You are on page 1of 13

ORIGINAL ARTICLE

Causes and Consequences of Occupational Fatigue:


Meta-Analysis and Systems Model
Ulises Techera, MS, Matthew Hallowell, PhD, Nathan Stambaugh, BS, and Ray Littlejohn, PhD

To address this need, we aimed to codify the literature related


Objective: The study objective was to statistically codify empirical
to occupational fatigue into a systems model that (1) identifies
occupational fatigue literature into a systems model depicting causal factors
causal factors of fatigue; (2) evaluates interrelationships among
of fatigue; interrelationships among causal factors; outcomes of fatigue; and
causal factors; (3) identifies outcomes of fatigue; and (4) evaluates
interrelationships among outcomes. Methods: The objective was achieved
the interrelationships among fatigue outcomes. Although there are
via a comprehensive and systematic review of existing literature and the
several hundred studies of fatigue, we built the model using only
first statistical meta-analysis of occupational fatigue. Results: The drivers
empirical and validated findings.
of fatigue with the greatest effect sizes include sleep deprivation and
In addition, a meta-analysis was performed using published
work environment factors such as noise, vibration, and temperature. The
data and formal statistics. Before the current paper, no formal or
most significant outcomes of fatigue include short-term cognitive
comprehensive meta-analysis had been performed on the causes or
and physical degradation and, to a lesser extent, error, injury, and illness.
consequences of occupational fatigue. The meta-analysis allowed us
Conclusions: Although there is a great deal of occupational fatigue literature,
to use a single statistic (Cohen d) to explain the direction and
there is a dearth of replication and validations studies. Researchers are also
magnitude of relationships among causes and consequences of
encouraged to address knowledge gaps like the relationship between worker
fatigue. This value was computed based upon the effect sizes,
relationships and fatigue.
sample sizes, and variance reported in past empirical studies.
The meta-analysis complements the systems model, which focuses
E very day, more than 20% of the working population in the
United States experiences occupational fatigue,1 resulting in
$136.4 billion in lost productivity and health care costs each year.2
on illustrating the number of studies for each relationship by
explaining the strength of the relationships and their statistical
significance. Consequently, a revised model was created to illustrate
Fatigue diminishes the quality of life of individuals and effect sizes and directions of influence.
weakens the immune system. Despite these important implications,
fatigue is a common occurrence, making it difficult for people to FATIGUE DEFINITION AND CLASSIFICATION
recognize its consequences and understand various primary and Before reviewing literature on the causes and consequences
secondary causes. of fatigue, it is important to provide a pithy definition of the various
Unfortunately, despite its implications, fatigue is often types of fatigue. In general, fatigue is a condition of the individual
uncovered only after a major accident.3 For example, Harrington4 that is recognized as a decreased ability to perform activities at the
showed how fatigue provably played a role in the Three Mile Island, desired level due to lassitude or exhaustion of mental and/or
Chernobyl, and the Exxon Valdez oil spill events, as they took place physical strength.8,9 Humans commonly identify this state when
during the first hours of the morning after a night shift when the feeling tired or weary. The condition of fatigue is experienced by
levels of fatigue among workers were at their highest. The negative every person10 and, depending on its causes, can be classified as
outcomes of fatigue are not surprising, as fatigue decreases the acute or chronic.11
ability to process and react to new information and respond to Acute and chronic fatigue is distinguished from one another
hazards.5–7 by their principal causes and long-term effects. Acute fatigue is
Past literature has called for a deeper understanding of the experienced as a consequence of mental or bodily labor, emotional
causes and consequences of occupational fatigue.3 In pursuit of this stress, insufficient recovery, or a temporary illness.10 It is considered
knowledge, researchers have aimed to study specific causes or to be a normal regulatory response to adverse condition and affects
consequences of fatigue using rigorous experimental methods. healthy people. Typically, acute fatigue can be relieved by quality
Despite the relatively large body of knowledge in the field of rest, sleep, appropriate diet, and exercise.11,12 Chronic fatigue, on
occupational fatigue, there is no single resource that catalogs and the contrary, can manifest as a side effect of severe illnesses or
organizes empirical literature or that integrates the findings in a treatments such as rheumatoid arthritis, diabetes, multiple sclerosis,
meaningful and comprehensive manner. Thus, researchers may find radiation exposure, and chemotherapy.9,11 The symptoms of chronic
it difficult to identify new theoretical points of departure and fatigue are similar to those produced by acute fatigue except that
cumbersome to perform high-quality literature reviews. they are experienced constantly and cannot be alleviated by rest
alone.13 In most occupational environments, the primary concern is
acute fatigue, which is the focus of this paper.
In addition to the distinction between acute and chronic,
From the Department of Civil, Environmental, and Architectural Engineering, fatigue can also be classified as muscular or mental. Mental fatigue
University of Colorado at Boulder (Drs Techara, Hallowell, Stambaugh); and
Lockheed Martin Engineering Management Program, UCB 428, Boulder, CO is associated with a decreased motivation to continue performing a
(Dr Littlejohn). current activity and a sensation of weariness.7 Mental fatigue
This project was funded by ELECTRI International under award 2014-03. decreases the ability to process and respond to information, ulti-
ELECTRI International encourages free publishing of all research results mately diminishing competency, productivity, and error avoid-
and does not request review or approval before publication. This organization
does not require formal acknowledgment. ance.5–7 Alternatively, muscular fatigue is a reduction in the
The authors have no conflicts of interest. physical ability to exert a force or perform a task. It is often a
Address correspondence to: Matthew Hallowell, PhD, UCB 428, 1111 Engin- result of either metabolic or neural decay over the course of an
eering Drive, Boulder, CO 80309-0428 (matthew.hallowell@colorado.edu). extended physical activity. Muscular fatigue is most often caused by
Copyright ß 2016 American College of Occupational and Environmental
Medicine high-intensity work, long duration of work, or improper work
DOI: 10.1097/JOM.0000000000000837 posture. Blue collar workers who are often required to perform

JOEM  Volume 58, Number 10, October 2016 961

Copyright © 2016 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited
Techera et al JOEM  Volume 58, Number 10, October 2016

heavy and repetitive tasks are more prone to muscular fatigue.14 a shift. In fact, they argue that sleep patterns before work are a better
Although muscular and mental fatigue can be distinguished from indicator of occupational fatigue than actual observations at work.
one another, they are often experienced in concert to varying
degrees and proportions. Thus, in literature and in professional Mental Exertion
practice, fatigue is often modeled and discussed in a general form. Some causes of fatigue can be purely mental. Mental exer-
When possible, we will distinguish between muscular and mental tion, defined as sustained cognitive activity that requires extraordi-
fatigue for precision; otherwise, we will refer to fatigue as a nary mental effort, is the principal cause of mental fatigue.26
general condition. According to the National Aeronautics and Space Administration
(NASA) task load index, tasks are susceptible to mental exertion
CAUSES OF FATIGUE when they require very high levels of mental demand, have sig-
In order to build a comprehensive systems model of fatigue, nificant consequences associated with failure, require significant
implied causal factors were identified, based on previous research, effort on behalf of the individual, and involve frustration.27 In
and the relationships among them were modeled in an organized addition, workers in jobs that require immediate cognitive responses
fashion. To increase the internal validity of the resulting systems (eg, emergency situations), repetition, and problem-solving that
model, our review is limited to research-based studies that provide requires resources beyond rote memory are particularly vulnerable.
empirical evidence for conclusions and we focus our attention on Air traffic control operators, first-responders, and engineers who
causal factors and relationships that have been validated by multiple support space flight27 are examples of occupations with high levels
studies. In addition, to reduce complexity and avoid repetition, of cognitive demand.28 The impact of mental fatigue in these
factors that share similar root causes were codified into single, industries can be compounded by decreased or interrupted sleep
fundamental causal factor that is aptly labeled. It should be noted as described in the Sleep Deprivation section.
that this initial systems model includes all empirical evidence even Mental fatigue can be recognized by the individuals them-
if the data reported in the study did not have sufficient detail to be selves as they experience difficulty in focusing attention, making
included in the subsequent meta-analysis. This was done to illustrate timely decisions, planning, and responding to stimuli.29– 33 In
the focal areas of past fatigue research. addition to subjective indicators of mental fatigue, Lorist et al26
used brain scanning technology, such as the electro encephalogram,
Sleep Deprivation to directly measure mental lassitude and unequivocally correlated
Sleep deprivation is the most commonly discussed and mental exertion with mental fatigue.
arguably the most significant cause of both muscular and mental
fatigue. It affects between 15% and 20% of US adults, and is Muscular Exertion
becoming an increasing concern in our modern society due to Muscular exertion, a purely physical occurrence, is a primary
various schedule irregularities and conflicting priorities that com- cause of localized fatigue.12,13,34 The link between muscular exer-
promise sleep.15–17 tion and localized muscular fatigue has been studied in numerous
Belenky et al18 produced a comprehensive analysis of the ways. For example, fatigue is caused by the repetitive contraction of
fatigue-related consequences of different sleep routines. In particu- a specific muscle through electrical impulses (dynamic exertion), or
lar, they studied three different sleep restriction conditions: a mild by keeping a specific muscle contracted (static exertion).35,36
sleep restriction (7 hours of sleep per night), a moderate sleep Beyond the static and dynamic state of exertion, there is either a
restriction (5 hours), and a severe sleep restriction (3 hours or less). metabolic or central nervous system degradation that explains the
The results were compared with control group, which obtained onset of muscular fatigue. For example, a fatigued quadriceps muscle
8 hours of sleep during the course of the study. They discovered that consumes much more energy than a relaxed muscle, although meta-
sleeping less than 8 hours generated negative outcomes. In particu- bolic energy production remains unchanged.37 This results in expo-
lar, those who experienced only mild or moderate sleep deprivation nential cumulative fatigue generation.
showed a reduction in their mental and physical performance, which Although mental and muscular exertion yield obvious and
stabilized after a few days but still was under the normal level of distinct impacts on mental and muscular fatigue, respectively, other
performance. Furthermore, those who experienced severe sleep factors affect both mental and muscular capabilities. Some of these
deprivation showed an ongoing degradation that did not stabilize factors interact producing a compounding effect on the development
even after a few days. Several authors agree that a single night of of general fatigue.
partial sleep deprivation will generate a reduction in performance
that can last several days.19,20 Interestingly, Belenky et al18 noted no Work Load Characteristics
significant improvements in fatigue for workers who obtain more Assigned occupational work load, defined as the product of
than 8 hours of sleep compared with the control group. physical demand and time, has been strongly linked to fatigue.38
Typically, sleep is compromised by early morning and night Ribet and Derriennic39 found that heavy workloads negatively
shifts that not only disrupt the circadian rhythm of the individual but impact sleep and, thus, interfere with the recovery process accumu-
also affect their social and family life.21 Such adverse effects are lating fatigue debt in the individual. In addition, high levels of
most prevalent the first night after a period of rest when typical work required workload without the option for lower-effort alternatives
shifts recommence.22 In addition to fatigue, sleep disruption causes cause an effort-to-reward imbalance that can ultimately lead to
impaired physical performance, decreased work satisfaction, mental fatigue.40,41 Gander et al9 found that time on task alone was
increased stress, and interrupted future sleep patterns.6,23 Unless inadequate for fatigue prediction as accurate predictions require
this cycle is broken and workers recover completely from the consideration of time on task, cognitive and physical demands,
previous work day, the sleep deprivation process can become repetition, and scheduled breaks; among other factors.
chronic, ultimately resulting in serious long-term mental and
physical health problems.24 Overtime and Long Work Hours (LWH)
Sleep quantity and quality is important, as the mind and body As defined by the Fair Labor Standards Act (1938), overtime
of human beings need rest to avoid lassitude.25 Sleep deprivation is in the US is the amount of work time in a week that exceeds
seen as such a strong precursor of fatigue that Dawson et al25 claim 40 hours. Overtime can be accomplished either by working more
that fatigue can be accurately predicted by simply measuring the than 8 hours per day in a five-day-per-week schedule or by having
amount of sleep that a worker experiences in the 24 to 48 hours before fewer days off from work per week.

