Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2024 Mar 12;33(6):e14186. doi: 10.1111/jsr.14186

Disruptions to sleep and circadian rhythms are associated with poorer athlete mental health in female, but not male, elite Australian Rules footballers

Luis Mascaro 1,2, Josh Leota 1,2, Daniel Hoffman 1,2, Shantha M W Rajaratnam 1,2, Sean P A Drummond 1,2, Elise R Facer‐Childs 1,2,✉
PMCID: PMC11596983  PMID: 38471498

Summary

Elite athletes are vulnerable to sleep and circadian disruption and associated mental health symptoms. This study aimed to investigate sex differences in sleep, circadian rhythms, and mental health, as well as the moderating role of sex in the prediction of mental health, among male professional and female semi‐professional elite athletes. Participants were 87 elite Australian Rules football (ARF) athletes (43% female; mean [standard deviation] age 24.0 [4.1] years). Participants completed baseline questionnaires, 2 weeks of sleep/wake monitoring via actigraphy, and a circadian phase assessment (dim‐light melatonin onset [DLMO]). Cross‐sectional data were collected in training‐only Australian Football League (AFL) Men's and Women's pre‐season periods, with 53 providing data in two pre‐seasons. Female athletes, relative to males, reported poorer mental health (a higher athlete psychological strain score), had a later mid‐sleep time (by 28 min), reported a greater preference towards eveningness, and displayed a later circadian phase (by 33 min). For female athletes, lower sleep efficiency and lower sleep regularity were associated with poorer mental health. For female athletes, there were U‐shaped relationships between both morningness–eveningness and phase angle (interval between sleep onset and DLMO time) and mental health. No significant relationships were found for male athletes. In summary, elite female ARF athletes reported poorer mental health, relative to males, especially when experiencing sleep or circadian disruption. Lifestyle factors associated with sex differences in ARF professionalism (scheduling, finances, supports) may contribute to these findings. Programmes to improve sleep, circadian alignment, and mental health among female semi‐professional elite athletes should be strongly considered.

Keywords: actigraphy, elite athletes, melatonin, psychological strain, sex differences

1. INTRODUCTION

The high prevalence of poor mental health in elite athlete populations highlights the critical need for further research and evidence‐based interventions (Reardon et al., 2019). Environmental factors that influence mental health include exercise, light, and social cues (Facer‐Childs et al., 2019; Haynes et al., 2016). These same exposures influence sleep and circadian health, which compound the strain on mental health (Foster, 2020). Circadian rhythms refer to behaviours that follow about 24‐h rhythms, such as muscle strength and aerobic capacity (Brager et al., 2022). The timing of endogenous melatonin (a sleep‐regulating hormone) secretion is one key marker of circadian phase (Czeisler et al., 1999); however, this has never been measured objectively in a group of male and female athletes. Considering sex differences in objective sleep and circadian phase, alongside group differences in environmental exposures, may help tailor athlete recommendations for optimal mental health.

Consistent with data from the general population, female athletes report poorer mental health compared to their male counterparts (Walton et al., 2021). However, sex differences in athlete sleep are unclear, likely due to the under representation of female athletes in sport science research (Cowley et al., 2021) and the difficulty in disentangling the relative influence of biological sex from environmental influences (i.e., the athlete lifestyle). Female adults generally report a more morning preference, have an earlier sleep–wake timing, and have an earlier melatonin peak relative to male adults (Cain et al., 2010; Roenneberg et al., 2004). Females with sleep or circadian rhythm disruption may have a greater risk of poor mental health relative to males (Carpenter et al., 2021); however, this is yet to be established in elite athletes.

In Australian Rules football (ARF), elite female athletes have been found to sleep later, shorter, and less efficiently relative to elite male athletes (Roberts et al., 2021). In contrast to the male athletes (competing in the ‘Australian Football League’ [AFL]), female ARF elite athletes (competing in the ‘AFL Women's’ [AFLW]) train mostly in the evening (Black, 2023), which may explain their later sleep–wake timing. Furthermore, both the AFL and AFLW are elite competitions (i.e., include the best athletes in their field), though AFLW athletes compete semi‐professionally (average salary AUD $46,280 in 2022) and AFL athletes have full‐time, professional status (average salary AUD $372,224 in 2021; Dixon, 2022). Both groups require a substantial time commitment for competition, training, and recovery; however, the pay difference necessitates many female athletes work a second job while the male athletes are afforded greater daytime access to training and recovery facilities. The demands on AFLW athletes may extend beyond what their semi‐professional, part‐time status allows, and the consequences of this potential inequity remain unclear. Further research may reveal the implications of whether professional status exacerbates or alters sex differences in sleep, circadian, and mental health in elite athletes.

The present study investigated sex differences in sleep, circadian timing, and mental health in elite ARF athletes, and whether sex moderated the relationship between sleep and circadian rhythms and mental health. It was hypothesised that: (i) female athletes will report poorer mental health, have poorer objective sleep, have later circadian timing, and have greater circadian misalignment relative to male athletes; and (ii) relationships of sleep and circadian disruption to mental health will be stronger among female athletes.

