Travel & Places

The influence of production systems on self-reported arousal, sleepiness, physical exertion and fatigue?consequences of increasing mechanization

Description
The present study examined the capability of a real-time assessment routine to sort out the individual impact of three production systems on psychological activation as measured by the Stress-/Energy Inventory, Karolinska Sleepiness Scale, Borg's
Published
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Transcript
  Copyright © 2003 John Wiley & Sons, Ltd. The influence of productionsystems on self-reportedarousal, sleepiness,physical exertion andfatigue—consequences of increasing mechanization Roger Persson,* ,†,1 Anne Helene Garde, 2 Åse Marie Hansen, 2 Palle Ørbæk 1 andKerstina Ohlsson 11 Division of Occupational and Environmental Medicine, Department of LaboratoryMedicine, Lund University Hospital, Sweden 2 National Institute of Occupational Health, Copenhagen, Denmark Summary The present study examined the capability of a real-time assessment routine to sort out the indi-vidual impact of three production systems on psychological activation as measured by the Stress- /Energy Inventory, Karolinska Sleepiness Scale, Borg’s CR-10 Perceived Exertion scale, and theSwedish Occupational Fatigue Inventory. Sixteen women between the ages of 26 and 57 years(mean 43 years) rotated in a counterbalanced order between three production systems: A, B and C. The systems produced the same goods but clearly differed in degree of automation and ergonomic demands. The results show that work at the most automated production system, C,on average generated lower energy index scores and higher sleepiness scores compared to theoldest system, A. Clear weekly and diurnal patterns were found for most rating measures. To con-clude, the increasing automation of the production systems is reflected in psychological activationby dampening feelings of positively valued high activity states and increased sleepiness. The expres-sions of a weekly and diurnal psychological activation, indicates that the subjects are able tounwind both during the weekend and after work, and suggest that the present method is suitablefor studying the immediate psychological adaptation to the social and physical work environment.Copyright © 2003 John Wiley & Sons, Ltd. *Correspondence to: Roger Persson, Division of Occu-pational and Environmental Medicine, Lund UniversityHospital, SE-22185 Lund, Sweden. Tel: + 46 46 177287. Fax: + 46 46 177285.E-mail: Roger.Persson@ymed.lu.se Stress and Health Stress and Health  19 : 163–171 (2003)Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/smi.967 Received 2 December 2002Accepted 3 March 2003 Key Words time-series; psychological assessment; repetitive work; diary; logbook Introduction The use of self-report diaries, or logbooks, in occupational health research has typicallyfocused on assessing exposure to chemical sub-stances or work tasks (e.g. Jensen, Murer, Olsen,  & Christensen, 1995; Viikari-Juntura et al.,1996). Methods for measuring the psychologicaleffects of work are insufficiently developed, particularly regarding time-series measurement.Time-series utilization of self-report diaries, orlogbooks, may provide an inexpensive prospec-tive and real-time account of the individuals’immediate psychological adaptation towards thesocial and physical environment. In the presentcase, it was expected that such measurementcould supplement and shed new light on existingtechnical data concerning muscle activity, andbody postures and movements, obtained fromdifferent production systems in a parquet floor-ing factory (I. Balogh, K. Ohlsson, S. Skerfving,G.-Å. Hansson, & T. Engström, unpublisheddata). The production systems in question madethe same goods but represent technological devel-opments that have resulted in distinct work con-ditions due to increasing mechanization. Anotherreason for the present investigation was to test aroutine involving frequent, and in part parallel,assessment of psychological and endocrine mea-sures. This approach partly gained inspirationfrom sleep research, which has developed time-series methods feasible for field settings (e.g.Kecklund, Åkerstedt, & Lowden, 1997). Theprimary aim here was to examine the capabilityof a real-time assessment routine to sort out theindividual impact of three production systems onpsychological activation. Other aims were tostudy the psychological activation over the weekand during days, with the purpose of examiningto what degree the subjects unwind during theweekend and after work, and to evaluate thesubject’s compliance to the present measurementschedule. Subjects and methods Subjects Sixteen women working on an afternoon shift(14.00 to 22.00 hours, including breaks) partici-pated. The mean age was 43 years (range 26 to57 years). These volunteers, identified in collabo-ration with the employer, showed low mental distress in the General Health