962 ß 2016 American College of Occupational and Environmental Medicine

Copyright © 2016 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited
JOEM  Volume 58, Number 10, October 2016 Causes and Consequences of Occupational Fatigue

The occurrence of overtime is a common problem of indus- substance abuse or physically and mentally demanding recreational
trialized environments.42 More specifically, in Europe, about 20% activities.9,25 Other common threats to recovery time are irregular or
of the working population experiences at least 5 hours of overtime split shifts in which a worker has between 2 and 4 hours break before
weekly and 13% of the full-time employees in Europe work at least returning to work. Such breaks are usually insufficient to effectively
10 hours of overtime a week.43 In the US, the situation is similar recover from fatigue because of common commute times.
where about 26% of the male working population and 11% of the
female working population work at least 10 hours of overtime At-Work Environmental Factors
weekly. In addition, the amount of overtime has been growing over Noise, light intensity, vibration, and temperature are all
the past 50 years, especially among women due to their increasing environmental factors linked to fatigue.53 Noise has been found
presence in the workplace.44 to cause fatigue through overstimulation.54,55 For example, Kjell-
The fatigue-related consequences of overtime are directly berg et al56 showed that airplane mechanics and boat patrol crews
connected to the perceived benefits and conditions under which experienced mental fatigue as a result of high duration of exposure
overtime takes place. In this regard, there are three different types of to low frequency noise. In addition, extreme temperatures (ie, below
overtime situations: 35F and above 95F) increase the rate of muscular fatigue.57,58
Similarly, Park and Gotoh59 found a significant positive correlation
1. Mandatory, low reward, low autonomy, and highly demanding between poor lightning and fatigue and Jiao et al60 found a positive
overtime. In this case, the worker is forced to work overtime in correlation between vibration and fatigue.
an environment that the worker cannot control, required to
perform highly demanding activities, or does not feel that the Social Environment at the Workplace
increased payment or future time off are sufficient compen- Most of the working population spends one-third to one-half
sation. This kind of overtime is associated with the most severe of their waking hours at work. Relationships with coworkers,
mental fatigue, dissatisfaction with the job, job burnout, negative managers, and subordinates play a major role in the development
work/home interaction, and slow recovery.45,46 of mental fatigue.61 Some specific psychosocial experiences can
2. Mandatory, high-reward overtime. The second kind of overtime exacerbate mental fatigue. For example, workplace abuse, defined
is that which is performed solely because the extra payment is as daily harassment from coworkers or abusive supervision for
desired. This kind of overtime encourages people to work up to a prolonged periods of time, can cause severe mental fatigue.62,63
point that exceeds their healthy state of functioning in order to Similarly, Hardy et al64 found that conflicts and adversarial situ-
earn more money, causing relatively high levels of fatigue.9 ations at work develop high levels of mental fatigue, even if they are
3. Voluntary overtime. This overtime is not driven solely by the essential attributes of the work type (eg, negotiations). Other social
economic reward but also by personal fulfillment. When it is factors that strongly influence the onset of mental fatigue are the
limited by the workers themselves, and it is performed with a perception of low autonomy, high emotional demands, and job
high level of autonomy, overtime does not appear to have insecurity.47
significant negative consequences, causing very low levels of
both muscular and mental fatigue within natural physiological Emotional Predisposition and Distress
limitations.47 Park et al48 concluded that 20 hours of overtime Fatigue impacts each individual differently depending on
per week is the point after which fatigue recovery becomes their emotional predisposition and is often exacerbated under dis-
extremely difficult, even when overtime is voluntary. tress. Emotional disposition pertains to the propensity of an indi-
vidual to experience negative emotions such as fear and sadness.65
Extended shifts can also cause fatigue. In contrast to Bültmann et al61 performed a field study with 11,020 workers from
overtime, which involves an accumulated number of work-hours 45 different companies, and discovered that those who are more
in a week, extended shifts or long working hours (LWHs) refer to likely to experience negative emotions are more likely to encounter
single work shift that exceeds 8 hours of work. It is important to mental fatigue than individuals who are more emotionally resilient.
notice that LWH do not necessarily imply overtime. Some sched- Furthermore, in a study of 16,139 employees from the public sector,
ules such as 10-hour shifts for 4 days a week involve LWH but Ala-Mursula et al66 showed that the onset of fatigue is accelerated
not overtime. under stressful conditions, especially for individuals who are
LWH studies showed that memory efficiency decreases emotionally vulnerable.
during cognitive activities when uninterrupted time on task exceeds Table 1 summarizes these key causal factors of fatigue and
3.5 hours.49 Once 3.5 hours of continuous work have been accumu- provides the references associated with each factor.67–72 In addition,
lated, mental acuity can only be maintained at the expense of extra when possible, the impacts of each factor on mental, muscular, and
mental exertion. The prolongation of such effort will cause mental general fatigue are identified.
fatigue.49 LWH also causes fatigue by exposing the individual to
other at-work stressors such as noise, inadequate lighting, extreme RELATIONSHIPS AMONG CAUSES OF FATIGUE
temperatures, and other environmental and social factors, when Causal factors of fatigue can compound one another when
present, for an additional period of time.48,50 The fatigue impact of they co-occur. Interestingly, many of the interrelationships
LWH depends on the activity performed. If the activity requires described in the literature relate to the organization of work by
mental effort, it will generate mental fatigue; otherwise, it will management. For example, work load, overtime, and extended shifts
generate muscular fatigue or a combination of both. tend to interact with each other and with other causal factors. Thus,
these managerial factors can be addressed through mindful
Incomplete Recovery work organization.
Recovery is the process of reversing the negative effects of
mental and muscular exertion to return to a pre-fatigued state. Acute Impacts of Increased Work Load on Other Causal
fatigue occurs when there is inadequate time to rest and recover Factors
from a work period.12,51,52 Incomplete recovery can result from As work load increases with additional or more demanding
inadequate time off from work, misclassifying nonrecovery time as activities, the extra mental and physical exertion can accelerate
recovery time (eg, considering commute time as recovery), and lassitude and, when extreme, it can lead to subsequent sleep
inadequate use of the given recovery periods by the worker such as deprivation.38 These consequences can affect all types of workers;

ß 2016 American College of Occupational and Environmental Medicine 963

Copyright © 2016 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited
Techera et al JOEM  Volume 58, Number 10, October 2016

TABLE 1. Causes of Fatigue


Factor Factor Definition Impacted Factor References
6,15,16,18–24,67
Sleep deprivation Sleep deprivation refers to a loss in the amount of consecutive hours of sleep. GEN
There are different degrees of sleep deprivation, which Belenky et al
classify as mild SD (7 h of time in time in bed), moderate SD (5 h of TIB),
severe SD (3 h of TIB), or total SD (no sleep at all).
18,26,49,70
Mental exertion Sustained cognitive activity that requires extraordinary mental effort. (Lorist MEN
et al 2005; Meijman and Schaufeli 1996)
12,13,34–37,68,69
Muscular exertion Exhaustion of the muscle due to an extent period of sustained tension or MUS
repetitive activity.
40
Work load High physical or mental demands at work. MEN
45
GEN
49
Overtime and LWH As defined by (FLSA 1938) ‘‘overtime’’ is the amount of time worked that MEN
45–48,50
exceeds 40 h of work a week. LWH is defined as working more than GEN
8 hours on a single shift.
25
Incomplete recovery Recovery is the process of reverting or reversing the negative effects of job GEN
demand to return to a pre work state.
56,71
Work environment Noise, light intensity, vibration, and temperature are all environmental factors MEN
53,57,58,66
linked to fatigue. GEN
61,62,64,72
Social environment The quality and characteristics of worker relationships with peer and MEN
supervisors, as well as the perceived freedom at work.
61,65
Emotional predisposition Emotional disposition pertains to the level of fear, stress, or overall attitude a MEN
worker has toward a certain task or job.

