23rd International Symposium on Shiftwork and Working Time: Towards a Global Consensus

ABSTRACT

The Working Time Society (WTS), and the International Commission on Occupational Health (ICOH) Scientific Committee on Shiftwork and Working Time, are twin organisations focused on conducting research, and informing practice, regarding the impact of work hours in general, and shiftwork in particular, on the efficiency, productivity, safety, well-being, health, and biological rhythms, of employees. Every 2–3 years since 1969, the WTS and ICOH have conducted a series of international symposia in Europe, Asia, Australia, North America, and South America. The purpose of these symposia is to provide a forum for the exchange of knowledge, and the discussion of contested issues, with researchers, employee representatives, regulators, and employers. The most recent symposium in this series – the 23rd International Symposium on Shiftwork and Working Time, entitled “Toward a Global Consensus” – was held on 19–23 June 2017, at Yulara, Australia, near Uluru. Since 2004, Chronobiology International has released a special issue after each symposium, and that tradition continues with a special issue that includes 17 contributions based on a selection of the 128 papers that were presented at the most recent symposium. Here, we provide an overview of the papers that comprise the special issue, and we briefly comment on the implications of the findings for shiftworkers and their employers.

KEYWORDS:

• Shift work
• night work
• work schedules
• circadian rhythms
• sleep
• sleep restriction
• sleep deprivation
• sleep detection
• health outcomes
• fatigue countermeasures

Introduction

The Working Time Society (WTS), and the International Commission on Occupational Health (ICOH) Scientific Committee on Shiftwork and Working Time, are twin organisations focused on conducting research, and informing practice, regarding the impact of work hours in general, and shiftwork in particular, on the efficiency, productivity, safety, well-being, health and biological rhythms, of employees. Every 2–3 years since 1969, the WTS and ICOH have conducted a series of international symposia in Europe, Asia, Australia, North America and South America, to provide a forum for the exchange of knowledge, and the discussion of contested issues, with researchers, employee representatives, regulators and employers. Traditionally, the symposia have been organised with reference to guiding principles that were first specified by Joseph Rutenfranz. These principles – the ‘Rutenfranz Rules’ – can be summarised as: (i) hold the symposium in a remote location, (ii) include participants from a mixture of disciplines, (iii) include a good proportion of early-career researchers, (iv) have fairly brief presentations from as many participants as possible, (v) take breaks and meals together to foster informal exchanges, (vi) include some group outings to encourage bonding and (vii) share the information that is presented with non-participants.

The most recent symposium in this series – the 23rd International Symposium on Shiftwork and Working Time, entitled ‘‘Toward a Global Consensus’’ – was held on 19–23 June 2017, at Yulara, Australia, near Uluru (fulfilling the first Rutenfranz Rule – hold the symposium in a remote location). The symposium was jointly organised by the WTS, ICOH’s Scientific Committee on Shiftwork and Working Time, and CQUniversity’s Appleton Institute for Behavioural Science. During the symposium, a total of 128 papers were presented in oral and poster sessions focused on the effects of shiftwork on physical and mental health, interactions between the sleep/wake and circadian systems, strategies and interventions to counteract the negative effects of shiftwork, biomathematical models of fatigue risk, and relationships between light, circadian rhythms and chronotype. A detailed program is available on the WTS website (www.workingtime.org).

In addition to the symposium’s 128 standard papers, a series of consensus papers were drafted by subject-matter experts and circulated to all WTS members prior to the symposium, then discussed in separate sessions at the symposium, and will be refined based on those discussions and published in a special issue of an open-access international scientific journal focussed on occupational health and safety – Industrial Health. These papers will provide consensus statements regarding current issues related to shiftwork, including the effects of shiftwork on physical and mental health, psychosocial stressors associated with shiftwork, interventions to improve circadian adaptation to shiftwork, effects of shiftwork on family and community, regulatory approaches to reducing the risks associated with shiftwork, training and education for employees and employers alike, risk-based approaches to fatigue management in shiftwork, non-diurnal work schedules and their impact on threshold limits to xenobiotics exposure, and individual differences in tolerance to shiftwork.

