Health anxiety

Health anxiety is a persistent fear of being seriously ill that survives the evidence against it. The defining feature is not the fear — almost everyone has had a frightening week waiting for a result — but what happens after the answer arrives. In health anxiety the answer closes the question for hours rather than months, and then the question reopens unchanged.

It is made of three ordinary processes, all of which everybody has: what attention lands on inside the body, what the mind concludes from it, and what gets done afterwards to settle the matter. None of the three is a malfunction on its own. The pattern is in how they chain together, and in the fact that the third one feeds the first.

This belongs here, second, before any of the rest of it, because everything after it can be misread as the claim that the symptoms are invented. They are not. Fear of illness has never protected anyone from illness, and nothing on this page is evidence about whether you are ill.

A Swedish national cohort published in 2023 followed 4,129 people diagnosed with hypochondriasis against 41,290 matched comparisons. Mortality in that group was higher, not lower — all-cause hazard ratio 1.69, with deaths from natural causes elevated and mean age at death five years earlier. Deaths by suicide were markedly more common, which should settle any suggestion that this is a comic or trivial complaint.

So the rule, stated plainly: a symptom that is new, or persistent, or changing deserves one proper look from a doctor. One, done properly. The distinction this entry is about is not between checking and not checking. It is between checking once and accepting the answer, and checking and then continuing to check. Anxiety cannot be used to rule anything out, in either direction: an anxious person with a new symptom has a new symptom, and the reasoning that says it is probably just stress is a guess being made in the wrong direction. How stress shows up in the body covers the other half of that — real physical symptoms produced by real systems, which is why "it is stress" has to be a conclusion reached after the alternatives, not instead of them.

They come apart, and a person can be high on one and low on another, which is why "worrying about your health" is too coarse to be useful.

Noticing. A body under surveillance is a noisy body. Everybody's produces a constant stream of twinges, flutters, skipped beats and odd little sensations; what varies is how much of it reaches awareness. Sustained attention lowers that threshold, so the scanning produces more material for itself. This is the sense described in interoception, and the useful finding there transfers: what matters is less how accurately the body is read than the gap between accuracy and confidence.

Interpreting. The sensation gets read, and read as evidence. Salkovskis and Warwick's account, which is still the spine of the clinical models, puts the misinterpretation at the centre: an ordinary bodily event is taken as a sign of serious disease, which raises arousal, which produces more bodily events. The worst explanation arrives first and feels like the most responsible one.

Settling it. Searching, checking, examining the same patch of skin, asking someone, booking an appointment, or refusing to. This is the part that keeps the pattern alive, and it is the subject of reassurance seeking.

The chain runs in a circle rather than a line. Each completed loop raises the resting level of alarm slightly, which lowers the noticing threshold, which produces the next sensation sooner. People usually read the shortening interval as the illness progressing. It is the loop doing what reinforced loops do.

This is the part most descriptions get wrong, and it changes who recognises themselves on the page.

DSM-5 distinguishes a care-seeking from a care-avoidant presentation of illness anxiety disorder. In practice the division is much less tidy than that: in the sample discussed in a 2024 review, a quarter were care-seeking, a seventh were care-avoidant, and the majority moved between the two.

Avoidance buys the identical relief on a slower schedule. Not booking the appointment closes the question by refusing to open it. The anxiety drops. The loop is reinforced in exactly the same way, by exactly the same mechanism. The difference is only in which door gets used.

Which is why the avoidant version goes unchallenged for years. From outside it reads as not minding, sometimes as an admirable lack of fuss, while the other version acquires a reputation for being difficult. The person who has not been to a doctor in eleven years because of what might be found is as deep inside this as the person on their fourth opinion in a month, and is in more danger, because the one proper look has not happened.

The practical consequence worth knowing: it is possible to appear to have solved this by force of will, having simply switched poles. Nothing about the loop has changed when that happens. Only the direction of the thing that closes the question.

It is the word people type, so it is worth taking apart once.

Hypochondriac is not a clinical term and has not been one for a long time. Its surviving use is as an insult between people who are not ill, and nothing it implies is accurate: this is not performance, not a bid for attention, not weakness, and not the same thing as worrying a lot in general. The name itself is a fossil of a discarded theory — the hypochondrium is the region under the ribs, where the complaint was once thought to originate.

The manuals went different ways with it, and knowing that saves confusion when the term turns up anyway. DSM-5 removed hypochondriasis in 2013 and distributed what it covered between two descriptions: somatic symptom disorder, where there are distressing symptoms, and illness anxiety disorder, where the fear is of illness that has not shown up. ICD-11 kept the name and reclassified it among the obsessive-compulsive and related disorders, which is an argument about machinery rather than about vocabulary, and a defensible one: the resemblance to checking compulsions is strong.

So a reader who looks this up gets two different official answers, and neither manual is using the word the way a family member does across a dinner table. Health anxiety is the description that is accurate, usable, and not an accusation, and it is the only one used elsewhere on this page.

This is not encouragement. It is the single most load-bearing fact on the page, and it is the one least often said, because health anxiety gets treated as a character trait to be endured more than almost anything else in the category.

A systematic review of fourteen randomised trials found a large effect for cognitive behavioural therapy adapted to health anxiety, against waiting lists, treatment as usual, medication and other psychological therapies, with the effect holding at six and twelve months. In the largest trial in the area, 444 health-anxious patients recruited from hospital medical clinics were randomised to a short course or to standard care, and the advantage was still measurable eight years later. Short course: single figures of sessions, not years.

Very little in this field has a record that clean. What it does not mean is that anything on this page constitutes the treatment, and the mechanics of how that therapy is done are not ours to give.

Two neighbouring pages are worth knowing apart from this one. Reassurance seeking is the behaviour, across every subject it attaches to, and intolerance of uncertainty is what is actually being chased — certainty rather than information, which is why no examination finishes the job, since examinations return probabilities and none of them returns zero. The long version of the mechanism, with the follow-up study that measured how fast medical reassurance drains, is in why reassurance makes health anxiety worse.

The scope note. Noticing the body and being frightened by it are universal, and a frightening fortnight after a relative's diagnosis is not a disorder; the pattern described here is one that has been running for months and has started to cost time, money, sleep or relationships. Nothing here is a diagnosis, and a questionnaire cannot make one — anxiety screening explains why the arithmetic does not allow it. The health anxiety test reads the three parts separately, because they come apart and each points somewhere different.

The first sign that the loop is loosening is not confidence about the body. It is a sensation that gets noticed, read as nothing in particular, and then forgotten by the evening.

Talking to someone about this is not an overreaction, and it does not require having worked out what to say first. And if the thoughts have moved from fearing illness to not wanting to be here, that is not the subject of this page and it does not wait for an appointment. Talk to someone today, a doctor, an emergency service, or a crisis line in your country.