Noise Health Home 

[Download PDF]
Year : 1998  |  Volume : 1  |  Issue : 1  |  Page : 47--55

Fatigue after work in noise - an epidemiological survey study and three quasi-experimental field studies

Anders Kjellberg1, Per Muhr2, Bjorn Skoldstrom1,  
1 National Institute for Working Life, Solna, Sweden
2 F10B, AM, �ngelholm, Sweden

Correspondence Address:
Anders Kjellberg
National Institute for Working Life, S-171 84 Solna


The contribution of noise exposure to fatigue at work was studied in a survey study and three field studies. The survey study was based on a questionnaire covering symptoms and work place exposure answered by 50 000 state employees. Noise exposure was also estimated from their type of job and self-rated noise exposure. Fatigue and headache were found to be more common among the noise exposed groups even after control for the effects of other critical variables. Study 2 compared reaction times before and after a week's work in high noise exposure and one in low exposure exposure in a group of aeroplane mechanics. Reaction times were prolonged after work in the noise week, whereas an opposite trend was seen in the control week. Study 3 showed a gradual increase of reaction times during a week of noise exposure in a group of aeroplane technicians. Study 4 compared reaction times and subjective fatigue among naval crews on a day with low and a day with high noise exposure. In one of the studied boat types the development of fatigue during the work day was accentuated on the day with high exposure.

How to cite this article:
Kjellberg A, Muhr P, Skoldstrom B. Fatigue after work in noise - an epidemiological survey study and three quasi-experimental field studies.Noise Health 1998;1:47-55

How to cite this URL:
Kjellberg A, Muhr P, Skoldstrom B. Fatigue after work in noise - an epidemiological survey study and three quasi-experimental field studies. Noise Health [serial online] 1998 [cited 2023 Dec 4 ];1:47-55
Available from:

Full Text


Fatigue is one of the most common complaints at work. Largely, this fatigue is a natural consequence of the work tasks, but the physical environment may also contribute to this fatigue. It is, for example, a common opinion that noise makes work more tiring, but very few studies have been made of this possible noise effect. Laboratory studies have repeatedly demonstrated that performance may be impaired after work in noise even if no effect was observed during the exposure (Cohen, 1980; Glass & Singer, 1972). However, very few studies of such effects have been performed in occupational contexts.

There are at least three ways in which noise may have fatiguing effects. First, noise may contribute to a general over-stimulation. Secondly, monotonous noise has been found to have sleep-provoking effects (Landstrom & Lundstrom, 1985; Landstrom & Lofstedt, 1987). Thirdly, noise may make a task more difficult and tiring to perform, for example by masking important acoustic information.

There is some evidence that complaints about fatigue are more common among workers exposed to high levels of noise (McDonald & Ronayne, 1989) and recently Melamed and Bruhis (1996) reported a study on textile workers exposed to noise during one week with and one without hearing protectors. They found subjective effects such as fatigue and irritability but also a higher cortisol level after the week without hearing protectors.

In the first of the present series of studies the association between noise exposure and fatigue was analysed in a large database containing self�reported health and occupational exposure data. The remaining three studies were field studies of fatigue before and after work using subjective and performance indicators of fatigue. The second and third studies dealt with aeroplane mechanics exposed to high noise level, and the fourth study was conducted on ships from the coastal fleet, where exposure levels were lower.


A large data base was used to study the connection between noise exposure and three complaints: fatigue, headache and negative affect. Headache was included since noise has often been mentioned as a factor triggering headache (Philips & Hunter, 1981), and is also often considered to have this effect in work places (McDonald & Ronayne, 1989; Ohrstrom, Bjorkman & Rylander, 1979). However, we know of no study that has demonstrated a higher headache frequency among workers exposed to noise. Negative affect is characterised by symptom such as irritated, depressed and ill at ease. In one study the exposed persons themselves reported that noise had such an effect (McDonald & Ronayne, 1989), but there are reasons to doubt that any causal relation exists between these types of symptoms and noise exposure (Stansfeld, Clark, Jenkins & Tarnopolsky, 1985).

