Skip to main content
NCBI home page
European Journal of Sport Science logoLink to European Journal of Sport Science
. 2024 Jun 14;24(8):1176–1185. doi: 10.1002/ejsc.12143

Sleep deprivation and recovery: Endurance racing as a novel model

Dwayne L Mann 1,✉, Cassandra L Pattinson 1, Alicia Allan 1, Liam St Pierre 2, Sally Staton 3, Karen Thorpe 3, Kalina Rossa 1, Simon S Smith 1
PMCID: PMC11295088  PMID: 38874812

Abstract

The aim of this study was to investigate sleep‐wake behavior and gain insights into perceived impairment (sleep, fatigue, and cognitive function) of athletes competing in two international multi‐day adventure races. Twenty‐four athletes took part across two independent adventure races: Queensland, Australia and Alaska, USA. Individual sleep periods were determined via actigraphy, and racers self‐reported their perceived sleep disturbances, sleep impairment, fatigue and cognitive function. Each of these indices was calculated for pre‐, during‐ and post‐race periods. Sleep was severely restricted during the race period compared to pre‐race (Queensland, 7:46 [0:29] vs. 2:50 [1:01]; Alaska, 7:39 [0:58] vs. 2:45 [2:05]; mean [SD], hh:mm). As a result, there was a large cumulative sleep debt at race completion, which was not ‘reversed’ in the post‐race period (up to 1 week). The deterioration in all four self‐reported scales of perceived impairment during the race period was largely restored in the post‐race period. This is the first study to document objective sleep‐wake behaviors and subjective impairment of adventure racers, in the context of two geographically diverse, multi‐day, international adventure races. Measures of sleep deprivation indicate that sleep debt was extreme and did not recover to pre‐race levels within 1 week following each race. Despite this objective debt continuing, perceived impairment returned to pre‐race levels quickly post‐race. Therefore, further examination of actual and perceived sleep recovery is warranted. Adventure racing presents a unique scenario to examine sleep, performance and recovery.

Keywords: cognitive impairment, fatigue, sleep restriction, sleep‐related impairment

Highlights

  • Adventure, or expedition, racing is an ultra‐endurance sport, carried out over multiple days, which presents an extreme test of athletic performance under conditions of severe sleep restriction and deprivation. However, limited research has been conducted to examine sleep and perceived impairment of athletes who choose this sport.

  • Using actigraphy, the team‐based nature of adventure racing was evident with severe sleep restriction synchronized between team members during the race.

  • There was significant sleep debt accumulated across the race and was not recovered by the end of the measurement period. Furthermore, perceived impairment returned to pre‐race levels quickly after race completion, despite the remaining sleep debt objectively observed.

  • This study posits that adventure racing presents a unique context to study sleep, performance, and recovery.

1. INTRODUCTION

Adventure racing, also known as expedition racing, is an ultra‐endurance, multi‐disciplinary team sport. Typically, teams of four follow a route or self‐navigate to checkpoints across a course involving some combination of running, hiking/treking, mountain biking (MTB) and kayaking in remote wilderness locations. This form of competition involves considerable mental and physical stamina during the extended race periods (up to 10 days), as teams contend with extreme sleep deprivation, fatigue, unpredictable weather conditions and tough terrain (https://www.sleepmonsters.com/features.php?article_id=51, accessed April 30, 2024). The extended duration of high intensity physical activity combined with severe sleep restriction is unparalleled in any other sport.

During the race, teams self‐select sleep breaks. The timing and duration of these breaks are a compromise between perceived sleep need, race checkpoints, terrain, forced stops and optimization of individual and team performance. The most competitive teams attempt to minimize the time spent asleep, whilst maintaining the cognitive and physical performance necessary to finish. This trade‐off between sleep and race progression may superficially appear as mutually exclusive elements; however, both are necessarily intertwined. For example, self‐navigation through variable and often dangerous terrain on unmarked paths requires substantial cognitive capacity, possible only with sufficient rest. Failure to determine the best route is not only a tactical disadvantage to race progress but may also lead to precarious or dangerous situations. Anecdotal evidence suggests that many racers experience ‘Sleep Monsters’, a colloquial term in adventure racing which describes hallucinations likely induced by sleep deprivation (Miller et al., 2022; Waters et al., 2018). These subjective reports indicate that athletes experience sleep‐dependent cognitive deficits during the race; however, the frequency, duration and timing of sleep opportunities during adventure racing has not previously been documented. After the race finishes, sleep is vital for recovery processes. These processes include the need to discharge the sleep debt accumulated across the race period and an increased need for sleep to promote physiological and cognitive recovery. Evidence from other endurance sports suggests that these recovery needs might not be met and that longer duration events pose greater risks of sleep disturbance following event completion (Calogiuri et al., 2017). Furthermore, sleep prior to the start of the race may also be very important. Strategies such as ensuring sufficient sleep prior to the race, also known as ‘sleep banking’ (Arnal et al., 2015; Rupp et al., 2009), may mitigate the impacts of sleep deprivation, but this has not been previously described in the adventure racing context.

