Skip to content
MyFreud

Workplace Stress: What the Evidence Shows

Why control matters more than workload, what job strain does to your physical health, and which fixes work when the problem is the job, not your resilience.

5 min read

Pop-art illustration of a woman resting her head on one hand at a desk in front of a laptop, with a colleague working at a desk behind her.

Key takeaways

  • Demand alone predicts stress poorly. Karasek's finding was that high demand becomes harmful mainly when it meets low control, which is why a busy job you run can feel better than a quiet one you do not.
  • A second model adds the thing the first one missed: effort that goes unrewarded. Siegrist described the imbalance between what you put in and what comes back as status, security, pay or recognition.
  • The health effects are physical and measurable. A pooled analysis of individual participant data across European cohorts found job strain associated with a modest but real increase in coronary heart disease risk.
  • Burnout is classified as an occupational phenomenon, not a medical condition, and the World Health Organization was deliberate about that. It is something a job does to a person, which places the problem in the work rather than the worker.
  • Individual coping skills help and cannot fix a structural problem. Where the cause is workload, understaffing or a manager, the evidence favours changing those, and the honest personal question becomes what you will do about a job that is not going to change.

The most useful finding in the research on workplace stress is that how much work you have predicts your health surprisingly badly, and how much say you have over it predicts it rather well. [karasek-1979-demand-control] That single result explains why a punishing job you control can feel survivable while an undemanding one you do not can be corrosive, and it changes what you should be looking at when you try to work out what is wrong. This guide covers what the evidence shows, what it does to the body, and which fixes have support behind them.

Demand is not the problem, demand without control is

Karasek’s demand-control model is the foundation of this field, and its claim is specific: psychological demand becomes harmful mainly in combination with low decision latitude, meaning little say over how and when the work is done. High demand with high control is the profile he called active work, and it does not carry the same cost.

This is why “I am not even that busy” is such a common and misleading self-report. The person doing moderate volume on somebody else’s schedule, unable to reorder tasks, pause, or choose method, is in the quadrant the model predicts trouble in. The question that separates a hard job from a harmful one is not how much, but how much of it you decide.

What the first model missed

Siegrist added the piece that demand and control do not capture: the balance between effort put in and reward coming back. [siegrist-1996-eri] Reward here is broader than pay, covering recognition, status, security and prospects, and the imbalance is what produces the specific bitterness of working hard for something that never arrives.

The two models pick up different jobs. Demand-control describes the assembly line and the call centre. Effort-reward imbalance describes the person on a fixed-term contract doing work they are not credited for, or the one repeatedly passed over after being told to be patient. Between them they cover most of what people mean when they say a job is making them ill.

Which features of a job the two models pick up Illustrative
0 25 50 75 100 Relative contribution 3 Volume of work 8 Say over method 7 Say over timing 6 Ability to pause
0 25 50 75 100 Relative contribution 5 Effort put in 8 Recognition 7 Job security 6 Prospects

A schematic of the emphases in Karasek (1979) and Siegrist (1996). Heights show which features each model treats as central, not measured effect sizes.

What it does to the body

The cardiovascular findings are the most solid part of this literature. Kivimäki and colleagues pooled individual participant data from European cohort studies and found job strain associated with an increased risk of coronary heart disease, an increase that was modest in size but consistent across cohorts. [kivimaki-2012-jobstrain]

Separately, long working hours have been linked to ischaemic heart disease and stroke, with Pega and colleagues estimating a substantial global burden of disease attributable to them. [pega-2021-longhours] The size of these effects is worth keeping in proportion: they are real and they are not individually enormous, which is the correct way to hold most epidemiological findings.

The effects most people actually notice are smaller and earlier. Sleep that breaks up at four in the morning, digestive symptoms, headaches, catching every minor illness going round, and a Sunday evening that starts feeling like Monday. Our guides to the physical symptoms of stress and why you get ill after stress cover the mechanisms.

Burnout is classified as something the job does

The World Health Organization includes burnout in ICD-11 as an occupational phenomenon resulting from chronic workplace stress that has not been successfully managed, and states explicitly that it is not classified as a medical condition. That wording was chosen carefully and it locates the cause in the work.

Two practical consequences follow. A doctor is unlikely to write burnout as a diagnosis and may record depression, anxiety or an adjustment reaction instead, which can feel like being disbelieved and is not. And the definition itself argues against the standard response of sending the exhausted person on a resilience course, since it names the workplace as the thing that has gone wrong. Our guide to burnout and depression covers the overlap, which is substantial.

Which quadrant is your job in?

Tick anything true of an ordinary week. This is about the shape of the job rather than how you are coping with it.

