Women and Misogyny

The womb, in this account, is an organ with appetites and grievances. When its appetites go unmet it walks. As it walks it presses on other organs and produces the symptoms the physician then catalogs.

Three things to notice. First, the symptom belongs to the womb, not to the woman. The woman is the corridor the womb walks through. Second, the cure follows from the diagnosis: marriage, pregnancy, intercourse. The remedy returns the womb to its proper occupation. Third, the entire causal chain is unverifiable. The womb does not, in fact, wander. The diagnosis was not built on observation. It was built on a presumption about what a woman's body wanted and what would happen if it did not get it.

The presumption is the artifact. Twenty-three centuries of clinical revision have not retired it; they have reupholstered it. The womb stops walking. The nerves take over. The nerves are replaced by the unconscious. The unconscious is replaced by the brain network. Each successor framework discards the anatomy of the previous one and preserves the operation: the woman's symptom is described by a clinical authority who possesses a more reliable account of her body than she does.

Jorden wrote his treatise to defend Elizabeth Jackson, a woman accused of bewitching a fourteen-year-old girl named Mary Glover. The girl had convulsions, blindness, and loss of speech. Jorden's medical argument was that the symptoms were natural in origin, caused by the suffocation of the mother. The patient was hysterical, not bewitched. Elizabeth Jackson was nevertheless convicted.

Jorden's treatise is often celebrated by historians as a step toward medicalization, an early replacement of supernatural explanation with natural explanation. It is also, considered carefully, an early example of the move this gallery documents. The patient's body remains the site of a force the patient does not control. The force has been relocated from external (witchcraft) to internal (the suffocating womb). The authority required to read the force has shifted from the cleric to the physician. The patient's own account of what is happening to her is not, in either framework, the relevant testimony.

The phrase Jorden coined, the suffocation of the mother, will travel. It survives as "the mother" in seventeenth-century English vernacular for the next century. It contributes the word hysteria, from the Greek hystera for womb, to the diagnostic vocabulary of the eighteenth, nineteenth, and most of the twentieth centuries. The phrase outlives the anatomy it was built on. The anatomy was wrong. The phrase still works because it does not, finally, refer to anatomy. It refers to a relation between clinician and patient.

The Iconographie was a catalog of women. Photographed at the Salpêtrière hospital under the direction of Jean-Martin Charcot, the women were posed, prompted, lit, and sometimes hypnotized into the stages of attack that Charcot's typology had already specified. The photographs were then offered as evidence for the typology that had produced them.

The procedure has a name now. Reactivity. The act of measurement modifies the thing being measured. In a clinical context the act of categorizing modifies the patient who has been told the category exists, who has watched other patients perform the category in the lecture hall, who has been rewarded for performing it correctly, and who is then photographed in the act of performing it. The historian Georges Didi-Huberman, whose 1982 book on the Iconographie remains the standard scholarly account, traced this loop carefully through the case of Augustine, the patient whose image appears most frequently in the volumes.

What the gallery wants the visitor to see is not Charcot's bad faith. Whether Charcot believed his own categories is a question for biography. What the gallery wants the visitor to see is the structural move. The clinical authority specifies the category. The patient is induced to inhabit the category. The category is then evidenced by the patient's inhabitation of it. The patient's own account of what is happening to her is, throughout this loop, the one thing not consulted. Augustine eventually escaped the Salpêtrière disguised as a man. She is not in the third volume.

The 1980 revision of the DSM moved hysteria, the diagnostic name, out of the manual. The 1980 revision did not move the diagnosis. It renamed it.

Conversion disorder, as DSM-III defined it, required that the patient's symptom "cannot be explained by any physical disorder or known pathophysiological mechanism." The criterion sounds like a careful exclusion. In practice it functioned as an absorption. Any symptom the clinician could not explain became eligible for the diagnosis. The patient whose multiple sclerosis had not yet been detected, the patient whose lupus presented atypically, the patient whose endometriosis had been dismissed for seven years, the patient whose autoimmune condition the lab had not yet learned to find: all became candidates. The diagnosis did not require evidence of a psychological mechanism. It required absence of a physical one. The absence was supplied by the limits of the clinician's current knowledge.

