Workers self-pacing in hot conditions: A case study

Applied Ergonomics 1985, 16.2, 85-90

Workers self-pacing in hot conditions: A case study Ph. Mairiaux and J. Malchaire Industrial Hygiene and Work Physiology Unit, Universit~ Catholique de Louvain, BP 3038, B-1200 Bruxelles, Belgium

It is not always possible in hot environments, to determine safe work-rest regimens based upon heat stress criteria. Is it then a good health policy to rely upon the self-pacing of the workers? To address this question, we observed the spontaneous work-rest cycles of seven masonry workers allocated to the maintenance of a float-glass furnace and the workers' heart rate and rectal temperature were continuously monitored. Results showed that the regimens adopted by the workers were poorly related to physiological parameters. The duration of each heat exposure was significantly related to the worker aerobic capacity, but not to the heart rate level reached at the end of the working period. Due to self-pacing of the task, heart rates and rectal temperatures remained within acceptable limits for all workers but one, even though these workers were rather old and had a rather low physical capacity. These favourable results may be ascribed to a spontaneous limitation of the effective working time to about 25% of the shift duration and to the fact that subjects worked, in pairs alternately, to perform the task. It is concluded that self-regulation of the work-rest cycles can be an effective means to protect the workers exposed to hot conditions from an excessive physiological strain, providing that the task has no urgent character and does not involve productivity incentives, and that the workers are well trained to their job.

Keywords: Thermal stress, heart rate, working conditions (physical) I ntroduction The prevention of heat disorders among workers exposed to hot ambient conditions has led to the formulation of various heat stress criteria. In the USA, the ACGIH has proposed threshold limit values (TLV) for heat stress which are based on the WBGT index. These TSVs were adopted in Belgium, as part of the general regulations aiming at the workers' health protection. The Belgian regulation makes precise work-rest regimens compulsory as soon as the WBGT value reaches 30" 1°C, 26"8°C and 25.1°C for light, moderate and heavy workloads respectively, and theoretically prohibits people from working when WBGT values are above 33°C, 32°C and 31 "5°C respectively. However, these work-rest regimens are only applicable when the subject works and rests in the same ambient conditions. This is, of course, not the case in extreme conditions. When work is performed and rest is taken in different environments and when the working conditions are above the permissible exposure limits, the TLVs cannot be used and alternative methods have been suggested to determine the optimal work-rest regimen to lower the risk of heat-induced illnesses.

0003-6870/85/02 0085-06 $03.00© Buttetworth& Co (Publishers)Ltd

The required sweat rate index is currently proposed as a new standard (ISO, 1983). It provides heat exposure limits based on three physiological criteria: maximal sweat rate, body heat storage and dehydration level. The use of this index requires the determination of the metabolic rate, the clothing insulation value and the environmental parameters. Kamon (1979) has suggested that cycles of work for hot ambient conditions should be designed on the basis of the expected heart rate increase induced by the work-load and the ambient parameters. Both approaches may have some drawbacks when applied to a hot situation. The first difficulty arises from the frequent variations in the heat load, for these methods are in fact best suited to steady and non-fluctuating ambient conditions. Heat protective clothing provides a second difficulty: neither the required sweat rate index nor the method proposed by Kamon takes into account the wearing of such special clothing. Therefore, in these situations, one is usually bound to rely upon the workers' spontaneous work-rest regimens. The value of this was opposed by Brouha (1960) who advocated the managerial control of the work-rest schedule on a

Applied Ergonomics

June 1985

85

physiological basis while for others (Millican et al, 1981), the workers' self-pacing seems to be one of the most effective measures to prevent, the occurrence of heat illnesses. The present study, conducted in a glass factory, provided an opportunity to re-examine this question, by observing the spontaneous work-rest schedules of a team of masonry workers. The aim of the study was to assess the effectiveness of this self-pacing as regards the limitation of the heatinduced physiological strain.

Methods

Workers Seven masonry workers, from a team of eight, participated in the study. Six of them volunteered to undergo a maximal exercise test on a cycle ergometer for the determination of their maximal oxygen uptake (VO: max). The physical characteristics of these workers and their length of time on the job are given in Table 1. All workers except No 4 were well trained to their job. Worker No 2 was previously allocated, for 25 years, to another hot department of the same plant. At the time of the study, all workers could be considered as heat-acclimatised. The standard protective equipment of the workers consisted of heat-resistant aluminised vest, trousers and cap, wooden shoes and gloves. Some workers also wore goggles, and a scarf wrapped around the neck and the mouth. Depending upon the ambient conditions, the workers used the whole protective equipment or some parts of it only.

