Mental health first aid is sold to the workforce and taken by volunteers

I’ve always said mental health first aid wasn’t any good because it had no evidence. So I had it checked properly.

That was slightly wrong, and the accurate version is worse for it. It’s sold to employers as a way to improve the wellbeing of the workforce. It’s taken up by volunteers. And where anyone has measured the workforce around them, nothing changed.

What it’s sold as

The programme’s owner calls it “one of the most powerful tools an organisation can put in place benefiting people and business”.¹ Its campaign page wants it “mandated in every organisation, just like physical first aid”.²

Its workplace page puts a return of £4.70 for every £1 next to the course, credited to Deloitte.³ That’s Deloitte’s average across everything employers spend on mental health and wellbeing, and the report doesn’t mention mental health first aid once.⁴

And employers seem to buy it on those terms. In the one study that asked, the most common reason given, by about a third of respondents, was that their organisation had seen or expected mental health problems among staff. Those were people involved in the training saying what they thought their organisation wanted. The authors concluded that the drive underneath was a productive workforce.⁵

Who turns up

In most of the studies I found, it was voluntary, and it filled with people who wanted to go. In that same study, one interviewee said the people on the course “have generally volunteered to be on that course, rather than people that have been asked to go on that course, or told to go on that course”.⁵ Some organisations capped the places. In the first workplace trial, the most common reasons for going were being “just interested”, something at work, and duty as a citizen.⁶ In a French survey of civil servants, most gave work reasons.⁷ In a UK schools trial, 8.6 per cent of staff did the two-day course.⁸

So a small share of staff get trained, mostly because they want to. That’s who the course reaches.

What the course does for them

It works on them, as far as courses go. Across 18 trials and nearly 6,000 people, what trainees know about helping someone with a mental health problem went up straight after the course, and was still up at six months, by less.⁹ Grade that moderate.

The catch is that the trials randomised volunteers. The gains are real for people who chose to be there. They tell you nothing about the colleague who’d never have signed up, who is presumably the one you were worried about.

What happens to everyone else

This is what employers are paying for, and it’s where the evidence goes quiet or runs the wrong way.

Cochrane looked at the mental health of the people around trainees and found the course may make little or no difference, with very uncertain evidence.¹⁰ In the only UK workplace trial, across nine organisations, employees where the course was adopted were less likely to seek formal help and reported lower social wellbeing. Few said a first aider had supported them at all.¹¹ NICE reviewed the evidence for its guideline on mental wellbeing at work and found it didn’t show any effect on mental wellbeing.¹² In the schools trial, teachers’ wellbeing showed no difference.¹³

All of that is very low certainty. Nine organisations, short follow-ups, and I could only read the UK trial’s abstract. It hasn’t been shown to work for the workforce, and the few times anyone has looked, nothing changed. That’s as far as the evidence goes, and it’s far enough.

Where my bet went

My bet was that it mostly attracts people who already believe in mental health, so it adds very little. I had that checked too, and I lost about half of it.

Nobody has compared the people who sign up with the colleagues who don’t, so the bet is untested. What there is fits it only partly. Most say they go for work reasons or because they’re interested. In the first workplace trial, the volunteers’ own mental health was worse than the population average,⁶ so plenty were going through it themselves. That could cut either way.

The better explanation is how few people the course reaches. The UK schools trial didn’t rely on volunteers. Staff nominated the colleagues they thought would suit the role, and those were the people trained. A year later, 6.1 per cent of teachers had used the peer support service, and teachers’ wellbeing showed no difference.⁸¹³ In one organisation’s survey, six of 21 first aiders hadn’t had a single conversation in a year, and every one of them said most or all of their conversations were with people they already knew.¹⁴

In the one trial that trained the people colleagues picked, the workforce still didn’t move, because most of it never went to them. That’s my reading of indirect evidence, not a measured finding, but it fits better than my bet did.

What I’d do with this

I wouldn’t scrap it. It costs about £325 a head at the recommended price,¹⁵ and people who go on it know more afterwards.

I’d stop expecting a course for volunteers to do the job of an organisation looking after its staff. HSE describes doing a stress risk assessment and acting on it as a legal duty.¹⁶ Mental health trained first aiders are one option its first aid guidance says you might choose.¹⁷ If the training is the plan, there isn’t one.