The patient population that received the diagnosis was, predictably, female by a wide margin. Subsequent reviews would put the female-to-male ratio at roughly two to one, with some studies considerably higher. The clinical literature treated this asymmetry as a feature of the disorder. The gallery's curatorial position is that the asymmetry is the artifact, not the disorder, and that conversion disorder in DSM-III is the moment when hysteria's structural move (relocate the symptom from the body to the mind) was rewritten in language modern enough to survive the rest of the century.

The 2013 revision was meant to retire the previous framework. The criterion that the symptom "cannot be explained by any physical disorder" was abandoned. The new criteria explicitly permit the diagnosis to coexist with a confirmed medical illness. The change was advertised as overcoming the mind-body dualism the older diagnosis had embedded.

What replaced the old criterion is, on inspection, the artifact this gallery exists to display. A patient now qualifies for somatic symptom disorder if she has one or more distressing somatic symptoms and if her clinician judges her thoughts about those symptoms to be "disproportionate," her anxiety "high," or her time and energy "excessive." The criteria do not specify a threshold for any of these adjectives. The threshold is the clinician's professional judgment. The diagnostic question becomes whether the clinician considers the patient's concern about her own body to be appropriately scaled.

Allen Frances, who chaired the DSM-IV task force and is therefore an unusually credible critic of the DSM-5 revisions, calculated in 2013 that the new criteria would produce a false-positive rate of roughly seven percent among generally healthy people. Frances's concern was that the criteria as written would absorb large numbers of patients whose vigilance about a serious condition was rational, not excessive. The concern has not, in the intervening twelve years, been resolved by the literature.

The gallery's curatorial position is that the operation is the same as Charcot's. The clinical authority specifies what counts as appropriate concern about a woman's body. The woman who exceeds the specification becomes the diagnosis. The patient whose endometriosis is on its fourth year of delay, the patient whose autoimmune workup keeps coming back inconclusive, the patient whose pain her doctors cannot account for: each is now eligible for a diagnosis that was designed to retire the diagnosis she would have received in 1990. The vocabulary updates. The premise does not.

A 2025 case report. A nineteen-year-old woman. Seven years of presumed epilepsy. Multiple antiseizure medications, inconclusive MRI, inconclusive EEG. The diagnosis revised in the seventh year. The report's framing notes, accurately, that functional neurological disorder is "frequently misdiagnosed" and has a "disproportionate impact on women."

The report then does something the gallery wants the visitor to notice. It traces the lineage of its own diagnostic category back to "uterine disturbances" and to the term hysteria. The lineage is named in print, in 2025, by clinicians applying the diagnostic category, as part of the introduction to a case in which the category's contemporary form was applied to a young woman for seven years of presumed epilepsy that turned out to be something other than what she was being treated for.

The gallery does not display this artifact to suggest the report's authors are wrong about functional neurological disorder. The artifact is on display because it makes the gallery's argument for the gallery. A 2025 case report in a peer-reviewed journal acknowledges that the diagnostic category being applied is the contemporary descendant of the wandering womb, that the category continues to be applied disproportionately to women, that it continues to be misapplied with consequence, and that the lineage runs unbroken across the period the gallery's other five exhibits document.

The wing's coda closes with the observation that two hundred years of medical revision have not retired this pattern. The 2025 report, by tracing its own lineage in its opening paragraph, supplies that observation in the words of the present clinical literature itself. The exhibit ends the gallery and the wing's longest historical reach lands here, in 2025, with the medical institution acknowledging in print that it has been performing a version of the same operation since the second century.