2. METHODS

2.1. Participants

Participants were recruited from an elite ARF club in Victoria, Australia. Eligibility criteria included that they were currently on the active playing list, free from concussion or long‐term injury while completing the study, and were aged ≥18 years at the commencement of data collection. Of the 89 elite athletes potentially available, 87 athletes were eligible, provided written informed consent, and completed the study. The study protocol was repeated during two consecutive pre‐seasons, with only 53 providing a second datapoint due to changes in eligibility between timepoints (e.g., squad turnover or injury), resulting in a total of 140 participant datasets. In elite ARF, biologically male athletes compete in the AFL and biologically female athletes compete in the AFLW. In the present study, biological sex was self‐reported (57% male, 43% female) and information on gender (a socially constructed identity) was not collected. The larger proportion of male athletes was expected based on larger squad sizes in the AFL. There was no sex difference in age, at a mean (standard deviation [SD]) of 24.02 (4.10) years (p = 0.68). The male athletes were taller (mean [SD] height 188.93 [7.57] cm) and weighed more (mean [SD] 86.69 [7.76] kg) relative to the female athletes (mean [SD] height 171.46 [7.48] cm; weight 66.97 [9.04] kg; all p < 0.001). Ethics approval was obtained from the Monash University Human Research Ethics Committee (approval number: 2020‐26361‐50920).

2.2. Procedure

The study was conducted twice with each of the AFL (February 2021, February 2022) and AFLW (November 2020, November 2021) squads separately during their respective pre‐seasons (i.e., training schedules enforced, but no official matches or travel associated with jet lag). Each study was an approximately 2‐week protocol which included: (a) baseline mental health and chronotype self‐report questionnaires; (b) 2 weeks of at‐home objective sleep monitoring; and (c) one at‐home circadian phase assessment completed about half‐way through the at‐home sleep monitoring period.

2.3. Measures

2.3.1. Mental health

The Athlete Psychological Strain Questionnaire (APSQ; Rice et al., 2020a) measures athlete‐specific mental health. The scale consists of 10 items (e.g., ‘Over the past month, it was difficult to be around teammates’) answered on a 5‐point Likert scale from 1 (‘None of the time’) to 5 (‘All of the time’). Higher total scores indicate poorer mental health. The APSQ has shown very good discriminant and convergent validity (Rice et al., 2020b). In the present sample, the internal consistency was high (α = 0.88).

2.3.2. Sleep

The GENEActiv actigraphy device (ActivInsights Ltd., Cambridgeshire, UK) was used to collect wrist‐worn rest‐activity data (accelerometry collected at 30 Hz and stored in 30‐s epochs). Participants on average wore the watch for 14 ± 2 nights. Accelerometry data were processed using the Cicada toolbox (v0.10.4 [beta]; Rick Wassing, 2022) in MATLAB R2019b (MathWorks, Natick, MA, USA). The intervals within which the GGIR algorithm (at medium sensitivity) detected sleep were based on self‐reported lights‐off/‐on times from daily sleep diaries. When diary reported times were missing or appeared a clear reporting error, algorithm‐estimated intervals were used to guide interval limits.

Total sleep time (TST; sleep duration [h]) and sleep efficiency (SE; the proportion of time spent asleep while in bed) from actigraphy were used as measures of sleep quantity and quality, respectively. Analysing actigraphy data in elite athletes has been validated as a suitable alternative to the ‘gold‐standard’ polysomnography (Sargent et al., 2016).

2.3.3. Circadian‐related timing

The reduced Morningness–Eveningness Questionnaire (rMEQ; Adan & Almirall, 1991) was used to measure preference towards morningness or eveningness (i.e., chronotype). The rMEQ consists of five items answered on a 4‐ or 5‐point Likert scale. For example, ‘During the first half an hour after you wake up in the morning, how do you feel?’ is on a 4‐point scale from 1 (‘Very tired’) to 4 (‘Very refreshed’). Higher total scores indicate greater morningness preference. The reduced version demonstrates strong discriminant and external validity based on objective activity (Natale et al., 2006). In the present sample, the internal consistency was adequate (α = 0.72).

Mid‐sleep time (MST; the midpoint between sleep onset and offset) from actigraphy was used to measure circadian‐related timing via objective rest‐activity patterns.

The at‐home circadian phase assessment protocol was adapted from Stone et al. (2021). Participants remained in dim light (<10 lux) for 6 h, from 5‐h pre‐ to 1‐h post‐habitual sleep time (individual habitual sleep times for the assessment determined by the sleep diaries collected ~7 days prior). Participants were instructed to remain seated and refrain from eating or drinking for 20 min prior to an hourly saliva sample self‐collected via Salivette (Sarstedt, Nümbrecht, Germany). Adherence was monitored via video with researchers. Samples remained on ice (−20°C) until radioimmunoassay of melatonin levels for each sample with the lowest limit of quantitation at 1 pg/mL. From the assessment, objective circadian timing was measured via dim‐light melatonin onset (DLMO), calculated as the linear interpolation across the values immediately below and above the threshold of 4 pg/mL (Stone et al., 2021).

2.3.4. Circadian alignment

Circadian alignment was measured using phase angle and sleep regularity index. Phase angle was calculated as the difference between average sleep onset (from actigraphy) and DLMO time. The sleep regularity index (SRI; Phillips et al., 2017) measures the probability an individual will be in the same sleep–wake state 24‐h apart, averaged across the recording period. Scores range from 0 to 100, with higher scores representing greater consistency between days. The SRI was calculated using the ‘sleepreg’ R package (version 1.3.5) (Windred et al., 2021).