Questionnaire-30(median = 1.61, Q1–Q3 = 1.52–1.74; Goldberg& Williams, 1988) and the State-Trait AnxietyInventory-Trait scale ( M  = 1.54, CI 95% = 1.39–1.68; Spielberger, Gorsuch, Lushene, Vagg,& Jacobs, 1983). According to the demand–control–support model (Johnson, 1986; Karasek& Theorell, 1990), nine subjects experienced apassive work situation (i.e. low demands and lowdecision latitude); six subjects experienced jobstrain (i.e. high demands and low decision lati-tude) and one subject experienced an active worksituation (i.e. high demands and high decision latitude). Subjects received 2h salary for com-pleting the baseline questionnaires outside workhours. Production systems and work description The production systems process small woodenslats for parquet flooring (length 202 to 456mm;width 49 to 66mm; thickness 10mm, weight 70to 300g). All systems produce the same goods;require highly trained workers, demand intenseconcentration and rapid qualified decisions, andthe subjects rotate between work assignmentsevery half hour. Production system A. System A is the oldestand most manual. In a first step slats are auto-matically fed to the system. Thereafter, the systemengages nine subjects, organized along two par-allel conveyor belts. The slats pass by at a rate of 2 slats per second. The first task, carried out whileseated, is to sort out defective slats by visualinspection. The next two tasks, carried out whileseated, are to visually determine the quality of theslats and to identify edge defects. Depending onquality and type of edge defect, the slats are placedon sideboards, one on each side of the worker. Thefourth task is carried out standing and focuses onfurther inspection of the edges by manually rotat-ing a bundle of 30 to 40 slats at the same time.Non-approved slats are sorted out whereasaccepted slats are loaded onto benches. A finaltask is to walk between the lines, and collect theslats from the sideboards and load them. Production system B. System B engages fivesubjects and represents a technologically inter-mediate step. At the first workstation slats aremanually fed to the system by carrying slats frombenches and loading them into a cassette. The fol-lowing four workstations are analogous to systemA and involve inspection of slats by focusing onquality and edges while seated or standing. Production system C. System C is the newestproduction system and the most automated. R. Persson et al  . Copyright © 2003 John Wiley & Sons, Ltd. Stress and Health  19 : 163–171 (2003)164  Industrial robots feed the system with slats.Clearly defective slats are then sorted out by computerized video scanning. The manual labouris performed by nine subjects, seated along threeparallel conveyor belts. At the first station at each belt, inspection focuses on the edges of theslats by mechanically rotating them by use of ajoystick-operated device. At the following twostations inspection focuses on the quality of theslats that pass by at a rate of 3 slats per second.Quality grading is marked by slightly changingthe position of the slats on the conveyor belt.Finally, the slats are loaded onto benches by anindustrial robot. Measures Borg CR-10 scale. This scale measures thedegree of perceived physical exertion according to a scaling method developed by Borg (1982,1998). The scale has 12 steps, some of which areverbally anchored. Higher values indicate greaterperceived physical exertion. In the present casethe subjects were told to rate their experiences of exertion as a reflection of their feelings of car-diovascular load. Karolinska Sleepiness Scale (KSS; Åkerstedt &Gilberg, 1990). The KSS assesses the alertnessor sleepiness experienced by subjects, on a 9-point scale with verbal anchors on every secondstep: (1) very alert, (3) alert, (5) neither alert norsleepy, (7) sleepy, but no difficulty remainingawake, and (9) extremely sleepy, fighting sleep.Sleepiness is here viewed as a symptom defined asthe momentarily perceived relative strength of two drives, a wake drive and a sleep drive. The Stress/Energy Inventory (Kjellberg &Iwanowski, 1989). The Stress/Energy Inven-tory measures feelings of arousal in two dimen-sions: stress and energy, each comprising sixitems. The stress dimension covers negatively-valued high-activity states to positively-valuedlow-activity states. The energy dimension covers positively-valued high-activity states tonegatively-valued low-activity states. Items are re-sponded to on 6-point scales (score ranges from0 to 5) with verbal anchors reflecting the inten-sity at which people experience a specific adjec-tive (e.g. relaxed, tense, stressed, etc.). Higherscores indicate more experienced stress or energy. The Swedish Occupational Fatigue Inventory-20 (SOFI-20; Åhsberg, Gamberale, & Kjellberg,1997). The SOFI-20 measures work-relatedfatigue in five dimensions: lack of energy, phy-sical exertion, physical discomfort, lack of motivation, and sleepiness. Each