Human Fatigue in Prolonged Mentally Demanding Work-Tasks: an Observational Study in the Field

Mississippi State University Scholars Junction

Theses and Dissertations Theses and Dissertations

1-1-2013

Human Fatigue in Prolonged Mentally Demanding Work-Tasks: An Observational Study in the Field

Shaheen Ahmed

Follow this and additional works at: https://scholarsjunction.msstate.edu/td

Recommended Citation Ahmed, Shaheen, "Human Fatigue in Prolonged Mentally Demanding Work-Tasks: An Observational Study in the Field" (2013). Theses and Dissertations. 2513. https://scholarsjunction.msstate.edu/td/2513

This Dissertation - Open Access is brought to you for free and open access by the Theses and Dissertations at Scholars Junction. It has been accepted for inclusion in Theses and Dissertations by an authorized administrator of Scholars Junction. For more information, please contact [email protected]. Automated Template C: Created by James Nail 2011V2.02

Human fatigue in prolonged mentally demanding work-tasks:

an observational study in the field

Shaheen Ahmed

A Dissertation Submitted to the Faculty of Mississippi State University in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in Industrial and Systems Engineering in the Department of Industrial and Systems Engineering

Mississippi State, Mississippi

August 2013

Shaheen Ahmed

2013

Human fatigue in prolonged mentally demanding work-tasks:

an observational study in the field

Shaheen Ahmed

Approved:

______Kari Babski-Reeves Janice DuBien Associate Professor and Graduate Associate Professor Coordinator Mathematics and Statistics Industrial and Systems Engineering (Minor Professor) (Director of Dissertation)

______Burak Eksioglu Adam C. Knight Associate Professor Assistant Professor Industrial and Systems Engineering Kinesiology (Committee Member) (Committee Member)

______Lesley Strawderman Heather E. Webb Assistant Professor Assistant Professor Industrial and Systems Engineering Kinesiology (Committee Member) (Committee Member)

______Royce O. Bowden Interim Dean of the Bagley College of Engineering

Name: Shaheen Ahmed

Date of Degree: August 17, 2013

Institution: Mississippi State University

Major Field: Industrial and Systems Engineering

Major Professor: Kari Babski-Reeves, Ph.D.

Title of Study: Human fatigue in prolonged mentally demanding work-tasks: an observational study in the field

Pages in Study: 171

Candidate for Degree of Doctor of Philosophy

Worker fatigue has been the focus of research for many years. However, there is limited research available on the evaluation and measurement of fatigue for prolonged mentally demanding activities.

The objectives of the study are (1 )to evaluate fatigue for prolonged, mentally demanding work-tasks by considering task-dependent, task-independent and personal factors, (2) to identify effective subjective and objective fatigue measures, (3) to establish a relationship between time and factors that affect fatigue (4) to develop models to predict fatigue.

A total of 16 participants, eight participants with western cultural backgrounds and eight participants with eastern cultural backgrounds, currently employed in mentally demanding work-tasks (e.g., programmers, computer simulation experts, etc.) completed the study protocols. Each participant was evaluated during normal working hours in their workplace for a 4-hour test session, with a 15-minute break provided after two hours.

Fatigue was evaluated using subjective questionnaires (Borg Perceived Level of Fatigue

Scale and the Swedish Occupational Fatigue Index (SOFI)); and objective measures

(change in resting heart rate and salivary cortisol excretion). Workload was also assessed using the NASA-TLX. Fatigue and workload scales were collected every 30 minutes, cortisol at the start and finish of each 2-hour work block, and heart rate throughout the test session.

Fatigue significantly increased over time (p-value <0.0001). All measures, except cortisol hormone, returned to near baseline level following the 15-minute break (p-value

<0.0001). Ethnicity was found to have limited effects on fatigue development. Poor to moderate (Rho = 0.35 to 0.75) significant correlations were observed between the subjective and objective measures. Time and fatigue load (a factor that impacts fatigue development) significantly interact to explain fatigue represented by a hyperbolic relationship. Predictive models explained a maximum of 87% of the variation in the fatigue measures.

As expected, fatigue develops over time, especially when considering other factors that can impact fatigue (e.g. hours slept, hours of work), providing further evidence of the complex nature of fatigue. As the 15-minute break was found to reduce all measures of fatigue, the development of appropriate rest breaks may mitigate some of the negative consequences of fatigue.

DEDICATION

This dissertation is dedicated to my mother, my sister, and my wife.

ACKNOWLEDGEMENTS

I would like to thank and acknowledge many people for their splendid support, outstanding advice, and great inspiration until the successful end.

First, I would like to thank my advisor, Dr. Kari Babski-Reeves. She has been advising me since 2007 right after my admission at MSU. From that day, she never told me what to do and how to do it. Most of the time, we were barely able to agree on anything that I remember. Dr. Babski’s Socrates method of mentoring always has given me opportunities to think differently and to discover my own path. During graduate school, she has been an extraordinary mentor, an exceptional parent and a good friend.

Dr. DuBien is one of the most influential teachers in my life who has helped me to understand the beauty of statistics. She has given selflessly of her time anytime I have visited her. Her experience, expertise and enthusiasm on design of experiments guided me to accomplish an excellent dissertation. She has given a tremendous input into my dissertation, which I will never be able to repay by just saying, “Thank you”! I was so motivated by her that I have selected design of experiments in human factors and ergonomics as one of my future research interests.

Dr. Eksioglu was the first person I met at MSU. He was extremely friendly and approachable during the conversion. I did not even realize that he was an associate professor at the end of the conversation when I asked him "How should I address professors here at MSU?" He replied, "You can call me Dr. Eksioglu!" From that day, he

iii

has been extremely supportive and an excellent mentor. He has been a great motivation for me to pursue my PhD and academia as my career. I would like to thank him for his tremendous support during my graduate school.

Dr. Knight is another extremely friendly committee member. When I was in the deep sea of my dissertation proposal, I could not figure out that it was too ambitious to accomplish. Dr. Knight is the first committee member who was very conscious about the scope of the dissertation. He advised me to narrow down focus to a particular area. That great piece of advice has helped to accomplish my dissertation within a reasonable period of time and effort.

Dr. Strawderman is one of the most organized and highly professional people I know. We have worked on a couple of projects together which I have enjoyed a lot. I have learned many great things from her. For example, she taught me to write my first peer reviewed article. She has been so supportive since then, during my MS thesis, finally during my dissertation. She has also promptly edited my CV during my job applications.

I would like to thank her for the great advice during my graduate study at MSU.

Dr. Heather Webb has contributed a lot to my dissertation, especially the design and analysis of my dissertation experiment regarding cortisol, which has been proven to be a measure of human fatigue. She has spent hours in the lab to help me analyze the saliva cortisol concentration. She has also spent many hours to help me interpret the results from the saliva cortisol analysis. In addition to the cortisol hormone research, I have gotten a lot of great advice from her.

Dr. Farhana Tasmin, my wife, a great medical doctor has provided me wonderful support during my graduate study. As a doctor, she has demonstrated many basics about

the functions of our body and mind. For example, she has explained the function of autonomic nervous system and endocrine system and their effect on the human body and mind, particularly when they become imbalanced. I have gotten many great suggestions, especially regarding the design and analysis of experiments, data analysis and the interpretation of results. Her splendid support at home has helped me to handle stress during my dissertation. Thank you so much for everything!

I acknowledge my study participants who volunteered their time without any sort of compensation. I could not complete my dissertation without their generosity. They were exceptionally motivated and actually helped me find more participants. Within a short period of time, I did find all my study participants. I would especially thank Dr.

Domenico "Mimmo" Parisi, the Director of the National Strategic Planning & Analysis

Research Center. He provided permission to collect data from his center. Dr. David

Thomson, Professor of the Department of Aerospace Engineering at MSU, who has been significantly cooperative and promptly helped me to find enough participants for the study.

For their extreme support, I would also like to mention, including but not limited to, Mohammad Faridul Alam (Farid), Tanmay Bhowmik, Soumya Bhoumik, Mohammad

Refatul Islam, Dr. Satish Ganji, Dr. Khaled Hassan, Mohammad Marufuzzaman, Dr.

Kylie Nash, Dr. Jibonananda Sanyal, and Andrew Staps. Last, but most importantly, to my parents, my sister, my wife, my family and friends who have been consistently providing me excellent support to earn a PhD, thank you!

TABLE OF CONTENTS

DEDICATION ...... ii

ACKNOWLEDGEMENTS ...... iii

LIST OF TABLES ...... ix

LIST OF FIGURES ...... xii

ACRONYMS ...... xiv

CHAPTER

I. INTRODUCTION ...... 1

1.1 Objective ...... 5

II. LITERATURE REIVEW ...... 6

2.1 Definition of Fatigue ...... 6 2.2 Measurement of fatigue ...... 10 2.2.1 Subjective measurement of fatigue ...... 11 2.2.1.1 Rating of perceived exertion scale ...... 15 2.2.1.2 Swedish Occupational Fatigue Inventory (SOFI) ...... 17 2.2.2 Objective measurement of fatigue ...... 18 2.2.2.1 Heart rate & heart rate variability ...... 22 2.2.2.2 Saliva cortisol concentration ...... 23 2.2.3 Quantifying physical and cognitive fatigue simultaneously ...... 29 2.2.4 Fatigue and performance...... 30 2.3 Factors that affect fatigue ...... 31 2.3.1 Sleep and shift work...... 32 2.3.2 Workload...... 32 2.3.3 Time ...... 35 2.3.4 Rest breaks ...... 36 2.3.5 Gender and age ...... 38 2.3.6 Ethnicity ...... 39 2.4 Fatigue predictive models ...... 40 2.4.1 Two-step quantitative model to predict fatigue ...... 45

III. METHOD ...... 49 vi

3.1 Experimental Design ...... 49 3.1.1 Population model for the experiment ...... 49 3.2 Independent variables ...... 51 3.2.1 Duration of the study ...... 52 3.3 Dependent variables ...... 52 3.3.1 Subjective measures of fatigue ...... 52 3.3.1.1 Modified Borg CR-10 scale to measure fatigue ...... 52 3.3.1.2 Swedish Occupational Fatigue Inventory (SOFI) ...... 53 3.3.2 Subjective measure of workload ...... 53 3.3.3 Objective measures of fatigue ...... 54 3.3.3.1 Change in Heart rate (∆HR) ...... 54 3.3.3.2 Saliva cortisol concentration ...... 55 3.3.3.2.1 Weighted saliva cortisol concentration ...... 56 3.3.3.2.2 Data cleaning method for saliva cortisol concentration to measure fatigue ...... 56 3.4 Participants ...... 57 3.5 Power analysis ...... 61 3.6 Procedure ...... 63 3.7 Data analysis ...... 64 3.7.1 Analysis of variance ...... 64 3.7.2 Correlations ...... 65 3.7.3 Regression Analysis ...... 65

IV. RESULTS ...... 68

4.1 Descriptive statistics ...... 68 4.2 Effect of ethnicity and time ...... 68 4.3 Effect of task-independent and personal factors ...... 77 4.4 Justification for the hypothesized hyperbolic relationship between time and a factor that causes fatigue ...... 78 4.5 Correlation analysis ...... 83 4.5.1 30-minute-block correlation...... 83 4.5.2 Two-hour-block correlations ...... 84 4.6 Fatigue predictive models ...... 86

V. DISCUSSION ...... 92

5.1 Ethnicity ...... 92 5.2 Effect of time ...... 94 5.3 Prediction of fatigue ...... 95

VI. LIMITATIONS AND FUTURE STUDIES ...... 97

6.1 Statistical power and sample size ...... 97 6.2 Study protocol ...... 97 6.3 Subjective instruments ...... 98 vii

6.4 Effect of ethnicity ...... 99 6.5 Effect of time ...... 99

VII. CONCLUSION ...... 101

REFERENCES ...... 104

APPENDIX

A. DEMOGRAPHIC QUESTIONNAIRE IN FATIGUE STUDY ...... 131

B. MODIFIED BORG SCALE ...... 136

C. MODIFIED SWEDISH OCCUPATIONAL FATIGUE INVENTORY ...... 138

D. WORKLOAD ASSESSMENT INSTRUMENT, NASA-TLX ...... 140

E. RANDOMIZATION OF THE DIMENSIONS OF SUBJECTIVE SCALES ...... 142

F. DESCRIPTIVE STATISTICS ...... 147

G. TUKEY ADJUSTED POST-HOC DETAILS...... 151

H. CORRELATION PROCEDURES ...... 160

viii

LIST OF TABLES

2.1 Definitions of Fatigue ...... 8

2.2 Fatigue scales (DeLuca, 2005) ...... 13

2.3 Borg ratings of perceived exertion (RPE) scale (G. Borg, 1970b)...... 15

2.4 Borg category ratio (CR10) scale (G. A. Borg, 1982)...... 16

2.5 Swedish Occupational Fatigue Inventory Scale (SOFI) and the type of fatigue ...... 18

2.6 Objective measures of fatigue ...... 20

2.7 Definitions for mental workload ...... 34

3.1 Expected mean squares for ANOVA ...... 51

3.2 Group wise demographic statistics ...... 58

3.3 Frequency table for logical groupings ...... 59

3.4 Overall demographic statistics ...... 60

3.5 Demographic statistics by ethnicity and working shift ...... 60

3.6 Demographic statistics by ethnicity within a working shift...... 62

3.7 Post-hoc power analysis ...... 63

4.1 ANOVA for Borg, SOFI, NASA and ∆HR ...... 69

4.2 Tukey adjusted tests of effect slices for Borg and SOFI ...... 70

4.3 Tukey adjusted post-hoc for ∆HR and NASA ...... 72

4.4 Tukey adjusted post-hoc for saliva cortisol ...... 72

4.5 Effect of time and task-independent and personal variables on Borg and SOFI ...... 78

4.6 Summary of stepwise regression as a proof of the hyperbolic relationship ...... 79

4.7 Parameter estimates from the stepwise regression as a proof of the hyperbolic relationship...... 80

4.8 Overall correlation matrix ...... 83

4.9 Correlation matrix for Indian participants ...... 84

4.10 Correlation matrix for Western participants ...... 84

4.11 Correlation matrix for fatigue and saliva cortisol measures ...... 85

4.12 Summary of stepwise regression for the first-two-hour session ...... 87

4.13 Parameter estimates of stepwise regression for the first-two-hour session ...... 88

4.14 Summary of Stepwise Regression for the rest-break session ...... 89

4.15 Parameter estimates of stepwise regression the rest-break session ...... 89

4.16 Summary of Stepwise Regression for the second-two-hour session ...... 90

4.17 Parameter estimates of stepwise regression for the second-two-hour session ...... 91

E.1 Randomization of the dimensions of subjective scales ...... 143

F.1 Descriptive statistics for Borg, SOFI, NASA and ∆HR by time ...... 148

F.2 Descriptive statistics for Borg, SOFI, NASA and ∆HR by Ethnicity ...... 149

F.3 Descriptive statistics for Borg, SOFI, NASA and ∆HR by working shift ...... 149

F.4 Descriptive statistics for saliva cortisol concentration (CRT) and weighted CRT ...... 150

F.5 Descriptive statistics for CRT and weighted CRT by time...... 150

F.6 Descriptive statistics for CRT and weighted CRT by ethnicity ...... 150

F.7 Descriptive statistics for CRT and weighted CRT by working shift ...... 150

G.1 Tukey adjusted post-hoc for change in heart rate and NASA ...... 152

G.2 Tukey adjusted post-hoc for change in Borg and SOFI with respect to sleep ...... 153

G.3 Tukey adjusted post-hoc for change in Borg and SOFI with respect to weekly working hours in primary occupation ...... 154

G.4 Tukey adjusted post-hoc for change in Borg and SOFI with respect to total weekly working hours in all occupations ...... 155

G.5 Tukey adjusted post-hoc for change in Borg and SOFI with respect to perceived fatigue at the end of the day ...... 156

G.6 Tukey adjusted post-hoc for change in Borg and SOFI with respect to Exercise ...... 157

G.7 Tukey adjusted post-hoc for change in Borg and SOFI with respect to daily rest ...... 158

G.8 Tukey adjusted post-hoc for change in Borg and SOFI with respect to Shift ...... 159

H.1 The CORR procedure at time = 0 minute ...... 161

H.2 The CORR procedure at time = 30 minutes ...... 162

H.3 The CORR procedure at time = 60 minutes ...... 163

H.4 The CORR procedure at time = 90 minutes ...... 164

H.5 The CORR procedure at time = 120 minutes ...... 165

H.6 The CORR procedure at time = 135 minutes ...... 166

H.7 The CORR procedure at time = 165 minutes ...... 167

H.8 The CORR procedure at time = 195 minutes ...... 168

H.9 The CORR procedure at time = 225 minutes ...... 169

H.10 The CORR procedure at time = 255 minutes ...... 170

H.11 The CORR procedure over time ...... 171

LIST OF FIGURES

2.1 Function of Hypothalamic-Pituitary-Adrenal Axis ...... 24

2.2 Diurnal salivary cortisol level ...... 25

2.3 An example of the change in cortisol levels in mentally demanding tasks...... 26

2.4 Trend in salivary cortisol concentrations between weekdays (left) and weekends (right)...... 27

2.5 Visualization of Yerkes-Dodson Law...... 33

2.6 Loss of strength with fatigue ...... 36

2.7 Causes of fatigue by Grandjean's ...... 38

2.8 Conceptual model of fatigue and performance in healthcare workers ...... 42

2.9 Bayesian network of fatigue variables ...... 43

2.10 Potential endogenous and exogenous variables that may be linked with fatigue...... 44

2.11 Factors that affect fatigue (step one of the two-steps model) ...... 46

2.12 Conceptual hyperbolic relationship of time and fatigue load ...... 47

4.1 Time and ethnicity significantly interact to affect fatigue in both scales...... 71

4.2 Change in resting heart and NASA scores over time ...... 73

4.3 Change in resting heart rate and NASA scores over ethnicity...... 73

4.4 Change of raw and normalized saliva cortisol concentration over time ...... 75

4.5 Change of raw and normalized saliva cortisol concentration over Ethnicity ...... 75

4.6 Change in MAUCI and MAUCG by ethnicity ...... 76 xii

4.7 Change in MAUCI and MAUCG by ethnicity ...... 76

4.8 Contour plot for Borg and SOFI ratings as a function of : workload and time [(a) and (b)], ∆HR and time [(c) and (d)], and CRT and time [(e) and (f) ...... 82

xiii

ACRONYMS

Acronym Definition

T or Time = Experimental clock time in hours, unless mentioned otherwise.

1= 2= 3= 4= 5= 6= 7= 8= 9= 10=

0:00 0:30 1:00 1:30 2:00 2:15 2:45 3:15 3:45 4:15

Eth = Ethnicity

A = Age

W = Weekly working hours in primary occupation

TW = Total weekly working hours in all occupations

EDF = Perceived fatigue at the end of a regular working day

MMF = Perceived fatigue on Monday morning

Ex = Weekly exercise frequency

Sl = Hours of daily sleep

DR = Daily rest after work in hours

Sh = Working shift, 1=morning shift, 2= afternoon shift

B or Borg = Perceived fatigue measured in one-dimensional fatigue scale

named after Borg, Borg scale

S or SOFI = Perceived fatigue measured in multi-dimensional fatigue scale

know as Swedish Occupational Fatigue Inventory (SOFI)

NASA = Perceived workload measured in NASA-TLX

xiv

∆HR = Change in resting heart rate (bit per minute)

A∆HR = Average change in resting heart rate during each two-hour session

CRT = Saliva cortisol concentration ( )

NCRT = Normalized saliva cortisol concentration ( )

CRTM = Saliva cortisol concentration ( ) during early morning

CRTR = Saliva cortisol concentration ( ) during relaxation

AUCI = area under the curve with respect to increase for salivary cortisol

AUCG = area under the curve with respect to ground for salivary cortisol

MAUCI = area under the curve with respect to increase for normalized

salivary cortisol.

MAUCG = area under the curve with respect to increase for normalized

salivary cortisol

Fatigue Load (FL) = Any quantitative factor (e.g. change in resting heart rate, workload,

daily sleep, etc.) that affect fatigue is defined as “fatigue load” in

this dissertation

CHAPTER I

INTRODUCTION

In the United States, 37.9% of workers reported fatigue, and 65.7% of those reported health-related lost of productive time compared with 26.4% of those without fatigue. Workers with fatigue cost employers $136.4 billion annually, which is $101.0 billion higher compared with workers without fatigue (Ricci, Chee, Lorandeau, & Berger,

2007). The prevalence of fatigue is growing every year. For example, in the Netherlands, in the workplace, 50% of women and 33% of men reported fatigue in 2008 as compared to 38% of women and 24% of men 15 years before (M. A. Boksem & Tops, 2008). The increase in fatigue could be associated with the increase in mentally demanding jobs, and/or sedentary jobs, which require less physical activity (M. A. Boksem & Tops, 2008).

In the United States, service industries held approximately 70% of jobs in 2008, which are mostly sedentary with less physical but not necessarily less mental demand

("U.S. Bureau of Labor Statistics," 2013). In the 48 years between 1960 and 2008, approximately 30% of jobs were converted from moderate physical activity to sedentary jobs ("U.S. Bureau of Labor Statistics," 2013). In the same period, task-dependent energy consumption decreased 140 calories for men and 124 calories for women per day ("U.S.

Bureau of Labor Statistics," 2013), which has been considered as the primary cause of mean weight gain of the U.S. population (Church et al., 2011). In the Netherlands, a 4.7 hour per week increase in sedentary work-tasks has been observed between 1975 and 1

2005; the non-occupational sedentary period was found unchanged though (van der Ploeg et al., 2013).

In the literature, studies have been observed to evaluate fatigue purely based on performance (Linsey M. Barker & Nussbaum, 2011). However, No significant differences in performance were observed for vigilance (screening) tasks between sitting and standing (Drury et al., 2008; Ohlinger, 2009). Most mentally demanding work-tasks are designed to be performed in sitting positions, which makes the jobs even more sedentary. Prolonged sitting multiplies the odds for mortality irrespective of physical activities (van der Ploeg Hp, 2012). Lack of physical activities either in the occupation or in non-occupation boost the risk for bad health consequences (Mork, Vasseljen, &

Nilsen, 2010; A. H. Taylor & Dorn, 2006). For example, a sedentary job with low physical demands significantly contributes to central and total obesity (Choi et al., 2010), which has been considered as the etiology of many life-threatening diseases (Bray, 2004;

Gilson, Burton, Van Uffelen, & Brown, 2011). Moreover, prolonged sitting has also been observed as the primary cause of fatigue in lumbar and truck muscles, which may add up to the overall body fatigue (Areeudomwong et al., 2012; van Dieën, Westebring-van der

Putten, Kingma, & de Looze, 2009). Sitting over an extended period of time introduces prolonged static postures resulting in discomfort (El Falou et al., 2003; Pietri et al., 1992) and muscle fatigue even for low exertion activities such as 5% Maximum Voluntary

Contraction (MVC) (Sjogaard, Kiens, Jorgensen, & Saltin, 1986) and 2%MVC (van

Dieën, et al., 2009). Fatigue has been reported throughout the literature for these low physically demanding work-tasks (Blangsted, Sjøgaard, Madeleine, Olsen, & Søgaard,

2005; Kroemer, 1997; Sjøgaard, Lundberg, & Kadefors, 2000).

Many of these sedentary tasks, including computer work-tasks such as programming and simulation, are substantially mentally demanding, which introduces mental fatigue (M. A. Boksem & Tops, 2008). Many physically demanding activities in industries have been automated, which have converted most physical workload into mental workload. Elimination of physical workload does not necessarily reduce the task demands on workers (Moore, 2000) because the total amount of time in a working day has not been reduced. Individuals reported fatigue at the end of the day (M. A. Boksem &

Tops, 2008; DeLuca, 2005) even though the task was designed to be performed without experiencing fatigue. For example, programming and simulation impose substantially higher mental demands with very little physical activities (Sjøgaard, et al., 2000), which establishes a perfect imbalance of the use of body resources. This type of imbalance job or work-task may contribute to the development of cognitive fatigue, physical fatigue and total fatigue (Sjøgaard, et al., 2000).

Peer-reviewed articles on fatigue have increased 90% over the past decade

(Friedberg, 2013). Most of the fatigue research has been focused primarily on the population with medical conditions. Only a few studies have been conducted on fatigue in the workplace. Moreover, studies on occupational fatigue have been focused on the effect of sleep disorder and shift work, which has been considered as the primary cause of fatigue. Poor design of jobs/work-tasks could be a reason for sleep disorders (Torbjörn

Åkerstedt, Fredlund, Gillberg, & Jansson, 2002). Therefore, sleep disturbance could be considered as an indirect objective measure of fatigue in the workplace. In addition to a measure like sleep disturbance, other variables related to a work-task could be more useful and direct in predicting fatigue. Studies on fatigue in the workplace with low

physical, but significantly mentally demanding work-tasks, have rarely been found throughout the literature. Hence, only a few variables related to a work-task have been found throughout the literature to measure fatigue objectively.

Some variables that could affect fatigue in the workplace include workload, duration of work-task and rest breaks. To prevent task-dependent musculoskeletal disorders at seated workstations, such as low back problems, individuals should move or change their positions every 30 minutes or less (Babski-Reeves, Stanfield, & Hughes,

2005). However, typically, breaks in the workplace are 15 minutes after two-hours of work, 30 minutes to an hour for lunch after four hours of work, and a second 15-minute break after six hours of work. While this break allocation is clearly insufficient (Mital,

Bishu, & Manjunath, 1991), it is reasonable to assume that this traditional break schedule may not hold for office situations where workers generally work continuously, even with a lunch break (Balci & Aghazadeh, 2003; Galinsky, Swanson, Sauter, Hurrell, &

Schleifer, 2000; Henning, Jacques, Kissel, Sullivan, & Alteras-Webb, 1997; Lindegard et al., 2012; Toomingas, Forsman, Mathiassen, Heiden, & Nilsson, 2012). Therefore, the effect of the duration of work tasks on fatigue is an important consideration for work- tasks with low physical, but high mental demand (Beynon, Burke, Doran, & Nevill, 2000;

Dababneh, Swanson, & Shell, 2001; Toomingas, et al., 2012).

The context-specific definitions and measurement techniques have substantially limited the opportunity to translate the findings from one context into another. Moreover, a limited number of studies in relation to sedentary computer work-tasks, especially programming and simulation types of jobs, makes it difficult to transform the findings from other areas to these low physically, but high mentally demanding work-tasks. For

example, the hyperbolic relationship of muscle fatigue with respect to the muscle power and time could hold for overall physical, cognitive or total fatigue with respect to workload and time.

1.1 Objective

The objective of this research is to understand human fatigue in prolonged mentally demanding work-tasks, for example, programming and simulation. The specific aims are given below:

Specific Aim 1: Evaluation of overall fatigue by studying task-independent and personal factors (e.g. exercise and rest) and task-dependent (perceive workload and physiological changes) information

Specific Aim 2: Subjective evaluations of fatigue utilizing:

1. One-dimensional fatigue measuring scale

2. Multi-dimensional fatigue measuring scale

3. Workload over time

Specific Aim 3: Objective evaluations of fatigue utilizing:

1. Saliva cortisol concentration over time

2. Change in resting heart rate over time

Specific Aim 4: To test the effectiveness of the hyperbolic interaction only relationship between time and a factor that affects fatigue, similar to the established hyperbolic relationship of muscle fatigue with respect to muscle power and time.

Specific Aim 5: Develop models to assess fatigue quantitatively utilizing task- dependent, task-independent and personal factors.

CHAPTER II

LITERATURE REIVEW

Relevant scholarly articles were searched in many databases, including,

ScienceDirect, Elsevier, PubMed, MedlinePlus, JSTOR, InderScience, ACM, IEEE and

Taylor & Francis. The review of literature is provided in between page 7 and page 49.

2.1 Definition of Fatigue

A number of definitions for fatigue exist in the literature (Table 2.1), though, a consensus definition of fatigue has not been identified (Barofsky & Legro, 1991;

Christley, Duffy, & Martin, 2010; Eidelman, 1980; Hashimoto, 1992; Komaroff &

Joncas, 1991; Lynch, Main, & Seth, 1991; Matthews, Lane, & Manu, 1988; Mills &

Young, 2008; Noy et al., 2011; Reeves et al., 2005; Sullivan, Pedersen, Jacks, &

Evengard, 2005; Tiesinga, Dassen, & Halfens, 1996). One reason might be the multiple sources or causes of fatigue such as physiological, psychological, social, environmental, etc. (Manu, Lane, & Matthews, 1992). Many definitions of fatigue have been developed specific to the area of interest, for example, the definition of fatigue for multiple sclerosis

(Filippi & Rocca, 2007; Krupp, 2003; Mills & Young, 2008) or for cancer (B. F. Piper &

Cella, 2010). Although a consensus definition of occupational fatigue has not been reached, many subscribe to the theory that fatigue is a multi-dimensional construct; meaning that it is a complex, multi-causal, nonspecific and subjective phenomenon

(Linsey Marinn Barker, 2009; Tiesinga, et al., 1996). Fatigue has been recognized as acute and chronic fatigue, muscle fatigue, mental fatigue and psychomotor fatigue (D.

Dawson et al., 2011). Occupation specific fatigue such as driving fatigue, visual fatigue, etc. has also been observed in the literature. However, the definition of occupational fatigue is not well documented in the literature. Table 2.1 shows a short list of definitions for fatigue. Many definitions of fatigue fail to follow the essential features of a definition of fatigue, including the following (Job & Dalziel, 2001).

1. The definition should identify fatigue as a hypothetical construct, not a

performance outcome per se;

2. The definition should not identify performance decrement as fatigue;

3. The definition should identify the cause of the state of the person;

4. The definition should reflect, as far as possible in logical limits, the

meaning ascribed to the term by the general population.

5. On the grounds of following conventional use (Point 4), fatigue should

include state arising in the central nervous system (CNS) and the muscles,

but not in the sensing neurons (such as in the retina and associated

neurons).

6. In consequence of these features of an appropriate definition, the

definition should allow a distinction between fatigue and related

phenomena.

In contrast to the medical field where a wealth of literature exists, few definitions of fatigue have emerged for the healthy working population. Research on fatigue in industrial settings has been progressing since the early 1900s (Brouha & Ball, 1948;

Clements, 1926; Crowden, 1930; E. L. Fisk, 1928; E. Grandjean, 1979; "Industrial

Fatigue," 1925; "Industrial Fatigue and Ventilation," 1921; "Industrial Fatigue: Work of the Research Board," 1926; Manson, 1916; "Reducing Industrial Fatigue," 1919).

Table 2.1 Definitions of Fatigue

Developer/Reference Definition The North American Nursing The self recognized state in which an individual experiences an Diagnosis Association overwhelming sustained sense of exhaustion and decreased capacity for (Carpenito-Moyet, 2006) physical and mental work. (Ream & Richardson, 1997) A subjective, unpleasant symptom which incorporates total body feeling ranging from tiredness to exhaustion, creating an unrelenting overall condition which interferes with individuals' ability to function to their normal capacity. (Aaronson et al., 1999) The awareness of a decreased capacity for physical and/or mental activity due to an imbalance in the availability, utilization, and/or restoration of resources needed to perform activity. (Shen, Barbera, & Shapiro, Fatigue is an overwhelming sense of tiredness, lack of energy and a feeling 2006) of exhaustion, associated with impaired physical and/or cognitive functioning; which needs to be distinguished from symptoms of depression, which include a lack of self-esteem, sadness and despair or hopelessness. (Brown, 1994) Psychological fatigue is defined as subjectively experienced disinclination to continue the task. (Hancock & Verwey, 1997) Fatigue is an individuals’ multi-dimensional physiological-cognitive state associated with stimulus repetition which results in prolonged residence beyond a zone of performance comfort (Job & Dalziel, 2001) Fatigue refers to the state of an organism’s muscles, viscera, or central nervous system, in which prior physical activity and/or mental process, in the absence of sufficient rests, results in insufficient cellular capacity or system wide energy to maintain the original level of activity and / or processing by using normal resources. (Williamson et al., 2011) Fatigue is biological drive for recuperative rest (Gander et al., 2011) Fatigue is the inability to function at the desired level due to incomplete recovery from the demands of prior work and other waking activities. Acute fatigue can occur when there is inadequate time to rest and recover from a work period. Cumulative (chronic) fatigue occurs when there is insufficient recovery from acute fatigue over time. Recovery from fatigue, i.e., restoration of function (particularly of cognitive function), requires sufficient good quality sleep.

Due to the rapid growth of service sectors in the post-industrialized world

(Soubbotina & Sheram, 2000), recent fatigue studies have also focused on service sectors, for example, health care (Linsey Marinn Barker, 2009). Although research on

industrial fatigue focused initially on physical activities, such as heavy load handling

(Chapman, 1990), psychological fatigue was equally perceived for these physically demanding tasks (Collier, 1943). It is still not known today, whether fatigue is primarily because of physical or psychological aspects, or both (DeLuca, 2005). A concept of total fatigue considering both physical and mental fatigue is proposed to define fatigue in the industrial context or for the healthy working population (Babski-Reeves & Crumpton,

1999). Later, a formalized definition of fatigue for nurses based on the concept of total fatigue is proposed (Linsey Marinn Barker, 2009).

“Total fatigue is a state comprised of at least two dimensions: mental fatigue and

physical fatigue. Mental and physical fatigue dimensions are present in nurses

exposed to excessive mental and physical demands through their work tasks and

schedules. These fatigue dimensions contribute to a state of total fatigue, which

over time can result in these workers not being able to function at their normal

capacity and can lead to an increased risk for injury or medical error”

The concept of total fatigue or the definition of total fatigue is appropriate if a work-task consists of both mental and physical demand simultaneously (e.g. nursing).

Many work-tasks in today’s world have significantly higher mental demand as compared to physical demand (e.g. programming and simulation). Neither a suitable definition for mentally demanding work-tasks nor a generalized definition of fatigue has been established yet.

Another missing dimension in most definitions of fatigue is associated with the low level of arousal and the on-set of boredom occurring in non-challenging jobs with poor quality of supervision and very little control over the job (Finkelman, 1994). For

example, data entry jobs which are neither physically nor mentally demanding may increase the risk of a very low level of both physical and mental arousal (Finkelman,

1994). Both overuse and underuse of resources could create fatigue (Finkelman, 1994).

Because the objective of this dissertation is to study fatigue, a generalized definition for fatigue is proposed here:

Fatigue is the imbalance created by either underuse or overuse of body resources, which causes a drive for balance by maintaining a healthy lifestyle, including food habit, daily sleep, proper rest, workload, exercise, etc.

2.2 Measurement of fatigue

There are various objective and subjective methods by which to measure fatigue.

Objective measures tend to focus on changes in physiological human responses (e.g., heart rate variability or decreased levels of muscle contraction force) (Stokes, Cooper, &

Edwards, 1988). Subjective measures of fatigue utilize questionnaires to obtain workers’ perceptions of their fatigue level. However, subjective fatigue measures can be poorly correlated with physiological responses (Berrios, 1990), raising the question of validity for objective and/or subjective fatigue measures. The seemingly lack of direct relationship between subjective perceptions and physiological changes or responses may have led to the vast collection of fatigue measures (Linsey Marinn Barker, 2009;

Barofsky & Legro, 1991). A brief review on both subjective and objective measures of fatigue is provided in section 2.2.1

2.2.1 Subjective measurement of fatigue

Many questionnaires (instruments) have been developed to measure fatigue in populations with and without medical conditions Table 2.2. Some of these instruments are one-dimensional while some are multidimensional. Commonly assessed fatigue characteristics have been negative feelings or perceptions. Some questionnaires have also included positive fatigue perceptions (e.g. fatigue after exercise) such as Swedish

Occupational Fatigue Inventory (SOFI) (Elizabeth Åhsberg & Gamberale, 1998). Most fatigue questionnaires solicit the degrees to which an individual perceives the fatigue dimension(s). One primary difficulty in synthesizing fatigue research is the diversity of the questionnaires in terms of length:

1. questionnaire length can range from a single question (e.g., the CR-10

scale (G. Borg, 1990)) to 83 questions (Stein, Martin, Hann, & Jacobsen,

1998);

2. answer format (e.g., Yes/No responses (Chalder et al., 1993) ), Likert-type

scales of various lengths (e.g, (Elizabeth Åhsberg & Gamberale, 1998)),

visual analog scales (e.g.,(Lee, Hicks, & Nino-Murcia, 1991));

3. time period for assessment (now, (Kogi, Saito, & Mitsuhashi, 1970)), past

few hours (Schwid, Covington, Segal, & Goodman, 2002), past week

(Hann, Denniston, & Baker, 2000), past four weeks (J. D. Fisk et al.,

1994), duration of illness onset (Grohar-Murray, Becker, Reilly, & Ricci,

1998); and

4. scoring methodology (e.g. factor analysis for scales with multiple

questions (Elizabeth Åhsberg & Gamberale, 1998; Kogi, et al., 1970; B. F.

Piper et al., 1998; Schwartz, Jandorf, & Krupp, 1993), mean, median, minimum and maximum for single dimension scales (G. A. Borg, 1982) ).

Despite these differences, subjective measures of fatigue remain one of the most frequently used techniques for a variety of applications.

Table 2.2 Fatigue scales (DeLuca, 2005)

Name of Scale Author, Year Initial Specified Fatigue Item Item Time Frame Population Subscales Length Scoring Fatigue Japanese (Kogi, Healthy, Drowsiness and dullness, 30 Yes/No Now Symptom et al., 1970) Cancer projection of physical Checklist English (Haylock disintegration, difficulty in & Hart, 1979) concentration Borg Rating of (G. Borg, 1970b) Healthy Rating of perceived exertion Single 6-20* Now Perceived item Exertion (RPE) scale Borg Category (G. Borg, 1982; Healthy Rating of perceived exertion Single 0-10** Now Ratio Scale G. A. Borg, item (CR-10) 1982) Piper Fatigue (B. F. Piper, Cancer Behavioral/severity, 22 items 0-10 One item asks for Scale (PFS) 1990; B. F. Piper, effective meaning, sensory, (+5 short duration et al., 1998; B. F. cognitive/mood answer) Piper et al., 1989) Fatigue (Krupp, LaRocca, MS, Lupus, None 9 1-7 Not stated past 2 Severity Scale Muir-Nash, & healthy weeks Steinberg, 1989) appropriate*** Single Item (Krupp, et al., MS, Lupus, None 1 Visual Not stated Visual 1989) healthy analogue Analogue Scale scale (VAS) Visual (Lee, et al., 1991) Sleep Energy, fatigue 18 Visual Not stated Analogue Scale disordered and analogue for Fatigue healthy scale (VAS-F) Fatigue (Schwartz, et al., Lyme, CFS, Fatigue severity, situation- 29 1-7 Past 2 weeks Assessment 1993) Lupus, MS, specific, consequences of Instrument Dysthymia, fatigue, responds to (FAI) healthy rest/sleep Fatigue Scale (Chalder, et al., Primary care Physical, mental 14 YES/No Not stated (FS) 1993) patients Checklist (J. H. Vercoulen CFS Subjective experience of 24 7-point Not stated Individual et al., 1994) fatigue, concentration, scale strength (CIS) motivation, physical activity Fatigue Impact (J. D. Fisk, et al., MS Physical, cognitive, 21 (short 0-4 Past 4 weeks Scale (FIS) 1994) psychosocial form: 5 items) * Designed to increase linearly with workload & heart rate. Some items anchored with verbal expressions ( e.g., very light = 9, very hard = 17). ** Borg designed this category scale to display ratio properties. Some items are again anchored with verbal expression. *** Krupp, personal communication. **** In Swedish. English translations of each item provided, but not validated. ***** A variation on the Chalder Fatigue Scale.