2.4. Statistical analysis

Data analysis was completed using R (R Core Team, 2022) and statistical significance was set at p < 0.05 for all analyses. Missing data on questionnaires (1%), actigraphy (1%), and circadian phase (10%) were not imputed.

Linear mixed models were used to calculate sex differences in mental health (APSQ), sleep (TST, SE), circadian‐related timing (rMEQ, MST, DLMO), and circadian alignment (phase angle, SRI). Each model included participant identifier as a random intercept, ‘sex’ (dummy coded, 0 = male, 1 = female) as the main predictor, and ‘timepoint’ (dummy coded, 0 = baseline, 1 = 1‐year follow‐up) as a covariate to control for change as a function of time.

Linear mixed models were used to examine whether sex moderated the relationship between sleep or circadian health and mental health. Each model included participant identifier as a random intercept, timepoint as a covariate, and an interaction term between sex and the sleep/circadian factor. A non‐significant interaction term was dropped from the model. Each circadian‐related timing variable and phase angle was initially entered as a quadratic polynomial based on the hypothesis that timing skewed towards very early or very late could both be associated with similar mental health outcomes (Lewy et al., 2006). A non‐significant quadratic term was replaced with the original term.

For any significant interactions, estimated marginal means analysis and visualisations were used to determine the nature of the interaction. For significant quadratic interactions, original values were mean‐centred and squared to facilitate the estimated marginal means analysis. For all significant interactions, the Johnson–Neyman technique (Johnson & Fay, 1950) was used to determine the range of values of the predictor for which there is a significant sex difference in the outcome.

2.4.1. Post hoc analyses

To provide further information on athlete mental health, the three sub‐scales of the APSQ (‘Self‐Regulation’, four items; ‘Performance’, four items; ‘External Coping’, two items) were calculated and analysed for sex differences using linear mixed models (including participant identifier as a random intercept, sex as the main predictor, and timepoint as a covariate).

To explore the potential roles of evening zeitgebers, the 3‐h window prior to sleep onset was more closely examined using the wrist‐worn devices. Using the default thresholds from the ‘Cicada’ toolbox, the epoch‐by‐epoch accelerometry data were classified as either sustained inactivity, light activity, or moderate–vigorous activity, and the light data (from the silicon photodiode sensor on the device face) as either dim light, moderate light, or bright light. The proportion of time spent in each activity or light condition was calculated per evening, averaged across the study period per participant, and analysed for sex differences using linear mixed models (including participant identifier as a random intercept, sex as the main predictor, and start‐time of pre‐sleep window and timepoint as covariates).

3. RESULTS

Female athletes reported poorer mental health (higher APSQ) relative to male athletes (B = 3.47 ± 1.13, p = 0.003). There were no sex differences in TST (B = 12.05 ± 6.73, p = 0.08) and SE (B = 0.58 ± 1.10, p = 0.60). Female athletes had later circadian‐related timing on all three measures relative to male athletes (rMEQ, B = −1.69 ± 0.68, p = 0.02; MST, B = 0.44 ± 0.13, p < 0.001; DLMO, B = 0.47 ± 0.20, p = 0.02). There were no sex differences in phase angle (B = −0.09 ± 0.18, p = 0.63) and SRI (B = 1.41 ± 1.35, p = 0.30). See Table 1 and Figure 1.

TABLE 1.

Descriptive statistics for the sample and split by sex.

Variable, mean (SD) Whole sample Male athletes Female athletes
APSQ score 15.59 (5.78) 14.09 (4.03) 17.66 (7.09) **
TST, h ± min 7.11 (33.19) 7.02 (31.5) 7.22 (34.47)
SE, % 82.38 (5.16) 82.07 (4.87) 82.79 (5.55)
rMEQ score 16.23 (3.36) 16.99 (2.86) 15.19 (3.73) *
MST, clock time ± min 3:07 a.m. (0:38) 2:55 a.m. (0:29) 3:23 a.m. (0:43) ***
DLMO, clock time ± min 8:28 p.m. (0:56) 8:14 p.m. (0:52) 8:47 p.m. (0:55) *
Phase angle, h ± min 2.49 (47) 2.55 (46) 2.42 (48)
SRI score 80.25 (7.13) 79.69 (7.56) 80.97 (6.53)
Open in a new tab

Note: Presented are raw mean (SD). Sex differences in each variable investigated via separate linear mixed models for each outcome, with sex as a fixed effect, timepoint as a covariate, and participant identifier as a random effect. Significant sex differences are bolded and indicated by stars.

Abbreviations: APSQ, Athlete Psychological Strain Questionnaire; DLMO, dim‐light melatonin onset; MST, mid‐sleep time; rMEQ, reduced Morningness–Eveningness Questionnaire; SE, sleep efficiency; SRI, Sleep Regularity Index; TST, total sleep time.

*

p < 0.05.

**

p < 0.01.

***

p < 0.001.

FIGURE 1.