The association between cognitive performance and sleep restriction across a multi‐day adventure race is unknown. Laboratory studies of sleep restriction typically involve monotonous tasks, completed by sedentary participants. However, these studies consistently show a considerable decrease in cognitive performance (e.g., attention, memory, decision making and visuomotor skills) resulting from extended sleep restriction and deprivation (Alhola et al., 2007; Frenda et al., 2016). Adventure racing, unlike typical laboratory environments, features motivation, teamwork, strategy and choice, which may interact to sustain performance in the context of sleep deprivation which has not been well described. As such, studying perceived cognitive functioning, sleep and fatigue in these athletes presents an opportunity to examine the effects of prolonged sleep restriction in an applied, naturalistic setting. Therefore, this study aims to characterize the sleep‐wake behavior and perceived impairment from athletes competing in multi‐day adventure races.

2. METHODS

2.1. Study design and setting

Ethical approval was obtained from the Human Research Ethics Committee at Queensland University of Technology (HREC: 1500000447). Twenty‐four athletes were recruited using convenience sampling from teams taking part in two international adventure races. All participants completed hardcopy surveys that included questions about their demographics, training, race and injury history, as well as self‐report measures of sleep disturbance, sleep‐related impairment, fatigue and cognitive function at three points classified as pre‐race, race and post‐race recovery. Participants also wore an accelerometer for at least 3 days pre‐race, during the race, and up to 7 days (range 0–7 days) post‐race. The two race events (both part of the 2015 Adventure Racing World Series) differed in both geographical location, season and format; as such, each race will be presented separately.

2.1.1. Race 1 (Queensland, Australia)

The ‘8th international XPD Expedition Race’ consisted of 12 race legs (1: Kayak 8 km, 2: Trek 30 km, 3: Kayak 60 km, 4: Trek 50 km, 5 MTB 50 km, 6: Trek 32 km, 7: MTB 53 km, 8: Kayak 70 km, 9: MTB 48 km, 10: MTB 145 km, 11: Trek 48 km, 12: MTB 50 km) over 644 km. The fastest time to complete the course was 5 days, 10 h and 33 min; the last team to complete finished in 9 days, 8 h and 6 min. All team members had to pass through each checkpoint to finish the race; therefore, it was a fixed race distance for all teams, with variable finishing times between the teams. There were no time cut‐offs for stages; however, there were some periods in which participants had to wait at a checkpoint for morning daylight before being able to begin the next stage (e.g., ocean kayaking). This race took place in early August with an average day length of 11 h and 24 min (average sunrise 6:38 a.m.; average sunset 6:01 p.m.). The average daily temperature range was 11.9–26.1°C.

2.1.2. Race 2 (Alaska, USA)

‘Expedition Alaska—2015’ consisted of 14 race legs (including whitewater and open water kayak, packraft, Eklutna glacier traverse, trek and MTB) over a planned 560 km course through the Kenai Peninsula of Alaska. This course had rolling time cut‐offs starting at 33 h into the race. If teams had not passed a certain point on the course (i.e., the end of the current stage), they either took a shorter course or were moved forward on the course by the organizers. No team completed the full length of the total course; however, all teams were on the course for approximately the same amount of time. The race finished for all teams the afternoon of day 7, followed by a recovery night, then an ‘epilogue’ leg on the 8th day (a 5 km trek with an elevation gain of approximately 3000 feet [914 m]). This race took place in late June/early July with an average of 18 h and 48 min of daily sunlight across the data collection period (average sunrise 3:46 a.m.; average sunset 10:35 p.m.) and no complete darkness. The average daily temperature range was 9.8–17.8°C.

2.2. Participant involvement

One of the authors (LSP) competed in both races. He aided in the design of the study, recruitment and data collection. Analysis was conducted separately with appropriate blinding to ensure none of the people reporting the results were aware of the teams and people involved. All athletes in this study were competitive on the international stage and considered to have elite level fitness.

2.2.1. Race 1 (Queensland)

Nine athletes (8 males, 1 female) from three teams wore actigraphs and completed the subjective survey measures at three points. Mean age of participants was 38.78 years (SD = 7.61, range: 31–54 years), and on average, participants trained 14 h per week (SD = 2.67, range: 11–20 h per week, comprised of Foot 5 h, Bike 5.4 h, Paddle 2 h and Other 1.6 h) ahead of the event and had completed 1.8 (range: 0–5) similarly arduous races previously. Nationalities represented included Brazilian, Australian and New Zealand, with some teams including a mix of nationalities. The teams studied finished 6th, 7th and 10th.

2.2.2. Race 2 (Alaska)

Fourteen athletes (10 males, 4 females) from four teams wore actigraphs and completed the subjective measures. The mean age of participants was 35.36 years (SD = 6.77, range: 26–53 years). On average, participants trained 14.68 h per week (SD = 4.56, range: 6–24 h per week, comprised of Foot 5.1 h, Bike 6.4 h, Paddle 2.1 h and Other 1.2 h) ahead of the event had completed 7.8 (range 1–30) similarly arduous races previously. Nationalities represented included United States, Australian and New Zealand, with some teams including a mix of nationalities. The teams studied finished 1st, 2nd, 3rd and 5th.