Task description The task was performed on a fixed morning-shift basis from 6 am to 12 noon. Salaries were determined on an 8 hrshift basis to compensate for the environmental conditions of the task; no other productivity incentive was given. The workers were in charge of the cleaning of the vault and the repairing of the walls of two float glass furnaces. The nature and the frequency and duration of the masonry work were dependent upon the circumstances, whilst the cleaning of the vault was repeated systematically day after day. By means of a 2 m long pipe connected to compressed air, the workers had to blow the dust deposited on top of the furnace

Table 1: Workers' characteristics

Worker

86

Age (yrs)

Length on the job (yrs)

Weight (kg)

V O 2 max HR max (I/min) (beats/min)

1

46

7

89

2-71

183

2

46

1

76

2-80

178

3

51

17

65

1"95

185

4

37

0-3

72

2"34

200

5

43

8

65

2"50

195

6

48

15

79

2-64

188

7

24

7

83

-

-

Applied Ergonomics

June 1985

Fig. 1

View of the burner's vaults and the workers' gangway from the middle of the furnace main vault.

progressively towards the central part of the vault, where it was collected and evacuated. During the blowing of the dust, the workers stayed on a gangway located above the burner's vaults (see Fig. 1). This task was most often carried out by a pair of workers: one blowing the dust and exposed to the heat, the other carrying the pipe but remaining at some distance in a zone where the heat was less severe. After each working period, the workers went down and rested beside the furnace. A 35 min rest pause was scheduled at 8 am during which the workers had a cold snack together in a small canteen. Apart from that 'snack-pause', the workrest cycles were completely free, the only requirement being to complete the cleaning of both furnaces in five consecutive shifts. The cleaning task could thus be considered as an actual self-paced task with no productivity incentive.

Measurements Continuous ECG recordings were obtained by means of three chest electrodes connected to a Medilog (Oxford Instruments) tape recorder carried at the waist. These recordings were analysed using a Digital MINC 11-23 computer to obtain instantaneous and averaged heart rate (HR) values for each working and resting period, together with cumulative HR histograms. A rectal thermistor probe (YSI 701) was introduced 11 cm beyond the anal sphincter. Readings of rectal temperature (Tre) level were made, during the resting periods only, by means of a Digitec (model 5810) thermometer. The very high ambient temperatures at the workplace did not allow the measurement of oxygen uptake during work. Energy expenditure during the blowing and sweeping of dust was estimated to be 220 W (0-65 1/rain ~O 2) from the combined assessment of body position, body motion and arm work. The heaviest work corresponded to the climbing of the stairs from the foot to the top of the furnace. During each of the observation shifts, dry and wet bulb temperatures (T a, Twb ) were measured at a height of 1-2 m from the ground using a psychrometer Haenni; for technical reasons, it was not possible to record globe temperatures above 65°C.

Work analysis

T a was nearly constant at 21°C. The averaged temperatures T a measured at the top of the furnace A are described in Fig. 2. Globe temperatures were higher than 65°C and WBGT values were above 34°C at any point of the vault. Configuration of furnace B was identical but T a levels were, on average, 15°C to 20°C lower. The ambient vapour pressure calculated from the recorded Twb values varied from 1-0 kPa to 1.6 kPa. As shown by Fig. 2, T a varied considerably along the furnace. On the two lateral gangways, where the workers were mostly standing during their job, the temperature could vary by more than 20°C from one point to another 2 m apart.

The pattern of activity of these workers was observed during 11 working shifts. The masonry task was observed during two shifts, and the cleaning task during nine shifts. Three workers were monitored on two occasions while performing the same task. The observation time corresponded to the effective working time but was somewhat shorter than the 6 h duration of the shift as some time was needed to fit the ECG and Tre recording system to the worker at the beginning of the shift and to remove it at the end.

Results

Fig. 3 illustrates the evolution over a whole shift of heart rate and rectal temperature for subjects 1 and 3 while working together above the vault. The HR patterns of the two workers were not identical, as they took turns in performing the task and in assisting from some distance in a less heat exposed location. The rectal temperature variation of worker 1 involved two phases of increase separated by a steady level during the snack pause. When work was resumed, Tre rapidly levelled off to reach a near steady-state until the end of the shift. The establishment of such a steady-state in the second half of the shift was observed in all observations but one.