Table 2.2 (continued)

Name of Scale Author, Year Initial Specified Fatigue Item Item Time Frame Population Subscales Length Scoring Myasthenia Gravis (Grohar-Murray, Myasthenia Perception of fatigue, 26 1-5 In general Fatigue Scale et al., 1998) gravis task avoidance, since illness (MGFS) observable motor onset signs or symptoms Multidimensional (Belza, 1995) Rheumatoid Degrees, severity, 15 1-10 One item Assessment of Fatigue arthritis distress, impact on asks for (MAF) activities of daily duration living Multidimensional (Smets, Garssen, Students, General fatigue, 20 1-7 Not stated Fatigue Inventory Bonke, & De physicians, physical fatigue, Haes, 1995) cancer, CFS, mental fatigue, soldiers reduced motivation, reduced activity Swedish (Elizabeth Healthy persons Lack of energy, 25 0-10 At present Occupational Fatigue Åhsberg & in 16 different physical exertion, Inventory (SOFI) Gamberale, 1998) occupation physical discomfort, lack of motivation, sleepiness Multi-component (Paul, Beatty, MS, myasthenia Mental, physical 15 0-5 At present, Fatigue Scale Schneider, Blanco, gravis & compared & Hames, to recent 1998)***** past Multidimensional (Stein, et al., Cancer Global, somatic, 83, (short 0-4 Last week Fatigue Symptom 1998) effective, behavioral, form: 30 Inventory (MFSI) cognitive symptoms items) of fatigue Fatigue Descriptive (Iriarte, MS Spontaneous mention 5 0-3 Note stated Scale (FDS) Katsamakis, & de of fatigue, antecedent Castro, 1999) conditions, frequency, impact on life Fatigue Symptom (Hann, et al., Cancer Intensity, duration, 13 0-10 Past week Inventory (FSI) 2000) impact on quality of life Rochester Fatigue (Schwid, et al., MS Lassitude (reduced 12 (1 item, Visual Past 2 hours Diary (RFD) 2002) energy) 12Xover analogue 24 hours) IOWA Fatigue Scale (Hartz, Bentler, & Primary care Cognitive, fatigue, 11 5-point scale (IFS) Watson, 2003) patients energy, productivity Child Fatigue Scale (Hockenberry et Children with Lack of energy, not 14 Frequency Past week (CFS) al., 2003) cancer (also able to function, (yes/no), versions for altered mood intensity (1- parents & staff) 5) * Designed to increase linearly with workload & heart rate. Some items anchored with verbal expressions ( e.g., very light = 9, very hard = 17). ** Borg designed this category scale to display ratio properties. Some items are again anchored with verbal expression. *** Krupp, personal communication. **** In Swedish. English translations of each item provided, but not validated. ***** A variation on the Chalder Fatigue Scale.

Two fatigue assessment scales such as CR-10 scale (G. Borg, 1982; G. B. E.

Borg, 1987) and SOFI (Elizabeth Åhsberg & Gamberale, 1998; E. Åhsberg, Gamberale,

& Gustafsson, 2000) for subjective fatigue measure have been widely used both in industries and in laboratories. A brief literature review for both scales is provided in section 2.2.1.1 and 2.2.1.2.

2.2.1.1 Rating of perceived exertion scale

Rating of Perceived Exertion (RPE) Scale is used to measure the perception of feeling for physical load that is imposed on an individual (G. Borg, 1970b). The modified

CR-10 scale known as CR10 (Gunnar Borg, 1998) is commonly used to measure ratings of perceived exertion. CR10 means category ratio scale between 0 and 10, although the initial development of the scale was a category scale with no zero at the beginning of the scale.

Table 2.3 Borg ratings of perceived exertion (RPE) scale (G. Borg, 1970b).

6 No exertion at all 7 8 Extremely light 9 Very Light 10 11 Light 12 13 Somewhat hard 14 15 Hard(heavy) 16 17 Very hard 18 19 Extremely hard 20 Maximal exertion

Both category scale and CR10 scales are given in Table 2.3 and Table 2.4 respectively. The number values in Table 2.3 represent heart rates as ten times of the number values such as 6 for 60 beats per minute, 7 for 70 beats per minute, 20 for 200 beats per minute and so on. Instead of absolute values of heart rate comparison, Borg later developed the category ratio (CR10) scales to measure perceived exertion based on individual’s perception (G. A. Borg, 1982) which is given in Table 2.4. The 0 in CR10 represents no perceived exertion, and 10 represents the maximum perceived exertion. The last dot (.) after the number 10 represents the absolute maximum or highest possible exertion (G. A. Borg, 1982).

Table 2.4 Borg category ratio (CR10) scale (G. A. Borg, 1982).

0 Nothing at all “No Perception” 0.3 0.5 Extremely weak Just noticeable 1 Very weak 1.5 2 Weak Light 2.5 3 Moderate 4 5 Strong Heavy 6 7 Very strong 9 10 Extremely strong “Max Perception” Absolute Maximum Highest possible

Studies have observed that sitting discomfort or perceived exertion is related to sitting fatigue (Helander & Zhang, 1997; Uenishi, Tanaka, Yoshida, Tsutsumi, &

Miyamoto, 2002). Moreover, the most widely used fatigue measuring scales (Swedish 16

Occupational Fatigue Inventory (SOFI)) have measured perceived exertion to predict physical fatigue (Elizabeth Åhsberg, Garnberale, & Kjellberg, 1997). Many studies claim that increasing the duration of study would introduce fatigue in their study participants

(Ahmed, 2010; Cham & Redfern, 2001). Therefore, perceived exertion over time can be a measure for fatigue (Cham & Redfern, 2001; Orlando & King, 2004). Borg (1982) scales have been used to measure fatigue in many previous studies, especially for prolonged and mentally demanding activities (Bansevicius, Westgaard, & Jensen, 1997; Cham &

Redfern, 2001). Instead of perceived exertion, perceived fatigue can be solicited to measure unidirectional subjective fatigue (Bansevicius, et al., 1997).

2.2.1.2 Swedish Occupational Fatigue Inventory (SOFI)

The Swedish Occupational Fatigue Inventory (SOFI) is the most widely used instrument to measure fatigue developed due to work tasks in industries and laboratories for healthy population. The validity and reliability of SOFI have been observed over a wide range of populations (Ada, Chetwyn, & Jufang, 2004; E. Åhsberg, et al., 2000;

González Gutiérrez, Jiménez, Hernández, & López López, 2005; Johansson, Ytterberg,

Back, Holmqvist, & von Koch, 2008), tasks (E. Åhsberg, et al., 2000; Linsey M. Barker

& Nussbaum, 2011; Muller, Carter, & Williamson, 2008) and shift work (Karlson et al.,

2006). Moreover, the scale has been proven valid and reliable to measure both mental and physical fatigue simultaneously (E. Åhsberg, et al., 2000; Elizabeth Åhsberg, et al., 1997;

Linsey M. Barker & Nussbaum, 2011). The right three columns in Table 2.5 represent the long form of the SOFI scale consisting of a total of 25 dimensions/expressions while the short form is reduced to five dimensions by factor analysis (Elizabeth Åhsberg, et al.,

1997). 17

The left column in Table 2.5 describes the concept of total fatigue comprising both physical and mental fatigue practiced by some researchers (Babski-Reeves &

Crumpton, 1999; Linsey Marinn Barker, 2009).

Table 2.5 Swedish Occupational Fatigue Inventory Scale (SOFI) and the type of fatigue

Type of Perceived Dimension Sub-dimension Range: 0 to 10 (“not at all” Fatigue to very high degrees) Breathing heavily Out of breath Physical Exertion Taste of blood Sweaty Palpitations Physical Fatigue Aching Hurting Physical Discomfort Stiff joints Numbness Tense muscles Uninterested Passive Lack of Motivation Indifferent Lack of Initiative Mental Fatigue Listless Sleepy Yawns Sleepiness Drowsy Fall asleep Lazy Overworked Spent Total Fatigue Lack of energy Drained Worn out Exhausted

2.2.2 Objective measurement of fatigue

In contrast to the subjective measure of fatigue, fatigue is difficult to measure objectively. Some researchers have argued that fatigue can only be measured validly by

using self-reported fatigue (Hancock & Desmond, 2001). Physiological changes are linked to the development of both cognitive and physical fatigue or vice versa (Satoshi et al., 2009). However, the association between objective physiological changes and perceived fatigue have not been observed to be significant, especially for low intensity physically demanding tasks (de Looze, Bosch, & van Dieen, 2009). Subtle changes in physiology for a short period of time may multiply over time to produce a significant change over a prolonged period of time. Nevertheless, these low physically and highly mentally demanding work-tasks have not been studied for a long period of time to measure the effect of fatigue. Many of the objective measures that have been used in the previous studies may not be generalized to other contexts. For example, mental fatigue has been observed to be correlated with eye blink rate in general (Stern, Boyer, &

Schroeder, 1994), information processing speed in patients with chronic fatigue syndrome (J. H. M. M. Vercoulen et al., 1998), level of physical activity over a two-week period in patients with chronic fatigue syndrome (Bazelmans, Bleijenberg, Vercoulen, van der Meer, & Folgering, 1997), a cognitive decline in short term memory for healthy subjects (D. van der Linden, M. Frese, & T. F. Meijman, 2003; van der Linden, Frese, &

Sonnentag, 2003). Physical fatigue has been found to be associated with muscle fatigue

(Vollestad, 1997), heart rate and heart rate variability (Bricout, Dechenaud, & Favre-

Juvin, 2010). Some objective measures for different types of fatigue are listed in Table

2.6. Most of these objective measures are significantly contextual and should not be generalized to other contexts.

Table 2.6 Objective measures of fatigue

Objective Reference Subject Subject Type of Study Task Task Findings Measure Age Fatigue Duration Type intensity (years) Measured Heart rate, (Hurum, 44 CFS 12 – 18 CFS 24h No task Low At night (sleep), HR, mean systolic & Sulheim, Patient & 52 arterial blood pressure and diastolic Thaulow, & Healthy diastolic blood pressure blood Wyller, 2011) were significantly higher pressure in CFS patients as compared with controls (p < 0.01). During daytime, HR was significantly higher among CFS patients (p < 0.05), whereas blood pressures were equal among the two groups. Physical (Ward, 1941) 600M & General Industrial Low to increased pulse rate, low variables 1200F with 4 Objective Task moderate blood pressure, pallor, percent CFS Measure tremor, and weight loss Heart rate (Hancock & 16 20-24 Total 3h Nursing high Heart rate was Desmond, Fatigue Task significantly affected by 2000a) (mental the simulated nursing task. and However, heart rate physical) variability was not found significant Oxygen (Hancock & 25 18-24 Physical Exercise High Fatigue, work decrement, consumptioDesmond, Fatigue and endurance were not n rate 2001) reflected in oxygen consumption rates. Interface (de Looze, et 12M & 15F 20-30 Comfort 15min Sitting, low Interface pressure was pressure al., 2009) and driving found more related to Discomfor comfort than discomfort t In-chair (McArdle, 1M & 7F 23-45 Sitting 2h Sitting, Low In-chair movement movement Katch, & Katch, Discomfor driving significantly increases over using 2010) t (or time interface Comfort) pressure Pupil (Chi & Lin, 10 18-32 Visual 20-60min Visual high Significant relationship diameter, 1998) fatigue Display between subjective rating eye Terminal of visual fatigue and the movement pupil diameter and eye velocity, movement Left & right (L. E. Hughes, 9M & 9F 18-33 45min Typing Low to Time pressure and force forearm Babski- Task high increase muscle activities. muscle Reeves, & (ECU & Smith- FCU) Jackson, 2007) Muscle (Kimura, Sato, 6M Localized 2h Typing, Medium Significant relationship activities in Ochi, Hosoya, Muscle 1Kg load to high established between upper trape & Sadoyama, Fatigue was subjective Rating of ziusmuscle 2007) (indirect placed on Perceived Exertion (RPE) measure) wrist and muscle activity

Table 2.6 (continued)

Objective Reference Subject Subject Type of Study Task Type Task Findings Measure Age Fatigue Duration intensity (years) Measured Back (McLean, 15F & Localized 1h 20min Regular low No significant change in muscles Tingley, 3M Muscle work in cervical paraspinal Scott, & Fatigue computers extensors, the lumbar Rickards, erector spinae, the upper 2000) trapezius, and the forearm extensors. (regular computer activities were performed) Upper arm (Seghers, 8M & 19-39 Discomfort 1h30min Regular Low to No statistical result and Jochem, & 8F VDT medium provided between shoulder Spaepen, muscle fatigue and muscles 2003) EMG from Right m. trapezius pars descendens, Right m. deltoı¨deus pars anterior, Right m. splenius capitis, Right m. sternocleideomasoideus, Left m. trapezius pars descendens, Left m. deltoı¨deus pars anterior, Left m. splenius capitis, Left m. sternocleidomastoid FCU, ECU (Gerard, 16M 22-57 Discomfort, 90min Typing Low to No objective muscle Armstrong, localized tasks medium fatigue observed, but Martin, & muscle significant increase in Rempel, fatigue discomfort which 2002) accounted as fatigue for the study Interface (Porter, Gyi, 8M & Mean 40 Discomfort 2.5h Real Road Low No significant pressure & Tait, 10F with std Driving relationship between 2003) =12 discomfort and subjective rating of discomfort Interface (Kyung & 12M & 20-35 Comfort & 15min Laboratory Low Interface pressure was pressure Nussbaum, 15F Discomfort Simulation more associated with 2008) comfort than discomfort Muscle (Babski- 4M & 18-33 Localized 2h VDT task low No significant difference activity in Reeves, et 4F Muscle was found for a L1, L5 and al., 2005) Fatigue particular sitting C7 position. However, significant difference was observed between different sitting positions.

Two of the objective measures, including change in resting heart rate and saliva cortisol concentration have been used in many previous studies to measure fatigue objectively, which are discussed in section 2.2.2.1 and 2.2.2.2.

2.2.2.1 Heart rate & heart rate variability

Both heart rate (Causse, Sénard, Démonet, & Pastor, 2010) and heart rate variability (Ahsan, Herbert, Toshio, & Marimuthu, 2010; Tiller, McCraty, & Atkinson,

1996) have been used to measure factors such as workload (Blain, Meste, Blain, &

Bermon, 2009) and stress (Causse, et al., 2010), which causes fatigue (Dorrian, Baulk, &

Dawson, 2011; Grech, Neal, Yeo, Humphreys, & Smith, 2009; Hancock & Desmond,

2000b), especially for the prolonged periods of work tasks (Dorrian, et al., 2011; Sood,

Nussbaum, & Hager, 2007). Heart rate variability can differentiate the activities between the parasympathetic and sympathetic nervous systems (Ahsan, et al., 2010) by which the functions for autonomic nervous system can be monitored (Tiller, et al., 1996). For example, the imbalance of the autonomic system can be detected by heart rate variability

(Karita, Nakao, Nishikitani, Nomura, & Yano, 2006), which is an indication of fatigue

(Masaaki Tanaka et al., 2011). Increase in sympathetic activities and decrease in parasympathetic activities have been recently determined as the underlying cause of daily fatigue (Jiao, Li, Chen, & Wang, 2005; M. Tanaka, Mizuno, Tajima, Sasabe, &

Watanabe, 2009; Masaaki Tanaka, et al., 2011). More specifically, prolonged cognitive load increases sympathetic activity while decreasing parasympathetic activity significantly resulting in mental fatigue (Mizuno et al., 2011). Because the activities of the autonomic nervous system are changed based on task demands, measuring sympathetic and parasympathetic activities by using heart rate (Hurum, et al., 2011) and 22

heart rate variability (Collet, Averty, & Dittmar, 2009) analysis can be used as an objective measure for fatigue (Boneva et al., 2007). Moreover, the heart rate measure is very effective for tasks that have both mental and physical parts (e.g. sports game, badminton, soccer, cricket, etc.) (Bricout, et al., 2010). Heart rate and heart rate variability have been observed to be a good measure for physical and mental fatigue, respectively (Bricout, et al., 2010). Many previous studies have found heart rate and heart rate variability as sensitive measures for fatigue (Jiao, et al., 2005; Yamamoto, LaManca,

& Natelson, 2003; Yoshiuchi, Quigley, Ohashi, Yamamoto, & Natelson, 2004).

2.2.2.2 Saliva cortisol concentration

Cortisol hormone has been observed as a stress hormone, regulated by hypothalamic-pituitary-adrenal axis (HPA) (Chrousos, 1995; Chrousos Gp, 1992).

Hypothalamus secretes corticotropin releasing hormone (CRH) according to the information received, including other hormones, serotonin and dopamine levels, immune system, and cortisol hormone itself. Whereas, the pituitary gland releases hormones such as adrenocorticotropic hormone, (ACTH) acts on the adrenal glands to produce the cortisol hormone, which facilitates balancing the functions of the body (Figure 2.1) (Hall

& Guyton, 2011; Saladin, 2008).

The cortisol hormone is secreted in four phases for healthy individuals with no medical conditions as given below (Weitzman et al., 1971):

1. Phase 1: A 6-hr period of “minimal secretory activity” (4 hr before and 2

hr after lights out);

2. Phase 2: A 3-hr period called “preliminary nocturnal secretory episode”

(3rd to 5th hr of sleep); 23

3. Phase 3: A 4-hr period, the “main secretory phase” (6, 7, 8 hr of sleep and

1st hr after awakening); and

4. Phase 4: The 11 hr of “intermittent waking secretory activity.

Figure 2.1 Function of Hypothalamic-Pituitary-Adrenal Axis

Picture Reference: left (Costanzo, 2010; Gross & Winstead, 2009) right (Papadopoulos & Cleare, 2012). Dotted and solid lines represent negative and positive feedback respectively.

Figure 2.2 Diurnal salivary cortisol level

Picture Reference: (Sephton et al., 2003).

Usually the cortisol level is lowest around midnight and highest at about half-an- hour after waking and can be visualized in Figure 2.2 (Dockray, Bhattacharyya, Molloy,

& Steptoe, 2008; Kirschbaum & Hellhammer, 1994; Sephton, et al., 2003).

In addition to responding to stress, the cortisol hormone regulates body functions, including the circulatory system, immune system, nervous system, metabolism of fats, carbohydrates, and proteins (Hall & Guyton, 2011). The trends in cortisol concentration follow a diurnal pattern with the levels being the maximum in the morning and minimum at midnight (Figure 2.2). Many studies have observed that individuals usually experience an increase in fatigue over time, which is highest at the end of a working day . Therefore,

the salivary cortisol concentration could be negatively correlated with fatigue if individuals are not conditioned.

Figure 2.3 An example of the change in cortisol levels in mentally demanding tasks

Picture reference: (Jens C. Pruessner, Hellhammer, & Kirschbaum, 1999).

Figure 2.4 Trend in salivary cortisol concentrations between weekdays (left) and weekends (right).

Picture reference: Solid and empty circles represent high and low stress, respectively (A. Dahlgren, Kecklund, & Akerstedt, 2005).

The trend of diurnal cortisol secretions is altered by the nature of work-tasks to cope with the challenges imposed by the tasks (A. Dahlgren, et al., 2005; Jens C.

Pruessner, et al., 1999; H. Webb et al., 2008; H. E. Webb et al., 2011). In contrast to the usual logarithmic decline in cortisol concentrations (Figure 2.4), it could rise or flatten under physical or mental challenges or both (Figure 2.3) (Greig, Marchant, Lovell,

Clough, & McNaughton, 2007; Jens C. Pruessner, et al., 1999; H. Webb, et al., 2008).

Flattening of cortisol secretions is still considered higher relative to the usual cortisol concentration.

Workload has been observed to affect the levels of cortisol concentration in the early morning (A. Dahlgren, Akerstedt, & Kecklund, 2004; A. Dahlgren, et al., 2005; A.

Dahlgren, Kecklund, & Akerstedt, 2006). Morning cortisol concentration for patients with chronic fatigue syndromes (CFS) has been observed to be significantly low,

meaning the restoration of cortisol levels has not been achieved due to the HPA axis dysregulation (Cleare et al., 2001; Demitrack et al., 1991; Nater et al., 2008). Hyper activities of HPA axis have also been observed for patients with CFS to cope with the challenges imposed by diseases or other factors (Gottschalk et al., 2005).

Studies have shown that significantly higher cortisol responses occur during mentally demanding work tasks performed for a prolonged period of time (Bohnen,

Houx, Nicolson, & Jolles, 1990; Engelmann et al., 2011). Therefore, changes in cortisol hormone level as an objective measure for fatigue have been observed (Adam, Hawkley,

Kudielka, & Cacioppo, 2006; Chida & Steptoe, 2009; Kumari et al., 2009; Nozaki et al.,

2009; J. C. e. a. Pruessner, 1997; Rubin, Hotopf, Papadopoulos, & Cleare, 2005a). The cortisol concentration for both populations with and without medical conditions has been observed to be significantly sensitive with respect to fatigue, either induced by diseases or work tasks or some other factors (Chida & Steptoe, 2009). Moreover, cortisol secretion by the hypothalamic-pituitary-adrenal (HPA) axis determines “hypercortisolemia” which is connected with low sleep efficiency and fatigue; while “eucortisolemia” or

“hypocortisolemia” is associated with high sleep efficiency and objective sleepiness

(Vgontzas, Bixler, & Chrousos, 2006). Because sleep is significantly related to fatigue and the secretion of cortisol is affected by the sleep quality, cortisol concentration should be a reliable measure of fatigue (Anna Dahlgren, Kecklund, & Åkerstedt, 2005;

Strickland, Morriss, Wearden, & Deakin, 1998).

Hence, salivary cortisol concentration can be a good biochemical measure to assess negative health consequences (H. Webb, et al., 2008) resulting in fatigue over time

(A. Dahlgren, et al., 2005, 2006), especially for mentally demanding work-tasks (Bohnen,

et al., 1990). Moreover, work-tasks which are less physically demanding (e.g. seated work-tasks) and highly mentally challenging (e.g. programming and computer simulation) alter the HPA axis activities, which changes the salivary cortisol concentration (H. E. Webb, et al., 2011). Therefore, salivary cortisol concentrations can be used as an objective measure in studies related to fatigue.

2.2.3 Quantifying physical and cognitive fatigue simultaneously

A primary limitation in the literature is the lack of studies that have quantified physical and mental fatigue simultaneously. Interesting findings were reported from those studies that have quantified both mental and physical fatigue simultaneously. For example, Liu et al. found that brain activity changes (changes in both electromyographic and magnetic resonance imaging signals) when working under muscle fatigue (Liu et al.,

2003); meaning that physical fatigue or extraneous physical activities change cognitive functions (Fukuda et al., 1994; LaManca et al., 1998). A recent study has also deemed that cognitive fatigue impairs physical performance (Marcora, Staiano, & Manning,

2009). These findings demand studies that quantify both mental and physical fatigue regardless of the structure of the tasks, such as primarily physical in nature, primarily cognitive in nature, or mixed (Linsey M. Barker & Nussbaum, 2011). Moreover, it is still unknown that whether fatigue is primarily because of physical or psychological variables, or both (DeLuca, 2005). Therefore, fatigue should be studied as a whole rather than localized.

2.2.4 Fatigue and performance

Many previous studies have observed significant performance declines due to fatigue (Bartley & Chute, 1947; Hockey & Earle, 2006; Huang et al., 2009; R. G. Hughes

& Clancy, 2008; Patterson & Yealy, 2010; Welford, 1968). Studies have also found relationships between prolonged activities and the decline in performance over an extended period of time (van der Linden & Eling, 2006; D. van der Linden, M. Frese, &

T. F. Meijman, 2003). For example, the Central Nervous System (CNS) has to work harder to sustain postural stability when fatigued (Kanekar, Santos, & Aruin, 2008).

However, the relationship between fatigue and performance has not been observed to be strongly correlated (Daniel, 1970; D. van der Linden, M. Frese, & T. F. Meijman, 2003) because often different physiological compensation is required in fatigue states to maintain the similar level of performance (Kanekar, et al., 2008; Rai, Foing, & Kaur,

2012; Robert & Hockey, 1997; Scott & Earnest, 2011; Dimitri van der Linden, Michael

Frese, & Theo F. Meijman, 2003). Performance has been observed to be unaffected by psychomotor tasks when subjects are required to maintain a certain level of performance; strategies have been changed to sustain the same levels of performance though (Kanekar, et al., 2008; Selen, Beek, & van Dieën, 2007). Therefore, many previous studies have not found significant relationships between the subjective fatigue and objective measures of fatigue, including working memory (S. K. Johnson, Lange, DeLuca, Korn, & Natelson,

1997), short-term memory (Susan K. Johnson, Deluca, Diamond, & Natelson, 1998;

Krupp & Elkins, 2000), executive function, complex attention (Krupp & Elkins, 2000), vigilance (Paul, et al., 1998), verbal fluency, and verbal memory (Krupp & Elkins, 2000).

As compared to a non-fatigued state, higher energy is required to perform the same

activities while an individual is in a fatigue state (Scott & Earnest, 2011). Higher physiological compensation is required to sustain performance when fatigued (Gates &

Dingwell, 2010; Hockey & Earle, 2006; Rai, et al., 2012; Robert & Hockey, 1997; Selen, et al., 2007). When resources are replenished, a performance decline is obvious

(Wickens, 2002, 2008). Some recent studies have also observed both performance decline and physiological compensation (e.g. change in muscle activities, CNS responses)

(Huysmans, Hoozemans, van der Beek, de Looze, & van Dieën, 2008; Kanekar, et al.,

2008). Therefore, fatigue will either affect performance or increase physiological cost or both. Most studies, in general, have observed that performance is affected by fatigue

(Linsey M. Barker & Nussbaum, 2011).

2.3 Factors that affect fatigue

Many factors that affect the development of fatigue, include gender, age, ethnicity, marital status, sleep hours, occupation, time spent in the occupation, weekly working hours, shift schedule, shift length, work setting, daily rest after work, other responsibilities, exercise, food habit and families. The top three factors, including sleep and shift, workload and the duration of work-tasks, have been considered as the statistically significant predictor of fatigue in the workplace. US National Health

Interview survey conducted between 2004 and 2008 reveals that shorter sleep and longer work-hours, which significantly contribute to fatigue, increase the risk of task-dependent injury (Lombardi, Folkard, Willetts, & Smith, 2010). Rest breaks within workdays, between workdays and in the weekends, if not taken properly, could substantially accumulate fatigue over time (Hooff, Geurts, Kompier, & Taris, 2007). In addition to the disturbance of sleep and workload (or work-hours), rest breaks have been considered as 31

one of the most effective tools to recover from fatigue accumulation (Bakker, Demerouti,

Oerlemans, & Sonnentag, 2013; Binnewies, Sonnentag, & Mojza, 2009; Sonnentag,

Binnewies, & Mojza, 2008; Sonnentag & Fritz, 2007). In regard to the rest-breaks, off- work activities could be another type of work but not the primary occupation, have been considered significantly effective to recover from fatigue (ten Brummelhuis & Bakker,

2012). Factors that affect fatigue are discussed here.

2.3.1 Sleep and shift work

Average daily sleep of an individual was significantly higher in 1910 (9 hours per night) than today (7 to 7.5 hours per night) (Coren, 1997). A recent poll by the National

Sleep Foundation revels that 29% felt sleepy at work and 36% have fallen asleep or nodded off while driving ("sleep in America Poll," 2008). Extensive research has been performed for sleep-related factors (e.g. sleep deprivation) in relation to the development of fatigue (Alison, Jill, & Adam, 2011; D. Dawson & Fletcher, 2001). Change in working shift causes sleep disturbance due to the change of circadian rhythm (Torbjörn Åkerstedt,

2003), which has also proven to be one of the reasons for fatigue (Torbjörn Åkerstedt, et al., 2002; Torbjörn Åkerstedt & Wright Jr, 2009; Östberg, 1973). Poor sleep has been proven to be one of the most important predictors of fatigue as compared to many other factors, including work load, gender, and exercise (T. Åkerstedt et al., 2004).

2.3.2 Workload

Workload has been proven to be one of the primary causes of fatigue in working population (Maarten AS Boksem, Meijman, & Lorist, 2006; Dorrian, et al., 2011;

Finkelman, 1994; Guastello et al., 2013; MacDonald, 2003). The limited ability to

process information is associated with performance decline (Eppler & Mengis, 2004;

Speier, Valacich, & Vessey, 1999; Wickens, 2008). In addition to information overload, information underload can also cause monotony, resulting in poor performance (Young &

Stanton, 2002) and fatigue (Finkelman, 1994) and increase the risk for negative health consequences (Frankenhaeuser & Gardell, 1976). One of the first explanations of performance under overload, and underload of information is given by the Yerkes-

Dodson Law of arousal and performance (Figure 2.5) (Yerkes & Dodson, 1908).

According to Yerkes-Dodson law, an optimal level of arousal by manipulating mental workload is required to achieve maximum performance. To overcome the situations of underload and overload, an optimum level of automation has been proposed so that the highest performance could be achieved.

Figure 2.5 Visualization of Yerkes-Dodson Law. 33

The term mental workload, however, is difficult to determine. No consensus definition of mental workload has been found in the literature (Table 2.7). Despite the disagreement on the definitions of mental workload, some common characteristics of mental workload are summarized as given below (Hacker, 1998):

1. Mental workload is associated with the task demands

2. Mental workload is conceptualized by cognitive information processing

which integrates mental processes, representations of work tasks, and

states of personal characteristics such as consciousness, mood, etc.

3. Mental workload is associated with multidimensional characteristics of

task requirements such as the design of tasks, individual behavior towards

the task performance, psychosocial aspects, etc.

Table 2.7 Definitions for mental workload

Source Definition (Wilson & “Mental workload refers to the portion of operator information Eggemeier, processing capacity or resources that is actually required to meet 1991) system demands.” (Gopher & “Mental workload may be viewed as the difference between the Donchin, 1986) capacities of the information processing system that are required for task performance to satisfy performance expectations and the capacity available at any given time.” (Kramer, “The cost of performing a task in terms of a reduction in the capacity Sirevaag, & to perform additional tasks that use the same processing resource.” Braune, 1987)

The complex nature of mental workload has also been understood in the development of two most commonly used scales to assess perceived workload (Gary &

Thomas, 1988; Hart & Staveland, 1988). The subjective workload assessment technique

(SWAT) developed by Gary and Thomas (1988) focuses on three characteristics of mental workload which are summarized as temporal effort, mental effort and stress.

These three components of mental workload have been determined to be significantly correlated with fatigue, for example, time (Aho, 2007), mental effort (Linsey M. Barker

& Nussbaum, 2011), and stress (Causse, et al., 2010). Many studies have determined a significant relationship between mental demand and fatigue (Ada, et al., 2004; D. van der

Linden, M. Frese, & S. Sonnentag, 2003). The other widely used scale NASA-TLX developed by Hart and Staveland (1988) includes two additional dimensions such as physical demand and performance to assess workload. The multidimensional and complex natures of workload include many elements that also significantly affect the development of fatigue (Hancock & Desmond, 2001).

2.3.3 Time

Fatigue has been reported at the end of a regular working day, and fatigue increase over time. Duration of work-tasks or amount of hours spent in occupation at the workplace is one of the primary factors affecting fatigue significantly (El Falou, et al.,

2003; Jensen, 2003; Østensvik, Veiersted, & Nilsen, 2009).

Figure 2.6 Loss of strength with fatigue

Figure 2.6 demonstrates that time interacts with load hyperbolically to affect muscle fatigue either peripheral or central, which has been well established throughout the literature (Hill, Poole, & Smith, 2002; Monod & Scherrer, 1965; Poole, Ward,

Gardner, & Whipp, 1988; Vanhatalo, Fulford, DiMenna, & Jones, 2010). Does this relationship hold for cognitive or total fatigue?

2.3.4 Rest breaks

Breaks have been proven to be one sensitive during the workday to reduce fatigue and other health consequences significantly. Breaks can help to minimize fatigue by at least by introducing (1) reduced stress and promoted enjoyment, (2) increased health awareness and facilitated behavior change, and (3) enhanced workplace social interaction

(W. C. Taylor et al., 2013). However, a clear understanding of how and when these 36

breaks should be introduced has not been studied enough. Is it the choice of individuals or is it assigned by the employers? The current practice in industries and service sectors recommend pre-schedule rest breaks after two hours, which may or may not be suitable with individuals’ preferences. Moreover, studies show that micro breaks are more important to maintain performance and manage negative health consequences, including fatigue (Dorion & Darveau, 2013; Henning, et al., 1997). In addition to the recommended breaks, self-selected micro breaks could be effective to manage fatigue (Tucker, 2003).

Recent studies show that self-selected rest breaks improve performance and reduce fatigue (Davy & Göbel, 2013). Because in many cases, for example, professional drivers can detect fatigue and the time that they need to take a rest break (Williamsonl, Friswelll,

Grzebieta, & Olivier, 2013). The mixed findings regarding rest break schedules, effective though, indicate that recommended pre-schedule breaks, self-selected rest breaks or frequent micro breaks, could be efficient for managing fatigue (Arlinghaus et al., 2012;

Williamson & Friswell, 2013).

Proper rest breaks between workdays and weekends are as important as breaks within the workday (Hooff, et al., 2007). One of the first attempts to define and understand fatigue in the workplace was developed primarily by focusing on the recovery of fatigue (Eo Grandjean, 1968). Figure 2.7 illustrates the causes of fatigue, which accumulates over time if not recovered respectably (Eo Grandjean, 1968).

Figure 2.7 Causes of fatigue by Grandjean's

Picture reference: (Griffith & Mahadevan, 2011)

2.3.5 Gender and age

The effect of gender has been observed to be significant in physical fatigue

(Linsey Marinn Barker, 2009; Billaut & Bishop, 2012; Kent-Braun, Ng, Doyle, & Towse,

2002; Laurent et al., 2010). Working females experience substantially elevated levels of perceived fatigue, when they have more responsibilities such as household-work, do not have time to take exercise, and with other negative factors (T. Åkerstedt, et al., 2004;

Karlqvist, Tornqvist, Hagberg, Hagman, & Toomingas, 2002; Loge, Ekeberg, & Kaasa,

1998; Steele et al., 1998). For an example, fatigue has been reported significantly higher by working women from the Indian subcontinent (the study only included Pakistani,

Indian and Bangladeshi). These women are often responsible for more household work as compared to their spouse (Bhui et al., 2011). Another example of women reporting fatigue significantly higher than men is associated with the lower levels of education and

occupational status (Jason et al., 1999). Similar results have also been observed in different age groups to affect physical fatigue significantly (Bilodeau, Henderson, Nolta,

Pursley, & Sandfort, 2001; Fell & Williams, 2008). Nevertheless, mixed results have been reported for mental and total fatigue with respect to age and gender (de Jong,

Candel, Schouten, Huijer Abu-Saad, & Courtens, 2005). No significant gender differences were observed in either mental or total fatigue, but for physical fatigue, in healthy nursing population (Linsey Marinn Barker, 2009).

2.3.6 Ethnicity

A few studies have been published comparing ethnic groups with respect to fatigue (Dinos et al., 2009). Moreover, the categorizations of ethnic group have not been performed methodically to determine the effect of ethnicity alone on fatigue (Dinos, et al., 2009; Jason La & et al., 1999; Njoku, Jason, & Torres-Harding, 2005). Studies have found a significant difference between the ethnic minority and majority with respect to fatigue, which must not be considered as the independent effect of ethnicity (Dinos, et al.,

2009; Jason, et al., 1999; Steele, et al., 1998). Nonetheless, the minority reported less fatigue when the studies are controlled by demographics, including only education and age (Cordero, Loredo, Murray, & Dimsdale, 2012). Studies have also reported no significant difference in ethnicity with respect to fatigue (Bhui, et al., 2011; Buchwald,

Manson, Pearlman, Umali, & Kith, 1996; Yennurajalingam, Palmer, Zhang, Poulter, &

Bruera, 2008). In contrast to ethnicity alone, socioeconomic status, unemployment, and being classified as minority significantly affects fatigue (R. R. Taylor, Jason, & Jahn,

2003). Moreover, fatigue studies comprising demographic and ethnicity, very important

though, have not been researched enough to reach to substantial conclusions (Di Milia et al., 2011; Noy, et al., 2011).

2.4 Fatigue predictive models

Many reliable models exist for localized muscle fatigue throughout the literature.

Unlike muscle fatigue, few models exist to measure physical and cognitive fatigue resulting in total body fatigue. One of the first models to quantify total body fatigue for healthy working populations was proposed by Babski-Reeves and Crumpton (1999).

Equation 2.1 illustrates the first model proposed to quantify the total body fatigue

(Babski-Reeves & Crumpton, 1999).

푖푗푘 = 1 푖푗푘 + 2 푖푗푘 + 3 푖푗푘 + 4 푖푗푘 + 5 푖푗푘 (2.1)

Where:

Overall fatigue level

Relative weighted value of each fatigue indicator

= Change in heart rate membership value

Tone task reaction time membership value

Level of tiredness membership value

Number of mental fatigue symptoms reported membership value

Number of physical fatigue symptoms reported membership value

And:

i = participant number

j = testing time

k = testing session number 40

The model was also validated for work-tasks in industrial (e.g. data entry operators and workers in manufacturing industries) and service sectors (e.g. nurses)

(Babski-Reeves & Crumpton, 1999; Babski-Reeves K., Crumpton-Young L., Riley J.,

Nitcavic L., & Gentry H., 2000). The model could predict fatigue 52.5% accurately for nursing work-tasks. Another model to assess total fatigue experienced by nurses in healthcare industries is provided in Figure 2.8 (Linsey Marinn Barker, 2009). Both models by Babski-Reeves (1999) and Barker (2009) are significantly task specific and both models simply include task variables to quantify fatigue. Barker’s model is highly associated with the impact of fatigue on performance, which is still equivocal and considerably task specific, (Ackerman & Kanfer, 2009; van der Linden & Eling, 2006; D. van der Linden, M. Frese, & T. F. Meijman, 2003; D. van der Linden, M. Frese, & S.

