Health Anxiety: Why Reassurance Makes It Worse, What the Mortality Research Actually Found, and Treatment That Works

Tags: Health Anxiety, Hypochondriasis, Remote Treatment, Mortality Research, Mental Health

Health Anxiety: Why Reassurance Makes It Worse, What the Mortality Research Actually Found, and Treatment That Works

In brief

This article explores a Swedish study on health anxiety, detailing how reassurance can worsen symptoms, findings on mortality rates, and effective remote treatments.

Key takeaways

  • Reassurance in health anxiety can lead to worse mental health outcomes.
  • A Swedish study found individuals with hypochondriasis die earlier, not from feared diseases.
  • Effective remote treatments are available for managing health anxiety.
A person using a phone in a dark room at night Health Anxiety

Health Anxiety: Why Reassurance Makes It Worse, What the Mortality Research Actually Found, and the Treatment That Works Remotely

A Swedish study followed 4,129 people diagnosed with hypochondriasis. They died about five years earlier than matched peers — but their cancer rate was no higher at all. What they died of instead changes how this condition should be understood.

You find the lump, or the flutter, or the patch of skin that looks different than it did. You search it. The first result is reassuring and the fourth one is not, so you keep going, and somewhere around midnight you are reading a forum post by someone whose symptoms started exactly like yours. You book the appointment. The doctor examines you, orders a test, and the test is normal — and for about four days you feel genuinely fine.

Then you notice something else.

Health anxiety is among the most dismissed conditions in medicine. The word most people reach for is "hypochondriac," which is a punchline rather than a diagnosis. That framing is wrong on the facts, and the most striking evidence against it comes from a study published in JAMA Psychiatry that followed thousands of people with the diagnosis and found they died meaningfully earlier than their matched peers — just not of anything they were afraid of.

70 vs 75Mean age at death, hypochondriasis vs matched comparison group 0.99Hazard ratio for cancer death — no excess risk whatsoever 2.33 / 1.52Hazard ratios for respiratory and circulatory death 76%Responded to internet-delivered CBT in a randomized trial

What Is Health Anxiety, Clinically?

Short answer

Health anxiety is preoccupation with having or developing a serious illness, out of proportion to any actual evidence, lasting at least six months and causing real distress or disruption. In DSM-5 it's split across two diagnoses: illness anxiety disorder, where physical symptoms are mild or absent and the fear itself is the problem, and somatic symptom disorder, where there are significant physical symptoms accompanied by disproportionate worry. "Hypochondriasis" is no longer a standalone diagnosis — roughly three-quarters of people who met the old criteria now fall under somatic symptom disorder.

Estimates put health anxiety at somewhere between 1.3% and 10% of the population depending on the definition used, and 3% to 8% in general medical settings. It affects men and women about equally. And it shows up far more often in primary care and specialist clinics than in psychiatric ones — which is the central practical problem, because the places people take it are not the places equipped to treat it.

One feature deserves its own paragraph because almost nobody knows about it. The diagnosis has two specifiers, and they point in opposite directions:

SubtypeWhat it looks likeWhy it gets missed
Care-seeking Frequent appointments, repeated tests, multiple opinions, persistent requests for scans and bloodwork It doesn't get missed — it gets labelled. This is the person who becomes known at the clinic, often unkindly
Care-avoidant Avoids doctors entirely. Won't go for screening, puts off appointments for years, doesn't want to know Looks like indifference or neglect rather than fear. These patients are invisible to the health system until something is advanced

The care-avoidant subtype matters enormously for what comes next, and it is the piece most often left out of articles like this one. Not everyone with health anxiety is in the waiting room. Some are avoiding it so successfully that nobody has any idea they're frightened.

The Mortality Finding — and What It Actually Means

Short answer

A Swedish nationwide matched-cohort study published in JAMA Psychiatry followed 4,129 people diagnosed with hypochondriasis and 41,290 matched comparisons for a mean of 7.7 years. Mean age at death was 70.0 years versus 75.1. All-cause mortality hazard ratio was 1.84, falling to 1.69 after adjusting for socioeconomic factors. The excess was concentrated in circulatory disease, respiratory disease, and external causes. Cancer mortality showed no excess at all — hazard ratio 0.99.

