When Symptoms Get Called Psychological

Why We Care · 13 October 2026

A 2026 survey of 6,607 people with autoimmune diseases found women more often reported being told their illness was psychological. The idea that the mind makes the symptom runs back through Freud to Charcot's ward, and follow-up studies show how often such labels held.

Key takeaways

In Melanie Sloan's 2026 survey of 6,607 people with autoimmune diseases, 43 per cent of women and 33 per cent of men reported that their illness was first attributed to a psychiatric cause.
After adjusting for age, disease group, continent and household finances, women in that survey had about 1.7 times the odds of men of reporting a psychosomatic misdiagnosis.
The Modern Mind records that Anna O., declared totally cured by the talking cure in 1882, was admitted to a sanatorium within a month.
Jon Stone's 2005 review of 27 studies found misdiagnosis of unexplained motor and sensory symptoms fell from 29 per cent in the 1950s to 4 per cent from the 1970s.
Sex gaps in diagnosis vary by condition: a 2015 UK study of 18,618 cancer patients found women waited longer in six cancers, with a pooled difference of about nine days.

This article is general information about research on how physical symptoms come to be attributed to the mind, and on delays in diagnosing autoimmune disease. It is not medical advice and does not replace a consultation with a qualified clinician. Medical review:.

A survey of 6,607 people with autoimmune diseases, published in eClinicalMedicine in 2026, found that women were more likely than men to report that their illness had first been put down to a psychiatric or psychosomatic cause. Among the women, 43 per cent reported a psychiatric misdiagnosis, against 33 per cent of the men. For psychosomatic explanations the figures were 34 and 22 per cent. Women's answers also clustered at longer waits: the single most common interval from first symptoms to diagnosis was under three months for men and one to two years for women, out of several time bands. The study was led by Melanie Sloan at the University of Cambridge. The pattern it describes has a long history.

Charcot and the talking cure

Peter Watson's The Modern Mind opens its account of the twentieth century with Freud, and on p. 13 it gives the back story. After qualifying, Freud won a scholarship to study in Paris under Jean-Martin Charcot, who ran an asylum for women with incurable nervous disorders. Watson credits Charcot with showing that hysterical symptoms could be induced under hypnosis, and reports it without qualification. The inference was tempting. If suggestion could produce a symptom, perhaps the mind could produce one unaided.

The same page tells the case that founded the method. From December 1880 the Viennese physician Josef Breuer treated Bertha Pappenheim, known to readers as "Anna O.". Watson lists sleepwalking, paralysis, a split personality and a phantom pregnancy. Breuer found that letting her talk at length about her symptoms made them disappear, and she herself named this the "talking cure". By June 1882 she ended treatment "totally cured". Watson adds at once, in parenthesis, that she is now known to have been admitted to a sanatorium within a month.

Freud moved from hypnosis to free association, and from there, in Watson's telling on pp. 13 to 14, to the unconscious and repression. Richard von Krafft-Ebing said Freud's account of hysteria "sounds like a scientific fairy tale". Watson's own complaint, on p. 27, is that Freud saw himself as a scientist yet offered no method by which a sceptic could be shown the unconscious existed. One detail on pp. 106 to 107 complicates the caricature. In 1905 Freud reported finding paternal syphilis in more than half of his severe cases of hysteria. The man who made the mind an explanation still went looking for the body.

The man who made the mind an explanation still went looking for the body.

Following up hysteria diagnoses

The label was never reserved for women. In chapter 9 Watson argues that the sheer volume of shell-shock cases among soldiers in the First World War did more than argument to change British medical opinion of Freud, and he credits W. H. R. Rivers with early studies of the condition (p. 155). What was rarely done in that era was to check systematically, years later, whether the patients labelled hysterical had in fact been ill with something else.

That check came in the second half of the century. In a 1965 paper in the BMJ, E. Slater reported a follow-up of patients diagnosed with hysteria between 1951 and 1955. A 2005 systematic review in the same journal, led by the Edinburgh neurologist Jon Stone, records that 32 of the 97 patients Slater traced had been misdiagnosed. A third of the diagnoses failing on follow-up cast doubt on the label itself.

Stone's review then gathered 27 studies published since 1965, covering 1,466 adults with motor or sensory symptoms that no disease explained, followed for a median of five years. The misdiagnosis rate fell from 29 per cent for patients diagnosed in the 1950s to 17 per cent in the 1960s and 4 per cent from the 1970s onwards. The authors judged that the fall probably reflected better study design, since the early studies were of poor quality, rather than the arrival of brain scanning. The decline held regardless of patients' age and sex. In the studies since 1970, a diagnosis of this kind has usually held.