To fulfil the seventh Rutenfranz Rule – share the information that is presented with non-participants – selected papers from previous symposia have been published in edited books and scientific journals. Since 2004, Chronobiology International has released a special issue after each symposium, and that tradition continues with this special issue that includes 17 contributions based on papers that were presented at the 23rd International Symposium on Shiftwork and Working Time. Below, we provide an overview of the papers that comprise the special issue and briefly comment on the implications of their findings.

Effects of total sleep deprivation and partial sleep restriction

At the most basic level, there are at least four stages in completing a cognitive task – perceiving, encoding, processing and responding. Honn et al. (2018) examined whether the encoding of information – one of the upstream components of cognition – is impaired by a full night of sleep deprivation. Participants completed a semantic matching task during the daytime after being awake for either 7 h (n = 31) or 31 h (n = 38). Sleep-deprived participants took longer than control participants to decide whether or not a pair of words belonged to the same semantic category, but the time taken for the encoding aspect of the task did not statistically differ between the groups. The authors conclude that the impairment in cognition caused by a night of sleep deprivation must occur downstream of the encoding of information. A next step is to determine whether or not semantic encoding is affected by the type of sleep loss that is most often experienced by shiftworkers – multiple consecutive days of partial sleep restriction.

Matthews et al. (2018) conducted a pilot study with six participants, using a repeated-measures design, to determine whether having a conversation while driving after a night of sleep restriction impairs performance. Participants had just 4 h in bed during the night before completing a 20-minute drive in a simulator during the afternoon. While driving, participants either sat in silence with a ‘passenger’ sitting behind them, had a conversation with a ‘passenger’ sitting behind them, or had a conversation with a researcher using a hands-free telephone. An analysis of the variance indicated that there was no effect of condition, but effect size analyses indicated that, compared to the control condition, lane deviation was significantly greater in both the passenger conversation condition (small effect size) and the phone conversation condition (large effect size). These results suggest that the cognitive demands of conducting a telephone conversation while sleep deprived may impair driving performance. Although allowable in most jurisdictions, this study indicates that it may be prudent for people who are sleep deprived, such as shiftworkers, not to talk on a phone while driving. In case it is unavoidable to phone during a commute (e.g., an emergency call), it is advisable to bring the car to a stop.

Validation of sleep-detection technologies

Sargent et al. (2018) examined the validity of a commercially available wearable sleep detection device (FitBit) using polysomnography for comparison. Such a comparison is pertinent, given that several devices, i.e., phones, smart watches, activity trackers, already include functions to assess sleep parameters derived from quantifications of physical activity, heart rate, electrodermal activity, etc. For many of these devices, the presence and timing of sleep periods can either be done by automatic detection or by manual entry. In this case, the device automatically detected 24 of 30 × 7–9-h night-time sleeps (80%) and 6 of 20 × 1–2-h daytime naps (30%). When automatic detection was used, total sleep time was overestimated by 51.5 minutes for night-time sleeps and by 3.9 minutes for daytime naps. When manual entry was used, total sleep time was overestimated by 7.0 minutes for night-time sleeps and 2.5 minutes for daytime naps. Even if wearable devices are being used with shiftworkers to collect large and representative data sets, detection failure rates of 20% for night-time sleeps and 70% for daytime naps hardly appear acceptable. Hence, in most other studies with rather small sample sizes, it would be advisable to establish the presence and timing of sleep periods using methods that are independent of the automatic detection functions, such as event markers or sleep diaries.