Sit with that cancer number for a second, because it's the whole point. Cancer is the single most common focus of health anxiety. These are people who spent years checking for it, reading about it, asking to be tested for it. And they died of it at exactly the same rate as everyone else.

What They Died Of Instead

0.99Cancer — no excess risk. The thing most feared was not the thing 1.52Circulatory disease 2.33Respiratory disease 2.43External causes 4.14Suicide — though this lost significance after adjusting for depressive and anxiety disorders

The authors' own summary was that these individuals face increased mortality "mainly from potentially preventable causes," and that clinicians should be aware they are at risk of death by suicide.

Their stated caveats deserve equal weight. There was no primary care data. Undetermined deaths were counted as suicides, and the authors note it isn't possible to know whether all of them were. Limited statistical power forced some causes to be grouped. And because the condition is underdiagnosed, some people who actually had it were almost certainly sitting in the comparison group — which, as the authors put it, means the true excess mortality "may be even higher than reported."

So how does a condition defined by fear of illness end up shortening life without raising cancer risk? The plausible mechanisms are mundane and all of them are addressable. Chronic anxiety has cardiovascular consequences. Smoking rates and other health behaviors differ. The care-avoidant group misses screening and routine management that would have caught the circulatory and respiratory problems early. And untreated anxiety disorders carry their own risk — which is where the suicide figure comes in.

What this research does not support is the reading that health anxiety is harmless because the cancer never came. It supports the opposite: that this is a serious condition with measurable consequences, and that the consequences run through exactly the kind of ordinary preventable illness that routine medical care is good at catching.

Do not use this article to talk yourself out of seeing a doctor. Health anxiety and real disease are not alternatives — people with health anxiety get sick at normal rates, and the mortality data above points to preventable causes, which means routine care matters more for this group rather than less. Nothing here suggests your symptoms are imaginary or that a new or changing symptom should be ignored. Get it checked. The argument in this article is about what happens in the weeks after a normal result, not about whether to seek one. A quiet hospital waiting area with empty seats Health anxiety presents far more often in primary care and specialist clinics than in psychiatric ones — which is part of why it so often goes untreated.

Why Does Reassurance Make It Worse?

Short answer

Because relief is a teacher. When you check, search, or get reassured, the anxiety drops — fast. Your brain registers that the checking is what made it drop, and files the checking as necessary. The next time a sensation appears, the urge arrives sooner and stronger, and the amount of reassurance needed to settle it goes up. The short-term relief is genuine. It's also the mechanism keeping the whole thing running.

This is the single most useful idea in this article, because it explains something that otherwise makes no sense: why a normal test result helps for four days instead of forever.

1. NoticeA sensation registers — a flutter, an ache, a mark. Bodies produce these constantly. 2. InterpretIt gets read as a sign of something serious rather than as noise. 3. CheckSearch it, examine it, ask someone, book an appointment, request a test. 4. ReliefAnxiety drops sharply. This is real and it feels like resolution. 5. The costThe brain learns the check was what saved you. Next time the threshold is lower.

In clinical terms, checking and reassurance-seeking are safety behaviors — actions that reduce distress in the moment while preventing you from ever learning that the distress would have passed on its own. The same mechanism appears in panic disorder, where carrying a water bottle or sitting near exits does identical work, and the exposure therapy literature consistently finds that safety behaviors blunt treatment benefit.

Which leads somewhere counterintuitive. The most loving thing a family member can do is often not to answer the question for the fortieth time. That isn't coldness, and it shouldn't be delivered coldly — but a partner who reliably provides reassurance has become part of the loop. This is something worth negotiating together, ideally with a clinician, rather than one person unilaterally going quiet.

Is This the Same as OCD or Panic Disorder?

Short answer

No, though they're close relatives and get confused constantly. Health anxiety centers on the belief that you are ill now. OCD's health-related obsessions tend to focus on becoming ill in the future, usually alongside obsessions about other things. Panic disorder is episodic — acute surges in which bodily sensations get read as a medical emergency — rather than the persistent background fear of health anxiety. People commonly have more than one.