2.3. Measures

2.3.1. Actigraphy

Participants wore an ActiGraph GT3X tri‐axial accelerometer watch (version 4.4.0) and were instructed to keep the watch on at all times. Data were downloaded via ActiLife software (version 6.13.3), and activity counts (movement activity per unit time greater than software defined threshold) were analyzed in 60‐s epochs using R version 3.3.2. Partially recorded days (i.e., those before the first night of sleep and following the last recorded night of sleep) were discarded from analysis, as were suspected periods of non‐wear. Due to practical constraints, sleep diaries were not feasible; instead, a combination of algorithmic and expert visual scoring were used to identify sleep periods. The current data were characterized by highly disrupted sleep schedules during races, and as such, a standardized automated detection of sleep periods did not accurately identify sleep periods. Therefore, a two‐step process was applied. First, the Tudor‐Locke algorithm (Tudor‐Locke et al., 2014) with default settings was used to detect candidate sleep periods from the full recording for each participant. Second, manual scoring via visual inspection of the data identified additional candidate sleep periods. Only those sleep periods meeting the following criteria were included in subsequent analysis: (1) begin and end on epochs with ‘0’ activity count and (2) bordered by five consecutive ‘0’ activity count epochs. Within accepted sleep periods, the Cole–Kripke (Jean‐Louis et al., 2001) sleep scoring algorithm was applied, which assesses the likelihood of sleep. Importantly, this method captures sustained inactivity indicative of restfulness, which may include both sleep and nap periods, greater than 15 min duration. Once sleep periods were confirmed, sleep debt was calculated. We determined baseline 24‐h sleep/wake fraction using sleep data from the pre‐race period and applied this throughout the race and post‐race periods (per minute of Wake = −1 × sleep/wake fraction, per minute of Sleep = 1‐sleep/wake fraction).

2.3.2. Survey

Sleep disturbance was measured using the PROMIS Sleep Disturbance Short Form 8a instrument, relating to quality and satisfaction with sleep (e.g., ‘I had a problem with my sleep’) on a 1‐5 rating scale (e.g., ‘Not at all—Very much’), with higher scores indicating greater sleep disturbance. Sleep‐related impairment was measured using all 16 items from the PROMIS Sleep‐Related Impairment bank relating to sleep impacting upon usual functioning (e.g., ‘I was sleepy during the daytime’) on a 1–5 response scale (e.g., ‘Not at all—Very much’), possible scores range from 16 to 80, with higher scores indicating greater impairment (Yu et al., 2011). Fatigue was measured using the PROMIS Fatigue Short Form 8a measure that asks participants to rate their level of fatigue (e.g., ‘How run‐down did you feel on average?’) on a 1–5 scale (e.g., ‘Not at all—Very much’), with higher scores indicating more fatigue (Yu et al., 2011). The 8‐item NEURO‐QOL Cognitive Function Short Form (Cella et al., 2012; Iverson et al., 2021) was used to measure participants' perception of their capacity for thinking and completing cognitive tasks (e.g., ‘I had trouble concentrating’) on a 1–5 scale (e.g., ‘Never—Very often (several times a day)’), with lower scores indicating compromised cognitive function.

2.4. Statistical analysis

Survey data were analyzed using IBM SPSS Version 23. Self‐report outcome measures (sleep disturbances, sleep‐related impairment, fatigue and cognitive function) were assessed using repeated‐measures Analysis of variance comparing rated outcomes across time points. Continuous values are presented as mean and standard deviation. There were a very small number (0.7%) of missing responses in the self‐report data, missing at random, and as such, imputation using expectation maximization was conducted. To account for multiple comparisons, Sidak adjustments were made for all pairwise post hoc comparisons. Statistical significance was considered when p 〈 0.05, while measures of effect size were assessed using partial eta squared (ηp 2). Figures were developed using MATLAB (R2018b, MathWorks).

3. RESULTS

3.1. Sleep‐wake behavior

For both race events described, sleep‐wake behavior varied considerably between the pre‐race and race periods, showing the anticipated severe sleep restriction throughout the entire race duration. Post‐race sleep‐wake behavior demonstrated similar patterns to pre‐race, with only marginal increases in sleep duration above the baseline period, which did not appear sufficient to restore the sleep debt accumulated during the race period. Figure 1 illustrates sleep‐wake patterns and cumulative sleep debt across the pre‐race, race and post‐race periods for all individuals from one team competing in Race 1 (Queensland).

FIGURE 1.

FIGURE 1

Open in a new tab

The sleep characteristics of individuals within one adventure race team, across the pre‐race, race and post‐race periods during Race 1 (Queensland). In each panel, the solid black vertical lines indicate the race start and end times, while the dotted vertical lines indicate 12 p.m. each day. The top panel shows the sleep periods for each individual as shaded horizontal bars; the bottom panel illustrates the cumulative sleep debt across time for each of the individual team member. The sleep debt is illustrated as a linear model, where wake time accumulates ‘sleep debt’ at the rate of 1‐sleep/wake fraction (reference to individual pre‐race sleep and wake times) for each minute of wake, while sleep ‘restores’ the debt at the rate of 1‐sleep/wake fraction, for each minute of sleep.

