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.
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
Most of this list describes low control, unrewarded effort, or both. Individual coping strategies will help you tolerate it and are unlikely to fix it, which is worth knowing before you spend a year trying.
A few of these are present in most jobs. Worth noticing which one you would most want changed, because that is the useful thing to raise.
Little of this is familiar. If you still feel wrung out, that is worth looking at separately rather than assuming the job is the cause.
The screener measures how stressed you have felt recently. It says nothing about whether the cause is your job, which is what this check is for.
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.