E

F

1

ecTfitLt

otactr

atsamapcssvi

0d

Accident Analysis and Prevention 45S (2012) 1– 5

Contents lists available at SciVerse ScienceDirect

Accident Analysis and Prevention

j ourna l h o mepage: www.elsev ier .com/ locate /aap

ditorial

atigue research in 2011: From the bench to practice

a b s t r a c t

Over the last 20 years, academic, industry and community stakeholders have been meeting at a biennialscientific conference to discuss fatigue-related research and policy in the transportation, resources andhealth sectors. During this period, the research conducted around the world has progressed substantially:we now better understand the basic processes of sleep and circadian physiology that underpin perfor-mance; we better understand that fatigue risk management in the absence of any discussion about sleepis fruitless at worst and inadequate at best; and we are improving the capacity of models and other tech-nologies to assist us to predict, monitor, identify, minimise and mitigate fatigue-related risk. At the same

time however, the relationship between performance on simple cognitive tasks in laboratory settingsand performance on complex tasks required to operate efficiently and safely in the workplace, remains astumbling block. This special issue brings together fifteen papers that cover the range of areas in the fieldof fatigue research and challenges us as researchers, regulators, industry representatives and communitymembers to continue the work of managing the risk of fatigue.

century regulatory and industrial policy architectures remain our

. Introduction

In this special edition of Accident Analysis and Prevention, theditors have chosen a group of papers presented at a recent fatigueonference held in March 2011 in Fremantle, Western Australia.hese papers showcase recent developments in fatigue science andatigue-risk mitigation in the transportation, resources and healthndustries. The conference is one of the longest running interna-ional conferences focused on fatigue. It was first convened byaurence Hartley in 1992 and has been held every 2–3 years sincehen in either Australia or the United States.

The conference remains relatively unique as it brings together allf the key stakeholders in the area to discuss fatigue in a collabora-ive multi-disciplinary forum. Scientists, industry representativesnd regulators share their thirst for knowledge and discuss theauses and consequences of fatigue as well as the latest regula-ory and technological developments for reducing fatigue-relatedisk.

The special issue presents 15 papers that illustrate the growthnd diversity of research in the field over the last two decades. Sincehe first meetings of this group nearly 20 years ago, our under-tanding of the basic science of fatigue has increased significantlynd community attitudes have changed fundamentally. We have aore detailed understanding of the basic physiology of the sleep

nd circadian systems and the effects of fatigue on simple taskerformance have been well documented. We have an emergingapacity to model the way in which work schedules shape sub-equent sleep and wakefulness and the effect on sleepiness and

imple task performance. Most recently, we have developed someery promising technologies that enable us to estimate monitor,dentify and protect fatigued workers who are at risk.

001-4575/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.oi:10.1016/j.aap.2011.09.016

© 2011 Elsevier Ltd. All rights reserved.

During those 20 years of research, community, government andregulatory attitudes to fatigue have changed significantly, espe-cially within the developed world. In 1992, when the conferencefirst convened, many in the general population struggled to under-stand what the fuss was about. Regulatory authorities often sawfatigue as an industrial issue rather than a safety concern. In thatera, it was still considered reasonable to address employee con-cerns about fatigue through the use of penalty payments and/orover-time premiums. Fatigue was seen primarily as one of the socialcosts of shift work.

Now, two decades on, fatigue is a clearly identified hazard underOH&S legislation and, in many developed countries, organisationsare routinely required to develop and implement a fatigue riskmanagement system with the same degree of rigour and proce-dural complexity as they would for long-accepted hazards such ascarcinogenic chemicals or manual handling.

On the other hand, some areas of the field have remainedremarkably static. Our understanding of the effects of fatigue oncomplex task performance at the individual or team level has notmoved forward anywhere near as fast. As a field, we continue toover-simplify the effects of fatigue. Many still equate real worldtask performance with simplistic measures of response times orhand eye co-ordination despite significant evidence to the contrary(Gander et al., 2008; Dawson et al., 2011a).