Who waits longer for diagnosis

Wider data on sex and diagnosis point in more than one direction. David Westergaard and colleagues at the University of Copenhagen analysed all public and private hospital admissions in Denmark from 1994 to 2015, covering about 6.9 million patients, in a 2019 paper in Nature Communications. Where age at first hospital diagnosis differed between the sexes, women were usually older. The exceptions were cancers, diseases of the blood and immune system, and diseases of the genitourinary system. The authors also caution that age at first hospital diagnosis should not be read as age at onset, so their data describe when people reach a hospital code rather than how long their symptoms went unexplained.

Cancer data give a smaller and patchier gap. A 2015 analysis of 18,618 patients in UK primary care records, led by Nafees Din, reported longer intervals from first symptoms to diagnosis for women in some of the cancers that affect both sexes. Pooled across the cancers studied, the difference was about nine days.

Some conditions affecting only women show far longer waits. A 2012 study of 171 women with confirmed endometriosis in Austria and Germany reported a median of 10.4 years from first symptoms to diagnosis, and 74 per cent had received at least one wrong diagnosis on the way. Misdiagnosis and patients' own tendency to treat painful periods as normal both lengthened the delay.

Reading the autoimmune survey

Sloan's team ran an online survey between December 2024 and March 2025, open to adults whose autoimmune or inflammatory rheumatic disease was confirmed in clinical correspondence. The sample was 91 per cent female, 70 per cent based in the UK and 88 per cent White. After adjusting for age, disease group, continent and household finances, women had about one and a half times the odds of men of reporting a psychiatric misdiagnosis, and about 1.7 times the odds for a psychosomatic one. Men were slightly more likely to report having been diagnosed with a different physical disease, 48 against 44 per cent, a difference that did not reach statistical significance.

Disease type also mattered. In the survey, some conditions, including type 1 diabetes and rheumatoid arthritis, had lower rates of psychological misattribution. The authors list their own limits plainly. Every misdiagnosis was perceived and reported by the patient, and some may have been a doctor's first working guess rather than a firm verdict. The survey may have drawn people with worse experiences, and no response rate could be calculated. Women in the sample also reported more anxiety diagnoses, 24 against 16 per cent, which might lead a clinician to read later symptoms through a psychological lens. The authors call the work exploratory.

The two bodies of evidence sit side by side without contradiction. Stone's review concerns patients given a deliberate diagnosis of unexplained motor or sensory symptoms and then followed for years, and it finds that the diagnosis was rarely overturned. Sloan's survey concerns patients recalling what they were told while their disease was still taking shape.

The two bodies of evidence sit side by side without contradiction.

Our reading is that the sex gap is probably real in direction but that its size cannot be read straight off a self-selected sample, and the survey cannot say whether bias, prior anxiety diagnoses or the way diseases present is doing the work. The authors' own conclusion, that psychological explanations can arrive too early, sits closer to the data than any claim about doctors disbelieving women. The history in The Modern Mind shows how attractive an early, tidy explanation could be, even to expert judgement.

Preparing for an appointment

None of the studies above tested what patients should do, so what follows is general information rather than evidence-based guidance. It draws loosely on two points in the survey paper: some conditions were less often misattributed than others, and the authors urge that diagnostic decisions be revisited as new information emerges.

1. A symptom diary. Dates, duration, what made things better or worse, and how the pattern has changed. Photographs of rashes, swelling or colour changes that come and go turn a passing sign into one a clinician can see.

2. A short list of relevant history and earlier explanations you have been given.

3. A plain question: what else could this be, and what would make you think again?

4. A request for what to watch for, and when to come back.

5. A note that second opinions exist if problems persist.

A psychological explanation can be correct, and an anxiety diagnosis can sit alongside a physical disease. The survey's authors suggest that early labels deserve a second look as new information emerges. A clear record of what happened, and when, gives any clinician a better chance of seeing the whole illness rather than its first chapter.

Common questions

Are women more likely to have symptoms dismissed as psychological?

In Melanie Sloan's 2026 survey of 6,607 people with autoimmune diseases, 43 per cent of women and 33 per cent of men reported a psychiatric misdiagnosis before their autoimmune diagnosis. After adjustment, women had about one and a half times the odds. The data are self-reported, the sample was mostly female and British, and the authors describe the findings as exploratory.

How often does a diagnosis of hysteria turn out to be wrong?

Jon Stone's 2005 BMJ review of 27 studies and 1,466 patients found misdiagnosis rates of 29 per cent for diagnoses made in the 1950s, falling to 4 per cent from the 1970s onwards. The authors attributed most of the fall to better-designed studies rather than to brain scanning.

Do women wait longer for a diagnosis?