FIGURE 1

Open in a new tab

Raw data distributions for each measure plotted as violin plots with bootstrapped means and confidence interval error bars for the male and female groups separately. Asterisks indicate significant differences between groups; * p < 0.05, ** p < 0.01, *** p < 0.001. DLMO, dim‐light melatonin onset; SRI, Sleep Regularity Index.

Sex did not interact with TST to predict mental health, and TST did not predict mental health (B = −0.01 ± 0.01, p = 0.49). There was a significant interaction between sex and SE in the prediction of mental health (B = −0.49 ± 0.19, p = 0.01). Only among female athletes was lower SE associated with poorer mental health (B = −0.36 ± 0.13, p = 0.01). Only at <85% SE was there a sex difference in mental health (Figure 2a).

FIGURE 2.

FIGURE 2

Open in a new tab

Four models with significant interactions, indicating different relationships between sleep and circadian rhythms to mental health based on sex. When plots are viewed online, the slope for female athletes is indicated in dark orange, and that for male athletes in dark blue. Dashed vertical lines indicate threshold values yielded from Johnson–Neyman analyses. (a) <85% SE there is a sex difference in APSQ; (b) for rMEQ values <13.4 and >19.0 there is a sex difference in APSQ; (c) for all values of phase angle there remains a sex difference in APSQ; (d) below an SRI of 85 there is a sex difference in APSQ. APSQ, Athlete Psychological Strain Questionnaire; rMEQ, reduced Morningness–Eveningness Questionnaire; SE, sleep efficiency; SRI, Sleep Regularity Index.

There was a significant interaction between sex and rMEQ when entered as a quadratic polynomial in the prediction of mental health (B = 0.21 ± 0.06, p < 0.001). Only among female athletes was there a U‐shaped association between rMEQ and mental health (B = 0.20 ± 0.04, p < 0.001), i.e., those closer to ‘definitely morning’ or ‘definitely evening’ were at the greatest risk of poor mental health. Only for rMEQ values of <13.4 and >19.0 was there a sex difference in mental health (Figure 2b).

Sex did not interact with MST to predict mental health, and MST did not predict mental health (B = 1.42 ± 0.81, p = 0.08). Sex did not interact with DLMO to predict mental health, and DLMO did not predict mental health (B = 0.47 ± 0.53, p = 0.38).

There was a significant interaction between sex and phase angle when entered as a quadratic polynomial in the prediction of mental health (B = 1.73 ± 0.79, p = 0.03). Only among female athletes was there a U‐shaped association between phase angle and mental health (B = 1.60 ± 0.53, p = 0.003), i.e., those with the smallest or longest phase angles (i.e., largest circadian misalignment) were at the greatest risk of poor mental health. For all participant phase‐angle values there was a sex difference in mental health (Figure 2c).

There was a significant interaction between sex and SRI in the prediction of mental health (B = −0.33 ± 0.13, p = 0.01). Only among female athletes was lower SRI associated with poorer mental health (B = −0.42 ± 0.10, p < 0.001). Only below an SRI of 85 was there a sex difference in mental health (Figure 2d).

3.1. Post hoc analyses

Female athletes reported more strain regarding Self‐Regulation (mean [SD] 7.53 [3.41]) relative to male athletes (mean [SD] 5.78 [SD] 1.92; p = 0.002). Female athletes reported more strain regarding Performance (mean [SD] 7.72 [3.62]) relative to male athletes (mean [SD] 5.99 [2.01]; p = 0.003). There was no sex difference in strain regarding External Coping (mean [SD] in female athletes 2.40 [1.31]; male athletes, 2.33 [0.78]; p = 0.87).

Female athletes, relative to male athletes, spent 21% less time in sustained inactivity (43% versus 64%) and 23% more time in light activity (54% versus 31%) during the 3‐h pre‐sleep window (all p < 0.001). There was no sex difference in time spent in moderate–vigorous activity (3% for females versus 5% for males; p = 0.36). Female athletes, relative to male athletes, spent 2% less time in dim light (84% versus 86%) and 2% more time in moderate light (13% versus 11%) during the 3‐h pre‐sleep window (all p = 0.03). There was no sex difference in time spent in bright light (3% for both females and males; p = 0.48).

4. DISCUSSION

The present study found that female athletes playing elite ARF are at a greater risk of poor mental health and show later circadian‐related timing relative to male athletes. We also found that specific sources of sleep and circadian disruption are associated with mental health only among female athletes, and not male athletes.

This elite athlete sample differed on more than just sex: the male athletes compete in ARF professionally (full‐time) while the female athletes compete semi‐professionally (part‐time). The AFLW athletes are paid significantly (roughly eight times) less than AFL athletes (Dixon, 2022). Consequences of this pay inequity for female athletes may include financial strain, juggling second jobs, managing studies, constrained training schedules, fewer specialised staff supports, limited access to resources, and contract insecurity (O'Brien et al., 2023). These economic, psychosocial, and environmental factors provide important context to the discussion of our findings.