Sonnentag, 2003), rather than what causes fatigue and how it should be assessed (Linsey

M. Barker & Nussbaum, 2011; Barker Steege & Nussbaum, 2013).

Figure 2.8 Conceptual model of fatigue and performance in healthcare workers

Picture reference: (Linsey Marinn Barker, 2009).

Another excellent model, Swedish Occupational Fatigue Inventory (SOFI) described in Section 2.2.1.2 and Table 2.5 (E. Åhsberg, et al., 2000; Elizabeth Åhsberg, et al., 1997) has been widely used to quantify fatigue induced by physical work-task, mental work-task or both in the workplace (Barker Steege & Nussbaum, 2013). The

SOFI model by Åhsberg only assesses the current states of fatigue levels without considering the cause of the development of fatigue, which is one of the primary criteria to define or assess fatigue.

Fatigue is highly complex and comprised of many factors besides just the work- task. For example, sleep and variables associated with sleep have been proven to be more effective in predicting fatigue than the variables associated with the work-task itself

(James C. Miller, 2005; James C Miller & Eddy, 2008). Therefore, numerous models to predict fatigue have been developed by utilizing sleep variables.

Because fatigue is complex and affected by many factors, including age, work- rest break, sleep deprivation, motivational factors, coping strategies, total time spent in occupation and circadian disruptions (Gawron, French, & Funke, 2001), a comprehensive model should be developed to assess fatigue. Few such models have been found throughout the literature. One of the conceptual and comprehensive models is depicted in

Figure 2.9 (Qiang, Lan, & Looney, 2006).

Figure 2.9 Bayesian network of fatigue variables

Picture reference: (McLauglin, 2007).

Figure 2.10 depicts a recent theoretical model and it incorporates an extensive list of factors that could potentially affect fatigue in the workplace (Di Milia, et al., 2011).

This model includes both factors associated within the body and outside the body.

However, little has been discussed in the model about the workload, and the amount of time spent in the work-task to affect fatigue, rather the model is highly focused on personal traits, demographic and other job related factors (Di Milia, et al., 2011).

Figure 2.10 Potential endogenous and exogenous variables that may be linked with fatigue.

Picture Reference: (Di Milia, et al., 2011)

A recent review of models to predict fatigue in the workplace only discovered models with sleep as the primary input (D. Dawson, et al., 2011), which may predict fatigue for jobs with shift work but not regular daytime mentally demanding jobs where other variables could be even more pertinent.

As the physically demanding jobs are decreasing, mentally demanding jobs are also increasing (M. A. Boksem & Tops, 2008; van der Ploeg, et al., 2013). Nevertheless, a model to predict fatigue induced by mentally demanding work-tasks has yet to be created (M. A. Boksem & Tops, 2008).

Many models have been developed in the medical field for patients with chronic fatigue syndrome due to different types of disease (B. Piper, 1989; Stein, et al., 1998).

However, generalization of these models to occupational ergonomics needs further investigation. Moreover, none of these models include time (the running clock during the working day) factor to predict fatigue; time spent in occupation has been proven to be one of the most important contributing factors to the development of fatigue in the workplace (Bansevicius, et al., 1997; Dembe, Erickson, Delbos, & Banks, 2005; Härmä,

2006; Schaufeli, Taris, & Van Rhenen, 2008; Smith, Folkard, Tucker, & Macdonald,

1998; Van der Hulst, 2003).

2.4.1 Two-step quantitative model to predict fatigue

Most models currently existing within the literature are inadequate to explain fatigue over time in the workplace due to at least the two following reasons:

1. Comprehensiveness; not including all or a sufficient number of factors that

could predict fatigue in the workplace; and not explaining a significant

amount of variation.

2. Time (the running clock during the day); job-related fatigue increases over

time during the working day. Therefore, an interactive relationship between

time and the other factors that cause fatigue could substantially affect

occupational fatigue, similar to the significant hyperbolic relationship (Figure 45

2.6) between time and effort in neuromuscular fatigue (Monod & Scherrer,

1965; Vanhatalo, et al., 2010).

Figure 2.11 Factors that affect fatigue (step one of the two-steps model)

Figure 2.11 illustrates the first step of the proposed two-steps model to address the limitations in the current models to predict fatigue in the workplace. In the first step, a comprehensive list of factors that affect fatigue in the workplace is identified. The variables to predict fatigue can be considered as quantitative except ethnicity (Figure

2.11). The time in the model must be interpreted as a running clock during the working

shift, which supposedly interacts with other factors in the model to affect fatigue. In the second step, a quantitative mathematical relationship is hypothesized to predict fatigue

(equation (2.1) and Figure 2.12). The potential hyperbolic relationship is depicted in

Figure 2.12.

Figure 2.12 Conceptual hyperbolic relationship of time and fatigue load

In addition to muscle fatigue, the hyperbolic relationship for overall physical, cognitive or total fatigue has been conceptualized in many previous studies. In comparison to lower intensity, the higher intensity of a factor must increase fatigue

quicker over time, which inherently indicates an interaction only relationship where main effects of the individual factors are meaningless (Anna Dahlgren, et al., 2005). The mathematical relationship of the resultant interaction for fatigue is given in equation

(2.1).

푥푦 = 푐표푛푠푡푎푛푡 (2.1)

Where,

x = Time spent in the workplace

y = fatigue load = quantitative factors (e.g. change in resting heart rate, workload, daily sleep, etc.) that affect fatigue is defined as “fatigue load” in this dissertation

constant in the equation can be called an iso-fatigue constant, which is fixed for a particular situation or working condition.

CHAPTER III

METHOD

3.1 Experimental Design

An observational study in the field was performed to evaluate fatigue in prolonged, mentally demanding work tasks. However, the strategy of data collection has resulted in a repeated measure design of experiment where participants were randomly selected from two populations (Indian and Westerner), and each participant was measured over a four-hour time period. Ethnicity (Eth) was considered as a between- subjects factor and experimental clock time, or simply time (T), was considered as a within-subjects factor. Task-independent and personal factors, including the working shift (Sh); weekly exercise frequency (Ex); hours of daily sleep (Sl); hours of rest after work (DR); weekly working hours in primary occupation (W); total weekly working hours all of occupations (TW); and fatigue perceived at the end of a regular working day

(EDF) were also considered as between-subject factors.

3.1.1 Population model for the experiment

The population model provides the primary guidelines for statistical analysis of variance. Therefore, the population means model in equation (3.1) was developed.

𝑌 = + 𝛼 + 휀 + τ + 𝛼휏 + 𝑒 푖푗푘 푖 푘(푖) 푗 푖푗 푖푗푘 (3.1)

Where: 49

i=1,2; j=1, 2,……, 9, 10; k=1, 2,……, 7, 8

th th th 𝑌 Response within i group, at j time point for the k subject

Overall mean

th 𝛼 Effect of i ethnic group subject to

𝛼

휀 Error associated with subjects nested in ethnic groups, ,

independent and identically distributed

th τ Effect of j time point subject to

th th 𝛼휏 The interaction effect for the i ethnic group and j time point subject to

𝛼휏 𝛼휏

𝑒 Experimental error, , independent and identically distributed

The observations 𝑌 for the repeated measures model in Equation (3.1) have the properties described in Equation (3.2) to (3.5):

𝐸 𝑌푖푗푘 = … + 𝛼푖 + τ푗 + 𝛼휏 푖푗 (3.2)

2 2 2 2 𝑌푖푗푘 = 𝛾 = 휀 + (3.3)

2 2 ′ 𝑌 , 𝑌 ′ = 푘 ≠ 푘 푖푗푘 푖푗 푘 휀 (3.4)

2 ′ ′ 𝑌푖푗푘 , 𝑌 ′ ′ = 0 푖 ≠ 푖 푗 ≠ 푗 푖 푗 푘 (3.5)

Table 3.1 describes the expected mean squares, which are utilized to create appropriate statistical F-tests. An expected mean square table was created based on the table method algorithm (Kutner, 2005).

Table 3.1 Expected mean squares for ANOVA

Source of Variation Mean Square Expected Mean Square

Ethnicity, A MSA 𝛼 푡 푡 푎

Time, T MST 휏 푎 푡

Interaction A*T MS(AT) 𝛼

푎 푡

Subjects (S) (within factor A) MSEa 푡

Error MSEb

Between-subject factor ethnicity and within subject factor time (ethnicity and time are fixed, subjects are random). a = levels of ethnicity factor = 2, t = levels of time factor = 10, r = numbers of subjects with in group = 8.

3.2 Independent variables

Two independent variables (time and ethnicity) were studied. The time variable consisted of 10 levels, including two baseline assessments at the beginning of each two- hour session and 4 assessments during each two-hour session. Ethnicity consisted of two levels, Westerner and Indian (Asian) populations. Task-independent and personal factors identified previously (Sh, Ex, Sl, DR, W, TW, and EDF) were also studied as independent variables as these factors have been identified in the literature as affecting fatigue (see CHAPTER II).

3.2.1 Duration of the study

As earlier studies have determined a significant relationship between the duration of work tasks and fatigue (Drew Dawson, Chapman, & Thomas, 2012; Trinkoff et al.,

2011), the duration of the study is highly relevant. Therefore, a four-hour study duration simulated one half of a traditional work day and was used in this research.

3.3 Dependent variables

Multiple dependent measures, including both objective and subjective fatigue measures were studied. The data collection procedures, instrumentations and data cleaning procedures are discussed for each dependent variable in respective sections below.

3.3.1 Subjective measures of fatigue

Two subjective instruments, the Modified Borg CR-10 scale (Borg Scale) and the

Swedish Occupational Fatigue Inventory (SOFI) were used to measure participants’ subjective perceptions of fatigue.

3.3.1.1 Modified Borg CR-10 scale to measure fatigue

Both the Rating of Perceived Exertion (RPE) scale and the Category Ratio (CR-

10) scale have been widely used to measure both perceived exertion and overall fatigue

(E. Åhsberg, et al., 2000; G. Borg, 1970a). A modified Borg CR-10 (modified because perceived overall fatigue was solicited instead of perceived exertion) scale was used to measure perceived fatigue every 30 minutes over a four-hour study period. A total of 10 assessments were performed including the baseline measurements at the beginning of each two-hour session before and after a short 15-minute break. The scales were

displayed as they appear in APPENDIX B (Borg) and APPENDIX C (SOFI). Participants rated their perceived fatigue for specific body parts presented in random order

(APPENDIX E). Perceived fatigue was collected for (1) leg, (2) buttock, (3) lower back,

(4) upper back (5) shoulder- neck, (6) eyes, and (7) whole body. A total fatigue score for each 30-minute block was calculated by adding fatigue ratings for each body part, including the whole body (G. A. Borg, 1982; Loge, et al., 1998).

(3.6)

3.3.1.2 Swedish Occupational Fatigue Inventory (SOFI)

The Swedish Occupational Fatigue Inventory (SOFI) used in this study is given in

APPENDIX C (E. Åhsberg, et al., 2000; Elizabeth Åhsberg, et al., 1997). The short version of SOFI was used, and participants completed the survey every 30 minutes. A total multi-dimensional fatigue score for each 30-minute block was calculated by adding the fatigue ratings for five dimensions of SOFI (E. Åhsberg, et al., 2000; Loge, et al.,

1998).

(3.7)

3.3.2 Subjective measure of workload

Subjective perceptions of workload were measured using the NASA-TLX. While fatigue and workload are generally considered two distinctly different concepts, they

have been found to be related in previous studies. A total workload score for each 30- minute block was measured by adding the scores for six dimensions of NASA-TLX equation (3.8) (Hart & Staveland, 1988; Loge, et al., 1998). Similar studies have not identified any significant difference between weighted and un-weighted scores of NASA scores (DiDomenico, 2003; Ikuma, Nussbaum, & Babski-Reeves, 2009). Therefore, simple un-weighted scores will be used to calculated total workload measured by NASA-

TLX.

(3.8)

3.3.3 Objective measures of fatigue

Two objective measures, (1) change in heart rate (Duchon, Smith, Keran, &

Koehler, 1997) and (2) saliva cortisol concentration (Rai, et al., 2012), were collected.

3.3.3.1 Change in Heart rate (∆HR)

A Polar RS 400 heart-rate monitor (Polar Electro Oy, Professorintie 5, Fl-90440

Kempete, Finland; www.polar.fi) was used to measure heart rate continuously at a sampling rate of 1Hz. Raw heart rate data was downloaded to the Polar Pro-Trainer 5 software (Polar Electro Oy, Professorintie 5, Fl-90440 Kempete, Finland; www.polar.fi) for analysis at a later time.

The heart-rate monitor was placed across the chest so that the sensor sits right of the sternum. A wrist watch was worn on either hand or placed on the working desk to

minimize interference, but close enough to the chest sensor for continuous heart rate monitoring. To start the experiment, resting heart rate was calculated in a sitting position while participants were requested to sit back and relax until they reached a steady state resting heart rate defined to be 2 consecutive heart rate readings within 5 bpm. This procedure took 2 to 5 minutes. After recording resting heart rate, the heart rate wrist watch clock was started to begin the experiment. Average heart rate was also calculated during the steady state condition by collecting thee heart rate readings. Change in heart rate (HR) was used in all analyses. To compute HR, task heart rate was averaged for each 30-minute block and the resting heart rate was subtracted from the average, heart rate for the 30-minute block.

3.3.3.2 Saliva cortisol concentration

Six saliva samples; four samples during the experiment, one sample during early morning (30 minutes after waking up) and another sample during a non-work day

(between 2:00 and 3:00PM on a Sunday afternoon); were collected. The four samples collected during the experiment were collected before and after each two-hour session.

Samples were analyzed according to the ELISA technique (Salimetrics, State College,

PA). Participants were asked to chew a clinical cotton gum for a minute to take the saliva sample, which was then stored in a test tube and kept in an ice box. Once the experimental session was completed, the saliva samples were refrigerated at -100C until they were needed for analysis.

3.3.3.2.1 Weighted saliva cortisol concentration

In addition to raw saliva cortisol concentration, normalized saliva cortisol concentration weighted by morning cortisol was also computed. To eliminate individual differences, saliva cortisol concentrations were normalized by the saliva cortisol concentration taken during early-morning according to equation (3.9).

표 푚푎 푖푧𝑒 푎 푖푣푎 𝐶표 푡푖푠표 𝐶표푛푐𝑒푛푡 푎푡푖표푛 푏푎푠𝑒 표푛 푚표 푛푖푛 푐표 푡푖푠표 , 𝐶

𝐸푎 푦 표 푛푖푛 𝐶표 푡푖푠표 , 𝐶 푎 푖푣푎 𝐶표 푡푖푠표 𝐶표푛푐𝑒푛푡 푎푡푖표푛, 𝐶 = 𝐸푎 푦 표 푛푖푛 𝐶표 푡푖푠표 , 𝐶 (3.9)

3.3.3.2.2 Data cleaning method for saliva cortisol concentration to measure fatigue

Two types of data manipulation, Area Under the Curve with respect to ground

(AUCG) and increase (AUCI), were performed to analyze the cortisol concentrations in

푛 1

퐴𝑈𝐶𝐼 = 퐴𝑈𝐶𝐺 푚1 푡(푖) saliva samples (Fekedulegn et al., 2007) ((3.10) and 푖=1

(3.11)). These two measures of salivary cortisol concentration were obtained by using the method developed by Pruessner et al. (J. C. Pruessner, Kirschbaum, Meinlschmid, &

Hellhammer, 2003). AUCG estimates total cortisol secretion during the entire session and predicts the mean cortisol secretion, while AUCI measures the sensitivity of the

Hypothalamus-Pituitary-Adrenal (HPA) axis activity over time (Edwards, Clow, Evans,

& Hucklebridge, 2001; Fekedulegn, et al., 2007; Schmidt-Reinwald et al., 1999).

푛 1 푚(푖+1) + 푚푖 퐴𝑈𝐶 = 푡 𝐺 2 (푖) 푖=1 (3.10)

푛 1

퐴𝑈𝐶𝐼 = 퐴𝑈𝐶𝐺 푚1 푡(푖) 푖=1 (3.11)

Equation reference: (J. C. Pruessner, et al., 2003)

3.4 Participants

Sixteen self-reported healthy participants with no medical conditions (back pain, shoulder or neck pain, buttock pain, or headache) and 20/20 natural or corrected eye vision volunteered for the study (descriptive statistics are presented in Table 3.2). No other exclusion criteria were used. Eight Indian and eight Western participants were randomly selected for the study. Four participants from each ethnicity were observed

(completed the experiment) during morning hours (between 8:00AM and 12:00PM

(noon)). Another four participants from the same ethnicity were observed during afternoon hours (between 1:00PM and 5:00PM).

To aid in later analyses, groupings for self-reported task-independent and lifestyle factors were created and used in later analyses due to the small sample size. Logical groupings were made to test the effects of task-independent and personal variables on perceived fatigue scores. The logical groupings are provided in Table 3.3. Overall demographic statistics are provided in Table 3.2. The mean age of participants was 28.69 years . Mean daily hours of sleep were reported to be within a normal sleeping range ( ). Detailed demographic information can be found from Table 3.2 to Table 3.5. In comparison to the Westerners, Indian participants reported higher weekly working hours in the primary occupation as well as in all occupations. End of the day fatigue on a regular working day was also observed to be

higher for Indians than Westerners. Monday morning fatigue, weekly exercise frequency, and hours of daily rest after work were reported to be similar by both ethnic groups.

Table 3.2 Group wise demographic statistics

Variable Value Frequency Percent Variable Value Frequency Percent A 23 2 12.50 TW 35 1 6.25 24 1 6.25 44 1 6.25 25 3 18.75 48 1 6.25 26 1 6.25 50 3 18.75 29 1 6.25 54 1 6.25 30 2 12.5 56 2 12.50 31 3 18.75 60 2 12.50 32 1 6.25 64 1 6.25 35 1 6.25 70 3 18.75 39 1 6.25 74 1 6.25 Eth I 8 50.00 EDF 1 1 6.25 W 8 50.00 2 2 12.50 Sl 6.5 5 31.25 3 4 25.00 7.5 10 62.50 4 5 31.25 9.5 1 6.250 5 3 18.75 W 30 1 6.25 7 1 6.25 40 6 37.50 MMF 0 14 87.50 50 6 37.50 2 2 12.50 60 3 18.75 Ex 0 5 31.25 DR 1 1 6.25 2 1 6.25 2 7 43.75 3 6 37.50 3 8 50.00 4 4 25.00 A = Age of a participant in years, Sl = Hours of daily sleep, W = Weekly working hours in primary occupations, TW = Total weekly working hours in all occupation, EDF = End of the day fatigue, MMF = Monday morning fatigue, Ex = Weekly exercise frequency and DR = Daily rest in hours

Table 3.3 Frequency table for logical groupings

Variable Grouping Frequency Percent 23-29 8 50 Age in years (A) 30 2 12.5 31-39 6 37.5 Ethnicity (Eth) I 8 50 W 8 50 Hours of daily sleep (Sl) 6.5 5 31.25 7.5 10 62.5 9.5 1 6.25 Weekly working hours in the primary occupation (W) 30-40 7 43.75 50 6 37.5 60 3 18.75 Total weekly working hours in all occupations (TW) 35-48 3 18.75 50-56 4 25 56-74 9 56.25 Fatigue at the end of a regular working day (EDF) 1-2 3 18.75 3-4 9 56.25 5-7 4 25 Monday morning fatigue (MMF) 0 14 87.5 2 2 12.5 Weekly exercise frequency (Ex) 0 5 31.25 2-3 7 43.75 4 4 25 Daily rest after work (DR) 1-2 8 50 3 8 50

Table 3.4 Overall demographic statistics

Number of Mean Std Min Max Subjects Dev Age (years) 16 28.69 4.43 23 39 Sleep (hours) 16 7.31 0.73 6.5 9.5 Hours worked weekly 16 46.88 8.48 30 60 Total weekly working hours in all 16 56.94 10.51 35 74 occupations End of the day fatigue 16 3.69 1.41 1 7 Monday morning fatigue 16 0.25 0.66 0 2 Weekly exercise frequency 16 2.25 1.61 0 4 Daily rest after work (hours) 16 2.44 0.61 1 3

Table 3.5 Demographic statistics by ethnicity and working shift

Eth Variable N Mean Std Mini Maxi Sh Mean Std Minim Maxi Dev mum mum Dev um mum I Age (year) 8 29.25 2.35 25 32 M 30.75 4.69 23 39 Sleep (hour) 8 7.00 0.50 6.5 7.5 7.50 0.87 6.5 9.5 Weekly working hours 8 50.00 8.71 40 60 43.75 8.62 30 60 Total weekly working 8 63.00 8.89 50 74 54.13 11.43 35 74 hours End of the day fatigue 8 4.38 1.33 3 7 3.75 1.40 2 7 Monday morning fatigue 8 0.25 0.67 0 2 0.50 0.87 0 2 Weekly Exercise 8 2.13 1.70 0 4 2.13 1.70 0 4 frequency Daily rest after work 8 2.25 0.67 1 3 2.50 0.50 2 3 (hour) W Age (year) 8 28.13 5.77 23 39 A 26.63 2.97 23 31 Sleep (hour) 8 7.63 0.79 6.5 9.5 7.13 0.49 6.5 7.5 Weekly working hours 8 43.75 7.00 30 50 50.00 7.12 40 60 Total weekly working 8 50.88 8.29 35 64 59.75 8.68 48 70 hours End of the day fatigue 8 3.00 1.13 1 4 3.63 1.42 1 5 Monday morning fatigue 8 0.25 0.67 0 2 0.00 0.00 0 0 Weekly Exercise 8 2.38 1.50 0 4 2.38 1.50 0 4 frequency Daily rest after work 8 2.63 0.49 2 3 2.38 0.70 1 3 (hour) Note: Eth=Ethnicity, I=Indian, W=Westerner, Sh=Working shift, M=Morning, A=Afternoon

Table 3.5 shows demographic statistics by ethnicity within a working shift, indicating a balanced design of experiments with respect to both ethnicity and working shift. Most variables are comparable within each ethnicity group by working shift.

However, a few trends were observed. For example, weekly working hours in primary occupations were observed to be higher for Indian participants (Morning

and afternoon ) as compared to western participants

(Morning and afternoon ) in both working shifts. Similar observations were also seen for total weekly working hours in all occupations.

3.5 Power analysis

Power analysis is given in Table 3.7. The lowest power obtained from the study was for the saliva cortisol concentration for non-normalized data. However, normalized saliva cortisol concentration provides power of more than 90% considering that fact that if the interactions are not obtained significant. Strong power was obtained for the time variable. However, normalized cortisol produced more power for ethnicity and time.

Table 3.6 Demographic statistics by ethnicity within a working shift

Indian Westerner Sh P Vari N Mean Std Mini Maxi Mean Std Mini Maxi able Dev mum mum Dev mum mum M 4 Sl 40 7.25 0.44 6.5 7.5 7.75 1.10 6.5 9.5 W 40 45.00 8.77 40 60 42.50 8.40 30 50 TW 40 58.50 9.96 50 74 49.75 11.24 35 64 EDF 40 4.25 1.66 3 7 3.25 0.84 2 4 MMF 40 0.50 0.88 0 2 0.50 0.88 0 2 Ex 40 1.75 1.81 0 4 2.50 1.52 0 4 DR 40 2.25 0.44 2 3 2.75 0.44 2 3 A 40 29.50 2.32 26 32 32.00 5.99 23 39 A 4 Sl 40 6.75 0.44 6.5 7.5 7.50 0.00 7.5 7.5 W 40 55.00 5.06 50 60 45.00 5.06 40 50 TW 40 67.50 4.39 60 70 52.00 3.20 48 56 EDF 40 4.50 0.88 3 5 2.75 1.32 1 4 MMF 40 0.00 0.00 0 0 0.00 0.00 0 0 Ex 40 2.50 1.52 0 4 2.25 1.50 0 4 DR 40 2.25 0.84 1 3 2.50 0.51 2 3 A 40 29.00 2.38 25 31 24.25 0.84 23 25 Sh = Working shift, M = Morning, A = Afternoon, P = Participants, A = Age of a participant in years, Sl = Hours of daily sleep, W = Weekly working hours in primary occupations, TW = Total weekly working hours in all occupation, EDF = End of the day fatigue, MMF = Monday morning fatigue, Ex = Weekly exercise frequency and DR = Daily rest in hours

Table 3.7 Post-hoc power analysis

DV Source Ethnicity Time Borg 0.997 >0.999 SOFI 0.985 0.9950 ∆HR 0.290 >0.999 NASA 0.41 <0.999 CRT 0.597 0.954 NCRT 0.973 0.798

AUCI 0.427 0.242

AUCG >0.999 0.965

MAUCI >0.999 0.990

MAUCG >0.999 0.992 Borg = one-dimensional fatigue scores measured in Borg scale, SOFI = multidimensional fatigue scores measured in Swedish Occupational Fatigue Inventory, NASA = workload measured in NASA-TLX, ∆HR = change in heart rate (bit per minute), CRT=saliva cortisol concentration, NCRT=Normalized saliva cortisol concentration by morning cortisol, AUCI= area under the curve with respect to increase for salivary cortisol, AUCG= area under the curve with respect to ground for salivary cortisol, MAUCI= area under the curve with respect to increase for normalized salivary cortisol, and MAUCG= area under the curve with respect to increase for normalized salivary cortisol.

3.6 Procedure

Each participant was given a verbal and written description of the experiment and was required to complete an Informed Consent document approved by the Institutional

Review Board (IRB) for Research Involving Human Subjects at Mississippi State

University. Participants were asked to complete a demographic questionnaire

(APPENDIX A) after the informed consent procedure. The heart-rate monitor was then attached according to manufacturer guidelines, and a resting heart rate assessment was conducted. The first saliva sample and all baseline subjective assessments were collected 63

just prior to the start of the first two hours of testing. At each 30-minute interval within each two-hour testing block, the subjective fatigue and workload assessments were collected. After the end of the first two hours of testing, a 15-minute break was provided, and all measures were collected. Procedures for the first-two-hour test session were replicated for a second-two-hour testing session.

3.7 Data analysis

Appropriate descriptive statistics were calculated for all dependent variables with respect to all independent variables. ANOVA was conducted to determine the effect of all time independent variables and their two-way interaction with time on all dependent measures. Moreover, interaction means were analyzed using Tukey Post-Hoc LSD analyses when appropriate. In addition, stepwise regression and spearman correlations were also performed. Statistical analyses were performed using Statistical Analysis

System (SAS) version 9.3, and were considered significant at a significance level of 0.05.

3.7.1 Analysis of variance

Analysis of variance (ANOVA) for each dependent variable was performed according to the population model described in (3.1). All F-statistics were calculated utilizing the expected mean square table described in Table 3.1. Tukey’s adjusted least squares means post-hot analyses were performed for all significant findings when appropriate. Post-hoc analysis was performed using the principle of hierarchy, which suggests ignoring the main effects if an interaction term is significant.

3.7.2 Correlations

Spearman's rank correlation coefficient or Spearman's rho is considered more appropriate for Likert scale type data. All subjective assessment instruments used in this study could be considered as Likert scale data. For example, on a 0 to 10 scale, the response from a participant is perceived more as ranks than purely numeric. Therefore,

Spearman’s rank correlation coefficient was used for all correlation analyses. Correlation matrices were generated for all pairs of dependent variables. A raw correlation matrix was developed using the 30-minute block data. In addition to the overall correlation, correlation matrices were generated for each assessment time point separately to determine the micro-relationships. Because cortisol samples were collected before and after each two-hour session, a correlation matrix was also created utilizing data corresponding to the two-hour session.

3.7.3 Regression Analysis

Stepwise regression was used to develop fatigue predictive models. All significant variables and interactions identified through ANOVA were potential variables for the fatigue prediction model. A significance level for entry and significance level to stay for the stepwise procedure was set to 0.15. Model performance was assessed by using adjusted R2 values. Task-independent and personal data that were artificially grouped for ANOVA were returned to their raw form for the model building exercise.

For example, the total number of hours worked per week was no longer categorical, but rather the total number of reported hours worked was used as the input to the regression model.

The population model used to determine the hyperbolic relationship is given in equation ((3.12). However, population model given in Equation (3.13) was used to develop the fatigue predictive models.

𝑌 = + + + + 휀 0 1 2 2 12 2 (3.12)

푘 푘

𝑌 = 0 + 1 + 푖 푖 + 1푖 푖 + 휀 푖=2 푖=2 (3.13)

Where,

𝑌 Perceived fatigue measured either in Borg or SOFI

= Intercept

’s = parameters

T = Time, the running clock during the experiment

= A factor that causes fatigue, for example workload

k = number of variables.

휀 Error , independent and identically distributed

Utilizing both dependent and independent variables, 6 models were developed to predict subjective fatigue ratings (Borg and SOFI). Because a 15-minute rest-break was provided after the first two-hour session of the total four-hour observation, period measures such as change in resting heart rate, workload and fatigue could be significantly reduced, which were opposite to the first two-hour and second two-hour sessions.

Moreover, the trends of change in fatigue during the first and second session could be significantly different. Generally, a predictive model for a shorter period of time is considered more efficient than for a wider period of time. Therefore, fatigue predictive

models were developed for the first two-hour session, the rest-break, and the second two- hour session.

CHAPTER IV

RESULTS

4.1 Descriptive statistics

A written summary of descriptive statistics is provided here. Detailed descriptive statistics tables are provided in the APPENDIX Table F.1 through Table F.7. Perceived total fatigue ratings for both scales (Borg and SOFI) increased over time. The rate of fatigue development was greater for the second two hours of testing. However, following the 15-minute break perceived fatigue levels were reported to be at or near the initial baseline level. A similar trend was found for perceived workload ratings.

Indian participants reported higher levels of perceived fatigue and workload, had a larger ∆HR and excreted more saliva cortisol. These findings were more pronounced if the testing occurred in the afternoon, with the exception of saliva cortisol concentration which was observed to be relatively flat over time.

4.2 Effect of ethnicity and time

Table 4.1 describes the results from repeated measure analyses of variance to determine significant effects of ethnicity, time and their interaction. Ethnicity and time were found to significantly interact to affect Borg ( ) and

SOFI ratings ( ).

Table 4.1 ANOVA for Borg, SOFI, NASA and ∆HR

Dependent Variables Effect Num Den DF F Pr > F DF Value Borg Ethnicity 1 14 26.86 0.0001 Time 9 126 9.51 <0.0001 Ethnicity*Time 9 126 2.03 0.0412 SOFI Ethnicity 1 14 28.22 0.0001 Time 9 126 6.9 <0.0001 Ethnicity*Time 9 126 3.28 0.0013 NASA-TLX Ethnicity 1 14 13.68 0.0024 Time 9 126 61.33 <0.0001 Ethnicity *Time 9 126 1.04 0.4091 ∆HR Ethnicity 1 14 7.77 0.0145 Time 9 126 51.47 <0.0001 Ethnicity *Time 9 126 0.74 0.6692 CRT Ethnicity 1 10 4.03 0.0726 Time 3 30 5.08 0.0058 Ethnicity *Time 3 30 1.43 0.2539 NCRT Ethnicity 1 10 20.07 0.0012 Time 3 30 5.03 0.0061 Ethnicity *Time 3 30 2.77 0.0585 Eth 1 10 3.92 0.0759 AUCI Time 1 10 1.96 0.1919 Ethnicity*Time 1 10 0.17 0.6861 Ethnicity 1 10 6.69 0.0271 AUCG Time 1 10 3.13 0.1074 Ethnicity*Time 1 10 0.12 0.7385 Ethnicity 1 10 11.85 0.0063 MAUCI Time 1 10 4.94 0.0504 Ethnicity*Time 1 10 0.75 0.4057 Ethnicity 1 10 27.02 0.0004 MAUCG Time 1 10 7.71 0.0196 Ethnicity*Time 1 10 1.33 0.2751 Borg = one-dimensional fatigue scores measured in Borg scale, SOFI = multidimensional fatigue scores measured in Swedish Occupational Fatigue Inventory, NASA = workload measured in NASA-TLX, ∆HR = change in heart rate (bit per minute), CRT=saliva cortisol concentration, NCRT=Normalized saliva cortisol concentration by morning cortisol, AUCI= area under the curve with respect to increase for salivary cortisol, AUCG= area under the curve with respect to ground for salivary cortisol, MAUCI= area under the curve with respect to increase for normalized salivary cortisol, and MAUCG= area under the curve with respect to increase for normalized salivary cortisol.

Table 4.2 Tukey adjusted tests of effect slices for Borg and SOFI

Line Borg Ethnicity T Line SOFI Ethnicity T LSMEAN LSMEAN A 19.25 Indian 10 A 12.25 I 10 B 8.37 Westerners 10 B 4.00 W 10 A 17.62 Indian 9 A 10.87 I 9 B 8.25 Westerners 9 B 4.12 W 9 A 13.25 Indian 8 A 8.62 I 8 B 6.87 Westerners 8 B 2.62 W 8 A 11.00 Indian 5 A 7.62 I 7 B 4.37 Westerners 5 B 2.37 W 7 A 9.50 Indian 7 A 4.75 I 5 B 6.75 Westerners 7 B 2.62 W 5 D 7.25 Indian 4 D 3.62 I 4 D 4.12 Westerners 4 D 2.85 W 4 D 5.37 Indian 3 D 2.87 W 3 D 4.00 Westerners 3 D 2.87 I 3 D 4.37 Indian 2 D 2.37 W 2 D 3.37 Westerners 2 D 2.37 I 2 D 3.12 Indian 6 D 1.87 I 1 D 2.62 Westerners 6 D 1.37 W 1 D 2.87 Indian 1 D 1.62 I 6 D 1.75 Westerners 1 D 1.37 W 6

Figure 4.1 depicts the significant interaction effect of time and ethnicity on perceived fatigue. Indian participants reported significantly higher perceived fatigue ratings for both scales over time, with larger increases in fatigue occurring in the second two-hour period.

Tukey adjusted least square means were compared between groups for each data collection time point by applying “slice” statistical techniques in the “proc mixed” procedure in SAS (Table 4.2). Table 4.2 shows that Indian participants experienced significantly elevated perceived fatigue for both scales at data collection time points 5, 7,

8, 9 and 10. Similarly, slicing by ethnicity demonstrates that Indians perceived significantly higher rating of fatigue during the entire four-hour session.

Figure 4.1 Time and ethnicity significantly interact to affect fatigue in both scales.

Table 4.4 provides the post-hoc analysis for saliva cortisol concentration. In contrast to the raw cortisol (CRT, AUCI, and AUCG) normalized saliva cortisol concentrations by morning cortisol (NCRT, MAUCI, and MAUCG) were significantly affected by the ethnicity and time in most cases. Indian participants significantly experienced elevated saliva cortisol concentration.

Table 4.3 Tukey adjusted post-hoc for ∆HR and NASA

Effect Levels of LSMEANS Line Levels of the LSMEANS Line the Variable Variable CHR NASA T 3 15.19 A 9 28.75 A T 9 14.25 A 10 28.63 A T 7 14.06 A 8 26.06 AB T 2 13.56 A 7 26.00 AB T 8 13.56 A 5 24.63 AB T 5 13.00 A 4 24.31 AB T 4 12.94 A 3 22.69 BC T 10 12.56 A 2 21.88 C T 1 0.00 B 6 0.94 D T 6 0.00 B 1 0.69 D Ethnicity Indian 11.63 A Indian 22.04 A Ethnicity Westerner 10.20 B Westerner 18.88 B Means with same letters are insignificant.

Table 4.4 Tukey adjusted post-hoc for saliva cortisol

Effect Levels of the LSMEANS Line Levels of the LSMEANS Line variables variables NCRT AUCG Ethnicity I 0.93 A I 24.25 A W 0.31 B W 12.00 B NCRT CRT Time 1 1.08 A 1 0.18 A 10 0.53 B 10 0.06 B 6 0.49 B 6 0.06 B 5 0.39 B 5 0.05 B

MAUCG MAUCI Ethnicity I 185.38 A I 106.75 A W 58.75 B W 27.69 B Time 10 155.88 A 10 92.75 A 5 88.25 B 5 41.69 B CRT=saliva cortisol concentration, NCRT=Normalized saliva cortisol concentration by morning cortisol, AUCG= area under the curve with respect to ground for salivary cortisol, MAUCI= area under the curve with respect to increase for normalized. salivary cortisol, and MAUCG= area under the curve with respect to increase for normalized salivary cortisol. Means with same letters are insignificant.

Figure 4.2 Change in resting heart and NASA scores over time

Figure 4.3 Change in resting heart rate and NASA scores over ethnicity 73

Figure 4.3 illustrates the ∆HR and NASA scores over ethnicity. Both scores were found to be significantly higher for Indian participants as compared to westerners. In contrast to NASA and ∆HR, saliva cortisol concentrations were significantly higher at the beginning of the experiments (at the first assessment point) compared to the other three measurements (Table 4.4). Indian participants secreted significantly higher cortisol as compared to the Westerners (Figure 4.5). As expected, normalized saliva cortisol concentrations were observed to be more sensitive during the work-tasks for both raw and cleaned by standardized methods as described in section 3.3.3.2.2.

Figure 4.6 shows that both area under the curve with respect to increase (AUCI) and ground (AUCG) for normalized saliva cortisol concentration show similar significant results. As compared to Indian participants, cortisol was significantly lower for Western participants. Figure 4.7 depicts that both MAUCI and MAUCG were increased in the second-two-hour session as compared to the first one.

Figure 4.4 Change of raw and normalized saliva cortisol concentration over time

Figure 4.5 Change of raw and normalized saliva cortisol concentration over Ethnicity

Figure 4.6 Change in MAUCI and MAUCG by ethnicity

Figure 4.7 Change in MAUCI and MAUCG by ethnicity

4.3 Effect of task-independent and personal factors

In addition to ethnicity, which was the primary independent variable of interest, many other variables that affect fatigue were also considered. Table 4.5 shows the

ANOVA results for Borg and SOFI ratings. Bold types in Table 4.5 indicate significant results. As an illustration, working hours in primary occupation significantly affected

Borg , and SOFI ratings

Few interactions were found to be significant. Slicing by time for each assessment point demonstrates that the perceived fatigue ratings increase as TW increases. Similar effects were observed for EDF.

