9/11 data points above the line The two points below the line correspond to a CAF of 11. Figure 3. Comparison of observed to predicted time for the indirect data (mean with whisker for 5 th percent ile). Observations to the right of the identity line are protective predictions.
23 Discussion Though there exists several methods for determining worker exposure under conditions of heat stress, none is universally accepted for extreme conditions. The USF SET equation was an attempt to consider environmental conditions through the widely accepted WBGT, the work dema nds through metabolic rate and clothing effects through the Clothing Adjustment Factor (CAF) While the equation was based on extensive data over five levels of heat stress and three clothing ensembles, the work demands were limited to a narrow range aro und 3 8 0 W. The method was not validated against other data. That was the purpose of the current study. It was clear from Figures 1 and 2 that there were extensive under predictions of a safe exposure time compared to the observed times That is, direc t data collected in the same laboratory over a 10 year period with different participants produced much different results. This was further supported by the indirect data gather ed from reports in the literature. While there might be errors in how the dat a were extracted and interpreted from the literature, the overall pattern was similar Using 171 individual trial data points from both the direct and indirect data, 32 of 171 times the USF SET equation predicted a safe exposure time that was less than ei ther the actual observed tolerance time of the worker in the trial or the protective time attributed to the trial, or about 19% of the time The USF SET equation was based on metabolic rates near 380 W. The average metabolic rate for the direct data was 3 35 W and it was 377 W for the indirect data. Given the overall closeness of the average metabolic rates among the SET, direct and
24 indirect data, metabolic rate might have been a contributor through wider variations from the mean. There was some indicatio n that the further the lower the metabolic rate the great er the loss of protection. This suggested that the SET equation is too sensitive to changes in metabolic rate from 380 W. Looking at high CAF clothing, 20 of the 30 points falling below the line corresponded to a CAF of 11 C WBGT. It was interesting to note that the ensembles assigned a CAF of 11 C WBGT were the same or similar to the vapor barrier ensemble used in the USF SET study, but the design of the study used a much lower CAF (6.5 C WBG T) knowing that the value of 11 C WBGT would be protective. The reason for the difference was due to the sensitivity of vapor barrier clothing to the prevailing water vapor pressure. This observation about the differences between observed and predicted would suggest that the USF SET does not handle the clothing differences very well and that much higher value s are required to obtain necessary protection. Thus far, the USF SET over estimates safe exposure time in heat stress conditions and for this reason its u se should be limited and double checked against another form of assigning safe exposure time Additional research is needed to determine what changes or adjustments to the USF safe exposure time equation will produce protective predicted safe expos ure time s using this equation.
25 References 1. Baker, Bruce and John LaDue. (2010, May 1) How Heat Stress Affects Performance. Occupational Health and Safety Retrieved December 4, 2010 from http://ohsonline.com/articles/2010/05/01/how heat stress affects performance.aspx?admgarea=ht.HeatStressQuenchers 2. Childs, Michael (2009, July 14) UGA researchers to study when heat risks rise for prep football players University of Georgia Office of Publi c Affairs. Retrieved December 4, 2010 from http://www.uga.edu/news/artman/publish/090714_HeatInjuries.shtml 3. Carter, III, Robert, Cheuvront, Samuael N., Williams, Jeffrey O., Kolka, Margaret A., Stephenson, Lou A., Sawka, Michael N., and Paul J. Amoroso. Epidemiology of Hospitalizations and Deaths from Heat Illness: Res ults Medicine and Science in Sports and Exercise 37(8):1338 1344 (2005) http://www.medscape.com/viewarticle/512284_3 4. Smith, Sandy. EHS Today. Retrieved December 4, 2010 from http://ehstoday.com/news/ehs_imp_35502/ 5. Wallace, Robert F. The Effects of Continuous Hot Weather Training on Risk of Exertional Heat Illness. Medicine & Science in Sports & Exercise 37(1):84 90, January 2005. 