The present study found no sex differences in sleep duration or quality via actigraphy, surprisingly in contrast to Roberts et al. (2021) who found elite female ARF athletes had lower sleep duration and quality. The semi‐professionalism of the AFLW was hypothesised to contribute to a sex difference in sleep due to the likely need to balance evening training schedules with other non‐athlete roles that supplement financial or vocational security (O'Brien et al., 2023). The finding of no sex difference contrasts literature in both athletes (Vlahoyiannis et al., 2021) and the general population (Carrier et al., 2017; Mong & Cusmano, 2016) that indicate females typically sleep longer and more efficiently according to actigraphy—even if mechanisms for this remain uncertain. Outside of a descriptive comparison between studies, future research could specifically investigate whether female elite semi‐professional ARF athletes are at a greater risk of poor sleep relative to their non‐ARF or non‐athlete peers. Accounting for lifestyle factors such as work, family, and free time has been shown to reduce the magnitude of sex differences in sleep (Burgard & Ailshire, 2013). It is possible that semi‐professional female ARF athletes have lifestyle factors that reduce sleep duration and quality comparably to that of full‐time professional males.

The female athletes in the present study had later circadian‐related timing subjectively (preference), objectively (rest‐activity), and endogenously (melatonin). Findings in the general population indicate female adults generally have more advanced sleep and melatonin rhythms relative to males (Cain et al., 2010; Roenneberg et al., 2004). However, circadian‐delaying consequences of the semi‐professional lifestyle were hypothesised for our female athletes in‐part due to the predicted sleep–wake delay attributed to their later training times (Lastella et al., 2015). Later exposure of exercise and light can increase evening activation and delay circadian rhythms (Facer‐Childs et al., 2019; Haynes et al., 2016). The post hoc analyses demonstrated that the female athletes spent less time inactive and less time in dim light in the 3‐h prior to sleep. It is likely that morning ARF training contributed to advanced circadian timing among males, while evening training (and thus evening activity and light) contributed to delayed circadian timing among females in the present study. It is also possible the distributions of chronotypes in the male and female squads represent self‐selection (Lastella et al., 2016) or even habituation to the training schedule. Habituation may also explain why elite athlete groups have similar synchrony of their behavioural and biological timing (i.e., no sex difference in circadian alignment).

Our sample of female athletes reported a greater risk of poor mental health relative to male athletes, which is consistent with previous literature (Reardon et al., 2019; Walton et al., 2021). Our finding therefore supports existing calls for promoting female athlete mental health, although sex differences in social desirability, mental health stigma, and symptom reporting must also be considered (suggesting male athletes may under‐report; Poucher et al., 2021). The post hoc analyses demonstrated that difficulties with self‐regulation and concerns regarding performance are driving the group difference in athlete mental health, and there was no sex difference in the use of external coping strategies. Therefore, our findings suggest mental health support for female semi‐professional athletes could be targeted towards coping with motivational stress and sport‐specific worries. Although further qualitative information was not collected, factors associated with the semi‐professionalism of the AFLW, such as contract insecurity, constrained training schedules, working second jobs, and having fewer staff supports (O'Brien et al., 2023), may have exacerbated distress.

Importantly, the present study demonstrates sleep and circadian disruption are uniquely associated with poor mental health among female semi‐professional, but not male professional, elite athletes. Low SE was associated with poor mental health among female athletes, which aligns with the established link between sleep and mental health symptoms (Foster, 2020). Circadian misalignment via phase angle (too short or too long) and low sleep regularity was associated with poor mental health among female athletes. A mismatch in timing between endogenous cycles of sleep–wake signalling and exogenous exposures (e.g., light, temperature, exercise) is known to negatively impact mood (Foster, 2020). Female athlete chronotype (closer to extreme morning or extreme evening) was associated with poorer mental health. Evening types typically are at greater risk of poor mental health due to struggling with social jet lag (a discrepancy between preferred and obligated rest‐activity scheduling; Caliandro et al., 2021). Given the female athletes in the present study trained in the evening, the morning types may also have experienced social jet lag thereby increasing their risk of poor mental health. Sleep and circadian disruption being associated with poorer mental health only among females is somewhat consistent with the general population, though study designs are highly varied (Carpenter et al., 2021). As the main sex differences in the present study are unique from the extant literature, future research should consider whether the level of mental health risk for female semi‐professional athletes experiencing sleep or circadian disruption (i.e., the significant sex moderation) is similar or worse relative to that for female adults in the general population.

Interestingly, sleep and circadian rhythms were unrelated to male athlete mental health in our sample. Professional athletes have greater access to support structures and resources compared to their semi‐professional counterparts. An athlete's full‐time status may also reduce financial strain, allowing the athlete to concentrate fully on their sporting career. Support systems, greater access to resources, and financial stability may therefore help mitigate the harmful effects of sleep and circadian disruption on mental health.

This study may contain some limitations. First, the cross‐sectional nature of the data tempers causal inferences on the effects of sleep and circadian disruption on mental health. Due to the potential bidirectional links between sleep, circadian rhythms, and mental health (Foster, 2020), future research should test causational and mechanistic models using experimental and intervention studies; e.g., via specific zeitgebers to the circadian clock such as light, activity, and meal intake. Second, despite the focus on sex differences, findings are not necessarily attributable to biological sex alone. Controlled biomarker studies are needed to ascertain if genetically associated sex‐expression differences, such as the menstrual cycle or hormone levels, moderate the relationships between sleep or circadian rhythms and mental health. Future research can also include clinical interviews assessing mental health and a measure on social desirability to address the possible influences of sex differences in reporting tendencies and athlete‐specific conceptualisations of mental health. Third, the male and female ARF athletes differed markedly in training schedules, contract status, and financial situations (Black, 2023; Dixon, 2022). Though these differences reflect the true conditions these groups are typically exposed to, future research could investigate their relative and specific contributions to sleep, circadian, and mental health, as well as account for demographic considerations such as family/household makeups, education status, and other occupations. The sex differences in training schedules and pay currently in the ARF are likely the case for other national sporting leagues that prioritise greater funding for male/men's professional teams; however, future research should investigate if our findings generalise to other sporting codes.

