An efficacy trial of a brief group based, single session intervention delivered through Australian community sport clubs to reduce mental illness stigma and increase help-seeking intention

ABSTRACT

The Australian NRL’s State of Mind programme was evaluated; specifically, a 1.5-hour intervention for increasing knowledge of mental illness, reducing attitudinal components of mental illness stigma and increasing help-seeking propensity in a sample drawn from amateur rugby league clubs in Queensland. The National Rugby League (NRL) is the pre-eminent rugby league club competition in Australia and New Zealand. Participants, 16 years of age and older, completed a pre- and post-intervention survey capturing four mental illness knowledge measures, four attitudinal measures of stigma and one measure of help-seeking propensity. Results demonstrated that 1.5 weeks after intervention, the 1.5-hour targeted intervention associated with significant meaningful improvements in mental health knowledge, attitudinal components of mental illness stigma and help-seeking intention.

KEYWORDS:

• Help-seeking
• mental health literacy
• mental illness
• sport
• stigma

Mental health continues to present as a major challenge in Australia, with associated substantial negative implications and burden on the economy, health services, state and federal budgets, as well as individuals’ quality of life. Within the Australian context, the financial impacts of mental ill health were conservatively estimated by the Productivity Commission (2020) to be between $200–220 billion per year. It is likely that the impacts of COVID-19 will have added to such estimates. Globally, depression and anxiety’s impact on productivity costs the international economy US$ 1 trillion yearly; further, the loss of productivity due to all mental health conditions is projected to reach US$6 trillion annually by 2030 (Lancet Global Health, 2020). In 2017–2018 an estimated one in five (20.1%), or 4.8 million, Australians had a mental or behavioural condition representing an increase from 18% in 2014–15. Estimates also include that in the same financial year (2017–2018) 3.2 million Australians (13.1%) had an anxiety-related condition and 1 in 10 (10.4%) had depression or feelings of depression (Australian Bureau of Statistics, 2017–18). Associated with mental ill-health, intentional self-harm in 2016 accounted for over one-third of deaths (35.4%) among Australians 15–24 years of age, over a quarter of deaths (28.6%) among those 25–34 years of age, and 16.0% of those people 35–44 years of age (Australian Bureau of Statistics, 2016).

The mechanisms that contribute to the high prevalence of mental health concerns in developed nations, such as Australia, are complex, as such are factors associated with effective management and treatment of mental health challenges. An acknowledged key contributor to effective management of mental health issues is early detection and treatment (Costello, 2016). However, research indicates that many people do not access the mental healthcare they need (Gulliver et al., 2012; Wright et al., 2006). Proposed barriers to help-seeking behaviour include low mental health literacy, cost, and fear of being stigmatized (Coates et al., 2018; N. J. Reavley & Jorm, 2011; Wright et al., 2006). As highlighted by the World Health Organization, ‘the single most important barrier to overcome in the community is the stigma and associated discrimination towards persons suffering from mental and behavioural disorders’ (World Health Organisation, 2001, p. 98). Unfortunately, despite mental health stigma receiving increasing attention, attitudes towards mental health help-seeking have reportedly become increasingly negative (Mackenzie et al., 2014). Accordingly, to better ensure that people access the mental health they need, national and international calls have been made for actions and strategies within communities to enhance mental health literacy and reduce mental health stigma (Coates et al., 2018; Jorm, 2012; Sane Australia, 2013).

Stigma is clearly a barrier to people’s mental health help-seeking behaviours, with large implications for effective early mental health detection and treatment. Stigma, defined by Goffman (1963, p. 13) as ‘an attribute that is socially discrediting’, is cited as one of the key deterrents to help-seeking behaviour by contributing to low self-esteem, rejection, and isolation for those suffering from a mental illness (Kitchener et al., 2017; P. W. Corrigan et al., 2014; World Health Organisation, 2014). Stigma is often conceptualized as a combination of knowledge, attitudinal, and behavioural components that manifest as stereotypes, prejudice, and discrimination, respectively (Fox et al., 2017; Hanisch et al., 2016; P. W. Corrigan et al., 2014). Inaccurate knowledge about mental illness and available treatments, such as ‘people with mental illness are dangerous’ or ‘they can’t be helped’ are examples of stereotypes (Gronholm et al., 2017; Hanisch et al., 2016). Prejudice, such as disgust and other negative emotions, society at large can hold towards people with mental illnesses represents the attitudinal component of stigma (Thornicroft, 2007). The behavioural component encompasses discrimination and refers to actions that disadvantage people with mental illness (Hinshaw & Cicchetti, 2000). The stigma surrounding mental illness at the person-level (as opposed to structural or institutional) is termed public stigma (stigma endorsed by the general public) or self-stigma (the internalized acceptance of the stereotypes, prejudice and discrimination endorsed by the public; Pescosolido & Martin, 2015).