Not uniformly, and the evidence is indirect. Westergaard's 2019 analysis of about 6.9 million Danish patients found women usually received a first hospital diagnosis at an older age, with exceptions including immune system diseases. A 2015 UK study of 18,618 cancer patients found longer diagnostic intervals for women in some cancers, with a pooled difference of about nine days.

Sources

Sloan, M., Taylor, S., Cox, B., Naidu, K. K., Modi, R., Diment, W., Tayabali, S., Piper, M., Brimicombe, J., Kaul, A., & D’Cruz, D. (2026). 'Sex and disease differences in patients' diagnostic experiences in autoimmune diseases: an international cross-sectional study.' eClinicalMedicine, 104259. doi:10.1016/j.eclinm.2026.104259. Full text read: online survey 7 December 2024 to 7 March 2025; 6,607 adults with clinically confirmed autoimmune or inflammatory rheumatic disease, 91.2 per cent female, 69.7 per cent UK, 87.8 per cent White; psychiatric misattribution 43.3 vs 32.7 per cent, psychosomatic 33.9 vs 21.5 per cent (both p < 0.001); other physical disease 47.6 vs 43.5 per cent (p = 0.131); adjusted OR 1.46 (95% CI 1.14 to 1.87) psychiatric and 1.74 (1.31 to 2.32) psychosomatic; modal time to diagnosis under 3 months (males, 21.2 per cent) and 1 to 2 years (females, 15.7 per cent); anxiety co-diagnosis 24.2 vs 16.1 per cent; authors describe the work as exploratory and list self-report, selection and recall limitations.

Stone, J., Smyth, R., Carson, A., Lewis, S., Prescott, R., Warlow, C., & Sharpe, M. (2005). “Systematic review of misdiagnosis of conversion symptoms and ‘hysteria’.” BMJ, 331(7523), 989. doi:10.1136/bmj.38628.466898.55. Abstract read, and the full text consulted for the table entry on Slater (1965): 27 studies, 1,466 patients, median follow-up five years; misdiagnosis 29 per cent (1950s), 17 per cent (1960s), 4 per cent (1970s, 1980s, 1990s); decline independent of age, sex and symptom duration and probably due to improved study quality rather than CT scanning. Slater's diagnoses made 1951 to 1955, 32 of 97 followed-up patients misdiagnosed.

Slater, E. (1965). “Diagnosis of ‘hysteria’.” British Medical Journal, 1(5447), 1395-1399. doi:10.1136/bmj.1.5447.1395. Not read directly (no abstract; PMC scan blocked); figures used only as reported in the full text of Stone et al. (2005).

Westergaard, D., Moseley, P., Sørup, F. K. H., Baldi, P., & Brunak, S. (2019). “Population-wide analysis of differences in disease progression patterns in men and women.” Nature Communications, 10, 666. doi:10.1038/s41467-019-08475-9. Full text read: Danish National Patient Registry, all public and private hospital admissions 1994 to 2015, 6,909,676 patients; 986 diagnoses with a sex difference in age at first hospital diagnosis, women older in the majority; exceptions neoplasms, blood and immune system diseases, genitourinary diseases; authors note age at first hospital diagnosis should not be confused with age of onset.

Din, N. U., Ukoumunne, O. C., Rubin, G., Hamilton, W., Carter, B., Stapley, S., & Neal, R. D. (2015). “Age and gender variations in cancer diagnostic intervals in 15 cancers: analysis of data from the UK Clinical Practice Research Datalink.” PLoS ONE, 10(5), e0127717. doi:10.1371/journal.pone.0127717. Abstract read: 18,618 patients aged 40 and over; longer diagnostic interval for females in six gender non-specific cancers; pooled difference for females 8.9 days (95% CI 5.5 to 12.2).

Hudelist, G., Fritzer, N., Thomas, A., Niehues, C., Oppelt, P., Haas, D., Tammaa, A., & Salzer, H. (2012). “Diagnostic delay for endometriosis in Austria and Germany: causes and possible consequences.” Human Reproduction, 27(12), 3412-3416. doi:10.1093/humrep/des316. Abstract read: 171 patients with histologically confirmed endometriosis; median interval from symptom onset to diagnosis 10.4 years; 74 per cent received at least one false diagnosis; misdiagnosis and normalisation of dysmenorrhoea prolonged the delay.

Watson, P. (2001). The Modern Mind: An Intellectual History of the 20th Century. HarperCollins (Perennial edition 2002). Read in full by this publication, with a reliability audit. Ch. 1, pp. 13 to 14 (Charcot, Breuer, Anna O., free association, Krafft-Ebing); p. 27 (Watson's criticism of Freud's method); pp. 106 to 107 (Freud's 1905 syphilis observation); ch. 9, p. 155 (shell shock, W. H. Rivers). The audit notes Charcot's induced symptoms are given without qualification.