Similarly, many of our frameworks for controlling or regulat-ing fatigue-related risk remain essentially unchanged over thatsame period. Despite 30 years of research to the contrary, 19th

preferred tool for reducing fatigue-related risk in most industries.Rules originally designed to reduce physical fatigue by controllingshift and break times along with aggregate working hours continue

2 s and P

tvotiCI

otdwtiItw

tdtt(pf

2

rToDwo2(bo

gtltciaia

wtpewptpmoo

aab

Editorial / Accident Analysi

o dominate policy frameworks for fatigue. Indeed, it is only in theery recent past that the primacy of sleep and circadian physiol-gy in mediating fatigue-related risk (rather than work and restimes) has been acknowledged by those responsible for public pol-cy formulation (House of Representatives Standing Committee onommunications, Transport and the Arts, 2000; (IATA et al., 2011;

CAO, 2011).It is only with the wisdom of hindsight that one can look back

ver the 20-year history of this conference and see the secularrends in research and policy. It is only by taking a meta-view of theiscipline that we can see those areas where we have moved for-ard rapidly (e.g. bio-mathematical modeling and monitoring) and

he areas where we have remained at a standstill (e.g. understand-ng the link between fatigue and complex work place behaviors).t is against this backdrop that we have selected a group of papershat represent the most novel or the most intriguing papers thatere presented at the conference.

The editors have selected these papers because we believehey best represent the diversity of current research and policyevelopment. They illustrate the diversity of methodologies andechniques now being used in the discipline. And finally, they illus-rate the ways in which different cultures across the world havere)interpreted the diversity of scientific opinions and regulatoryractice to formulate their own unique approaches to reducingatigue related risk at the local level.

. Bio-mathematical models of fatigue

The first group of papers in this special issue is drawn from theapidly expanding field of bio-mathematical modeling of fatigue.his field first emerged in the early part of the last decade. Basedn the pivotal theoretical work of Borbely and Daan (Borbely, 1982;aan et al., 1984), several groups around the world developed soft-are that was used to estimate the likely levels of fatigue based

n the timing and duration of sleep and wakefulness (Mallis et al.,004; Van Dongen, 2004). In their initial developmental phase1997–2005), early fatigue models were parameterized using lab-ased data and produced outputs indicating the likely mean fatiguef a group of workers.

In the last few years the limitations of using lab-based data androup means has been identified as a significant issue reducingheir validity and limiting further expansion of their use in regu-atory frameworks (Dawson et al., 2011b). In this issue we presenthree recent studies that provide strong empirical evidence for theritique of early model use, provide new approaches that will signif-cantly increase the potential validity of bio-mathematical modelsnd give us a better understanding of the extent to which inter- andntra- individual differences will constrain the predictive reliabilitynd validity of these models.

Darwent et al. (2012) in their paper expand upon their recentork demonstrating the extent to which social factors modify the

iming and duration of sleep–wake behavior over and above theurely physiological (e.g. sleep and circadian factors). They havexpanded their initial work on using the timing and duration ofork schedules to predict sleep–wake behavior in long-haul airlineilots and to present a more general model that can be appliedo a broader set of work places and occupations. This work willrovide significant support for the idea that second-generation bio-athematical modeling can provide a significant improvement in

ur capacity to design schedules that provide an adequate sleeppportunity.

Van Dongen et al. (2012) present a very elegant paper teasingpart the relative contribution of individual differences in sleepnd circadian physiology, from the statistical variability in neuro-ehavioural performance while fatigued. Their data clearly show

revention 45S (2012) 1– 5

the relative contribution of individual differences in homeostaticand circadian processes to overall variability and the significantpotential to improve bio-mathematical models by ‘tuning’ themto the individual worker. More controversially, their paper hintsat the possibility that these differences may reflect genetic differ-ences at the physiological level. What is particularly exciting aboutthis line of investigation is the idea that these genetic differencesmight form the basis of identifying workers who are most at risk andensuring that they are protected from the adverse consequences offatigue-related error. Should this turn out to be the case, we havethe possibility of an entire new set of fatigue-risk managementtools.