GEN, general fatigue; MEN, mental fatigue; MUS, muscular fatigue.

however, women below 49 years of age who come from a high consequences of fatigue can be connected via mechanisms other
socioeconomic status are particularly vulnerable.45 In addition, Van than fatigue, particularly with mental disorders and long-term health
Der Hulst and Geurts45 demonstrated that when workload is concerns. However, this review did not include these secondary
increased because of additional physical demands, the effects could relationships.
remain resident for up to a day after the exertion period. Below is a comprehensive review of empirical literature that
With regard to mental fatigue, Boksem and Tops73 found that has linked occupational fatigue with specific mental and physical
increased work load is a significant mental stressor when the reward consequences. Some of these consequences are subtle or acute,
received by the worker is perceived as insufficient to compensate for while others can manifest on delay or can be chronic. Understanding
the extra effort required. Furthermore, Ribet and Derriennic39 the interrelationships among consequences of fatigue is critical to
discovered that the pressure to hurry is the principal psychosocial modeling the true impact that fatigue can have on an individual.
occupational risk factor affecting sleep deprivation.
Mood Changes
Impacts of Overtime and LWH on Other Causal Mood, related to short-term emotions and feelings, can be
Factors assessed by numerous variables such as tension, anger, vigor,
The synergistic effects of overtime and LWH magnify the sadness, anxiety, and depression. Mood is typically measured
effects of fatigue caused by other factors. In fact, most of the subjectively through the Profile of Mood States.78 According to
aforementioned factors as discussed by the respective researchers the prevailing literature, fatigue results in strong negative changes to
in the context of extended periods of work or exertion, implicitly mood states, which can change emotional stability, increase anxiety,
suggesting that LWH or overtime are pre-requisite factors. and eventually lead to long-term depression.6,79–84 In addition, a
When the work requirements are mandated and workers have negative mood or emotional state has been shown to decrease
low autonomy, lengthened work periods become especially complex cognitive performance, clouding risk perception.85
problematic.21,48,50,74,75 Scott et al82 studied the mood-related consequences of
When workers are required to work overtime or LWH, there fatigue arising from 30 hours of sleep deprivation and concluded
is a natural reduction in recovery because fewer hours are available that tension and anger were not affected but that feelings of
to rest. On many occasions, overtime and LWH are combined with depression and sadness manifested quickly, even after only one
night or early morning shifts that disrupt circadian rhythms and missed sleep cycle. Mikulincer et al86 studied the timing of mood-
interfere with family life.15–17,21,74,76 This affects the individual related change arising from fatigue and found that fatigued indi-
physically, emotionally, and mentally, slowing down the recovery viduals experienced relatively high levels of anxiety between 0400
process.16,21,45,46,77 and 0800 hours and relatively low levels between 1600 and
Table 2 identifies and provides references for research of the 2000 hours. Interestingly, using a series of complex and controlled
interrelationships among the causal factors of fatigue. experiments, Kajtna et al84 found that the fatigue-induced mood
changes compromise an individual’s ability to concentrate on
CONSEQUENCES OF FATIGUE a stimulus.
Fatigue research tends to focus on either causal factors or
outcomes of fatigue. Causal factors and outcomes are rarely Cognitive Degradation
modeled together. Here, implied causes and consequences of fatigue Cognitive functions, classified as complex or basic, are
are analyzed exclusively underlying their relationship with fatigue essential for effective completion of work and for healthy life
as the main variable of interest. Occasionally, some causes and outside of work. Complex cognitive functions include, but are

964 ß 2016 American College of Occupational and Environmental Medicine

Copyright © 2016 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited
JOEM  Volume 58, Number 10, October 2016 Causes and Consequences of Occupational Fatigue

mentally demanding tasks, lassitude of concentration and attention


TABLE 2. Relationships Among Causes of Fatigue
are experienced similarly by a fatigued individual regardless of
Factor Impacted Factor References causal factors. In more recent years, fatigue has been studied in
different naturalistic tasks such as driving and conflict tests.13,93 As
Work load Sleep deprivation 38,39 hypothesized, these studies demonstrated that fatigue decreased
40
Mental exertion processing of stimulus, concentration capacity, and made individ-
38
Muscular exertion uals more prone to error.
46
Overtime and LWH
15–17,21,74,76
Overtime and LWH Sleep deprivation
21,48–50
Reaction Time
Mental exertion When reacting to a stimulus, the human body must perceive,
21,50
Muscular exertion
16,21,45,46,77 process, and physically react to the stimulus. Previously, we dis-
Incomplete recovery
cussed the reduced attention capacity resulting from fatigue. In
addition, reaction time, defined as the time between a stimulus onset
and the response to such stimulus, is highly compromised
not limited to, the ability to plan, perceive risk, and make decisions by fatigue.
under uncertainty. Basic cognitive functions include attention, The prevailing method for measuring reaction time involves
vigilance, and response to stimuli.87 When studying the impact using the psychomotor vigilance test (PVT), which records the
of fatigue on cognitive performance, researchers typically manip- reaction time to a visual stimulus.94 Angus and Heslegrave88 used
ulate degrees of sleep deprivation and measure performance with PVT to study the impact of fatigue on reaction time and found that
standardized cognitive activities.19,20,67,88 the reaction time of fatigued workers increased by 24% when
Studies on complex cognitive performance involve sophis- compared with well-rested workers. Furthermore, higher levels
ticated experimental methods and the use of advanced technologies. of fatigue induced by a second night without sleep resulted in a
Researchers have focused on the impacts of acute fatigue on verbal 57% increase in reaction time. Similar studies found more modest
fluency and communication, decision-making capability, creative but still significant increases in reaction time.95,96
thinking, planning, executive control, and novelty performance.
Overwhelmingly, data support that all of these complex cognitive Physical Degradation and Pain
functions are significantly deteriorated even with modest fatigue The physical degradation caused by fatigue can be significant
levels.6,89 and pervasive. For example, Johnston et al68 tested static and
With the help of electroencephalographic (EEG) tools, func- dynamic balance of fatigued individuals and found drastic increases
tional neuroimaging, and psychophysiological assessments, in the risk of falls compared with nonfatigued individual. In addition
researchers have found that the prefrontal cortex (PFC), which to reduced physical stability, fatigue can result in reduced muscular
governs the executive functions of the brain, is the locus of strength causing slower movement and increased number of
deterioration when an individual is fatigued.29,90 Unlike basic errors.34 The long-term impacts of physical degradation primarily
functions that can be rote and automatic, complex tasks require involve musculoskeletal disorders of the neck, shoulders, and
uncompromised PFC functionality.89,90 Thus, the brain is highly back.14 Acute, reduced balance was observed in fatigued subjects
susceptible to even modest fatigue levels when executive functions for at least 15 minutes after prolonged exertion of cervical muscles,
are required to make complex decisions.49,83,91 In an occupational leading to longer-term localized muscular fatigue.69,97 The implica-
environment, this translates to a differential in potential con- tion is that seemingly short-term effects of acute fatigue can remain
sequences depending on the tasks performed by different workers resident for longer periods of time than may be intuitive.
with the same levels of fatigue.
In addition to complex cognitive functions, other more basic Illnesses
and automatic cognitive functions can be compromised. One of the Fatigue has also shown to have significant long-term impacts
earliest and most comprehensive investigations of the relationship on the human body. For example, researchers have found evidence
between fatigue and cognitive performance found that a single night that night-shift workers experience higher levels of fatigue and,
of sleep deprivation decreased cognitive performance by 30%.88 consequently, present gastrointestinal disorders, irregular bowel
Furthermore, their findings showed that cognitive performance was activity such as constipation and diarrhea, bowel pain, and even
reduced by an additional 30% during the second 24-hour period of ulcers at a rate that is nearly double their day-worker counter-
sleep deprivation and remained stable at 40% of the original parts.82,98,99 Furthermore, Steenland and Fine,100 Bøggild and
cognitive performance after 48 hours without sleep. This study Knutsson,101 and Knutsson99 found that shift workers present a
was empirically validated by Van Dongen and Dinges,19 Lim and 40% higher risk of suffering cardiovascular disease or myocardial
Dinges,20 and Zhang and Liu67 who later replicated the experiment infarction than day-time workers. Although gastrointestinal and
with a more heterogeneous sample of participants and a more cardiovascular conditions are the most common occupational ill-
diverse set of cognitive tasks and fatigue scenarios, finding that nesses related to fatigue,102 some postulate that fatigue has greater
cognitive degradation arising from a single night of total sleep consequences among those who are pregnant or have diabetes.99
deprivation fluctuates between a 10% and a 30%.
Human Error and Injuries
Attention and Concentration As previously discussed, lassitude of cognitive functions
The PFC helps an individual to concentrate attention on a task decrease recall, reaction time, ability to plan, and ability to respond
or stimulus. As the functionality of the PFC is compromised by to stimuli. In addition, lassitude of the muscular system decreases
fatigue, both static and dynamic functions are affected. Researchers the ability to physically execute a plan. Thus, it is not surprising that
have studied the impact of fatigue on concentration and attention for researchers have found evidence of a direct relationship between the
many years, starting with the seminal work with 17,000 participants degree of fatigue and rates of human error.7,29,103,104 Human error is
by Yoshitake.92 In this field study, three different dimensions of said to occur when an individual takes unintentional actions or is
fatigue were identified: drowsiness and dullness, inability to con- unable to execute planned functions and an undesired outcome is
centrate, and physical discomfort. Even though the inability to realized. This was exemplified in the well-publicized Chernobyl and
concentrate was experienced most strongly by workers with Three-Mile Island accidents wherein human errors contributed to

