What the data actually shows
The clearest evidence that stress can lower your defences comes from controlled infection experiments. In a well-known line of research, Sheldon Cohen measured people's psychological stress and then deliberately exposed them to a cold virus under controlled conditions. People reporting higher stress were more likely to actually develop colds, which is strong evidence that stress can impair the immune response rather than just correlating with illness.
Over longer timescales, chronic stress is linked to inflammation, cardiovascular risk, and slower wound healing. Bruce McEwen's concept of 'allostatic load' captures the idea that the body's stress response is adaptive in bursts but causes cumulative wear when it stays switched on. Persistent activation of stress systems is associated with the kind of background inflammation and strain that contributes to heart disease and other chronic conditions.
But the relationship is conditional, not automatic. Acute, short-term stress can be benign or even helpful, mobilising energy and focus for a challenge. Whether stress harms health appears to depend heavily on how chronic it is, how much control the person feels, and how they cope. The data describes a risk factor that interacts with circumstances, not a simple cause-and-effect switch.
Why this feels different from how it actually is
Stress feels either more or less dangerous than the data says depending on which story you have absorbed. The dramatic version — that stress is a silent killer and any tension is toxic — overshoots, because it ignores how much short-term, controllable stress is harmless or even useful. The dismissive version — that it is 'all in your head' — undershoots, because the immune and cardiovascular effects are measurable and real.
It also feels uncertain because the effects are statistical and delayed. Stress rarely makes you visibly ill the next day in a way you can trace, and most stressed people do not get seriously sick, so the connection is easy to dismiss. The harm shows up as shifted odds across a population and accumulated load over years, which is exactly the kind of pattern individual experience is bad at perceiving.
And worrying about whether stress is harming you can itself feel like more stress. Beliefs about stress matter, and the looping anxiety of 'is my stress making me sick?' is part of why the honest, calibrated picture — real risk, not certain doom — is more useful than either extreme.
'Stress makes you sick' is true as a risk statement and false as a certainty.
What the research says to do about it
The most supported target is chronicity and control rather than the presence of stress at all. Reducing how long stress runs without relief, and increasing your sense of control over stressful situations, maps onto the factors the research most consistently links to whether stress harms health. Trying to eliminate all stress is neither realistic nor the point.
The well-established health basics double as stress buffers. Regular physical activity, adequate sleep, and strong social connection are each independently linked to better health and to a more resilient stress response. They are the highest-confidence levers available and matter regardless of how stressed you are.
For stress that is persistent or overwhelming, evidence-based approaches such as therapy and structured stress-management programmes have support. This is also the clearest signal in this whole topic that it is worth talking to a qualified clinician — persistent stress and unexplained physical symptoms both deserve real assessment, not self-management alone.
What the research says does not help
Trying to live a completely stress-free life does not help and is not what the research recommends. Some stress is normal and even adaptive, and treating every moment of pressure as a health emergency tends to add anxiety rather than remove risk. The target is chronic, uncontrolled stress, not stress itself.
Dismissing stress as purely psychological and therefore harmless is equally unsupported. The immune and cardiovascular effects are real and measurable, so 'just toughen up' ignores a genuine, if probabilistic, risk factor.
Generic 'just relax' advice and one-off wellness fixes rarely touch chronic stress, because the harmful version is defined by its persistence and lack of control. A single spa day or breathing app does not change a months-long situation that feels inescapable; the durable changes are to circumstances, coping, and support, ideally with clinical input where needed.
The harm shows up as shifted odds across a population and accumulated load over years — exactly the pattern individual experience is bad at perceiving.
What this looks like in real life
A hard week that passes
A tense stretch before a deadline — short-lived and within your control — is the kind of stress that is often harmless and can even sharpen focus. Treating that ordinary pressure as a health emergency tends to add anxiety rather than remove risk; the target is chronic, uncontrolled stress, not stress itself.
Months with no relief or control
Stress that grinds on for months without relief or a sense of control is the kind the research most consistently links to harm. A single spa day or breathing app does not change a months-long situation that feels inescapable; the durable changes are to circumstances, coping and support — ideally with clinical input where needed.
Real numbers in context
The strongest single piece of evidence here is experimental rather than a tidy statistic. In Cohen's controlled-exposure studies, people under higher psychological stress were meaningfully more likely to develop colds after the same deliberate virus exposure — direct evidence that stress can weaken immune defences, not just coincide with illness.
For the longer-term effects, the honest framing is shifted odds rather than precise numbers. Chronic stress is associated with higher inflammation, raised cardiovascular risk, and slower healing through the cumulative 'allostatic load' McEwen described, but the size of the risk for any one person depends on duration, control, coping, and many other health factors. Be wary of confident, exact claims that stress causes a specific disease in a specific share of people; the evidence supports a real risk factor, not a deterministic cause.
Educational only — not medical advice. For persistent stress or physical symptoms you are concerned about, see a qualified clinician.