dimensionencompasses four items that are responded to on7-point scales (score ranges from 0 to 6) with verbally anchored endpoints ‘not at all’ and ‘to avery high degree’. Higher scores indicate moreintense symptoms of fatigue. Since there is nogenerally accepted gold standard for measuringfatigue it should be noted that the current defini-tion sees fatigue as a multidimensional phenom-enon that incorporates both mental and physicalfeatures. Procedure One week before the study, the subjects com-pleted a 24-h training cycle involving logbooks,urine and saliva sampling. The administration of the logbooks followed the principles outlined byBorg (1998). Thus time was spent on discussingand explaining the items with the respondenteven before the training session. On the first dayin the assessment week, the subjects completed,under supervision, a basic questionnaire. The log-books were completed during a single week(Table I). Assessment points were chosen to coverkey aspects of the 24-h cycle as well as workhours (1) awakening (2) just before work at 14.00hours, (3) during work at 16.00 hours, (4) duringwork at 20.00 hours and (5) just before going tobed or in bed. Fatigue ratings were, however, onlymade at bedtime. From Tuesday to Friday and on Sunday, parallel assessment was made withendocrine variables (A.-H. Garde, Å.-M. Hansen,R. Persson, P. Ørbæk, & K. Ohlsson, unpublisheddata). With few exceptions, subjects rotated be-tween production systems in a counterbalancedorder. On workdays, some of our staff memberswere in an adjacent room to receive urine andsaliva samples, and to answer questions concern-ing the logbooks. Ethics All participants gave written consent to partici-pate. The Ethics Committee of Lund Universityapproved the study, LU 690-99. Influence of production systems on self-reports Copyright © 2003 John Wiley & Sons, Ltd. Stress and Health  19 : 163–171 (2003)165  Statistical analysis Values of P below 0.05 were considered statisti-cally significant. Categorical predictors were day,time and production system. Dependent variableswere the inventory scores. Only days including allassessment points were analysed. Due to unsys-tematic loss of data the general linear mixedmodels module in SPSS 11.0.1 was used to esti-mate overall treatment effects and treatmenteffects at separate assessment points (SPSS Inc.,1998). A repeated measures model was speci-fied and an iterative model-fitting strategy wasadopted. Models were first fitted using a max-imum likelihood (ML) method but the finalmodel was solved using the restricted maximumlikelihood (REML) method. For all analyses, afirst order autoregressive and a compound sym-metry covariance structure was tested. The com-pound symmetry structure provided in all cases abetter fit, so it was generally applied. Residualplots were used to identify potential outliers anddeviations from normality assumptions. No out-liers or deviations from normality were detected.Statistical inference for the mean structure of datawas made with fixed effects models and approx-imate Type III F  -tests and t  -tests, including anoverall intercept. Production system A, Sunday,and bedtime were generally selected as first-handreference levels. Results are typically expressed asdeviations from reference levels in raw scores aswell as deviations from reference levels in stan-dard deviation units: Z -scores. The main questionwas to evaluate the overall difference between theproduction systems. The comparisons studyingthe effects of day and time on the inventoryscores, utilized full-factorial models in whichinteraction effects and main effects only wereretained if statistically significant. Results Comparison Between Production Systems Self-reported arousal. Production systeminfluenced the energy index scores (  p  < 0.001) butday and time did not. Work at system C gener-ated on average lower scores compared to systemA (  p  = 0.002; Table II). No influence from pro-duction system, day or time was observed on thestress index score. Sleepiness and physical exertion. Both pro-duction system (  p  = 0.009) and time (  p  = 0.001)influenced the sleepiness scores. Work at systemC generated on average higher sleepiness scoresthan work at system A (  p  = 0.005). Sleepinessratings were lower at 14.00 hours compared to20.00 hours. Only time influenced the physicalexertion ratings (  p  = 0.002), which on averagewere 0.51 points lower at 14.00 hours comparedto 20.00 hours (  p  < 0.001; CI 95% = - 0.78; - 0.23; Z  = - 0.68, CI 95% = - 0.68; - 0.20). Fatigue. Production system influenced theSOFI-20 lack of energy ratings (  p  = 0.010). Workat system B generated scores which were onaverage 0.50 points lower (  p  = 0.034) than thosegenerated at system A (CI 95% = - 0.96; - 0.04; Z = - 0.39, CI 95% = - 0.76; - 0.03). Scores for lackof energy were 0.80 points higher for work atsystem C (  p  = 0.011) compared to system B (CI95% = 0.21; 1.39; Z = 0.63, CI 95% = 0.16; 