Detailed post-hoc analysis can be found in APPENDIX G. Generally the post-hoc results show that an increase in weekly working hours in a participant’s primary occupations, total weekly working hours across all occupations, and end of day fatigue

(EDF), resulted in increased perceived fatigue ratings. However, increases in daily sleep

(Sl), rest after work (DR), and weekly exercise frequency reduce perceived fatigue ratings.

Table 4.5 Effect of time and task-independent and personal variables on Borg and SOFI

Borg SOFI Effect Num DF Den DF F Value Pr > F F Value Pr > F Sl 1 13 16.56 0.0013 11.80 0.0044 T 9 117 12.09 <0.0001 7.89 <0.0001 Sl*T 9 117 1.05 0.4022 1.02 0.4319 W 2 13 53.43 <0.0001 8.60 0.0042 T 9 117 16.39 <0.0001 6.27 <0.0001 W*T 18 117 1.29 0.2095 0.27 0.9988 TW 2 13 72.68 <0.0001 11.80 0.0012 T 9 117 8.54 <0.0001 4.10 0.0001 TW*T 18 117 2.14 0.0083 1.03 0.4351 Ex 2 13 10.76 0.0018 0.69 0.5205 T 9 117 10.08 <0.0001 5.91 <0.0001 Ex*T 18 117 1.01 0.4537 0.21 0.9998 DR 1 14 19.80 0.0006 1.42 0.2540 T 9 126 9.46 <0.0001 5.68 <0.0001 DR*T 9 126 0.48 0.8870 0.14 0.9983 Sh 1 14 36.53 <0.0001 41.39 <0.0001 T 9 126 10.28 <0.0001 7.32 <0.0001 Sh*T 9 126 0.40 0.9338 0.93 0.5016 EDF 2 13 45.22 <0.0001 22.61 <0.0001 T 9 117 12.48 <0.0001 7.87 <0.0001 EDF*T 18 117 2.10 0.0099 1.53 0.0917 Note: Borg = one-dimensional fatigue scores measured in Borg scale, SOFI = multidimensional fatigue scores measured in Swedish Occupational Fatigue Inventory, T=experimental clock time, W=weekly work hours in primary occupation, TW=total work including primary and other occupations, EDF= end of the day fatigue, Ex= Exercise, DR=daily rest after work, Sh=morning or afternoon shift of the day. Bold faces indicate significant results.

4.4 Justification for the hypothesized hyperbolic relationship between time and a factor that causes fatigue

Based on the previous literature described in Figure 2.12 and section 2.3.3, it was hypothesized that time and fatigue load follows a hyperbolic functional relationship. To investigate this inherent relationship, stepwise regressions were performed using time,

fatigue load (workload, change in heart rate, and mean cortisol concentration level) and their interaction as predictors of fatigue. The stepwise regression analysis results are given in Table 4.6. Only the interaction term between time and a factor (e.g. workload) that causes fatigue was kept by the stepwise regression model in all six cases (Table 4.6).

Table 4.7 shows the parameter estimates and the associated statistical tests. All interaction terms were observed to be significant with p-value ranges between 0.02 and less than 0.0001.

Table 4.6 Summary of stepwise regression as a proof of the hyperbolic relationship

Step Variable Entered Variable Removed Model R2 F Value Pr > F

Borg 1 T*NASA 0.4284 118.40 <0.0001

Borg 1 T*∆HR 0.3880 100.16 <0.0001

Borg 1 T*CRT 0.2247 6.38 0.0193

SOFI 1 T*NASA 0.2634 56.50 <0.0001

SOFI 1 T*∆HR 0.3468 83.88 <0.0001

SOFI 1 T*CRT 0.2499 7.33 0.0129

T=Time in minute, NASA=perceived workload measured in NASA-TLX, ∆HR=change in resting heart rate, and CRT=normalized saliva cortisol concentration.

Table 4.7 Parameter estimates from the stepwise regression as a proof of the hyperbolic relationship

Parameter Standard Type II F Pr

Estimate Error SS Value > F

Borg Intercept 2.122320 0.622770 315.05 11.61 0.0000

T*NASA 0.001670 0.000154 3211.99 118.40 <0.0001

Borg Intercept 2.156460 0.660390 309.70 10.66 0.0000

T*∆HR 0.003270 0.000327 2909.16 100.16 <0.0001

Borg Intercept 4.800850 1.757420 301.90 7.46 0.0100

T*CRT 0.038750 0.015340 257.95 6.38 0.0200

SOFI Intercept 1.516660 0.468070 160.89 10.50 0.0000

T*NASA 0.000869 0.000116 865.86 56.50 <0.0001

SOFI Intercept 0.995270 0.451730 65.97 4.85 0.0300

T*∆HR 0.002050 0.000224 1139.89 83.88 <0.0001

SOFI Intercept 2.350820 0.853790 68.49 7.58 0.0100

T*CRT 0.000077 0.000028 66.22 7.33 0.0100

T=Time in minute, NASA=perceived workload measured in NASA-TLX, ∆HR=change in resting heart rate, and CRT=normalized saliva cortisol concentration.

In figure 4.8, the color gradient on the response surface represents the increase in fatigue from blue to red. The stronger contrast between the color of the fitted surface and an observation indicates higher residual at that point. An interesting result could be mentioned that the asymptotic relationship along the time coordinate is more obvious than along the Y-axis (a factor that affect fatigue), meaning that an individual will not

report fatigue if there is no workload, change in resting heart, or increase in saliva cortisol. However, someone will report fatigue within a fraction of time if the workload is too high for that person.

Figures 4.8 a and b depict the significant functional relationship between workload ratings and time on reported Borg Ratings (Figure 4.8a) and SOFI ratings

(Figure 4.8b). The contour lines represent the predicted responses by the model. The figure clearly indicates that workload and time significantly interact to affect perceived fatigue (i.e., higher fatigue ratings as a result of increasing workload and/and or time).

Similar trends were found for Borg and SOFI ratings as a function of HR and time

(Figure 4.8c and d respectively) and CRT and time (Figures 4.8 e and f), respectively.

(a) (b)

(b) (d)

(e) (f)

Figure 4.8 Contour plot for Borg and SOFI ratings as a function of : workload and time [(a) and (b)], ∆HR and time [(c) and (d)], and CRT and time [(e) and (f)

4.5 Correlation analysis

4.5.1 30-minute-block correlation

Table 4.8 shows the correlation analysis for four response variables, which were measured every 30-minute. All response variables were significantly correlated, though the strength of the correlation ranged from poor to strong. It is of interest to note that subjective rating correlations were moderate to strong amongst themselves, objective fatigue measures were in general moderate to strong amongst themselves, but subjective ratings with ∆HR in general were poor to moderate. Also, of note was that correlations between the response variables were observed to be stronger for the Indian participants than Westerners (Table 4.9 and Table 4.10).

As time was considered one of the primary variables of this study, correlation matrices were developed for each data collection time point (given in APPENDIX H).

The strength of the correlations between the response variables continued to increase with time and were highest at the end of each two-hour session, especially at the end of the second-two-hour session (APPENDIX H).

Table 4.8 Overall correlation matrix

Borg SOFI NASA ∆HR Borg 1.0000 0.7762 0.6398 0.3787 <0.0001 <0.0001 <0.0001 SOFI 1.0000 0.5151 0.3566 <0.0001 <0.0001 NASA 1.0000 0.5371 <0.0001 ∆HR 1.0000

Spearman Correlation Coefficients, N = 160; Prob > |r| under H0: Rho=0 Borg=Borg scale rating, SOFI=SOFI scale rating, NASA=NASA-TLX rating, ∆HR=change in resting heart rate.

Table 4.9 Correlation matrix for Indian participants

Borg SOFI NASA ∆HR Borg 1.0000 0.8119 0.6276 0.4510 <0.0001 <0.0001 <0.0001 SOFI 1.0000 0.4662 0.3729 <0.0001 0.0007 NASA 1.0000 0.4958 <0.0001 ∆HR 1.0000

Spearman Correlation Coefficients, N = 80; Prob > |r| under H0: Rho=0 Borg=Borg scale rating, SOFI=SOFI scale rating, NASA=NASA-TLX rating, ∆HR=change in resting heart rate.

Table 4.10 Correlation matrix for Western participants

Borg SOFI NASA ∆HR Borg 1.0000 0.6233 0.5621 0.2377 <0.0001 <0.0001 0.0337 SOFI 1.0000 0.3746 0.2642 0.0006 0.0179 NASA 1.0000 0.4940 <0.0001 ∆HR 1.0000

Spearman Correlation Coefficients, N = 80; Prob > |r| under H0: Rho=0 Borg=Borg scale rating, SOFI=SOFI scale rating, NASA=NASA-TLX rating, ∆HR=change in resting heart rate.

4.5.2 Two-hour-block correlations

Saliva cortisol samples were collected before and after each two-hour session, which produced four raw data points. No correlations were observed between fatigue and the raw cortisol secretion measures (CRT and NCRT). However, the standardized data cleaning process uses area under the curve with respect to increase (AUCI) and ground

(AUCG) generates only one data point at the end of each two-hour session. Table 4.11 shows the results from the two-hour-block correlation analysis using the data obtained

from the standardized data cleaning process mentioned in section 3.3.3.2.2. Bold-faced values in the Table 4.11 are considered as significant correlations. In general, Borg ratings were moderately correlated with AUCG measures. Perceived workload was observed to be correlated with AUCI and AUCG. Further, cortisol measures tended to be moderately correlated with each other.

Table 4.11 Correlation matrix for fatigue and saliva cortisol measures

Borg SOFI NASA ∆HR CRT NCRT AUCI AUCG MAUCI MAUCG

1.0000 0.7834 0.6142 0.3044 0.1758 0.1866 0.2959 0.4238 0.4400 0.3512 Borg <.0001 0.0014 0.1481 0.4111 0.3826 0.1603 0.0391 0.0314 0.0925 1.0000 0.4980 0.27681 0.2237 0.0451 0.3057 0.4559 0.3378 0.2636 SOFI 0.0133 0.1904 0.2934 0.8343 0.1464 0.0251 0.1065 0.2132 1.0000 0.1644 0.0369 0.0514 0.4697 0.5373 0.3203 0.2271 NASA 0.4427 0.8641 0.8115 0.0206 0.0068 0.1271 0.286 1.0000 0.2182 0.0962 -0.1382 0.0024 0.0033 -0.0160 ∆HR 0.3058 0.6547 0.5195 0.991 0.9877 0.941 1.0000 0.4940 0.0394 0.4055 0.2588 0.4474 CRT 0.0141 0.8551 0.0493 0.222 0.0284 1.0000 -0.3021 -0.0871 0.4075 0.7529 NCRT 0.1514 0.6856 0.0481 <.0001 1.0000 0.8756 0.6304 0.2769 AUC I <.0001 0.001 0.1903 1.0000 0.6438 0.4208 AUC G 0.0007 0.0406 1.0000 0.8758 MAUC I <.0001 1.0000 MAUC G Spearman Correlation Coefficients, N = 24; Prob > |r| under H0: Rho=0 CRT = raw saliva cortisol concentration, NCRT = raw saliva cortisol concentration weighted by early morning cortisol concentration, AUCI = area under the curve with respect to increase for salivary cortisol, AUCG = area under the curve with respect to ground for salivary cortisol, MAUCI = area under the curve with respect to increase for normalized salivary cortisol, and MAUCG = area under the curve with respect to increase for normalized salivary cortisol.

4.6 Fatigue predictive models

Stepwise regression was used to develop a fatigue predictive model for each two hour test block as well as the 15-minute rest period for Borg and SOFI fatigue ratings. A total of 10 variables and their interactions, resulting in a total of 21 variables were considered for inclusion. Table 4.12, Table 4.14 and Table 4.16 provide the detailed summary of the stepwise regression procedures, while Table 4.13, Table 4.15 and Table

4.17 provide the summary of parameter estimates and associated statistics for each block of time.

Predictions in Borg were observed to be moderate to strong with R2 values ranging between the minimum during the rest break (R2 value = 0.76) and maximum during the second-two-hour session (R2 value = 0.87). However, poor to moderate R2 values were observed for SOFI prediction. In contrast to the first-two-hour models, second-two-hour models had a larger number of interaction variables.

Table 4.12 Summary of stepwise regression for the first-two-hour session

Step Variable Variable Number Partial Model F Value Pr > F Entered Removed Vars In R2 R2 Borg 1 TW 1 0.3463 0.35 41.32 <0.0001 2 T*SH 2 0.2359 0.58 43.49 <0.0001 3 T*DR 3 0.0471 0.63 9.66 0.0027 4 T*EDF 4 0.0492 0.68 11.49 0.0011 5 Sh 5 0.0230 0.70 5.71 0.0194 6 Sl 6 0.0187 0.72 4.89 0.0301 7 W 7 0.0097 0.73 2.59 0.1119 8 DR 8 0.0210 0.75 5.98 0.0169 9 ETH 9 0.0288 0.78 9.16 0.0035 SOFI 1 Sh 1 0.2936 0.29 32.43 <0.0001 2 T*EDF 2 0.1397 0.43 18.99 <0.0001 3 DR 3 0.0474 0.48 6.94 0.0102 4 Sl 4 0.0779 0.56 13.24 0.0005 5 TW 5 0.0954 0.65 20.40 <0.0001 6 W 6 0.0292 0.68 6.74 0.0114 7 ETH 7 0.0324 0.72 8.21 0.0054 8 EDF 8 0.0299 0.75 8.35 0.0051 9 T*TW 9 0.0084 0.75 2.39 0.1265 10 T*EDF 8 0.0002 0.75 0.05 0.8217 11 T*W 9 0.0206 0.77 6.39 0.0137 T = running clock during the experiment (min), ETH = ethnicity (0 for Westerners and 1 for Indians), Sl = hours of daily sleep, W = weekly working hours in primary occupation in hours, TW = total weekly working hours in all occupations, EDF = perceived fatigue at the end of a regular working day, Ex = weekly exercise frequency, DR = daily rests in hours after work, Sh = working shift (1 for morning and 2 for afternoon), ∆HR = change in resting heart rate, NASA = workload measured in NASA-TLX.

Table 4.13 Parameter estimates of stepwise regression for the first-two-hour session

Variable Parameter Standard Type II SS F Value Pr > F Estimate Error Borg Intercept -10.4171 5.6034 19.89 3.46 0.0672 ETH -2.5916 0.8563 52.73 9.16 0.0035 Sl 1.2935 0.5005 38.44 6.68 0.0118 W -0.5077 0.1237 97.00 16.85 0.0001 TW 0.5212 0.0951 172.76 30.02 <0.0001 DR -2.9559 0.8095 76.74 13.33 0.0005 Sh 3.8928 0.9283 101.22 17.59 <0.0001 T*EDF 0.0168 0.0030 176.31 30.63 <0.0001 T*DR -0.0181 0.0065 43.83 7.62 0.0074 T*SH 0.0167 0.0104 15.01 2.61 0.1108 SOFI Intercept -15.5097 3.0363 44.09 26.09 <0.0001 ETH -1.8537 0.4772 25.50 15.09 0.0002 Sl 2.0820 0.2763 95.97 56.79 <0.0001 W -0.2292 0.0878 11.50 6.81 0.0111 TW 0.2022 0.0673 15.24 9.02 0.0037 EDF 0.8076 0.1719 37.28 22.06 <0.0001 DR -2.6298 0.4084 70.08 41.47 <0.0001 Sh 3.7351 0.3698 172.43 102.04 <0.0001 T*W -0.0022 0.0009 10.80 6.39 0.0137 T*TW 0.0021 0.0007 14.72 8.71 0.0043 T = running clock during the experiment (min), ETH = ethnicity (0 for Westerners and 1 for Indians), Sl = hours of daily sleep, W = weekly working hours in primary occupation in hours, TW = total weekly working hours in all occupations, EDF = perceived fatigue at the end of a regular working day, Ex = weekly exercise frequency, DR = daily rests in hours after work, Sh = working shift (1 for morning and 2 for afternoon), ∆HR = change in resting heart rate, NASA = workload measured in NASA-TLX.

Table 4.14 Summary of Stepwise Regression for the rest-break session

Step Variable Variable Number Partial Model F Value Pr > F Entered Removed Vars In R2 R2 Borg 1 TW 1 0.3185 0.32 14.02 0.0008 2 T*∆HR 2 0.2700 0.59 19.02 0.0001 3 Sh 3 0.0923 0.68 8.09 0.0082 4 ∆HR 4 0.0520 0.73 5.25 0.0300 5 DR 5 0.0291 0.76 3.17 0.0865 SOFI 1 Sh 1 0.2742 0.27 11.34 0.0021 2 T*SH 2 0.1557 0.43 7.92 0.0087 3 EDF 3 0.0861 0.52 4.98 0.0338 4 Sl 4 0.0469 0.56 2.90 0.1003 T = running clock during the experiment (min), Sl = hours of daily sleep, TW = total weekly working hours in all occupations, EDF = perceived fatigue at the end of a regular working day, DR = daily rests in hours after work, Sh = working shift (1 for morning and 2 for afternoon), ∆HR = change in resting heart rate.

Table 4.15 Parameter estimates of stepwise regression the rest-break session

Variable Parameter Standard Type II SS F Value Pr > F Estimate Error Borg Intercept -7.4424 4.5811 22.84 2.64 0.1163 TW 0.1719 0.0561 81.40 9.41 0.0050 DR -1.6373 0.9190 27.47 3.17 0.0865 Sh 2.9544 1.1056 61.79 7.14 0.0128 ∆HR -14.1586 5.9717 48.64 5.62 0.0254 T*∆HR 0.1207 0.0495 51.37 5.94 0.0220 SOFI Intercept -13.6583 5.8198 22.44 5.51 0.0265 Sl 1.0630 0.6247 11.80 2.90 0.1003 EDF 0.8968 0.3130 33.46 8.21 0.0080 Sh 15.3482 3.9111 62.75 15.40 0.0005 T*SH -0.0933 0.0301 39.20 9.62 0.0045 T = running clock during the experiment (min), Sl = hours of daily sleep, TW = total weekly working hours in all occupations, EDF = perceived fatigue at the end of a regular working day, DR = daily rests in hours after work, Sh = working shift (1 for morning and 2 for afternoon), ∆HR = change in resting heart rate.

Table 4.16 Summary of Stepwise Regression for the second-two-hour session

Step Variable Variable Number Partial Model F Value Pr > F Entered Removed Vars In R2 R2 Borg 1 T*TW 1 0.5627 0.56 100.35 <0.0001 2 T*EX 2 0.1321 0.69 33.31 <0.0001 3 T*SH 3 0.0519 0.75 15.58 0.0002 4 T*ETH 4 0.0308 0.78 10.38 0.0019 5 ETH 5 0.0192 0.80 6.99 0.0100 6 T*DR 6 0.0122 0.81 4.65 0.0343 7 ∆HR 7 0.0215 0.83 9.11 0.0035 8 T*NASA 8 0.0107 0.84 4.79 0.0319 9 T*EDF 9 0.0083 0.85 3.88 0.0529 10 EDF 10 0.0074 0.86 3.55 0.0639 11 T*W 11 0.0097 0.87 4.94 0.0295 SOFI 1 T*ETH 1 0.3454 0.35 41.15 <0.0001 2 T*SH 2 0.2498 0.60 47.50 <0.0001 3 ∆HR 3 0.0497 0.64 10.64 0.0017 4 T*TW 4 0.0168 0.66 3.71 0.0577 5 T*W 5 0.0198 0.68 4.60 0.0352 6 ETH 6 0.0105 0.69 2.50 0.1183 7 Sl 7 0.0132 0.71 3.23 0.0766 T = running clock during the experiment (min), ETH = ethnicity (0 for Westerners and 1 for Indians), Sl = hours of daily sleep, W = weekly working hours in primary occupation in hours, TW = total weekly working hours in all occupations, EDF = perceived fatigue at the end of a regular working day, Ex = weekly exercise frequency, DR = daily rests in hours after work, Sh = working shift (1 for morning and 2 for afternoon), ∆HR = change in resting heart rate, NASA = workload measured in NASA-TLX.

Table 4.17 Parameter estimates of stepwise regression for the second-two-hour session

Variable Parameter Standard Type II SS F Value Pr > F Estimate Error Borg Intercept 4.0300 3.1471 15.39 1.64 0.2047 ETH -9.1319 3.6843 57.67 6.14 0.0157 T*ETH 0.0458 0.0190 54.30 5.78 0.0189 EDF -2.9658 1.1685 60.47 6.44 0.0134 T*W -0.0017 0.0008 46.42 4.94 0.0295 T*TW 0.0022 0.0006 117.96 12.57 0.0007 T*EDF 0.0207 0.0064 98.85 10.53 0.0018 T*EX -0.0068 0.0015 198.09 21.10 <0.0001 T*DR -0.0205 0.0043 208.44 22.20 <0.0001 T*SH 0.0207 0.0046 191.64 20.41 <0.0001 ∆HR 0.3967 0.0881 190.24 20.27 <0.0001 T*NASA -0.0010 0.0004 68.61 7.31 0.0087 SOFI Intercept 7.6699 5.4673 19.76 1.97 0.1650 ETH -6.4717 3.0760 44.44 4.43 0.0389 T*ETH 0.0660 0.0161 169.43 16.88 0.0001 Sl -1.0700 0.5955 32.42 3.23 0.0766 T*W 0.0011 0.0005 42.52 4.23 0.0432 T*TW -0.0014 0.0005 98.10 9.77 0.0026 T*SH 0.0201 0.0039 263.88 26.28 <0.0001 ∆HR 0.2479 0.0651 145.76 14.52 0.0003 T = running clock during the experiment (min), ETH = ethnicity (0 for Westerners and 1 for Indians), Sl = hours of daily sleep, W = weekly working hours in primary occupation in hours, TW = total weekly working hours in all occupations, EDF = perceived fatigue at the end of a regular working day, Ex = weekly exercise frequency, DR = daily rests in hours after work, Sh = working shift (1 for morning and 2 for afternoon), ∆HR = change in resting heart rate, NASA = workload measured in NASA-TLX.

CHAPTER V

DISCUSSION

The objective of this dissertation was to quantify human fatigue during prolonged mentally demanding work-tasks. An observational study in the field was conducted involving study participants performing either computer programming or simulation in the workplace on a daily basis and the study participants were assessed over a 4 hour session in the morning or the afternoon.

5.1 Ethnicity

ANOVA results demonstrated that all response variables (Borg, SOFI, ∆HR, workload and NCRT) were affected by ethnicity and time. As expected, fatigue increased over time for both populations. More interestingly, ethnicity significantly interacted with time to affect fatigue with Indian participants experiencing fatigue faster than Westerners.

Generally, ethnic minority groups report significantly higher fatigue because of socioeconomic status, unemployment, and being classified as minority (R. R. Taylor, et al., 2003). When these factors are adjusted for, either no difference (Buchwald, et al.,

1996; Yennurajalingam, et al., 2008) or less (Cordero, et al., 2012) fatigue was reported by the ethnic minority groups. Being international students, most Indian participants of this study may face extra challenges, for instance, VISA issues, legal status issues, cultural shock, language, maintaining an outstanding academic record, and homesickness.

These additional challenges perhaps explain why Indians experienced increased fatigue rates for participants in the study.

This study observed that both groups had to sustain a constant production of cortisol over time, which is usually decreased over the course of the day if humans are not conditioned by some demand, for example, prolonged cognitively demanding work- tasks (Bohnen, et al., 1990; Engelmann, et al., 2011; Weitzman, et al., 1971). The trend of cortisol concentration observed in this study cannot be considered as following the normal diurnal patter depicted in Figure 2.4, which indicates elevated stress (H. E. Webb, et al., 2011) and workload (Anna Dahlgren, et al., 2005) resulting in fatigue over time corresponding to the previous studies (Adam, et al., 2006; Chida & Steptoe, 2009; Rubin,

Hotopf, Papadopoulos, & Cleare, 2005b). Moreover, significantly higher cortisol hormone secretion was measured for those participants who reported higher fatigue.

The area under the curve with respect to ground, which has been considered as a measure of fatigue in previous literature, was significantly correlated with Borg and

SOFI. Nevertheless, the poor correlations between cortisol and fatigue (Borg = 0.42,

SOFI=0.46) indicate that the cortisol hormone does not explain fatigue completely.

Similar findings were observed for the other two measures (workload and change in resting heart rate). These findings indicate that task-dependent variables were not enough to explain the cause of fatigue because of the complex nature of fatigue.

This study also found that lack of sleep, exercise, rest-breaks and daily rest after work were significant factors that affected fatigue negatively. This implies that fatigue accumulated over the previous workdays carried over to the next day (Torbjörn

Åkerstedt, 2003; T. Åkerstedt, et al., 2004; Torbjörn Åkerstedt & Wright Jr, 2009;

Östberg, 1973). This study also observed higher workload during the day and/or week significantly and negatively affected fatigue, which are task-dependent variables considered to be primary causes of fatigue in the workplace (Torbjörn Åkerstedt, et al.,

2002; Dorrian, et al., 2011). In addition to the higher workload, if there is an opportunity to recover (e.g. rest-break, daily rest), the development of fatigue become even worse

(Torbjörn Åkerstedt, 2003; Torbjörn Åkerstedt, et al., 2002; T. Åkerstedt, et al., 2004;

Torbjörn Åkerstedt & Wright Jr, 2009; Eo Grandjean, 1968).

5.2 Effect of time

As expected, this study observed that fatigue increased over time. Interestingly, in the regression analysis process, time significantly interacted with most factors in the study, indicating that the time trends in the fatigue measures are dependent on other factors. Similar to localized muscle fatigue curves (Rohmert’s curves) (Hill, et al., 2002;

Monod & Scherrer, 1965), previous studies have hypothesized that there could be a possible hyperbolic relationship between time and fatigue load (El Falou, et al., 2003;

Jensen, 2003; Østensvik, et al., 2009). For example, the standardized data cleaning process for repeated measure cortisol hormone using the area under the curve over time

(J. C. Pruessner, et al., 2003), which is basically the interaction between cortisol response and time to explain total fatigue. This study has utilized three factors, including (1) change in resting heart rate, (2) workload measured by NASA-TLX, and (3) saliva cortisol concentration separately with time to determine the hyperbolic relationship between fatigue load and time to explain total fatigue. In all three cases, the only term left in the stepwise regression process was the interaction term, which could justify the possible hypothesized hyperbolic relationship between a fatigue load and time to explain 94

total fatigue. The estimated parameters were highly significant with R2 values ranging from 0.22 to 0.43. For a single variable to explain human fatigue, R2 values could be considered good because of the complex nature of fatigue caused by many intrinsic and extrinsic factors (Di Milia, et al., 2011).

In this study, more interestingly, the asymptotes parallel to both axes indicate that an individual would not report fatigue for a long period of time if there were no fatigue load. In contrast, an individual would report fatigue within a very short period of time if the fatigue load were too high for that person. The observed hyperbolic relationship could be elaborated on as “the higher intensity of a fatigue load induces fatigue quicker and vice versa.” The relationship is depicted in Figure 2.12.

5.3 Prediction of fatigue

In general, Borg ratings were predicted better when compared to the multidimensional fatigue score (SOFI) across all models, meaning that Borg is catching more information than SOFI to measure fatigue. The Borg scale was used for seven different body parts, including the whole body, and the scale was measuring overall fatigue for each of those body parts. In contrast to the Borg, SOFI scale does not directly measure fatigue; rather it solicits perceived responses to its five dimensions, including physical discomfort, physical exertion, sleepiness, lack of energy and lack of motivation.

These five dimensions were hypothesized to be the cause of total fatigue, which may partially be true for the work-tasks studied in this dissertation.

Although perceived fatigue increased over time, surprisingly no models kept time as a main effect; rather, a number of variables that interacted with time were retained in the models (especially for the second-two-hour model). Also, the second-two-hour 95

models kept more task-dependent variables such as workload and change in resting heart rate and their interactions with time. In contrast to the second-two-hour models, the first- two-hour models only utilized the task-independent and personal factors with minimum interaction terms. Therefore, time interactions were more important in predicting fatigue in the second two-hour session. This implies that at the onset of work, task-independent and personal factors (such as lack of sleep) are the primary drivers of fatigue, but as the workday continues, work related factors (such as changes in workload or work tasks) are the primary drivers of fatigue. This relationship is further explained by the hyperbolic relationship found between time and fatigue loading, which indicates that later in the day, smaller changes in work related factors have a larger impact on the amount of fatigue experienced by workers.

CHAPTER VI

LIMITATIONS AND FUTURE STUDIES

6.1 Statistical power and sample size

As power is defined by the strength of a statistical test to reject a false null hypothesis, false claim could be made by the test if power is poor. An excellent post-hoc power of the study was observed for most variables in this dissertation. However, the limited number of sample size in each category of the task-independent and personal factors did not allow testing in the ANOVA for a wider classification of these variables.

As the shorter categorizations of the task-independent and personal factors were observed to be significant, each of these task-independent and personal variables with larger numbers of categories could be studied in the future.

6.2 Study protocol

Although this dissertation was an observational study that performs the testing in the field, this study followed the break schedules that are practiced in industrial and service sectors today. Many participants of this study reported that they usually would take a break whenever they needed because the nature of their jobs (research jobs in programming or simulation at a University) allows them to do so. As expected, the 15- minute break in the middle of the four-hour session helped to reduce perceived fatigue, workload and change in resting heart rate significantly. However, a slight upward trend

of saliva cortisol (not significant though) was also observed during the requested 15- minute break, which was introduced by the researcher. Perhaps, this requested 15-minute break interrupted their regular break schedules resulting in unnecessary stress, which could be the reason for the slightly higher salvia cortisol concentrations. Cortisol concentration was significantly higher to start the experiment as compared to the other three measurements taken during the experiment. People feel uncomfortable and stressed if they are being watched, which could explain the significant increase in cortisol production at the beginning of this observational study. However, the deviation from the typical diurnal cortisol hormone during the study demonstrates that participants were affected by the task as well.

6.3 Subjective instruments

In the beginning of the experiment, most study participants reported many “No

Fatigue” ratings. Either participants really had nothing to report, or they reported no fatigue because they thought this was the appropriate response. Regardless, perhaps acclimating participants to the scales may be warranted to ensure correct response collection.

Most participants reported nearly identical workload and fatigue levels at the start of the second 2-hour session (following the break) as they did at the start of testing, as has been reported in previous studies (Babski-Reeves K., et al., 2000; Dababneh, et al.,

2001; W. C. Taylor, et al., 2013). Further, reduced fatigue and workload has been associated with improved worker productivity and employee well-being (Dababneh, et al., 2001; Galinsky, et al., 2000; Henning, et al., 1997). In this study, a forced 15-minute break following 2 hours of work was used, when in actuality, these participants indicated 98

that they typically took breaks as needed. Regardless of the timing of breaks, it appears that workers will benefit from reduced fatigue with the introduction of breaks. Future studies should focus on identifying optimal or differences in the effectiveness of timed vs random breaks.

Although the subjective measures were more associated with the work-tasks, the physiological responses may not be solely associated with the work-tasks, and could have been influenced by a number of outside factors. This could explain why the correlations between the objective and subjective measure were poor to moderate. Debate on the gold standard (subjective vs objective) fatigue measures continues, though cortisol levels appear to have promise and were relatively well correlated with subjective perceptions of fatigue.

6.4 Effect of ethnicity

This study found significant differences between Indian and Western populations.

Indian participants in the study were all international students. Being an international student, the study participant may face some additional challenges that the westerners who did not face. Control studies in the future addressing this issue may find interesting results.

6.5 Effect of time

The new finding of this study suggests that the interaction between a factor that causes fatigue (fatigue load) and time significantly affects fatigue, in addition to the apparent increase in fatigue over time. Future studies should focus on these interactions not just the time effect alone, because the main effect becomes irrelevant when a

significant interaction is presented. This study observed that as the time increased, especially in the second-two-hour session, time started to play a major role in explaining fatigue. Therefore, the duration of the study could be an important variable in assessing fatigue in the workplace.

100

CHAPTER VII

CONCLUSION

The objective of this study was to evaluate human fatigue in prolonged mentally demanding work-tasks, for example, computer programming and simulation. An observational study in the field was performed to determine the effect of mentally demanding work-tasks on fatigue. Based on the 16 participants, a few conclusions are summarized below:

1. As compared to Westerners, Indian participants experienced significantly

higher ∆HR, workload, salivary cortisol concentration and perceived fatigue

ratings. This conclusion must be interpreted carefully. For example, all Indian

participants of the study were international students who constantly face many

issues, including cultural, being a non-resident alien in a foreign country, food

habit, etc. that may explain the significant increase in all measures.

2. As expected, fatigue increases over time. More interestingly, the interaction

between time and fatigue load (a factor that causes fatigue) was observed to

be significant and to maintain a reasonably hyperbolic relationship. This

relationship demonstrates that people would experience peak fatigue faster if

the intensity of a factor that causes fatigue increases and vice versa.

3. Workload significantly affects the perception of fatigue measured in both one-

dimensional and multidimensional scales. Participants who rated the workload 101

higher during the experiment also perceived elevated fatigue. Moreover,

perceived fatigue in both scales was significantly increased for participants

who had higher weekly working hours in their primary occupations and higher

total weekly working hours in all occupations.

4. Inability to recover from fatigue is still proven to be the substantial cause of

fatigue accumulation. For example, this study observed that daily sleep, rest

on weekends, rest-breaks during work, and daily rest after work significantly

affected the perception of fatigue measured in one-dimensional Borg and

multi-dimensional SOFI scales.

5. Participants without any weekly exercise reported significantly elevated

fatigue. Regular physical exercise could be an important remedy to recover

from fatigue accumulated due to these low physical/high mental work-tasks.

6. Generally, poor correlations between subjective and objective measures of

fatigue suggest using subjective instruments over objective measures if a good

predictive model comprising objective measures, task-independent and

personal information, is not utilized. However, moderate and significant

correlations were observed between perceived fatigue and the interaction of

time and objective fatigue load (e.g. change in resting heart rate). The strength

of the correlation was observed better for Indians, as compared to Westerners.

The poor correlation also indicated that a single factor is not enough to

explain the causes of the complex nature of fatigue.

102

7. Fatigue predictive models indicated that the interactions between time and

other variables that cause fatigue are more prominent than their main effects.

Therefore, duration of study is very important in fatigue research.

103

REFERENCES

Aaronson, L. S., Teel, C. S., Cassmeyer, V., Neuberger, G. B., Pallikkathayil, L., Pierce, J., . . . Wingate, A. (1999). Defining and measuring fatigue. Journal of Nursing Scholarship, 31(1), 45-50.

Ackerman, P. L., & Kanfer, R. (2009). Test length and cognitive fatigue: an empirical examination of effects on performance and test-taker reactions. J Exp Psychol Appl, 15(2), 163-181. doi: 10.1037/a0015719 2009-09189-006 [pii]

Ada, W. S. L., Chetwyn, C. H. C., & Jufang, H. (2004). Structural stability and reliability of the Swedish occupational fatigue inventory among Chinese VDT workers. [Article]. Applied Ergonomics, 35, 233-241. doi: 10.1016/j.apergo.2004.02.004

Adam, E. K., Hawkley, L. C., Kudielka, B. M., & Cacioppo, J. T. (2006). Day-to-day dynamics of experience--cortisol associations in a population-based sample of older adults. Proc Natl Acad Sci U S A, 103(45), 17058-17063. doi: 0605053103 [pii] 10.1073/pnas.0605053103

Ahmed, S. (2010). Effect of backrest angle on operator discomfort, from http://library.msstate.edu/etd/show.asp?etd=etd-04052010-014822

Aho, K. A. (2007). Acceleration and Time Pathologies. Time & Society, 16(1), 25-42. doi: 10.1177/0961463x07074100

Ahsan, H. K., Herbert, F. J., Toshio, M., & Marimuthu, P. (2010). Association of cardiac autonomic neuropathy with alteration of sympatho-vagal balance through heart rate variability analysis. [Article]. Medical Engineering and Physics, 32, 161-167. doi: 10.1016/j.medengphy.2009.11.005

Åhsberg, E., & Gamberale, F. (1998). Perceived fatigue during physical work: an experimental evaluation of a fatigue inventory. International Journal of Industrial Ergonomics, 21(2), 117-131. doi: 10.1016/s0169-8141(96)00071-6

Åhsberg, E., Gamberale, F., & Gustafsson, K. (2000). Perceived fatigue after mental work: An experimental evaluation of a fatigue inventory. [print]. Ergonomics, 43(2), 252-268.

104

Åhsberg, E., Garnberale, F., & Kjellberg, A. (1997). Perceived quality of fatigue during different occupational tasks Development of a questionnaire. International Journal of Industrial Ergonomics, 20(2), 121-135. doi: 10.1016/s0169- 8141(96)00044-3

Åkerstedt, T. (2003). Shift work and disturbed sleep/wakefulness. Occupational Medicine, 53(2), 89-94. doi: 10.1093/occmed/kqg046

Åkerstedt, T., Fredlund, P., Gillberg, M., & Jansson, B. (2002). Work load and work hours in relation to disturbed sleep and fatigue in a large representative sample. Journal of psychosomatic research, 53(1), 585-588. doi: http://dx.doi.org/10.1016/S0022-3999(02)00447-6

Åkerstedt, T., Knutsson, A., Westerholm, P., Theorell, T., Alfredsson, L., & Kecklund, G. (2004). Mental fatigue, work and sleep. Journal of psychosomatic research, 57(5), 427-433. doi: http://dx.doi.org/10.1016/j.jpsychores.2003.12.001

Åkerstedt, T., & Wright Jr, K. P. (2009). Sleep loss and fatigue in shift work and shift work disorder. Sleep medicine clinics, 4(2), 257.

Alison, D. B., Jill, D., & Adam, F. (2011). Some Vocal Consequences of Sleep Deprivation and the Possibility of “Fatigue Proofing” the Voice With Voicecraft® Voice Training. [Article]. Journal of Voice, 25, 447-461. doi: 10.1016/j.jvoice.2010.10.020

Areeudomwong, P., Puntumetakul, R., Kaber, D. B., Wanpen, S., Leelayuwat, N., & Chatchawan, U. (2012). Effects of handicraft sitting postures on lower trunk muscle fatigue. Ergonomics, 55(6), 693-703. doi: 10.1080/00140139.2012.658086

Arlinghaus, A., Lombardi, D. A., Courtney, T. K., Christiani, D. C., Folkard, S., & Perry, M. J. (2012). The effect of rest breaks on time to injury - a study on work-related ladder-fall injuries in the United States. Scand J Work Environ Health, 38(6), 560-567. doi: 10.5271/sjweh.3292 3292 [pii]

Babski-Reeves, K., & Crumpton, L. L. (1999). An Evaluation of Total Body Fatigue. Paper presented at the Proceedings of the Eighth Annual Institute of Industrial Engineering Research Conference.