 Materials and Methods

Study group

A large data base was used containing questionnaire data on exposure and health from several hundred thousand persons all over Sweden collected by the occupational health care services for state employees. Data from 61 832 subjects collected 1985-89 from three occupational groups (transport and communication; manufacturing industry, machine maintenance; service work) were included in the study. The three occupations were chosen to guarantee that a large part of the study group was exposed to noise. Another reason was to get a group that was fairly homogeneous regarding what is to be meant by being exposed to noise.

Noise exposure

Each person answered a question whether he was regularly exposed to noise at work. A validation study had shown that about 50 percent of those who were regularly exposed to noise above 70 dB(A) reported themselves to be exposed and that 95 percent of those exposed to lower levels reported themselves to be unexposed.

For the 88 occupational groups including more than 50 per-sons (altogether 57 481 per-sons) two occupational hygienists estimated equivalent sound level at work into three classes: 80 dB(A), (4021 persons).The five groups containing 345 persons where the hygienists did not agree were excluded from the analyses.

In the analyses of the association between noise exposure and complaints a combination of the two exposure indicators was used. From the group estimated to have an exposure below 60 dB(A) the 7642 persons reporting themselves to be exposed were excluded. Missing data in any of the confounding variables included in the analyses meant that the group was further reduced to 41442 persons (80 dB(A): 3252 persons).

Indicators of fatigue and other complaints

The answer to the yes-no question 'Do you often feel tired without any apparent cause?' was used as the indicator of fatigue. Headache was indicated by the question 'Do you often have headache?'.

A third indicator, negative affect, was defined as an affirmative answer to one of the following five questions: Do you often feel 1. unconcentrated; 2. restless or tense; 3. irritated or impatient; 4. anxious or nervous; 5. depressed, ill at ease or sad.

Statistical analysis Logistic regression analyses were performed of the association between noise and the three complaints.

Several variables were included in the analyses as possible confounders: Gender, age, work schedule (daytime work, two shift, three shift work, night work), different indicators of physical work load (heavy lifting, sedentary work, repetitive motions, difficult work positions, all yes/no-questions), bad climate conditions (yes/no), work load (mean of two questions about how often there is too much to do and how often work demands are too high) and work involvment (mean of two questions about how interesting and varying they condiderd their tasks to be). The strength of the associations was expressed as odds ratios (OR) with the group exposed to 2. AEROPLANE MECHANICS: HIGH AND LOW EXPOSURE WORKING WEEK

 Materials and Methods

Fatigue and reaction time performance were assessed in a group of twenty-four male aeroplane mechanics from the Swedish Air Force during two weeks: one week of high noise exposure while working on the runway and one week of low exposure at their base. A more comprehensive report of this study is given by Kjellberg, Skoldstrom, Andersson and Lindberg (1996).

Noise exposure during work was continuously monitored by a dosimeter (Larson & Davis Model 700 Dosimeter) worn with the microphone close to the ear by the mechanic. Wind speed, air temperature and number of takeoffs were also registered each day during the period.

An adjective check list containing 16 subscales was used to obtain scale values in three subjective dimensions: Energy, Stress and Wakefulness (Kjellberg & Bohlin, 1974; Kjellberg & Iwanowski, 1989). Subjects also answered questions about work load and other working condition during the day.

The simple reaction time test is part of the SPES computer test battery (Gamberale, Iregren & Kjellberg, 1990). The signal is a square that appears on the screen at irregular intervals (2.5�5 sec). The subject's task is to press the space bar on the keyboard as fast as possible once the square appears.

Each subject participated during one week on the runway (high noise exposure) and one at the base (low exposure).

The main task during the high exposure week on the runway was to make the aeroplanes ready for takeoff and serve them after landing. Refuelling was the responsibility of another group, and the mechanics were therefore not exposed to jet fuel fumes to any notable extent. During the base week they mainly worked with check-ups and repairing.

