What Actually Works in Workplace Health? A UK Evidence Guide for Employers

hercuwise

9 hours ago

9 min read

By Dr Suzanne Barr, MEd RD PhD, Director of HercuWise and Visiting Senior Research Fellow, King's College London. Last updated: September 2026.

TLDR / Summary: no single wellbeing product fixes workplace health. The strongest UK evidence supports a layered approach. Fix the working conditions that cause harm first (psychosocial and behavioural risk management), then equip line managers to act, then give people real control over how they work, backed by accessible support such as occupational health.

Search for "the most effective workplace health interventions" and you will find plenty of confident lists. Very few tell you how strong the evidence behind each item actually is.

Why does this matter right now?

Work-related stress, depression or anxiety affected an estimated 964,000 workers in Great Britain in 2024/25, according to the Health and Safety Executive. That is more than half of all work-related ill health. It also cost 22.1 million working days, an average of roughly 23 days per case (HSE figures, via Personnel Today).

The cost side is just as stark. Deloitte puts the cost of poor mental health to UK employers at £51 billion a year, with presenteeism the largest single part (Deloitte, 2024).

So spending is not the problem - spending on the wrong things is.

The evidence at a glance

Approach	What it reliably changes	Strength of evidence	The honest caveat
Psychosocial risk management (HSE Management Standards)	Identifies and reduces the causes of work-related stress	Recommended by HSE and NICE as the foundation	Only works if the survey leads to action
Line manager mental health training	Managers' knowledge, attitudes and supportive behaviour	Good for manager outcomes; limited for employee health	Managers need time and authority to act
Employee-controlled flexible working	Some health outcomes, including sleep and mental health	Tentative (small, older studies)	Employer-driven "flexibility" can do harm
Physical activity and sit-stand desks	Time spent sitting	Low quality	Health and productivity gains are unproven
Occupational health and EAPs	Access to counselling, adjustments, return to work	Mixed; much of it industry-reported	Awareness is low; cannot fix organisational causes
Individual mindfulness, resilience sessions	Unclear	A large UK study found no benefit	That study was cross-sectional

1. Start with the work itself: psychosocial risk management

The HSE Management Standards are the UK's framework for managing work-related stress. They cover six areas:

Demands: workload, work patterns, the working environment
Control: how much say people have over their work
Support: resources and help from managers and colleagues
Relationships: preventing conflict and dealing with unacceptable behaviour
Role: clear responsibilities, no conflicting roles
Change: how organisational change is managed and communicated

NICE's guideline on mental wellbeing at work (NG212) reaches the same conclusion from a different direction. It recommends a tiered approach, with organisation-level action as the foundation, individual approaches on top, and targeted support for those who need it. It applies to micro and small employers as well as large ones.

Why start here? Because burnout, persistent absence and turnover are often symptoms of how work is designed, not personal failings. A stress survey that gets filed away does nothing. The principle is simple: measure, act, review, and tell staff what changed.

2. Train line managers, but be clear about what that buys you

Managers set workloads, handle change and are usually the first to notice someone struggling. Training them is one of the most practical levers a smaller organisation has. The evidence needs reading carefully though. A 2018 systematic review and meta-analysis found manager training improved managers' knowledge, attitudes and self-reported supportive behaviour. Evidence on whether it improves employees' mental health was still preliminary.

There is one encouraging trial - in a cluster randomised controlled trial in an Australian fire and rescue service, four hours of manager training was linked to a significant drop in work-related sickness absence. The researchers estimated a return of nearly $10 for every $1 spent. That is one organisation, in one sector, so treat it as promising rather than proven. Good manager training helps people to:

Notice changes in behaviour, attendance or performance, without trying to diagnose
Start a supportive, non-judgemental conversation
Talk about workload and priorities
Understand reasonable adjustments
Know when to signpost to HR, occupational health or other support
Recognise their limits. Managers are not therapists.

And it needs organisational backing. Training a manager who has no time or authority to change anything is a waste of both.

3. Flexible working helps when employees hold the control

Flexible working is often described as "good for wellbeing" in general terms. The research is more specific than that. A Cochrane review found that arrangements increasing workers' own control, such as self-scheduling of shifts, tended to improve health outcomes. Arrangements driven by the organisation's interests, such as involuntary part-time work or fixed-term contracts, showed neutral or negative effects. The authors called the evidence tentative, as the studies were limited.

So the better question is not "do we offer flexible working?" It is "do people have meaningful, fair control over when and how they work?"

The law is moving in the same direction. Employees have had a day-one right to request flexible working since April 2024. Under the Employment Rights Act 2025, employers will only be able to refuse where it is reasonable to do so, with changes expected from autumn 2027 (DLA Piper summary). Managers handling requests will need to explain their reasoning properly. Another reason to train them, but one caution - Flexibility cannot fix chronic understaffing or unrealistic workloads. It sits inside good job design, not in place of it.

4. Physical activity: useful, but keep expectations modest

Active breaks, walking meetings and sit-stand desks can help, especially in sedentary roles. A Cochrane review found sit-stand desks reduced sitting at work by about 84 to 116 minutes a day in the short term. But the evidence was rated low quality, and there is no evidence yet on long-term effects. If a supplier promises productivity gains from standing desks, ask to see the data. The research does not support that claim as settled.