The title is a literary reference. The shepherd boy who cried wolf was, in the original fable, a liar whose lies caused his community to discount his cries even when the wolf finally came. The paper's argument is that women in pain are subject to a similar institutional discount, applied not because individual women have lied but because the institution has assumed, prior to the encounter, that women are unreliable narrators of their own bodies.

The paper assembled the evidence available in 2001. Postoperative pain studies showing women received less analgesic medication than men reporting equivalent pain. Emergency room studies showing women with abdominal pain waited longer than men for assessment. Cancer pain studies showing differential undertreatment. Speculation, in the medical literature itself, that women's pain reports were less accurate, that women complained more, that men's stoicism made their complaints more credible. Hoffmann and Tarzian noted that the speculation traveled in the literature without empirical support, and that the speculation was acting as the explanation for the disparity it described.

The paper has been cited several thousand times in the twenty-five years since publication. The disparities it documented have been replicated, refined, and re-documented across multiple jurisdictions. The institutional response has been, by the paper's own subsequent telling, modest. A 2024 review noted that the disparities Hoffmann and Tarzian described in 2001 remained measurable in the present clinical record. The gallery's curatorial position is that this paper is the wing's most cited single artifact and the most thoroughly ignored.

Endometriosis affects roughly one in ten women of reproductive age. The average diagnostic delay is seven years. The most recent French cohort puts the delay at ten. The Cyprus cohort reports a mean of seven years with an interquartile range of fifteen and a half. The United Arab Emirates cohort reports a mean delay of eleven and a half years, rising to twenty for unmarried women.

The numbers are not a single statistic but a finding reproduced across geographies, decades, and study designs. The mechanism is not contested. Patients report severe pain. Clinicians attribute the pain to anxiety, to depression, to gastrointestinal complaints, to the imagined exaggeration of menstrual symptoms, to the supposition that severe pain is the normal experience of menstruation. The disease, which can in fact be visualized on laparoscopy, goes unvisualized because the laparoscopy is not ordered. The laparoscopy is not ordered because the pain has been routed through an explanation that does not require it.

The intersectional axis intensifies the delay. The AMA Journal of Ethics article cited above notes that Black patients face the same diagnostic dismissal at higher rates. The combination of gender bias and racial bias produces diagnostic delays that, in cohort studies, are measured in additional years rather than additional months. The artifact is not the delay itself. The artifact is the institutional acceptance of the delay as a chronic feature of the diagnostic process for this condition, which it has been since the condition was named in the nineteenth century.

In 2016, four researchers at the University of Virginia surveyed 222 white medical students and residents about their beliefs concerning biological differences between Black and white patients. The survey items were drawn from the historical record of nineteenth-century racial pseudoscience: that Black people's skin is thicker, that Black people's nerves are less sensitive, that Black people's blood coagulates more quickly. The items were false. Roughly half the medical students and residents endorsed at least one.

The endorsement was not, by itself, the study's finding. The finding was that endorsement predicted lower pain ratings when the surveyed clinicians were asked to assess the pain of a Black patient versus a white patient with equivalent presentation, and predicted less accurate treatment recommendations for the Black patient. The pseudoscience traveled through the survey into the clinical judgment the students would, on graduation, apply to patients.

This exhibit sits in the women's wing because the population most affected by the compounding of these biases is Black women. The 1997 Todd study referenced widely in the subsequent literature found that Black patients in emergency rooms with isolated long-bone fractures were significantly less likely than white patients to receive analgesia, despite reporting comparable pain. The same study, broken out by sex, found Black women in particular received the least. The 2016 PNAS paper provides one mechanism. The mechanism is older than the medical students surveyed and was reproduced in the cohort being trained to practice medicine in their lifetimes.

The gallery's curatorial position is that this exhibit cannot be retired by improved diversity training, by intentional bias-reduction protocols, or by the personal goodwill of individual clinicians, important as those are. The exhibit can only be retired by an institutional response commensurate with the finding, which is that medical students in the United States are still being taught, formally or by absorption, beliefs about Black patients that originated in the literature defending slavery. The finding is a quarter of a century younger than Hoffmann and Tarzian. The institutional response has been comparable in scope.