The final paper in this section, by Dorrian et al. (2012) chal-lenges some of the ideas put forward by Van Dongen et al. (2012).The authors argue that individual differences, while present, maynot be consistent across time. That is they may be state ratherthan trait variables. Using long haul aviation data where the samepilots undertook the same routes and ‘lay-overs’ on different occa-sions, the authors clearly demonstrate that individual pilots will usedifferent sleep–wake strategies on different occasions. The impli-cation of these data is that the suggestion that bio-mathematicalmodels can be ‘tuned’ to individual differences to improve validitywhere the data have been derived from cross-sectional studies maybe premature. Considered together with recent studies of mathe-matical modeling and banking sleep (McCauley et al., 2009; Ruppet al., 2009), these data also hint at the idea that the variabilitycurrently attributed to individual differences may well reflect theuse of a limited set of sleep–wake strategies on different occasions.If this is indeed the case, it carries interesting implications for the‘genetic’ argument for explaining variability currently attributed toindividual differences (for review, see King et al., 2009).

3. Work/rest schedules

The second group of papers explores the link between workschedules, sleep–wake behavior and fatigue in work place settings.As discussed above, there is an increasing awareness of the discrep-ancies between the timing and duration of sleep–wake behavior inlaboratory environments and what happens in work place settings.A better understanding of the factors that modulate the decision togo to bed or get up and/or the recuperative value of sleep in dif-ferent settings will be critical if we are to continue to improve theways in which we regulate working time arrangements. Moreover,this type of data will be critical if we are to continue to improvethe validity and reliability of bio-mathematical models of fatigueacross a broad set of occupational and demographic settings.

In the first of these papers, Roach et al. (2012) report the effectsof early starts on the rosters of short-haul pilots. The data clearlyindicate that early starts reduce the amount of sleep obtained inthe previous 12 h and that early starts are associated with higherself-reported fatigue. Critically, the data demonstrate the extent towhich early starts can increase risk independent of the length of theproceeding flight and duty periods. This work supports the recentICAO initiatives (IATA et al., 2011; ICAO, 2011) and carries signifi-cant implications for air safety regulators around the world who donot yet incorporate time-of-day factors into the risk assessment ofworking time arrangements in aviation.

The complexity of the social and operational phenomena medi-ating sleep–wake behavior in pilots is further explored by Holmeset al. (2012). In their paper, ultra-long-range pilots travellingbetween Houston and Doha were monitored for sleep–wake pat-

terns, fatigue and compliance with FRMS recommendations onoptimum sleeping times while on layover. The most interest-ing aspect of this study is the fact that despite poor compliance(64%) with recommended sleeping times, pilots reported relatively

and P

not‘cbwdsdedci

swcaowqo

trtiafritidq

4

ohrcpefft2

Fadmdairg1(msc

Editorial / Accident Analysis

ormal levels of sleep per 24 h presumably through increased usef in-flight napping. The key lesson to be drawn from this study ishe refractory nature of human behavior in the field in response toexpert’ recommendations and the extent to which social factorsan over-ride the biological factors thought to drive sleep–wakeehavior. The implications of this study are profound for thoseorking in policy areas where guidelines and ‘expert’ recommen-ations form a central element of the fatigue risk managementystem. The data reinforce the importance of appreciating theiversity of work place strategies that exist and the extent to whichmployees will exploit behavioural plasticity to manage risk. Theata also indicate the extent to which we may have ‘privileged’hrono-biological perspectives and the need to develop guidelinesn conjunction with employees.

This theme is further explored by Paterson et al. (2012) in theirtudy of rail workers. In this study, they show that rail workersho have dependents gain less sleep in a given break than their

olleagues who do not. This result is not surprising; dependentsnd their demands will compete with sleep during non-work peri-ds. However, the study further reinforces the fact that there areide range of factors that can potentially influence the amount and

uality of sleep during break periods and the relative simplicity ofur current predictive models.