If you’ve already got first aiders, find out whether anyone knows who they are, whether anyone has asked them for anything, and how they’re doing themselves.

What this can’t tell you

The searching and first grading were done by AI agents, and a random 30 per cent of the sources was re-checked by separate verifiers. I re-read every passage the evidence review quotes. Most trials tested the Australian course rather than the version sold here. I’m a benefits broker with an MSc, not a guideline panel.

What would change my mind

A proper UK trial across a few dozen organisations, measuring the wellbeing and help-seeking of the whole workforce. A study of who volunteers, against who doesn’t, would settle the other half of my bet. If either comes back the other way, this note gets a dated correction.

Until then, I’ll stop saying it has no evidence. It has evidence that volunteers learn something, and none yet that the workforce is better for it.

Footnotes

1. MHFA England. Workplace Mental Health First Aid Standards. mhfaengland.org. Read 18 October 2026. ↩

2. MHFA England. Set the standard. Mandate the change. mhfaengland.org. Read 18 October 2026. On 9 October 2026 its chief executive wrote that its position is to build the evidence before asking government to legislate: mhfaengland.org. ↩

3. MHFA England. Mental health courses for workplaces. mhfaengland.org. Read 18 October 2026. ↩

4. Deloitte (2024). Mental health and employers: the case for employers to invest in supporting working parents and a mentally healthy workplace. deloitte.com. ↩

5. Narayanasamy, M., Geraghty, J., Coole, C., Nouri, F., Thomson, L., Callaghan, P. and Drummond, A. (2018). MENtal health first aid in The wORkplace (MENTOR): a feasibility study. IOSH research report, University of Nottingham. iosh.com. ↩ ↩²

6. Kitchener, B.A. and Jorm, A.F. (2004). Mental health first aid training in a workplace setting: a randomized controlled trial. BMC Psychiatry, 4, 23. doi:10.1186/1471-244X-4-23. ↩ ↩²

7. Lebrat, M., Carrouel, F., du Sartz de Vigneulles, B., et al. (2026). Assessment of knowledge, attitudes, and behaviors toward others’ mental disorders to improve mental health first aid: cross-sectional study in a French workplace. JMIR Public Health and Surveillance, 12, e93207. doi:10.2196/93207. ↩

8. Fisher, H., Harding, S., Bell, S., et al. (2020). Delivery of a Mental Health First Aid training package and staff peer support service in secondary schools: a process evaluation of uptake and fidelity of the WISE intervention. Trials, 21, 745. doi:10.1186/s13063-020-04682-8. ↩ ↩²

9. Morgan, A.J., Ross, A. and 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), e0197102. doi:10.1371/journal.pone.0197102. ↩

10. Richardson, R., Dale, H.E., Robertson, L., Meader, N., Wellby, G., McMillan, D. and Churchill, R. (2023). Mental Health First Aid as a tool for improving mental health and well-being. Cochrane Database of Systematic Reviews, CD013127. doi:10.1002/14651858.CD013127.pub2. ↩

11. Atanda, O., Reavey, P., Durcan, G., Carter, T., O’Shea, N., Brown, S., Wood, K., Vangeli, E., White, S. and Callaghan, P. (2026). A cluster randomised controlled trial of mental health first aid (MHFA) in UK workplaces. Journal of Public Mental Health. doi:10.1108/JPMH-03-2026-0051. Abstract only. ↩

12. National Institute for Health and Care Excellence (2022). Mental wellbeing at work (NG212). Evidence review A: organisational universal-level approaches. nice.org.uk. ↩

13. Kidger, J., Turner, N., Hollingworth, W., et al. (2021). An intervention to improve teacher well-being support and training to support students in UK high schools (the WISE study): a cluster randomised controlled trial. PLoS Medicine, 18(11), e1003847. doi:10.1371/journal.pmed.1003847. ↩ ↩²

14. Summerfield et al. (2025). An evaluation of Mental Health First Aid Officer utilisation across an international, interdisciplinary oncology group. Journal of Medical Radiation Sciences. doi:10.1002/jmrs.70021. ↩

15. MHFA England. Mental Health First Aid course page, recommended price. mhfaengland.org. Read 18 October 2026. ↩

16. Health and Safety Executive. Work-related stress: stress risk assessment. hse.gov.uk. ↩

17. Health and Safety Executive. First aid at work: what employers need to do. hse.gov.uk. ↩