The fatigue ratings were made before and after work each day Monday till Thursday during the two study weeks. The reaction time test was only performed before and after work on Thursday during the two weeks.



The A-weighted equivalent sound level during the runway weeks varied between 95 and 100 dB(A). During periods of work with the aeroplanes the equivalent sound levels for ten minutes' periods were 100 to 110 dB(A), with maximum levels up to 138 dB(A). The equivalent sound level at the base was approximately 20 dB(A) lower. Since all mechanics wore ear protectors, the weekly dose fell just below 85 dB(A), if the protectors worked as specified by the manufacturer.


[Figure 1],[Figure 2] show mean AM and PM scores for the two noise conditions. Both Wakefulness and Energy were lowered from the AM to the PM rating in the runway week. In the base week the energy rating was stable over the day whereas Wakefulness scores increased in the afternoon. This effect was reflected as a significant interaction between time of day and noise condition for both scales (F 1/21 =4.50 and 5.77, respectively, p 3/63 =3.50, p 1/19 =17.74, p 19 =2.69, p=0.015). During the runway week there was a non-significant trend of reaction times being longer in the afternoon (F 1/19 =1.88, p=0.19) compared to the morning.


The results confirmed the hypothesis that the mechanics felt more sleepy and less energetic during the week of work on the runway than after repairing work at their companies. This effect was most evident in the afternoon after work and increased during the week. Thus, there seems to have been a cumulative effect on fatigue during the runway week.

Furthermore, reaction times were prolonged at the end of the week on the runway and this effect was apparent already in the morning before noise exposure.

Fatigue ratings were neither correlated with workload nor with boredom, air temperature or wind speed. The nearly identical stress ratings during the two weeks indicate that the mental load was about the same.

Alternative explanations of the effects were not supported and noise exposure therefore seems to be the most plausible explanation.


In the previous study, self-reports indicated an increasing fatigue effect during the high exposure runway working week. As reaction times were measured only during the last day, no corresponding analysis was possible of these data. However, the longer reaction times already obtained in the morning measurements indicate a cumulative effect on reaction times also. The aim of the present study was to test whether such a gradual increase of reaction times occurred among aeroplane mechanics working on the runway.

 Materials and Methods

Sixteen male aeroplane mechanics were followed during a week at the runway using the same methods as in the first study, with the difference that reaction times and ratings were collected before and after work each day from Monday to Friday.


Equivalent sound levels varied between 96 and 102 dB (with maximum levels between 120 - 140 dBA). All mechanics wore hearing protectors. Reaction times were gradually prolonged during the week ([Figure 5], F 4,60 =4,38, p 1,15 =2,09, p=.17).

Ratings of wakefulness and energy differed systematically between days but presented no regular trend over the week.


The gradual prolongation of reaction times supported the interpretation of the difference in reaction times in the morning as a result of an accumulated fatigue effect during the week. However, the absence of a control condition means that these resultson theirn own cannot be taken as evidence for the fatiguing effects of noise. It should also be noted that reaction times were always longer after work than in the morning; this effect agrees with the results from the previous study but is in a direction opposite to what is expected under normal conditions as a result of the diurnal rhythm.

The absence of any cumulative effect on rated fatigue may reflect that ratings more than reaction times are affected by changes between days of specific work conditions. If so reaction times reflect changes of a person's state that are less easily affected by changing environmental characteristics. Such effects of changing working condition between days were more likely in this study than in the first one, since all data were collected during one week, whereas data collection in the first study was spread over six weeks.


In an attempt to generalise the findings of the aeroplane mechanics studies, a study was conducted of crews of three ships from the coastal fleet, where the sound levels are considerably lower and the working conditions are also different in many other ways.