Baulk and Fletcher (2012) further reinforce this idea showinghat self-reported sleep quality and quantity in truck drivers on theoad and at home is different. The amount of sleep and its recupera-ive value is rated higher for home-based sleep than sleep obtainedn the cab. On the other hand the objective data obtained fromctigraphy and sleep diaries indicates that there are few if any dif-erences. Again, it is hardly surprising that truck drivers subjectivelyate home-based sleep better than in-cab sleep. What is fascinat-ng, however, is that there is little objective evidence in the data seto support this. This further suggests the importance of the phys-cal and socio-cultural sleep environment, not just the timing anduration of sleep, in determining subjective perceptions of sleepuality.

. Policy and practice

The third group of papers is related to the translation of a bodyf basic research into policy and practice. In recent years, regulatorsave begun to understand the complexity of the safety issues withespect to fatigue and the need to address them in a systematic andomprehensive manner (Gander et al., 2011). In recent years, manyolicy makers in the transport area have finally acknowledged thextent to which limits to working time arrangements over-simplifyatigue risk management. Indeed the recent ICAO policy is a pro-ound endorsement of the need for policy makers to step back andake a much broader view of fatigue risk management (IATA et al.,011; ICAO, 2011). This conference was no exception.

In the first of these papers, Cabon et al. (2012) outline a pilotRMS program for French regional airlines. What is proposed is

fundamental shift away from the more traditional flight anduty-time limitations (FTL) that have characterised fatigue-riskitigation in the past. They describe an approach that is multi-

imensional and includes a range of risk controls including thed-hoc use of bio-mathematical models to analyse schedules anddentify at-risk practices that are currently permitted under FDLegulations. Most importantly, they introduce a central idea fromeneral safety management theory to the FRMS process (Reason,990). That is, the idea of monitoring both systems performance

through the linkage of bio-mathematical models and incident/near

iss data) as well as individual performance (through web-basedurveys and in-flight observation of flight crew). This paper indi-ates that fatigue-related risk can be effectively excised from its

revention 45S (2012) 1– 5 3

roots in industrial and labor law and re-integrated into a moregeneral risk and safety management system. In addition, the papersuggests that airline operators and unions are willing to try suchapproaches and have the capacity to implement this type ofapproach to fatigue risk management successfully.

The second paper in this group is of particular note. Mohamedet al. (2012) present an analysis of the likely impacts of fatigueregulation on the bus industry in Malaysia. In response to commu-nity concerns about the elevated level of risk and reduced safetyassociated with overnight bus travel, there has considerable pres-sure to ban night-time bus travel to improve road safety. Theauthors report the results of a community consultation process anda systems-based analysis of the data. What is exceptional aboutthis paper is the sophisticated understanding of the complexityof introducing fatigue regulations into a complex socio-technicalsystem and the considerable likelihood that well intentioned but ill-informed regulation might produce unintended and/or paradoxicaloutcomes. This paper shows the down-stream complexity associ-ated with changes in transport policy and the need for regulatorsto understand the non-linear impact of policy in this domain. Thereis much in the analysis that those involved in policy regulation inmore developed countries could learn from an approach that rightlyconstructs fatigue-related risk as just one element of a complexnon-linear system where reductions in risk on one dimension of thesystem may increase overall risk through equilibration behaviourin other system variables.

5. Laboratory simulations

In the fourth group of studies, the authors present a set of labo-ratory studies utilizing sleep restriction under conditions of forceddesynchrony (Aschoff and Wever, 1981) to understand how sleeptime, sleep debt and the circadian system interact to mediate sub-jective fatigue, simple and complex task performance and foodchoice when people are sleep deprived. These papers were cho-sen for their methodological rigor and their capacity to tease apartthe relative contributions of each of the factors in mediating thechanges in a complex suite of dependent variables.

In the first of these papers, Ferguson et al. (2012) report the rel-ative effects of sleep duration, sleep debt and circadian phase onsubjective fatigue. In the second of these papers, Matthews et al.(2012) try to tease apart the relative effects on simulated drivingperformance and in the third study by Heath et al. (2012) the effecton snack preference was explored. In the first two studies, aside forthe expected main effects of sleep and circadian variables, the keyfinding of interest was the complexity of the relationship betweensleep, sleep debt and circadian phase on subjective fatigue andsimulated driving behavior. The interaction effects are not insub-stantial and these findings carry significant implications for thoseresponsible for formulating public policy in the area of fatigue. Inthe third study, Heath et al. (2012) provide some extremely noveldata separating the effects of sleep loss and circadian phase onfood choice. Along with the pioneering work of Van Cauter andcolleagues (Van Cauter et al., 1991; Spiegel et al., 1999) this studyshows further evidence that the sleep loss and circadian disruptionassociated with shift work disturb appetite regulation in ways thatmay predispose shift workers to obesity and the subsequent healthproblems associated with unhealthy weight gain.

