Psychosocial work environment: a practical guide to measuring and improving it

When the Swedish Work Environment Authority (Arbetsmiljöverket) reports on working conditions, the headlines tend to focus on what is visible, such as the fact that Sweden still has almost one workplace death a week, with no improvement since the 2000s. The psychosocial work environment is not visible in the same way. Stress, unclear expectations and victimisation leave no cordons behind them, yet they are part of the same employer responsibility, and most organisations still measure them only once a year. The question worth asking is: how is any of it actually followed up?

This guide covers what the psychosocial work environment really is, where to find the formal requirements, and how to build a measurement practice that catches problems before they turn into sick leave and resignations.

What is the psychosocial work environment?

The psychosocial work environment is the part of the working environment that concerns how work is organised and how people feel and interact at work. It differs from the physical work environment, which covers premises, machinery and chemicals. It includes workload, influence over your own work, support from managers and colleagues, the social climate, clarity about roles and expectations, and the presence of conflict and victimisation.

At the Swedish Work Environment Authority, Prevent and Suntarbetsliv you will more often meet the term organisational and social work environment (OSA, from the Swedish "organisatorisk och social arbetsmiljö"). In practice it means the same thing: the organisational part is about how work is managed, allocated and resourced, the social part about interaction, collaboration and social support. When you see "OSA" in regulations, surveys and training, it is the psychosocial work environment that is meant.

OSA and the legal requirements: where to find them and what the work centres on

An employer's responsibility for the work environment does not stop at physical risks; it also covers the organisational and social work environment. The exact requirements are set out in the Authority's provisions on the organisational and social work environment. We deliberately do not reproduce them in detail here: the rules are changing, and the current wording at the Authority is always what applies. One example of that change: according to Suntarbetsliv, the Swedish government presented a new work environment strategy in June 2026 that includes removing sanctions for OSA breaches. What that means in practice remains to be seen, which is all the more reason to stay up to date.

What OSA work centres on in practice, however, is clear. Prevent's course on OSA in practice covers stress, ways of working, bullying and conflict, and its materials also highlight recovery and managers' responsibility for the work environment. That is where the risks arise, and that is where your measurement needs coverage.

Both the Authority and Suntarbetsliv also point to systematic work environment management (SAM, from the Swedish "systematiskt arbetsmiljöarbete") as the framework for how this work should be done. Whatever the exact wording of the provisions, the logic of a systematic approach stays the same: investigate the work environment, assess the risks, act on them and check that the actions worked. It is a continuous cycle, not an annual event. This is where many organisations go wrong: they treat the investigation as done once the annual survey is complete, but have no way of spotting risks that emerge in between.

Why the annual OSA survey is not enough

The classic OSA survey, such as the tool of the same name from Prevent, is a good instrument for a structured snapshot. The problem is not the survey itself but the frequency. An annual measurement has three built-in weaknesses:

1. It measures too late. Unhealthy workloads build up over weeks and months. A risk that emerges in February is not caught until the autumn survey, if the employee is still there by then.
2. It gives a snapshot, not a trend. An average once a year says nothing about direction. Is the team's stress level going up or down? Without a time series you cannot know, and without direction you cannot prioritise.
3. It separates measurement from action. By the time results land three months after the survey, the situation has often changed, and the link between answers and action becomes unclear to employees. That erodes trust in the whole process, a pattern we have written more about in Survey fatigue is a leadership problem.

Systematic work environment management is continuous by nature. An annual survey is an annual check, not a system.

The conclusion is not to scrap the in-depth annual survey but to complement it. Short, frequent pulse measurements or quarterly deep dives provide the ongoing signal. In-depth surveys provide the detail when the signal shows that something needs a closer look. We compare the approaches in detail in Annual employee engagement surveys vs pulse surveys.

What should you measure? Six areas that capture the psychosocial work environment

For continuous measurement to be useful in OSA work, the questions need to cover the areas where risks actually arise. Six areas give a good foundation:

• Workload and stress. Do employees feel that demands are reasonable given their time and resources? The balance between demands and resources is the core of all OSA work, and often the first place problems show.
• Recovery. Do employees have time to wind down between shifts? Signals of poor recovery often come before signals of stress.
• Clarity. Does everyone know what is expected, who decides what and what to prioritise when not everything can get done? Lack of clarity is one of the most common roots of both stress and conflict.
• Leadership and support. Do employees get the support they need from their immediate manager? The manager is the one who has to act on risks day to day, so you need to know how that support is experienced.
• Social climate and psychological safety. Do people dare to say what they think, raise problems and ask for help? An unsafe climate also makes every other metric less reliable.
• Bullying and victimisation. Ask the question explicitly and regularly. People who are targeted rarely answer in an annual survey with a long lag, but may do so in an anonymous, recurring measurement. We explain why anonymity is crucial in Why is anonymity important in employee surveys?

How to build a continuous measurement practice, step by step

1. Start from the systematic cycle. Investigate, assess risks, act, follow up. Your measurement practice should feed every step, not be a standalone HR activity.
2. Choose frequency by purpose. More frequent measurement, either short pulse measurements every other week or quarterly follow-ups, can capture trends and early signals.
3. Guarantee anonymity and explain how it works. Without confidence in anonymity you get polite answers, not true ones.
4. Give the results to managers, not only to HR. Teams are where risks arise and get dealt with. Managers need to see their own team's trend in real time, with HR as support rather than middleman.
5. Set thresholds for action. Decide in advance what happens when an area drops: who looks at it, within what time, and when it is escalated. A signal with no recipient is not a risk assessment.
6. Document actions and follow up in the data. Follow-up is the last step of the cycle, and with continuous measurement you can actually see whether an action worked, instead of waiting a year for the next survey.

From measurement to action: where OSA work is decided

No regulator is impressed by a beautiful dashboard. What counts, for the organisation and the people in it, is that risks are found and dealt with. Here continuous measurement is not just a compliance tool but a leadership tool: when a manager sees their team's temperature on workload fall week by week, the conversation can happen while the problem is still small. Small frictions left to grow become expensive, as we have shown in The hidden cost of workplace friction.

That is exactly the logic Winningtemp is built on: short, research-based temperature measurements that continuously capture workload, leadership and psychological safety, among other things, anonymously and in real time, so that both managers and HR can act on direction instead of reacting to last year's average. But the principle stands on its own: whatever the tool, continuous measurement with clear ownership of action is what makes OSA work genuinely systematic.

If you need to go deeper, there is good material from Suntarbetsliv, a joint initiative of trade unions and employer organisations that offers free tools and training with a focus on municipalities, regions and the public sector, and from Prevent, which offers OSA training and survey tools.

What to do on Monday morning

• Map your current cycle. How often do you investigate the psychosocial work environment today, and how long does it take to get from answer to action? If the answers are "once a year" and "several months", you know where the gap is.
• Check coverage. Are you measuring the areas where risks actually arise (workload, recovery, clarity, bullying and conflict), or only general job satisfaction?
• Read the current provisions. Go to the Swedish Work Environment Authority's pages on the organisational and social work environment and check your work against the wording in force, especially now that a new work environment strategy is about to change the playing field.
• Define the chain of action. Write down who sees the results, who acts and within what time. That is the difference between a survey and systematic work environment management.
• Pilot continuous measurement in part of the organisation. Compare after a quarter: how many risks did you catch that the annual survey would have missed?