P. W. Corrigan et al. (2014) proposed a relationship between mental illness stigma at the person level (public stigma and self-stigma) and help-seeking where higher levels of stigma lead to lower levels of help-seeking intentions. P. W. Corrigan et al. (2014) also accounted for three moderating variables: the knowledge, culture, and networks of people that can influence both stigma and help-seeking behaviour. A lack of knowledge about mental illness is thought to lead to prejudicial attitudes, discriminatory behaviour, and an under-recognition of mental illness symptoms (Thornicroft, Rose, Kassam, & Sartoius, 2007). Differences in culture (referring to shared beliefs, values, and norms of a specific racial or ethnic group) have been associated with differences in the form stigma takes (Abdullah & Brown, 2011). One’s network, comprising their family, friends, and colleagues can also impact their help-seeking behaviour. Closer networks consisting of those with more direct experience of mental illness are more likely to be a resource that promotes help-seeking behaviour (P. W. Corrigan et al., 2014).

Targeted stigma reduction programmes have been used successfully to reduce stigma in important groups and organizations such as health care providers and police officers who have frequent contact with people experiencing mental illness (Gronholm et al., 2017; Hanisch et al., 2016). Mental health literacy programmes such as the Mental Health First Aid (MHFA) programme, which contains a stigma reduction element, have also been found to successfully reduce stigma (Hadlaczky et al., 2014; Kitchener & Jorm, 2002; Svensson & Hansson, 2014). More recently, targeted stigma reduction programmes have been used in groups that are known to be at risk of higher levels of stigma and exhibit less help-seeking behaviour; perhaps because of the poor rates of male help-seeking, men and masculine environments have become a focus.

Sporting clubs in Australia have traditionally been a community hub and, especially for males, an important social network (Hughes, 2012). An example of previous research conducted through a community hub, in rural Victoria MHFA training was delivered to 36 coaches and club leaders (35 males) of Australian Rules Football (AFL) clubs in response to a number of suicides in the community (Pierce et al., 2010). The aim was to improve the coaches’ ability to assist players, and it was hoped that aspects of the training would be disseminated through the clubs. Whilst evaluation found improvements in knowledge, attitudinal and behavioural measures of the coaches and club leaders, no significant improvements were found in the wider playing community that did not receive the training (n = 275; Pierce et al., 2010).

More recently, a mental health literacy and resilience intervention named Ahead of the Game has been studied across a number of nonelite organized sports clubs in Australia (e.g. soccer, Australian Rules football, rugby league, and basketball; Published across multiple studies including Hurley et al., 2018, 2021; Vella et al., 2021). This intervention includes targeted adolescent (player), parent, and coach components, with a mixture of short 45-min workshops and an additional 11 self-paced online module for coaches. Importantly, the intervention was designed using a Community-Based Participatory Research (CBPR) framework, meaning that engagement and collaboration were central design pillars. Findings have been promising, with improvements in outcomes, such as depression and anxiety literacy, intentions to seek help from formal sources, confidence to seek mental health information, resilience and even well-being for the adolescent component (Vella et al., 2021). Similar findings were reported for the pilot of the parental component (Hurley et al., 2018) wherein the experimental group demonstrated greater increases in depression literacy, anxiety literacy, knowledge of help seeking options, and confidence to assist an adolescent experiencing a mental health disorder, compared to the control group. Favourably, changes in the experimental group were maintained at 1-month follow-up. Replication of the parental pilot did not support all components, but nevertheless important improvements, such as increased likelihood of seeking formal help for themselves, increased confidence and knowledge to help someone experiencing a mental health disorder, reduced psychological distress and greater perceived support from other parents within their sport club, were documented compared to the control group (Hurley et al., 2021). Taken together, these findings suggest that a brief educational intervention delivered in community sports clubs can improve components of adolescent players’ and parents’ mental health literacy.

Despite the MHFA programme and other programmes run within Australia achieving significant improvements in knowledge and attitude variables related to stigma and help-seeking (e.g. the workplace programme Looking After Wellbeing at Work [LWW] and a psychoeducational programme used by the Royal Navy, Trauma Risk Management [TRiM]), these programmes typically take at least 2 days to deliver (Gould et al., 2007; Moffit et al., 2014). This poses challenges and feasibility issues when attempting to reach a large number of participants, for example, in amateur sporting clubs due to time and fiscal constraints. Perhaps because of this, MHFA training tends to be delivered in workplaces and professional contexts and often the participants in evaluation studies are predominantly female and hail from care provision sectors, such as health, ageing, family, and community services (Hadlaczky et al., 2014).