 Core fearTime courseCharacteristic behavior
Health anxiety "I already have something serious and it hasn't been found" Persistent and enduring, by definition six months or more Body checking, searching, repeated testing — or total avoidance of medical care
OCD "I will contract or cause something" — typically future-oriented Intrusive and recurrent, usually alongside non-health obsessions Compulsions, often washing, checking or mental rituals
Panic disorder "This is happening right now and I'm dying" Discrete attacks peaking within minutes, plus worry about the next one Avoiding places and activities that produce the sensations

The distinction isn't academic. Treatments overlap substantially — all three respond to exposure-based CBT — but the specific targets differ, and getting the formulation right shapes what you actually practise. It's also why a careful diagnostic evaluation is worth more than it sounds for something that can feel so obvious from the inside.

What About Searching Symptoms Online?

Short answer

Searching doesn't cause health anxiety, but it is close to a perfect accelerant for it. Research on what's sometimes called cyberchondria describes escalating searches that increase rather than reduce distress. The structural problem is that the internet is an infinite reassurance source that can never finish the job — there is always one more result, and the rare terrible explanation is always available somewhere.

Two features make it unusually effective at feeding the loop. First, search engines surface the serious alongside the benign without weighting by likelihood, so a headache returns both dehydration and brain tumour with no sense of base rates. Second, it's available at 2 a.m., when no doctor is, and when the capacity to evaluate what you're reading is at its lowest.

The useful intervention is rarely "stop searching," which almost nobody manages as a flat prohibition. It's treating searching as the safety behavior it is and working on it deliberately, which is exactly what treatment does.

What Actually Works?

Short answer

Exposure-based cognitive behavioral therapy is the best-supported treatment, and the evidence for delivering it remotely is unusually strong. In a randomized non-inferiority trial of 204 adults published in JAMA Psychiatry in 2020, internet-delivered CBT was non-inferior to twelve weekly face-to-face sessions — 76% versus 74% response — while requiring about 10 minutes of therapist time per patient per week instead of 46. The authors concluded it "should be considered a first-line intervention for health anxiety."

That trial is worth describing properly, because it is rare for a remote treatment to be tested head-to-head against the in-person version and come out level.

Participants had a principal diagnosis of DSM-5 somatic symptom disorder or illness anxiety disorder, recruited through a primary care clinic in Stockholm, and were randomly assigned to twelve weeks of therapist-guided internet CBT or twelve weekly individual sessions. On the primary outcome the difference was effectively zero. Net societal costs were lower for the internet arm, mainly through reduced therapist time. Adverse events were comparable — 20% versus 18% — with no serious adverse events.

The honest caveats, which the authors raise themselves: the sample was highly educated, face-to-face showed slightly better adherence and somewhat larger long-term effects on general anxiety and depression, and at twelve months there was a 2.4-point difference favouring face-to-face that was clinically but not statistically significant — the trial wasn't powered to test non-inferiority at follow-up.

Exposure and response prevention

The core of treatment. Deliberately sitting with the uncertainty and the sensation without checking, searching or seeking reassurance — so the fear gets a chance to fall on its own.

Working on the interpretation

Not positive thinking, but learning to weigh likelihood properly — that the same sensation has a hundred benign explanations and that uncertainty is not the same as danger.

A negotiated reassurance plan

Agreeing in advance how family will respond to the question, so nobody is improvising and nobody is cast as the villain for declining to answer.

Treating what else is there

Depression and other anxiety disorders commonly co-occur, and the suicide signal in the mortality data largely tracked them. Treating those is not a detour.

Medication where indicated

SSRIs have evidence in health anxiety, and are particularly worth discussing when depression or another anxiety disorder is also present. Best alongside the behavioral work.

A sane medical plan

Agreeing with a physician on what routine care and screening you will have, and on what basis new symptoms get assessed — which protects the care-avoidant as much as the care-seeking.

That last card is the one that distinguishes good treatment from bad. The goal is not to stop you going to the doctor. It's to replace reactive, anxiety-driven testing with a plan you and a physician agree on in advance — so that appropriate care happens on schedule rather than at 2 a.m., and so the mundane preventable things the mortality research points to actually get caught.