 Materials and Methods

Two conventional patrol boats (displacement 150 tons) and an experimental ship model (displacement 140 tons), which utilised the hovercraft technique, were included in the study. Twenty-nine persons took part in the study: eight persons from each patrol boat and thirteen from the experimental ship. The same methods and general design were used as in the studies of aeroplane mechanics. However, in addition to the morning and afternoon measurements, measurements were also conducted around noon.

For each subject these measurements were made during a day at sea (high noise exposure) and on a day working on the boat at the quay.

Wind speed and wave height were registered during the days at sea, and subjects were asked about the use of medication against sea sickness.


During the high exposure days at sea the equivalent noise level at both types of ships was about 80 dB(A). The control day noise level was between 50 and 65 dB(A). The low frequency part of noise was stronger on the experimental ship than on at the patrol boats.

During the testing days at sea the wind force was 5-10 m/s and the wave height 0,5-1,0 m on the patrol boats and 3-5 m/s and 2,22 =3,49, p 2,54 =10,42, p Acknowledgements

This paper was presented at the 3rd European Conference on Protection Against Noise (PAN), 12-15 March 1998, Stockholm Sweden organised and supported by the European Commission BIOMED 2 concerted action PAN (Contract BMH 4-CT96-0110).[13]


1Cohen, S. (1980). Aftereffects of stress on human performance and social behavior: A review of research and theory. Psychological Bulletin, 88, 82-108.
2Gamberale, F., Iregren, A., & Kjellberg, A. (1990). SPES: Assessing the effects of the work environment on man with computerized performance testing. In Computer aided ergonomics. W. Karwowsky, A. M. Genaidy, & S. S. Asfour (Eds.), (pp. 381-396). London: Taylor & Francis.
3Glass, D. C., & Singer, J. E. (1972). Urban stress: experiments on noise and social stressors. New York: Academic Press.
4Kjellberg, A., & Bohlin, G. (1974). Self-reported arousal: Further development of a multi-factorial inventory. Scandinavian Journal of Psychology, 15, 285-292.
5Kjellberg, A., & Iwanowski, S. (1989). Stress/Energi-formularet: Utveckling av en metod for skattning av sinnesstamning i arbetet [The Stress/Energy Questionnaire. Development of a method for mood ratings at work] (Report 1989:26): National Institute of Occupational Health, Solna, Sweden.
6Kjellberg, A., Skoldstrom, B., Andersson, P., & Lindberg, L. (1996). Fatigue effects of noise among airplane mechanics. Work and Stress, 10, 62-71. Landstrom, U., Bystrom, M., & Nordstrom, B. (1985). Changes in wakefulness during exposure to noise at 42 Hz, 1000 Hz and individual EEG frequencies. Journal of Low Frequency Noise and Vibration, 4, 27-33.
7Landstrom, U., & Lundstrom, R. (1985). Changes in wakefulness during exposure to whole body vibration. Electroencephalography and Clinical Neurophysiology, 61, 411-415.
8Landstrom, U., & Lofstedt, P. (1987). Noise, vibration and changes in wakefulness during helicopter flight. Aviation, Space, and Environmental Medicine, 58, 109-118.
9McDonald, M., & Ronayne, T. (1989). Jobs and their environment: the psychological impact of work in noise. Irish Journal of Psychology, 10, 39-55.
10Melamed, S., & Bruhis, S. (1996). The effects of chronic noise exposure on urinary cortisol, fatigue, and irritability. Journal of Occupational and Environmental Medicine, 38, 252-256.
11Philips, C., & Hunter, M. (1981). Pain behaviour in headache sufferers. Behaviour Analysis and Modification, 4, 257-266.
12Stansfeld, S. A., Clark, C. R., Jenkins, L. M., & Tarnopolsky, A. (1985). Sensitivity to noise in a community sample: I Measurement of psychiatric disorder and personality. Psychological Medicine, 15, 243-254.
13Ohrstrom, E., Bjorkman, M., & Rylander, R. (1979). Subjective evaluation of work environment with special reference to noise. Journal of Sound and Vibration, 65, 241-249.