The observations were carried ou,t during the months of August and September. The air temperature outside varied from 12°C to 2 I°C (September) and 29°C (August). Air temperatures measured inside at the resting place ranged from 17°C to 30°C, and varied by 5 to 7°C from 6 am to midday. In the canteen, where the workers stayed at 8 am,

Air temperatures °C

The spontaneous work-rest regimens during the 11 shifts of observation are detailed in Fig. 4. From top to bottom of the graph are figured the nine shifts involving the cleaning task and the two shifts (~) involving masonry work. A preliminary survey of the results showed that the physiological monitoring procedures at the start of the shift had interacted with the usual working pattern before the 8 am rest pause. The time left after the fitting of the electrodes was often too short to allow two separate working periods to take place before this pause and therefore the worker tried to carry out an equivalent amount of work in one longer working period. This was considered to be an experimental bias, and the analysis of the work-rest regimens was restricted to those following the 8 am rest pause.

___l Furnace vault 5:3

62

--1

64

68

90

82

~--

J

Fig. 2

Averaged air temperatures (Ta) measured at various locations of furnace A vaults. Temperatures above the gas burners are noted by stars on the graph.

175

FC, beats/rain

75

E .

.

.

38.0

Tre, °C

37.0

......

';:

E

.

.

.

.

.

.

.

.

.

.

.

.

m

.

.

.

.

.

Subj I J

.... " ~ -

N

I 8

. . . . . . .

....

Fig. 3

be""_': 'n . . . . I 7

.

Fig. 4 shows that the partition of work and rest periods was not systematic at all but varied from worker to worker, and for the same worker from one shift to another. This variability in work-rest patterns was examined with regard to possible physiological correlates: the worker fitness, the working HR level and the HR level at the end of pauses.

s"b2._s .....

N

I 9

N

Time (h)

NI I0

B

I II

B

Heart rate variations of workers 1 and 3 during the cleaning task; dashed line for HR level at 125 beats/min. Rectal temperature variation of worker 1 in the same conditions. The working periods above the vaults are figured by hatched areas.

Applied ~Ergonomics • J u n e 4 9 8 , ~ j

.

87

s~{o)

El

{b)

I~1

B

s~

I~

~

~

:-

Et

EYJ

[El

fl I 8

B

~

I~

F~

i 7

Fig. 4

I~

F~i

s3

s7

B

B

FJ _ _ B

F~

B

J 9

lower degree tor workers 2, 4 and 5. For workers 1 and 3, the resting period durations appeared to be independent of the working HRs.

B

B

F~

B ] I0

_

Et

B / 12h

I It

Observed work-rest regimens among the seven workers. For each shift, the length of the line corresponds to the observation time and the working periods are figured by hatched areas.

The effectiveness of the spontaneous work-rest regimens in terms of physiological strain limitation can be evaluated on the basis of data in Tables 2 and 3. Table 2 gives, for each worker, the HR evolution throughout the successive working and resting periods. No creeping up of the HR level could be observed among the workers, worker 6 (shift b) excepted. Table 3 summarises the overall results of the HR and Tre monitoring over the shift. The total heat exposure duration ranged from 9% to 35% of the shift. As regards the cleaning task, this ratio seemed to be fairly constant, whatever the worker, at about 25%. For subject 1, the heat exposure duration was longer when he worked alone (shift b) than while he was paired with another worker (shift a). For subject 4 (shift b), the working duration was longer as it included non-typical subsidiary tasks in less severe environmental conditions. Peak HRs during exposure to heat (HR 1%) ranged from 141 to 170 beats/min. The mean HR level averaged over the whole shift exceeded 1 t 0 beats/ rain on three occasions: shifts a and b for worker 4 (cleaning task) and shift b for worker 6 (masonry work). The rectal temperature increase from the start to the end of the shift was limited to about I°C for all workers except worker 6 (shift b). At the end of the shift, rectal temperatures reached more than 38°C for workers 4 and 6.