6. Excertion al Heat Stroke Return to Duty/Return to Play: Conference Proceedings. Curr. Sports Med. Rep. 9(5), 314 321 (2010). 7. Heat Related Deaths Among Crop Workers --United States, 1992 -2006 CDC MMWR Weekly June 20, 2008 / 57(24);649 653. Accessed online Decem ber 4 2010 ( http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5724a1.htm ) 8. Bonauto, MD, MPH, David, Burke, MD, Brian, Rauser, PE, Edmund, and Robert Anderson, CIH, CSP, MSPH. Heat related Illness in Washington State, State fund 2004. Technical Report Number 59 1 2006. Safety and Health Assessment and Research for Prevention (SHARP) Program Washington State Department of Labor and Industries. January 2006. Retriev ed December 4, 2010 from http://www.lni.wa.gov/Safety/Research/files/HeatRelatedIllness.pdf 9. Margaret Wan, MSPH. Occupational Exposure to Hot Environments: Florida Workers Need Help. Florida Public Health Review 2004:1:53 55. 10. Budd, Grahame M., Wet bulb globe temperature (WBGT) its history and its limitations. J. Sci. Med. Sport 11(1):20 32 (January 2008). 11. Rastogi S. K., Gupta, B. N. and T. Husain. Wet bulb globe temperature index: a predictor of physiological strain in hot environments. Occup. Med 42:93 97 (1992). 12. Parsons, Ken. Heat Stress Standard ISO 7243 and its Global Application. Ind. Hlth 44:368 379 (2006)
26 13. Lee, Chin Hyong, B.E., M.S., M.S. in I.E. Effects of Wick contamination and Thermal Component Variation on Thermal Indices. Texas Tech University Doctoral Dissertation. May 1996. Accessed online December 4, 2010 from http://etd.lib.ttu.edu/theses/availa ble/etd 03252009 31295004913736/unrestricted/31295004913736.pdf 14. International Organization for Standardization (ISO): Ergonomics of the thermal environment -Analytical determination and interpretation of heat stress using calculation of the predicted hea t strain ( ISO 7933:2004) [Standard]. Geneva: ISO, 2004. 15. Malchaire, J., Piette, A., Kampmann, B., Mehnert, P., Gebhardtu, H., Havenith, G., Den Hartog, E., Holmer, I., Parsons, I., Alfano, G., and B. Griefahn. Development and Validation of the Predicted Heat Strain Model. Ann. Occup. Hyg. 45(2):123 135, (2001) 16. Brake, Derrick John. The Deep Body Core Temperatures, Physical Fatigue and Fluid Status of Thermally Stressed Workers and the Development of Thermal Work Limit as an Index of heat Stress. Curtin Un iversity of Technology, School of Public Health Doctoral Dissertation, Australia, 2002. 17. Miller, Veronica S., and Graham P. Bates. The Thermal Work Limit Is a Simple Reliable heat Index for the Protection of Workers in Thermally Stressful Environments. Ann. Occup. Hyg 51(6):553 561. 18. Bell, C. R., Hellon, R. F., Hiorns, R. W., Nicol, P. B., and Provins, K. A. Safe exposure of men to severe heat. J. Appl. Physiol 20(2):288 292 (1965) 19. Dasler A.R. Heat Stress, Work function and physiological heat Exposure Limits in Man. Defense Technical Information Center. Naval Medical Research Institute, Bethesda, February 1977. 20. Bernard, Thomas E. and Candi D. Ashley. Short Term Heat Stress Exposure Limits Based on Wet Bulb Globe Temperature Adjusted for Clothing and Metabolic Rate. J. Occup. Enviro. Hyg. 6:632 638 (2009) 21. Doerr, Don and Ken Cohen. Unpublished Data. Physiological testing of chemical HAZMAT suits. Medical and Environmental Management Divisio n, Kennedy Space Center, August 2010. 22. Zhang, Yang, Bishop, Phillip A., Green, James Matthew, Richardson, Mark T., Schumacker, and E. Randall. Evaluation of a Carbon Dioxide personal Cooling Device for Workers in Hot Environments. J. Occup. Environ. Hyg 7: 389 396 (2010). 23. Bishop, Phillip A., Nunneley, Sarah A., Constable, Stefan H. Comparisons of Air and Liquid Personal Cooling for Intermittent heavy work in Moderate Temperatures. Am. Ind. Hyg. Assoc. J 52(9):393 397 (1991). 24. Muir, Ian H., Bishop, Phillip A. and Paul Ray. Effects of a Novel ice Cooling Technique on Work in protective Clothing at 28 C, 23 C, and 18 C WBGTs. Am. Ind. Hyg. Assoc. J 60:96 104 (1999). 25. Hostler, David, Gallagher Us., Michael, Goss, Fredric L., Seitz, Jennifer R., Reis, Steven E., Robertson, Robert J., Northington, William E., and Joe Suyama. The effect of hyperhydration on physiological and perceived strain during treadmill exercise in personal protective equipment. Eur. J. Appl. Physiol 105:607 613 (2009)