0 of 6 ticked

What actually helps

The uncomfortable finding is that interventions aimed at the individual perform worse than interventions aimed at the work. Resilience training, mindfulness sessions and stress-management workshops produce small effects that tend to fade; changes to workload, staffing, scheduling and control perform better and last longer.

That is not an argument for doing nothing personally, and the individual measures with the best support are unglamorous: protecting sleep, taking actual breaks, and building a stopping point into the evening that is enforced by something other than willpower. Our guides to nervous system regulation and stress management techniques cover the specifics.

It is an argument for being clear-eyed about which problem you have. If the cause is understaffing, no amount of breathing practice resolves it, and the honest question becomes what you are going to do about a job that is not going to change. That question is uncomfortable, and it is more useful than a wellbeing app.

When to speak to someone

Speak to a doctor if low mood or anxiety has persisted for more than two weeks, if sleep has been disturbed for a month or more, if you are drinking more to wind down, or if physical symptoms have appeared with no other explanation. Bring specifics: when it started, what an ordinary week looks like, what has changed at work.

Say what your work actually involves rather than your job title. Hours, control, and whether you can stop are the details that make the difference to what gets suggested, and they are the details people most often leave out.

How MyFreud can help

The argument about whether it is the job or you is usually settled by a record rather than by reflection. Tracking mood daily shows what happens across a week, on leave, and after a change at work, which is exactly the comparison that memory handles badly and a chart handles well.

Frequently asked questions

What is job strain?

Job strain is the combination of high psychological demand with low decision latitude, meaning little control over how and when the work gets done. The concept comes from Karasek, and its value is that it explains something workload alone does not: why two people doing similarly demanding work can have very different stress outcomes. The person who decides the order of their own tasks, can pause, and has some say in method fares better than the person doing the same volume to somebody else's schedule. If you are trying to work out what is wrong with a job, control is usually a more revealing question than hours.

Is burnout a medical diagnosis?

No, and the distinction is deliberate rather than a technicality. The World Health Organization includes burnout in ICD-11 as an occupational phenomenon, defined as resulting from chronic workplace stress that has not been successfully managed, and specifically states it is not classified as a medical condition. Practically this means a doctor is unlikely to write burnout as a diagnosis, and may instead record depression, anxiety or an adjustment reaction if those are present. It also means the definition points at the workplace, which is where the evidence suggests the more effective interventions sit.

Can workplace stress actually make you physically ill?

The cardiovascular evidence is the strongest. Kivimäki and colleagues pooled individual participant data from European cohort studies and found job strain associated with an increased risk of coronary heart disease. The increase was modest in relative terms and the authors were careful about that, but it was consistent, and it held after accounting for a range of other factors. Long working hours have separately been linked to ischaemic heart disease and stroke in work by Pega and colleagues. The everyday effects are less dramatic and more familiar: disturbed sleep, digestive symptoms, headaches and frequent minor illness.

Do workplace wellbeing programmes work?

The evidence is much weaker than the volume of provision suggests, and the pattern in it is informative. Interventions aimed at the individual, such as resilience training, mindfulness apps and stress-management workshops, show small effects that fade. Interventions that change the work itself, such as workload, staffing, scheduling and the amount of control people have, perform better. This is an awkward finding for employers because the first kind is cheap and the second is not, which is part of why the first kind is what most people are offered.

How do I know if it is the job or me?

A useful test is what happens on holiday and what happened in previous roles. If a week away restores you and the feeling returns within days of going back, that points at the job. If the same pattern followed you through several different roles with different managers, that points at something more portable, which may be treatable in its own right. The two also combine: a demanding job is harder to survive while depressed, and being persistently under-recovered makes an ordinary job feel unmanageable. It is not always one or the other, and a doctor can help separate them.

References

  1. 1.Karasek RA ( 1979). Job demands, job decision latitude, and mental strain: implications for job redesign. Administrative Science Quarterly.
  2. 2.Siegrist J ( 1996). Adverse health effects of high-effort/low-reward conditions. Journal of Occupational Health Psychology.
  3. 3.Kivimäki M, Nyberg ST, Batty GD, Fransson EI, Heikkilä K, Alfredsson L, et al. ( 2012). Job strain as a risk factor for coronary heart disease: a collaborative meta-analysis of individual participant data. The Lancet.
  4. 4.Pega F, Náfrádi B, Momen NC, Ujita Y, Streicher KN, Prüss-Üstün AM, et al. ( 2021). Global, regional, and national burdens of ischemic heart disease and stroke attributable to exposure to long working hours. Environment International.