For Race 1 (Queensland), the average sleep duration per day across all participants decreased severely from pre‐race to during the race (7:46 [0:29] vs. 2:50 [1:01]; mean [SD], hh:mm). Average sleep duration during post‐race recovery was 8:25 [1:10] (hh:mm). The team‐based nature of the sport is evident through near identical individual sleep periods (both time‐of‐day and duration) within teams during the race, not observed before and after the race.

Similar patterns were noted for athletes competing in Race 2 (Alaska); however, there were greater variability in sleep patterns within the race period. Average sleep duration per day was severely decreased from pre‐race to during the race (7:39 [0:58] vs. 2:45 [2:05]; mean [SD], hh:mm). Sleep duration varied between teams within the post‐race recovery period; however, the average across all teams (from available data) was 8:23 [1:21] (hh:mm). Further, the post‐race period included one night of sleep prior to the compulsory epilogue race; as such, sleep may have been restricted on the first ‘recovery’ night.

These data also demonstrate that sleep debt continued to accumulate across the race, with recovery of this estimated debt being variable for each team member. Our data indicated that no competitor had fully ‘recovered’ this estimated debt by 7‐days post‐race.

3.2. Perceived impairment

Differences in self‐reported sleep disturbances, sleep‐related impairment, fatigue and cognitive function across the race periods are presented in Table 1 and Figure 2. Sleep‐related impairment and cognitive function followed similar patterns across both races, with significantly higher ratings of impairment and lower ratings of cognitive function during the race than reported at either pre‐ or post‐race. For Race 1 (Queensland), despite significantly shorter sleep durations recorded during the race, there were no statistically significant differences reported across time for sleep disturbances, and although fatigue differed across the race stages, pairwise post‐hoc comparisons were not significant. However, for Race 2 (Alaska), significant differences were found for each outcome. Participants reported significantly less sleep disturbances post‐race compared to pre‐race and race periods. Fatigue significantly differed across all time points with rating of fatigue increasing during the race and not returning to pre‐race levels at post‐race recovery. Finally, participants reported a significant reduction in cognitive functioning during the race compared to both pre‐ and post‐race recovery.

TABLE 1.

Objective sleep duration, self‐reported sleep disturbances, sleep‐related impairment, fatigue and cognitive function across the race periods and group level statistics.

Pre‐race Race Post‐race ANOVA
F‐value p‐value Partial eta squared
Race 1—Queensland
Sleep duration (SD) [HH:MM] 7:46 (0:29) 2:50 (1:01) 8:25 (1:10) ‐ ‐ ‐
Sleep disturbance (SD) 16.22 (2.95) 17.78 (7.03) 16.33 (2.78) 0.36 0.705 0.043
Sleep‐related impairment (SD)†,‡ 29.33 (7.04) 45.56 (5.70) 33.00 (7.55) 24.94 <0.001 0.757
Fatigue (SD) 16.00 (7.09) 21.33 (3.08) 17.11 (5.86) 4.93 0.022 0.381
Cognitive function (SD)†,‡ 33.00 (5.68) 26.89 (5.56) 32.33 (4.15) 4.83 0.023 0.376
Race 2—Alaska
Sleep duration (SD) [HH:MM] 7:39 (0:58) 2:45 (2:05) 8:23 (1:21) ‐ ‐ ‐
Sleep disturbance (SD)‡,§ 18.43 (5.29) 20.71 (5.04) 12.64 (3.69) 35.97 <0.001 0.857
Sleep‐related impairment (SD)†,‡ 33.71 (10.40) 57.93 (10.00) 39.00 (12.00) 30.61 <0.001 0.702
Fatigue (SD)†,‡,§ 15.00 (5.60) 28.14 (5.89) 21.64 (7.02) 17.77 <0.001 0.577
Cognitive function (SD)†,‡ 34.43 (4.22) 22.50 (6.30) 30.50 (7.21) 21.86 <0.001 0.627
Open in a new tab

Note: Symbols denoting significant pairwise comparison, †Pre‐race versus Race, ‡Race versus Post‐race and §Pre‐race versus Post‐race. Higher scores for sleep disturbance, sleep‐related impairment and fatigue indicate greater sleep disturbance, greater impairment and more fatigue, respectively. Lower scores for the cognitive function indicate the compromised cognitive function.

Abbreviation: ANOVA, Analysis of variance.

FIGURE 2.

FIGURE 2

Open in a new tab

Self‐reported sleep disturbance, sleep impairment, fatigue and cognitive functioning across the three race periods. Higher scores for sleep disturbance, sleep‐related impairment and fatigue indicate greater sleep disturbance, greater impairment and more fatigue, respectively. Lower scores for the cognitive function indicate the compromised cognitive function. The vertical bar at each data point shows standard deviation. Pairwise statistics are shown on the plot. Group statistics are shown in Table 1.