1.09). Comparison between days off and days at work  Self-reported arousal. Both day (  p  < 0.001)and time (  p  < 0.001) influenced the stress index R. Persson et al  . Copyright © 2003 John Wiley & Sons, Ltd. Stress and Health  19 : 163–171 (2003)166Table I.The sampling schedule for salivary cortisol (S), catecholamines and cortisol in urine (U) and self-reportlogbooks (L).Training day: 1TuesdayWednesdayThursdayFridaySaturdaySundayweek prior to testingWake upS/U/LS/U/LS/U/LS/U/LS/U/LS/U/LS/U/L + 30minSSSSS—S + 45minSSSSS—S1400S/U/LS/U/LS/U/LS/U/LS/U/LLS/U/L1600S/U/LS/U/LS/U/LS/U/LS/U/LLS/U/L2000S/U/LS/U/LS/U/LS/U/LS/U/LLS/U/LBedtimeS/U/LS/U/LS/U/LS/U/LS/U/LLS/U/L —, no sampling. Salivary cortisol is always sampled first when several samples are collected.  scores. Lower stress ratings were made on Sundaythan on the workdays (Figure 1, Table III). Stressratings at bedtime were lower than ratings duringwork but higher than ratings in the morning.Only time influenced the energy index ratings (  p < 0.001). At bedtime the energy ratings werelower than during the day but not different frommorning levels. Sleepiness and physical exertion. Time influ-enced the sleepiness ratings (  p  < 0.001), whichwere higher at bedtime than during the day (TableIV). Both day (  p  < 0.001) and time (  p  < 0.001)influenced the physical exertion score. The physi-cal exertion ratings were lower on Sunday com-pared to the workdays. At bedtime the physicalexertion ratings were higher than in the morning,and just prior to work at 14.00 hours. Fatigue. Day had an effect on the SOFI-20 lackof energy (  p  < 0.001) and physical discomfortratings (  p  < 0.001), which were both lower onSunday compared to the workdays (Table V). Compliance The stress/energy inventory had a compliance rateof 95 per cent. The sleepiness and physical exer-tion items had a compliance rate of 96 per cent.The five SOFI-20 scales, the sleep quality indexand the awakening index had a compliance rateof 97 per cent. Discussion The present study shows that increasing mecha-nization of production systems is reflected in psy-chological effect measures. Primarily, it has beenobserved that work at the most automated pro-duction system C, dampens the feelings of positively-valued high-activity states and increasesleepiness when compared to the most manualproduction system A. Observably, the impact of production system C is relatively larger regardingthe dampening of feelings of positively-valuedhigh-activity states compared to the impact onsleepiness. These observations are in agreementwith I. Balogh et al.’s (unpublished data) studywhich showed less muscle activity and move-ments at production system C, as well as with ourassessment of endocrine activity made in parallelwith the present ratings (A.H. Garde et al.,unpublished data), which showed lower concen-trations of urinary adrenaline and noradrenalineat system C. The results also seem to corroboratethe concept of somnificity, which emphasizes thatthe subjects’ posture and activity level (both phy-sical and mental) is of major importance in deter-mining our sleep and wake drives ( Johns, 2002).Interestingly, A.H. Garde et al. (unpublisheddata) report that both the positively- and negatively-valued high-activity states (i.e. energyand stress scores) were associated with higheradrenaline levels whereas low cortisol levels wereassociated with higher sleepiness scores. How-ever, when the self-report data and endocrine Influence of production systems on self-reports Copyright © 2003 John Wiley & Sons, Ltd. Stress and Health  19 : 163–171 (2003)167Table II.Comparisons between production systems, the energy index and sleepiness scores: results from modelsincluding production system (3 levels), day (4 levels) and time (3 levels). REML estimates and t  -tests, using acompound symmetry covariance structure. Energy index scoreSleepiness scoreRaw- D (95% CI) Z - D (95% CI)  p -valueRaw- D (95% CI) Z - D (95% CI)  p -valueIntercept3.64 (3.36; 3.92)——3.19 (2.59; 3.78)——P-system A00—00—P-system B0.02 ( - 0.12; 0.15)0.02 ( - 0.21; 0.27)0.8240.03 ( - 0.25; 0.32)0.02 ( - 0.20; 0.27)0.811P-system C  - 0.25 ( - 0.38; - 0.11)  - 0.44 ( - 0.68; - 0.20)  < 0.0010.41 (0.13; 0.70)0.34 (0.10; 0.57)0.005Tuesday  - 0.14 ( - 0.28; 0.00)  - 0.25 ( - 0.49; 0.00)0.0500.21 ( - 0.08; 0.50)0.17 ( - 0.07; 0.40)0.166Wednesday  - 0.11 ( - 0.24; 0.03)  - 0.19 ( - 0.43; 0.06)0.1350.36 (0.07; 0.64)0.29 (0.01; 0.52)0.017Thursday  - 0.04 ( - 0.18; 0.10)  - 0.01 ( - 0.31; 0.18)0.6090.15 ( - 0.14; 0.44)0.12 ( - 0.12; 0.35)0.317Friday00—00—14000.00 ( - 0.11; 0.12)0.00 ( - 0.20; 0.22)0.944  - 0.48 ( - 0.72; - 0.23)  - 0.39 ( - 0.59; - 0.19)  < 0.00116000.02 ( - 0.01; 0.14)0.03 ( - 0.18; 0.24)0.752  - 0.24 ( - 0.49; - 0.01)  - 0.19 ( - 0.39; 0.01)0.059200000—00—Raw- D , difference from reference level in raw scores. Z - D , difference from reference level in Z scores.
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x