7 Things Worth Bringing to a First Appointment

  1. How long it's been going on. Six months is the diagnostic threshold, so the timeline matters more than the current intensity.
  2. What you actually do about it. Searching, body checking, asking for reassurance, booking appointments — or avoiding all of it. Be specific; this is the treatment target.
  3. Which subtype you are. Care-seeking and care-avoidant need different plans, and the avoidant version is easy to under-report.
  4. What it's costing you. Hours lost, relationships strained, appointments made, screening skipped.
  5. What tests you've already had. So nobody repeats them, and so the pattern is visible.
  6. Any low mood or hopelessness. Depression travels with this often, and it changes the urgency.
  7. Whether you've had thoughts of suicide. The mortality research makes this worth asking directly, even when it feels unrelated to a fear of illness.

Where to Find Reliable Information

Frequently Asked Questions

How do I know if I have health anxiety or a real undiagnosed illness?

You don't settle that by reading, and you shouldn't try to. The honest answer is that these aren't mutually exclusive — people with health anxiety develop real illnesses at ordinary rates, which is exactly why the advice is to get symptoms assessed rather than to reason your way out of them. What points toward health anxiety is the pattern: repeated normal results that reassure only briefly, fear that moves from one illness to the next, and a six-month-plus course. A clinician can hold both possibilities at once, which is harder to do alone at midnight.

Why does a normal test result only help for a few days?

Because the relief it provides is the thing that keeps the cycle going. A normal result drops your anxiety sharply, which teaches your brain that getting tested is what resolved the danger — so the next sensation triggers the urge sooner and needs more reassurance to settle. This is why reassurance behaves like a short-acting medication with a tolerance effect rather than like a cure, and it's the core thing treatment targets.

Does health anxiety mean I'm more likely to get seriously ill?

Not of the thing you're afraid of. In the Swedish cohort, cancer mortality was identical between groups — hazard ratio 0.99. But overall mortality was higher, concentrated in circulatory disease, respiratory disease, external causes and suicide, which the authors described as "mainly potentially preventable." The practical reading is the opposite of fatalism: routine care, screening and treating the anxiety itself all matter more for this group, not less.

Should I stop googling my symptoms?

As a flat rule it rarely survives contact with a frightening sensation, so treating it as a willpower problem usually fails. It's more useful to recognise searching as a safety behavior — something that reduces distress now at the cost of making the next episode worse — and to work on reducing it deliberately as part of treatment, which is a standard part of exposure-based CBT. Most people find a negotiated reduction more achievable than an outright ban.

My partner keeps asking me to reassure them. What should I do?

Repeated reassurance feels like help and functions as fuel, so continuing indefinitely isn't kind in the long run. But going quiet without warning lands as rejection and usually backfires. The better route is to agree a plan together, ideally with a clinician involved: what you'll say, what you won't, and what you'll do instead — which is often sitting with them through the discomfort rather than resolving it. That's a joint decision, not something to impose.

Can this be treated by telehealth?

Unusually well, and this is one of the few conditions where there's direct head-to-head evidence. In the 2020 JAMA Psychiatry trial, therapist-guided internet CBT was non-inferior to twelve weekly in-person sessions, with 76% versus 74% response and far less therapist time required. Psychiatric evaluation and medication management also work by video. The authors argued remote delivery should be considered first-line, partly because it widens access to a treatment most people never get offered at all.

If You Need Help Right Now

If you're having thoughts of harming yourself, call or text 988 to reach the 988 Suicide & Crisis Lifeline, available 24/7 across the United States, or chat at 988lifeline.org. The mortality research above found an elevated suicide risk in this group, which is a reason to treat these thoughts as worth saying out loud rather than as separate from the health fears.

If you have symptoms that could indicate a medical emergency — chest pain, difficulty breathing, signs of a stroke — call 911. Having health anxiety does not mean a serious symptom is less likely to be serious.