Impact of shiftwork on sleep/wake behaviour

Riethmeister et al. (2018) examined the sleep/wake behaviour of 60 employees working 14 x 12-h day shifts (07:00–19:00h) on offshore oil rigs in the Dutch Central North Sea. While offshore, the workers spent less time in bed and obtained less sleep than at home. Interestingly though, the proportion of time in bed that was converted into sleep (i.e., sleep efficiency) was higher while sleeping offshore than at home. These results suggest that less sleep is obtained while offshore than at home, not because the conditions on the offshore rigs interfere with sleep, but because the workers spend less time in bed while offshore. Therefore, interventions to increase total sleep time while offshore are probably better directed at increasing time in bed rather than modifying the sleep environment.

Bordreau et al. (2018) examined sleep/wake patterns, cognition, and alertness in 18 marine pilots working in the Saint Lawrence River seaway in Canada over 16–21 days each. Consistent with the results of laboratory-based studies, both basic cognitive function and subjective alertness were affected by time-of-day and the duration of prior wake, and both were lowest in the morning after a night of work. On workdays, the pilots obtained 18% less sleep overall, but the proportion of sleep obtained during the daytime (07:00–23:00h) compared to the night-time (23:00–07:00h) was 89% higher, than on rest (work-free) days. In situations such as this, where shiftworkers may not always be able to obtain a sufficient amount of sleep during the night, they should be encouraged to supplement nocturnal sleep with daytime naps. Such daytime napping should be supported with appropriate facilities, effective policies and procedures, and competency-based training and education.

Karhula et al. (2018) compared the sleep/wake behaviour of 154 permanent night workers with that of 6486 rotational shiftworkers and 2672 non-shiftworkers (day workers) from six hospital districts, using data from the Finnish Public Sector Survey. The proportion of all shifts that were worked at night by permanent night workers, rotational shiftworkers, and day workers was 93%, 11% and 0%, respectively. Despite having to work at night, and presumably having to sleep during the daytime, the amount of sleep obtained by permanent night workers (7.45 h/day) was greater than that obtained by rotational shiftworkers (7.22 h/day) and day workers (7.17 h/day). If possible, it would be interesting to obtain more information from these permanent night workers to determine whether their capacity to cope with night work is due to physiological predispositions (e.g., long circadian period) and/or due to behavioural strategies (e.g., polyphasic sleep). This information could be used to select and/or train others to cope better with night work.

Impact of shiftwork on mental and physical health

Garde et al. (2018) describe a database – the Danish Working Hour Database (DWHD) – that has been sourced from the payroll data of Danish employees who predominantly work in public hospitals. The database is comprehensive – it contains 190,438,405 shifts worked by 345,260 individuals across a 5–6-year period, with information about (i) the timing of regular shifts, overtime shifts, and on-call shifts, (ii) absence from work due to sickness, holidays, maternity leave, care days, and work-related injuries, and (iii) demographic factors such as gender, date of birth, job type, and department/hospital. The database should prove to be an invaluable resource for those conducting research regarding the effect of work hours both on short-term and long-term health outcomes. Already, researchers are using the database to investigate the impact of particular aspects of shiftwork on sickness, work-related injury, mental disorders, pregnancy outcomes, and cardiovascular disease.

Holanda et al. (2018) examined the relationship between exposure to night work and indicators of metabolic syndrome in shiftworkers. A total of 60 nurses were divided into three groups based on whether they currently did night work (n = 30), previously did night work (n = 18), or had never done night work (n = 12). Compared to the ‘never exposed’ and ‘previously exposed’ groups, the ‘currently exposed’ group obtained less sleep on workdays, they were more likely to have higher levels of triglycerides, and they were more likely to have higher blood pressure. The prevalence of metabolic syndrome was higher in this total sample of nurses (32%) than in the general population (25%), but the prevalence of metabolic syndrome did not significantly differ between the three study groups. These preliminary results with a small number of observations suggest that exposure to night work may not further increase the risk of metabolic syndrome associated with shiftwork.