The Calderone study has the rare quality of being structurally clean. Same operation, same hospital, same nursing staff, same prescribing protocols. The only variable that mattered was the patient's sex. Women got sedatives. Men got analgesia. Sedatives do not, mechanically, treat pain. The two patient populations had received different things in response to the same input.

The result was published in 1990 in Sex Roles, a peer-reviewed journal at the intersection of social psychology and sociology. The finding was cited in Hoffmann and Tarzian eleven years later as one of the cleanest demonstrations in the literature of differential pain management by sex. The finding has been replicated, with variations in the specific medication and the specific procedure, across thirty-five years of subsequent literature.

What the Calderone finding reveals, structurally, is the substitution at the heart of the wing's argument. The institutional response to a woman in pain has historically substituted the management of her emotional state for the management of her pain. The pain is not treated. The expression of the pain is suppressed. The clinical record then reflects a patient whose pain has been addressed, which it has not been, and whose distress has diminished, which is true only in the sense that her capacity to articulate the distress has been chemically reduced.

The structural move is the move this wing's first gallery documented at the diagnostic level. The pain is real. The clinician's response is to act on the patient's mind rather than her body. The Calderone study shows the move operating in the pharmacological log. The substitution is observable, measurable, and reproducible. It does not require the speculation about women's emotions that the literature of 1990 was still comfortable supplying.

Fibromyalgia occupies an unusual position in the contemporary medical record. The American College of Rheumatology recognizes it as a diagnosable condition. The Food and Drug Administration has approved three medications for its treatment. Most insurers reimburse for its management. A substantial portion of practicing physicians, surveyed across multiple jurisdictions in the past fifteen years, continue to express doubt that the condition is real.

The condition is real. The neurological literature on central sensitization, the imaging literature on pain-processing differences in fibromyalgia patients, and the longitudinal studies on the condition's natural history all converge on a coherent clinical entity. What persists alongside the evidence is the institutional habit of treating fibromyalgia patients as if the condition were not real. The patient is described as drug-seeking. The patient is described as somatizing. The patient is described as having a "soft" diagnosis. The patient is described, with regularity that suggests a category rather than an individual judgment, as exaggerating.

Roughly ninety percent of fibromyalgia patients are women. The condition's mistreatment is therefore a mistreatment of women patients, applied at scale, in a system that simultaneously recognizes and disbelieves the condition. The patient encounters a clinician who has been trained that the condition exists and may not personally believe that the condition exists. The patient experiences this contradiction as institutional dismissal. A 2025 ethnographic study of patient narratives, drawing on more than two thousand patient posts and comments collected between 2019 and 2024, found that the dominant theme was the patient's need to perform legitimacy persistently and across many clinical encounters before the diagnosis was taken seriously.

The artifact this exhibit displays is not a single document. It is the persistence of the gap. A condition the institution officially recognizes is unofficially dismissed by a meaningful fraction of the institution's practitioners. The patients pay the cost of the gap in delays, in mistrust, in undertreatment, and in the labor of self-advocacy that the wing's other galleries also document. The gallery's curatorial position is that fibromyalgia is, structurally, hysteria's contemporary mirror: a diagnostic category whose recognition does not protect its patients from the diagnostic suspicion the category was supposed to retire.

The wing's wall labels have, until this exhibit, displayed artifacts from the institutional record: medical journals, diagnostic manuals, peer-reviewed empirical studies, the published literature of medicine writing about itself. This exhibit displays an artifact from the patient record: a survey aggregation, reported in patient-facing health media, in which ninety-four percent of patients with chronic invisible illness reported experiencing what the patient community now calls medical gaslighting.

The term is recent. Its widespread use in patient communities began around 2018 and accelerated sharply after 2020. The term names something the patient population had been describing for decades without a shared vocabulary: the institutional pattern of being told, by clinicians, that one's symptoms are not what one experiences them to be. The patient leaves the encounter doubting her own perception of her body. The pattern is gendered, racial, and class-correlated. Women, Black patients, fat patients, disabled patients, and patients without economic resources to seek second opinions experience it disproportionately.