5. CONCLUSION

The present study suggests that female elite ARF athletes, relative to males, are at a greater risk of poor mental health, particularly when experiencing low SE or regularity, circadian misalignment, or a somewhat extreme chronotype. The AFLW is currently a semi‐professional league that hopes to be full‐time professional by 2026 (Dixon, 2022), and other Australian sporting leagues are also committed to closing the pay gap between male and female athletes. The findings of this study provide compelling evidence for the need to assess sleep and circadian factors in female sports, and support sleep quality, sleep regularity, and circadian alignment as intervention targets to improve mental health.

AUTHOR CONTRIBUTIONS

Luis Mascaro: Investigation; data curation; formal analysis; visualization; writing – original draft; project administration. Josh Leota: Investigation; formal analysis; writing – review and editing. Daniel Hoffman: Investigation; data curation; software; writing – review and editing. Shantha M. W. Rajaratnam: Conceptualization; writing – review and editing. Sean P. A. Drummond: Conceptualization; writing – review and editing. Elise R. Facer‐Childs: Conceptualization; investigation; writing – review and editing; data curation; visualization; funding acquisition; supervision; methodology; resources.

FUNDING INFORMATION

Elise R. Facer‐Childs has received funding from the Department of Industry, Innovation and Science (Australian Government, ICG000899 and ICG001546), and St Kilda Football Club, and is currently supported by the Monash Lung and Sleep Institute, and a Science Industry Endowment Fund (SIEF) Ross Metcalf STEM + Business Fellowship administered by the Commonwealth Scientific Industrial Research Organisation (CSIRO). The funders did not influence the content of this manuscript.

CONFLICT OF INTEREST STATEMENT

Elise R. Facer‐Childs and Daniel Hoffman declare they are practitioners in elite sports. Elise R. Facer‐Childs has received research support or consultancy fees from the Monash Lung and Sleep Institute, Tempur Australia, Team Focus Ltd, British Athletics, Australian National Football League, Australian National Rugby League, Collingwood Football Club, Melbourne Storm Rugby Club, and Henley Business School, which are not related to this paper. Shantha M.W. Rajaratnam has served as a Programme Leader for the Cooperative Research Centre (CRC) for Alertness, Safety and Productivity, Australia; is a Director and Chair of the Sleep Health Foundation; has received grants from Vanda Pharmaceuticals, Philips Respironics, Cephalon, Rio Tinto, BHP Billiton and Shell; and has received equipment support and consultancy fees through his institution from Optalert, Compumedics, Teva Pharmaceuticals, Roche and Circadian Therapeutics, which are not related to this paper. Sean P.A. Drummond has received funding from the National Health and Medical Research Council (NHMRC) and the United States Department of Defence, received consultancy fees from Avecho Biotechnology Ltd, and participated on an advisory board for Zelda Therapeutics, which are not related to this paper. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

ACKNOWLEDGEMENTS

The authors would like to express gratitude to the athletes who participated in the study and those who supported their involvement. Luis Mascaro and Josh Leota receive financial support from the Australian Government through Research Training Programme scholarships. Open access publishing facilitated by Monash University, as part of the Wiley ‐ Monash University agreement via the Council of Australian University Librarians.

Mascaro, L. , Leota, J. , Hoffman, D. , Rajaratnam, S. M. W. , Drummond, S. P. A. , & Facer‐Childs, E. R. (2024). Disruptions to sleep and circadian rhythms are associated with poorer athlete mental health in female, but not male, elite Australian Rules footballers. Journal of Sleep Research, 33(6), e14186. 10.1111/jsr.14186