Gifkins et al. (2018) conducted semi-structured interviews with experienced nurses (n = 12) and inexperienced nurses (n = 9) to examine whether food choices and eating pattern were affected by exposure to shiftwork. Both groups reported having increased food cravings, caffeine consumption, snacking, and dehydration, during night shifts compared to day shifts. Experienced nurses were more likely than inexperienced nurses to plan ahead and to consume home-made meals at work, but paradoxically, they were also more likely to skip meals at work in case of too high workload.

Vincent et al. (2018) investigated the potential benefits to health of breaking up prolonged sitting when sleep restricted, using a repeated-measures design with six adult males in the laboratory. On two separate occasions, participants had 3 days with 5 h in bed each night followed by a 3-h postprandial glucose response test each morning. During the 3 days of sleep restriction, participants spent most of their time sitting in a chair, either with no extra physical activity (inactive) or 3 minutes of treadmill walking at 3 km/h every hour (active). There were no main effects of day or condition on the area under the curve for glucose, and no interaction effect. Given that prolonged sitting while being sleep restricted did not impair glucose metabolism, it was not possible to test the potentially beneficial effects of light-intensity walking. Nevertheless, examining the efficacy of such interventions is a worthy pursuit given that shiftworkers who work at night on average are more likely than non-shiftworkers to have sedentary occupations, obtain less sleep, and have poorer health.

Impact of on-call work arrangements

The requirement to be on-call can be disruptive for salaried and volunteer personnel in the emergency services industry (e.g., fire, ambulance, police). Jay et al. (2018) interviewed 24 women from two emergency service agencies to examine the impact of being on-call during the night. These 24 women stated that being on-call disturbed their sleep, made them fatigued, made them feel that work was relentless, and disrupted their family life. They also stated that their coping strategies included being supported by family, friends and colleagues, and ensuring that they planned and prepared in advance for on-call periods. Ideally, the knowledge generated by this qualitative study can be used to lower barriers to entry, and inform training and education programs, for women in the emergency service industry. It is not yet clear whether the effects of being on-call, and the effectiveness of particular coping strategies, are similar for women and men.

Sprajcer et al. (2018) conducted a laboratory-based study with 24 participants, using a repeated-measures design, to examine whether being on-call during the night increased anxiety and/or interfered with sleep. On three separate nights, the perceived likelihood of being called during sleep was manipulated by telling participants either that they will ‘definitely not be called’, that they ‘may be called’, or that they ‘definitely will be called’. On the ‘will be called’ night, participants felt more anxious before going to bed compared to the ‘may be called’ and ‘will not be called’ nights, but the total amount of sleep obtained was similar on all three conditions. These results suggest that the potentially disruptive influence on sleep of being on-call is probably not related to the perceived likelihood of being called, which in turn, rules it out as a factor that could be manipulated to improve the sleep of shiftworkers.

Use of work breaks and napping as a countermeasure to fatigue

Wilson et al. (2018) conducted a survey to examine the efficacy of workplace break opportunities in 1285 healthcare workers at two hospitals in the US Pacific Northwest. The respondents worked day shifts of 8–12 h duration starting at 06:00–08:00h (66%), 12-h night shifts starting at 18:00–20:00h (13%), or other types of shifts (21%). The 69% of respondents who took a break of at least 30 minutes during shifts reported greater levels of sleepiness than the 31% of respondents who did not take such breaks. The authors highlight the difficulty of interpreting cross-sectional data. It could be inferred that taking a break at work increases sleepiness, but an equally valid, and potentially more likely explanation, is that people who are sleepy are more likely to take a break at work than others.

Centofanti et al. (2018) surveyed 130 nurses and midwives regarding their use of two common fatigue countermeasures – napping and caffeine. Those 16% of respondents who napped during night shifts obtained less sleep at home than others, which indicates either that obtaining less sleep at home increases the need to nap at work, or that napping at work reduces the need to sleep at home. Given that the nappers obtained 3 h less sleep at home than others, but the maximum opportunity to nap was only 45 minutes, the former explanation is the most plausible. On average, respondents had been doing shiftwork already for 19 years, and over that time, the prevalence of high caffeine consumption (i.e., > 400mg/day) increased from 15% to 33%. This indicates either that shiftwork encourages caffeine consumption, or that caffeine consumption increases with age. This paper highlights interesting associations that could be pursued using alternative research designs to determine the presence and/or direction of causality.