6. Effects of fatigue

The final group of studies looks at novel effects of fatigue. In thefirst of these studies, Michael et al. (2012) report a potential salivarybio-marker for fatigue. This is a very exciting development as it isone of the first compounds reported that may have the potential

4 s and P

tmdrthpttebttTo

lfrewiaestfcooftot

tlebftHflfb(fcoftctRr

7

astai

Editorial / Accident Analysi

o form the basis of what many believe is the ‘holy grail’ in fatigueanagement – a biological test for fatigue similar to those used for

etecting drug and alcohol impairment. In this study the authorseport the presence of two small molecular weight peptides, onehat increases and one that decreases in response to moderate toigh intensity physical activity. It is important to note that theseeptides are produced in response to physical exercise rather thanime awake and presumably reflects muscular fatigue rather thanhe drowsiness or sleepiness associated with mental fatigue. Nev-rtheless there is the exciting possibility that these peptides maye associated with time awake as well as physical activity. Untilhese studies have been undertaken, it is impossible to rule outhat we may have finally discovered the fatigue equivalent of %BAC.his seems an area of research that would be just too tempting toverlook even if it appears a long shot.

The paper by Michael et al. (2012) was not the only paperooking at the potential to develop a more objective measure ofatigue. Several authors have previously suggested the use of postu-al balance tests as a measure of fatigue-related impairment (Moradt al., 2007). Such a test could be potentially used in the sameay the Romberg test had been used to determine alcohol related

mpairment in drunk drivers before breathalyzers were commonlyvailable (Romberg, 1853). Postural balance tests have not, how-ver, been validated for use in measuring fatigue impairment. In theecond study in this group of papers, Sargent et al. (2012) reporthe use of a postural balance measure to measure the effects ofatigue induced by sleep restriction in a forced desynchrony proto-ol. This would enable them to tease apart the relative contributionf sleep and circadian effects and provide some initial assessmentn the potential for postural balance to be used as measure ofatigue-related impairment. Unfortunately it would appear that inhis study it may not be suited for use as a road-side testing technol-gy. While the test was sensitive to differences in circadian phasehere were not significant effects due to prior wake.

The final study in this group is a very interesting study exploringhe ways in which fatigue might influence subtle aspects of air-ine pilot behavior and increase risk attributable to fatigue. Druryt al. (2012) present a study of 300-odd flight sectors in which pilotehavior and flight deck communications were recorded and ratedor the degree of Heightened Emotional Activity (HEA) in responseo threats. The authors reported that there were higher levels ofEA associated with reduced sleep in the 24 and 48 h prior toying. This paper shows the ways in which we are increasingly

ocusing on the very subtle ways in which fatigue influences ourehavior in ways that increase the probability of a latent errorReason, 1990). This marks a move away from the initial studies ofatigue on performance that have focused on relatively gross psy-hometric measures of impairment such as falling asleep, lapsesr delayed response times. This paper signals one of the excitinguture directions of fatigue related research – an understanding ofhe ubiquitous and complicated ways in which fatigue influencesomplex task performance in the work place. In this paper we seehe seminal work of safety theorists such as Helmreich (2000) andeason (1990) beginning to permeate the groundwork of earlieresearch in the field.

. Conclusions

Through a better understanding of the effects of sleep restrictionnd circadian disruption, we will be able to design human–machine

ystems that are more resilient to human error and better ableo reduce the likelihood that fatigue-related errors will lead toccident or injury, but without unnecessarily reducing productiv-ty and efficiency. I hope you enjoy reading this special issue of

revention 45S (2012) 1– 5

Accident Analysis and Prevention as much as I have enjoyed writingits editorial.