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

REFERENCES

  1. Adan, A. , & Almirall, H. (1991). Horne & Östberg morningness‐eveningness questionnaire: A reduced scale. Personality and Individual Differences, 12(3), 241–253. 10.1016/0191-8869(91)90110-W [DOI] [Google Scholar]
  2. Black, S. (2023). Happy days: The training shift that indicates a new AFLW era. AFL. https://www.afl.com.au/news/1002628/happy-days-the-training-shift-that-indicates-a-new-aflw-era. Updated 2023 May 29. Accessed June 25, 2023.
  3. Brager, A. J. , Gordon, J. A. , Rouska, A. , Moore, B. , & Mysliwiec, V. (2022). Circadian advantages in elite athletes. Current Sleep Medicine Reports, 8(4), 187–192. 10.1007/s40675-022-00239-0 [DOI] [Google Scholar]
  4. Burgard, S. A. , & Ailshire, J. A. (2013). Gender and time for sleep among U.S. adults. American Sociological Review, 78(1), 51–69. 10.1177/0003122412472048 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cain, S. W. , Dennison, C. F. , Zeitzer, J. M. , Guzik, A. M. , Khalsa, S. B. S. , Santhi, N. , Schoen, M. W. , Czeisler, C. A. , & Duffy, J. F. (2010). Sex differences in phase angle of entrainment and melatonin amplitude in humans. Journal of Biological Rhythms, 25(4), 288–296. 10.1177/0748730410374943 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Caliandro, R. , Streng, A. A. , van Kerkhof, L. W. M. , van der Horst, G. T. J. , & Chaves, I. (2021). Social jetlag and related risks for human health: A timely review. Nutrients, 13(12), 4543. 10.3390/nu13124543 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Carpenter, J. S. , Crouse, J. J. , Scott, E. M. , Naismith, S. L. , Wilson, C. , Scott, J. , Merikangas, K. R. , & Hickie, I. B. (2021). Circadian depression: A mood disorder phenotype. Neuroscience and Biobehavioral Reviews, 126, 79–101. 10.1016/j.neubiorev.2021.02.045 [DOI] [PubMed] [Google Scholar]
  8. Carrier, J. , Semba, K. , Deurveilher, S. , Drogos, L. , Cyr‐Cronier, J. , Lord, C. , & Sekerovick, Z. (2017). Sex differences in age‐related changes in the sleep‐wake cycle. Frontiers in Neuroendocrinology, 47, 66–85. 10.1016/j.yfrne.2017.07.004 [DOI] [PubMed] [Google Scholar]
  9. Cowley, E. S. , Olenick, A. A. , McNulty, K. L. , & Ross, E. Z. (2021). “Invisible Sportswomen”: The Sex Data Gap in Sport and Exercise Science Research. Women in Sport and Physical Activity Journal, 29(2), 146–151. 10.1123/wspaj.2021-0028 [DOI] [Google Scholar]
  10. Czeisler, C. A. , Duffy, J. F. , Shanahan, T. L. , Brown, E. N. , Mitchell, J. F. , Rimmer, D. W. , Ronda, J. M. , Silva, E. J. , Allan, J. S. , Emens, J. S. , Dijk, D. J. , & Kronauer, R. E. (1999). Stability, precision, and Near‐24‐hour period of the human circadian pacemaker. Science, 284(5423), 2177–2181. 10.1126/science.284.5423.2177 [DOI] [PubMed] [Google Scholar]
  11. Dixon, E. (2022). AFLW players get 94% pay rise in new CBA. SportsPro. https://www.sportspromedia.com/news/aflw‐cba‐collective‐bargaining‐agreement‐2022‐player‐salaries/ Updated 2022 May 20. Accessed June 25, 2023
  12. Facer‐Childs, E. R. , Middleton, B. , Skene, D. J. , & Bagshaw, A. P. (2019). Resetting the late timing of ‘night owls’ has a positive impact on mental health and performance. Sleep Medicine, 60, 236–247. 10.1016/j.sleep.2019.05.001 [DOI] [PubMed] [Google Scholar]
  13. Foster, R. G. (2020). Sleep, circadian rhythms and health. Interface Focus, 10(3), 20190098. 10.1098/rsfs.2019.0098 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Haynes, P. L. , Gengler, D. , & Kelly, M. (2016). Social rhythm therapies for mood disorders: An update. Current Psychiatry Reports, 18(8), 75. 10.1007/s11920-016-0712-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Johnson, P. O. , & Fay, L. C. (1950). The Johnson‐Neyman technique, its theory and application. Psychometrika, 15(4), 349–367. 10.1007/BF02288864 [DOI] [PubMed] [Google Scholar]
  16. Lastella, M. , Roach, G. D. , Halson, S. L. , & Sargent, C. (2015). Sleep/wake behaviours of elite athletes from individual and team sports. European Journal of Sport Science, 15(2), 94–100. 10.1080/17461391.2014.932016 [DOI] [PubMed] [Google Scholar]
  17. Lastella, M. , Roach, G. D. , Halson, S. L. , & Sargent, C. (2016). The Chronotype of elite athletes. Journal of Human Kinetics, 54(1), 219–225. 10.1515/hukin-2016-0049 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Lewy, A. J. , Lefler, B. J. , Emens, J. S. , & Bauer, V. K. (2006). The circadian basis of winter depression. Proceedings of the National Academy of Sciences of the United States of America, 103(19), 7414–7419. 10.1073/pnas.0602425103 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Mong, J. A. , & Cusmano, D. M. (2016). Sex differences in sleep: Impact of biological sex and sex steroids. Philosophical Transactions of the Royal Society B, 371(1688), 20150110. 10.1098/rstb.2015.0110 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Natale, V. , Esposito, M. J. , Martoni, M. , & Fabbri, M. (2006). Validity of the reduced version of the Morningness‐Eveningness questionnaire. Sleep and Biological Rhythms, 4(1), 72–74. 10.1111/j.1479-8425.2006.00192.x [DOI] [Google Scholar]
