Skating on thin ice? Mental health and well-being in women’s ice hockey

Population

This study adopted a purposive sampling with the eligibility criteria of playing in the SDHL. The players were recruited through their respective team managers. Forty-nine per cent (n=90) of the players had another nationality than Swedish.17 The mean age in the sample population was 22.3 years (SD 4.8 years) and ranged from 16 to 35 years. Eighty per cent (n=145) reported having dual careers and 30% (n=54) reported being injured.

Design

A cross-sectional online survey in Swedish and English was distributed to all SDHL players through their respective team managers. Of the 199 players in SDHL, 182 from all nine teams completed the survey (response rate 91%). Data were collected in February 2023 during the competitive season (before the play-offs started). Before the data collection took part all clubs in SDHL were informed about the study. After that information, a link to the survey was distributed from the researchers to the clubs with instructions to forward the link to the players in the clubs. A video was also recorded and distributed from the researchers to the population of players. In the video, information about the study and the ethical aspects, relevant for the study, were presented. Some of the clubs allocated time in connection to practice for the players, who agreed to take part, to complete the survey while other teams only distributed the link and asked the players to complete the survey in their free time. The participants gave informed consent before taking part in the study. Participation was voluntary, and participants could withdraw from the study at any time without receiving negative consequences. The study was conducted in accordance with the Declaration of Helsinki.18 The Strengthening the Reporting of Observational Studies in Epidemiology checklist was used to guide the reporting of the study.

Patient and public involvement

This study was performed in collaboration with the Swedish Ice Hockey Association, which wanted an increased understanding of MH promotion and prevention. The Ice Hockey Association were involved in designing the study and assisted during the different steps in the research process. More specifically, staff from the Ice Hockey Association were involved in the design of the study. Also, they facilitated contact with the clubs as well as with the league organisation. The study was founded by the Swedish Ice Hockey Association, but they had no access to the data and were not involved in the data analysis phase.

Measures

The survey consisted of questions on demographic variables (age, injury, dual careers). The variable age was reported as the number of ages. The variable injury was reported as having an injury or not. Dual careers refer to combining one’s sports career with education or work and were reported as having dual careers. The survey also consisted of the following validated self-report questionnaires:

Generalized Anxiety Disorder-7 was used to assess the severity of anxiety symptoms.19 Respondents scored seven items on how often they, during the last 2 weeks, experienced different states. The responses were given on a 4-point scale ranging from 0 to 3, ‘not at all’ to ‘nearly every day’. The scores were categorised into severity levels: minimal to mild (0–9 points) and moderate to severe (10–21 points) based on the clinical cut-off score of 10 points.

Patient Health Questionnaire-9 was used to assess the severity of depression symptoms.20 Respondents scored nine items on how often they, during the last 2 weeks, experienced different states. The responses were given on a 4-point scale ranging from 0 to 3, ‘not at all’ to ‘nearly every day’. The scores were categorised into severity levels: none to mild (0–9 points) and moderate to severe (10–27 points) based on the clinical cut-off score of 10 points.

The Athlete Sleep Screening Questionnaire is an athlete-specific sleep screening questionnaire.21 The questionnaire contains 16 questions, of which five questions assess sleep difficulties. Responses to questions on sleep difficulties were scored between 0 and 3 or 4 points, depending on the question. Total scores were categorised into severity levels: none to mild (0–7 points) and moderate to severe (8–17 points) based on the clinical cut-off score of 8 points.

The Mental Health Continuum Short Form was used to assess positive MH.22 Respondents scored 14 items on how often they, during the last 2 weeks, have experienced different states of well-being, both emotional, social (ie, ‘hedonic’) and psychological (ie, positive functioning). The responses were given on a 6-point scale ranging from 0 to 5, ‘never’ to ‘every day’. If respondents scored 4 or 5 on three hedonic items and six functioning items they were categorised as ‘flourishing MH’, which is referred as the ideal state.12 The remaining respondents were categorised as ‘non-flourishing MH’.

The Swedish Acceptance and Action Questionnaire assessed PF.23 Respondents scored six items on a 7-point scale ranging from 1 to 7, ‘never true’ to ‘always true’. The sum of scores was divided by the number of items. Lower scores indicate higher PF. No cut-offs were applied for this continuous variable.

The Multidimensional Scale of Perceived Social Support was used to assess PSS.24 The scale covers three dimensions of social support: family, friends and significant others. Respondents scored 12 items on a 7-point scale ranging from 1 to 7, ‘very strongly disagree’ to ‘very strongly agree’. Higher scores indicate higher levels of PSS. The sum of scores was further divided by the number of items. No cut-offs were applied for this continuous variable.

Statistical analysis

IBM SPSS (V.28.0) was used for the statistical analysis. Descriptive statistics of the dichotomised MH variables and their coexistence were calculated and presented as numbers and percentages. Multivariate binary logistic regression analyses were conducted to investigate associations between the dichotomised MH variables (positive MH, anxiety, depression, sleep difficulties) as dependent variables, and age, dual careers, injury, PSS and PF, respectively, as independent variables. Associations for each MH variable were calculated and presented separately, including ORs and a 95% CI.

Mental health

Among the respondents (n=182) in this study, 41.2% (n=75) reported reaching the clinical cut-off for at least one of the questionnaires (ie, anxiety, depression or sleep difficulties). More specific prevalence of symptoms of anxiety, depression and sleep difficulties is presented in table 1. Moreover, 10 of the players (5.5%) reported suicidal intentions. Comorbidities (ie, reaching clinical cut-off for multiple questionnaires) were reported by 19.2% (n=35) of the respondents. More details on the prevalence of comorbidities are presented in table 2.

The results on positive MH show that, among the respondents (n=181), 60.2% (n=109) reported having flourishing MH and that 39.8% (n=72) did not reach the criteria for flourishing MH. Additionally, 40.3% (n=73) reported flourishing MH without reaching clinical cut-off levels for symptoms of sleep difficulties, depression or anxiety. The remaining respondents were either not flourishing (39.7%, n=72) or flourishing with MHP symptoms (19.9%, n=36; table 3).

Associations with potential risk, protective factors and promotive factors

The results from the logistic regression analyses showed that lower PF was associated with higher odds of symptoms of anxiety (OR=2.46, 95% CI 1.66 to 3.65), depression (OR=2.75, 95% CI 1.80 to 4.18) and sleep difficulties (OR=1.76, 95% CI 1.26 to 2.45). Lower PF was also associated with lower odds of flourishing MH (OR=0.49, 95% CI 0.34 to 0.71). Higher PSS was associated with higher levels of flourishing MH (OR=3.49, 95% CI 1.93 to 6.31) and lower sleep difficulties (OR=0.49, 95% CI 0.30 to 0.80). PSS was not statistically significantly associated with anxiety or depression. No statistically significant associations were found for age, dual careers and injury. A more detailed summary of the results from the logistic regression analysis is described in table 4.

Mental health

In the present study, 41.2% of the players reported moderate or higher symptoms of at least one MHP (anxiety, depression or sleep difficulties). Compared with the prevalence of comorbidities at 19.2%, this illustrates that comorbidities were represented in almost half of the cases. This can highlight the importance of acknowledging comorbidities of anxiety, depression and sleep difficulties, which are MH states with strong positive relations shown in recent research on elite athletes.11 Notably, out of the 19.2% with comorbidities, 14.8% had sleep difficulties, indicating that sleep difficulties are a big part of comorbid cases and should be addressed. If not, comorbid MHP symptoms such as anxiety and depression will unlikely dissolve.1

Symptoms of sleep difficulties were most frequently reported, with a prevalence rate of 29.7% for moderate or severe symptom levels. This is somewhat higher than the percentages from a systematic review of athletes from various sports and contexts, showing a prevalence rate of sleep disturbances at 26% among female and male athletes.25 The result from the present study adds to the existing empirical evidence on the common prevalence of sleep problems in high-level athletes and the need to target them.10 Within the league it is a mix of games played between lunch time (game starts and 12:00) and the evening (19:00). There is also a mix of overnight stays versus travel back home the same night when playing away games. For the present study population, addressing sleep difficulties can be of certain interest due to the inevitable travel associated with competition, which in previous consensus statements has been presented as a primary risk factor for sleep disturbances. Additionally, it is highlighted in the consensus statements that sleep in elite athletes is an essential part of both their mental and physical health.10

For anxiety symptoms, 20.9% of the respondents in the present study reported moderate or higher levels. For symptoms of depression, 18.1% reported moderate or higher levels. The percentages from the present study correspond to a previous Swedish study, which included a similar population consisting of female players in top leagues in soccer, handball and ice hockey. The COVID-19 pandemic was present in that study, and COVID-19-related worries were associated with anxiety and depression among the athletes.2 COVID-19 restrictions were not present when the present study was conducted. Instead, it was conducted during the competitive season, which can come with different stressors that may affect MH, such as travel, external expectations and media.26 Also, performance-related stressors might be of particular importance given the time of the season when the study was conducted.27

Among the respondents in the present study, 60.2% reported flourishing MH, meaning experiencing a good MH with optimal functioning, including emotional, social and psychological well-being.12 22 The percentage in the present study is notably higher than in a Canadian study, showing 44–45%.28 It should be noted that the population in the Canadian study were both male and female student-athletes from a wide variety of sports, in comparison with the present study with athletes from a specific sport without the specific university context, meaning the contextual factors differ between the populations which could influence the MH prevalence.29 Even though the percentages from the present study could be interpreted as relatively high, it should be noted that the ISSP states that the ideal state is experiencing flourishing MH without symptoms of MHP (ie, symptoms of anxiety, depression or sleep problems),12 which was reported among about 40.3% in the present study. As follows, 59.7% were not reaching the ideal state of MH, either due to not having a flourishing MH or the presence of symptoms of MHP.

Associations with risk, protective factors and promotive factors

In this study, PF was the variable with the most associations with MH and the symptoms of MHP. More specifically, higher levels of PF were associated with flourishing MH, and lower levels of PF were associated with symptoms of depression, anxiety and sleep difficulties. This indicates that PF can be suggested as a promotive and protective factor in this context. The negative association between PF and anxiety corresponds with previous research with male elite soccer players in Portugal participating in an intervention inspired by acceptance and commitment therapy programmes.16 Other research has further shown that PF enhances MH directly, both decreasing anxiety and depression, as well as enhancing social functioning,30 which may explain the associations found in the present study. The findings in this study can add to recent research highlighting the relevance of working towards promoting PF in athlete populations. For instance, a randomised controlled trial study with female student-athletes has shown that participation in interventions to increase PF reduces psychological, emotional and behavioural concerns.31

The results of the present study also showed that higher PSS was associated with higher odds of flourishing MH. Similarly, a recent Danish study showed that elite athletes with above-average mental well-being received higher support than athletes with low mental well-being.32 The associations presented in the present study can be understood through recent research with a German working population where it was suggested that social support facilitates the enforcement of mental well-being, even when multimorbidity is present.33 The association between lower PSS and symptoms of sleep difficulties was a more unexpected finding. However, this correlation corresponds with recent research on Dutch elite athletes.34 The association between PSS and sleep is theorised by other scholars with a suggestive explanation that ‘social support may influence sleep through the meaning that people derive from their supportive social ties and how meaning promotes better mental health’35 (p 153). The present study supports previous research on elite athletes presenting low social support as a risk factor and high social support as a protective factor for MH,14 as well as its positive association with mental well-being,32 suggesting that it can also be understood as a promotive factor.

In the discussion of protective factors, it is important to state that even if we, in the current study, have focused on athlete (PF) as well as interpersonal (social support) aspects, organisational as well as societal factors also influence both the risk for MHP as well as the athletes’ chances of coping with potential stressors. It is, therefore, important for different stakeholders to work actively with strategies that could help the athlete to cope with the potential stressors to facilitate health and performance.

Clinical implications

This study mapped MH among players in the SDHL. The prevalence of symptoms of anxiety, depression and sleep difficulties reported in this study should be acknowledged, considering that the mapping and early identification of MHP is an important prevention strategy.12 Likewise, the prevalence rates also indicate where further assessment and management are needed. Moreover, this study gives insight into the prevalence of a subideal MH state, indicating a promotion gap. Additionally, this study identifies PF and PSS as protective and promotive factors that could be beneficial to prevent MHP and promote MH. The findings in this study can be valuable for promoting MH in this elite athlete population, which in turn can be understood as a key resource for a healthy and prosperous sports system.29 Different initiatives to facilitate MH are important to consider for stakeholders.13 Such initiative could, for example, be to educate stakeholders about both how to facilitate MH and how to identify signs of MHP. Also, for the organisation to screen for MHP, using, for example, the IOC Sport Mental Health Assessment Tool 136 to detect athletes who need help can be of great importance. In this process, it is also important for the organisation to have connection with psychologists to refer athletes who need treatment for MHP. This can, in turn, be valuable, considering that sport is a powerful platform for promoting MH and gender equality, especially within high-level sports where such work can serve as a model for society at large.1 37

Limitations and strengths

Considering the cross-sectional design of this study, causal relationships cannot be determined. This study only investigated the point prevalence of MH status during the competitive season. It should be noted that the prevalence could differ during the off-season. Therefore, it can be suggested that future research observing MH status and associated factors over time is needed. A strength of this study was the high response rate, contributing to higher quality and accuracy of the data, making the sample much more representative. Another strength is that both MI and MH were investigated, contributing to a fuller picture of the MH status of the studied population.