References

Aschoff, J., Wever, R.A., 1981. The circadian system of man. In: Aschoff, J. (Ed.),Handbook of Behavioral Neurology: Biological Rhythms. Plenum, New York, pp.311–348.

Baulk, S.D., Fletcher, A., 2012. At home and away: measuring the sleep of Australiantruck drivers. Accident Analysis and Prevention 45, 36–40.

Borbely, A.A., 1982. A two process model of sleep regulation. Human Neurobiology1, 195–204.

Cabon, P., Deharvengt, S., Grau, J.Y., Maille, N., Berechet, I., Mollard, R., 2012. Researchand guidelines for implementing fatigue risk management systems for theFrench regional airlines. Accident Analysis and Prevention 45, 41–44.

Daan, S., Beersma, D., Borbely, A.A., 1984. Timing of human sleep: recovery pro-cess gated by a circadian pacemaker. American Journal of Physiology 246,R161–R183.

Darwent, D., Dawson, D., Roach, G.D., 2012. A model of shiftworker sleep/wakebehaviour. Accident Analysis and Prevention 45, 6–10.

Dawson, D., Chapman, J., Thomas, M.J.W., 2011a. Fatigue-proofing: a new approachto reducing fatigue-related risk using the principles of error management. SleepMedicine Reviews, doi:10.1016/j.smrv.2011.05.004.

Dawson, D., Noy, I., Harma, M., Akerstedt, T., Belenky, G., 2011b. Modelling fatigueand the use of fatigue models in work settings. Accident Analysis and Prevention43, 549–564.

Dorrian, J., Darwent, D., Dawson, D., Roach, G.D., 2012. Predicting pilot’s sleep duringlayovers using their own behaviour or data from colleagues: implications forbiomathematical models. Accident Analysis and Prevention 45, 17–21.

Drury, D.A., Ferguson, S.A., Thomas, M.J.W., 2012. Restricted sleep and negative affec-tive states in commercial pilots during short haul operations. Accident Analysisand Prevention 45, 80–84.

Ferguson, S.A., Paech, G.M., Sargent, C., Darwent, D., Kennaway, D.J., Roach, G.D.,2012. The influence of circadian time and sleep dose on subjective fatigue rat-ings. Accident Analysis and Prevention 45, 50–54.

Gander, P., Hartley, L., Powell, D., Cabon, P., Hitchcock, E., Mills, A., Popkin,S., 2011. Fatigue risk management: organizational factors at the regula-tory and industry/company level. Accident Analysis and Prevention 43, 573–590.

Gander, P., Millar, M.M., Webster, C., Merry, A., 2008. Sleep loss and perfor-mance of anaesthesia trainees and specialists. Chronobiology International 25,1077–1091.

Heath, G., Roach, G.D., Dorrian, J., Ferguson, S.A., Darwent, D., Sargent, C., 2012. Theeffect of sleep restriction on snacking behaviour during a week of simulatedshiftwork. Accident Analysis and Prevention 45, 62–67.

Helmreich, R.L., 2000. On error management: lessons from aviation. British MedicalJournal 320, 781–785.

Holmes, A., Al-Bayat, S.S., Hilditch, C., Bourgeois-Bougrine, S., 2012. Sleep and sleepi-ness during an ultra long-range flight operation between the Middle East andUnited States. Accident Analysis and Prevention 45, 27–31.

House of Representatives Standing Committee on Communications, Transport andthe Arts, 2000. Beyond the Midnight Oil: Managing Fatigue in Transport. TheParliament of the Commonwealth of Australia.

International Air Transport Association (IATA), International Civil Aviation Organi-zation (ICAO), International Federation of Air Line Pilots’ Associations (IFALPA),2011. Fatigue Risk Management Systems: Implementation Guide for Operators.

International Civil Aviation Organization (ICAO), 2011. Fatigue Risk ManagementSystems: implementation Guide for Regulators.