Discussion

The duration of each working period was plotted against the worker VO 2max (ml kg/min). This analysis involved only eight shifts where the usual pair-work organisation was observed. The duration of each heat exposure was significantly related to the worker physical fitness by y = 1.065 x - 23"2 (R = 0.63, df = 32, p < 0-001). However, this relationship contributed a mere 40% to the total variance of the working period duration. The duration of the rest pauses was not related to that of the preceding working period but appeared to be positively correlated with the working HR levels for worker 6 (p < 0-05) and to a

The main question addressed by this case-study was the degree of health protection provided by the workers' selfregulation during exposure to hot ambient conditions. Owing to the continuous monitoring of HR and Tre, four criteria of strain could be considered! a mean HR level above 110 beats/rain, a rectal temperature above 38°C at the end of the shift (WHO 1969) and a rising trend in HR, or in Tre, with the exposure. Out of the seven workers monitored, two (Nos 4 and 6) presented one or more of these strain criteria. The four criteria were observed together on only one occasion, for worker 6 during shift b. However, these particular

TabW 2: Heartratelevelsobserved atthe end ofthe working(HRw) and resting periods(HR r)

Shift

1

a

120

b b

2

136

136

130

145

147

148

127

151

126

127

147

140

121

126

a

166

154

147

152

b

175

137

164

135

127

142

a

149

b

3 4 5 6 7

811

HR w (beats/min)

Worker

AINllied Etllmtomi¢=

HR r (beats/min) 89

95

95

100

101

97

79

81

88

97

103

103

107

120

110

118

111

138

85

89

168

142

104

103

129

151

156

165

161

98

109

120

116

117

147

132

144

125

93

98

92

97

June 1985

136

127 174

126

140

91

92

89 109

105

Table 3: Results of the HR and Tre monitoring

Mean HR Mean HR working HR whole periods 1% shift ATre (beats/min) (beats/rain) (beats/rain) (°C)

Worker

Shift

Heat exposure duration Working time

1

a

0"24

119

141

101

1.0

38"0

b

0"33

133

153

112

0.9

38"0

a

0"25

-

-

-

0"5

37"6

2

b

0"22

119

148

93

0"4

37"5

0"22

123

145

103

-

-

a

0.21

144

169

121

0-8

38"1

b

0"35

138

170

123

0.7

38"2

3

4 5 6 7

Tref (°C)

0"28

113

142

97

0.7

37-6

a

0"22

139

163

111

1.1

38"3

b

0"28

13R

164

116

1.4

38"7

0"09

132

152

102

0.7

37"8

HR 1% is the level which HR was observed during 1% of the monitoring time ATre corresponds to the difference between Tre recorded at the beginning of the shift and the level observed at the end (Tre f)

results may be ascribed to the 'observer effect', reported in other studies (Meursing, 1974; Malchaire et al, 1984) as this subject on this occasion, while performing masonry work, was clearly eager to demonstrate his capabilities by working at a higher pace and taking shorter rest periods than usual. This worker did work normally while sweeping dust during shift a. During this shift, a high Tre was also observed but not associated with other signs of excessive strain. In worker 4, the HR and rectal temperature limits were consistently exceeded during both shifts. Nevertheless, no creeping up of either HR or Tre was observed during the two exposures. The poor heat tolerance of this worker may tentatively be linked with his physical fitness which was low both in absolute and age-related terms. Therefore, it may be stated that the observed work-rest regimens have allowed the heat-induced strain to remain within acceptable limits for six workers out of seven. Similar observations were reported in a group of Indian workers (Parikh et al, 1978). Other observations made by Wyndham (1973) and Gertner et al (1984) suggested that workers were able to reduce their work output in warm ambient conditions in order to regulate the circulatory strain, even though this adaptation sometimes meant a decrease in their salary. In this study, the effectiveness of the workers' self-regulation system looked fairly good when considering the ages and physical capacities of the workers. Older men are known to exhibit greater strain in very hot climates than younger ones (Lind et al, 1970). Moreover, the workers studied were not especially fit subjects; their maximal oxygen uptake per kg body weight ranged from 85% (worker 4) to 111% (worker 2) of the mean expected values for non-athletic males (Binkhorst e t al, 1966). Two main reasons can be suggested to explain these favourable results. First, the heat exposure duration was limited, on average, to 25% of the working time. In the