ß 2016 American College of Occupational and Environmental Medicine 965

Copyright © 2016 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited
Techera et al JOEM  Volume 58, Number 10, October 2016

the events, including breaking protocol by disabling emergency illustrates the strength and direction of the effect sizes among
cooling systems, turning off automatic blocking devices, and not these factors.
following standard operating procedures during emergency
response. Although not always connected to fatigue, these errors MODEL LIMITATIONS
and violations prove to be more common after when workers are The systems model has numerous limitations. First, the
fatigued.29,103 model is built completely from past research and the structure
Table 3 summarizes these key outcomes of fatigue. They are and content of the model was derived from what has been studied
briefly described and references are provided for each factor.105–114 in the past. Thus, the model reflects any limitations associated with
In addition, for consistency with Table 1, the consequences of the topics of investigation to date. Second, the completeness of the
fatigue are linked to the corresponding causal type of fatigue. model is commensurate with the completeness of the literature
review. Our analysis mainly considered research published in
RELATIONSHIPS AMONG OUTCOMES OF FATIGUE English in top occupational safety and health, applied psychology,
There are some consequences of fatigue that exacerbate other cognitive psychology, medical, industrial systems, engineering,
outcome factors, which can magnify or perpetuate the effects of sleep, and neuroscience journals. To ensure completeness, we also
acute fatigue. These factors are most often cited in medical, cross-referenced all references of past papers to ensure that popular
neuroscience, and psychology research because they are related studies were not omitted. Nevertheless, it is possible that some
to the function (or dysfunction) of the human body. research was missed and that the model could be improved. Third,
we made the implicit assumption that the mechanics of fatigue are
Impacts of Cognitive Degradation on Other the same regardless of ethnic background, culture, and occupation.
Outcomes The body of research affirms this assumption; however, it remains
As cognition degrades, a plethora of negative outcomes arise. just that an assumption that has yet to be validated. Fourth, our
From a mental standpoint, cognitive fatigue reduces complex and review and the model only include the direct causes and con-
basic cognitive function, negatively affects mood, and increases sequences of fatigue. There may be a plethora of factors that
reaction time. From primary impacts of fatigue arise a host of indirectly affect fatigue or are an indirect outcome of fatigue
secondary outcomes. For example, weakened basic and complex through some mediating variable. These interactions are out of
cognitive functions increase human error, injuries, and decrease the the scope of this research. Fifth, the model does not include direct
quality and productivity of work completed.29,93,115 Also, mood connections between causal and outcome factors without fatigue as
changes often cause the affected worker to become more irritable, a mediator. Future researchers may wish to study these direct
thereby degrading workplace, social, and family relationships.116 relationships as a means to identify when outcomes are falsely
Finally, a delayed reaction time increases the rate of human error attributed to fatigue and when other mediating variables may be at
and increases safety risk.6,93 It should be noted that most authors play. Finally, our descriptions of each causal factor and outcome of
agree that basic cognitive functions are more resilient to fatigue than fatigue were abbreviated. There is a wealth of information on
complex cognitive functions and mood. However, even modest research methods, nuances of findings, and implications that we
levels of fatigue can affect basic cognitive functions. simply could not include within the scope of one paper. Rather, our
aim was to produce a single-source systems model of the many
Impacts of Physical Degradation on Other antecedents and consequences of fatigue and to acknowledge the
Outcomes key research used to build a valid model. The Meta-Analysis
Approach section offers a formal statistical analysis of published
After an intense or prolonged period of physical stress,
data in order to measure and model the effect sizes of the relation-
human bodies experience muscular fatigue. This physical degra-
ships shown in Fig. 1.
dation is accompanied by a less noticeable, but equally important,
reduction in motor function34,68 that slows physical reactions. The META-ANALYSIS APPROACH
results include decrease balance and decrease coordination, making
As our main objective was to summarize the body of knowl-
a worker prone to human errors and injuries.30,34 Dramatic
edge related to the causes and consequences of occupational fatigue,
reductions in balance in particular have shown to result in errors
we complemented the systems model with a formal analysis of the
and injuries.97
magnitude and direction of the relationships. To achieve this
objective, we performed a statistical meta-analysis. When formally
Impacts of Illnesses on Other Outcomes performed, a meta-analysis allows one to codify the effect sizes
When fatigue contributes to illness, the effects of the illness, from multiple studies into a single statistic. As effect sizes are
in turn, often reinforce fatigue thereby causing a downward cycle. independent of the sample size and have a reportable standard error,
Jason et al12 found that between 5% and 8% of the general they can be compared and aggregated across multiple studies. Such
population experience the fatigue-illness cycle for 1 to 5 months. analysis allowed us to identify the factors that have the largest
impact on fatigue and the factors that are most impacted by fatigue.
SYSTEMS MODEL We used four basic steps in our meta-analysis as suggested
After analyzing the causes and consequences of fatigue, we by Lipsey and Wilson118 and Field and Gillett.119 The first step,
aimed to represent the findings in a systems model to illustrate studies selection, involved a comprehensive literature review of
the strengths and gaps in the current body of knowledge. Figure 1 studies that reported sufficient empirical data to calculate an effect
illustrates the complexity of the relationships between causes and size. In addition, in this step, each study is evaluated according to
consequences of fatigue, where fatigue is the mediating factor. certain criteria of quality and methodology appropriates for their
The relationships among factors are represented by arrows and inclusion in the meta-analysis. Although each study included in the
the volume of research developed in that line of inquiry is systems model reported empirical evidence, many studies did not
represented by the thickness of the line. Therefore, it becomes report sufficient detail of their analysis and, thus, were not included
easy to distinguish mature areas of research from those where in the meta-analysis. The second step, studies coding, involved
additional research is needed. All the information presented in the extracting the data needed to compute an effect size for each
model is a summary of the information presented in Tables 1 to relationship of interest. The result of this step is a matrix of effect
4.117 It should be noted that the subsequent meta-analysis sizes reported by all scientific studies for each relationship. The

966 ß 2016 American College of Occupational and Environmental Medicine

Copyright © 2016 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited
JOEM  Volume 58, Number 10, October 2016 Causes and Consequences of Occupational Fatigue

TABLE 3. Consequences of Fatigue


Fatigue Impacted Factor Impacted Factor Description References
7,26,29–32,85,88,94,95,96,105
MEN Cognitive degradation Cognitive functions refer to both complex and basic cognitive
functions. Complex cognitive activities are not the result of a
natural impulse or a learned behavior, as these activities
require a high degree of processing information in order to be
performed (Boonstra et al 2007). Basic cognitive functions
have been defined as attention, concentration, reaction time,
and well learned behavior. (Jovanović et al 2012)
45,106–108
Illnesses Illnesses refer to the affections that the individual can develop as
a result of an extended period of fatigue. Fatigue debilitates
our immune system and causes stress that opens the door for
health problems.
7,29,30
Error or Injuries Error or Injuries refers to any kind of procedure or action that is
not perform in the correct way, due to fatigue, and that
causes some kind of loss or damage to the workers.
23,79–84,86,95
GEN Bad mood Mood comprehends numerous variables such as tension, anger,
vigor, anxiety, and depression. An important number of
studies utilize the Profile of Mood State to assess mood.
6,7,13,14,18–20,23,29,49,83,88–93,95,96,97
Cognitive degradation See above
14,34,92,109
Physical degradation Physical degradation refers to any bodily symptom that impairs
the normal functioning of the body, such as pain,
somnolence, and lack of strength.
38,53,98–102,110–114
Illnesses See above.
4,21,50,113
Error or Injuries See above
34,68,69
MUS Physical degradation See above
106
Illnesses See above

GEN, general fatigue; MEN, mental fatigue; MUS, muscular fatigue.