The institutional literature has been slow to adopt the term, though it has begun to. The medical-education literature has begun to address it. The bioethics literature has begun to address it. Whether the rate at which medical gaslighting is experienced will decline in response is a question for the next decade. The artifact this exhibit displays is the gap between the rate at which patients report the experience (ninety-four percent in the cited survey, with similar magnitudes in adjacent surveys) and the rate at which the institution acknowledges it. The gap is the artifact.

The exhibit closes the gallery on a methodological note. The wing's evidence base is the published institutional literature. The exhibit is a reminder that the institutional literature is, on the questions this wing documents, the secondary source. The primary source is the patient population whose accumulated experience the institutional literature has historically dismissed, and which has, in the past decade, begun to organize, name, and publish itself as the literature the gallery has been waiting for.

Bernadine Healy was, at the time of publication, the first woman to direct the National Institutes of Health. The editorial accompanied two empirical papers in the same issue of the New England Journal of Medicine, both documenting that women hospitalized for coronary heart disease received less aggressive care than men with comparable disease.

Healy named the pattern. She named it after a literary character whose only path to education ran through the disguise of male identity. The naming was deliberate. The argument was that women in cardiology were receiving adequate care only when their disease, their symptoms, and their bodies could be made to resemble those of men closely enough that the male-trained clinician recognized what he was looking at. The disguise was not a metaphor. The clinical literature had been written from male reference patients. The female patient who did not present in the male manner was, by definition, atypical, and therefore less likely to be diagnosed, less likely to be treated, and more likely to die from the underdiagnosis and undertreatment.

The editorial was published on July 25, 1991. Thirty-five years later, the museum is still cataloging restatements of its central finding. The exhibits that follow are not refutations of Healy's argument. They are confirmations of it, published at intervals of roughly a decade, in venues of comparable prestige, by authors who have read Healy and continue to find the disparity she described. Each restatement treats the finding as if it were a discovery. The gallery's curatorial position is that the finding has been discovered. It has not been adopted.

The word is everywhere in the cardiovascular literature. Atypical. The female presentation is called atypical not because the symptoms are rare in women but because they are different from the male presentation. The word does not describe the symptoms. The word describes the relation of the symptoms to a reference patient who is, by convention, male.

Roughly half the population presents with what the literature calls atypical symptoms when having a heart attack. The other half, the typical half, is the half who matches the diagnostic framework. The framework was built on male reference patients, on male-dominated clinical trials, and on the male presentation of coronary disease. The framework treats its own incompleteness as a property of the patients who do not match it.

A clinician trained on this framework looks for crushing substernal chest pain radiating to the left arm. The patient who presents with nausea, jaw discomfort, mid-back pain, and unexplained fatigue is, in this framework, presenting atypically. The clinician's first hypotheses follow from the framework: gastrointestinal distress, anxiety, musculoskeletal complaint, perimenopausal symptoms. The cardiac workup, when it occurs, occurs later. By the time the troponin returns elevated, hours have passed, and the time to intervention has narrowed.

The word is the artifact this gallery names most directly. To call a presentation atypical is to assert a typology in which it is the deviation. The artifacts that follow in this gallery describe the downstream consequences of the typology. The gallery's curatorial position is that the word should have been retired in 1991. It has not been retired. It is reproduced, by 2024, in every systematic review the gallery cites.

Troponin is a protein. It is released into the bloodstream when heart muscle is damaged. Elevated cardiac troponin is, by the criteria the European Society of Cardiology and the American College of Cardiology jointly published in 2000, sufficient to diagnose a heart attack. The criterion is biochemical. The criterion does not consult the patient's symptoms, the patient's demographic profile, or the patient's resemblance to the reference patient. The criterion consults the protein.