In 2015, the Australian National Rugby League (NRL) in partnership with Lifeline, Kids Helpline, Headspace, and The Blackdog Institute commenced a campaign to reduce stigma towards mental illness and stimulate help-seeking behaviour within the rugby league playing community (NRL, 2022). In consultation with their coalition partners, the NRL decided to focus the campaign on stigma around depression, substance use, and suicide. In 2017, 89% of registered club participants were men, making help-seeking a priority (NRL, 2022). In the year to September 2016, the Australian Sports Commission estimated that 247,883 Australian adults and children played rugby league in Australia at club level. To be able to reach a substantial portion of such a large population, there is a need to determine whether brief group-based, single session interventions delivered through sport community clubs can reduce mental illness stigma and increase help-seeking intentions. An integral aspect of the NRL and partners’ mental health help-seeking campaign, a group-based, 1.5-hour mental health stigma reduction intervention was created to be delivered to amateur rugby league clubs throughout Queensland, Australia. The intervention was based on education and contact principles. Education refers to replacing myths, stereotypes and inaccuracies in beliefs about mental illness with accurate knowledge (Knifton et al., 2009). Contact approaches challenge stigmatizing attitudes by incorporating and utilizing direct interactions with people with mental illness or people that have recovered from mental illness (P. Corrigan, 2004; P. W. Corrigan et al., 2001). Education and contact are often used in targeted interventions to address stigma at the person-level, and contact has been found to enhance the effectiveness of educational approaches (Gould et al., 2007). With reference to culture and networks outlined in P. W. Corrigan et al. (2014), research consistently implicates traditional masculine ideals and norms, such as stoicism and self-reliance, as well as avoidant coping strategies such as substance use as barriers to help-seeking in males (Lynch et al., 2018; Oliffe et al., 2016; Player et al., 2015). Therefore, the content of the intervention for a specific target population needs to fully consider and acknowledge these traditional ideals and norms. This NRL and partners’ State of Mind programme intervention is the focus of this research paper. The broader State of Mind programme also included elements such as club action plans; developed and implemented by club leaders with NRL support, recognition as a State of Mind club, use of current and former elite players with lived experience of mental ill health, and targeted media campaigns coinciding with peak viewer times such as State of Origin matches.

Research aims

Acknowledging the key advances already made by previous research yet based on the limitations in our current understanding, the aim of this research was to assess whether mental health knowledge, mental health stigma attitude, and help-seeking intention significantly improve from pre- to post-participation in a single session intervention delivered through a community sport group. While behaviour is also recognized as an import component of stigma, due to the methodological requirements to achieve valid and reliable measures of actual behaviour, rather than self-reported expected and actual behaviour, this investigation focussed upon mental health knowledge, mental health stigma attitude, and help-seeking intention aspects of stigma.

Materials and methods

Design

A data linkage approach was adopted where data previously collected before and after NRL and partners’ 1.5-hour group-based mental health stigma reduction intervention sessions were interrogated. To address the research question of how mental health knowledge, stigma attitude and help-seeking intention differed from pre- to post-invention, a within-participation design was used with no control group. The absence of a control group was based on ethical reasoning associated with the potential detrimental effects of withholding the invention from selected participants.

Recruitment

The sample comprised of members of amateur rugby league clubs being coaches, committee members, managers, senior players, referees, and board members. The participants were required to be 16 years of age and older. The recruitment of such a broad range of club membership positions was purposeful to enable the examination of the intervention efficacy across the breadth of such a community group. Ethical approval for analysis of de-identified data was obtained from the Human Research Ethics Committee of University of XXXXX. Ethical approval for the original data collection was obtained through a peer-review process between researchers at yourtown and the NRL. Attendance at the voluntary intervention sessions was promoted by the NRL to ‘grass root’ club presidents and secretaries who organized the hosting of the event within their own clubs; the sessions were held through 40 rugby league clubs throughout Queensland, Australia. These 40 clubs were recruited as were the first 40 clubs to take part in the intervention series. Regional/remote clubs with greater prevalence of First Australians took part in a more culturally appropriate version of the intervention at a later time. For this reason, such clubs were not included in this sample. Grass root sport clubs are generally considered the foundation for sport in Australia, they generally cater for mass participation across all ages and are predominantly managed by volunteers. To help facilitate club members’ attendance and engagement with the sessions, the NRL encouraged clubs to invite senior club members and senior players.

Attendance at the intervention session was independent of agreeing to take part in the research; people were welcome to attend the sessions without any expectation they would take part in the research. Recruitment for research participants was conducted at the start of the intervention session. Placed on each chair of the session location (usually, the rugby league clubhouse), was a research pack containing research participant information sheet, informed consent, pre-intervention survey, envelope and pen. Attendees were welcome to sit at any available seats and were encouraged to peruse the pack as they waited for the session to start. As part of the introduction to the session, presenters also described the purpose of the research, what participation would involve and the informed consent content and process; there were also opportunities for further questions regarding potential participation. The voluntary and anonymous nature of participation was highlighted, as was the right to withdraw from the research without reason or consequence. Attendees who volunteered to participate in the research, indicated their agreement via signing the informed consent and placing it in the provided research pack envelope. To ensure that attendees did not feel coerced into participating through social desirability affects, those who did not want to participate were instructed to also place the informed consent back into the envelope, but unsigned. As such, others in the room would not be able to see who had and had not agreed to participate as both behaviours appeared similar. Only the research team had access to the signed informed consents.

Procedure

Following informed consent to take part, participants were then invited to complete a pre-intervention paper-based survey, before sealing it in the provided envelope. Non-participating attendees also sealed their non-completed survey in their provided envelope in the same fashion to prevent overt indications of their decision not to take part. All envelopes were then collected. The delivery of the 1.5-hour session then commenced. At the conclusion of the presentation, participants who had agreed to take part in the research and supplied their email address were informed that a link would be emailed to them within 1 week inviting them to complete the online follow-up survey. The mean average duration from session to completion time of post-intervention surveys was 1.5 weeks.

Intervention design

The intervention, called the State of Mind programme, was based on education and contact principles and designed to be delivered in a single 1.5-hour session. The rationale for the design of the intervention was to reduce stigma towards mental illness and stimulate help-seeking behaviours within the rugby league playing community. The educational element of the presentation included information providing a snapshot of mental health facts in Australia, signs, symptoms, and causes of mental ill health (depression and suicide), what to do if participants or someone they knew were ‘not OK’, tips for maintaining good mental health, and a list of available supports in the community. The material was based on the research and resources provided by the Black Dog Institute, headspace, Kids Helpline (yourtown) and Lifeline. The intervention began with a 5-min welcome video outlining the aims of the overall State of Mind campaign featuring current NRL players sharing stories about mental well-being. A second 5-min video was then screened sharing the story of ‘lived experience’ with mental illness and seeking help as told by a current NRL player. The delivery of the educational material followed, incorporating the PowerPoint presentation as a central focus. The final video included clips of past and present NRL players (both men and women) sharing their personal tips for mental wellbeing and how they incorporate those practices into their daily lives. Throughout the presentation, questions from the audience were encouraged at all times and the presenter regularly asked questions of the audience. The interested reader is directed to this website for further information https://www.nrl.com/community/state-of-mind/tools-and-resources/

Measures

Researchers developed a bespoke survey, using items from published scales wherever possible and creating bespoke items where necessary, to capture four knowledge elements of mental health, four attitudinal elements of stigma and one measure of help-seeking intentions, which the training intervention was specifically designed to address (see Table 1). In line with feedback from the NRL who considered the language of published items to be too formal for their constituents, the language and phrasing of some items was adapted to be more user-friendly. In some instances, we had to make compromises between fully validated scales with strong psychometric properties yet selecting measures that were appropriate for the participants, such as being shorter in length or with language suitable for basic literary levels. Although a risk to the confidence of the findings, this was essential for showing due respect to our participants and maintaining trust with our research partners. The survey then underwent initial psychometric testing on a sample of four clubs (n = 47). In accordance with the intended administration methods, the pre-intervention survey was administered in person as a paper-based measure, with the post-intervention online survey being administered via an email link sent to participants within 1 week of completing the first survey.

Table 1. Survey-dependent variables with pre- and post-intervention internal consistency scores.

Category	Dependent Variables
Mental health knowledge	Mental health general knowledge†
	Mental health signs and symptoms knowledge (.808, .948)
	MHFA for depression and substance abuse knowledge (.628, .610)
	MHFA for suicide knowledge (.619, .676)
Stigma attitudes	Self-efficacy to support‡
	Public stigma prejudice (.541, .691)
	Help-seeking propensity (.761, .722)
	Prejudice to help-seeking (.819, .853)
Help-seeking intentions	Help-seeking intentions‡
Descriptive information	Gender
	Age
	Role at club
	Previous mental health training

Note: Pre- and post-intervention internal consistency Cronbach α in parenthesis. †No internal consistency Cronbach α calculated as is a knowledge test. ‡No internal consistency Cronbach α calculated as scale has only one item.

The final pre-intervention survey consisted of 43 items, including demographic information, assessing knowledge and attitudinal components of stigma. The final post-intervention survey consisted of 39 items assessing knowledge and attitudinal components of stigma without demographic information as summarized in Table 1.

Mental health knowledge

Mental health general knowledge

General knowledge of the prevalence of mental health concerns in Australia was measured by seven multiple-choice items based on information developed by headspace presented in the PowerPoint presentation. Answers were summed with a minimum score of 0 (low general knowledge) and a maximum score of 7 (high general knowledge). Example item includes ‘What percentage of Australians have a mental illness in any given year?’

Mental health signs and symptoms knowledge

Participants’ knowledge of some key signs of depression and risk of suicide was measured by asking participants to rate six statements (signs and symptoms of mental illness provided by headspace) as to how likely they thought the statement represented a sign or symptom of mental illness. These statements were taken verbatim from the PowerPoint slide contents delivered to participants in the intervention and designed by headspace. Example items include: ”To what extent do you think it is likely that the following are signs and symptoms of a mental ill health? Not enjoying, or not wanting to be involved in things that you would normally enjoy’. Answers were made on a 4-point Likert scale (from O’connor & Casey, 2015, used with permission) ranging from 1 (very unlikely) to 4 (very likely) and summed up with a minimum possible score of 6 (low knowledge) and a maximum score of 24 (high knowledge). Scoring was reversed based on the valence of the question. Using a 4-point Likert scale was preferred over a true/false or multiple-choice response scale to extract more detailed information.

MHFA for depression and substance use knowledge

This variable measured participants’ knowledge of appropriate mental health first aid for depression and substance abuse. Participants were invited to read a vignette used by N. J. Reavley and Jorm (2011) describing a person experiencing depression and substance use. Participants were asked to the extent to which they would enact actions in response to the vignette scenario described in five statements (developed by headspace as part of PowerPoint material) on a 5-point Likert scale from 1 (yes, I’d do this) to 5 (no, I wouldn’t do this). Example items include: ’Are you likely to suggest any of the following options: “Talk to him firmly about getting his act together”’. The response to the fifth statement was reverse scored and all scores summed, resulting in a minimum score of 5 (low MHFA knowledge) and a maximum of 25 (high MHFA knowledge).

MHFA for suicide knowledge

This variable measured participants’ knowledge of appropriate mental health first aid for suicide. Participants rated how likely they would be to enact six different MHFA steps for suicide (supplied by Lifeline) on a 5-point Likert scale from 1 (not likely at all) to 5 (very likely). Scores were summed, resulting in a possible minimum score of 6 (low MHFA to suicide) and a maximum score of 30 (high MHFA to suicide). Examples of MHFA steps include: ‘Ask them directly are you thinking about suicide?’ and ‘Stay with them or find someone else reliable to stay with them’.

Stigma attitudes

Self-efficacy to support

This variable measured participants’ evaluation of their confidence to provide appropriate MHFA support to a person experiencing depression and substance abuse. Participants rated a single item based on how confident they would be to provide support to someone like that described in the previous vignette on a 4-point Likert scale from 1 (very confident) to 4 (not confident at all). Responses were reverse scored so that high scores indicated high confidence and low scores indicated low confidence to provide support.

Public stigma prejudice

With reference to the same vignette, participants rated how strongly they agreed with five statements describing the attitudinal component of public stigma used by N. J. Reavley and Jorm (2011). Responses were made on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Scores were summed, resulting in a minimum score of 5 (low public stigma prejudice) to a maximum score of 25 (high public stigma prejudice).

Help-seeking propensity

Help-seeking propensity refers to the extent to which individuals believe they are willing and able to seek professional help (Mackenzie et al., 2004). Participants rated their level of agreement with five statements from the Help-Seeking Propensity subscale (Cronbach’s α = .76for full scale, Mackenzie et al., 2004). Scores were summed with a minimum score of 5 (having a low propensity to seek help for a mental health concern) and a maximum score of 25 (having a high propensity to seek help for a mental health concern). The Help-Seeking Propensity subscale is one of the three subscales from Mackenzie et al. (2004) Inventory of Attitudes Toward Seeking Mental Health Services (IASMHS).

Prejudice to help-seeking

Prejudice to help-seeking was measured by asking participants to indicate their agreement with three prejudicial statements towards help-seeking from the Psychological Openness subscale from the IASMHS (Cronbach’s α = .82, Mackenzie et al., 2004). Agreement was indicated using the same 5-point Likert scale described previously. All scores were summed, resulting in a minimum possible score of 3 (low prejudice to help-seeking) and a maximum possible score of 15 (high prejudice to help-seeking).

Help-seeking intentions

The help-seeking intentions variable was designed to capture participants’ professional help-seeking behavioural intentions. This question was based on a series of questions used by Mackenzie et al. (2004) in a community survey designed to cross-validate the IASMHS scale. Participants rated their likelihood to seek help on a 7-point Likert scale from 1 (very unlikely) to 7 (very likely).

Demographic information

The final four questions of the survey assessed participants’ gender, age, role at club and previous mental health training.

Data analysis

Surveys were delivered to the research team, and all data were entered in the Statistical Package for Social Scientists (SPSS, IBM Corp, 2021) for analysis. To ensure data were inputted correctly, cross-checks were performed by members of the research team. Cases with more than 15% missing data or more than 50% missing data on a given measure were excluded. For cases with less than 15% missing data on one measure, participants’ mean score on that measure was used as a substitute. Prior to statistical analysis, data were checked for normality and the relevant assumptions were met. Normality was assessed through visual inspection of histograms and normal P–P plots. To assess whether mental health knowledge and mental health stigma attitude significantly improved from pre-to-post intervention, two separate repeated measures one-way (trial: pre- and post-intervention) multiple analysis of variance (MANOVA) were conducted on the mental health knowledge-dependent variables and again for the stigma attitudes-dependent variables (see Table 1). Significant MANOVA trial effects were further integrated by univariate test of the same design. A paired sample t-test was used to assess whether help-seeking intentions significantly improved from pre- to post-participation. For all analyses, significance was set at the 95% level of confidence. Partial-eta squared was used to determine the effect sizes; of small (η_p² = .01) medium (η_p² = .06) or large (η_p² = .14) (Cohen, 1988).

Results

Participants

The pre-intervention sample consisted of 619 participants (70% male, M_age = 37 years, SD_age = 11.3 years, range 16–70 years). The post-intervention sample consisted of 164 participants (65% male, Mage = 40 years, SD_age = 10.2 years, range 16–68 years). The completion rate for both pre- and post-intervention surveys was 26.5%. Final participant numbers for each of the three analyses varied slightly dependent upon missing data: Health knowledge MANOVA n = 128; stigma attitudes MANOVA n = 134; and help-seeking intention paired sample t-test n = 151. With the exception of participants’ roles within the clubs, between groups analyses indicated that the pre- and post-intervention samples did not differ at p < .05 on any collected measures. With regard to role within club, there were notably fewer players in the post-intervention sample (5%) compared to the pre-intervention sample (18%). A potential explanation for this difference offered by the club presidents, was that players may have had less time or had more competing commitments than non-players. There were no significant differences in stigma attitudes-dependent variables at pre-intervention between those participants who completed and did not complete the post-intervention survey (p < .05).

Scale internal consistencies

Pre- and post-intervention Cronbach’s α analysis for each of the multi-item scales, with the exception of pre-intervention public stigma prejudice, demonstrated satisfactory (>.6) internal reliabilities (Nunnally & Bernstein, 1994; Pallant, 2001; see Table 1). With regard to the pre-intervention public stigma prejudice scale, based on the small number of items within the scale (five) which is acknowledged as contributing to conservative Cronbach’s α findings (Nunnally & Bernstein, 1994), the decision was made to accept the scale, but note the potential limited internal consistency and thus consider associated findings with caution.

Health knowledge

The MANOVA results for health knowledge (n = 128) demonstrated a significant difference between pre- and post-intervention scores with a large effect size, F (4, 124) = 73.48, p < .001, Wilks’ Lambda =.30, η_p² = .70. Subsequent univariate analysis revealed each of the four dependent variables (mental health general knowledge; mental health signs and symptoms knowledge; MHFA for depression and substance abuse knowledge; and MHFA for suicide knowledge) significantly improved from pre- to post-intervention with large effect size (Table 2).

Table 2. Descriptive and inferential univariate statistical results for mental health knowledge and stigma attitude-dependent variables.

Variable	Pre-M (SD)	Post-M (SD)	F	df	p	ƞp^2
Mental Health Knowledge† (n = 128)
General Knowledge	3.67 (1.34)	5.37 (1.09)	153.537*	1, 127	<.001	.547
Knowledge Signs Symptoms	20.66 (2.37)	22.03 (2.91)	24.731*	1, 127	<.001	.161
MHFA Depression	20.39 (4.12)	23.50 (2.42)	89.073*	1, 127	<.001	.412
MHFA Suicide	23.52 (3.35)	26.43 (2.68)	101.797*	1, 127	<.001	.445
Stigma Attitudes‡ (n = 134)
Self-Efficacy to Support	2.58 (0.84)	3.10 (0.70)	53.191*	1, 133	<.001	.286
Public Stigma Prejudice	20.03(2.82)	21.24 (2.60)	22.035*	1, 133	<.001	.142
Help-Seeking Propensity	18.12 (3.46)	20.51 (2.90)	62.819*	1, 133	<.001	.321
Prejudice to Help Seeking	12.45 (2.13)	13.29 (1.68)	21.825*	1, 133	<.001	.141

Note. †Significant MANOVA, F (4, 124) = 73.48, p < .001, Wilks’ Lambda =.30, η_p² = .70. ‡Significant MANOVA, F (4, 130) = 27.729, p < .001, Wilks’ Lambda =.540, η_p² = .46. *p < .001.

Stigma attitudes

The MANOVA results for stigma attitudes (n = 134) also demonstrated a significant difference between pre- and post-intervention scores with a large effect size, F (4, 130) = 27.729, p < .001, Wilks’ Lambda =.540, η_p² = .46. Subsequent univariate analysis revealed that each of the four dependent variables (self-efficacy to support; public stigma prejudice; help-seeking propensity; and prejudice to help-seeking) significantly improved from pre- to post-intervention with large effect size (Table 2).

Help-seeking intention

In terms of help-seeking intention, results from the paired sample t-test (n = 151) demonstrated a significant increase in scores from pre-intervention (M = 5.46, SD = 1.48) to post-intervention (M = 5.89, SD = 1.49) with medium effect size, t (150) = −3.06, p = .003, 95% CI [−.71, −.15], η_p² = .06.

Discussion

Our interpretation of the results from the MANOVAs and the paired sample t-test shows that a 1.5-hour targeted intervention can achieve significant and meaningful reductions in a range of variables designed to capture mental health knowledge, attitudinal components of mental illness stigma and help-seeking intention. However, it is important to note that these reductions were recorded on average 1.5 weeks after the intervention; further research would better inform the longevity of these changes. These observed reductions are in line with evaluations of a range of other stigma reduction and help-seeking interventions (Breslin et al., 2017; Bu et al., 2020; Confectioner et al., 2021). For example, a review of 13 randomized controlled trials of stigma reduction interventions found an educational approach had a small effect on changing attitudes and behavioural intentions whilst a contact approach had a small effect on behavioural intentions but a medium-sized effect on attitudes (P. W. Corrigan et al., 2012). Whereas a systematic review of five studies (1,239 athletes) examining mental health literacy interventions’ effect on athletes found that mental health knowledge and help-seeking attitudes increased, and stigma reduced (Bu et al., 2020). Likewise, a systematic review of 10 studies focusing on mental health awareness interventions in athletes and coaches established that stigma was reduced and mental health knowledge generally increased (Breslin et al., 2017). The MHFA programme has been found to generate up to medium-sized effects on changes in knowledge and small effects on changes in attitudes (Hadlaczky et al., 2014). In the present intervention, effect sizes were all moderate to high.

Significant changes observed in both mental health knowledge and attitude measures of stigma would be expected when using an intervention incorporating both an education and contact approach (Gould et al., 2007). Whilst the use of bespoke items in the present study allows a high number of variables of interest to be captured from a relatively brief survey, direct in-depth comparison to similar studies is difficult. A review of mental illness stigma literature from 2004 to 2010 found that over 400 measures of mental illness stigma had been used; two-thirds of which had not undergone any systematic psychometric evaluation (Fox et al., 2017). The present study offers some insight into areas where future research could focus, especially given the value of a brief intervention. It lends further support to the findings of the educational workshop component of the Ahead of the Game (Hurley et al., 2018, 2021; Vella et al., 2021). However, the use of psychometrically validated measures and ideally a control sample is recommended.

The current study did not examine gender differences within the sample. Addis and Mahalik (2003) believe that a sex-differences approach is limited because it does not consider the inter- and intra-individuality of men and women. A gender differences explanation considers male help-seeking as a product of masculine-gender role socialization whereby males learn gendered roles from a young age about what it is to be male. In this way, it is the adherence to traditional masculine norms of stoicism and self-reliance rather than being male in and of itself that underlies poorer rates of help-seeking (Addis & Mahalik, 2003). Measures such as the Male Role Norms Inventory – Revised (MRNI-R) and the Conformity to Masculine Norms Inventory (CMNI) have been used in the previous research to find that endorsements of aspects of traditional masculine norms were associated with the avoidance of help-seeking (Wimer & Levant, 2011). Nevertheless, men’s conformity to traditional masculine norms is variable, and men are not a homogenous group, as well as, other factors can impact help-seeking behaviours, such as the severity of mental health symptoms, cultural background, and sexual orientation (Sagar-Ouriaghli et al., 2019). Moreover, it may be more apt to consider sports culture as needing consideration wherein help-seeking and mental illness are viewed negatively and full schedules are not conducive to help-seeking behaviour (Jones et al., 2022). It has also been suggested that being too focused on problems associated with masculinity can overshadow positive or adaptive masculinity traits (Sagar-Ouriaghli et al., 2019). Indeed, Sagar-Ouriaghli et al. (2019) systematic review of men and help-seeking recommends that interventions on help-seeking reframe such behaviour with male values and as a sign of strength.

The present intervention made use of current and former NRL players to deliver positive messages of recovery from mental illness through seeking help. It was hoped that using male role models, who are generally perceived to be ‘tough’ and resilient elite athletes, to deliver these messages could help challenge some pervasive traditional masculine ideals. Qualitative feedback from an online campaign aimed at AFL players and construction workers reported that video stories delivered by AFL players were the most engaging aspect of the campaign (Australian Football League Players’ Association, 2017). The inclusion of a measure to assess adherence to masculine norms is needed to assess such an effect in this study. Further, the inclusion of qualitative feedback would help to triangulate quantitative data and highlight which aspects of the intervention were considered most useful and engaging by participants. Research that investigates the impact of the elite athletes with lived experience components of the intervention may also be warranted. Interestingly, a review of Australian initiatives (N = 61) to reduce stigma around mental illness found the involvement of people with lived experience, delivering content, occurred frequently and was a strength of Australian approaches (Morgan et al., 2021). Notably, a large review of anti-stigma programmes found they were most effective when they involved people with lived experience (Thornicroft et al., 2022). Further, it has been proposed that by improving the understanding and treatment of elite athletes’ mental health, help-seeking within the general community may also increase as a follow-on effect (Walton et al., 2019).

In addition to those discussed, other limitations exist in the present study. Missing data at Time 1 due to incomplete pre-intervention surveys was a problem in this study and as such the results should be interpreted with caution: For instance, the large drop-out rate from n = 619 at pre-intervention to n = 164 at post-intervention presents problems with potential bias in the reported results. It is possible that those who completed the survey at post-intervention were motivated to do so because they found it useful.

The potential for demand characteristics to have exerted influence on results also needs to be considered. As intervention aims were made explicitly clear to participants, it is possible participants answered surveys in a manner they believed was in keeping with the intervention aims. Future studies should include a group-delay control to explore whether demand characteristics had an impact. Of consideration is also the delayed data collection of approximately 1.5 weeks after the intervention, which may have caused emergencies and random errors but was required from a practical standpoint and requested by the NRL, so as not to delay club members from continuing on with their business for the evening. Finally, in line with suggestions from similar studies, including additional follow-up surveys would provide further insight into whether the results obtained in this study can be maintained several months later (Gronholm et al., 2017; Hanisch et al., 2016; Knifton et al., 2009).

Conclusion

The present study provides partial support that in the short term, at least a 1.5-hour intervention achieves changes in measures of mental health knowledge, mental illness stigma, and help-seeking intentions in a masculine environment. Whilst experimental research is needed in this area to more accurately determine the efficacy of the State of Mind programme (i.e. comparison to a control group, standardized measures, immediate post-survey measurement and longer term follow-up measurements), the brief intervention demonstrated improvement in knowledge measures, attitude measures of stigma, and help-seeking intentions is encouraging. Results from this study suggest brief interventions delivered through community sports groups have great potential to meaningfully promote positive attitudes, improve knowledge towards mental illness, and help-seeking intentions. Notably, a comprehensive, expert Lancet Commission report in 2022 on ending stigma and discrimination in mental health stated recommendations for anti-stigma initiatives/programmes, including several elements that were core aspects of the NRL’s State of Mind approach (Thornicroft et al., 2022). These include initiatives being adapted to the local context and involving those with lived experience, the use of aspirational figures (in this case rugby league players), having a reproducible approach, drawing on evidence-based interventions, using appropriate language and focusing on professional, cultural, and social values pertinent to the setting that underlies the potential stigma. Given that one of the ultimate consequences of mental illness stigma can be suicide, any improvement in this area is of great significance and impact.

Disclosure statement

The authors report that there are no known competing interests to declare.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Notes on contributors

Claire Margaret Ryan

Claire Margaret Ryan holds a Doctor of Philosophy in organisational psychology and is a registered psychologist. Claire conducts applied research and delivers interventions to change organisational culture and systems to make workplaces safer and more effective. She recently worked with an Australian charity to deliver program strategy, organisational development, evaluations, research, and advocacy. She leads the business direction of Authentic Lives Psychology and has a keen interest in helping people reach thier potential.

Campbell S Innes

Campbell S Innes is a registered psychologist in Queensland Australia having obtained undergraduate qualifications from the University of the Sunshine Coast and post graduate qualifications from the University of Queensland. His interests are trauma, depression and anxiety.

Lee Kannis-Dymand

Lee Kannis-Dymand is a researcher and Senior Lecturer in Clinical Psychology at University of the Sunshine Coast, Australia. His research is in awe, visitor and user experience (human nature/wildlife interactions; gaming), environmental psychology, and performance psychology. He co-authored the self-help book How to Deal with Anxiety.

Jonathan Mason

Jonathan Mason is Professor of Clinical Psychology at the Cairnmillar Institute in Melbourne, Australia. He is co-director of the Centre for Health, Wellbeing & Disability. His research, clinical and teaching interests include topics relating to disability, evidenced-based practice in clinical psychology, sexual health, and burnout.

Geoff P Lovell

Geoff P Lovell is an adjunct Associate Professor from the University of the Sunshine Coast, Australia. He is currently Programme Director at Kaplan Open learning for Online MSc Psychology programmes delivered in partnership with the University of Liverpool, UK. His main research interest is the psychology of sport, physical activity, and health.