For manual, shift-based, driving or customer-facing roles, the priorities are often different: sensible scheduling, recovery time, safe task design and decent rest facilities. The aim is to make healthy behaviour easier, not to add another expectation. The UK Chief Medical Officers' physical activity guidelines are a useful reference for what "enough" movement looks like.

5. Occupational health and EAPs: valuable, but only if people use them

Occupational health advises on adjustments, rehabilitation and return to work. Employee Assistance Programmes (EAPs) offer confidential counselling and referral. Both have a real place but the weak spot is uptake. One industry survey found that while 79% of employers provide an EAP, only 27% of employees knew one existed (Employee Benefits, 2024). A service nobody knows about delivers close to nothing.

Usage data also needs care. Low use may mean poor awareness or low trust. High use may mean people trust it, or it may signal significant unmet need. Your own data tells you more than any average.

Most importantly: counselling cannot fix an unsustainable workload. Referring someone to an EAP when the cause is organisational simply moves the problem.

What about wellbeing apps, mindfulness and resilience training?

This is where the evidence is weakest. A 2024 University of Oxford study of 46,336 workers across 233 UK organisations found that people taking part in individual-level interventions (resilience training, mindfulness, wellbeing apps and similar) were no better off on wellbeing measures than those who did not. The author suggested organisational changes, such as scheduling, management practices and job design, were more promising. The study was cross-sectional, so it cannot prove these tools never work. But it is a strong signal not to lead with them.

Where does accessible digital learning fit?

Digital learning is useful for giving consistent, flexible access to health education, particularly for SMEs and for 'hard-to-reach' dispersed, hybrid or shift-based teams. It works best when it is:

Evidence-informed and honest about limitations
Practical, focused on what people can actually apply
Accessible on mobile and across working patterns
Embedded, with managers and organisational action behind it
Evaluated for learning, confidence and behaviour change

An online stress module can help someone recognise overload and raise it with their manager. If the root cause is excessive workload, the organisation still has to fix the workload. That is the difference between health education and prevention.

What should businesses, including SMEs do first?

Much of the research comes from large organisations with HR teams and occupational health contracts. SMEs usually have less time, smaller budgets and no dedicated wellbeing lead. That does not mean doing nothing. It means doing a few things well.

A proportionate starting point:

Run a confidential staff survey or facilitated discussion on the six Management Standards areas
Pick two or three priorities and name who owns each action
Give managers short, practical guidance on supportive conversations and workload
Make sure everyone knows what support exists and how to reach it
Add targeted digital learning where it fills a gap
Review in six months and tell staff what changed

Policy is heading this way too. The government's Keep Britain Working review (November 2025) called for employers to take a bigger role in prevention and early intervention, and launched a three-year vanguard phase to test what works across organisations of different sizes.

How should employers measure whether it's working?

Overclaiming is common. Attendance at a webinar doesn't prove it improved productivity and liking a course doesn't prove it reduced absence.

This matters for return on investment. Deloitte's 2024 review of 26 studies estimated an average return of £4.70 for every £1 invested in workplace mental health (Deloitte). It's an average across varied studies, not a guarantee for your organisation. Deloitte also found early, organisation-wide approaches tended to return more than later, reactive support. Set a baseline, pick the outcomes that matter, and track them over time.

So, what actually works?

The evidence points to a prevention-first, whole-organisation approach:

Understand the risks. Start with psychosocial risk and working conditions.
Listen, then act. Consultation only counts if something visibly changes.
Equip managers. Skills plus the time and authority to use them.
Design work well. Workload, control, role clarity, flexibility, communication.
Make support accessible. Occupational health, EAPs and adjustments people know about and trust.
Use digital learning purposefully. To extend access, not to replace organisational change.
Design for everyone. Shift workers, frontline staff, remote workers, disabled employees.
Evaluate honestly. Be clear what you measured and what it shows.

Workplace wellbeing is not about finding the next product that promises healthier, happier, more productive staff. It is about building work where people have a fair chance of staying well, spotting problems early, and getting the right support when they need it.

Frequently asked questions

What is the most effective workplace health intervention? There isn't a single one. UK guidance from NICE and HSE supports organisation-level action on working conditions as the foundation, with manager training, flexible working and accessible support layered on top.

Do workplace wellbeing programmes improve productivity? Sometimes, but the evidence is mixed. Deloitte's review found an average return of £4.70 per £1 on workplace mental health, but results vary widely by intervention. Productivity is less consistently measured than absence or wellbeing.

Is manager mental health training worth it? For improving how managers respond, yes, the evidence is good. For improving employee health directly, the evidence is still limited, though one trial found reduced work-related sickness absence.

Do wellbeing apps work? Independent evidence is thin. A 2024 Oxford study of over 46,000 UK workers found no wellbeing benefit from individual-level interventions such as apps and resilience training.

Are employers legally required to manage work-related stress? Employers have a legal duty to assess and manage health risks at work, including stress. HSE's Management Standards are the recommended way to do this.

Where should a small business start? With an honest look at what is driving absence and stress in your workplace, usually through a short staff survey against the six Management Standards areas, then two or three owned actions.

Want an honest view of your own workplace?

Most organisations already have some of the pieces above. What is usually missing is a clear, evidence-based picture of what is working, what isn't, and what to fix first. That is what a HercuWise workplace health audit gives you, followed by practical advisory support and, where it fits, evidence-based digital learning for your teams.

Book a workplace health audit conversation →

About the author: Dr Suzanne Barr is the Director of HercuWise, and a Visiting Senior Research Fellow at King's College London.