In the Hull and East Yorkshire data, the criterion was consulted inconsistently. Forty-six percent of men with elevated troponin received the diagnosis the criterion specified. Thirty-three percent of women with elevated troponin received the diagnosis the criterion specified. The remaining patients had elevated troponin, met the published criterion, and did not receive the diagnosis. The sex of the patient was, in this cohort, more predictive of receiving the diagnosis than the biomarker the criterion was built on.

This finding is structurally important to the gallery because it disposes of the explanation the cardiovascular literature has historically offered for the disparity. The explanation has been that women present with atypical symptoms that make recognition harder. The Kilpatrick data shows the disparity persisting after the biochemical recognition has already occurred. The troponin was elevated. The lab had already reported the result. The clinician then, knowing what the lab had reported, declined to enter the diagnosis the lab result supported, at a rate substantially higher for women than for men.

The atypical-symptoms argument cannot account for this. The lab result is not a symptom. The lab result was, by 2000, sufficient on its own. The gallery's curatorial position is that the Kilpatrick study quantifies the artifact the previous exhibits have described qualitatively: that the disparity in cardiovascular care for women is not, in finally, about symptoms. It is about the interpretive disposition of the clinician who receives the symptoms, the lab values, and the patient, and decides what to do with all three.

Twenty years after Healy named the Yentl syndrome, the cardiologist Noel Bairey Merz wrote her own editorial. She wrote it for the European Heart Journal. She was responding, as Healy had, to two empirical papers in the same issue, both documenting the disparity Healy had described in 1991. Bairey Merz was, at the time of writing, director of the Barbra Streisand Women's Heart Center at Cedars-Sinai. The center was named, deliberately, after the actress who had played Yentl.

The 2011 editorial documented the same pattern at finer granularity. Stable women with coronary disease were receiving fewer aspirin prescriptions and fewer ACE inhibitor prescriptions than stable men. Women with acute coronary syndrome were receiving fewer ACE inhibitors, fewer beta blockers, and fewer statins. Women with acute coronary syndrome underwent fewer angiograms, fewer revascularizations, fewer cardiac catheterizations, and fewer coronary artery bypass grafts than men. They died more often. The differential was measured in absolute numbers, percentages, and survival.

The editorial did not claim to be discovering this pattern. The editorial claimed to be documenting it again, twenty years on, with finer data and the same conclusion. The pattern had not yielded to two decades of awareness campaigns, of women's heart health months, of the named center at Cedars-Sinai, of the editorials and the empirical studies that had cited Healy thousands of times in the intervening years. The pattern had been thoroughly described. The gallery's curatorial position is that thorough description has not been the obstacle.

The editorial closed with a call for further action. The call was unspecific. The wing's lobby observed that the medical literature has been making this case against itself for two hundred years. The case is most concentrated in this gallery, where the same finding has been reported with what is functionally a metronomic cadence: 1991, 2002, 2011, 2021, 2024, 2026. The exhibits that follow continue the metronome.

In 2022, a team of cardiologists published a case report. They titled it after Healy's 1991 editorial. The case was a woman who had come to the emergency department with epigastric pain. The pain was attributed to gastrointestinal complaints. The patient was discharged. She returned. She was discharged a second time. The myocardial infarction was eventually caught. The case report describes what the authors call, in their own words, the Yentl syndrome.

The case report is unusual not in its clinical content but in its framing. The authors did not present a novel finding. They presented an instance of a pattern named thirty-one years earlier, in a journal that has by some measures the most prestige in the field, by a physician who became director of the NIH. They were publishing the case report because the pattern persists at the bedside despite thirty-one years of institutional acknowledgment that the pattern persists.

This is an unusual rhetorical position for a case report to occupy. The standard form is to describe an unusual presentation, an instructive complication, a diagnostic puzzle solved. The Cureus case report describes none of those. It describes a woman discharged twice with abdominal pain who turned out to have ischemic heart disease, and the authors, in framing the case, invoke the 1991 editorial as if it were still useful to remind their readers of it. The framing is the artifact. A case report in 2022 invoking the 1991 editorial as relevant background suggests that the editorial's argument has not landed durably enough to retire the case the report describes.