A Practical First Step

Health anxiety is hard to describe in a first appointment, partly because saying it out loud tends to feel like confessing to being unreasonable. A score is easier to hand across than a story. Our free, confidential screening tools include the GAD-7 for anxiety and the PHQ-9 for depression — neither is specific to health anxiety, but both are useful, and depression co-occurs often enough that screening for it genuinely changes the plan.

A screening isn't a diagnosis, and nobody sees the result but you. What it does is get the conversation started at the second sentence instead of the tenth.

How We Approach This

East Coast Telepsychiatry provides psychiatric evaluation, diagnosis and medication management by video — which for this condition means working out whether what you have is illness anxiety disorder, somatic symptom disorder, OCD, panic disorder or some combination, and treating the depression or anxiety disorder that so often sits alongside it. We won't treat your symptoms as imaginary, and we won't tell you to stop seeing your physician.

One honest note on scope. The best-evidenced treatment here is exposure-based CBT, which is therapy rather than prescribing. Our board-certified and licensed providers handle the psychiatric side and coordinate with therapists who deliver that work — and given how strong the remote-delivery evidence is for this particular condition, it's worth asking any therapist you approach whether they do exposure-based work for health anxiety specifically. You can see our providers, the conditions we treat, and what care costs before booking anything.

Related Reading

Conditions We TreatAnxiety, panic, OCD, depression and more Free ScreeningsGAD-7, PHQ-9 and other validated tools Our ProvidersBoard-certified and licensed clinicians Insurance & CostWhat's covered and what care actually costs FAQsHow telepsychiatry works, start to finish More ArticlesEvidence-based mental health education

The Relief Keeps Running Out Because Reassurance Was Never the Treatment

If normal results only help for a few days, that pattern is recognised, well studied, and treatable — and the evidence says it treats well remotely.

Schedule an Evaluation Explore Support Options

Sources & Further Reading

  1. Mataix-Cols D, Isomura K, Sidorchuk A, et al. "All-Cause and Cause-Specific Mortality Among Individuals With Hypochondriasis." JAMA Psychiatry, 2024;81(3):284–291 (published online December 13, 2023). 4,129 cases, 41,290 matched comparisons, mean 7.7 years follow-up.
  2. Axelsson E, Andersson E, Ljótsson B, Björkander D, Hedman-Lagerlöf M, Hedman-Lagerlöf E. "Effect of Internet vs Face-to-Face Cognitive Behavior Therapy for Health Anxiety: A Randomized Noninferiority Clinical Trial." JAMA Psychiatry, 2020;77(9):915–924.
  3. Hedman E, et al. "Internet-delivered exposure-based cognitive–behavioural therapy and behavioural stress management for severe health anxiety: randomised controlled trial." The British Journal of Psychiatry.
  4. Illness Anxiety Disorder — DSM-5 criteria, subtypes and prevalence (PsychDB clinical reference), including the reclassification of DSM-IV hypochondriasis.
  5. PubMed Central record for the mortality study, for the full text and tables.
  6. Blakey SM, Abramowitz JS. "The effects of safety behaviors during exposure therapy for anxiety" (PDF) — the review behind the reassurance-as-safety-behavior framing.
  7. National Institute of Mental Health. Anxiety Disorders.
  8. Anxiety and Depression Association of America. Health Anxiety.
  9. International OCD Foundation. Health Anxiety.
  10. SAMHSA. National Helpline — 1-800-662-HELP (4357).

This article is for general educational purposes and does not constitute medical advice, diagnosis, or treatment, and must not be used to decide that a physical symptom does not need medical assessment. Health anxiety and physical illness frequently coexist; people with health anxiety develop medical conditions at ordinary rates, and the mortality research described here points to largely preventable causes — meaning routine medical care and screening remain important. The mortality findings are observational, describe populations rather than individuals, and carry the limitations the study authors state, including the absence of primary care data and uncertainty about classification of some deaths. Only a qualified clinician can diagnose any of the conditions described. Do not start, stop, or change any prescribed medication on your own. If you are having thoughts of harming yourself, call or text 988; for a possible medical emergency, call 911.

Helpful next steps

Share This Article

Loading East Coast Telepsychiatry...