4. DISCUSSION

This is the first study to investigate the sleep‐wake behavior and perceived impairment from athletes competing in two, multi‐day adventure races. This observational study aimed to characterize subjective sleep disturbance, sleep impairment, fatigue and cognitive functioning during pre‐race, race and post‐race periods in addition to quantifying actual levels of sleep deprivation typically accrued during these race events. Through this investigation, we gathered valuable insights into the extreme extent of sleep deprivation during races and the discovery that the accumulated debt is not fully restored after 1 week of post‐race recovery period. In line with previous research on sleep deprivation, perceived sleep‐related impairment (Stojanoski et al., 2019; Thompson et al., 2022) and cognitive function (Khan et al., 2023) scores were significantly worse during race periods compared to pre‐race and post‐race periods. Fatigue increased during the race events, while sleep disturbance was only marginally worse during race events. Overall, the accumulated sleep deprivation and impact upon perceived sleep‐related measures (such as cognitive function) have implications for future research across both physical and psychological performance.

The two races we studied differed substantially in duration, format, length, terrain and environment, which is reflected in the results; however, there were some striking similarities.

4.1. Sleep‐wake behavior

In both races, competitors experienced unparalleled levels of chronic volitional sleep restriction throughout the race duration and accumulated considerable sleep debt. Unsurprisingly, individual sleep patterns within teams during the race were highly synchronized. The duration of sleep per day increased marginally in the post‐race period compared to the pre‐race period; however, critically, the observed increase did not facilitate restoration of sleep debt to baseline levels within our recording duration.

While ‘sleep banking’ has been suggested as a potential strategy to mitigate impairment from chronic sleep deprivation (Rupp et al., 2009), we did not observe any systematic behavior to this effect. In the pre‐race period before both of the races studied, sleep patterns varied within and between individuals; however, there was a trend toward shorter sleep duration for the sleep period immediately prior to race start. While most of the competitors we studied were familiar with adventure racing, it is possible that anticipation of the upcoming race, combined with necessary preparations, contributed to this reduced sleep duration.

For each competitor, we modeled sleep parameters from the pre‐race ‘stable’ sleep periods. As such, the sleep debt that accumulates throughout each race was personalized to ‘normal’ sleep periods, as determined from the available data. While this may be affected by travel requirements, it reasonably captures the baseline sleep/wake ratio, which was then applied throughout the recording duration. Individual sleep periods within a team were highly coordinated during the race (to capitalize on strategic sleep opportunities) yet demonstrated divergence in their accumulated sleep debt, highlighting interindividual variability in responses to sleep loss (Tkachenko et al., 2018).

Post‐race sleep duration was often influenced by numerous commitments (e.g., travel, family and work); however, the sleep attained reflects that permitted within each competitor's naturalistic environment. Interestingly, the recovery from sleep debt occurred slowly during the post‐race period and with considerable variability between competitors. As modeled, those individuals with reduced baseline sleep/wake ratio recover sleep debt slightly quicker than those with larger baseline sleep/wake ratio. This may reflect individual sleep requirements and further help to describe trait‐like differences in sleep‐related impairment (Van Dongen et al., 2004).

Sleep‐wake behavior was likely influenced by daily scheduling throughout all stages (pre‐race, race and post‐race). Unsurprisingly, daily ‘scheduling’ overtly affected sleep‐wake behavior during the race period necessarily resulting in extreme sleep restriction. It is possible that certain race activities are more conducive to improved sleep quality (when opportunity presents), while other activities potentially lead to more disturbed sleep. However, this is likely to differ between individuals based on personal activity preferences. Interestingly, during Race 2 (Alaska) where there was no complete darkness, we observed teams resting at times out of phase with the diurnal cycle. This contrasts with Race 1 (Queensland) where there was a clear day/night light cycle and teams tended to rest at night and early morning (dark) times. It is possible that the disrupted light/dark cycle contributes to both scheduling and significant decline in self‐reported Fatigue (pre‐race to race for Race 2, not significant in Race 1). Daily scheduling also likely contributes to sleep parameters in both pre‐race and post‐race conditions. Many participants noted symptoms of jetlag during the pre‐race period and return travel mixed with ‘usual life’ in the post‐race period. Despite these interruptions to sleep in the pre‐ and post‐race periods, the profound interruption to sleep during race period remains distinct.

4.2. Perceived impairment

Sleep‐related impairment, which captures compromised daytime functioning through sleepiness, was significantly worsened during the race period compared to pre‐ and post‐race periods. This pattern of heightened sleep‐related impairment during the race period was significant across both Race 1 and Race 2. In agreement with previous literature, a similar pattern of greater compromise during the race period (i.e., acute chronic sleep deprivation) was also seen for the cognitive function (Filardi et al., 2020; Saugy et al., 2013; Thun et al., 2015). It is possible that adventure racing athletes become ‘desensitized’ or their self‐awareness of changes in cognitive function is diminished during races, given that prolonged periods of sleep deprivation in a natural environment setting, under conditions of near constant exertion, culminating in a significant lack of recovery time, are a routine part of adventure racing.