Romyn et al. (2018) conducted a laboratory-based study to examine the efficacy of daytime napping to supplement night-time sleep in athletes (very fit shiftworkers!). Twelve well-trained male soccer players completed three conditions in a counterbalanced order: 9 h in bed overnight followed by no daytime nap (9 + 0); 8 h in bed overnight followed by a 1-h daytime nap (8 + 1); and 7 h in bed overnight followed by a 2-h daytime nap (7 + 2). The total amount of sleep obtained in each of the three conditions was similar, i.e., 8.1 h (9 + 0), 8.2 h (8 + 1), and 8.0 h (7 + 2). Using the same data set, Lastella et al. (2018) examined how well the soccer players were able to estimate the amount of sleep that they obtained during the 1- and 2-h afternoon nap opportunities. Compared to polysomnography – the current gold standard – participants on average underestimated the amount of sleep that they obtained in both the 1-h nap (56 vs. 45 minutes, respectively) and in the 2-h nap (104 vs. 82 minutes, respectively). Taken together, these results indicate that healthy young adults are able to supplement sleep at night with a daytime nap, but that they might underestimate the amount of sleep obtained during the nap.

Roach et al. (2018) examined whether the angle of recline of a seat affected both the quantity and quality of sleep, using a repeated-measures design with six healthy adults. Participants attended the laboratory on three occasions, and each time they had a 4-h sleep opportunity in a bed (02:00–06:00h), 7 h of wake, and then a 4-h sleep opportunity in a seat (13:00–17:00h). The only difference between conditions was the back angle of the seat to the vertical during the seat-based sleep periods: 20 degrees (upright), 40 degrees (reclined), and 90 degrees (flat). Compared to the flat seat, the reclined seat resulted in similar amounts of total sleep and slow-wave sleep, but 37% less REM sleep; and the upright seat resulted in 29% less total sleep, 30% less slow-wave sleep, and 79% less REM sleep. These data are particularly relevant for long-haul transportation industries (i.e., road, rail, air, sea) where shiftworkers have access to on-board rest facilities during duty periods. The data could be used by companies that employ shiftworkers, or indeed by passengers, to compare the sleep-related benefits, with the financial costs, of supplying, or paying for access to, different types of seats.

Agenda for future research

We hope that the results published in this special edition, together with the results and recommendations published in the Working Time Society’s consensus papers in Industrial Health, will inform and spark future research within the following framework:

• elucidate the underlying pathophysiology of health and sleep problems in night- and shift workers;

• identify key job stressors and risk factors to safety in the context of night- and shift work, especially at the level of the individual;

• validate interventions – at the level of the individual – to improve adaptation processes combining physiological, behavioural and socio-cultural approaches in the context of work 4.0;

• update regulatory approaches aimed at reducing the risks associated with shift work, which includes training and education of both employees and employers;

• examine the combined effects of environmental and organisational factors affecting health outcomes of night- and shift workers;

• evaluate the safety of current xenobiotic exposure limits for night- and shift workers.

Declaration of interest

In cases where a guest editor was an author of a manuscript submitted to this Special Issue, editorial responsibility for that paper was assigned to either a guest editor with no conflict of interest or to a member of Chronobiology International’s Editorial Board.

Acknowledgements

We thank the Editor-in-Chief, Prof. Francesco Portaluppi, and the publisher, Taylor & Francis, for their support of this Special Issue of Chronobiology International. We also acknowledge the financial support provided to the 23rd International Symposium on Shiftwork and Working Time by Integrated Safety Support, Safe Work Australia, the International Commission on Occupational Health, and the Working Time Society. Finally, we are very grateful to all of those who contributed their time and expertise to this Special Issue by reviewing submitted manuscripts.