present case, this low work productivity had no incidence on the workers' wages. The other factor was most likely the pair-work organisation observed during the cleaning task. This organisation provided, along with an evident safety factor, the opportunity to share the heat exposure period depending on each individual's physical fitness and heat tolerance. This hypothesis cannot be verified from the limited number of observations collected. Nevertheless, the low cardiac strain observed in worker 3 who was the oldest worker and had the lowest VO2 max seems to support this hypothesis, together with the comparison of shift a and shift b results of worker 1. This worker exhibited a higher cardiac strain when carrying out his task alone (shift b) than when working in a team with another worker (shift a). Our results showed that the work-rest regimens actually adopted by the workers were poorly related to physiological parameters. The worker physical fitness was the only parameter significantly to affect the duration of each working period. During a self-paced task in controlled laboratory conditions, Vogt et al (1983) had observed that the end of each working period was related to a critical HR level, this being near constant for a given pattern of workrest cycles. In this study, such a phenomenon was never observed among the seven workers (see Table 2). Thus, the results did not throw much light on how a worker determined his own work-rest regimen. In this respect, an important role may have been played by perceptual factors, such as thermal discomfort sensations in relation to the skin wettedness underneath the clothing, or the subjective balance between the time allowed to complete the task and the amount of work still to be done. However, the study was not designed to explore these factors as it was orientated towards the assessment of physiological strain levels at the request of the plant occupational physician.

Applied Ergonomics

June 1985

89

What can be the meaning of such results with regard to a health protection policy for jobs in hot conditions? Previous observations made in hot industries showed that the work-rest regimens spontaneously adopted by workers were sometimes neither efficient in terms of work productivity nor in terms of physiological strain reduction (Brouha, 1960; ANACT, 1979). In some conditions, the workers were prone to work in a very hot environment, i e, inside a furnace, for a period as long as possible in order to reduce the total number of heat exposures (ANACT, 1979). Although such a behaviour has not been observed by other authors (Parikh et al, 1978), the possibility must be taken into account when considering the adoption of free work-rest schedules. In conclusion, the results of the present study cannot be extended beyond their conditions of observation. Spontaneous work-rest regimens should not be recommended when the task has an urgent character or does involve productivity incentives. In any case, free workrest cycles have to be associated with close supervisory surveillance, as emphasised by Millican et al (1981), and with other measures such as good acclimatisation procedures. In other situations, predetermined work-rest schedules may constitute a better solution to prevent heat stress disorders.

Gertner, A., Israeli, R., and Cassuto, Y. 1984 Ergonomics, 27, 135-146. Effects of work and motivation on the heart rates of chronic heat-exposed workers during their regular work shifts. ISO 1983 D6termination analytique et interpr6tation de la contrainte thermique fond~es sur le calcul de la sudation requise. Avent-projet de norme (DP 7933). International Standards Organisation.

Kamon, E. 1979 Ergonomics, 22, 427-440. Scheduling cycles of work for hot ambient conditions.

Lind, A.R., Humphreys, P.W., Collins, K.J., Foster, K., and Sweetland, K.F. 1970 JApplPhysiol, 28, 50-56. Influence of age and daily duration of exposure on responses of men to work in heat.

Malchaire, J., Wallemacq, M., Rogowsky, M., and Vanderputten, M. 1984 Ann Occup Hyg, 28, 189-193. Validity of oxygen consumption measurement at the workplace: what are we measuring?

Meursing, N.A. 1974 Remaking of the inner wall of a converter with oxygen. Research report No 583/74, European Community for Steel and Coal, Luxembourg.

References

Millican, R., Baker, R.C., and Cook, G.T. 1981 Am Ind Hyg Ass J, 42, 411-416. Controlling heat stress - administrative versus physical control.

ACGIIt 1977 Threshold limit values for physical agents. American Conference of Governmental Industrial Hygienists, Cincinnati, Ohio.

Parikh, DJ., Ghodasara, N.B., and Ramanathan, N.L. 1978 Eur J Appl Physiol, 40, 63-72. A special thermal stress

ANACT 1979 Formation pour l'am61ioration des conditions de travail et d6marche participative. Une exp6rience significative. Agence Nationale pour l'Am61ioration des Conditions de Travail, Paris.

Mairiaux, Ph. 1983 Ergonomics, 26, 1173-1185. Heart rate and spontaneous work-rest cycles during exposure to heat.

Binkhorst, R.A., Pool, J., Van Leeuwen, P., and Bouhuys, A. 1966 Intern Z angewPhysiol, 22, 10-18. Maximum

1969 Health factors involved in working under conditions of heat stress. Technical Report Series, No 412, Geneva.

oxygen uptake in healthy non-athletic males.

Brouha, L. 1960 'Physiology in industry'. Pergamon Press, Oxford.

90

AppliedErgonomics June 1985

problem in ceramic industry.

Vogt, J.J., Libert, J.P., Candas, V., Daull, F., and

World Health Organisation

Wyndham, C.H. 1973 Arch Sc Physiol, 27, A491-A497. The effects of heat stress upon human productivity.