third step, standardization, involves translating the effect sizes as a relative continuous score and studies that reported objective
between studies into common units so that they can be size measures such as reaction time in a PVT were naturally continuous.
aggregated. Finally, the fourth step, overall effect size computation, When the raw data were available, the following process was used.
involves computing a single effect size that combines the stand-
ardized data from multiple studies. The details associated with each X̄2  X̄1
Cohen0 s d ¼ (1)
step are described below. S pooled
Studies Selection
Where X̄1 is the expected value of the control group sample,
In order to be included in the meta-analysis, a study must (1)
X̄2 is the expected value of the treatment group sample, and Spooled is
contain results about a clear and specific relationship between a
the pooled standard deviation.
cause of fatigue and fatigue, or fatigue and a consequence of fatigue;
We used the method introduced by Lipsey and Wilson118 to
(2) involve a healthy working sample of subjects between 18 and 65
pool the standard error.
years of age and represent the working population (ie, studies
limited to athletes, children, sick, or disabled persons were omitted); sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
(3) include either an effect size or enough information to compute an n1 þ n2 d2
SE d ¼ þ (2)
effect size; and (4) involve five participants or more. Of the 105 n1 n2 2ðn1 þ n2 Þ
studies of occupational fatigue that report empirical data, 23 met all
four criteria.
It was also important to ensure that only valid and reliable Where SEd represents the standard error of the statistic d; n1 is
the sample size of the control group, n2 is the sample size of the
data were used in the analysis. From the 23 studies, a total of 64
effect sizes were obtained, including correlation coefficients (r), risk treatment group, and d is the effect size (Cohen d).
ratios (RRs), and odds ratios (ORs). However, not all these effect In order for a study to be included in our sample, the effect
sizes were acceptable for the analysis because they were not size and the standard deviation had to be reported. Sometimes these
variables are directly presented in a paper; other times, these
accompanied by proper demographic statistics (eg, sample size
of the control group). This restriction yielded a total of 29 inde- variables can be extracted from the results of a t test, an F-test,
pendent effect sizes. or an analysis of variance (ANOVA), to mention a few. As the effect
size of small samples has been shown to be biased by outliers, we
Studies Coding used a sample size correction introduced by Hedges.121 This
The coding strategy to extract effect sizes from a study correction, shown in the equation below, reduces the relative impact
depended on the target aggregate statistic. As all of the variables of studies that report sample sizes less than 20. When sample sizes
are greater than 20, the adjustment is nullify. The equation presented
of interest could be converted to a continuous scale and the standard
deviations could be pooled, Cohen d is the ideal statistic.120 Cohen d by Hedges121 is as follows:
yields information about the difference in means of two groups,  
3
measured in standard deviations. We were able to use Cohen d d0 ¼ 1  d (3)
because studies that reported the results of a questionnaire could be 4N  9

ß 2016 American College of Occupational and Environmental Medicine 967

Copyright © 2016 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited
Techera et al JOEM  Volume 58, Number 10, October 2016

For the case wherein RRs and ORs were reported with their
TABLE 4. Relationships Among Consequences of Fatigue
standard error, the statistics were standardized into a Cohen d using
Impacted Factor the following equation:
Factor Name References pffiffiffi
lnðORÞ 3
13,18 d OR ¼ (6)
Cognitive performance Illnesses p
7,16,24,30,49,70,83,89,90,93,103,104
Error or Injuries
14,97,98,117
Physical degradation Illnesses where dOR represents the Cohen d effect size calculated from
7,29,34,68,69,103,104
Error or Injuries
an OR.
Ultimately, this process yielded all affect sizes expressed as a
Cohen d with a corresponding 95% CI as suggested by Lipsey and
Wilson.118
where d’ represents the unbiased effect size and N is the total
sample (n1þn2).
Overall Effect Size Calculation
Standardization In this last step, the effect sizes for multiple studies
In some cases, effects sizes were reported by authors using a investigating the same relationship were aggregated into an
statistic other than Cohen d (eg, regression coefficient). These overall effect size. There are two basic ways of performing this
statistics were transformed into a Cohen d and a corresponding aggregation: the homogeneous effect meta-analysis and the ran-
confidence interval (CI) of 95% was created. When a correlation dom effect meta-analysis. The homogeneous effect meta-analysis
coefficient was reported, Cohen d was calculated using the follow- is only valid for exact replications of an experiment. Here, there
ing formula: were few replications and far more studies that used a variety of
methods and sample sizes. Thus, we used the random effect meta-
2r analysis. This method not only accounts for variability within a
d r ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffi (4)
1  r2 sample due to chance but also accounts for true variability among
samples from different studies. This ensured that each effect size
where dr represents the Cohen d effect size calculated from a was weighted according to the quality of the study from which it
correlation coefficient (r). was taken. Furthermore, random effect meta-analysis does not
The standard error was calculated as the standard error of a allow any study to dominate. The inverse variance weight method
Fisher distribution and then translated back with the hyperbolic was employed and each effect size (d) was weighted by the
function. This value was also transformed into a d effect size. inverse of its variance. An overall effect for each relationship was
calculated using Eq. (7). In some cases, when only one study
1 reported an effect size for a relationship, the effect was converted
SE Z r ¼ pffiffiffiffiffiffiffiffiffiffiffiffi (5) to Cohen d and reported for such relationship.
n3
P
wi d i
where SEZr represents the standard error of a correlation coefficient ES d ¼ P (7)
in a Fisher distribution and n is the sample size. wi

Emotional
Predisposition
Social
Environment

Cognitive
Mental Degradation
Work Exertion
Environment
MENTAL
!!
FATIGUE
Work Error
Overtime Load Injury
LWH
Bad
Sleep
Deprivation
!!
FATIGUE Mood Illnesses

MUSCULAR
Incomplete !!
FATIGUE Physical
Recovery Degradation
Muscular
Exertion

FIGURE 1. Systems model illustrating empirical literature on the relationships among causes and effects of fatigue. Dashed lines
represent the inter-relationships among the causes and consequences of fatigue and solid lines represent the immediate
relationships between causes, fatigue, and consequences. Line thickness corresponds directly to the number of studies devoted
to each relationship.

968 ß 2016 American College of Occupational and Environmental Medicine

Copyright © 2016 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited
JOEM  Volume 58, Number 10, October 2016 Causes and Consequences of Occupational Fatigue

TABLE 5. Effects of Causes of Fatigue


Relationship Source n.t. n.c. Cohen d 95% CI P
105,122
Sleep Deprivation – General Fatigue 15 15 1.24 1.50 3.98 0.188
29,49
Mental Exertion – Mental Fatigue 49 45 0.91 0.48 1.33 0.000
38
Workload – General Fatigue 18828 18828 0.25 0.23 0.28 0.000
45
Overtime and LWH – Mental Fatigue 325 325 0.43 0.11 0.75 0.000
38,45,46,48
Overtime and LWH – General Fatigue
57
Environment – General Fatigue 7 7 2.83 1.30 4.36 0.000
61,72
Social Environment – Mental Fatigue 30 30 0.30 0.19 0.42 0.000
11020 11020

One-tailed P values.
95% CI, 95% confidence interval.

where ESd represents the overall effect size, Wi stands for each where W 0i represents the random effect weights for each effect size,
individual effect size weight, and di denotes each individual SEd is the standard error of the statistic d, and t2 stands for the
effect size. random effect variance.
The standard error was computed using Eq. (8): With this new set of weights, the random effect sizes are
sffiffiffiffiffiffiffiffiffiffiffiffiffi calculating by applying Eq. (7) and substituting Wi by W 0i.
1
SE ES d ¼ P (8)
Wi RESULTS OF THE META-ANALYSIS
The application of our meta-analysis method yielded the
where SE ES d represents the overall effect size standard error and Wi results summarized in Tables 5 to 8.122,123 Note that when only one
stands for each individual effect size weight. study is reported on a specific relationship, the Cohen d in the tables
represents the effect size of that single study. In these tables, the
The equations presented above are sufficient for the fixed
column heading ‘‘n.t’’ represents the sample size of the treatment
effect meta-analysis; however, in order to compute a random effect
meta-analysis, three additional equations are required to evaluate group, ‘‘n.c’’ represents the sample size of the control group, and
95% CI represents the upper and lower bounds of the 95% CI. The
the homogeneity, random effect variance, and random effect
relationships derived from this meta-analysis are depicted in a
weights using Eqs. 9, 10, and 11, respectively.
revised systems model in Fig. 2. Here, the line thickness corre-
X sponds directly to Cohen d, the aggregated effect size.
Q¼ wi ðd i  ES d Þ2 (9)
DISCUSSION OF THE META-ANALYSIS
where Q represents the homogeneity statistic, di denotes each In a meta-analysis, an effect between 0 and 0.20 is considered
individual effect size, and ESd represents the overall effect size. small; between 0.20 and 0.50 is considered medium; and over 0.50,
The homogeneity statistic is used to calculate the variance preferable closer to 0.80, is considered large.120 Although most
associated with true differences among different studies, such results achieved levels of high statistical significance, some were
variance receives the name of random effect variance (t2) as shown weak, either because they did not achieve significance or because
in Eq. 10. their CI was too wide, even including negative values. For example,
the effect of sleep deprivation on fatigue was not statistically
QdfQ significant in our meta-analysis because, despite the large effect
t2 ¼ P P 2 (10) sizes reported, the CI for this relationship includes the possibility of
W
Wi  P i Wi a null effect. This was observed because one study in this domain
reported results that were not statistically significant.122 Intuitively,
some may think that a study that does not report statistical signifi-
where t2 represents the random effect variance, Q is the homogen-
cance should be discarded; however, a meta-analysis of all past
eity statistic, dfQ represents the degrees of freedom of Q, and Wi
research results must be considered for a relationship as long as they
stands for each individual effect size weight.
were obtained through high-quality methods.
With this random effect variance, a new set of weights for
The largest significant effect, the relationship between
each effect size can be calculated using Eq. 11.
environmental conditions at work such as noise, vibration, and
1 temperature and fatigue, was observed by González-Alonso
W 0i ¼ (11) et al57 in a single study. Such a strong finding with little previous
SE d þ t 2

TABLE 6. Effects Among Causes of Fatigue


Relationship Source n.t. n.c. Cohen d 95% CI P
45
Overtime – Incomplete recovery 323 323 0.65 0.36 0.95 0.000

One-tailed P value.
95% CI, 95% confidence interval.