  21. O'Brien, W. , Taylor, T. , Hanlon, C. , & Toohey, K. (2023). Exploring (semi) professionalization in Women's team sport through a continuum of care lens. Journal of Sport Management, 37(2), 129–140. 10.1123/jsm.2022-0068 [DOI] [Google Scholar]
  22. Phillips, A. J. K. , Clerx, W. M. , O'Brien, C. S. , et al. (2017). Irregular sleep/wake patterns are associated with poorer academic performance and delayed circadian and sleep/wake timing. Scientific Reports, 7(1), 3216. 10.1038/s41598-017-03171-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Poucher, Z. A. , Tamminen, K. A. , Kerr, G. , & Cairney, J. (2021). A commentary on mental Health Research in elite sport. Journal of Applied Sport Psychology, 33(1), 60–82. 10.1080/10413200.2019.1668496 [DOI] [Google Scholar]
  24. Reardon, C. L. , Hainline, B. , Aron, C. M. , Baron, D. , Baum, A. L. , Bindra, A. , Budgett, R. , Campriani, N. , Castaldelli‐Maia, J. M. , Currie, A. , Derevensky, J. L. , Glick, I. D. , Gorczynski, P. , Gouttebarge, V. , Grandner, M. A. , Han, D. H. , McDuff, D. , Mountjoy, M. , Polat, A. , … Engebretsen, L. (2019). Mental health in elite athletes: International Olympic Committee consensus statement (2019). British Journal of Sports Medicine, 53(11), 667–699. 10.1136/bjsports-2019-100715 [DOI] [PubMed] [Google Scholar]
  25. Rice, S. , Olive, L. , Gouttebarge, V. , Parker, A. G. , Clifton, P. , Harcourt, P. , Llyod, M. , Kountouris, A. , Smith, B. , Busch, B. , & Purcell, R. (2020b). Mental health screening: Severity and cut‐off point sensitivity of the athlete psychological strain questionnaire in male and female elite athletes. BMJ Open Sport & Exercise Medicine, 6(1), e000712. 10.1136/bmjsem-2019-000712 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Rice, S. M. , Parker, A. G. , Mawren, D. , Clifton, P. , Harcourt, P. , Lloyd, M. , Kountouris, A. , Smith, B. , McGorry, P. D. , & Purcell, R. (2020a). Preliminary psychometric validation of a brief screening tool for athlete mental health among male elite athletes: The athlete psychological strain questionnaire. International Journal of Sport and Exercise Psychology, 18(6), 850–865. 10.1080/1612197X.2019.1611900 [DOI] [Google Scholar]
  27. Roberts, S. S. H. , Falkenberg, E. , Stevens, A. , Aisbett, B. , Lastella, M. , & Condo, D. (2021). The sleep of elite Australian rules footballers during preseason: A comparison of men and women. International Journal of Sports Physiology and Performance, 16(5), 641–646. 10.1123/ijspp.2020-0340 [DOI] [PubMed] [Google Scholar]
  28. Roenneberg, T. , Kuehnle, T. , Pramstaller, P. P. , Ricken, J. , Havel, M. , Guth, A. , & Merrow, M. (2004). A marker for the end of adolescence. Current Biology, 14(24), R1038–R1039. 10.1016/j.cub.2004.11.039 [DOI] [PubMed] [Google Scholar]
  29. Sargent, C. , Lastella, M. , Halson, S. L. , & Roach, G. D. (2016). The validity of activity monitors for measuring sleep in elite athletes. Journal of Science and Medicine in Sport, 19(10), 848–853. 10.1016/j.jsams.2015.12.007 [DOI] [PubMed] [Google Scholar]
  30. Stone, J. E. , Phillips, A. J. K. , Chachos, E. , Hand, A. J. , Lu, S. , Carskadon, M. A. , Klerman, E. B. , Lockley, S. W. , Wiley, J. F. , Bei, B. , Rajaratnam, S. M. W. , & for the CLASS Study Team . (2021). In‐person vs home schooling during the COVID‐19 pandemic: Differences in sleep, circadian timing, and mood in early adolescence. Journal of Pineal Research, 71(2), e12757. 10.1111/jpi.12757 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Vlahoyiannis, A. , Aphamis, G. , Bogdanis, G. C. , Sakkas, G. K. , Andreou, E. , & Giannaki, C. D. (2021). Deconstructing athletes' sleep: A systematic review of the influence of age, sex, athletic expertise, sport type, and season on sleep characteristics. Journal of Sport and Health Science, 10(4), 387–402. 10.1016/j.jshs.2020.03.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Walton, C. C. , Rice, S. , Gao, C. X. , Butterworth, M. , Clements, M. , & Purcell, R. (2021). Gender differences in mental health symptoms and risk factors in Australian elite athletes. BMJ Open Sport & Exercise Medicine, 7(1), e000984. 10.1136/bmjsem-2020-000984 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Windred, D. P. , Jones, S. E. , Russell, A. , Burns, A. C. , Chan, P. , Weedon, M. N. , Rutter, M. K. , Olivier, P. , Vetter, C. , Saxena, R. , Lane, J. M. , Cain, S. W. , & Phillips, A. J. K. (2021). Objective assessment of sleep regularity in 60 000 UK biobank participants using an open‐source package. Sleep, 44(12), zsab254. 10.1093/sleep/zsab254 [DOI] [PubMed] [Google Scholar]

Associated Data

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

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Articles from Journal of Sleep Research are provided here courtesy of Wiley

RESOURCES