Babski-Reeves, K., Stanfield, J., & Hughes, L. (2005). Assessment of video display workstation set up on risk factors associated with the development of low back and neck discomfort. International Journal of Industrial Ergonomics, 35(7), 593- 604. doi: 10.1016/j.ergon.2005.01.006

105

Babski-Reeves K., Crumpton-Young L., Riley J., Nitcavic L., & Gentry H. (2000). An Investigation of the Relationship between Time Stress and Fatigue. Paper presented at the Proceedings of the Ninth Annual Institute of Industrial Engineering Research Conference.

Bakker, A. B., Demerouti, E., Oerlemans, W., & Sonnentag, S. (2013). Workaholism and daily recovery: A day reconstruction study of leisure activities. Journal of Organizational Behavior, 34(1), 87-107.

Balci, R., & Aghazadeh, F. (2003). The effect of work-rest schedules and type of task on the discomfort and performance of VDT users. Ergonomics, 46(5), 455-465. doi: 10.1080/0014013021000047557

Bansevicius, D., Westgaard, R. H., & Jensen, C. (1997). Mental Stress of Long Duration: EMG Activity, Perceived Tension, Fatigue, and Pain Development in Pain-Free Subjects. Headache: The Journal of Head and Face Pain, 37(8), 499-510. doi: 10.1046/j.1526-4610.1997.3708499.x

Barker, L. M. (2009). Measuring and modeling the effects of fatigue on performance specific application to the nursing profession, from http://scholar.lib.vt.edu/theses/available/etd-07232009-101727/

Barker, L. M., & Nussbaum, M. A. (2011). The effects of fatigue on performance in simulated nursing work. Ergonomics, 54(9), 815-829. doi: 10.1080/00140139.2011.597878

Barker Steege, L. M., & Nussbaum, M. A. (2013). Dimensions of fatigue as predictors of performance: A structural equation modeling approach among registered nurses. IIE Transactions on Occupational Ergonomics and Human Factors, 1(1), 16-30.

Barofsky, I., & Legro, M. W. (1991). Definition and measurement of fatigue. Rev Infect Dis, 13 Suppl 1, S94-97.

Bartley, S. H., & Chute, E. (1947). Fatigue and impairment in man. New York: McGraw- Hill Book Co.

Bazelmans, E., Bleijenberg, G., Vercoulen, J. H. M. M., van der Meer, J. W. M., & Folgering, H. (1997). The chronic fatigue syndrome and hyperventilation. Journal of psychosomatic research, 43(4), 371-377. doi: 10.1016/s0022-3999(97)00169-4

Belza, B. L. (1995). Comparison of self-reported fatigue in rheumatoid arthritis and controls. J Rheumatol, 22(4), 639-643.

Berrios, G. E. (1990). Feelings of fatigue and psychopathology: a conceptual history. Compr Psychiatry, 31(2), 140-151. doi: 0010-440X(90)90018-N [pii]

106

Beynon, C., Burke, J., Doran, D., & Nevill, A. (2000). Effects of activity-rest schedules on physiological strain and spinal load in hospital-based porters. Ergonomics, 43(10), 1763-1770. doi: 10.1080/001401300750004168

Bhui, K. S., Dinos, S., Morelli, M.-L., Khoshaba, B., Nazroo, J., Wessely, S., & White, P. D. (2011). Ethnicity and Fatigue: Expressions of Distress, Causal Attributions and Coping. Sociology, 1(4), 156-163.

Billaut, F., & Bishop, D. J. (2012). Mechanical work accounts for sex differences in fatigue during repeated sprints. European journal of applied physiology, 112(4), 1429-1436.

Bilodeau, M., Henderson, T. K., Nolta, B. E., Pursley, P. J., & Sandfort, G. L. (2001). Effect of aging on fatigue characteristics of elbow flexor muscles during sustained submaximal contraction. Journal of Applied Physiology, 91(6), 2654-2664.

Binnewies, C., Sonnentag, S., & Mojza, E. J. (2009). Daily performance at work: feeling recovered in the morning as a predictor of day‐level job performance. Journal of Organizational Behavior, 30(1), 67-93.

Blain, G., Meste, O., Blain, A., & Bermon, S. (2009). Time-frequency analysis of heart rate variability reveals cardiolocomotor coupling during dynamic cycling exercise in humans. American Journal of Physiology - Heart and Circulatory Physiology, 296(5), H1651-H1659. doi: 10.1152/ajpheart.00881.2008

Blangsted, A. K., Sjøgaard, G., Madeleine, P., Olsen, H. B., & Søgaard, K. (2005). Voluntary low-force contraction elicits prolonged low-frequency fatigue and changes in surface electromyography and mechanomyography. Journal of Electromyography and Kinesiology, 15(2), 138-148. doi: http://dx.doi.org/10.1016/j.jelekin.2004.10.004

Bohnen, N., Houx, P., Nicolson, N., & Jolles, J. (1990). Cortisol reactivity and cognitive performance in a continuous mental task paradigm. Biol Psychol, 31(2), 107-116. doi: 0301-0511(90)90011-K [pii]

Boksem, M. A., Meijman, T. F., & Lorist, M. M. (2006). Mental fatigue, motivation and action monitoring. Biological Psychology, 72(2), 123-132.

Boksem, M. A., & Tops, M. (2008). Mental fatigue: costs and benefits. Brain Res Rev, 59(1), 125-139. doi: S0165-0173(08)00071-4 [pii] 10.1016/j.brainresrev.2008.07.001

Boneva, R. S., Decker, M. J., Maloney, E. M., Lin, J.-M., Jones, J. F., Helgason, H. G., . . . Reeves, W. C. (2007). Higher heart rate and reduced heart rate variability persist during sleep in chronic fatigue syndrome: A population-based study. Autonomic Neuroscience, 137(1–2), 94-101. doi: http://dx.doi.org/10.1016/j.autneu.2007.08.002 107

Borg, G. (1970a). Perceived exertion as an indicator of somatic stress. Scandinavian Journal Of Rehabilitation Medicine, 2(2), 92-98.

Borg, G. (1970b). Perceived exertion as an indicator of somatic stress. Scand J Rehabil Med, 2(2), 92-98.

Borg, G. (1982). Ratings of perceived exertion and heart rates during short-term cycle exercise and their use in a new cycling strength test. Int J Sports Med, 3(3), 153- 158. doi: 10.1055/s-2008-1026080

Borg, G. (1990). Psychophysical scaling with applications in physical work and the perception of exertion. Scand J Work Environ Health, 16 Suppl 1, 55-58.

Borg, G. (1998). Borg's Perceived exertion and pain scales. Champaign, IL: Human Kinetics.

Borg, G. A. (1982). Psychophysical bases of perceived exertion. Med Sci Sports Exerc, 14(5), 377-381.

Borg, G. B. E. (1987). The Need of an additional constant in the psychophysical function for taste perception. Stockholm, Sweden: University of Stockholm, Dept. of Psychology.

Bray, G. A. (2004). Medical consequences of obesity. J Clin Endocrinol Metab, 89(6), 2583-2589. doi: 10.1210/jc.2004-0535 89/6/2583 [pii]

Bricout, V. A., Dechenaud, S., & Favre-Juvin, A. (2010). Analyses of heart rate variability in young soccer players: the effects of sport activity. Auton Neurosci, 154(1-2), 112-116. doi: S1566-0702(09)00574-8 [pii] 10.1016/j.autneu.2009.12.001

Brouha, L., & Ball, V. (1948). Research on industrial fatigue. Rev Can Biol, 7(3), 479- 483.

Brown, I. D. (1994). Driver Fatigue. Human Factors: The Journal of the Human Factors and Ergonomics Society, 36(2), 298-314. doi: 10.1177/001872089403600210

Buchwald, D., Manson, S. M., Pearlman, T., Umali, J., & Kith, P. (1996). Race and Ethnicity in Patients with Chronic Fatigue. Journal of Chronic Fatigue Syndrome, 2(1), 53-66. doi: doi:10.1300/J092v02n01_05

Carpenito-Moyet, L. J. (2006). Nursing diagnosis : application to clinical practice. Philadelphia: Lippincott Williams & Wilkins.

Causse, M., Sénard, J.-M., Démonet, J. F., & Pastor, J. (2010). Monitoring Cognitive and Emotional Processes Through Pupil and Cardiac Response During Dynamic Versus Logical Task. Applied Psychophysiology & Biofeedback, 35(2), 115-123. 108

Chalder, T., Berelowitz, G., Pawlikowska, T., Watts, L., Wessely, S., Wright, D., & Wallace, E. P. (1993). Development of a fatigue scale. J Psychosom Res, 37(2), 147-153.

Cham, R., & Redfern, M. S. (2001). Effect of Flooring on Standing Comfort and Fatigue. [Article]. Human Factors, 43(3), 381.

Chapman, C. B. (1990). The long reach of Harvard's Fatigue Laboratory, 1926-1947. Perspect Biol Med, 34(1), 17-33.

Chi, C.-F., & Lin, F.-T. (1998). A Comparison of Seven Visual Fatigue Assessment Techniques In Three Data-Acquisition VDT Tasks. Human Factors: The Journal of the Human Factors and Ergonomics Society, 40(4), 577-590. doi: 10.1518/001872098779649247

Chida, Y., & Steptoe, A. (2009). Cortisol awakening response and psychosocial factors: A systematic review and meta-analysis. Biological Psychology, 80(3), 265-278. doi: 10.1016/j.biopsycho.2008.10.004

Choi, B., Schnall, P. L., Yang, H., Dobson, M., Landsbergis, P., Israel, L., . . . Baker, D. (2010). Sedentary work, low physical job demand, and obesity in US workers. American journal of industrial medicine, 53(11), 1088-1101.

Christley, Y., Duffy, T., & Martin, C. R. (2010). A review of the definitional criteria for chronic fatigue syndrome. J Eval Clin Pract. doi: 10.1111/j.1365- 2753.2010.01512.x

Chrousos, G. P. (1995). The Hypothalamic–Pituitary–Adrenal Axis and Immune- Mediated Inflammation. New England Journal of Medicine, 332(20), 1351-1363. doi: doi:10.1056/NEJM199505183322008

Chrousos Gp, G. P. W. (1992). The concepts of stress and stress system disorders: Overview of physical and behavioral homeostasis. JAMA, 267(9), 1244-1252. doi: 10.1001/jama.1992.03480090092034

Church, T. S., Thomas, D. M., Tudor-Locke, C., Katzmarzyk, P. T., Earnest, C. P., Rodarte, R. Q., . . . Bouchard, C. (2011). Trends over 5 Decades in U.S. Occupation-Related Physical Activity and Their Associations with Obesity. PLoS ONE, 6(5), e19657. doi: 10.1371/journal.pone.0019657

Cleare, A. J., Miell, J., Heap, E., Sookdeo, S., Young, L., Malhi, G. S., & O’Keane, V. (2001). Hypothalamo-Pituitary-Adrenal Axis Dysfunction in Chronic Fatigue Syndrome, and the Effects of Low-Dose Hydrocortisone Therapy. Journal of Clinical Endocrinology & Metabolism, 86(8), 3545-3554. doi: 10.1210/jc.86.8.3545

109

Clements, E. C. (1926). Visual Problems in regard to Flying and Industrial Fatigue from a Service Standpoint. Proc R Soc Med, 19(War Sect), 15-23.

Collet, C., Averty, P., & Dittmar, A. (2009). Autonomic nervous system and subjective ratings of strain in air-traffic control. Applied Ergonomics, 40(1), 23-32. doi: 10.1016/j.apergo.2008.01.019

Collier, H. E. (1943). Outlines of industrial medical practice. London: Edward Arnold.

Cordero, E. D., Loredo, J. S., Murray, K. E., & Dimsdale, J. E. (2012). Characterizing Fatigue: The Effects of Ethnicity and Acculturation1. Journal of Applied Biobehavioral Research, 17(1), 59-78. doi: 10.1111/j.1751-9861.2012.00077.x

Coren, S. (1997). The sleepy society Encyclopedia Britannica Yearbook of Science and the Future (pp. 42–55). Chicago, IL: Encyclopedia Britannica.

Costanzo, L. S. (2010). Physiology. Philadelphia, PA: Saunders/Elsevier.

Crowden, G. P. (1930). Industrial Efficiency and Fatigue. Proc R Soc Med, 23(4), 479- 486.

Dababneh, A. J., Swanson, N., & Shell, R. L. (2001). Impact of added rest breaks on the productivity and well being of workers. [print]. Ergonomics, 44(2), 164-174.

Dahlgren, A., Akerstedt, T., & Kecklund, G. (2004). Individual differences in the diurnal cortisol response to stress. Chronobiol Int, 21(6), 913-922.

Dahlgren, A., Kecklund, G., & Akerstedt, T. (2005). Different levels of work-related stress and the effects on sleep, fatigue and cortisol. Scand J Work Environ Health, 31(4), 277-285. doi: 883 [pii]

Dahlgren, A., Kecklund, G., & Akerstedt, T. (2006). Overtime work and its effects on sleep, sleepiness, cortisol and blood pressure in an experimental field study. Scand J Work Environ Health, 32(4), 318-327. doi: 1016 [pii]

Dahlgren, A., Kecklund, G., & Åkerstedt, T. (2005). Different levels of work-related stress and the effects on sleep, fatigue and cortisol. Scandinavian journal of work, environment & health, 277-285.

Daniel, K. (1970). Remarks on attention control. Acta Psychologica, 33(0), 118-131. doi: 10.1016/0001-6918(70)90127-7

Davy, J., & Göbel, M. (2013). The effects of a self-selected nap opportunity on the psychophysiological, performance and subjective measures during a simulated industrial night shift regimen. Ergonomics, 56(2), 220-234. doi: 10.1080/00140139.2012.751459

110

Dawson, D., Chapman, J., & Thomas, M. J. W. (2012). Fatigue-proofing: A new approach to reducing fatigue-related risk using the principles of error management. Sleep medicine reviews, 16(2), 167-175. doi: 10.1016/j.smrv.2011.05.004

Dawson, D., & Fletcher, A. (2001). A quantitative model of work-related fatigue: background and definition. Ergonomics, 44(2), 144-163. doi: 10.1080/00140130119399

Dawson, D., Ian Noy, Y., Harma, M., Akerstedt, T., Belenky, G., Advancing, F., & Safety, R. (2011). Modelling fatigue and the use of fatigue models in work settings. Accident Analysis and Prevention, 43(2), 549-564. de Jong, N., Candel, M. J. J. M., Schouten, H. C., Huijer Abu-Saad, H., & Courtens, A. M. (2005). Course of mental fatigue and motivation in breast cancer patients receiving adjuvant chemotherapy. Annals of Oncology, 16(3), 372-382. doi: 10.1093/annonc/mdi095 de Looze, M., Bosch, T., & van Dieen, J. (2009). Manifestations of shoulder fatigue in prolonged activities involving low-force contractions. Ergonomics, 52(4), 428- 437. doi: 910771016 [pii] 10.1080/00140130802707709

DeLuca, J. (2005). Fatigue as a window to the brain. Cambridge, Mass.: MIT.

Dembe, A. E., Erickson, J. B., Delbos, R. G., & Banks, S. M. (2005). The impact of overtime and long work hours on occupational injuries and illnesses: new evidence from the United States. Occupational and Environmental Medicine, 62(9), 588-597.

Demitrack, M. A., Dale, J. K., Straus, S. E., Laue, L., Listwak, S. J., Kruesi, M., J.P., . . . Gold, P. W. (1991). Evidence for Impaired Activation of the Hypothalamic- Pituitary-Adrenal Axis in Patients with Chronic Fatigue Syndrome. Journal of Clinical Endocrinology & Metabolism, 73(6), 1224-1234. doi: 10.1210/jcem-73- 6-1224

Di Milia, L., Smolensky, M. H., Costa, G., Howarth, H. D., Ohayon, M. M., & Philip, P. (2011). Demographic factors, fatigue, and driving accidents: An examination of the published literature. Accident Analysis & Prevention, 43(2), 516-532. doi: 10.1016/j.aap.2009.12.018

DiDomenico, A. T. (2003). An investigation on subjective assessments of workload and postural stability under conditions of joint mental and physical demands. Virginia Polytechnic Institute and State University.

111

Dinos, S., Khoshaba, B., Ashby, D., White, P. D., Nazroo, J., Wessely, S., & Bhui, K. S. (2009). A systematic review of chronic fatigue, its syndromes and ethnicity: prevalence, severity, co-morbidity and coping. International Journal of Epidemiology, 38(6), 1554-1570. doi: 10.1093/ije/dyp147

Dockray, S., Bhattacharyya, M. R., Molloy, G. J., & Steptoe, A. (2008). The cortisol awakening response in relation to objective and subjective measures of waking in the morning. Psychoneuroendocrinology, 33(1), 77-82. doi: 10.1016/j.psyneuen.2007.10.001

Dorion, D., & Darveau, S. (2013). Do Micropauses Prevent Surgeon's Fatigue and Loss of Accuracy Associated With Prolonged Surgery? An Experimental Prospective Study. Annals of Surgery, 257(2), 256-259 210.1097/SLA.1090b1013e31825efe31887.

Dorrian, J., Baulk, S. D., & Dawson, D. (2011). Work hours, workload, sleep and fatigue in Australian Rail Industry employees. Applied Ergonomics, 42(2), 202-209. doi: 10.1016/j.apergo.2010.06.009

Drury, C. G., Hsiao, Y. L., Joseph, C., Joshi, S., Lapp, J., & Pennathur, P. R. (2008). Posture and performance: sitting vs. standing for security screening. Ergonomics, 51(3), 290-307. doi: 10.1080/00140130701628790

Duchon, J. C., Smith, T. J., Keran, C. M., & Koehler, E. J. (1997). Psychophysiological manifestations of performance during work on extended workshifts. International Journal of Industrial Ergonomics, 20(1), 39-49. doi: 10.1016/s0169- 8141(96)00030-3

Edwards, S., Clow, A., Evans, P., & Hucklebridge, F. (2001). Exploration of the awakening cortisol response in relation to diurnal cortisol secretory activity. Life Sci, 68(18), 2093-2103.

Eidelman, D. (1980). Fatigue: towards an analysis and a unified definition. Med Hypotheses, 6(5), 517-526. doi: 0306-9877(80)90105-X [pii]

El Falou, W., Duchêne, J., Grabisch, M., Hewson, D., Langeron, Y., & Lino, F. (2003). Evaluation of driver discomfort during long-duration car driving. Applied Ergonomics, 34(3), 249-255.

Engelmann, C., Schneider, M., Kirschbaum, C., Grote, G., Dingemann, J., Schoof, S., & Ure, B. M. (2011). Effects of intraoperative breaks on mental and somatic operator fatigue: a randomized clinical trial. Surg Endosc, 25(4), 1245-1250. doi: 10.1007/s00464-010-1350-1

112

Eppler, M., & Mengis, J. (2004). The Concept of Information Overload: A Review of Literature from Organization Science, Accounting, Marketing, MIS, and Related Disciplines. [Article]. Information Society, 20(5), 325-344. doi: 10. 1080/01972240490507974

Fekedulegn, D. B., Andrew, M. E., Burchfiel, C. M., Violanti, J. M., Hartley, T. A., Charles, L. E., & Miller, D. B. (2007). Area under the curve and other summary indicators of repeated waking cortisol measurements. Psychosom Med, 69(7), 651-659. doi: PSY.0b013e31814c405c [pii] 10.1097/PSY.0b013e31814c405c

Fell, J., & Williams, A. D. (2008). The effect of aging on skeletal muscle recovery from exercise: possible implications for the aging athlete. Journal of Aging and Physical Activity, 16(1), 97-115.

Filippi, M., & Rocca, M. A. (2007). Toward a definition of structural and functional MRI substrates of fatigue in multiple sclerosis. J Neurol Sci, 263(1-2), 1-2. doi: S0022- 510X(07)00474-1 [pii] 10.1016/j.jns.2007.07.010

Finkelman, J. M. (1994). A Large Database Study of the Factors Associated with Work- Induced Fatigue. Human Factors: The Journal of the Human Factors and Ergonomics Society, 36(2), 232-243. doi: 10.1177/001872089403600205

Fisk, E. L. (1928). Industrial Fatigue. Am J Public Health Nations Health, 18(12), 1465- 1469.

Fisk, J. D., Ritvo, P. G., Ross, L., Haase, D. A., Marrie, T. J., & Schlech, W. F. (1994). Measuring the functional impact of fatigue: initial validation of the fatigue impact scale. Clin Infect Dis, 18 Suppl 1, S79-83.

Frankenhaeuser, M., & Gardell, B. (1976). Underload and Overload in Working Life: Outline of a Multidisciplinary Approach. Journal of Human Stress, 2(3), 35-46. doi: 10.1080/0097840x.1976.9936068

Friedberg, F. (2013, 26 February 2013). New Journal: Fatigue: Biomedicine, Health and Behavior Retrieved May 26, 2013, from http://www.alphagalileo.org/ViewItem.aspx?ItemId=128849&CultureCode=en

Fukuda, K., Straus, S. E., Hickie, I., Sharpe, M. C., Dobbins, J. G., & Komaroff, A. (1994). The chronic fatigue syndrome: a comprehensive approach to its definition and study. International Chronic Fatigue Syndrome Study Group. Ann Intern Med, 121(12), 953-959.

Galinsky, T. L., Swanson, N. G., Sauter, S. L., Hurrell, J. J., & Schleifer, L. M. (2000). A field study of supplementary rest breaks for data-entry operators. Ergonomics, 43(5), 622-638. doi: 10.1080/001401300184297

113

Gander, P., Hartley, L., Powell, D., Cabon, P., Hitchcock, E., Mills, A., & Popkin, S. (2011). Fatigue risk management: Organizational factors at the regulatory and industry/company level. Accident Analysis & Prevention, 43(2), 573-590. doi: 10.1016/j.aap.2009.11.007

Gary, B. R., & Thomas, E. N. (1988). The Subjective Workload Assessment Technique: A Scaling Procedure for Measuring Mental Workload. [Article]. Advances in Psychology, 52, 185-218. doi: 10.1016/s0166-4115(08)62387-0

Gates, D. H., & Dingwell, J. B. (2010). Muscle fatigue does not lead to increased instability of upper extremity repetitive movements. Journal of Biomechanics, 43(5), 913-919. doi: 10.1016/j.jbiomech.2009.11.001

Gawron, V. J., French, J., & Funke, D. (2001). An overview of fatigue. In P. A. Hancock & P. A. Desmond (Eds.), Stress, workload, and fatigue (pp. (pp. 581-595). , xv, 682 pp). Mahwah, NJ, US: Lawrence Erlbaum Associates Publishers.

Gerard, M. J., Armstrong, T. J., Martin, B. J., & Rempel, D. A. (2002). The Effects of Work Pace on Within-Participant and Between-Participant Keying Force, Electromyography, and Fatigue. Human Factors: The Journal of the Human Factors and Ergonomics Society, 44(1), 51-61. doi: 10.1518/0018720024494757

Gilson, N. D., Burton, N. W., Van Uffelen, J. G., & Brown, W. J. (2011). Occupational sitting time: employees? perceptions of health risks and intervention strategies. Health Promotion Journal of Australia, 22(1), 38-43.

González Gutiérrez, J. L., Jiménez, B. M., Hernández, E. G., & López López, A. (2005). Spanish version of the Swedish Occupational Fatigue Inventory (SOFI): Factorial replication, reliability and validity. [Article]. International Journal of Industrial Ergonomics, 35(8), 737-746. doi: 10.1016/j.ergon.2005.02.007

Gopher, D., & Donchin, E. (1986). Workload--An examination of the concept Handbook of perception and human performance: Vol. 2. Cognitive processes and performance B2 - Handbook of perception and human performance: Vol. 2. Cognitive processes and performance. New York: Wiley.

Gottschalk, M., Kumpfel, T., Flachenecker, P., Uhr, M., Trenkwalder, C., Holsboer, F., & Weber, F. (2005). Fatigue and regulation of the hypothalamo-pituitary-adrenal axis in multiple sclerosis. Arch Neurol, 62(2), 277-280. doi: 62/2/277 [pii] 10.1001/archneur.62.2.277

Grandjean, E. (1968). Fatigue: its physiological and psychological significance. Ergonomics, 11(5), 427.

Grandjean, E. (1979). Fatigue in industry. Br J Ind Med, 36(3), 175-186.

114

Grech, M. R., Neal, A., Yeo, G., Humphreys, M., & Smith, S. (2009). An Examination of the Relationship Between Workload and Fatigue Within and Across Consecutive Days of Work: Is the Relationship Static or Dynamic? [Article]. Journal of Occupational Health Psychology, 14(3), 231-242. doi: 10.1037/a0014952

Greig, M., Marchant, D., Lovell, R., Clough, P., & McNaughton, L. (2007). A continuous mental task decreases the physiological response to soccer-specific intermittent exercise. British Journal of Sports Medicine, 41(12), 908-913.

Griffith, C. D., & Mahadevan, S. (2011). Inclusion of fatigue effects in human reliability analysis. Reliability Engineering & System Safety, 96(11), 1437-1447. doi: http://dx.doi.org/10.1016/j.ress.2011.06.005

Grohar-Murray, M. E., Becker, A., Reilly, S., & Ricci, M. (1998). Self-care actions to manage fatigue among myasthenia gravis patients. J Neurosci Nurs, 30(3), 191- 199.

Gross, A. K., & Winstead, P. S. (2009). Current controversies in critical illness-related corticosteroid insufficiency and glucocorticoid supplementation. Orthopedics, 32(9). doi: 10.3928/01477447-20090728-40 orthosupersite.com/view.asp?rID=43205 [pii] orthopedics.43205 [pii]

Guastello, S. J., Boeh, H., Gorin, H., Huschen, S., Peters, N. E., Fabisch, M., & Poston, K. (2013). Cusp catastrophe models for cognitive workload and fatigue: a comparison of seven task types. Nonlinear dynamics, psychology, and life sciences, 17(1), 23-47.

Hacker, W. (1998). Mental Workload. In J. M. Stellman (Ed.), Encyclopaedia of occupational health and safety: Geneva : International Labour Office, 1998.

4th ed. / editor-in-chief, Jeanne Mager Stellman ; senior assosicate editor, Michael McCann ; associate editors, Leon Warshaw, Carole Brabant ; senior editors, John Finklea ... [et al.] ; managing editor, Chantal Dufresne.

Hall, J. E., & Guyton, A. C. (2011). Guyton and Hall textbook of medical physiology. Philadelphia, Pa.: Saunders/Elsevier.

Hancock, P. A., & Desmond, P. A. (2000a). Stress, workload, and fatigue, from http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&db=n labk&AN=54795

Hancock, P. A., & Desmond, P. A. (2000b). Stress, Workload, and Fatigue, from https://login.proxy.library.msstate.edu/login?url=http://search.ebscohost.com/logi n.aspx?direct=true&db=nlebk&AN=54795&site=eds-live

Hancock, P. A., & Desmond, P. A. (2001). Stress, Workload, and Fatigue: Lawrence Erlbaum Associates, Inc. 115

Hancock, P. A., & Verwey, W. B. (1997). Fatigue, workload and adaptive driver systems. Accident analysis and prevention., 29(4).

Hann, D. M., Denniston, M. M., & Baker, F. (2000). Measurement of fatigue in cancer patients: further validation of the Fatigue Symptom Inventory. Qual Life Res, 9(7), 847-854.

Härmä, M. (2006). Workhours in relation to work stress, recovery and health. Scandinavian journal of work, environment & health, 502-514.

Hart, S. G., & Staveland, L. E. (1988). Development of NASA-TLX (Task Load Index):results of empirical and theoretical research. In P. A. Hancock & N. Meshkati (Eds.), Human mental workload. Amsterdam; New York; New York, N.Y., U.S.A.: North-Holland ; Sole distributors for the U.S.A. and Canada, Elsevier Science Pub. Co.

Hartz, A., Bentler, S., & Watson, D. (2003). Measuring fatigue severity in primary care patients. J Psychosom Res, 54(6), 515-521. doi: S0022399902006001 [pii]

Hashimoto, N. (1992). [Definition of the chronic fatigue syndrome and its issues]. Nippon Rinsho, 50(11), 2591-2599.

Haylock, P. J., & Hart, L. K. (1979). Fatigue in patients receiving localized radiation. Cancer Nurs, 2(6), 461-467.

Helander, M. G., & Zhang, L. (1997). Field studies of comfort and discomfort in sitting. Ergonomics, 40(9), 895-915. doi: 10.1080/001401397187739

Henning, R. A., Jacques, P., Kissel, G. V., Sullivan, A. B., & Alteras-Webb, S. M. (1997). Frequent short rest breaks from computer work: effects on productivity and well-being at two field sites. Ergonomics, 40(1), 78-91. doi: 10.1080/001401397188396

Hill, D. W., Poole, D. C., & Smith, J. C. (2002). The relationship between power and the time to achieve VO~ 2~ m~ a~ x. Medicine and science in sports and exercise, 34(4), 709-714.

Hockenberry, M. J., Hinds, P. S., Barrera, P., Bryant, R., Adams-McNeill, J., Hooke, C., . . . Manteuffel, B. (2003). Three instruments to assess fatigue in children with cancer: the child, parent and staff perspectives. J Pain Symptom Manage, 25(4), 319-328. doi: S0885392402006802 [pii]

Hockey, G. R. J., & Earle, F. (2006). Control Over the Scheduling of Simulated Office Work Reduces the Impact of Workload on Mental Fatigue and Task Performance. JOURNAL OF EXPERIMENTAL PSYCHOLOGY APPLIED, 12, 50-65.

116

Hooff, M. M., Geurts, S. E., Kompier, M. J., & Taris, T. (2007). Workdays, in-between workdays and the weekend: a diary study on effort and recovery. International Archives of Occupational and Environmental Health, 80(7), 599-613. doi: 10.1007/s00420-007-0172-5

Huang, Y. M., Chang, Y. J., Hsu, M. J., Chen, C. L., Fang, C. Y., & Wong, A. M. (2009). Errors in force generation and changes in controlling patterns following agonist muscle fatigue. J Appl Biomech, 25(4), 293-303.

Hughes, L. E., Babski-Reeves, K., & Smith-Jackson, T. (2007). Effects of psychosocial and individual factors on physiological risk factors for upper extremity musculoskeletal disorders while typing. Ergonomics, 50(2), 261-274. doi: 10.1080/00140130601049378

Hughes, R. G., & Clancy, C. M. (2008). Research linking nurses' work hours to errors prompts more state restrictions. AORN J, 87(1), 209-211. doi: S0001- 2092(07)00846-0 [pii] 10.1016/j.aorn.2007.12.020

Hurum, H., Sulheim, D., Thaulow, E., & Wyller, V. B. (2011). Elevated nocturnal blood pressure and heart rate in adolescent chronic fatigue syndrome. Acta Paediatr, 100(2), 289-292. doi: 10.1111/j.1651-2227.2010.02073.x

Huysmans, M. A., Hoozemans, M. J. M., van der Beek, A. J., de Looze, M. P., & van Dieën, J. H. (2008). Fatigue effects on tracking performance and muscle activity. Journal of Electromyography and Kinesiology, 18(3), 410-419. doi: 10.1016/j.jelekin.2006.11.003

Ikuma, L. H., Nussbaum, M. A., & Babski-Reeves, K. L. (2009). Reliability of physiological and subjective responses to physical and psychosocial exposures during a simulated manufacturing task. International Journal of Industrial Ergonomics, 39(5), 813-820.

Industrial Fatigue. (1925). Can Med Assoc J, 15(7), 737-740.

Industrial Fatigue and Ventilation. (1921). Br Med J, 1(3145), 533.

Industrial Fatigue: Work of the Research Board. (1926). Br Med J, 2(3429), 574-575.

Iriarte, J., Katsamakis, G., & de Castro, P. (1999). The Fatigue Descriptive Scale (FDS): a useful tool to evaluate fatigue in multiple sclerosis. Mult Scler, 5(1), 10-16.

Jason, L. A., Jordan, K. M., Richman, J. A., Rademaker, A. W., Huang, C.-F., Mccready, W., . . . Frankenberry, E. L. (1999). A Community-based Study of Prolonged Fatigue and Chronic Fatigue. Journal of Health Psychology, 4(1), 9-26. doi: 10.1177/135910539900400103

117

Jason La, R. J. A. R. A. W., & et al. (1999). A community-based study of chronic fatigue syndrome. Archives of Internal Medicine, 159(18), 2129-2137. doi: 10.1001/archinte.159.18.2129

Jensen, C. (2003). Development of neck and hand-wrist symptoms in relation to duration of computer use at work. [print]. Scandinavian Journal of Work Environment & Health, 29(3), 197-205.

Jiao, K., Li, Z., Chen, M., & Wang, C. (2005). Synthetic effect analysis of heart rate variability and blood pressure variability on driving mental fatigue. Journal Of Biomedical Engineering, 22(2), 343-346.

Job, R. E. S., & Dalziel, J. (2001). Defining Fatigue as a Condition of the Organism and Distinguishing It From Habituation, Adaptation, and behavior. In P. A. Hancock & P. A. Desmond (Eds.), Stress, Workload and Fatigue (pp. 468). Mahwah, New jersey: Lawrence Erlbaum Associates, Publishers.

Johansson, S., Ytterberg, C., Back, B., Holmqvist, L. W., & von Koch, L. (2008). The Swedish Occupational Fatigue Inventory in People with Multiple Sclerosis. Journal of Rehabilitation Medicine, 40(9), 737-743. doi: 10.2340/16501977-0236

Johnson, S. K., Deluca, J., Diamond, B. J., & Natelson, B. H. (1998). Memory Dysfunction in Fatiguing Illness: Examining Interference and Distraction in Short-term Memory. Cognitive Neuropsychiatry, 3(4), 269 - 285.

Johnson, S. K., Lange, G., DeLuca, J., Korn, L. R., & Natelson, B. (1997). The effects of fatigue on neuropsychological performance in patients with chronic fatigue syndrome, multiple sclerosis, and depression. Appl Neuropsychol, 4(3), 145-153. doi: 10.1207/s15324826an0403_1

Kanekar, N., Santos, M. J., & Aruin, A. S. (2008). Anticipatory postural control following fatigue of postural and focal muscles. Clinical Neurophysiology, 119(10), 2304-2313. doi: 10.1016/j.clinph.2008.06.015

Karita, K., Nakao, M., Nishikitani, M., Nomura, K., & Yano, E. (2006). Autonomic nervous activity changes in relation to the reporting of subjective symptoms among male workers in an information service company. [Article]. International Archives of Occupational & Environmental Health, 79(5), 441-444. doi: 10.1007/s00420-005-0062-7

Karlqvist, L., Tornqvist, E. W., Hagberg, M., Hagman, M., & Toomingas, A. (2002). Self-reported working conditions of VDU operators and associations with musculoskeletal symptoms: a cross-sectional study focussing on gender differences. International Journal of Industrial Ergonomics, 30(4–5), 277-294. doi: 10.1016/s0169-8141(02)00131-2

118

Karlson, B., Eek, F. C., Hansen, A. M., Garde, A. H., Osterberg, K., & Orbaek, P. (2006). Diurnal cortisol pattern of shift workers on a workday and a day off. Scandinavian Journal of Work Environment & Health(Suppl. 2), 27-34.

Kent-Braun, J. A., Ng, A. V., Doyle, J. W., & Towse, T. F. (2002). Human skeletal muscle responses vary with age and gender during fatigue due to incremental isometric exercise. Journal of Applied Physiology, 93(5), 1813-1823.

Kimura, M., Sato, H., Ochi, M., Hosoya, S., & Sadoyama, T. (2007). Electromyogram and perceived fatigue changes in the trapezius muscle during typewriting and recovery. European journal of applied physiology, 100(1), 89-96.

Kirschbaum, C., & Hellhammer, D. H. (1994). Salivary cortisol in psychoneuroendocrine research: Recent developments and applications. Psychoneuroendocrinology, 19(4), 313-333. doi: 10.1016/0306-4530(94)90013-2

Kogi, K., Saito, Y., & Mitsuhashi, T. (1970). Validity of three components of subjective fatigue feelings. Journal of Science of Labour, 46(5), 20.

Komaroff, A., & Joncas, J. H. (1991). Definition of the syndrome and differential diagnosis. Can Dis Wkly Rep, 17 Suppl 1E, 59-61.

Kramer, A. F., Sirevaag, E. J., & Braune, R. (1987). A Psychophysiological Assessment of Operator Workload During Simulated Flight Missions. Human Factors: The Journal of the Human Factors and Ergonomics Society, 29(2), 145-160. doi: 10.1177/001872088702900203

Kroemer, K. H. E. (1997). Design of the Computer Workstation. In M. Helander, T. K. Landauer & P. V. Prabhu (Eds.), Handbook of human-computer interaction. Amsterdam; New York: Elsevier.

Krupp, L. B. (2003). Fatigue in multiple sclerosis: definition, pathophysiology and treatment. CNS Drugs, 17(4), 225-234. doi: 1742 [pii]

Krupp, L. B., & Elkins, L. E. (2000). Fatigue and declines in cognitive functioning in multiple sclerosis. Neurology, 55(7), 934-939.

Krupp, L. B., LaRocca, N. G., Muir-Nash, J., & Steinberg, A. D. (1989). The fatigue severity scale. Application to patients with multiple sclerosis and systemic lupus erythematosus. Arch Neurol, 46(10), 1121-1123.

Kumari, M., Badrick, E., Chandola, T., Adam, E. K., Stafford, M., Marmot, M. G., . . . Kivimaki, M. (2009). Cortisol secretion and fatigue: associations in a community based cohort. Psychoneuroendocrinology, 34(10), 1476-1485. doi: S0306- 4530(09)00153-X [pii] 10.1016/j.psyneuen.2009.05.001

Kutner, M. H. (2005). Applied linear statistical models. Boston: McGraw Hill. 119

Kyung, G., & Nussbaum, M. A. (2008). Driver sitting comfort and discomfort (part II): Relationships with and prediction from interface pressure. International Journal of Industrial Ergonomics, 38(5–6), 526-538. doi: 10.1016/j.ergon.2007.08.011

LaManca, J. J., Sisto, S. A., DeLuca, J., Johnson, S. K., Lange, G., Pareja, J., . . . Natelson, B. H. (1998). Influence of exhaustive treadmill exercise on cognitive functioning in chronic fatigue syndrome. Am J Med, 105(3A), 59S-65S.

Laurent, C., Green, J., Bishop, P., Sjökvist, J., Schumacker, R., Richardson, M., & Curtner-Smith, M. (2010). Effect of gender on fatigue and recovery following maximal intensity repeated sprint performance. The Journal of sports medicine and physical fitness, 50(3), 243.

