Information for Interest | Pilot Advisory Notes | Restricted Access
Impact on Wives and Families | Work and Sleep Patterns | Implications for Fatigue Management
The work practices of Marine Pilots: a review
5.0 SLEEP, FATIGUE AND PERFORMANCE
Research relating to sleep loss, performance and fatigue can be broadly divided into two categories: (i) studies which have examined the effects of total sleep loss, and; (ii) studies which have investigated the impact of partial sleep deprivation. As marine pilots are more susceptible to experiencing chronic reductions in sleep rather than prolonged periods of total sleep loss, the following section briefly reviews the former category of studies and more thoroughly details research relating to the effects of partial sleep deprivation.
5.1 Total Sleep Loss
There is a large body of evidence in the literature indicating that fatigue, mood and performance are adversely affected under conditions of sustained wakefulness. How and colleagues (1994) examined the effects of total sleep deprivation in 20 male naval seamen and found significant reductions in mood and performance in tasks requiring cognitive, vigilance, psychomotor and to some extent, physical functions. Performance deteriorated in basically two stages, with an initial significant drop occurring after 36 to 42 hours of sleep deprivation, followed by a further continuous decline after 66 to 72 hours.
McCarthy and Waters (1997) documented reductions in attention demanding cognitive tasks in 71 male undergraduate students following 36 hours of sleep loss. Specifically, sleep deprived individuals were slower to attend to relevant environmental stimuli, exhibited less response to stimuli, lost interest in stimuli more rapidly and were slower and more variable in their processing of stimuli (McCarthy & Waters 1997).
Similarly, progressive reductions in vigilance performance have been reported during a 30 minute monotonous tracking task after one nights sleep deprivation (Bohnen & Gaillard 1994). Interestingly, performance on a time estimation task was not significantly affected; however the higher levels of stimulation and performance feedback associated with the latter task may have enabled subjects to overcome any effects of 1 night’s sleep loss for the 30 minute testing session (Bohnen & Gaillard 1994).
By examining the way in which mood and performance deteriorate over periods of sustained wakefulness, researchers have identified that both endogenous and exogenous factors are involved. For example, Babkoff and colleagues (1991a) assessed subjective ratings of sleepiness over 72 hours of sleep deprivation in 11 male subjects, and indicated that both accumulated sleep loss and circadian factors were significant in determining the subject’s estimates of sleepiness. Sleepiness ratings progressively increased over the sleepless period and showed distinct circadian oscillations, with the highest ratings of sleepiness occurring between 0200 and 0600 hours, while the lowest ratings were reported at 1000 hours and between 1800 and 2000 hours.
Likewise, Horne et al (1983) noted a significant stepwise decline in intrinsic capacity to detect signals on an auditory vigilance task during 60 hours of sleep loss. Performance fell sharply over usual sleeping periods and levelled out during the day. The lack of performance deterioration over the day was thought to be a reflection of daytime circadian improvements in performance counteracting performance declines due to sleep loss (Horne et al. 1983).
Recent work by Australian researchers (Dawson & Reid 1997) has attempted to quantify the level of cognitive psychomotor performance impairment induced by sustained wakefulness with alcohol intoxication. Forty subjects participated in the study involving two counter-balanced conditions. In one condition subjects remained awake for 28 continuous hours (from 0800 to 1200 hours the following day), while in the second condition subjects consumed 10-15 grams of alcohol at 30 minute intervals (from 0800 hours) until their mean blood alcohol concentration reached 0.10 percent. In both conditions cognitive psychomotor performance was measured by way of an unpredictable tracking task, at 30 minute intervals from the start of the session. The results indicated that performance deteriorated significantly in both conditions. By equating the level of performance impairment observed in both conditions it was shown that after 17 hours of sustained wakefulness, cognitive psychomotor performance had deteriorated to a level equivalent to a blood alcohol concentration of 0.05 percent (Dawson & Reid 1997). Furthermore, 24 hours of sustained wakefulness induced performance decrements equivalent to a blood alcohol concentration of 0.10 percent (Dawson & Reid 1997). Thus it was concluded by the authors that even relatively moderate amounts of sleep loss can produce fatigue-related performance impairments equivalent to, or greater than currently accepted levels for alcohol intoxication (Dawson & Reid 1997).
In summary, it is evident from the above results and findings from other studies (Angus & Heslegrave 1985; Angus et al. 1992; Bergstrom et al. 1973) that fatigue, mood and performance decrements are a common outcome of total sleep loss. The nature of tasks and influence of endogenous and exogenous factors affect the extent to which decrements become evident. However recent work suggests that even moderate levels of sleep loss can produce fatigue-related performance impairments equivalent to or greater than decrements induced by a blood alcohol concentration of 0.05 percent.
5.2 Partial Sleep Deprivation
Studies examining the effects of partial sleep deprivation on mood, fatigue and performance have produced variable results. When 16 young adults (average age, 23 years) had their sleep restricted to approximately 5 hours per night for 7 consecutive nights, mood and performance reductions became apparent after 1 to 2 nights (Dinges et al. 1997). Specifically, subscale scores for fatigue, confusion, tension, mental exhaustion and stress became elevated across the period of sleep restriction, while vigilance performance significantly deteriorated. Memory performance also showed a trend towards poorer performance across days of reduced sleep, but this pattern failed to reach statistical significance (Dinges et al. 1997). The results indicated that there was a cumulative effect on performance and mood, as further significant decrements were observed towards the end of the experiment. Following the period of sleep restriction, 2 nights of recovery sleep were required before a full recovery was achieved (Dinges et al. 1997).
Restricting sleep to 4 hours per night for 2 and 4 days duration has also been shown to cause performance reductions in cognitive, vigilance and memory tasks (Tilley & Wilkinson 1984; Wittersheim et al. 1992). Placement of the 4 hour block of sleep in either the first or second half of the night had no effect on the level of impairment experienced, suggesting that the performance decrements were a function of sleep loss per se rather than the changes in sleep composition which occurred due to the different timing of sleep (Tilley & Wilkinson 1984). Residual impairment in some tasks was still evident after a single night of recovery sleep (Wittersheim et al. 1992), thereby supporting the suggestion of Dinges et al (1997) that at least 2 nights are required before a complete recovery is achieved.
Furthermore, Morris and Miller (1996) reported that performance, as measured by error rates, significantly deteriorated over the first 3.5 hours of a 4 hour simulator flight in 10 experienced pilots, when an average of only 2.4 hours sleep was obtained during the previous night. Pre and post flight scores of subjective fatigue, workload and sleepiness increased, however only the first two of these measures reached statistical significance (Morris & Miller 1996). On the day after the simulated flight, subjects reported higher than normal levels of fatigue despite having had an extended period of sleep post testing (Morris & Miller 1996). This latter finding once again indicates that recovery from partial sleep deprivation is incomplete after 1 night of sleep.
In contrast to the above, Blagrove and colleagues (1995) examined the effects of chronic sleep reduction in 3 groups of young adults and noted no performance decrements in logical reasoning or auditory vigilance tasks, as compared to a control group. One group of subjects, whose sleep had been limited to 4.3 hours per night for 4 days, did however perform worse on a task requiring focused attention, thereby suggesting they were more easily distracted (Blagrove et al. 1995). Caution should be used in reviewing these results however, as the performance tests were only conducted 3 to 6 times per week for 5 to 20 minutes at a time. Hence, subjects were probably able to increase their effort and overcome the effects of sleep deprivation for the short duration of the testing sessions (Blagrove et al. 1995). Given that most jobs would involve a greater workload than that used in the investigation, the relevance of this study to the real world is questionable.
Additionally, Angus and colleagues (1992) reviewed a number of field studies which examined the performance of military personnel during sustained operations in which sleep was limited, and indicated that when an average of 4 or more hours sleep per day was attained, performance remained stable. It is possible however, that the high levels of motivation and dedication often exhibited by military personnel may have enabled the subjects to actively overcome the effects of sleep deprivation for the duration of the studies.
The weight of evidence therefore seems to suggest that partial sleep deprivation, for even relatively short periods of time, can have detrimental effects on mood, fatigue and performance in a wide range of tasks. In fact, a meta-analysis of the effects of sleep deprivation on performance (Pilcher & Huffcutt 1996) indicated that partial sleep deprivation had considerably greater negative effect on mood and cognitive performance than total sleep loss. It was also identified that sleep loss (of either a partial or continuous nature) caused greater decrements in cognitive performance as compared to motor performance (Pilcher & Huffcutt 1996). While these results should be interpreted cautiously as a significant number of primary studies could not be included in the analysis (Pilcher & Huffcutt 1996), they seem to support the general conclusion that partial sleep deprivation negatively affects mood, fatigue and performance. Two possible exceptions to this conclusion may be when tasks are only required to be performed for short periods of time at infrequent intervals (i.e. 3-6 times per week for 5-20 minutes at a time) or when subjects are highly motivated and dedicated to their work. Additionally, as most of the studies cited above used relatively young adults for subjects, it is questionable to what extent these results are applicable to older population groups.
In spite of these limitations, the effects of sleep deprivation seem to be cumulative in nature, as performance and mood became progressively worse as the duration of sleep deprivation continued. This finding is consistent with evidence suggesting that insufficient amounts of sleep obtained over several consecutive days leads to a cumulative sleep debt (Folkard 1996b; Gillberg 1995; Knauth 1993, 1996; Tilley et al. 1982; TSB 1997). It also appears that at least 2 nights of recovery sleep are required before full recovery from partial sleep deprivation is achieved. With regards to marine pilotage in the Great Barrier Reef region, this latter finding is somewhat disconcerting as preliminary work schedule data suggests there may be times when pilots do not have the opportunity for complete recovery between work assignments (Parker et al. unpublished observations).
5.3 Mechanisms of Performance Decrements with Sleep Loss
The effects of sleep deprivation are most evident in a higher prevalence of ‘lapsing’ or ‘microsleep’ (Dinges 1992a; Dinges & Kribbs 1991; Krueger 1989; Reinhart 1995). Lapsing involves periods of delayed or non response which tend to increase in frequency and duration as sleep deprivation continues and manifests as greater variability in performance and increased errors of omission (Dinges 1992a; Dinges & Kribbs 1991; Krueger 1989; Rosekind et al. 1996). It is most evident under boring and monotonous conditions (Dinges & Kribbs 1991), and often the sleep deprived individual is unaware of its occurrence (Reinhart 1995; SIRC 1996).
Sleeplessness has also been associated with a slowed response speed to new and previously encountered stimuli (Dinges 1992a; McCarthy & Waters 1997) and an increased tendency for false positive responding; that is, responding when no signal is present (Dinges 1992a). Additionally, memory problems may be experienced, time on task decrements are accentuated and self paced tasks tend to be carried out more slowly under conditions of sleep deprivation (Dinges 1992a; Dinges & Kribbs 1991).
Furthermore, an unusual manifestation of sleepiness reported amongst a small percentage of night nurses (Folkard et al. 1984) and air traffic control officers (Folkard & Condon 1987) is that of night shift paralysis. This form of paralysis appears to be a special type of sleep paralysis which presents as a temporary inability to respond to relevant stimuli, despite being alert and coherent (Folkard et al. 1984; Folkard & Condon 1987).
The extent to which any one of these features of performance impairment are evident under conditions of sleep deprivation is dependent upon task and personal characteristics, as well as the magnitude of sleeplessness (Babkoff et al. 1991b; Dinges 1992a; Dinges & Kribbs 1991). For example, knowledge of results (Steyvers & Gaillard 1993) and stimulating tasks (Dinges & Kribbs 1991) can, in some instances, improve performance of sleep deprived individuals. Tasks performed intermittently rather than continuously also show less degradation in performance with sleep loss (Angus & Heslegrave 1985). Furthermore, high levels of motivation can enable subjects to exert additional effort for brief periods of time to overcome performance decrements (de Vries-Griever & Meijman 1987; Dinges & Kribbs 1991; Hockey 1997).
5.4 Napping
Given that irregular work hours and periods of sustained wakefulness are at times unavoidable, the effectiveness of potential countermeasures, such as naps, have been examined. While the different study protocols and results obtained make it difficult to draw a general conclusion, naps appear to have a beneficial effect on the mood and performance of sleep deprived individuals when compared to no nap conditions (Angus et al. 1992; Bonnet 1991; Dinges 1992b; Gillberg et al. 1996; Naitoh et al. 1992; Rogers et al. 1989). Naps, however, should not be considered as a replacement for normal nocturnal sleep, as mood and performance after naps generally remained somewhat impaired compared to baseline levels (Angus et al. 1992; Naitoh et al. 1992; Rogers et al. 1989) and in some studies, beneficial effects were only evident for relatively short periods of time (Gillberg et al. 1996).
Naps taken prior to work or before the beginning of a sleepless period can be beneficial, as they reduce the duration of continuous wakefulness and help dissipate any sleep need which the person may have accumulated (Rosekind et al. 1996). A study examining the effects of 2, 4, and 8 hour naps taken before 52 hours of wakefulness revealed that performance and alertness in all of the nap conditions was improved in a dose-response fashion compared to a no nap control condition (Bonnet 1991). The beneficial effects of naps lasted for up to 24 hours, after which time there were no significant differences between the nap and no nap conditions (Bonnet 1991). The finding that longer naps improved performance and alertness to a greater extent than shorter naps supports the suggestion that sleep continuity is an important factor in the restorative function of sleep (Bonnet 1986).
When reviewing the above studies, it is important to recognise some of their limitations. For example, most of the studies employed relatively young subjects who were free from sleep disorders and had no pre-existing sleep debt or shiftwork exposure (Bonnet 1991; Gillberg et al. 1996) Additionally, sleep was often taken in laboratories where an optimal sleep environment was provided by controlling noise, temperature, lighting and other potentially disruptive factors (Bonnet 1991; Rogers et al. 1989). Hence, just how applicable some of these studies are to a specific population such as GBRP is not known.
The effects of sleep inertia also need to be considered when examining the usefulness of naps in countering the effects of sleep deprivation. Sleep inertia presents as a transient period of performance impairment that occurs immediately upon awakening, and is more severe when a person has been sleep deprived and when waking occurs during the first half of the night (Dinges 1992b; Dinges et al. 1985; Dinges & Kribbs 1991). The implications of sleep inertia on performance is an important factor to appraise when a person is required to wake quickly and respond to immediate performance expectations, as marine pilots may have to do.
It is therefore evident that further research examining the effectiveness of naps as a countermeasure for sleep loss is required before definitive conclusions can be reached. Among other things, there is a need to clarify at what time into a period of sleeplessness naps are most beneficial and how long naps should be. Additionally, the impact of such factors as a pre-existing sleep debt, shiftwork exposure and age also need to be examined.
5.5 Summary
In summary, the majority of evidence indicates that sleep deprivation causes a cumulative deterioration in mood, fatigue and performance, and that at least 2 nights of recovery sleep are required before these attributes are restored to baseline levels. When relating these findings to the work practices of marine pilots, a somewhat disturbing picture arises. Marine pilots frequently experience periods of sleep loss as a
result of their work schedules, and preliminary data indicates that on some occasions, the opportunity for complete recovery between work assignments may not be available. Hence, it is possible pilots may, at times, be susceptible to mood, fatigue and performance deterioration induced by inadequate sleep. For this reason it is vital that research investigating the value of potential countermeasures, such as naps, continues.