King, A.C., Belenky, G., Van Dongen, H.P.A., 2009. Performance impairment con-sequent to sleep loss: determinants of resistance and susceptibility. CurrentOpinion in Pulmonary Medicine 15, 559–564.

McCauley, P., Kalachev, L.V., Smith, A.D., Belenky, G., Dinges, D.F., Van Dongen, H.P.A.,2009. A new mathematical model for the homeostatic effects of sleep loss onneurobehavioral performance. Journal of Theoretical Biology 256, 227–239.

Mallis, M.M., Mejdal, S., Nguyen, T.T., Dinges, D., 2004. Summary of the key featuresof seven biomathematical models of human fatigue and performance. Aviation,Space and Environmental Medicine 75, A4–A14.

Matthews, R.W., Ferguson, S.A., Zhou, X., Kosmadopoulos, A., Kennaway, D.J., Roach,G.D., 2012. Simulated driving under the influence of extended wake, time of dayand sleep restriction. Accident Analysis and Prevention 45, 55–61.

Michael, D.J., Daugherty, S., Santos, A., Ruby, B.C., Kalns, J.E., 2012. Fatigue biomarkerindex: an objective salivary measure of fatigue level. Accident Analysis andPrevention 45, 68–73.

Mohamed, N., Mohd-Yusoff, M., Othman, I., Zulkipli, Z., Osman, M.R., Voon, W.S.,2012. Fatigue-related crashes involving express buses in Malaysia: will the pro-posed policy of banning the early-hour operation reduce fatigue-related crashesand benefit overall road safety? Accident Analysis and Prevention 45, 45–49.

Morad, Y., Azaria, B., Avni, I., Barkana, Y., Zadok, D., Kohen-Raz, R., Barenboim, E.,2007. Posturography as an indicator of fatigue due to sleep deprivation. Aviation,

Space and Environmental Medicine 78, 859–863.

Paterson, J.L., Dorrian, J., Clarkson, L., Darwent, D., Ferguson, S.A., 2012. Beyondworking time: factors affecting sleep behaviour in rail safety workers. AccidentAnalysis and Prevention 45, 32–35.

Reason, J., 1990. Human Error. Cambridge University Press, New York.

and P

R

RR

S

S

V

V

Editorial / Accident Analysis

oach, G.D., Sargent, C., Darwent, D., Dawson, D., 2012. Duty periods with early starttimes restrict the amount of sleep obtained by short-haul airline pilots. AccidentAnalysis and Prevention 45, 22–26.

omberg, M., 1853. Manual of Nervous Diseases of Man. Sudenham Society, London.upp, T.L., Wesensten, N.J., Bliese, P.D., Balkin, T.J., 2009. Banking sleep: realization

of benefits during subsequent sleep restriction and recovery. Sleep 32, 311–321.argent, C., Darwent, D., Ferguson, S.A., Roach, G.D., 2012. Can a simple balance task

be used to assess fitness for duty? Accident Analysis and Prevention 45, 74–79.piegel, K., Leproult, R., Van Cauter, E., 1999. Impact of sleep debt on metabolic and

endocrine function. Lancet 354, 1435–1439.an Cauter, E., Blackman, J.D., Roland, D., Spire, J.P., Refetoff, S., Polonsky, K.S., 1991.

Modulation of glucose regulation and insulin secretion by circadian rhythmicityand sleep. Journal of Clinical Investigation 88, 934–942.

an Dongen, H.P.A., 2004. Comparison of mathematical model predictions to exper-imental data of fatigue and performance. Aviation, Space and EnvironmentalMedicine 75, A15–A36.

revention 45S (2012) 1– 5 5

Van Dongen, H.P.A., Bender, A.M., Dinges, D.F., 2012. Systematic individualdifferences in sleep homeostatic and circadian rhythm contributions to neurobe-havioral impairment during sleep deprivation. Accident Analysis and Prevention45, 11–16.

Drew Dawson ∗

Centre for Sleep Research, University of SouthAustralia, GPO Box 2471, Adelaide, South Australia

5001, Australia

∗ Tel.: +61 8 8302 6624; fax: +61 8 8302 6623.E-mail address: drew.dawson@unisa.edu.au