Lee, K. A., Hicks, G., & Nino-Murcia, G. (1991). Validity and reliability of a scale to assess fatigue. Psychiatry Res, 36(3), 291-298.

Lindegard, A., Wahlstrom, J., Hagberg, M., Vilhelmsson, R., Toomingas, A., & Wigaeus Tornqvist, E. (2012). Perceived exertion, comfort and working technique in professional computer users and associations with the incidence of neck and upper extremity symptoms. BMC Musculoskelet Disord, 13(1), 38. doi: 1471- 2474-13-38 [pii] 10.1186/1471-2474-13-38

Liu, J. Z., Shan, Z. Y., Zhang, L. D., Sahgal, V., Brown, R. W., & Yue, G. H. (2003). Human brain activation during sustained and intermittent submaximal fatigue muscle contractions: an FMRI study. J Neurophysiol, 90(1), 300-312. doi: 10.1152/jn.00821.2002 00821.2002 [pii]

Loge, J. H., Ekeberg, Ø., & Kaasa, S. (1998). Fatigue in the general norwegian population: Normative data and associations. Journal of psychosomatic research, 45(1), 53-65. doi: http://dx.doi.org/10.1016/S0022-3999(97)00291-2

Lombardi, D. A., Folkard, S., Willetts, J. L., & Smith, G. S. (2010). DAILY SLEEP, WEEKLY WORKING HOURS, AND RISK OF WORK-RELATED INJURY: US NATIONAL HEALTH INTERVIEW SURVEY (2004–2008). Chronobiology International, 27(5), 1013-1030. doi: doi:10.3109/07420528.2010.489466

Lynch, S., Main, J., & Seth, R. (1991). Definition of chronic fatigue syndrome (CFS). Br J Psychiatry, 159, 439-440.

MacDonald, W. (2003). The Impact of Job Demands and Workload on Stress and Fatigue. Australian Psychologist, 38(2), 102-117. doi: 10.1080/00050060310001707107

Manson, J. S. (1916). INDUSTRIAL FATIGUE AND ITS CAUSES (Vol. 1, pp. 570).

Manu, P., Lane, T. J., & Matthews, D. A. (1992). Chronic fatigue syndromes in clinical practice. Psychother Psychosom, 58(2), 60-68. 120

Marcora, S. M., Staiano, W., & Manning, V. (2009). Mental fatigue impairs physical performance in humans. J Appl Physiol, 106(3), 857-864. doi: 91324.2008 [pii] 10.1152/japplphysiol.91324.2008

Matthews, D. A., Lane, T. J., & Manu, P. (1988). Definition of the chronic fatigue syndrome. Ann Intern Med, 109(6), 511-512.

McArdle, W. D., Katch, F. I., & Katch, V. L. (2010). Exercise physiology : nutrition, energy, and human performance. Baltimore, MD: Lippincott Williams & Wilkins.

McLauglin, J. L. (2007). Stress, fatigue and workload: Determining the combined affect on human performance. University of Central Florida Orlando, Florida.

McLean, L., Tingley, M., Scott, R. N., & Rickards, J. (2000). Myoelectric signal measurement during prolonged computer terminal work. J Electromyogr Kinesiol, 10(1), 33-45. doi: S1050-6411(99)00021-8 [pii]

Miller, J. C. (2005). Operational Risk Management of Fatigue Effects. (ADA434836). Brooks City-Base, Texas: United States Air Force Research Laboratory.

Miller, J. C., & Eddy, D. R. (2008). Operational risk management of fatigue effects II: DTIC Document.

Mills, R. J., & Young, C. A. (2008). A medical definition of fatigue in multiple sclerosis. QJM, 101(1), 49-60. doi: hcm122 [pii] 10.1093/qjmed/hcm122

Mital, A., Bishu, R. R., & Manjunath, S. G. (1991). Review and evaluation of techniques for determining fatigue allowances. International Journal of Industrial Ergonomics, 8(2), 165-178. doi: 10.1016/0169-8141(91)90017-g

Mizuno, K., Tanaka, M., Yamaguti, K., Kajimoto, O., Kuratsune, H., & Watanabe, Y. (2011). Mental fatigue caused by prolonged cognitive load associated with sympathetic hyperactivity. Behavioral and Brain Functions, 7(1), 17.

Monod, H., & Scherrer, J. (1965). THE WORK CAPACITY OF A SYNERGIC MUSCULAR GROUP. Ergonomics, 8(3), 329-338. doi: 10.1080/00140136508930810

Moore, J. E. (2000). ONE ROAD TO TURNOVER: AN EXAMINATION OF WORK EXHAUSTION IN TECHNOLOGY PROFESSIONALS. [Article]. MIS Quarterly, 24(1), 141-168.

Mork, P. J., Vasseljen, O., & Nilsen, T. I. (2010). Association between physical exercise, body mass index, and risk of fibromyalgia: Longitudinal data from the Norwegian Nord‐Trøndelag Health Study. Arthritis Care & Research, 62(5), 611-617.

121

Muller, R., Carter, A., & Williamson, A. (2008). Epidemiological diagnosis of occupational fatigue in a fly-in-fly-out operation of the mineral industry. Annals of Occupational Hygiene, 52(1), 63-72.

Nater, U. M., Maloney, E., Boneva, R. S., Gurbaxani, B. M., Lin, J. M., Jones, J. F., . . . Heim, C. (2008). Attenuated morning salivary cortisol concentrations in a population-based study of persons with chronic fatigue syndrome and well controls. J Clin Endocrinol Metab, 93(3), 703-709. doi: jc.2007-1747 [pii] 10.1210/jc.2007-1747

Njoku, M. G. C., Jason, L. A., & Torres-Harding, S. R. (2005). The Relationships among Coping Styles and Fatigue in an Ethnically Diverse Sample. Ethnicity & Health, 10(4), 263-278. doi: 10.1080/13557850500138613

Noy, Y. I., Horrey, W. J., Popkin, S. M., Folkard, S., Howarth, H. D., & Courtney, T. K. (2011). Future directions in fatigue and safety research. Accident Analysis & Prevention, 43(2), 495-497. doi: 10.1016/j.aap.2009.12.017

Nozaki, S., Tanaka, M., Mizuno, K., Ataka, S., Mizuma, H., Tahara, T., . . . Watanabe, Y. (2009). Mental and physical fatigue-related biochemical alterations. Nutrition, 25(1), 51-57. doi: S0899-9007(08)00316-X [pii] 10.1016/j.nut.2008.07.010

Ohlinger, C. M. (2009). The effect of active workstation use on measures of cognition, attention and motor skill. Miami University.

Orlando, A., & King, P. (2004). Relationship of Demographic Variables on Perception of Fatigue and Discomfort Following Prolonged Standing Under Various Flooring Conditions. Journal of Occupational Rehabilitation, 14(1), 63-76. doi: 10.1023/b:joor.0000015011.39875.75

Östberg, O. (1973). Interindividual differences in circadian fatigue patterns of shift workers. British Journal of Industrial Medicine, 30(4), 341-351. doi: 10.1136/oem.30.4.341

Østensvik, T., Veiersted, K. B., & Nilsen, P. (2009). A method to quantify frequency and duration of sustained low-level muscle activity as a risk factor for musculoskeletal discomfort. Journal of Electromyography and Kinesiology, 19(2), 283-294. doi: 10.1016/j.jelekin.2007.07.005

Papadopoulos, A. S., & Cleare, A. J. (2012). Hypothalamic-pituitary-adrenal axis dysfunction in chronic fatigue syndrome. [10.1038/nrendo.2011.153]. Nat Rev Endocrinol, 8(1), 22-32.

Patterson, P. D., & Yealy, D. M. (2010). Resident fatigue, distress, and medical errors. JAMA, 303(4), 329; author reply 330. doi: 303/4/329-a [pii] 10.1001/jama.2010.15

122

Paul, R. H., Beatty, W. W., Schneider, R., Blanco, C. R., & Hames, K. A. (1998). Cognitive and physical fatigue in multiple sclerosis: relations between self-report and objective performance. Appl Neuropsychol, 5(3), 143-148. doi: 10.1207/s15324826an0503_5

Pietri, F., Leclerc, A., Boitel, L., Chastang, J. F., Morcet, J. F., & Blondet, M. (1992). Low-back pain in commercial travelers. Scandinavian journal of work, environment & health, 18(1), 52-58.

Piper, B. (1989). Fatigue: current bases for practice. Key Aspects of Comfort: Management of Pain, Fatigue, and Nausea. New York: Springer Publishing Co, 187-198.

Piper, B. F. (1990). Piper fatigue scale available for clinical testing. Oncol Nurs Forum, 17(5), 661-662.

Piper, B. F., & Cella, D. (2010). Cancer-related fatigue: definitions and clinical subtypes. J Natl Compr Canc Netw, 8(8), 958-966. doi: 8/8/958 [pii]

Piper, B. F., Dibble, S. L., Dodd, M. J., Weiss, M. C., Slaughter, R. E., & Paul, S. M. (1998). The revised Piper Fatigue Scale: psychometric evaluation in women with breast cancer. Oncol Nurs Forum, 25(4), 677-684.

Piper, B. F., Rieger, P. T., Brophy, L., Haeuber, D., Hood, L. E., Lyver, A., & Sharp, E. (1989). Recent advances in the management of biotherapy-related side effects: fatigue. Oncol Nurs Forum, 16(6 Suppl), 27-34.

Poole, D. C., Ward, S. A., Gardner, G. W., & Whipp, B. J. (1988). Metabolic and respiratory profile of the upper limit for prolonged exercise in man. Ergonomics, 31(9), 1265-1279. doi: 10.1080/00140138808966766

Porter, J. M., Gyi, D. E., & Tait, H. A. (2003). Interface pressure data and the prediction of driver discomfort in road trials. Applied Ergonomics, 34(3), 207-214. doi: 10.1016/s0003-6870(03)00009-7

Pruessner, J. C., Hellhammer, D. H., & Kirschbaum, C. (1999). Low self-esteem, induced failure and the adrenocortical stress response. Personality and Individual Differences, 27(3), 477-489. doi: 10.1016/s0191-8869(98)00256-6

Pruessner, J. C., Kirschbaum, C., Meinlschmid, G., & Hellhammer, D. H. (2003). Two formulas for computation of the area under the curve represent measures of total hormone concentration versus time-dependent change. Psychoneuroendocrinology, 28(7), 916-931.

Pruessner, J. C. e. a. (1997). Free cortisol levels after awakening: a reliable biological marker for the assessment of adrenocortical activity. [10.1016/S0024- 3205(97)01008-4]. Life Sci., 61, 2539-2549. 123

Qiang, J., Lan, P., & Looney, C. (2006). A probabilistic framework for modeling and real-time monitoring human fatigue. Systems, Man and Cybernetics, Part A: Systems and Humans, IEEE Transactions on, 36(5), 862-875. doi: 10.1109/tsmca.2005.855922

Rai, B., Foing, B. H., & Kaur, J. (2012). Working hours, sleep, salivary cortisol, fatigue and neuro-behavior during Mars analog mission: Five crews study. Neuroscience Letters(0). doi: 10.1016/j.neulet.2012.03.067

Ream, E., & Richardson, A. (1997). Fatigue in patients with cancer and chronic obstructive airways disease: a phenomenological enquiry. International Journal of Nursing Studies, 34(1), 44-53. doi: 10.1016/s0020-7489(96)00032-6

Reducing Industrial Fatigue. (1919). Cal State J Med, 17(1), 1.

Reeves, W. C., Wagner, D., Nisenbaum, R., Jones, J. F., Gurbaxani, B., Solomon, L., . . . Heim, C. (2005). Chronic fatigue syndrome--a clinically empirical approach to its definition and study. BMC Med, 3, 19. doi: 1741-7015-3-19 [pii] 10.1186/1741- 7015-3-19

Ricci, J. A., Chee, E., Lorandeau, A. L., & Berger, J. (2007). Fatigue in the US workforce: prevalence and implications for lost productive work time. Journal of Occupational and Environmental Medicine, 49(1), 1-10.

Robert, G., & Hockey, J. (1997). Compensatory control in the regulation of human performance under stress and high workload: A cognitive-energetical framework. Biological Psychology, 45(1–3), 73-93. doi: 10.1016/s0301-0511(96)05223-4

Rubin, G. J., Hotopf, M., Papadopoulos, A., & Cleare, A. (2005a). Salivary cortisol as a predictor of postoperative fatigue. [10.1097/01.psy.0000161207.73744.4e]. Psychosom. Med., 67, 441-447.

Rubin, G. J., Hotopf, M., Papadopoulos, A., & Cleare, A. (2005b). Salivary cortisol as a predictor of postoperative fatigue. Psychosom Med, 67(3), 441-447. doi: 67/3/441 [pii] 10.1097/01.psy.0000161207.73744.4e

Saladin, K. S. (2008). Human anatomy. Boston: McGraw-Hill.

Satoshi, N., Masaaki, T., Kei, M., Suzuka, A., Hiroshi, M., Tsuyoshi, T., . . . Yasuyoshi, W. (2009). Applied nutritional investigation: Mental and physical fatigue-related biochemical alterations. [Article]. Nutrition, 25, 51-57. doi: 10.1016/j.nut.2008.07.010

Schaufeli, W. B., Taris, T. W., & Van Rhenen, W. (2008). Workaholism, Burnout, and Work Engagement: Three of a Kind or Three Different Kinds of Employee Well‐being? Applied Psychology, 57(2), 173-203.

124

Schmidt-Reinwald, A., Pruessner, J. C., Hellhammer, D. H., Federenko, I., Rohleder, N., Schurmeyer, T. H., & Kirschbaum, C. (1999). The cortisol response to awakening in relation to different challenge tests and a 12-hour cortisol rhythm. Life Sci, 64(18), 1653-1660.

Schwartz, J. E., Jandorf, L., & Krupp, L. B. (1993). The measurement of fatigue: a new instrument. J Psychosom Res, 37(7), 753-762. doi: 0022-3999(93)90104-N [pii]

Schwid, S. R., Covington, M., Segal, B. M., & Goodman, A. D. (2002). Fatigue in multiple sclerosis: current understanding and future directions. J Rehabil Res Dev, 39(2), 211-224.

Scott, C. B., & Earnest, C. P. (2011). Resistance Exercise Energy Expenditure is Greater with Fatigue as Compared to Non-Fatigue. Journal of Exercise Physiology Online, 14(1), 1-10.

Seghers, J., Jochem, A., & Spaepen, A. (2003). Posture, muscle activity and muscle fatigue in prolonged VDT work at different screen height settings. Ergonomics, 46(7), 714-730.

Selen, L., Beek, P., & van Dieën, J. (2007). Fatigue-induced changes of impedance and performance in target tracking. Experimental Brain Research, 181(1), 99-108. doi: 10.1007/s00221-007-0909-0

Sephton, S. E., Studts, J. L., Hoover, K., Weissbecker, I., Lynch, G., Ho, I., . . . Salmon, P. (2003). Biological and Psychological Factors Associated With Memory Function in Fibromyalgia Syndrome. HEALTH PSYCHOLOGY -HILLSDALE-, 22, 592-597.

Shen, J., Barbera, J., & Shapiro, C. M. (2006). Distinguishing sleepiness and fatigue: focus on definition and measurement. Sleep medicine reviews, 10(1), 63-76.

Sjogaard, G., Kiens, B., Jorgensen, K., & Saltin, B. (1986). Intramuscular pressure, EMG and blood flow during low-level prolonged static contraction in man. Acta Physiol Scand, 128(3), 475-484.

Sjøgaard, G., Lundberg, U., & Kadefors, R. (2000). The role of muscle activity and mental load in the development of pain and degenerative processes at the muscle cell level during computer work. European journal of applied physiology, 83(2), 99-105.

Sleep in America Poll. (2008) Retrieved May 23, 2013, from http://www.sleepfoundation.org/sites/default/files/2008%20POLL%20SOF.PDF

Smets, E. M., Garssen, B., Bonke, B., & De Haes, J. C. (1995). The Multidimensional Fatigue Inventory (MFI) psychometric qualities of an instrument to assess fatigue. J Psychosom Res, 39(3), 315-325. doi: 002239999400125O [pii] 125

Smith, L., Folkard, S., Tucker, P., & Macdonald, I. (1998). Work shift duration: a review comparing eight hour and 12 hour shift systems. Occupational and Environmental Medicine, 55(4), 217-229. doi: 10.1136/oem.55.4.217

Sonnentag, S., Binnewies, C., & Mojza, E. J. (2008). " Did You Have A Nice Evening?" A Day-Level Study on Recovery Experiences, Sleep and Affect. Journal of Applied Psychology, 93(3), 674-683.

Sonnentag, S., & Fritz, C. (2007). The recovery experience questionnaire: Development and validation of a measure for assessing recuperation and unwinding from work. Journal of Occupational Health Psychology, 12(3), 204.

Sood, D., Nussbaum, M. A., & Hager, K. (2007). Fatigue during prolonged intermittent overhead work: reliability of measures and effects of working height. Ergonomics, 50(4), 497-513. doi: 10.1080/00140130601133800

Soubbotina, T. P., & Sheram, K. (2000). Beyond economic growth : meeting the challenges of global development. Washington, D.C.: World Bank.

Speier, C., Valacich, J. S., & Vessey, I. (1999). The Influence of Task Interruption on Individual Decision Making: An Information Overload Perspective. Decision Sciences, 30(2), 337-360. doi: 10.1111/j.1540-5915.1999.tb01613.x

Steele, L., Dobbins, J. G., Fukuda, K., Reyes, M., Randall, B., Koppelman, M., & Reeves, W. C. (1998). The epidemiology of chronic fatigue in San Francisco. The American Journal of Medicine, 105(3, Supplement 1), 83S-90S. doi: http://dx.doi.org/10.1016/S0002-9343(98)00158-2

Stein, K. D., Martin, S. C., Hann, D. M., & Jacobsen, P. B. (1998). A multidimensional measure of fatigue for use with cancer patients. Cancer Pract, 6(3), 143-152.

Stern, J. A., Boyer, D., & Schroeder, D. (1994). Blink rate: a possible measure of fatigue. Hum Factors, 36(2), 285-297.

Stokes, M. J., Cooper, R. G., & Edwards, R. H. (1988). Normal muscle strength and fatigability in patients with effort syndromes. BMJ, 297(6655), 1014-1017.

Strickland, P., Morriss, R., Wearden, A., & Deakin, B. (1998). A comparison of salivary cortisol in chronic fatigue syndrome, community depression and healthy controls. Journal of Affective Disorders, 47(1–3), 191-194. doi: http://dx.doi.org/10.1016/S0165-0327(97)00134-1

Sullivan, P. F., Pedersen, N. L., Jacks, A., & Evengard, B. (2005). Chronic fatigue in a population sample: definitions and heterogeneity. Psychol Med, 35(9), 1337-1348. doi: S0033291705005210 [pii] 10.1017/S0033291705005210

126

Tanaka, M., Mizuno, K., Tajima, S., Sasabe, T., & Watanabe, Y. (2009). Central nervous system fatigue alters autonomic nerve activity. Life Sci, 84(7-8), 235-239. doi: S0024-3205(08)00495-5 [pii] 10.1016/j.lfs.2008.12.004

Tanaka, M., Mizuno, K., Yamaguti, K., Kuratsune, H., Fujii, A., Baba, H., . . . Watanabe, Y. (2011). Autonomic nervous alterations associated with daily level of fatigue. Behavioral And Brain Functions: BBF, 7, 46-46.

Taylor, A. H., & Dorn, L. (2006). Stress, fatigue, health, and risk of road traffic accidents among professional drivers: the contribution of physical inactivity. Annu. Rev. Public Health, 27, 371-391.

Taylor, R. R., Jason, L. A., & Jahn, S. C. (2003). Chronic fatigue and sociodemographic characteristics as predictors of psychiatric disorders in a community-based sample. Psychosomatic medicine, 65(5), 896-901.

Taylor, W. C., King, K. E., Shegog, R., Paxton, R. J., Evans-Hudnall, G. L., Rempel, D. M., . . . Yancey, A. K. (2013). Booster Breaks in the workplace: participants’ perspectives on health-promoting work breaks. Health Education Research. doi: 10.1093/her/cyt001 ten Brummelhuis, L. L., & Bakker, A. B. (2012). Staying engaged during the week: the effect of off-job activities on next day work engagement. J Occup Health Psychol, 17(4), 445-455. doi: 10.1037/a0029213 2012-18555-001 [pii]

Tiesinga, L. J., Dassen, T. W., & Halfens, R. J. (1996). Fatigue: a summary of the definitions, dimensions, and indicators. Nurs Diagn, 7(2), 51-62.

Tiller, W. A., McCraty, R., & Atkinson, M. (1996). Cardiac coherence: a new, noninvasive measure of autonomic nervous system order. Alternative Therapies In Health And Medicine, 2(1), 52-65.

Toomingas, A., Forsman, M., Mathiassen, S. E., Heiden, M., & Nilsson, T. (2012). Variation between seated and standing/walking postures among male and female call centre operators. BMC Public Health, 12(1), 154. doi: 1471-2458-12-154 [pii] 10.1186/1471-2458-12-154

Trinkoff, A. M., Johantgen, M., Storr, C. L., Liang, Y., Han, K., & Gurses, A. P. (2011). Nurses' work schedule characteristics, nurse staffing, and patient mortality. Nurs. Res. Nursing Research, 60(1), 1-8.

Tucker, P. (2003). The impact of rest breaks upon accident risk, fatigue and performance: A review. Work & Stress, 17(2), 123-137. doi: 10.1080/0267837031000155949

. U.S. Bureau of Labor Statistics. (2013, Wednesday, June 1st, 2011) Retrieved May 24, 2013, from http://www.vitamedica.com/sedentary-work-has-contributed-to-u-s- weight-gain/ 127

Uenishi, K., Tanaka, M., Yoshida, H., Tsutsumi, S., & Miyamoto, N. (2002). 2002-01- 0773 Driver's Fatigue Evaluation During Long Term Driving for Automotive Seat Development.

Van der Hulst, M. (2003). Long workhours and health. Scandinavian journal of work, environment & health, 171-188. van der Linden, D., & Eling, P. (2006). Mental fatigue disturbs local processing more than global processing. Psychol Res, 70(5), 395-402. doi: 10.1007/s00426-005- 0228-7 van der Linden, D., Frese, M., & Meijman, T. F. (2003). Mental fatigue and the control of cognitive processes: effects on perseveration and planning. Acta Psychol (Amst), 113(1), 45-65. doi: S0001691802001506 [pii] van der Linden, D., Frese, M., & Meijman, T. F. (2003). Mental fatigue and the control of cognitive processes: effects on perseveration and planning. Acta Psychologica, 113(1), 45-65. doi: 10.1016/s0001-6918(02)00150-6 van der Linden, D., Frese, M., & Sonnentag, S. (2003). The impact of mental fatigue on exploration in a complex computer task: rigidity and loss of systematic strategies. Hum Factors, 45(3), 483-494. van der Ploeg Hp, C. T. K. R. J. B. E. B. A. (2012). SItting time and all-cause mortality risk in 222 497 australian adults. Archives of Internal Medicine, 172(6), 494-500. doi: 10.1001/archinternmed.2011.2174 van der Ploeg, H. P., Venugopal, K., Chau, J. Y., van Poppel, M. N. M., Breedveld, K., Merom, D., & Bauman, A. E. (2013). Non-Occupational Sedentary Behaviors: Population Changes in the Netherlands, 1975–2005. American Journal of Preventive Medicine, 44(4), 382-387. doi: http://dx.doi.org/10.1016/j.amepre.2012.11.034 van Dieën, J. H., Westebring-van der Putten, E. P., Kingma, I., & de Looze, M. P. (2009). Low-level activity of the trunk extensor muscles causes electromyographic manifestations of fatigue in absence of decreased oxygenation. Journal of Electromyography and Kinesiology, 19(3), 398-406. doi: http://dx.doi.org/10.1016/j.jelekin.2007.11.010

Vanhatalo, A., Fulford, J., DiMenna, F. J., & Jones, A. M. (2010). Influence of hyperoxia on muscle metabolic responses and the power–duration relationship during severe-intensity exercise in humans: a 31P magnetic resonance spectroscopy study. Experimental Physiology, 95(4), 528-540. doi: 10.1113/expphysiol.2009.050500

128

Vercoulen, J. H., Swanink, C. M., Fennis, J. F., Galama, J. M., van der Meer, J. W., & Bleijenberg, G. (1994). Dimensional assessment of chronic fatigue syndrome. J Psychosom Res, 38(5), 383-392.

Vercoulen, J. H. M. M., Bazelmans, E., Swanink, C. M. A., Galama, J. M. D., Fennis, J. F. M., van der Meer, J. W. M., & Bleijenberg, G. (1998). Evaluating Neuropsychological Impairment in Chronic Fatigue Syndrome. [Article]. Journal of Clinical & Experimental Neuropsychology, 20(2), 144-156.

Vgontzas, A. N., Bixler, E. O., & Chrousos, G. P. (2006). Obesity-Related Sleepiness and Fatigue. Annals of the New York Academy of Sciences, 1083(1), 329-344. doi: 10.1196/annals.1367.023

Vollestad, N. K. (1997). Measurement of human muscle fatigue. J Neurosci Methods, 74(2), 219-227. doi: S0165-0270(97)02251-6 [pii]

Ward, R. V. (1941). Chronic fatigue symptoms among industrial workers. Canadian Public Health Journal, 32, 464-467.

Webb, H., Weldy, M., Fabianke-Kadue, E., Orndorff, G., Kamimori, G., & Acevedo, E. (2008). Psychological stress during exercise: cardiorespiratory and hormonal responses. European Journal of Applied Physiology, 104(6), 973-981. doi: 10.1007/s00421-008-0852-1

Webb, H. E., Fabianke-Kadue, E. C., Kraemer, R. R., Kamimori, G. H., Castracane, V. D., & Acevedo, E. O. (2011). Stress Reactivity to Repeated Low-Level Challenges: A Pilot Study. Applied Psychophysiology and Biofeedback, 36(4), 243-250.

Weitzman, E. D., Fukushima, D., Nogeire, C., Roffwarg, H., Gallagher, T. F., & Hellman, L. (1971). Twenty-four hour pattern of the episodic secretion of cortisol in normal subjects. The Journal of clinical endocrinology and metabolism, 33(1), 14-22.

Welford, A. T. (1968). Fundamentals of skill. London: Methuen.

Wickens, C. D. (2002). Multiple resources and performance prediction. Theoretical Issues in Ergonomics Science, 3(2), 159-177. doi: 10.1080/14639220210123806

Wickens, C. D. (2008). Multiple Resources and Mental Workload. Human Factors: The Journal of the Human Factors and Ergonomics Society, 50(3), 449-455. doi: 10.1518/001872008x288394

Williamson, A., & Friswell, R. (2013). Fatigue in the workplace: causes and countermeasures. Fatigue: Biomedicine, Health & Behavior, 1(1-2), 81-98. doi: 10.1080/21641846.2012.744581

129

Williamson, A., Lombardi, D. A., Folkard, S., Stutts, J., Courtney, T. K., & Connor, J. L. (2011). The link between fatigue and safety. Accident Analysis & Prevention, 43(2), 498-515. doi: 10.1016/j.aap.2009.11.011

Williamsonl, A., Friswelll, R., Grzebieta, R., & Olivier, J. (2013). WHAT DO WE TELL DRIVERS ABOUT FATIGUE MANAGEMENT? Paper presented at the Contemporary Ergonomics and Human Factors 2013: Proceedings of the International Conference on Ergonomics and Human Factors 2013, Cambridge, UK, 15-18 April 2013.

Wilson, G. F., & Eggemeier, F. T. (1991). Psychophysiological assessment of workload in multi-task environments. In D. L. Damos (Ed.), Multiple task performance. London [u.a.]: Taylor & Francis.

Yamamoto, Y., LaManca, J. J., & Natelson, B. H. (2003). A measure of heart rate variability is sensitive to orthostatic challenge in women with chronic fatigue syndrome. [print]. Experimental Biology and Medicine (Maywood), 228(2), 167- 174.

Yennurajalingam, S., Palmer, J. L., Zhang, T., Poulter, V., & Bruera, E. (2008). Association between fatigue and other cancer-related symptoms in patients with advanced cancer. Supportive Care in Cancer, 16(10), 1125-1130. doi: 10.1007/s00520-008-0466-5

Yerkes, R. M., & Dodson, J. D. (1908). The relation of strength of stimulus to rapidity of habit formation. Journ. Comp. Neruol.& Psychol., 459-482.

Yoshiuchi, K., Quigley, K. S., Ohashi, K., Yamamoto, Y., & Natelson, B. H. (2004). Use of time-frequency analysis to investigate temporal patterns of cardiac autonomic response during head-up tilt in chronic fatigue syndrome. Autonomic Neuroscience, 113(1–2), 55-62. doi: 10.1016/j.autneu.2004.05.001

Young, M. S., & Stanton, N. A. (2002). Attention and automation: new perspectives on mental underload and performance. [Article]. Theoretical Issues in Ergonomics Science, 3(2), 178-194. doi: 10.1080/14639220210123789

130

APPENDIX A

DEMOGRAPHIC QUESTIONNAIRE IN FATIGUE STUDY

131

Subject ID:

Please complete the following questions.

1. Gender: □ Female □ Male

2. Age: □

3. Please choose the race, ethnicity, or national group you most identify with:

□ African-American □ Multiracial □ Asian-American □ Native American

□ Caucasian □ Native Hawaiian or Other Pacific Islander □ Foreign national

□ Other □ Hispanic

4. Marital status: □ Single □ Separated□ Married □ Other□ Divorced

5. Number of dependents (e.g., children, elderly relatives, etc.):

□ 0□ 1 - 2□ 3 - 4□ More than 4

6. How many hours of sleep do you get, on average, per night?

□ 8 - 9 □ 5 - 6□ 7 - 8 □ Less than 5□ 6 - 7 □ More than 9

7. Do you have any kind of sleep disorder such as insomnia? □ Yes □ No If yes, please explain your sleep patterns in few words.

______

132

8. What is your highest educational degree?

□ High School □ Freshman□ Sophomore □ Junior□ Senior □

Four year bachelor degrees□ Senior □ Ph.D.□ Other

9. What is your current primary occupation? ______

10. On average, how many hours per week do you work per week in your primary occupation?

□ Less than 20 □ 21 - 40□ 41 - 60 □ 61 - 80□ More than 80

11. Do you have any other jobs (even including taking care of your dependent such as children, household work) besides your main occupation? □ Yes □ No, If yes, please list:

______

12. On average, how many hours do you work per week in all jobs combined?

□ Less than 20 □ 21 - 40□ 41 - 60 □ 61 – 80 □ More than 80

13. In what degrees, do you feel fatigue at the end of your regular working day?

□ Not all □ Low□ Moderate Low □ Moderate □ Moderate High

□ High □ Very High □ Extremely High

14. In what degrees, do you feel fatigue in the Monday morning (residual fatigue from previous week)?

□ Not all □ Low□ Moderate Low □ Moderate □ Moderate High

□ High □ Very High □ Extremely High

15. Do you work in addition to house hold work (e.g. laundry) in weekends? □ Yes

□ No If yes, how many hours in total per week do you work in weekends including household and other part time jobs? 133

□ Less than 4 hours □ 6 □ 8 □ 10

□ 12 □ 14 □ 16 □ More than 16 hours

16. How many times per week do you take exercise per week?

□ No Exercise □ one□ two □ Three

□ Four □ Five □ Six □ Seven

17. How many hours in total do you relax (e.g. watching TV, reading news paper, novels, etc.) or take rest (e.g. nap) in weekdays?

□ Less than ½ an hour □ 1 hour □ 2 hours □ 3 hours □ More than 3 hours

18. How many hours in total do you relax or take rest in weekends?

□ Less than 4 hours □ 6 □ 8 □ 10

□ 12 □ 14 □ 16 □ More than 16 hours

19. Do you have any medical condition? If yes, please specify______

20. Do you regularly take any medication? If yes, please specify______

21. Do you take any beverage as stimulant such as coffee, tea, energy drinks, caffeine, etc.?

If yes, please specify according to the following:

a. Just after waking up______

b. Between one and two hours of waking up______

c. Between two and four hours of waking up______

d. During lunch______

134

c. Between one and two hours of lunch______

d. Between two to four hours after lunch______

e. During dinner______

f. During Suffer______

22. Food diary for three days: Please provide as much details as possible including the names of the main dish, side dish, appetizer, beverage, etc.

Day Breakfast Lunch dinner

Monday

Wednesday

Friday

135

APPENDIX B

MODIFIED BORG SCALE

136

Instruction to use the scale: Think of how you feel now. To what extent do the expressions below describe how you feel? For every expression, answer spontaneously, and mark the number that corresponds to how you feel right now. The numbers vary between 0 (not at all) and 15 (to absolute maximum) (Ada, et al., 2004).

Perception of Fatigue for

ulder & Neck & ulder

Leg Buttock Back Lower Back Upper Sho Eyes body Whole 0 Nothing at all “No” 0 0 0 0 0 0 0 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.5 Extremely weak Just noticeable 0.5 0.5 0.5 0.5 0.5 0.5 0.5 1 Very weak 1 1 1 1 1 1 1 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 2 Weak Light 2 2 2 2 2 2 2 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 3 Moderate 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 5 Strong Heavy 5 5 5 5 5 5 5 6 6 6 6 6 6 6 6 7 Very strong 7 7 7 7 7 7 7 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9 10 Extremely strong “Max P” 10 10 10 10 10 10 10 Absolute Maximum Highest possible Note: Body parts will be randomized, while the rating will be fixed, automatically in each application. Online format of the Borg Category Ratio (CR10) Scale to measure fatigue adapted from original version (G. A. Borg, 1982).

137

APPENDIX C

MODIFIED SWEDISH OCCUPATIONAL FATIGUE INVENTORY

138

Dimension* Sub-dimension** Not at all Vey high Breathing heavily 0 1 2 3 4 5 6 7 8 9 10 Out of breath 0 1 2 3 4 5 6 7 8 9 10 Physical Exertion Taste of blood 0 1 2 3 4 5 6 7 8 9 10 Sweaty 0 1 2 3 4 5 6 7 8 9 10 Palpitations 0 1 2 3 4 5 6 7 8 9 10 Aching 0 1 2 3 4 5 6 7 8 9 10 Hurting 0 1 2 3 4 5 6 7 8 9 10 Physical Stiff joints 0 1 2 3 4 5 6 7 8 9 10 Discomfort Numbness 0 1 2 3 4 5 6 7 8 9 10 Tense muscles 0 1 2 3 4 5 6 7 8 9 10 Uninterested 0 1 2 3 4 5 6 7 8 9 10 Passive 0 1 2 3 4 5 6 7 8 9 10 Lack of Indifferent 0 1 2 3 4 5 6 7 8 9 10 Motivation Lack of Initiative 0 1 2 3 4 5 6 7 8 9 10 Listless 0 1 2 3 4 5 6 7 8 9 10 Sleepy 0 1 2 3 4 5 6 7 8 9 10 Yawns 0 1 2 3 4 5 6 7 8 9 10 Sleepiness Drowsy 0 1 2 3 4 5 6 7 8 9 10 Fall asleep 0 1 2 3 4 5 6 7 8 9 10 Lazy 0 1 2 3 4 5 6 7 8 9 10 Overworked 0 1 2 3 4 5 6 7 8 9 10 Spent 0 1 2 3 4 5 6 7 8 9 10 Lack of energy Drained 0 1 2 3 4 5 6 7 8 9 10 Worn out 0 1 2 3 4 5 6 7 8 9 10 Exhausted 0 1 2 3 4 5 6 7 8 9 10 Note: *Short from, the five dimensions (randomized between applications) will be used Every 30 minutes.

