Why mental health 'awareness' isn't enough anymore

Executive Summary

• Awareness campaigns have reduced stigma and made mental health a normal subject in UK workplaces, and that progress is important.
• Those gains have not lowered incidence. HSE data for 2024/25 records 964,000 workers with work-related stress, depression, or anxiety, 52% of all work-related ill health and 22.1 million lost working days.
• The driver is how work is designed, not how well people cope. The CIPD Good Work Index finds 1 in 4 UK workers say their job harms their mental health.
• Shifting the numbers requires building mental health into workload, manager behaviour, and job design, not bolting on awareness weeks and apps.

The numbers that should end the debate

The HSE figures for 2024/25 should settle any debate about whether the current approach works. 964,000 UK workers experienced work-related stress, depression, or anxiety. That accounts for 52% of all work-related ill health, more than half of every illness people attribute to their jobs.

The cost lands in lost time. Work-related stress, depression, and anxiety drove 22.1 million lost working days across the year. For most employers, absence at that scale shows up directly in productivity, in cover costs, and in the pressure it puts on the people who remain.

The trend matters more than any single number. Cases rose by 180,000 in a single year, a record high. Organisations have spent more on awareness, more on apps, and more on wellbeing initiatives over that same period, and the incidence climbed anyway.

A model that absorbs growing investment while the outcomes worsen is not under-resourced. It is pointed at the wrong target. The next sections show where that target should sit, and why awareness was never able to reach it.

What awareness has genuinely achieved, and where it runs out

The awareness era earned its place. A decade ago, most employees would not have told a manager they were struggling, and most managers would not have known how to respond. Stigma kept mental health off the agenda entirely. Awareness days, training sessions, and senior leaders speaking openly changed that. People talk about mental health at work now in a way they simply did not before, and that progress is real.

The trouble is that talking about a problem is not the same as reducing it. Stigma reduction is a precondition for action, not a substitute for it. You can measure how comfortable people feel raising mental health, and that number can climb every year while the incidence of work-related stress, depression, and anxiety climbs alongside it. Comfort and outcomes are not the same variable.

The CIPD's Good Work Index makes the gap impossible to ignore. One in four UK workers say their job actively harms their mental health. That figure points at the work itself, not at how openly people discuss it. No amount of conversation fixes a job that is poorly designed, overloaded, or stripped of control. Awareness opened the door to that conversation. It was never built to redesign the work waiting on the other side.

Why add-on support cannot carry the load

Employee assistance programmes, meditation apps, and wellbeing tools do real work. They give people in distress a place to turn, and for someone in crisis at 2am, an EAP helpline matters. The problem is not that they fail at what they do. Employers ask them to do something they were never built for.

These tools operate downstream of the cause. An EAP catches the worker who is already struggling. A meditation app helps the person who already feels overwhelmed. Neither touches the workload, the unclear role, or the manager whose behaviour produced the overwhelm in the first place. You can fund all three and still see your stress figures climb, because the conditions generating the harm sit upstream and untouched.

Deployed without a map of where risk sits, these tools reach everyone in theory and no one in particular in practice. A meditation app is offered to the whole workforce, then used mostly by the people who would have managed anyway. The team carrying the real risk, buried in an understaffed function with no autonomy and a manager under pressure, rarely shows up in the app usage data at all.

Most employers never ask what is driving the risk, and for which people. Until they answer that, add-on support stays a treatment for symptoms whose cause goes unexamined.

Reactive by default

Most employers manage mental health by watching absence data, which means they only see a problem once someone has already stopped working. Sickness absence is a lagging indicator. By the time a case registers in your figures, the person has been struggling for weeks or months, the work that triggered it has continued unchanged, and the cost has already landed. You are reading a receipt, not a warning.

That model makes prevention structurally impossible. When your trigger for action is the absence itself, you can only ever respond after the damage is done. You manage the symptom, the time off, the cover, the return-to-work meeting, while the condition that produced it stays exactly where it was. The next person on that team meets the same workload, the same manager, the same lack of control, and follows the same path.

Reacting to absence also frames mental health as an individual event rather than a pattern. One person off looks like one person's problem. Six people off the same team over a year is a design fault. Absence data rarely surfaces that pattern early enough to act on it, so organisations keep treating cases one at a time and never reach the cause.

What embedded support actually looks like

Embedded support starts with how managers behave in ordinary moments, because the manager relationship shapes more of an employee's daily experience than any wellbeing programme ever will. A manager who notices a workload spiking, asks before assuming, and adjusts deadlines without making someone justify their limits does more for mental health than a meditation app reaching the same person. Train managers to spot strain and give them the authority to act on it. That authority is a structural choice, not a soft skill.

Workload design carries the same weight. When you assign work, you decide whether a role is sustainable or quietly corrosive, and most overload happens because no one owns the total demand on a person across competing priorities. Build a clear view of who carries what, and treat persistent overload as a design fault rather than a personal failing.

Autonomy and role clarity work together. People cope with demanding work far better when they control how they do it and understand exactly what success looks like, and ambiguity on either point produces the low-grade stress that accumulates into absence. Define decision rights. Make expectations explicit. Push real control downward.

Psychological safety becomes a feature of work when people raise problems early without fear of being penalised, which only happens when leaders respond to bad news with curiosity rather than blame. None of these are programmes you launch. They are decisions you make every week about how work runs. A strategy document that sits apart from those decisions changes nothing.

Seeing risk before it becomes absence

Targeted prevention depends on knowing which teams, roles, or cohorts carry disproportionate risk, and most organisations cannot answer that question. They can tell you the company-wide absence rate. They cannot tell you that the claims team has been running at 20% above capacity for four months, or that engagement scores in one division have fallen while turnover intentions have climbed. The risk is concentrated, but the data is averaged, so the signal disappears.

Sickness absence is a lagging indicator. By the time someone goes off sick with stress, the conditions that caused it have been present for weeks or months, and the cost has already landed. Acting on absence data means acting after the damage is done.

Leading indicators sit upstream of that point. Workload pressure, low autonomy, declining manager support, and falling psychological safety all show up in survey and operational data before they show up in absence. Read at the cohort level rather than the company average, those signals tell you where pressure is building and for whom. You cannot prevent what you cannot see, and a single headline absence figure shows you almost nothing about where to act.

From awareness to architecture

Awareness was the right starting point, but it is the wrong place to stop. The organisations that will move the HSE numbers are the ones that treat mental health as a design problem rather than a communications one. They build it into how work is assigned, how managers operate, and how risk is identified before it shows up as absence. That work is harder than running an awareness week, and it is the only approach the evidence supports.

Real change starts with seeing where risk is building, in which teams and for what reasons, then acting on those specific drivers rather than rolling out the same support to everyone. Champion Health works with organisations to surface that risk at the cohort level and direct prevention to where it will count, so people leaders can act before the cost lands.