ß 2016 American College of Occupational and Environmental Medicine 969

Copyright © 2016 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited
Techera et al JOEM  Volume 58, Number 10, October 2016

TABLE 7. Effects of Fatigue


Relationship Source n.t. n.c. Cohen d 95% CI P
29,31
Mental Fatigue – Cognitive Degradation 48 44 2.27 2.59 7.13 0.180
106
Mental Fatigue – Illnesses 484 484 0.35 0.22 0.48 0.000
30
Mental Fatigue – Error and Injury 36 32 0.42 0.09 0.93 0.052
82,84
General Fatigue – Bad Mood 15 15 0.86 0.10 1.81 0.039
23,97
General Fatigue – Cognitive Degradation 1039 1039 1.00 0.91 1.09 0.000
111,114
General Fatigue – Illnesses 594 1328 0.03 0.07 0.13 0.271
50,97
General Fatigue – Error and Injury 10,804 10,805 0.38 0.05 0.81 0.042
69
Muscular Fatigue – Physical Degradation 16 16 0.99 0.25 1.73 0.004
106
Muscular Fatigue – Illnesses 232 232 0.41 0.00 0.82 0.000

One-tailed P values.
95% CI, 95% confidence interval.

TABLE 8. Effects Among Consequences of Fatigue


Relationship Source n.t. n.c. Cohen d 95% CI P
123
Illnesses – Cognitive Degradation 298 60 1.97 1.65 2.28 0.000

One-tailed P value.
95% CI, 95% confidence interval.

validation indicates the strong need for scientific replication. Other fatigue and these effects can be caused by a plethora of external
medium effects include the impact of mental exertion, overtime, and factors as well. The influence of fatigue on mood has been charac-
LWH on mental fatigue and the impact of social environment and terized as important with a large effect size, (d ¼ 0.86; 95%
workload on general fatigue. CI ¼ 0.10 to 1.81), but a distended CI diminishes confidence in
Only one study provided useful information about relation- this effect as zero is included within the bounds.
ships among causes of fatigue,45 showing a strong relationship
between overtime, LWH, and incomplete recovery. This study CONCLUSIONS AND RECOMMENDATIONS
confirms that overtime plays an important role as inhibitor of a The systems models presented in Fig. 1 clearly show that
proper recovery but, again, should be replicated by future research- there is much less research on the system of causal factors of fatigue
ers to confirm the effects. than on the system of outcomes of fatigue. This may be due to the
Most impacts of fatigue on physiology proved to be large. fact that the outcomes of fatigue are more readily identifiable and
This was especially true for the effect of general fatigue among measurable than the causal factors, which can be more difficult to
cognitive degradation (d ¼ 1.00; 95% CI ¼ 0.91 to 1.09) and mus- empirically measure. Also, the psychological, health, and medical
cular fatigue among physical degradation (d ¼ 0.99; 95% CI ¼ 0.25 research community that deals with the health and wellness aspects
to 1.73). This is not surprising given the relatively large body of of fatigue is quite large in comparison to the industrial systems,
knowledge focused on this domain. safety and health, and human factors research community that often
The effect sizes of the relationship between fatigue and error deals with the occupational causes of fatigue.
or illness were moderate to low, but with weak significance. These With respect to the causes of fatigue, Fig. 1 clearly reveals
results make sense, as errors and injuries are indirect results of acute some knowledge gaps that may be useful for framing new research.

Social
0.35
Environment
0.43
0.42*
1.97
Incomplete
Mental 0.91
Recovery
Exertion
MENTAL 2.27* Cognitive
0.65
Overtime and FATIGUE Degradation
LWH Work 0.25 1.00
Load
0.46
Illnesses
FIGURE 2. Systems model represents
0.86 * Bad Error or
FATIGUE
Mood Injury effect sizes (Cohen’s d values) for each
2.83
Work
Environment 0.30*
relationship. Arrows indicate direction
Sleep MUSCULAR and line thickness corresponds directly
Deprivation FATIGUE 0.03*
to the magnitude of the effect size from
0.99
Physical the meta-analysis. The () indicates a
Degradation
95% CI that includes zero or nonsigni-
ficance for such an effect size.

970 ß 2016 American College of Occupational and Environmental Medicine

Copyright © 2016 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited
JOEM  Volume 58, Number 10, October 2016 Causes and Consequences of Occupational Fatigue

For example, there is a wide range of research about the way that 14. Lipscomb JA, Trinkoff AM, Geiger-Brown J, Brady B. Work-schedule
characteristics and reported musculoskeletal disorders of registered nurses.
fatigue arising from sleep deprivation impacts workers, and also Scand J Work Environ Health. 2002;28:394–401.
about the way that irregular and extended shifts impact workers. 15. Webb WB, Agnew Jr HW. Are we chronically sleep deprived? Bull Psychon
However, there is a clear need for additional research about the way Soc. 1975;6:47–48.
that workload impacts muscular exertion and the recovery process. 16. Åkerstedt T. Psychological and psychophysiological effects of shift work.
Both have a logical foundation that could be supported through Scand J Work Environ Health. 1990;16:67–73.
rigorous research. These are simply a sample of obvious gaps in the 17. Bliwise DL, King AC, Harris RB, Haskell WL. Prevalence of self-
knowledge system. The reader may find additional connections reported poor sleep in a healthy population aged 50-65. Soc Sci Med.
upon which to build a future study. 1992;34:49–55.
With respect to the consequences of fatigue, even though 18. Belenky G, Wesensten NJ, Thorne DR, et al. Patterns of performance
degradation and restoration during sleep restriction and subsequent recov-
there is an abundant body of literature addressing the relationship ery: a sleep dose-response study. J Sleep Res. 2003;12:1–12.
between general fatigue and mood changes, there is no significant 19. Van Dongen HPA, Dinges DF. Sleep circadian rhythms, and psychomotor
research that investigates the relationship between exclusively vigilance. Clin Sports Med. 2005;24:237–249.
muscular or mental fatigue and mood changes. In addition, during 20. Lim J, Dinges DF. Sleep deprivation and vigilant attention. Ann N Y Acad
our research, we also found it strange that there is no substantial Sci. 2008;1129:305–322.
research about the way in which mood changes affect work 21. Folkard S, Tucker P. Shift work, safety and productivity. Occup Med.
relationships, which seems to be a reinforcing process. The relation- 2003;53:95–101.
ship between mood and cognitive functions is ripe for both con- 22. Roach GD, Fletcher A, Dawson D. A model to predict work-related fatigue
firmatory and seminal investigation. based on hours of work. Aviat Space Environ Med. 2004;75:A61–A69.
The meta-analysis of the causes and consequences of fatigue 23. Philibert I. Sleep loss and performance in residents and nonphysicians: a
illustrated in Fig. 2 revealed significant gaps in empirical evidence meta-analytic examination. Sleep. 2005;28:1392–1402.
of fatigue-related factors. In fact, only 23 studies provided sufficient 24. Tepas DI, Mahan RP. The many meanings of sleep. Work Stress. 1989;3:
93–102.
empirical data. Furthermore, there is a dearth of replication among
25. Dawson D, McCulloch K. Managing fatigue: it’s about sleep. Sleep Med
fatigue studies, indicating a strong need for scientific validation. Rev. 2005;9:365–380.
Future research is especially needed to confirm proposed relation- 26. Lorist MM, Boksem MA, Ridderinkhof KR. Impaired cognitive control and
ships among causal factors and among outcomes of fatigue. As the reduced cingulate activity during mental fatigue. Cogn Brain Res.
medical field has examined the relationship between fatigue and 2005;24:199–205.
physiology, the impact of fatigue on other, less direct outcomes such 27. Hartfirst, Sandra G, Staveland, Lowell E. Development of NASA-TLX
as work performance, is needed. (Task Load Index): results of empirical and theoretical research. Adv
This paper adds to the efforts of previous researchers to help Psychol. 1988;52:139–183.
formalize the study of fatigue among the working population. It can 28. Rubio S, Diaz E, Martin J, Puente JM. Evaluation of subjective mental
workload: a comparison of SWAT, NASA-TLX, and workload profile
serve as a point of departure for future research to evaluate what has methods. Appl Psychol. 2004;53:61–86.
been done, where extra research is needed, and how to organize such 29. van der Linden D, Frese M, Meijman TF. Mental fatigue and the control of
research to contribute to the enlargement of the current body cognitive processes: effects on perseveration and planning. Acta Psychol
of knowledge. (Amst). 2003;113:45–65.
30. van der Linden D, Frese M, Sonnentag S. The impact of mental fatigue on
REFERENCES exploration in a complex computer task: rigidity and loss of systematic
1. Kroenke K, Price RK. Symptoms in the community: prevalence, classifi- strategies. Hum Factors J Hum Factors Ergon Soc. 2003;45:483–494.
cation, and psychiatric comorbidity. Arch Intern Med. 1993;153:2474–2480. 31. Boksem MAS, Meijman TF, Lorist MM. Effects of mental fatigue on
attention: an ERP study. Cogn Brain Res. 2005;25:107–116.
2. Ricci JA, Chee E, Lorandeau AL, Berger J. Fatigue in the U.S. workforce:
prevalence and implications for lost productive work time. J Occup Environ 32. Boksem MA, Meijman TF, Lorist MM. Mental fatigue, motivation and
Med Am Coll Occup Environ Med. 2007;49:1–10. action monitoring. Biol Psychol. 2006;72:123–132.
3. Reiner BI, Krupinski E. Demystifying occupational stress and fatigue 33. van der Linden D, Eling P. Mental fatigue disturbs local processing more
through the creation of an adaptive end-user profiling system. J Digit than global processing. Psychol Res. 2006;70:395–402.
Imaging. 2011;25:201–205. 34. Grandjean E. Fatigue in industry. Br J Ind Med. 1979;36:175–186.
4. Harrington JM. Health effects of shift work and extended hours of work. 35. Christensen H, Søgaard K, Jensen BR, Finsen L, Sjøgaard G. Intramuscular
Occup Environ Med. 2001;58:68–72. and surface EMG power spectrum from dynamic and static contractions. J
5. Meijman TF, Schaufeli WB. Psychische vermoeidheid en arbeid [Occu- Electromyogr Kinesiol. 1995;5:27–36.
pational fatigue]. De Psycholoog. 1996;31:236–243. 36. Edwards RG, Lippold OCJ. The relation between force and integrated
6. Harrison Y, Horne JA. The impact of sleep deprivation on decision making: electrical activity in fatigued muscle. J Physiol. 1956;132:677–681.
a review. J Exp Psychol Appl. 2000;6:236–249. 37. Edwards RH, Hill DK, Jones DA. Heat production and chemical changes
during isometric contractions of the human quadriceps muscle. J Physiol.
7. Lorist MM, Boksem MAS, Ridderinkhof KR. Mental fatigue and task
1975;251:303–315.
control: planning and preparation. Psychophysiology. 2000;37:614–625.
38. Åkerstedt T, Fredlund P, Gillberg M, Jansson B. A prospective study of fatal
8. Hallowell MR. Worker fatigue: managing concerns in rapid renewal high-
occupational accidents-relationship to sleeping difficulties and occu-
way construction projects. Prof Saf. 2010;55:18–26.
pational factors. J Sleep Res. 2002;11:69–71.
9. Gander P, Hartley L, Powell D, et al. Fatigue risk management: organiz-
39. Ribet C, Derriennic F. Age, working conditions, and sleep disorders: a
ational factors at the regulatory and industry/company level. Accid Anal
longitudinal analysis in the French cohort E.S.T.E. V Sleep J Sleep Res Sleep
Prev. 2011;43:573–590. Med. 1999;22:491–504.
10. Aaronson LS, Teel CS, Cassmeyer V, et al. Defining and measuring fatigue.
40. Boksem MAS, Tops M, Kostermans E, De Cremer D. Sensitivity to
J Nurs Scholarsh. 1999;31:45–50.
punishment and reward omission: evidence from error-related ERP com-
11. Piper BF. Fatigue: current bases for practice. Management of Pain, Fatigue ponents. Biol Psychol. 2008;79:185–192.
and Nausea. Macmillan Education UK, 1989;187–198.
41. Hsiao H, Simeonov P. Preventing falls from roofs: a critical review.
12. Jason LA, Evans M, Brown M, Porter N. What is fatigue? Pathological and Ergonomics. 2001;44:537–561.
nonpathological fatigue. PM&R. 2010;2:327–331.
42. Kodz J, Davis S, Lain D, et al. Working long hours: a review of the evidence.
13. Brown ID. Driver fatigue. Hum Factors J Hum Factors Ergon Soc. (Employment Relations ResearchSeries ERRS16). London, United King-
1994;36:298–314. dom: Institute for Employment Studies; 2003.