139

APPENDIX D

WORKLOAD ASSESSMENT INSTRUMENT, NASA-TLX

140

Dimension Low High Definition of the dimension Mental 0 1 2 3 4 5 6 7 8 9 10 How much mental and perceptual activity was Demand required ( e.g., thinking, deciding, calculating, remembering, looking, searching, etc.) Was the task easy or demanding, simple or complex, exacting or forgiving? Physical 0 1 2 3 4 5 6 7 8 9 10 How much physical activity was required (e.g. Demand pushing, pulling, turning, controlling, activating, etc.)? Was the task easy or demanding, slow or brisk, slack or strenuous, restful or laborious? Temporal 0 1 2 3 4 5 6 7 8 9 10 How much time pressure did you feel due to the rate Demand or pace at which the tasks or task elements occurred? Was the pace slow and leisurely or rapid and frantic? Performance 0 1 2 3 4 5 6 7 8 9 10 How successful do you think you were in accomplishing the goals of the task set by the experimenter ( or yourself)? How satisfied were you which your performance in accomplishing these goals? Effort 0 1 2 3 4 5 6 7 8 9 10 How hard did you have to work (mentally and physically to accomplish your level of performance? Frustration 0 1 2 3 4 5 6 7 8 9 10 How insecure, discouraged, irritated, stressed and Level annoyed versus secure, gratified, content, relaxed and complacent did you feel during the task? Note: NAS-TLX adapted from the original version (Hart & Staveland, 1988)

141

APPENDIX E

RANDOMIZATION OF THE DIMENSIONS OF SUBJECTIVE SCALES

142

Table E.1 Randomization of the dimensions of subjective scales

BORG SOFI NASA S T LG BT LB UB SN ES WB PE PD LM S LE MD PD TD P E FL 1 1 2 1 3 7 4 6 5 2 1 3 5 4 2 1 3 6 4 5 1 2 1 2 7 3 6 5 4 1 2 5 3 4 1 2 6 3 5 4 1 3 1 7 2 5 4 3 6 1 5 2 4 3 1 6 2 5 4 3 1 4 7 2 5 4 3 6 1 5 1 4 3 2 6 1 4 3 2 5 1 5 2 5 4 3 7 1 6 1 4 3 2 5 2 5 4 3 6 1 1 6 5 4 2 7 1 6 3 4 3 2 5 1 4 3 2 6 1 5 1 7 5 3 7 1 6 4 2 3 2 5 1 4 4 2 6 1 5 3 1 8 3 7 1 6 4 2 5 2 5 1 4 3 3 6 1 5 4 2 1 9 7 1 5 3 2 4 6 5 1 4 3 2 6 1 5 3 2 4 1 10 2 6 4 3 5 7 1 1 5 3 2 4 1 5 3 2 4 6 2 1 5 3 2 4 6 1 7 4 2 1 3 5 5 3 2 4 6 1 2 2 4 3 5 6 2 7 1 3 2 4 5 1 3 2 4 5 1 6 2 3 3 5 6 2 7 1 4 2 3 4 1 5 3 4 5 2 6 1 2 4 4 5 2 6 1 3 7 3 4 2 5 1 4 5 2 6 1 3 2 5 5 2 6 1 4 7 3 4 2 5 1 3 4 2 5 1 3 6 2 6 2 6 1 5 7 3 4 2 4 1 3 5 2 5 1 4 6 3 2 7 6 1 5 7 3 4 2 4 1 3 5 2 5 1 4 6 2 3 2 8 1 5 7 3 4 2 6 1 4 5 2 3 1 5 6 3 4 2 2 9 4 7 2 3 1 6 5 4 5 2 3 1 4 6 2 3 1 5 2 10 6 2 3 1 5 4 7 5 2 3 1 4 6 2 3 1 5 4 3 1 2 3 1 5 4 7 6 2 3 1 5 4 2 3 1 5 4 6 3 2 2 1 5 4 7 6 3 2 1 4 3 5 2 1 4 3 6 5 3 3 1 5 4 7 6 3 2 1 3 2 5 4 1 4 3 6 5 2 3 4 5 4 7 6 3 2 1 3 2 5 4 1 4 3 6 5 2 1 3 5 5 7 6 4 3 1 2 3 5 4 2 1 4 6 5 3 2 1 3 6 7 6 4 3 1 2 5 5 4 3 2 1 6 5 4 3 1 2 3 7 7 4 3 1 2 5 6 5 4 3 1 2 6 4 3 1 2 5 3 8 4 3 1 2 5 7 6 4 3 1 2 5 4 3 1 2 5 6 3 9 4 1 2 5 7 6 3 3 1 2 4 5 3 1 2 4 6 5 3 10 1 2 4 7 5 3 6 1 2 3 5 4 1 2 4 6 5 3 4 1 1 4 7 5 2 6 3 1 3 5 4 2 1 3 6 4 2 5 4 2 3 7 5 1 6 2 4 2 5 3 1 4 3 6 4 1 5 2 4 3 7 4 1 6 2 3 5 5 3 1 4 2 6 4 1 5 2 3 4 4 4 1 7 2 3 5 6 4 1 5 2 3 4 1 6 2 3 5 4 5 1 7 2 4 5 6 3 1 5 2 3 4 1 6 2 3 4 5 4 6 7 1 4 5 6 2 3 5 1 2 3 4 6 1 3 4 5 2 4 7 2 5 6 7 3 4 1 1 3 4 5 2 1 4 5 6 2 3 4 8 4 5 6 2 3 1 7 3 4 5 1 2 4 5 6 2 3 1 4 9 5 6 3 4 2 7 1 4 5 2 3 1 4 5 2 3 1 6 4 10 6 4 5 2 7 1 3 4 2 3 1 5 5 3 4 2 6 1

143

Table E.1 (Continued)

BORG SOFI NASA S T LG BT LB UB SN ES WB PE PD LM S LE MD PD TD P E FL 5 1 4 5 2 6 1 3 7 3 4 2 5 1 4 5 2 6 1 3 5 2 5 3 6 1 4 7 2 4 2 5 1 3 4 2 5 1 3 6 5 3 3 6 1 4 7 2 5 2 4 1 3 5 3 5 1 4 6 2 5 4 5 1 3 6 2 4 7 4 1 3 5 2 5 1 3 6 2 4 5 5 1 4 6 3 5 7 2 1 3 5 2 4 1 3 5 2 4 6 5 6 4 6 2 5 7 1 3 2 4 1 3 5 3 5 2 4 6 1 5 7 6 2 4 7 1 3 5 4 2 3 5 1 5 2 4 6 1 3 5 8 2 4 7 1 3 6 5 2 4 5 1 3 2 4 6 1 3 5 5 9 3 7 1 2 6 5 4 3 5 1 2 4 3 6 1 2 5 4 5 10 7 1 2 5 4 3 6 5 1 2 4 3 6 1 2 5 4 3 6 1 1 2 6 5 3 7 4 1 2 5 4 3 1 2 5 4 3 6 6 2 1 6 5 2 7 4 3 1 4 3 2 5 1 5 4 2 6 3 6 3 6 5 1 7 4 3 2 4 3 1 5 2 5 4 1 6 3 2 6 4 6 1 7 5 3 2 4 4 1 5 3 2 5 1 6 4 3 2 6 5 1 7 6 4 2 5 3 1 5 4 3 2 1 6 5 3 2 4 6 6 6 5 3 1 4 2 7 5 4 2 1 3 6 5 3 1 4 2 6 7 6 4 1 5 3 7 2 5 3 1 4 2 5 3 1 4 2 6 6 8 5 2 6 4 7 3 1 3 1 4 2 5 4 1 5 3 6 2 6 9 2 6 5 7 4 1 3 1 4 3 5 2 2 5 4 6 3 1 6 10 6 4 7 3 1 2 5 4 3 5 2 1 5 4 6 3 1 2 7 1 4 7 3 1 2 5 6 4 5 3 1 2 4 6 3 1 2 5 7 2 7 4 1 2 5 6 3 5 3 1 2 4 6 3 1 2 4 5 7 3 4 1 2 5 6 3 7 3 1 2 4 5 4 1 2 5 6 3 7 4 1 2 4 6 3 7 5 1 2 4 5 3 1 2 4 5 3 6 7 5 2 4 6 3 7 5 1 1 3 4 2 5 1 3 5 2 6 4 7 6 3 5 2 6 4 1 7 2 4 1 5 3 3 5 2 6 4 1 7 7 4 2 5 3 1 7 6 4 2 5 3 1 4 2 5 3 1 6 7 8 3 5 4 2 7 6 1 2 4 3 1 5 2 4 3 1 6 5 7 9 5 4 2 7 6 1 3 3 2 1 5 4 4 3 2 6 5 1 7 10 4 2 7 5 1 3 6 3 2 5 4 1 4 2 6 5 1 3 8 1 2 7 5 1 4 6 3 2 5 4 1 3 2 6 4 1 3 5 8 2 7 5 1 3 6 2 4 5 3 1 2 4 6 4 1 3 5 2 8 3 6 1 3 7 2 4 5 4 1 3 5 2 5 1 3 6 2 4 8 4 1 3 7 2 5 6 4 1 3 5 2 4 1 3 6 2 4 5 8 5 2 7 1 4 6 3 5 2 5 1 3 4 2 6 1 4 5 3 8 6 7 1 4 6 2 5 3 5 1 3 4 2 6 1 3 5 2 4 8 7 1 4 6 2 5 3 7 1 3 5 2 4 1 4 6 2 5 3 8 8 4 6 2 5 3 7 1 3 5 1 4 2 3 5 1 4 2 6 8 9 6 2 5 4 7 1 3 4 1 3 2 5 5 2 4 3 6 1 8 10 3 6 5 7 1 4 2 2 4 3 5 1 2 5 4 6 1 3

144

Table E.1 (Continued)

BORG SOFI NASA S T LG BT LB UB SN ES WB PE PD LM S LE MD PD TD P E FL 9 1 6 5 7 2 4 3 1 4 3 5 1 2 5 4 6 1 3 2 9 2 5 7 2 4 3 1 6 4 5 1 3 2 5 6 2 4 3 1 9 3 7 3 5 4 2 6 1 5 2 4 3 1 6 2 4 3 1 5 9 4 3 6 4 2 7 1 5 2 4 3 1 5 3 5 4 2 6 1 9 5 6 3 2 7 1 5 4 4 3 2 5 1 5 3 2 6 1 4 9 6 3 2 7 1 5 4 6 3 2 5 1 4 3 2 6 1 5 4 9 7 2 6 1 4 3 5 7 2 5 1 4 3 2 6 1 4 3 5 9 8 6 1 3 2 4 7 5 5 1 3 2 4 5 1 3 2 4 6 9 9 1 4 2 5 7 6 3 1 3 2 4 5 1 3 2 4 6 5 9 10 3 1 4 7 5 2 6 2 1 3 5 4 3 1 4 6 5 2 10 1 1 3 7 4 2 5 6 1 3 5 4 2 1 3 6 4 2 5 10 2 2 7 3 1 4 5 6 2 5 3 1 4 2 6 3 1 4 5 10 3 7 3 2 4 5 6 1 5 2 1 3 4 6 2 1 3 4 5 10 4 3 2 4 5 6 1 7 2 1 3 4 5 3 2 4 5 6 1 10 5 2 4 5 6 1 7 3 2 3 4 5 1 2 3 4 5 1 6 10 6 4 5 6 2 7 3 1 2 3 4 1 5 3 4 5 1 6 2 10 7 5 6 3 7 4 1 2 3 4 1 5 2 4 5 2 6 3 1 10 8 5 3 7 4 1 2 6 4 2 5 3 1 5 3 6 4 1 2 10 9 3 7 5 1 2 6 4 3 5 4 1 2 3 6 4 1 2 5 10 10 7 5 1 2 6 3 4 5 3 1 2 4 6 4 1 2 5 3 11 1 6 1 2 7 3 4 5 4 1 2 5 3 5 1 2 6 3 4 11 2 1 2 7 4 5 6 3 1 2 5 3 4 1 2 6 3 4 5 11 3 1 7 3 5 6 2 4 1 5 2 3 4 1 6 3 4 5 2 11 4 7 2 4 5 1 3 6 5 2 3 4 1 6 2 4 5 1 3 11 5 2 4 5 1 3 6 7 2 4 5 1 3 2 4 5 1 3 6 11 6 3 4 1 2 6 7 5 3 4 1 2 5 3 4 1 2 5 6 11 7 3 1 2 6 7 5 4 3 1 2 4 5 3 1 2 5 6 4 11 8 1 2 6 7 4 3 5 1 2 4 5 3 1 2 5 6 4 3 11 9 1 5 6 3 2 4 7 1 4 5 3 2 1 5 6 3 2 4 11 10 5 6 3 2 4 7 1 4 5 2 1 3 4 5 2 1 3 6 12 1 5 3 2 4 7 1 6 4 2 1 3 5 5 3 2 4 6 1 12 2 4 2 5 7 1 6 3 3 2 4 5 1 3 2 4 6 1 5 12 3 3 5 7 2 6 4 1 2 3 5 1 4 2 4 6 1 5 3 12 4 4 7 2 5 3 1 6 3 5 1 4 2 4 6 2 5 3 1 12 5 7 2 4 3 1 5 6 5 2 4 3 1 6 2 4 3 1 5 12 6 2 4 3 1 5 6 7 2 4 3 1 5 2 4 3 1 5 6 12 7 3 2 1 4 5 7 6 3 2 1 4 5 3 2 1 4 5 6 12 8 2 1 4 5 7 6 3 2 1 3 4 5 2 1 3 4 6 5 12 9 2 4 5 7 6 3 1 1 2 3 5 4 1 3 4 6 5 2 12 10 4 5 7 6 3 1 2 2 3 5 4 1 3 4 6 5 2 1

145

Table E.1 (Continued)

BORG SOFI NASA S T LG BT LB UB SN ES WB PE PD LM S LE MD PD TD P E FL 13 1 5 7 6 4 1 2 3 3 5 4 2 1 4 6 5 3 1 2 13 2 7 6 4 1 2 3 5 5 4 3 1 2 6 5 4 1 2 3 13 3 7 5 1 3 4 6 2 5 4 1 2 3 6 4 1 2 3 5 13 4 6 1 4 5 7 3 2 4 1 2 3 5 5 1 3 4 6 2 13 5 1 4 5 7 3 2 6 1 3 4 5 2 1 4 5 6 3 2 13 6 3 4 6 2 1 5 7 3 4 5 2 1 3 4 6 2 1 5 13 7 4 6 2 1 5 7 3 3 5 2 1 4 3 5 2 1 4 6 13 8 6 2 1 5 7 3 4 4 2 1 3 5 5 2 1 4 6 3 13 9 3 1 6 7 4 5 2 2 1 4 5 3 2 1 5 6 3 4 13 10 2 6 7 4 5 3 1 1 4 5 2 3 1 5 6 3 4 2 14 1 6 7 3 5 2 1 4 4 5 2 3 1 5 6 3 4 2 1 14 2 7 3 5 2 1 4 6 5 3 4 2 1 6 3 5 2 1 4 14 3 4 6 2 1 5 7 3 3 5 2 1 4 3 5 2 1 4 6 14 4 5 2 1 4 7 3 6 4 2 1 3 5 5 2 1 4 6 3 14 5 2 1 4 7 3 5 6 2 1 4 5 3 2 1 4 6 3 5 14 6 1 4 7 2 5 6 3 1 3 5 2 4 1 3 6 2 4 5 14 7 4 7 1 5 6 3 2 2 5 1 3 4 3 6 1 4 5 2 14 8 7 1 5 6 4 2 3 5 1 3 4 2 6 1 4 5 3 2 14 9 1 5 6 4 2 3 7 1 4 5 3 2 1 5 6 4 2 3 14 10 5 6 4 2 3 7 1 4 5 3 1 2 4 5 3 1 2 6 15 1 6 5 3 4 7 2 1 4 3 1 2 5 5 4 2 3 6 1 15 2 5 3 4 7 2 1 6 4 2 3 5 1 5 3 4 6 2 1 15 3 4 5 7 3 2 6 1 3 4 5 2 1 3 4 6 2 1 5 15 4 5 7 3 2 6 1 4 3 5 2 1 4 4 6 3 2 5 1 15 5 7 4 3 6 2 5 1 5 3 2 4 1 6 3 2 5 1 4 15 6 5 4 7 3 6 1 2 3 2 5 1 4 4 3 6 2 5 1 15 7 4 7 3 5 1 2 6 3 5 2 4 1 4 6 3 5 1 2 15 8 7 3 4 1 2 5 6 5 3 4 1 2 6 3 4 1 2 5 15 9 3 4 1 2 5 7 6 3 4 1 2 5 3 4 1 2 5 6 15 10 4 2 3 5 7 6 1 3 1 2 4 5 3 1 2 4 6 5 16 1 3 4 5 7 6 2 1 1 2 3 5 4 2 3 4 6 5 1 16 2 3 5 7 6 2 1 4 2 3 5 4 1 3 4 6 5 2 1 16 3 4 7 5 2 1 3 6 3 5 4 2 1 4 6 5 2 1 3 16 4 7 4 2 1 3 5 6 5 4 2 1 3 6 4 2 1 3 5 16 5 4 2 1 3 5 6 7 4 2 1 3 5 4 2 1 3 5 6 16 6 2 1 3 4 5 6 7 2 1 3 4 5 2 1 3 4 5 6 16 7 1 2 3 4 5 6 7 1 2 3 4 5 1 2 3 4 5 6 16 8 1 2 3 4 5 6 7 1 2 3 4 5 1 2 3 4 5 6 16 9 2 3 4 5 6 7 1 1 2 3 4 5 1 2 3 4 5 6 16 10 2 3 4 5 7 1 6 1 2 3 4 5 2 3 4 5 6 1

146

APPENDIX F

DESCRIPTIVE STATISTICS

147

Table F.1 Descriptive statistics for Borg, SOFI, NASA and ∆HR by time

TIME N Obs Variable N Mean Std Dev Minimum Maximum 0:00 16 Borg 16 2.31 2.98 0 9 SOFI 16 1.63 2.09 0 6 NASA 16 0.69 1.58 0 6 ∆HR 16 0 0 0 0 0:30 16 Borg 16 3.88 3.12 0 9 SOFI 16 2.38 2.19 0 8 NASA 16 21.88 4.81 11 30 ∆HR 16 13.56 3.61 7 23 1:00 16 Borg 16 4.69 4.61 0 14 SOFI 16 2.88 2.7 0 9 NASA 16 22.69 4.25 17 32 ∆HR 16 15.19 3.97 10 22 1:30 16 Borg 16 5.69 5.1 0 14 SOFI 16 3.25 2.67 0 9 NASA 16 24.31 6.06 15 37 ∆HR 16 12.94 2.72 9 19 2:00 16 Borg 16 7.69 6.24 0 19 SOFI 16 3.69 2.94 0 11 NASA 16 24.63 6.53 15 37 ∆HR 16 13 2.99 7 20 2:15 16 Borg 16 2.88 3.42 0 9 SOFI 16 1.5 2.37 0 8 NASA 16 0.94 2.14 0 8 ∆HR 16 0 0 0 0 2:45 16 Borg 16 8.13 4.87 0 16 SOFI 16 5 4.58 0 18 NASA 16 26 6.75 18 41 ∆HR 16 14.06 3.57 10 24 3:15 16 Borg 16 10.06 6.71 0 22 SOFI 16 5.63 5.04 0 18 NASA 16 26.06 6.77 18 41 ∆HR 16 13.56 4 10 26 3:45 16 Borg 16 12.94 8.43 0 27 SOFI 16 7.5 6.03 0 24 NASA 16 28.75 6.63 18 42 ∆HR 16 14.25 4.12 8 23 4:15 16 Borg 16 13.81 9.25 0 28 SOFI 16 8.13 6.79 0 26 NASA 16 28.63 7.4 15 42 ∆HR 16 12.56 4.1 6 24 Borg = one-dimensional fatigue scores measured in Borg scale, SOFI = multidimensional fatigue scores measured in Swedish Occupational Fatigue Inventory, NASA = workload measured in NASA-TLX, and ∆HR = change in heart rate (bit per minute)

148

Table F.2 Descriptive statistics for Borg, SOFI, NASA and ∆HR by Ethnicity

Ethnicity Variable N Mean Std Dev Minimum Maximum Indian Borg 80 9.36 7.5 0 28 SOFI 80 5.65 5.49 0 26 NASA 80 22.04 12.37 0 42 ∆HR 80 11.63 6.88 0 26 Westerner Borg 80 5.05 5.4 0 20 SOFI 80 2.66 2.64 0 10 NASA 80 18.88 10.32 0 37 ∆HR 80 10.2 5.78 0 22 Borg = one-dimensional fatigue scores measured in Borg scale, SOFI = multidimensional fatigue scores measured in Swedish Occupational Fatigue Inventory, NASA = workload measured in NASA-TLX, and ∆HR = change in heart rate (bit per minute)

Table F.3 Descriptive statistics for Borg, SOFI, NASA and ∆HR by working shift

Shift Variable N Mean Std Dev Minimum Maximum Morning Borg 80 4.79 6.19 0 27 SOFI 80 2.4 3.01 0 13 NASA 80 20.04 12.19 0 42 ∆HR 80 9.91 5.66 0 22 Afternoon Borg 80 9.63 6.69 0 28 SOFI 80 5.91 5.13 0 26 NASA 80 20.88 10.75 0 38 ∆HR 80 11.91 6.9 0 26 Borg = one-dimensional fatigue scores measured in Borg scale, SOFI = multidimensional fatigue scores measured in Swedish Occupational Fatigue Inventory, NASA = workload measured in NASA-TLX, and ∆HR = change in heart rate (bit per minute)

149

Table F.4 Descriptive statistics for saliva cortisol concentration (CRT) and weighted CRT

Variable N Mean Std Dev Minimum Maximum Saliva cortisol concentration (μg/dL) 48 0.08 0.1 0.03 0.72 Normalized saliva cortisol concentration (μg/dL) 48 0.52 0.59 0.02 3.55

Table F.5 Descriptive statistics for CRT and weighted CRT by time

Experimental Variable N Mean Std Dev Minimum Maximum Clock Time 0:00 Saliva cortisol concentration, CRT (μg/dL) 12 0.16 0.18 0.04 0.72 Normalized CRT (μg/dL) 12 0.86 0.95 0.07 3.55 2:00 Saliva cortisol concentration (μg/dL) 12 0.05 0.01 0.03 0.08 Normalized CRT (μg/dL) 12 0.33 0.25 0.02 0.88 2:15 Saliva cortisol concentration (μg/dL) 12 0.06 0.02 0.03 0.1 Normalized CRT (μg/dL) 12 0.43 0.36 0.02 1.32 4:15 Saliva cortisol concentration (μg/dL) 12 0.06 0.03 0.03 0.12 Normalized CRT (μg/dL) 12 0.45 0.42 0.03 1.44

Table F.6 Descriptive statistics for CRT and weighted CRT by ethnicity

Ethnicity Variable N Mean Std Dev Minimum Maximum Indian Saliva cortisol concentration, CRT (μg/dL) 16 0.12 0.16 0.03 0.72 Normalized CRT (μg/dL) 16 0.93 0.82 0.23 3.55 Westerner Saliva cortisol concentration, CRT (μg/dL) 32 0.06 0.04 0.03 0.2 Normalized CRT (μg/dL) 32 0.31 0.25 0.02 0.88

Table F.7 Descriptive statistics for CRT and weighted CRT by working shift

Shift Variable N Mean Std Dev Minimum Maximum Morning Saliva cortisol concentration, CRT (μg/dL) 32 0.09 0.12 0.03 0.72 Normalized CRT (μg/dL) 32 0.6 0.68 0.03 3.55 Afternoon Saliva cortisol concentration, CRT (μg/dL) 16 0.06 0.04 0.03 0.2 Normalized CRT (μg/dL) 16 0.36 0.28 0.02 0.88

150

APPENDIX G

TUKEY ADJUSTED POST-HOC DETAILS

151

Table G.1 Tukey adjusted post-hoc for change in heart rate and NASA

∆HR NASA Effect T _T Estimate DF t Value Pr > |t| Adj P Estimate t Value Pr > |t| Adj P T 1 2 -13.56 126 -12.68 <0.0001 <0.0001 -21.19 -11.85 <0.0001 <0.0001 T 1 3 -15.19 126 -14.20 <0.0001 <0.0001 -22.00 -12.30 <0.0001 <0.0001 T 1 4 -12.94 126 -12.10 <0.0001 <0.0001 -23.63 -13.21 <0.0001 <0.0001 T 1 5 -13.00 126 -12.16 <0.0001 <0.0001 -23.94 -13.39 <0.0001 <0.0001 T 1 6 0.00 126 0.00 1.0000 1.0000 -0.25 -0.14 0.8890 1.0000 T 1 7 -14.06 126 -13.15 <0.0001 <0.0001 -25.31 -14.15 <0.0001 <0.0001 T 1 8 -13.56 126 -12.68 <0.0001 <0.0001 -25.38 -14.19 <0.0001 <0.0001 T 1 9 -14.25 126 -13.32 <0.0001 <0.0001 -28.06 -15.69 <0.0001 <0.0001 T 1 10 -12.56 126 -11.75 <0.0001 <0.0001 -27.94 -15.62 <0.0001 <0.0001 T 2 3 -1.63 126 -1.52 0.1312 0.8820 -0.81 -0.45 0.6504 1.0000 T 2 4 0.63 126 0.58 0.5600 0.9999 -2.44 -1.36 0.1753 0.9360 T 2 5 0.56 126 0.53 0.5998 1.0000 -2.75 -1.54 0.1266 0.8742 T 2 6 13.56 126 12.68 <0.0001 <0.0001 20.94 11.71 <0.0001 <0.0001 T 2 7 -0.50 126 -0.47 0.6409 1.0000 -4.13 -2.31 0.0227 0.3922 T 2 8 0.00 126 0.00 1.0000 1.0000 -4.19 -2.34 0.0208 0.3702 T 2 9 -0.69 126 -0.64 0.5215 0.9997 -6.88 -3.84 0.0002 0.0071 T 2 10 1.00 126 0.94 0.3516 0.9950 -6.75 -3.77 0.0002 0.0090 T 3 4 2.25 126 2.10 0.0374 0.5284 -1.63 -0.91 0.3653 0.9959 T 3 5 2.19 126 2.05 0.0429 0.5691 -1.94 -1.08 0.2807 0.9855 T 3 6 15.19 126 14.20 <0.0001 <0.0001 21.75 12.16 <0.0001 <0.0001 T 3 7 1.13 126 1.05 0.2948 0.9882 -3.31 -1.85 0.0663 0.7006 T 3 8 1.63 126 1.52 0.1312 0.8820 -3.38 -1.89 0.0614 0.6776 T 3 9 0.94 126 0.88 0.3824 0.9969 -6.06 -3.39 0.0009 0.0306 T 3 10 2.63 126 2.45 0.0155 0.3035 -5.94 -3.32 0.0012 0.0376 T 4 5 -0.06 126 -0.06 0.9535 1.0000 -0.31 -0.17 0.8616 1.0000 T 4 6 12.94 126 12.10 <0.0001 <0.0001 23.38 13.07 <0.0001 <0.0001 T 4 7 -1.13 126 -1.05 0.2948 0.9882 -1.69 -0.94 0.3472 0.9946 T 4 8 -0.63 126 -0.58 0.5600 0.9999 -1.75 -0.98 0.3297 0.9930 T 4 9 -1.31 126 -1.23 0.2220 0.9666 -4.44 -2.48 0.0144 0.2887 T 4 10 0.38 126 0.35 0.7264 1.0000 -4.31 -2.41 0.0173 0.3281 T 5 6 13.00 126 12.16 <0.0001 <0.0001 23.69 13.25 <0.0001 <0.0001 T 5 7 -1.06 126 -0.99 0.3224 0.9922 -1.38 -0.77 0.4434 0.9989 T 5 8 -0.56 126 -0.53 0.5998 1.0000 -1.44 -0.80 0.4230 0.9984 T 5 9 -1.25 126 -1.17 0.2447 0.9757 -4.13 -2.31 0.0227 0.3922 T 5 10 0.44 126 0.41 0.6832 1.0000 -4.00 -2.24 0.0271 0.4377 T 6 7 -14.06 126 -13.15 <0.0001 <0.0001 -25.06 -14.01 <0.0001 <0.0001 T 6 8 -13.56 126 -12.68 <0.0001 <0.0001 -25.13 -14.05 <0.0001 <0.0001 T 6 9 -14.25 126 -13.32 <0.0001 <0.0001 -27.81 -15.55 <0.0001 <0.0001 T 6 10 -12.56 126 -11.75 <0.0001 <0.0001 -27.69 -15.48 <0.0001 <0.0001 T 7 8 0.50 126 0.47 0.6409 1.0000 -0.06 -0.03 0.9722 1.0000 T 7 9 -0.19 126 -0.18 0.8611 1.0000 -2.75 -1.54 0.1266 0.8742 T 7 10 1.50 126 1.40 0.1632 0.9243 -2.63 -1.47 0.1446 0.9021 T 8 9 -0.69 126 -0.64 0.5215 0.9997 -2.69 -1.50 0.1354 0.8887 T 8 10 1.00 126 0.94 0.3516 0.9950 -2.56 -1.43 0.1544 0.9145 T 9 10 1.69 126 1.58 0.1171 0.8563 0.13 0.07 0.9444 1.0000

152

Table G.2 Tukey adjusted post-hoc for change in Borg and SOFI with respect to sleep

Borg SOFI Effect Sl T _Sl _T DF Estimate Standard Error t Value Adj P Estimate Standard Error t Value Adj P

Sl 6.5 7.5 13 3.53 0.8674 4.07 0.0013 2.21 0.6434 3.43 0.0044 T 1 2 117 -1.65 1.9396 -0.85 0.9975 -0.75 1.4387 -0.52 1.0000 T 1 3 117 -2.65 1.9396 -1.37 0.9349 -1.15 1.4387 -0.80 0.9985 T 1 4 117 -3.85 1.9396 -1.98 0.6112 -1.60 1.4387 -1.11 0.9826 T 1 5 117 -6.50 1.9396 -3.35 0.0349 -2.20 1.4387 -1.53 0.8777 T 1 6 117 -0.45 1.9396 -0.23 1.0000 0.00 1.4387 0.00 1.0000 T 1 7 117 -6.35 1.9396 -3.27 0.0436 -4.10 1.4387 -2.85 0.1324 T 1 8 117 -9.10 1.9396 -4.69 0.0003 -4.90 1.4387 -3.41 0.0296 T 1 9 117 -12.30 1.9396 -6.34 <.0001 -6.95 1.4387 -4.83 0.0002 T 1 10 117 -13.40 1.9396 -6.91 <.0001 -7.90 1.4387 -5.49 <.0001 T 2 3 117 -1.00 1.9396 -0.52 1.0000 -0.40 1.4387 -0.28 1.0000 T 2 4 117 -2.20 1.9396 -1.13 0.9801 -0.85 1.4387 -0.59 0.9999 T 2 5 117 -4.85 1.9396 -2.50 0.2792 -1.45 1.4387 -1.01 0.9913 T 2 6 117 1.20 1.9396 0.62 0.9998 0.75 1.4387 0.52 1.0000 T 2 7 117 -4.70 1.9396 -2.42 0.3220 -3.35 1.4387 -2.33 0.3790 T 2 8 117 -7.45 1.9396 -3.84 0.0074 -4.15 1.4387 -2.88 0.1218 T 2 9 117 -10.65 1.9396 -5.49 <.0001 -6.20 1.4387 -4.31 0.0014 T 2 10 117 -11.75 1.9396 -6.06 <.0001 -7.15 1.4387 -4.97 <.0001 T 3 4 117 -1.20 1.9396 -0.62 0.9998 -0.45 1.4387 -0.31 1.0000 T 3 5 117 -3.85 1.9396 -1.98 0.6112 -1.05 1.4387 -0.73 0.9993 T 3 6 117 2.20 1.9396 1.13 0.9801 1.15 1.4387 0.80 0.9985 T 3 7 117 -3.70 1.9396 -1.91 0.6639 -2.95 1.4387 -2.05 0.5657 T 3 8 117 -6.45 1.9396 -3.33 0.0376 -3.75 1.4387 -2.61 0.2265 T 3 9 117 -9.65 1.9396 -4.98 <.0001 -5.80 1.4387 -4.03 0.0038 T 3 10 117 -10.75 1.9396 -5.54 <.0001 -6.75 1.4387 -4.69 0.0003 T 4 5 117 -2.65 1.9396 -1.37 0.9349 -0.60 1.4387 -0.42 1.0000 T 4 6 117 3.40 1.9396 1.75 0.7626 1.60 1.4387 1.11 0.9826 T 4 7 117 -2.50 1.9396 -1.29 0.9543 -2.50 1.4387 -1.74 0.7717 T 4 8 117 -5.25 1.9396 -2.71 0.1833 -3.30 1.4387 -2.29 0.4010 T 4 9 117 -8.45 1.9396 -4.36 0.0011 -5.35 1.4387 -3.72 0.0111 T 4 10 117 -9.55 1.9396 -4.92 0.0001 -6.30 1.4387 -4.38 0.0011 T 5 6 117 6.05 1.9396 3.12 0.0670 2.20 1.4387 1.53 0.8777 T 5 7 117 0.15 1.9396 0.08 1.0000 -1.90 1.4387 -1.32 0.9469 T 5 8 117 -2.60 1.9396 -1.34 0.9419 -2.70 1.4387 -1.88 0.6845 T 5 9 117 -5.80 1.9396 -2.99 0.0938 -4.75 1.4387 -3.30 0.0403 T 5 10 117 -6.90 1.9396 -3.56 0.0186 -5.70 1.4387 -3.96 0.0049 T 6 7 117 -5.90 1.9396 -3.04 0.0822 -4.10 1.4387 -2.85 0.1324 T 6 8 117 -8.65 1.9396 -4.46 0.0008 -4.90 1.4387 -3.41 0.0296 T 6 9 117 -11.85 1.9396 -6.11 <.0001 -6.95 1.4387 -4.83 0.0002 T 6 10 117 -12.95 1.9396 -6.68 <.0001 -7.90 1.4387 -5.49 <.0001 T 7 8 117 -2.75 1.9396 -1.42 0.9193 -0.80 1.4387 -0.56 0.9999 T 7 9 117 -5.95 1.9396 -3.07 0.0768 -2.85 1.4387 -1.98 0.6140 T 7 10 117 -7.05 1.9396 -3.63 0.0146 -3.80 1.4387 -2.64 0.2108 T 8 9 117 -3.20 1.9396 -1.65 0.8205 -2.05 1.4387 -1.42 0.9170 T 8 10 117 -4.30 1.9396 -2.22 0.4514 -3.00 1.4387 -2.09 0.5416 T 9 10 117 -1.10 1.9396 -0.57 0.9999 -0.95 1.4387 -0.66 0.9997

153

Table G.3 Tukey adjusted post-hoc for change in Borg and SOFI with respect to weekly working hours in primary occupation

Borg SOFI Effect W T _W _T Estimate Standard t Value Adj P Estimate Standard t Value Adj P Error Error W 40 50 -4.10 0.7607 -5.39 0.0004 -1.7 0.6859 -2.48 0.0696 W 40 60 -8.48 0.9317 -9.11 <.0001 -3.0333 0.8400 -3.61 0.0093 W 50 60 -4.38 0.9317 -4.70 0.0014 -1.3333 0.8400 -1.59 0.2883 T 1 2 -1.72 1.6037 -1.07 0.9862 -0.7778 1.4460 -0.54 0.9999 T 1 3 -2.83 1.6037 -1.77 0.7542 -1.2222 1.4460 -0.85 0.9976 T 1 4 -4.22 1.6037 -2.63 0.2156 -1.6667 1.4460 -1.15 0.9777 T 1 5 -6.44 1.6037 -4.02 0.0041 -2.1667 1.4460 -1.50 0.8900 T 1 6 -0.50 1.6037 -0.31 1.0000 0.4444 1.4460 0.31 1.0000 T 1 7 -6.44 1.6037 -4.02 0.0041 -3.5 1.4460 -2.42 0.3244 T 1 8 -8.83 1.6037 -5.51 <.0001 -4.3889 1.4460 -3.04 0.0845 T 1 9 -12.28 1.6037 -7.66 <.0001 -6.3333 1.4460 -4.38 0.0011 T 1 10 -13.28 1.6037 -8.28 <.0001 -7.2778 1.4460 -5.03 <.0001 T 2 3 -1.11 1.6037 -0.69 0.9995 -0.4444 1.4460 -0.31 1.0000 T 2 4 -2.50 1.6037 -1.56 0.8645 -0.8889 1.4460 -0.61 0.9998 T 2 5 -4.72 1.6037 -2.94 0.1062 -1.3889 1.4460 -0.96 0.9938 T 2 6 1.22 1.6037 0.76 0.9989 1.2222 1.4460 0.85 0.9976 T 2 7 -4.72 1.6037 -2.94 0.1062 -2.7222 1.4460 -1.88 0.6806 T 2 8 -7.11 1.6037 -4.43 0.0009 -3.6111 1.4460 -2.50 0.2818 T 2 9 -10.56 1.6037 -6.58 <.0001 -5.5556 1.4460 -3.84 0.0076 T 2 10 -11.56 1.6037 -7.21 <.0001 -6.5 1.4460 -4.50 0.0007 T 3 4 -1.39 1.6037 -0.87 0.9971 -0.4444 1.4460 -0.31 1.0000 T 3 5 -3.61 1.6037 -2.25 0.4288 -0.9444 1.4460 -0.65 0.9997 T 3 6 2.33 1.6037 1.45 0.9064 1.6667 1.4460 1.15 0.9777 T 3 7 -3.61 1.6037 -2.25 0.4288 -2.2778 1.4460 -1.58 0.8570 T 3 8 -6.00 1.6037 -3.74 0.0106 -3.1667 1.4460 -2.19 0.4700 T 3 9 -9.44 1.6037 -5.89 <.0001 -5.1111 1.4460 -3.53 0.0204 T 3 10 -10.44 1.6037 -6.51 <.0001 -6.0556 1.4460 -4.19 0.0023 T 4 5 -2.22 1.6037 -1.39 0.9291 -0.5 1.4460 -0.35 1.0000 T 4 6 3.72 1.6037 2.32 0.3843 2.1111 1.4460 1.46 0.9046 T 4 7 -2.22 1.6037 -1.39 0.9291 -1.8333 1.4460 -1.27 0.9586 T 4 8 -4.61 1.6037 -2.88 0.1256 -2.7222 1.4460 -1.88 0.6806 T 4 9 -8.06 1.6037 -5.02 <.0001 -4.6667 1.4460 -3.23 0.0505 T 4 10 -9.06 1.6037 -5.65 <.0001 -5.6111 1.4460 -3.88 0.0066 T 5 6 5.94 1.6037 3.71 0.0118 2.6111 1.4460 1.81 0.7302 T 5 7 0.00 1.6037 0.00 1.0000 -1.3333 1.4460 -0.92 0.9954 T 5 8 -2.39 1.6037 -1.49 0.8935 -2.2222 1.4460 -1.54 0.8742 T 5 9 -5.83 1.6037 -3.64 0.0148 -4.1667 1.4460 -2.88 0.1237 T 5 10 -6.83 1.6037 -4.26 0.0017 -5.1111 1.4460 -3.53 0.0204 T 6 7 -5.94 1.6037 -3.71 0.0118 -3.9444 1.4460 -2.73 0.1760 T 6 8 -8.33 1.6037 -5.20 <.0001 -4.8333 1.4460 -3.34 0.0364 T 6 9 -11.78 1.6037 -7.34 <.0001 -6.7778 1.4460 -4.69 0.0003 T 6 10 -12.78 1.6037 -7.97 <.0001 -7.7222 1.4460 -5.34 <.0001 T 7 8 -2.39 1.6037 -1.49 0.8935 -0.8889 1.4460 -0.61 0.9998 T 7 9 -5.83 1.6037 -3.64 0.0148 -2.8333 1.4460 -1.96 0.6288 T 7 10 -6.83 1.6037 -4.26 0.0017 -3.7778 1.4460 -2.61 0.2247 T 8 9 -3.44 1.6037 -2.15 0.4987 -1.9444 1.4460 -1.34 0.9406 T 8 10 -4.44 1.6037 -2.77 0.1598 -2.8889 1.4460 -2.00 0.6024 T 9 10 -1.00 1.6037 -0.62 0.9998 -0.9444 1.4460 -0.65 0.9997

154

Table G.4 Tukey adjusted post-hoc for change in Borg and SOFI with respect to total weekly working hours in all occupations