Overall, the patterns of perceived impairment were consistent across each of the races; however, only in Race 2, did participants report statistically significant differences in perceived impairment, on all four impairment subscales, across the three race period. Indeed, Race 2 participants reported significantly less sleep disturbances post‐race compared to pre‐ and during the race period. Although objectively, participants were unable to fully ‘recover’ the lost sleep from during the race period, and this finding indicates that subjectively, participants felt that during the recovery period, their sleep may have been compressed, ‘deep’ and disproportionally refreshing (e.g., this may be due to a rebound in slow wave sleep) (Brunner et al., 1993; Webb et al., 1965). In apparent contrast to previous research, which has shown that after completing a race, athletes commonly report injuries, illnesses and disruptions to mood and sleep (Graham et al., 2021; Van Dongen et al., 2004); we observed trends toward improvement (though not necessarily complete restoration) in all self‐reported impairment measures.

Interestingly, given the significant reduction in sleep duration seen across both races, the fact that sleep disturbances pre‐ and during the race were not significantly different was not anticipated. However, this finding may be due to a number of factors including (a) the extreme need to consolidate sleep during the race period may have resulted in participants not experiencing sleep disturbances (e.g., reporting no troubles getting to sleep), (b) participants being unaware of their current level of sleep disturbances due to sleep deprivation (Chattu et al., 2018; Dinges, 1995), and/or c) a ‘floor effect’ in the survey instrument, for example, PROMIS Sleep disturbance queries ‘I had trouble getting to sleep’, may not be sensitive in extreme conditions. Indeed, increased sleep propensity is typically associated with fatigue (Ohayon, 2008), which was observed to increase during the race from pre‐race levels. It is well understood that environmental factors, such as noise, light, and temperature, impact negatively upon sleep quantity and quality (Johnson et al., 2018; Okamoto‐Mizuno et al., 2012). However, in the context of racing, reduced sleep quantity is most likely driven by competitive intent. We did not see a significant reduction in sleep quality (i.e., sleep disturbance was not significantly increased during race), suggesting that despite potentially sub‐optimal environmental sleeping conditions, adventure racers were able to maintain sufficient sleep quality (minimal sleep disturbances) during the race.

Finally, fatigue was the only measure which significantly differed across each of the race periods (in Race 2 only). While reported fatigue was greatest during the race, it remained elevated in the post‐race period (as compared to pre‐race). This also supports prior research which found that self‐reported alertness remained altered in a group of endurance sled‐dog racers (Calogiuri et al., 2017). Therefore, fatigue may be more closely related to relative functioning and recovery than the reports of sleep disturbances or sleep‐related impairment.

4.3. Limitations and research implications

We considered several limitations. (1) This is a relatively small sample of predominantly male adventure racers. Therefore, further studies with larger numbers of diverse participants are warranted. (2) In the absence of sleep diary data, the sleep periods in this study were set based on the recorded activity using established algorithms in combination with manual settings. As such, we are unable to determine indices of sleep onset latency. However, anecdotal evidence suggests sleep onset latency in this cohort is very low. (3) Future studies should consider commencing data collection earlier and follow‐up over a longer period. Many competitors traveled across time zones to the race locations and underwent mandatory training (e.g., glacier traverse) in the days prior to race start. Also, competitors report that disruptions to health and mood after race events may take up to 2 weeks to resolve, while pain and injuries may take much longer (Anglem et al., 2008; Calogiuri et al., 2017). (4) Additional aspects which would be of interest in future studies of these athletes may include, (a) injury tracking—the association between physical performance, cognitive components (i.e., balance, navigation), hazardous environmental terrains, and sleep deprivation would be of interest, (b) the effect of diet and pre‐training routines on performance and (c) understanding discrepancies between the anticipated race strategy and actual race strategy may help understand how sleep deprivation in extreme environments interacts with complex decision‐making and higher‐order cognitive faculties.

In addition to the interesting and varied potential scientific directions noted above, the current study also provides a useful base to develop practical applications. While we suggest caution with directly extrapolating our results to all athletes participating in adventure racing, the substantial decline in cognitive functioning we observed in the athletes we studied during the race period (compared with pre‐race period) may help guide event organizers when considering future race designs. It would appear prudent that latter race stages were prepared such that the penalty for failure is less severe; the extreme level of personal accomplishment with overcoming the challenges associated with endurance events should not outweigh competitor safety.

5. CONCLUSION

This study is the first to document sleep‐wake behaviors, perceived sleep, fatigue and cognitive function in the context of two geographically diverse, multi‐day, adventure races. The team nature of the sport was evident with sleep period entrainment. Sleep deprivation in both races was extreme, with this debt not fully reinstated by the end of the monitored post‐race recovery period. Across both races, sleep‐related impairment and cognitive function were significantly worse during race periods versus pre‐ and post‐race periods. However, only in Race 2 (Alaska) did participants report significant problems with fatigue (at all time‐points). Whether this is due to increased difficulty of the Alaska terrain and conditions or due to some other factor unable to be accounted for in this study remains to be seen. What is evident is that adventure racing is characterized by sustained physical exertion and significant sleep deprivation over multiple days, presenting a unique scenario for examining sleep and its relationship to performance.