ß 2016 American College of Occupational and Environmental Medicine 971

Copyright © 2016 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited
Techera et al JOEM  Volume 58, Number 10, October 2016

43. Pascal P, Damien M. Third European survey on working conditions 2000. 70. Okogbaa OG, Shell RL, Filipusic D. On the investigation of the neuro-
European Foundation for the Improvement of Living and Working Con- physiological correlates of knowledge worker mental fatigue using the EEG
ditions; 2001. signal. Applied Ergonomics. 1994;25:355–365.
44. Caruso CC. Possible broad impacts of long work hours. Ind Health. 71. Melamed S, Bruhis S. The Effects of Chronic Industrial Noise Exposure on
2006;44:531–536. Urinary Cortisol, Fatigue, and Irritability: A Controlled Field Experiment. J
45. van der Hulst M, Geurts S. Associations between overtime and psycho- Occup Environ Med. 1996;38:252–256.
logical health in high and low reward jobs. Work Stress. 2001;15:227–240. 72. Maslach C, Jackson SE. Burnout in organizational settings. Applied Social
46. Beckers DGJ, et al. Voluntary or involuntary? Control over overtime and Psychology Annual. 1984;5:133–153.
rewards for overtime in relation to fatigue and work satisfaction. Work 73. Boksem MAS, Tops M. Mental fatigue: costs and benefits. Brain Res Rev.
Stress. 2008;22:33–50. 2008;59:125–139.
47. Taris TW, Beckers DGJ, Dahlgren A, Geurts SAE, Tucker P. Overtime work 74. Axelsson J, Kecklund G, AAkerstedt T, Lowden A. Effects of alternating
and well-being: Prevalence, conceptualization and effects of working 8-and 12-hour shifts on sleep, sleepiness, physical effort and performance.
overtime. In: McIntyre S, Houdmont J, editors. Occupational health Scand J Work Environ Health. 1998;24(suppl 3):62–68.
psychology: European perspectives on research, education and practice. 75. van der Hulst M, Veldhoven M, van Beckers D. Overtime and need for
Nottingham University Press; 2007. 2:21-40. recovery in relation to job demands and job control. J Occup Health.
48. Park J, Kim Y, Chung HK, Hisanaga N. Long working hours and subjective 2006;48:11–19.
fatigue symptoms. Ind Health. 2001;39:250–254. 76. Baulk SD, Fletcher A, Kandelaars KJ, Dawson D, Roach GD. A field study
49. Meijman TF. Mental fatigue and the efficiency of information processing in of sleep and fatigue in a regular rotating 12-h shift system. Appl Ergon.
relation to work times. Int J Ind Ergon. 1997;20:31–38. 2009;40:694–698.
50. Dembe AE, Erickson JB, Delbos RG, Banks SM. The impact of overtime 77. De Wolff C, Drenth PJD, Henk T. A Handbook of Work and Organizational
and long work hours on occupational injuries and illnesses: new evidence Psychology: Volume 2: Work Psychology. 2. East Sussex, UK: Psychology
from the United States. Occup Environ Med. 2005;62:588–597. Press; 2013.
51. Swaen GMH, Amelsvoort LGPM, van Bültmann U, Kant IJ. Fatigue as a 78. McNair DM, Lorr M, Droppleman LF. Profile of mood states. San Diego,
risk factor for being injured in an occupational accident: results from the CA: Educational and Industrial Testing Service; 1981.
Maastricht Cohort Study. Occup Environ Med. 2003;60:i88–i92. 79. Pilcher JJ, Huffcutt AJ. Effects of sleep deprivation on performance: a meta-
52. Beurskens AJHM, Bültmann U, Kant I, Vercoulen JHMM, Bleijenberg G, analysis. J Sleep Res Sleep Med. 1996;19:318–326.
Swaen GMH. Fatigue among working people: validity of a questionnaire 80. Dinges DF, Pack F, Williams K, et al. Cumulative sleepiness, mood
measure. Occup Environ Med. 2000;57:353–357. disturbance and psychomotor vigilance performance decrements during a
53. Krause N, Lynch J, Kaplan GA, Cohen RD, Goldberg DE, Salonen JT. week of sleep restricted to 4-5 hours per night. Sleep J Sleep Res Sleep Med.
Predictors of disability retirement. Scand J Work Environ Health. 1997; 1997;20:267–277.
23:403–413. 81. Sagaspe P, et al. Effects of sleep deprivation on Color-Word, Emotional, and
54. Landström U, Lundström R. Changes in wakefulness during exposure to Specific Stroop interference and on self-reported anxiety. Brain Cogn.
whole body vibration. Electroencephalogr Clin Neurophysiol. 1985; 2006;60:76–87.
61:411–415. 82. Scott JP, McNaughton LR, Polman RC. Effects of sleep deprivation and
55. Landström U, Löfstedt P. Noise, vibration and changes in wakefulness exercise on cognitive, motor performance and mood. Physiol Behav.
during helicopter flight. Aviat Space Environ Med. 1987;58:109–118. 2006;87:396–408.
56. Kjellberg A, Muhr P, Skoldstrom B. Fatigue after work in noise: an 83. Boonstra TW, Stins JF, Daffertshofer A, Beek PJ. Effects of sleep depri-
epidemiological survey study and three quasi-experimental field studies. vation on neural functioning: an integrative review. Cell Mol Life Sci.
Noise Health. 1998;1:47–55. 2007;64:934–946.
57. González-Alonso J, Teller C, Andersen SL, Jensen FB, Hyldig T, Nielsen B. 84. Kajtna T, Štukovnik V, Grošelj LD. Effect of acute sleep deprivation on
Influence of body temperature on the development of fatigue during concentration and mood states with a controlled effect of experienced stress.
prolonged exercise in the heat. J Appl Physiol. 1999;86:1032–1039. Zdr Vestn. 2011;80:354–361.
58. Zivin JG, Neidell MJ. Temperature and the Allocation of Time: Implications 85. Tixier AJ-P, Hallowell MR, Albert A, van Boven L, Kleiner BM. Psycho-
for Climate Change. Cambridge, MA: National Bureau of Economic logical antecedents of risk-taking behavior in construction. J Constr Eng
Research; 2010. Manag. 2014;140:1–10.
59. Park S, Gotoh S. Research on office environment evaluation and worker 86. Mikulincer M, Babkoff H, Caspy T, Sing H. The effects of 72 hours of sleep
fatigue. J Therm Biol. 1993;18:565–570. loss on psychological variables. Br J Psychol. 1989;80:145–162.
60. Jiao K, Li Z, Chen M, Wang C, Qi S. Effect of different vibration 87. Jovanović M, Sporiš G, Šopar J, Harasin D, Matika D. The effects of basic
frequencies on heart rate variability and driving fatigue in healthy drivers. military training on shooting tasks in conditions of sleep deprivation.
Int Arch Occup Environ Health. 2004;77:205–212. Kinesiology. 2012;44:169–177.
61. Bültmann U, Kant I, Schröer C, Kasl S. The relationship between psycho- 88. Angus RG, Heslegrave RJ. Effects of sleep loss on sustained cognitive
social work characteristics and fatigue and psychological distress. Int Arch performance during a command and control simulation. Behav Res Methods
Occup Environ Health. 2001;75:259–266. Instrum Comput. 1985;17:55–67.
62. Tepper BJ. Consequences of abusive supervision. Acad Manage J;. 89. Harrison Y, Horne JA. One night of sleep loss impairs innovative thinking
2000;178–190. and flexible decision making. Organ Behav Hum Decis Process.
63. Zapf D, Gross C. Conflict escalation and coping with workplace bullying: a 1999;78:128–145.
replication and extension. Eur J Work Organ Psychol. 2001;10:497–522. 90. Jones K, Harrison Y. Frontal lobe function, sleep loss and fragmented sleep.
64. Hardy GE, Shapiro DA, Borrill CS. Fatigue in the workforce of national Sleep Med Rev. 2001;5:463–475.
health service trusts: levels of symptomatology and links with minor 91. Nilsson JP, et al. Less effective executive functioning after one night’s sleep
psychiatric disorder, demographic, occupational and work role factors. J deprivation. J Sleep Res. 2005;14:1–6.
Psychosom Res. 1997;43:83–92. 92. Yoshitake H. Three characteristic patterns of subjective fatigue symptoms.
65. St Clair Gibson A, Baden DA, Lambert MI, et al. The conscious perception Ergonomics. 1978;21:231–233.
of the sensation of fatigue. Sports Med. 2003;33:167–176. 93. Fogt DL, Kalns JE, Michael DJ. A comparison of cognitive performance
66. Ala-Mursula L, Vahtera J, Linna A, Pentti J, Kivimäki M. Employee decreases during acute, progressive fatigue arising from different concur-
worktime control moderates the effects of job strain and effort-reward rent stressors. Mil Med. 2010;175:939–944.
imbalance on sickness absence: the 10-town study. J Epidemiol Community 94. Dinges DF, Powell JW. Microcomputer analyses of performance on a
Health. 2005;59:851–857. portable, simple visual RT task during sustained operations. Behav Res
67. Zhang NZ, Liu H-TL. Effects of sleep deprivation on cognitive functions. Methods Instrum Comput. 1985;17:652–655.
Neurosci Bull. 2008;24:45–48. 95. Opstad PK, Ekanger R, Nummestad M, Raabe N. Performance, mood, and
68. Johnston R, Howard M, Cawley, Posse G. Effect of lower extremity clinical symptoms in men exposed to prolonged, severe physical work and
muscular fatigue on motor control performance. Med Sci Sports Exerc. sleep deprivation. Aviat Space Environ Med. 1978;49:1065–1073.
1998;30:1703–1707. 96. Naitoh P. Circadian cycles and restorative powers of naps. In: Johnson LC,
69. Yaggie J, Armstrong WJ. Effects of lower extremity fatigue on indices of Tepas DI, Colqhoun WP, Colligan MJ, editors. Biol Rhythms Sleep Shift
balance. Balance. 2004;312:2004. Work. New York: SP Medical and Scientific Books; 1981. p. 553–580.