Borg SOFI Effect TW T _TW _T DF Estimate Standard t Value Adj P Estimate Standard t Value Adj P Error Error TW 3548 5056 12 -1.93 1.0819 -1.78 0.2178 -2.75 0.9911 -2.77 0.0414 TW 3548 5674 12 -7.96 0.9766 -8.15 <.0001 -3.92 0.8946 -4.38 0.0024 TW 5056 5674 12 -6.03 0.7507 -8.03 <.0001 -1.17 0.6877 -1.70 0.2434 T 1 2 108 -1.37 1.7281 -0.79 0.9986 -0.63 1.5831 -0.40 1.0000 T 1 3 108 -1.81 1.7281 -1.04 0.9887 -0.93 1.5831 -0.58 0.9999 T 1 4 108 -2.54 1.7281 -1.47 0.9016 -1.19 1.5831 -0.75 0.9990 T 1 5 108 -3.91 1.7281 -2.26 0.4228 -1.58 1.5831 -1.00 0.9917 T 1 6 108 -0.33 1.7281 -0.19 1.0000 0.17 1.5831 0.11 1.0000 T 1 7 108 -5.03 1.7281 -2.91 0.1157 -3.09 1.5831 -1.95 0.6329 T 1 8 108 -6.31 1.7281 -3.65 0.0143 -3.56 1.5831 -2.25 0.4327 T 1 9 108 -8.43 1.7281 -4.88 0.0002 -5.26 1.5831 -3.32 0.0386 T 1 10 108 -9.07 1.7281 -5.25 <.0001 -5.67 1.5831 -3.58 0.0178 T 2 3 108 -0.44 1.7281 -0.25 1.0000 -0.30 1.5831 -0.19 1.0000 T 2 4 108 -1.17 1.7281 -0.68 0.9996 -0.56 1.5831 -0.36 1.0000 T 2 5 108 -2.54 1.7281 -1.47 0.9016 -0.95 1.5831 -0.60 0.9998 T 2 6 108 1.04 1.7281 0.60 0.9998 0.80 1.5831 0.50 1.0000 T 2 7 108 -3.66 1.7281 -2.12 0.5203 -2.46 1.5831 -1.56 0.8659 T 2 8 108 -4.94 1.7281 -2.86 0.1315 -2.93 1.5831 -1.85 0.7031 T 2 9 108 -7.06 1.7281 -4.08 0.0033 -4.63 1.5831 -2.92 0.1115 T 2 10 108 -7.70 1.7281 -4.46 0.0008 -5.04 1.5831 -3.18 0.0573 T 3 4 108 -0.73 1.7281 -0.42 1.0000 -0.27 1.5831 -0.17 1.0000 T 3 5 108 -2.10 1.7281 -1.22 0.9682 -0.66 1.5831 -0.42 1.0000 T 3 6 108 1.47 1.7281 0.85 0.9975 1.09 1.5831 0.69 0.9995 T 3 7 108 -3.22 1.7281 -1.86 0.6925 -2.17 1.5831 -1.37 0.9341 T 3 8 108 -4.50 1.7281 -2.60 0.2287 -2.63 1.5831 -1.66 0.8144 T 3 9 108 -6.62 1.7281 -3.83 0.0079 -4.33 1.5831 -2.74 0.1724 T 3 10 108 -7.27 1.7281 -4.21 0.0021 -4.74 1.5831 -2.99 0.0937 T 4 5 108 -1.37 1.7281 -0.79 0.9986 -0.39 1.5831 -0.25 1.0000 T 4 6 108 2.20 1.7281 1.28 0.9571 1.36 1.5831 0.86 0.9973 T 4 7 108 -2.49 1.7281 -1.44 0.9112 -1.90 1.5831 -1.20 0.9710 T 4 8 108 -3.77 1.7281 -2.18 0.4763 -2.36 1.5831 -1.49 0.8928 T 4 9 108 -5.89 1.7281 -3.41 0.0301 -4.06 1.5831 -2.57 0.2460 T 4 10 108 -6.54 1.7281 -3.78 0.0092 -4.47 1.5831 -2.82 0.1413 T 5 6 108 3.57 1.7281 2.07 0.5537 1.75 1.5831 1.11 0.9832 T 5 7 108 -1.12 1.7281 -0.65 0.9997 -1.51 1.5831 -0.95 0.9941 T 5 8 108 -2.40 1.7281 -1.39 0.9285 -1.97 1.5831 -1.25 0.9629 T 5 9 108 -4.52 1.7281 -2.61 0.2238 -3.68 1.5831 -2.32 0.3837 T 5 10 108 -5.17 1.7281 -2.99 0.0949 -4.08 1.5831 -2.58 0.2403 T 6 7 108 -4.69 1.7281 -2.72 0.1804 -3.26 1.5831 -2.06 0.5602 T 6 8 108 -5.97 1.7281 -3.46 0.0260 -3.72 1.5831 -2.35 0.3656 T 6 9 108 -8.09 1.7281 -4.68 0.0003 -5.43 1.5831 -3.43 0.0283 T 6 10 108 -8.74 1.7281 -5.06 <.0001 -5.83 1.5831 -3.68 0.0127 T 7 8 108 -1.28 1.7281 -0.74 0.9992 -0.46 1.5831 -0.29 1.0000 T 7 9 108 -3.40 1.7281 -1.97 0.6241 -2.17 1.5831 -1.37 0.9341 T 7 10 108 -4.05 1.7281 -2.34 0.3717 -2.57 1.5831 -1.63 0.8325 T 8 9 108 -2.12 1.7281 -1.23 0.9664 -1.70 1.5831 -1.08 0.9860 T 8 10 108 -2.77 1.7281 -1.60 0.8444 -2.11 1.5831 -1.33 0.9435 T 9 10 108 -0.65 1.7281 -0.38 1.0000 -0.41 1.5831 -0.26 1.0000

155

Table G.5 Tukey adjusted post-hoc for change in Borg and SOFI with respect to perceived fatigue at the end of the day

Borg SOFI Effect EDF T _EDF _T DF Estimat Standar t Value Pr > |t| Adj P Estimat Standar t Value Adj P e d Error e d Error EDF 13 34 12 -4.33 0.8322 -5.20 0.0002 0.0006 -1.49 0.7063 -2.11 0.1303 EDF 13 57 12 -9.30 0.9389 -9.91 <.0001 <.0001 -4.85 0.7969 -6.09 0.0001 EDF 34 57 12 -4.98 0.7528 -6.61 <.0001 <.0001 -3.36 0.6389 -5.26 0.0005 T 1 2 108 -1.57 1.5422 -1.02 0.3111 0.9906 -0.76 1.3090 -0.58 0.9999 T 1 3 108 -2.40 1.5422 -1.56 0.1222 0.8649 -1.10 1.3090 -0.84 0.9978 T 1 4 108 -3.44 1.5422 -2.23 0.0276 0.4409 -1.43 1.3090 -1.09 0.9845 T 1 5 108 -5.86 1.5422 -3.80 0.0002 0.0087 -2.00 1.3090 -1.53 0.8780 T 1 6 108 -0.38 1.5422 -0.24 0.8083 1.0000 0.07 1.3090 0.05 1.0000 T 1 7 108 -5.97 1.5422 -3.87 0.0002 0.0068 -3.93 1.3090 -3.00 0.0918 T 1 8 108 -8.33 1.5422 -5.40 <.0001 <.0001 -4.74 1.3090 -3.62 0.0157 T 1 9 108 -11.00 1.5422 -7.13 <.0001 <.0001 -6.44 1.3090 -4.92 0.0001 T 1 10 108 -11.99 1.5422 -7.77 <.0001 <.0001 -7.24 1.3090 -5.53 <.0001 T 2 3 108 -0.83 1.5422 -0.54 0.5901 0.9999 -0.33 1.3090 -0.25 1.0000 T 2 4 108 -1.88 1.5422 -1.22 0.2267 0.9683 -0.67 1.3090 -0.51 1.0000 T 2 5 108 -4.29 1.5422 -2.78 0.0064 0.1557 -1.24 1.3090 -0.94 0.9945 T 2 6 108 1.19 1.5422 0.77 0.4403 0.9988 0.83 1.3090 0.64 0.9998 T 2 7 108 -4.40 1.5422 -2.85 0.0052 0.1318 -3.17 1.3090 -2.42 0.3252 T 2 8 108 -6.76 1.5422 -4.39 <.0001 0.0011 -3.97 1.3090 -3.03 0.0846 T 2 9 108 -9.43 1.5422 -6.11 <.0001 <.0001 -5.68 1.3090 -4.34 0.0013 T 2 10 108 -10.42 1.5422 -6.75 <.0001 <.0001 -6.47 1.3090 -4.94 0.0001 T 3 4 108 -1.04 1.5422 -0.68 0.5009 0.9996 -0.33 1.3090 -0.25 1.0000 T 3 5 108 -3.46 1.5422 -2.24 0.027 0.4350 -0.90 1.3090 -0.69 0.9995 T 3 6 108 2.03 1.5422 1.31 0.1914 0.9482 1.17 1.3090 0.89 0.9965 T 3 7 108 -3.57 1.5422 -2.31 0.0225 0.3885 -2.83 1.3090 -2.16 0.4873 T 3 8 108 -5.93 1.5422 -3.85 0.0002 0.0075 -3.64 1.3090 -2.78 0.1567 T 3 9 108 -8.60 1.5422 -5.57 <.0001 <.0001 -5.35 1.3090 -4.09 0.0033 T 3 10 108 -9.58 1.5422 -6.21 <.0001 <.0001 -6.14 1.3090 -4.69 0.0003 T 4 5 108 -2.42 1.5422 -1.57 0.12 0.8608 -0.57 1.3090 -0.44 1.0000 T 4 6 108 3.07 1.5422 1.99 0.0491 0.6076 1.50 1.3090 1.15 0.9785 T 4 7 108 -2.53 1.5422 -1.64 0.1041 0.8259 -2.50 1.3090 -1.91 0.6624 T 4 8 108 -4.89 1.5422 -3.17 0.002 0.0591 -3.31 1.3090 -2.53 0.2672 T 4 9 108 -7.56 1.5422 -4.90 <.0001 0.0001 -5.01 1.3090 -3.83 0.0079 T 4 10 108 -8.54 1.5422 -5.54 <.0001 <.0001 -5.81 1.3090 -4.44 0.0009 T 5 6 108 5.49 1.5422 3.56 0.0006 0.0191 2.07 1.3090 1.58 0.8544 T 5 7 108 -0.11 1.5422 -0.07 0.9427 1.0000 -1.93 1.3090 -1.47 0.8991 T 5 8 108 -2.47 1.5422 -1.60 0.1119 0.8439 -2.74 1.3090 -2.09 0.5384 T 5 9 108 -5.14 1.5422 -3.33 0.0012 0.0375 -4.44 1.3090 -3.40 0.0312 T 5 10 108 -6.13 1.5422 -3.97 0.0001 0.0049 -5.24 1.3090 -4.00 0.0044 T 6 7 108 -5.60 1.5422 -3.63 0.0004 0.0152 -4.00 1.3090 -3.06 0.0801 T 6 8 108 -7.96 1.5422 -5.16 <.0001 <.0001 -4.81 1.3090 -3.67 0.0133 T 6 9 108 -10.63 1.5422 -6.89 <.0001 <.0001 -6.51 1.3090 -4.98 0.0001 T 6 10 108 -11.61 1.5422 -7.53 <.0001 <.0001 -7.31 1.3090 -5.58 <.0001 T 7 8 108 -2.36 1.5422 -1.53 0.1287 0.8767 -0.81 1.3090 -0.62 0.9998 T 7 9 108 -5.03 1.5422 -3.26 0.0015 0.0461 -2.51 1.3090 -1.92 0.6553 T 7 10 108 -6.01 1.5422 -3.90 0.0002 0.0062 -3.31 1.3090 -2.53 0.2672 T 8 9 108 -2.67 1.5422 -1.73 0.0867 0.7765 -1.71 1.3090 -1.31 0.9505 T 8 10 108 -3.65 1.5422 -2.37 0.0196 0.3551 -2.50 1.3090 -1.91 0.6624 T 9 10 108 -0.99 1.5422 -0.64 0.5239 0.9997 -0.79 1.3090 -0.60 0.9998

156

Table G.6 Tukey adjusted post-hoc for change in Borg and SOFI with respect to Exercise

Borg SOFI Effect Ex T _Ex _T Estimate Standard DF t Value Adj P Estimate Standard t Value Adj P Error Error Ex 0 4 3.31 1.1479 12 2.89 0.0340 -0.51 0.8866 -0.57 0.8375 Ex 0 23 2.85 0.9204 12 3.10 0.0233 0.80 0.7108 1.13 0.5151 Ex 4 23 -0.46 1.0847 12 -0.43 0.9056 1.31 0.8377 1.56 0.2982 T 1 2 -1.55 1.9268 108 -0.80 0.9984 -0.87 1.4881 -0.58 0.9999 T 1 3 -2.37 1.9268 108 -1.23 0.9660 -1.45 1.4881 -0.98 0.9930 T 1 4 -3.49 1.9268 108 -1.81 0.7260 -1.89 1.4881 -1.27 0.9587 T 1 5 -5.90 1.9268 108 -3.06 0.0790 -2.47 1.4881 -1.66 0.8140 T 1 6 -0.56 1.9268 108 -0.29 1.0000 0.05 1.4881 0.03 1.0000 T 1 7 -6.24 1.9268 108 -3.24 0.0486 -4.13 1.4881 -2.78 0.1583 T 1 8 -8.33 1.9268 108 -4.32 0.0014 -4.86 1.4881 -3.27 0.0453 T 1 9 -11.49 1.9268 108 -5.96 <.0001 -7.09 1.4881 -4.76 0.0003 T 1 10 -12.55 1.9268 108 -6.51 <.0001 -7.84 1.4881 -5.27 <.0001 T 2 3 -0.82 1.9268 108 -0.43 1.0000 -0.58 1.4881 -0.39 1.0000 T 2 4 -1.95 1.9268 108 -1.01 0.9911 -1.02 1.4881 -0.68 0.9996 T 2 5 -4.35 1.9268 108 -2.26 0.4242 -1.60 1.4881 -1.08 0.9859 T 2 6 0.98 1.9268 108 0.51 1.0000 0.92 1.4881 0.62 0.9998 T 2 7 -4.70 1.9268 108 -2.44 0.3141 -3.26 1.4881 -2.19 0.4694 T 2 8 -6.78 1.9268 108 -3.52 0.0213 -3.99 1.4881 -2.68 0.1946 T 2 9 -9.94 1.9268 108 -5.16 <.0001 -6.22 1.4881 -4.18 0.0023 T 2 10 -11.00 1.9268 108 -5.71 <.0001 -6.97 1.4881 -4.68 0.0003 T 3 4 -1.12 1.9268 108 -0.58 0.9999 -0.43 1.4881 -0.29 1.0000 T 3 5 -3.53 1.9268 108 -1.83 0.7134 -1.02 1.4881 -0.68 0.9996 T 3 6 1.81 1.9268 108 0.94 0.9948 1.50 1.4881 1.01 0.9911 T 3 7 -3.88 1.9268 108 -2.01 0.5928 -2.68 1.4881 -1.80 0.7348 T 3 8 -5.96 1.9268 108 -3.09 0.0724 -3.41 1.4881 -2.29 0.4047 T 3 9 -9.12 1.9268 108 -4.73 0.0003 -5.63 1.4881 -3.78 0.0092 T 3 10 -10.18 1.9268 108 -5.28 <.0001 -6.38 1.4881 -4.29 0.0015 T 4 5 -2.41 1.9268 108 -1.25 0.9624 -0.59 1.4881 -0.39 1.0000 T 4 6 2.93 1.9268 108 1.52 0.8810 1.93 1.4881 1.30 0.9518 T 4 7 -2.75 1.9268 108 -1.43 0.9156 -2.24 1.4881 -1.51 0.8861 T 4 8 -4.84 1.9268 108 -2.51 0.2746 -2.97 1.4881 -2.00 0.6017 T 4 9 -7.99 1.9268 108 -4.15 0.0026 -5.20 1.4881 -3.49 0.0231 T 4 10 -9.05 1.9268 108 -4.70 0.0003 -5.95 1.4881 -4.00 0.0044 T 5 6 5.34 1.9268 108 2.77 0.1604 2.52 1.4881 1.69 0.7965 T 5 7 -0.35 1.9268 108 -0.18 1.0000 -1.66 1.4881 -1.11 0.9823 T 5 8 -2.43 1.9268 108 -1.26 0.9598 -2.39 1.4881 -1.60 0.8433 T 5 9 -5.59 1.9268 108 -2.90 0.1184 -4.61 1.4881 -3.10 0.0714 T 5 10 -6.65 1.9268 108 -3.45 0.0265 -5.37 1.4881 -3.61 0.0164 T 6 7 -5.68 1.9268 108 -2.95 0.1049 -4.18 1.4881 -2.81 0.1472 T 6 8 -7.77 1.9268 108 -4.03 0.0039 -4.91 1.4881 -3.30 0.0414 T 6 9 -10.92 1.9268 108 -5.67 <.0001 -7.13 1.4881 -4.79 0.0002 T 6 10 -11.98 1.9268 108 -6.22 <.0001 -7.89 1.4881 -5.30 <.0001 T 7 8 -2.09 1.9268 108 -1.08 0.9855 -0.73 1.4881 -0.49 1.0000 T 7 9 -5.24 1.9268 108 -2.72 0.1790 -2.96 1.4881 -1.99 0.6105 T 7 10 -6.30 1.9268 108 -3.27 0.0447 -3.71 1.4881 -2.49 0.2848 T 8 9 -3.16 1.9268 108 -1.64 0.8265 -2.23 1.4881 -1.50 0.8912 T 8 10 -4.22 1.9268 108 -2.19 0.4713 -2.98 1.4881 -2.00 0.6002 T 9 10 -1.06 1.9268 108 -0.55 0.9999 -0.75 1.4881 -0.51 1.0000

157

Table G.7 Tukey adjusted post-hoc for change in Borg and SOFI with respect to daily rest

Borg SOFI Effect DR T _DR _T DF Estimate Standard t Value Adj P Estimate Standard t Value Adj P Error Error DR 3 12 13 -2.9482 0.8471 -3.48 0.0041 -0.5107 0.6455 -0.79 0.443 T 1 2 117 -1.6518 1.8941 -0.87 0.997 -0.7679 1.4434 -0.53 0.9999 T 1 3 117 -2.4821 1.8941 -1.31 0.9494 -1.3036 1.4434 -0.9 0.9961 T 1 4 117 -3.5 1.8941 -1.85 0.7034 -1.7143 1.4434 -1.19 0.9729 T 1 5 117 -5.6875 1.8941 -3 0.0909 -2.2232 1.4434 -1.54 0.8729 T 1 6 117 -0.5804 1.8941 -0.31 1 0.02679 1.4434 0.02 1 T 1 7 117 -6.0625 1.8941 -3.2 0.0536 -3.6786 1.4434 -2.55 0.2545 T 1 8 117 -8.1696 1.8941 -4.31 0.0014 -4.3214 1.4434 -2.99 0.0929 T 1 9 117 -11.205 1.8941 -5.92 <.0001 -6.4018 1.4434 -4.44 0.0008 T 1 10 117 -12.17 1.8941 -6.43 <.0001 -7.1339 1.4434 -4.94 0.0001 T 2 3 117 -0.8304 1.8941 -0.44 1 -0.5357 1.4434 -0.37 1 T 2 4 117 -1.8482 1.8941 -0.98 0.9931 -0.9464 1.4434 -0.66 0.9997 T 2 5 117 -4.0357 1.8941 -2.13 0.5101 -1.4554 1.4434 -1.01 0.9912 T 2 6 117 1.0714 1.8941 0.57 0.9999 0.7946 1.4434 0.55 0.9999 T 2 7 117 -4.4107 1.8941 -2.33 0.3788 -2.9107 1.4434 -2.02 0.5893 T 2 8 117 -6.5179 1.8941 -3.44 0.0267 -3.5536 1.4434 -2.46 0.3001 T 2 9 117 -9.5536 1.8941 -5.04 <.0001 -5.6339 1.4434 -3.9 0.0059 T 2 10 117 -10.518 1.8941 -5.55 <.0001 -6.3661 1.4434 -4.41 0.0009 T 3 4 117 -1.0179 1.8941 -0.54 0.9999 -0.4107 1.4434 -0.28 1 T 3 5 117 -3.2054 1.8941 -1.69 0.7976 -0.9196 1.4434 -0.64 0.9998 T 3 6 117 1.9018 1.8941 1 0.9915 1.3304 1.4434 0.92 0.9955 T 3 7 117 -3.5804 1.8941 -1.89 0.6755 -2.375 1.4434 -1.65 0.8228 T 3 8 117 -5.6875 1.8941 -3 0.0909 -3.0179 1.4434 -2.09 0.5377 T 3 9 117 -8.7232 1.8941 -4.61 0.0004 -5.0982 1.4434 -3.53 0.0202 T 3 10 117 -9.6875 1.8941 -5.11 <.0001 -5.8304 1.4434 -4.04 0.0037 T 4 5 117 -2.1875 1.8941 -1.15 0.9775 -0.5089 1.4434 -0.35 1 T 4 6 117 2.9196 1.8941 1.54 0.8724 1.7411 1.4434 1.21 0.97 T 4 7 117 -2.5625 1.8941 -1.35 0.9386 -1.9643 1.4434 -1.36 0.9364 T 4 8 117 -4.6696 1.8941 -2.47 0.2982 -2.6071 1.4434 -1.81 0.73 T 4 9 117 -7.7054 1.8941 -4.07 0.0033 -4.6875 1.4434 -3.25 0.047 T 4 10 117 -8.6696 1.8941 -4.58 0.0005 -5.4196 1.4434 -3.75 0.0098 T 5 6 117 5.1071 1.8941 2.7 0.1874 2.25 1.4434 1.56 0.8648 T 5 7 117 -0.375 1.8941 -0.2 1 -1.4554 1.4434 -1.01 0.9912 T 5 8 117 -2.4821 1.8941 -1.31 0.9494 -2.0982 1.4434 -1.45 0.9071 T 5 9 117 -5.5179 1.8941 -2.91 0.1136 -4.1786 1.4434 -2.89 0.1188 T 5 10 117 -6.4821 1.8941 -3.42 0.0282 -4.9107 1.4434 -3.4 0.03 T 6 7 117 -5.4821 1.8941 -2.89 0.119 -3.7054 1.4434 -2.57 0.2453 T 6 8 117 -7.5893 1.8941 -4.01 0.0041 -4.3482 1.4434 -3.01 0.0886 T 6 9 117 -10.625 1.8941 -5.61 <.0001 -6.4286 1.4434 -4.45 0.0008 T 6 10 117 -11.589 1.8941 -6.12 <.0001 -7.1607 1.4434 -4.96 0.0001 T 7 8 117 -2.1071 1.8941 -1.11 0.9825 -0.6429 1.4434 -0.45 1 T 7 9 117 -5.1429 1.8941 -2.72 0.1799 -2.7232 1.4434 -1.89 0.6779 T 7 10 117 -6.1071 1.8941 -3.22 0.0502 -3.4554 1.4434 -2.39 0.3391 T 8 9 117 -3.0357 1.8941 -1.6 0.8443 -2.0804 1.4434 -1.44 0.9114 T 8 10 117 -4 1.8941 -2.11 0.5231 -2.8125 1.4434 -1.95 0.6362 T 9 10 117 -0.9643 1.8941 -0.51 1 -0.7321 1.4434 -0.51 1

158

Table G.8 Tukey adjusted post-hoc for change in Borg and SOFI with respect to Shift

Borg SOFI Effect Sh T _Sh _T DF Estimate Standard t Value Adj P Estimate Standard t Value Adj P Error Error Sh 1 2 13 -4.15 0.8134 -5.11 0.0002 -3.48 0.5734 -6.08 <.0001 T 1 2 117 -1.69 1.8188 -0.93 0.9952 -0.80 1.2821 -0.63 0.9998 T 1 3 117 -2.49 1.8188 -1.37 0.9340 -1.32 1.2821 -1.03 0.9898 T 1 4 117 -3.62 1.8188 -1.99 0.6090 -1.74 1.2821 -1.36 0.9372 T 1 5 117 -5.63 1.8188 -3.10 0.0710 -2.18 1.2821 -1.70 0.7937 T 1 6 117 -0.60 1.8188 -0.33 1.0000 0.06 1.2821 0.05 1.0000 T 1 7 117 -6.25 1.8188 -3.44 0.0271 -3.70 1.2821 -2.88 0.1222 T 1 8 117 -8.20 1.8188 -4.51 0.0006 -4.32 1.2821 -3.37 0.0329 T 1 9 117 -11.32 1.8188 -6.22 <.0001 -6.32 1.2821 -4.93 0.0001 T 1 10 117 -12.21 1.8188 -6.71 <.0001 -7.01 1.2821 -5.47 <.0001 T 2 3 117 -0.80 1.8188 -0.44 1.0000 -0.52 1.2821 -0.40 1.0000 T 2 4 117 -1.93 1.8188 -1.06 0.9875 -0.94 1.2821 -0.73 0.9992 T 2 5 117 -3.95 1.8188 -2.17 0.4833 -1.38 1.2821 -1.07 0.9864 T 2 6 117 1.09 1.8188 0.60 0.9999 0.87 1.2821 0.68 0.9996 T 2 7 117 -4.56 1.8188 -2.51 0.2750 -2.89 1.2821 -2.26 0.4252 T 2 8 117 -6.51 1.8188 -3.58 0.0174 -3.52 1.2821 -2.74 0.1689 T 2 9 117 -9.63 1.8188 -5.30 <.0001 -5.52 1.2821 -4.30 0.0014 T 2 10 117 -10.52 1.8188 -5.78 <.0001 -6.21 1.2821 -4.84 0.0002 T 3 4 117 -1.13 1.8188 -0.62 0.9998 -0.42 1.2821 -0.33 1.0000 T 3 5 117 -3.14 1.8188 -1.73 0.7773 -0.86 1.2821 -0.67 0.9996 T 3 6 117 1.89 1.8188 1.04 0.9890 1.38 1.2821 1.08 0.9858 T 3 7 117 -3.76 1.8188 -2.07 0.5544 -2.38 1.2821 -1.85 0.7004 T 3 8 117 -5.71 1.8188 -3.14 0.0639 -3.00 1.2821 -2.34 0.3718 T 3 9 117 -8.83 1.8188 -4.85 0.0002 -5.00 1.2821 -3.90 0.0060 T 3 10 117 -9.71 1.8188 -5.34 <.0001 -5.69 1.2821 -4.44 0.0008 T 4 5 117 -2.02 1.8188 -1.11 0.9829 -0.44 1.2821 -0.34 1.0000 T 4 6 117 3.02 1.8188 1.66 0.8155 1.80 1.2821 1.41 0.9228 T 4 7 117 -2.63 1.8188 -1.45 0.9090 -1.96 1.2821 -1.53 0.8794 T 4 8 117 -4.58 1.8188 -2.52 0.2699 -2.58 1.2821 -2.01 0.5920 T 4 9 117 -7.71 1.8188 -4.24 0.0018 -4.58 1.2821 -3.57 0.0178 T 4 10 117 -8.59 1.8188 -4.72 0.0003 -5.27 1.2821 -4.11 0.0029 T 5 6 117 5.04 1.8188 2.77 0.1597 2.24 1.2821 1.75 0.7656 T 5 7 117 -0.62 1.8188 -0.34 1.0000 -1.52 1.2821 -1.18 0.9735 T 5 8 117 -2.56 1.8188 -1.41 0.9222 -2.14 1.2821 -1.67 0.8090 T 5 9 117 -5.69 1.8188 -3.13 0.0656 -4.14 1.2821 -3.23 0.0492 T 5 10 117 -6.57 1.8188 -3.61 0.0156 -4.83 1.2821 -3.77 0.0094 T 6 7 117 -5.65 1.8188 -3.11 0.0691 -3.76 1.2821 -2.93 0.1086 T 6 8 117 -7.60 1.8188 -4.18 0.0022 -4.38 1.2821 -3.42 0.0285 T 6 9 117 -10.72 1.8188 -5.90 <.0001 -6.38 1.2821 -4.98 <.0001 T 6 10 117 -11.61 1.8188 -6.38 <.0001 -7.07 1.2821 -5.52 <.0001 T 7 8 117 -1.95 1.8188 -1.07 0.9866 -0.63 1.2821 -0.49 1.0000 T 7 9 117 -5.07 1.8188 -2.79 0.1527 -2.63 1.2821 -2.05 0.5678 T 7 10 117 -5.96 1.8188 -3.27 0.0436 -3.31 1.2821 -2.58 0.2373 T 8 9 117 -3.13 1.8188 -1.72 0.7830 -2.00 1.2821 -1.56 0.8643 T 8 10 117 -4.01 1.8188 -2.20 0.4600 -2.69 1.2821 -2.10 0.5339 T 9 10 117 -0.88 1.8188 -0.49 1.0000 -0.69 1.2821 -0.54 0.9999

159

APPENDIX H

CORRELATION PROCEDURES

160

Table H.1 The CORR procedure at time = 0 minute

Spearman Correlation Coefficients, N = 16 Prob > |r| under H0: Rho=0 B S NASA ∆HR TNASA T∆HR

B 1.00000 0.78261 0.10262 . . . 0.0003 0.7053 . . .

S 0.78261 1.00000 -0.11338 . . . 0.0003 0.6759 . . .

NASA 0.10262 -0.11338 1.00000 . . . 0.7053 0.6759 . . .

∆HR ......

TNASA ......

T∆HR ......

161

Table H.2 The CORR procedure at time = 30 minutes

Spearman Correlation Coefficients, N = 16 Prob > |r| under H0: Rho=0 B S NASA ∆HR TNASA T∆HR

B 1.00000 0.71558 0.75041 0.68593 0.74538 0.70781 0.0018 0.0008 0.0034 0.0009 0.0022

S 0.71558 1.00000 0.51052 0.56534 0.49368 0.59286 0.0018 0.0433 0.0225 0.0520 0.0155

NASA 0.75041 0.51052 1.00000 0.47478 0.99926 0.48669 0.0008 0.0433 0.0631 <0.0001 0.0559

∆HR 0.68593 0.56534 0.47478 1.00000 0.46553 0.99704 0.0034 0.0225 0.0631 0.0692 <0.0001

TNASA 0.74538 0.49368 0.99926 0.46553 1.00000 0.47672 0.0009 0.0520 <0.0001 0.0692 0.0619

T∆HR 0.70781 0.59286 0.48669 0.99704 0.47672 1.00000 0.0022 0.0155 0.0559 <0.0001 0.0619

162

Table H.3 The CORR procedure at time = 60 minutes

Spearman Correlation Coefficients, N = 16 Prob > |r| under H0: Rho=0 B S NASA ∆HR TNASA T∆HR

B 1.00000 0.79849 0.69448 -0.34386 0.67346 -0.32887 0.0002 0.0028 0.1922 0.0042 0.2136

S 0.79849 1.00000 0.44605 0.02399 0.42112 0.02618 0.0002 0.0833 0.9297 0.1043 0.9233

NASA 0.69448 0.44605 1.00000 -0.16196 0.99778 -0.13047 0.0028 0.0833 0.5490 <0.0001 0.6301

∆HR -0.34386 0.02399 -0.16196 1.00000 -0.18977 0.99778 0.1922 0.9297 0.5490 0.4815 <0.0001

TNASA 0.67346 0.42112 0.99778 -0.18977 1.00000 -0.15902 0.0042 0.1043 <0.0001 0.4815 0.5564

T∆HR -0.32887 0.02618 -0.13047 0.99778 -0.15902 1.00000 0.2136 0.9233 0.6301 <0.0001 0.5564

163

Table H.4 The CORR procedure at time = 90 minutes

Spearman Correlation Coefficients, N = 16 Prob > |r| under H0: Rho=0 B S NASA ∆HR TNASA T∆HR

B 1.00000 0.82701 0.40747 0.08745 0.38187 0.07319 <0.0001 0.1172 0.7474 0.1444 0.7876

S 0.82701 1.00000 0.18840 0.21736 0.15304 0.20513 <0.0001 0.4847 0.4187 0.5715 0.4460

NASA 0.40747 0.18840 1.00000 0.26090 0.99553 0.25019 0.1172 0.4847 0.3291 <0.0001 0.3500

∆HR 0.08745 0.21736 0.26090 1.00000 0.20809 0.99925 0.7474 0.4187 0.3291 0.4393 <0.0001

TNASA 0.38187 0.15304 0.99553 0.20809 1.00000 0.19746 0.1444 0.5715 <0.0001 0.4393 0.4635

T∆HR 0.07319 0.20513 0.25019 0.99925 0.19746 1.00000 0.7876 0.4460 0.3500 <0.0001 0.4635

164

Table H.5 The CORR procedure at time = 120 minutes

Spearman Correlation Coefficients, N = 16 Prob > |r| under H0: Rho=0 B S NASA ∆HR TNASA T∆HR

B 1.00000 0.86328 0.57704 0.54444 0.55358 0.59331 <0.0001 0.0193 0.0292 0.0261 0.0154

S 0.86328 1.00000 0.48725 0.38685 0.45912 0.42717 <0.0001 0.0556 0.1388 0.0736 0.0989

NASA 0.57704 0.48725 1.00000 0.47972 0.99778 0.49628 0.0193 0.0556 0.0600 <0.0001 0.0506

∆HR 0.54444 0.38685 0.47972 1.00000 0.45837 0.99244 0.0292 0.1388 0.0600 0.0742 <0.0001

TNASA 0.55358 0.45912 0.99778 0.45837 1.00000 0.47504 0.0261 0.0736 <0.0001 0.0742 0.0630

T∆HR 0.59331 0.42717 0.49628 0.99244 0.47504 1.00000 0.0154 0.0989 0.0506 <0.0001 0.0630

165

Table H.6 The CORR procedure at time = 135 minutes

Spearman Correlation Coefficients, N = 16 Prob > |r| under H0: Rho=0 B S NASA ∆HR TNASA T∆HR

B 1.00000 0.51512 0.17307 . 0.17307 . 0.0412 0.5215 . 0.5215 .

S 0.51512 1.00000 -0.01280 . -0.01280 . 0.0412 0.9625 . 0.9625 .

NASA 0.17307 -0.01280 1.00000 . 1.00000 . 0.5215 0.9625 . <0.0001 .

∆HR ......

TNASA 0.17307 -0.01280 1.00000 . 1.00000 . 0.5215 0.9625 <0.0001 . .

T∆HR ......

166

Table H.7 The CORR procedure at time = 165 minutes

Spearman Correlation Coefficients, N = 16 Prob > |r| under H0: Rho=0 B S NASA ∆HR TNASA T∆HR

B 1.00000 0.26915 0.40701 0.36311 0.41852 0.34667 0.3134 0.1177 0.1669 0.1067 0.1884

S 0.26915 1.00000 0.04651 0.43736 0.04854 0.41748 0.3134 0.8642 0.0902 0.8583 0.1076

NASA 0.40701 0.04651 1.00000 -0.21021 0.99553 -0.20553 0.1177 0.8642 0.4346 <0.0001 0.4451

∆HR 0.36311 0.43736 -0.21021 1.00000 -0.21899 0.99553 0.1669 0.0902 0.4346 0.4152 <0.0001

TNASA 0.41852 0.04854 0.99553 -0.21899 1.00000 -0.20908 0.1067 0.8583 <0.0001 0.4152 0.4371

T∆HR 0.34667 0.41748 -0.20553 0.99553 -0.20908 1.00000 0.1884 0.1076 0.4451 <0.0001 0.4371

167

Table H.8 The CORR procedure at time = 195 minutes

Spearman Correlation Coefficients, N = 16 Prob > |r| under H0: Rho=0 B S NASA ∆HR TNASA T∆HR

B 1.00000 0.48889 0.27979 0.42573 0.27515 0.37930 0.0546 0.2939 0.1001 0.3024 0.1474

S 0.48889 1.00000 0.07869 0.13857 0.08383 0.08969 0.0546 0.7721 0.6088 0.7576 0.7412

NASA 0.27979 0.07869 1.00000 0.21974 0.99926 0.20090 0.2939 0.7721 0.4135 <0.0001 0.4556

∆HR 0.42573 0.13857 0.21974 1.00000 0.20905 0.99245 0.1001 0.6088 0.4135 0.4372 <0.0001

TNASA 0.27515 0.08383 0.99926 0.20905 1.00000 0.19254 0.3024 0.7576 <0.0001 0.4372 0.4750

T∆HR 0.37930 0.08969 0.20090 0.99245 0.19254 1.00000 0.1474 0.7412 0.4556 <0.0001 0.4750

168

Table H.9 The CORR procedure at time = 225 minutes

Spearman Correlation Coefficients, N = 16 Prob > |r| under H0: Rho=0 B S NASA ∆HR TNASA T∆HR

B 1.00000 0.61658 0.55030 0.14694 0.54102 0.16592 0.0110 0.0272 0.5871 0.0305 0.5391

S 0.61658 1.00000 0.20251 -0.06592 0.20236 -0.03941 0.0110 0.4519 0.8083 0.4523 0.8848

NASA 0.55030 0.20251 1.00000 0.03443 0.99926 0.06167 0.0272 0.4519 0.8993 <0.0001 0.8205

∆HR 0.14694 -0.06592 0.03443 1.00000 0.03291 0.99250 0.5871 0.8083 0.8993 0.9037 <0.0001

TNASA 0.54102 0.20236 0.99926 0.03291 1.00000 0.06013 0.0305 0.4523 <0.0001 0.9037 0.8249

T∆HR 0.16592 -0.03941 0.06167 0.99250 0.06013 1.00000 0.5391 0.8848 0.8205 <0.0001 0.8249

169

Table H.10 The CORR procedure at time = 255 minutes

Spearman Correlation Coefficients, N = 16 Prob > |r| under H0: Rho=0 B S NASA ∆HR TNASA T∆HR

B 1.00000 0.72272 0.52321 0.48218 0.51105 0.44709 0.0016 0.0375 0.0586 0.0431 0.0825

S 0.72272 1.00000 0.29564 0.34330 0.29173 0.32443 0.0016 0.2663 0.1930 0.2729 0.2202

NASA 0.52321 0.29564 1.00000 0.21850 0.99926 0.19362 0.0375 0.2663 0.4162 <0.0001 0.4724

∆HR 0.48218 0.34330 0.21850 1.00000 0.21089 0.99477 0.0586 0.1930 0.4162 0.4330 <0.0001

TNASA 0.51105 0.29173 0.99926 0.21089 1.00000 0.18607 0.0431 0.2729 <0.0001 0.4330 0.4902

T∆HR 0.44709 0.32443 0.19362 0.99477 0.18607 1.00000 0.0825 0.2202 0.4724 <0.0001 0.4902

170

Table H.11 The CORR procedure over time

Spearman Correlation Coefficients, N = 160 Prob > |r| under H0: Rho=0 B S NASA ∆HR TNASA T∆HR

B 1.00000 0.77622 0.63984 0.37874 0.60563 0.56927 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

S 0.77622 1.00000 0.51511 0.35658 0.52054 0.51823 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

NASA 0.63984 0.51511 1.00000 0.53713 0.80629 0.67106 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

∆HR 0.37874 0.35658 0.53713 1.00000 0.45837 0.62862 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

TNASA 0.60563 0.52054 0.80629 0.45837 1.00000 0.93597 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

T∆HR 0.56927 0.51823 0.67106 0.62862 0.93597 1.00000 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

171