AUTHOR CONTRIBUTIONS

Study design: Liam St Pierre, Sally Staton, Simon S. Smith, Cassandra L. Pattinson, Alicia Allan; Data analysis: Dwayne L. Mann, Alicia Allan, Cassandra L. Pattinson; Interpretation of results and preparation of the manuscript: all authors.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

ACKNOWLEDGMENTS

The authors would like to thank the race athletes for participation in this study.

Open access publishing facilitated by The University of Queensland, as part of the Wiley ‐ The University of Queensland agreement via the Council of Australian University Librarians.

DATA AVAILABILITY STATEMENT

De‐identified data will be shared upon reasonable request to the corresponding author.

REFERENCES

  1. Alhola, P. , and Polo‐Kantola P.. 2007. “Sleep Deprivation: Impact on Cognitive Performance.” Neuropsychiatric Disease and Treatment 3(5): 553–567. [PMC free article] [PubMed] [Google Scholar]
  2. Anglem, N. , Lucas S. J. E., Rose E. A., and Cotter J. D.. 2008. “Mood, Illness and Injury Responses and Recovery with Adventure Racing.” Wilderness and Environmental Medicine 19(1): 30–38. 10.1580/07-weme-or-091.1. [DOI] [PubMed] [Google Scholar]
  3. Arnal, P. J. , Sauvet F., Leger D., van Beers P., Bayon V., Bougard C., Rabat A., Millet G. Y., and Chennaoui M.. 2015. “Benefits of Sleep Extension on Sustained Attention and Sleep Pressure Before and During Total Sleep Deprivation and Recovery.” Sleep 38(12): 1935–1943. 10.5665/sleep.5244. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Brunner, D. P. , Dijk D.‐J., and Borbély A. A.. 1993. “Repeated Partial Sleep Deprivation Progressively Changes in EEG During Sleep and Wakefulness.” Sleep 16(2): 100–113. 10.1093/sleep/16.2.100. [DOI] [PubMed] [Google Scholar]
  5. Calogiuri, G. , Rossi A., Formenti D., and Weydahl A.. 2017. “Sleep Recovery in Participants After Racing in the Finnmarkslop – Europe's Longest Dog‐Sled Race.” The Journal of Sports Medicine and Physical Fitness 57(1–2): 103–110. 10.23736/s0022-4707.16.05922-3. [DOI] [PubMed] [Google Scholar]
  6. Cella, D. , Lai J. S., Nowinski C. J., Victorson D., Peterman A., Miller D., Bethoux F., et al. 2012. “Neuro‐QOL: Brief Measures of Health‐Related Quality of Life for Clinical Research in Neurology.” Neurology 78(23): 1860–1867. 10.1212/wnl.0b013e318258f744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chattu, V. , Manzar M. D., Kumary S., Burman D., Spence D., and Pandi‐Perumal S.. 2018. “The Global Problem of Insufficient Sleep and its Serious Public Health Implications.” Healthcare (Basel) 7(1): 1. 10.3390/healthcare7010001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dinges, D. F. 1995. “An Overview of Sleepiness and Accidents.” Journal of Sleep Research 4(S2): 4–14. 10.1111/j.1365-2869.1995.tb00220.x. [DOI] [PubMed] [Google Scholar]
  9. Filardi, M. , Morini S., and Plazzi G.. 2020. “Pre‐Race Sleep Management Strategy and Chronotype of Offshore Solo Sailors.” Nature and Science of Sleep 12: 263–269. 10.2147/nss.s241162. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Frenda, S. J. , and Fenn K. M.. 2016. “Sleep Less, Think Worse: The Effect of Sleep Deprivation on Working Memory.” J Appl Res Mem Cogn 5(4): 463–469. 10.1016/j.jarmac.2016.10.001. [DOI] [Google Scholar]
  11. Graham, S. M. , Martindale R. J. J., McKinley M., Connaboy C., Andronikos G., and Susmarski A.. 2021. “The Examination of Mental Toughness, Sleep, Mood and Injury Rates in an Arctic Ultra‐Marathon.” European Journal of Sport Science 21(1): 100–106. 10.1080/17461391.2020.1733670. [DOI] [PubMed] [Google Scholar]
  12. Iverson, G. L. , Connors E. J., Marsh J., and Terry D. P.. 2021. “Examining Normative Reference Values and Item‐Level Symptom Endorsement for the Quality of Life in Neurological Disorders (Neuro‐QoL) v2.0 Cognitive Function‐Short Form.” Archives of Clinical Neuropsychology 36(1): 126–134. 10.1093/arclin/acaa044. [DOI] [PubMed] [Google Scholar]
  13. Jean‐Louis, G. , Kripke D. F., Cole R. J., Assmus J. D., and Langer R. D.. 2001. “Sleep Detection with an Accelerometer Actigraph: Comparisons with Polysomnography.” Physiology & Behavior 72(1–2): 21–28. 10.1016/s0031-9384(00)00355-3. [DOI] [PubMed] [Google Scholar]