972 ß 2016 American College of Occupational and Environmental Medicine

Copyright © 2016 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited
JOEM  Volume 58, Number 10, October 2016 Causes and Consequences of Occupational Fatigue

97. Gosselin A, De Koninck J, Campbell KB. Total sleep deprivation and 111. Alfredsson L, Karasek R, Theorell T. Myocardial infarction risk and
novelty processing: implications for frontal lobe functioning. Clin Neuro- psychosocial work environment: an analysis of the male Swedish working
physiol. 2005;116:211–222. force. Soc Sci Med. 1982;16:463–467.
98. Segawa K, et al. Peptic ulcer is prevalent among shift workers. Dig Dis Sci. 112. Hansen J. Increased breast cancer risk among women who work predom-
1987;32:449–453. inantly at night. Epidemiology. 2001;12:74–77.
99. Knutsson A. Health disorders of shift workers. Occup Med. 2003;53:103–108. 113. Nylén L, Voss M, Floderus B. Mortality among women and men relative to
100. Steenland K, Fine L. Shift work, shift change, and risk of death from heart unemployment, part time work, overtime work, and extra work: a study
disease at work. Am J Ind Med. 1996;29:278–281. based on data from the Swedish twin registry. Occupat Environ Med.
2001;58:52–57.
101. Bøggild H, Knutsson A. Shift work, risk factors and cardiovascular disease.
Scand J Work Environ Health. 1999;25:85–99. 114. Liu Y, Tanaka H. Overtime work, insufficient sleep, and risk of non-fatal
acute myocardial infarction in Japanese men. Occupat Environ Med.
102. Iwasaki K, Sasaki T, Oka T, Hisanaga N. Effect of working hours on 2002;59:447–451.
biological functions related to cardiovascular system among salesmen in a
machinery manufacturing company. Ind Health. 1998;36:361–367. 115. Craig A, Cooper RE. Symptoms of acute and chronic fatigue. In: Smith AP,
Jones DM, editors. Handbook of Human Performance, Vol.3: State and
103. Craig A, Cooper RE. Symptoms of acute and chronic fatigue. Handbook of Trait. San Diego, CA: Academic Press Inc; 1992. 3:289-339.
human performance. Academic Press Ltd. 1992;3:289–339.
116. Williams KJ, Alliger GM, Role Stressors, Mood Spillover. and perceptions
104. Czeisler CA, Kronauer RE, Allan JS, et al. Association of sleep-wake habits of work-family conflict in employed parents. Acad Manage J. 1994;37:
in older people with changes in output of circadian pacemaker. Lancet. 837–868.
1992;340:933–936.
117. Knutsson A. Health disorders of shift workers. Occupat Med. 2003;53:
105. Gillberg M, Åkerstedt T. Sleep restriction and SWS-suppression: effects on 103–108.
daytime alertness and night-time recovery. J Sleep Res. 1994;3:144–151.
118. Lipsey MW, Wilson DB. Practical Meta-Analysis. ix. Applied social
106. Fredriksson K, Alfredsson L, Köster M, et al. Risk factors for neck and
research methods series Vol 49. Thousand Oaks, CA: Sage Publications,
upper limb disorders: results from 24 years of follow up. Occupat Environ
Inc; 2001.
Med. 1999;56:59–66.
119. Field AP, Gillett R. How to do a meta-analysis. Br J Math Stat Psychol.
107. Amick III BC, McDonough P, Chang H, Rogers WH, Pieper CF, Duncan G.
2010;63:665–694.
Relationship between all-cause mortality and cumulative working life
course psychosocial and physical exposures in the United States labor 120. Cohen J. Statistical Power Analysis for the Behavioural Sciences. New
market from 1968 to 1992. Psychosom Med. 2002;64:370–381. York: Lawrence Erlbaum Associates; 1988.
108. Kivimäki M, Leino-Arjas P, Luukkonen R, Riihimäi H, Vahtera J, Kirjonen 121. Hedges LV. Distribution theory for Glass’s estimator of effect size and
J. Work stress and risk of cardiovascular mortality: prospective cohort study related estimators. J Educ Behav Stat. 1981;6:107–128.
of industrial employees. Brit Med J. 2002;325:857–861. 122. Herscovitch J, Broughton R. Performance deficits following short-term
109. Nindl BC, Leone CD, Tharion WJ, et al. Physical performance responses partial sleep deprivation and subsequent recovery oversleeping. Can J
during 72 h of military operational stress. Med Sci Sports Exerc. Psychol Can Psychol. 1981;35:309.
2002;34:1814–1822. 123. Vercoulen JHMM, Swanink CMA, Fennis JFM, Galama JMD, van der Meer
110. Xu X, Ding M, Li B, Christiani DC. Association of rotating shiftwork with JWM, Bleijenberg G. Dimensional assessment of chronic fatigue syndrome.
preterm births and low birth weight among never smoking women textile J Psychosom Res. 1994;38:383–392.
workers in China. Occupat Environ Med. 1994;51:470–474.

ß 2016 American College of Occupational and Environmental Medicine 973

Copyright © 2016 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited