Dissertation – Mental Health First Aid Training in the Workplace

Abstract

Mental ill-health is a leading contributor to Australia’s burden of disease and an increasingly recognised workplace risk. Mental Health First Aid (MHFA) training, developed in Australia, teaches employees to recognise and respond to colleagues who may be experiencing mental health problems, yet questions remain about whether improved knowledge translates into changed attitudes and helping behaviour. This dissertation evaluates a workplace MHFA program using an explanatory sequential mixed methods design that combined a pre-post survey of 196 trained employees with 14 follow-up interviews. Quantitative results showed large gains in knowledge and helping confidence, a moderate increase in help-offering intentions, and smaller reductions in personal stigma and desired social distance. Interviews revealed greater confidence to begin supportive conversations, a reframing of distress as a health issue, uncertainty about the boundaries of the first-aider role, and heavy reliance on organisational follow-through. The study concludes that training is necessary but not sufficient, and that it must be embedded within workplace psychosocial risk management.

Introduction

Mental health has moved to the centre of the Australian workplace agenda. The Australian Bureau of Statistics estimates that more than one in five adults experiences a mental health condition in any given year, with anxiety and affective disorders the most common (Australian Bureau of Statistics [ABS], 2023), while the Australian Institute of Health and Welfare identifies mental ill-health as one of the largest sources of non-fatal disease burden nationally (Australian Institute of Health and Welfare [AIHW], 2023). Because most adults spend much of their waking lives at work, the workplace is both where distress becomes visible and where early, informal support can be offered.

The regulatory context has shifted decisively. The Productivity Commission (2020) estimated the cost of mental ill-health and suicide to the Australian economy at up to $70 billion a year, and identified workplaces as a priority setting for prevention and early intervention. Employer obligations have been sharpened accordingly: the primary duty of care under work health and safety law is now given practical content by a model code requiring organisations to identify and control psychosocial hazards such as high job demands and low support (Safe Work Australia, 2022), reinforced by the Fair Work Commission’s jurisdiction over workplace bullying (Fair Work Act 2009 (Cth)). National initiatives have encouraged employers to treat psychological health as a systems responsibility rather than an individual one (Beyond Blue, 2020).

Mental Health First Aid training is one of the most widely adopted workplace responses. Developed in Australia and now delivered internationally, the standard course teaches participants to use a structured action plan to support a person developing a mental health problem or experiencing a crisis, and to encourage appropriate professional and other help. Evaluations consistently report gains in knowledge and confidence, but effects on stigma and actual helping behaviour are more modest, and little Australian evidence integrates measured outcomes with trainees’ accounts of the role. This dissertation addresses that gap by evaluating a workplace MHFA program in the Australian setting. Three research questions guided the study:

1. To what extent does workplace MHFA training change employees’ mental health knowledge, helping confidence, stigmatising attitudes and help-offering intentions from before to after training?
2. How do trained employees experience and apply the mental health first-aider role in the workplace?
3. Which organisational conditions support or constrain the translation of training into help-offering behaviour?

Literature Review

Mental health literacy and the MHFA model

The conceptual foundation for MHFA is mental health literacy, defined as the knowledge and beliefs that help people recognise, manage and respond to mental health problems (Jorm, 2012). The premise is that community members, including co-workers, are often the first to notice that someone is struggling, and that equipping them to respond can reduce delays to care. The training operationalises this through a structured action plan that guides the helper to assess the situation and any risk of harm, listen without judgement, give reassurance and information, encourage appropriate professional help, and encourage other supports. The original public evaluation demonstrated that the course improved recognition of disorders, changed beliefs about treatment toward those of professionals, and increased confidence in providing help (Kitchener & Jorm, 2002).

Evidence on knowledge, attitudes and behaviour

The accumulated evidence is broadly positive but uneven across outcomes. A randomised controlled trial conducted in a workplace found that trained employees showed greater knowledge, more supportive attitudes and greater confidence than waitlisted colleagues, with some benefits sustained at follow-up (Kitchener & Jorm, 2004). Meta-analytic syntheses confirm the pattern at scale: training produces reliable improvements in knowledge and confidence, but effects on stigmatising attitudes are smaller and effects on helping behaviour smaller still, in part because behaviour is difficult to measure and opportunities to help are intermittent (Hadlaczky et al., 2014; Morgan et al., 2018). This gradient, in which cognitive outcomes shift more readily than attitudes and behaviour, is the central interpretive puzzle for any evaluation and frames the present study’s expectations.

The workplace as a setting and the evaluation gap

The workplace introduces conditions that generic evaluations rarely capture. Guidelines developed through Australian expert consensus emphasise that a workplace first aider operates within an organisational system of policies, managers and referral pathways, and that role clarity, confidentiality and support for the helper are essential to safe practice (Bovopoulos et al., 2016). This situates training as one control within a broader psychosocial risk-management framework rather than a standalone remedy, consistent with the model code’s expectation that hazards be managed systemically (Safe Work Australia, 2022). Two gaps follow. First, few Australian studies combine measured pre-post change with qualitative accounts of how the role is enacted. Second, the conditions that convert trained capability into offered help are under-examined. This dissertation addresses both.

Methodology

Research design

An explanatory sequential mixed methods design was adopted, in which a quantitative phase is followed by a qualitative phase that helps to explain the statistical results (Creswell & Plano Clark, 2018). The evaluation was structured around a program logic model, shown in Figure 1, which links training inputs and activities to short-term learning outcomes and to downstream behaviour in the workplace. The survey measured the learning outcomes (knowledge, confidence and stigma), while the interviews explored the behaviour level and the organisational conditions that the logic model treats as moderators. Integration occurred through purposive selection of interviewees from survey respondents and through a joint interpretation of statistical and thematic findings.

Participants and procedure

Approximately 210 employees across three Australian organisations, spanning professional services, local government and higher education, completed the accredited 12-hour course during the study period. Matched pre-training and post-training surveys were obtained from 196 participants, who form the analytic sample; the pre survey was administered immediately before the first session and the post survey within two weeks of completion. Of these participants, 61% identified as women, the mean age was 41.3 years (SD = 10.4), and 44% held a supervisory or managerial role. For the qualitative phase, a maximum-variation subsample of 14 participants was selected to span organisation, gender, seniority and whether they had yet used the training, and each took part in a semi-structured interview of 40 to 55 minutes.

Measures and analysis

Mental health knowledge was assessed with a 25-item quiz covering recognition, evidence-based support and referral. Helping confidence used a composite rated from 1 to 10, benchmarked against the participant’s sense of readiness to assist a colleague. Personal stigma was measured with the personal subscale of the Depression Stigma Scale, scored 0 to 36 so that lower scores indicate less stigma, and desired social distance with a five-item scale scored 0 to 15 in the same direction. Help-offering intention was rated from 0 to 10. Multi-item scales returned acceptable internal consistency, with Cronbach’s alpha between .78 and .90. Analysis used paired-samples t-tests and standardised mean change (Cohen’s d) computed with the pooled standard deviation. Interviews were audio-recorded, transcribed and analysed using reflexive thematic analysis (Braun & Clarke, 2021), moving from coding to the construction of themes across the dataset.

Ethical considerations

The study was approved by the administering university’s Human Research Ethics Committee, and procedures complied with the National Statement on Ethical Conduct in Human Research (National Health and Medical Research Council [NHMRC], 2023). Because the subject matter is sensitive, particular care was taken to protect participants: involvement was voluntary with written informed consent, responses were de-identified, interview participants received pseudonyms, and a distress protocol together with information about professional support pathways, including workplace employee assistance and general practitioners, was provided to everyone approached. The evaluation concerns the provision of initial support and encouragement to seek help; it does not position first aiders as clinicians, and no participant was asked to disclose a personal diagnosis.

Findings

Phase 1: Pre-post survey results

Every measure moved in the expected direction from before to after training, but the magnitude of change differed markedly, as shown in Table 1. Confidence to assist a colleague showed the largest shift, from a pre-training mean of 5.3 to 7.4 on the 10-point scale. Taking the pooled standard deviation as the square root of the averaged variances gives

SD_pooled = √[(1.8² + 1.5²) / 2] = √2.745 = 1.66, and

d = (7.4 – 5.3) / 1.66 = 2.1 / 1.66 = 1.27,

a large standardised change by conventional benchmarks. Knowledge showed a similarly clear gain (d = 0.88) and help-offering intention a moderate one (d = 0.75). By contrast, personal stigma and desired social distance shifted only modestly (d = 0.43 and 0.35 respectively). The pattern is instructive: the cognitive outcomes that training targets most directly moved furthest, whereas the attitudinal outcomes, which are more deeply held, proved more resistant to a single course.

Table 1: Pre-post change on knowledge, confidence, stigma, social distance and help-offering intention (N = 196)

Note. Higher scores indicate more knowledge, confidence and help-offering intention; for personal stigma and desired social distance, lower scores are favourable, so the negative mean change represents improvement. Cohen’s d is reported as a magnitude and computed with the pooled standard deviation.

Phase 2: Interview themes

Reflexive thematic analysis produced five themes, summarised in Table 2. The most consistent account was of new confidence to begin a conversation: participants described no longer avoiding a colleague who seemed to be struggling, but instead feeling able to ask a direct and caring question. Closely related was a reframing of distress as a health issue rather than a personal weakness. Where participants had used the training, they described drawing on the structured action plan to steady themselves, as one explained: “having the steps in my head stopped me freezing, I knew to listen first and then talk about getting proper help” (Participant 7). Two themes qualified this optimism. Many spoke of uncertainty about the limits of the role and of an emotional load in carrying it, worrying about “getting it wrong” or being drawn beyond their competence. Finally, participants emphasised that the value of training depended on organisational follow-through, including visible leadership, clear referral pathways and permission to step away from normal duties to help; without these, several felt the role risked becoming tokenistic.

Table 2: Themes from reflexive thematic analysis of interviews (n = 14)

Discussion

The survey answers the first research question with a clear gradient. Training produced a large increase in helping confidence and a substantial gain in knowledge, a moderate lift in help-offering intention, and only modest movement in personal stigma and social distance. This ordering aligns closely with the meta-analytic evidence that MHFA reliably improves knowledge and confidence while shifting stigma and behaviour more weakly (Hadlaczky et al., 2014; Morgan et al., 2018), and it is consistent with the workplace trial that first established these benefits (Kitchener & Jorm, 2004). The interpretation is not that the course fails on stigma, but that deeply held attitudes and enacted behaviour respond less to a single educational intervention than does declarative knowledge, precisely the pattern the mental health literacy framework would anticipate (Jorm, 2012).

The interviews address the second and third research questions and explain why measured intention may overstate real-world helping. Confidence and reframing were widely reported, echoing the survey, but two accounts complicate a simple success story. Uncertainty about role boundaries and the emotional load of supporting others suggest that willingness does not automatically become safe, sustained practice, and the expert workplace guidelines anticipate exactly this by insisting on role clarity, confidentiality and support for the helper (Bovopoulos et al., 2016). The strong emphasis on organisational follow-through reinforces the study’s central argument: help-offering is a property of the system as much as of the individual, enabled where leadership, referral pathways and permission to act were present and symbolic where they were absent.

These findings carry a regulatory as well as a practical reading. Under the model code, Australian employers must manage psychosocial hazards through a systematic process rather than a single initiative (Safe Work Australia, 2022), and the systems approach promoted nationally frames training as one element within a mentally healthy workplace, not a substitute for job design and support (Beyond Blue, 2020). A first-aid capability that is not embedded in that system is unlikely to realise its promise, and may even shift responsibility onto individual employees for a risk that is organisational in origin.

Four limitations qualify these conclusions. The quantitative phase used an uncontrolled pre-post design, so the gains, particularly in confidence, may be inflated by expectancy and the absence of a comparison group. All measures were self-reported, and help-offering was assessed as intention rather than observed behaviour, which the qualitative accounts suggest is an imperfect proxy. The post survey was administered soon after training, so the durability of effects and any decay over time are unknown. Finally, participants enrolled voluntarily and may hold more favourable attitudes than the wider workforce, limiting generalisation.

Implications and Conclusion

Three sets of implications follow:

• Management practice: organisations should define the first-aider role explicitly, provide ongoing supervision and debriefing to manage the emotional load, and ensure clear referral pathways such as employee assistance and general practice, so that willingness is matched by safe, supported practice.
• Organisational policy and regulation: MHFA should be positioned as one control within a psychosocial risk-management system, integrated with leadership commitment and job design in line with the model code (Safe Work Australia, 2022) and the national systems approach (Beyond Blue, 2020), rather than deployed as a standalone response.
• Research: controlled and longitudinal designs that follow trained employees over time and capture observed helping and organisational outcomes, drawing on Australian workforce data (ABS, 2023; AIHW, 2023), would test whether the gains reported here translate into a genuinely safer workplace.

In sum, workplace Mental Health First Aid training delivered its most reliable benefits where the literature predicts, in knowledge and confidence, with more modest movement in stigma and a moderate lift in the intention to help. The interviews showed that these capabilities become useful support only when the organisation clarifies the role, protects the helper and provides somewhere to refer people on. Training is therefore necessary but not sufficient. For Australian employers now accountable for foreseeable psychosocial risk, the task is to embed first-aid capability within a system of prevention and support, so that a trained colleague’s willingness to begin a difficult conversation is met by an organisation ready to see it through.

References

Australian Bureau of Statistics. (2023). National study of mental health and wellbeing, 2020-2022. ABS.

Australian Institute of Health and Welfare. (2023). Mental health: Prevalence and impact. AIHW.

Beyond Blue. (2020). Heads up: Guidance for a mentally healthy workplace. Beyond Blue.

Bovopoulos, N., Jorm, A. F., Bond, K. S., LaMontagne, A. D., Reavley, N. J., Kelly, C. M., Kitchener, B. A., & Martin, A. (2016). Providing mental health first aid in the workplace: A Delphi consensus study. BMC Psychology, 4, 41.

Braun, V., & Clarke, V. (2021). Thematic analysis: A practical guide. Sage.

Creswell, J. W., & Plano Clark, V. L. (2018). Designing and conducting mixed methods research (3rd ed.). Sage.

Fair Work Act 2009 (Cth).

Hadlaczky, G., Hokby, S., Mkrtchian, A., Carli, V., & Wasserman, D. (2014). Mental Health First Aid is an effective public health intervention for improving knowledge, attitudes, and behaviour: A meta-analysis. International Review of Psychiatry, 26(4), 467-475.

Jorm, A. F. (2012). Mental health literacy: Empowering the community to take action for better mental health. American Psychologist, 67(3), 231-243.

Kitchener, B. A., & Jorm, A. F. (2002). Mental health first aid training for the public: Evaluation of effects on knowledge, attitudes and helping behavior. BMC Psychiatry, 2, 10.

Kitchener, B. A., & Jorm, A. F. (2004). Mental health first aid training in a workplace setting: A randomized controlled trial. BMC Psychiatry, 4, 23.

Morgan, A. J., Ross, A., & Reavley, N. J. (2018). Systematic review and meta-analysis of Mental Health First Aid training: Effects on knowledge, stigma, and helping behaviour. PLOS ONE, 13(5).

National Health and Medical Research Council. (2023). National statement on ethical conduct in human research. NHMRC.

Productivity Commission. (2020). Mental health (Inquiry report No. 95). Australian Government.

Safe Work Australia. (2022). Model code of practice: Managing psychosocial hazards at work. Safe Work Australia.