  14. Johnson, D. A. , Billings M. E., and Hale L.. 2018. “Environmental Determinants of Insufficient Sleep and Sleep Disorders: Implications for Population Health.” Current Epidemiology Reports 5(2): 61–69. 10.1007/s40471-018-0139-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Khan, M. A. , and Al‐Jahdali H.. 2023. “The Consequences of Sleep Deprivation on Cognitive Performance.” Neurosciences 28(2): 91–99. 10.17712/nsj.2023.2.20220108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Miller, D. J. , Bianchi D., and Lastella M.. 2022. “Running on Empty: Self‐Reported Sleep/Wake Behaviour During Ultra‐Marathon Events Exceeding 100 Miles.” European Journal of Investigation in Health, Psychology and Education 12(7): 792–801. 10.3390/ejihpe12070058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Ohayon, M. M . 2008. “From Wakefulness to Excessive Sleepiness: What We Know and Still Need to Know.” Sleep Medicine Reviews 12(2): 129–141. 10.1016/j.smrv.2008.01.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Okamoto‐Mizuno, K. , and Mizuno K.. 2012. “Effects of Thermal Environment on Sleep and Circadian Rhythm.” Journal of Physiological Anthropology 31(1): 14. 10.1186/1880-6805-31-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Rupp, T. L. , Wesensten N. J., Bliese P. D., and Balkin T. J.. 2009. “Banking Sleep: Realization of Benefits During Subsequent Sleep Restriction and Recovery.” Sleep 32(3): 311–321. 10.1093/sleep/32.3.311. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Saugy, J. , Place N., Millet G. Y., Degache F., Schena F., and Millet G. P.. 2013. “Alterations of Neuromuscular Function After the World's Most Challenging Mountain Ultra‐Marathon.” PLoS One 8(6): e65596. 10.1371/journal.pone.0065596. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Stojanoski, B. , Benoit A., Van Den Berg N., Ray L. B., Owen A. M., Shahidi Zandi A., Quddus A., Comeau F. J. E., and Fogel S. M.. 2019. “Sustained Vigilance Is Negatively Affected by Mild and Acute Sleep Loss Reflected by Reduced Capacity for Decision Making, Motor Preparation, and Execution.” Sleep 42(1):zsy200. 10.1093/sleep/zsy200. [DOI] [PubMed] [Google Scholar]
  22. Thompson, K. I. , Chau M., Lorenzetti M. S., Hill L. D., Fins A. I., and Tartar J. L.. 2022. “Acute Sleep Deprivation Disrupts Emotion, Cognition, Inflammation, and Cortisol in Young Healthy Adults.” Frontiers in Behavioral Neuroscience 16: 945661. 10.3389/fnbeh.2022.945661. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Thun, E. , Bjorvatn B., Flo E., Harris A., and Pallesen S.. 2015. “Sleep, Circadian Rhythms, and Athletic Performance.” Sleep Medicine Reviews 23: 1–9. 10.1016/j.smrv.2014.11.003. [DOI] [PubMed] [Google Scholar]
  24. Tkachenko, O. , and Dinges D. F.. 2018. “Interindividual Variability in Neurobehavioral Response to Sleep Loss: A Comprehensive Review.” Neuroscience & Biobehavioral Reviews 89: 29–48. 10.1016/j.neubiorev.2018.03.017. [DOI] [PubMed] [Google Scholar]
  25. Tudor‐Locke, C. , Barreira T. V., Schuna J. M., Mire E. F., and Katzmarzyk P. T.. 2014. “Fully Automated Waist‐Worn Accelerometer Algorithm for Detecting Children's Sleep‐Period Time Separate from 24‐h Physical Activity or Sedentary Behaviors.” Applied Physiology Nutrition and Metabolism 39(1): 53–57. 10.1139/apnm-2013-0173. [DOI] [PubMed] [Google Scholar]
  26. Van Dongen, H. P. , Baynard M. D., Maislin G., and Dinges D. F.. 2004. “Systematic Interindividual Differences in Neurobehavioral Impairment From Sleep Loss: Evidence of Trait‐Like Differential Vulnerability.” Sleep 27(3): 423–433. [PubMed] [Google Scholar]
  27. Waters, F. , Chiu V., Atkinson A., and Blom J. D.. 2018. “Severe Sleep Deprivation Causes Hallucinations and a Gradual Progression Toward Psychosis with Increasing Time Awake.” Frontiers in Psychiatry 9: 303. 10.3389/fpsyt.2018.00303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Webb, W. B. , and H. W. Agnew, Jr. 1965. “Sleep: Effects of a Restricted Regime.” Science 150(3704): 1745–1747. 10.1126/science.150.3704.1745. [DOI] [PubMed] [Google Scholar]
  29. Yu, L. , Buysse D. J., Germain A., Moul D. E., Stover A., Dodds N. E., Johnston K. L., and Pilkonis P. A.. 2011. “Development of Short Forms From the PROMIS Sleep Disturbance and Sleep‐Related Impairment Item Banks.” Behavioral Sleep Medicine 10(1): 6–24. 10.1080/15402002.2012.636266. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

De‐identified data will be shared upon reasonable request to the corresponding author.

Articles from European Journal of Sport Science are provided here courtesy of Wiley

RESOURCES