The gallery's curatorial position is that case reports of this kind, of which the published literature contains dozens, are not exceptions to the cardiovascular trajectory women patients experience. They are samples from a population whose distribution has not meaningfully shifted in thirty-five years. Each case report's authors believe they are documenting an instance of a known problem. The gallery agrees. The known problem has been known for thirty-five years.

The gallery's closing exhibit moves from the diagnostic process, which the previous five exhibits documented, to the downstream outcomes. The 2024 study in the journal Heart, the peer-reviewed organ of the British Cardiovascular Society, established three findings that the gallery's argument predicted thirty-five years in advance.

First, women in England are less likely than men to be offered heart surgery for comparable disease. Second, the same disparity applies to Black patients and low-income patients, with the disparities compounding intersectionally. Third, when women do undergo heart surgery, their one-year survival is worse than that of men who underwent the same surgery. The differential cannot be explained by disease severity, comorbidity burden, or surgical complexity at the rates the data measure. The differential survives the multivariable adjustments the study performed.

This is the cumulative downstream cost of the trajectory the gallery has documented. The atypical-symptoms framing produces the initial dismissal. The discount on women's reports produces the diagnostic delay. The diagnostic delay produces the more advanced disease at presentation. The more advanced disease, combined with the persisting institutional pattern that offers women less aggressive intervention, produces the survival differential the 2024 Heart paper measures. Each step is observable. Each step is independently documented. The combined trajectory is the institutional response to women patients with heart disease in the year the wing opens.

The British Heart Foundation has, since 2019, published a series of reports under the title Bias and Biology, examining the gender, racial, and class disparities in cardiac care across the four nations of the United Kingdom. The reports document essentially the pattern Healy described in 1991 and Bairey Merz redocumented in 2011. The vocabulary has been updated. The premise has not. The gallery's curatorial position is that the wing's lobby statement, that the pattern is not yet finished, is most concretely evidenced here, in the outcome the 2024 paper measured in patients who were alive at the time the wing opened.

The Wisconsin Sleep Cohort Study is the founding population-based dataset for sleep-disordered breathing. The 1993 paper has been cited more than ten thousand times. The numbers it reported, 24% of men and 9% of women with an AHI of five or higher, became the prevalence figures the field built itself on. For the next decade, sleep apnea was understood, taught, and screened for as a condition that affected men at nearly three times the rate it affected women.

The Spanish cohort, designed differently, reported prevalence of 28% in women and 26.2% in men in roughly the same period. The Spanish numbers were closer to even. The difference between the cohorts was not biological. The difference was methodological: how arousals were scored, how hypopneas were counted, how the population was sampled. The Wisconsin numbers were nine percent because of how the Wisconsin study was designed. The literature that followed treated the Wisconsin numbers as the prevalence and the Spanish numbers as anomalous.

The downstream cost was not a measurement error. It was a clinical doctrine. A field that believed sleep apnea affected women at one third the rate it affected men screened for sleep apnea in women at one third the rate it screened for it in men. The undiagnosed cases accumulated. The Wisconsin investigators themselves, in subsequent work, estimated that up to 80% of moderate and severe OSA cases in their cohort remained undiagnosed by their physicians despite adequate access to care. The undiagnosed-in-women rate was higher than the undiagnosed-in-men rate.

The gallery's curatorial position is that the 1993 paper is one of the wing's clearest examples of a methodological choice becoming a clinical truth. The 9% figure is in the foundational sentence of every sleep apnea review article for the next thirty years. The methodological caveats appear, when they appear, in the limitations section.

Continuous positive airway pressure therapy was introduced in 1981 by Colin Sullivan in Sydney. The mask sealed against the patient's face and delivered pressurized air through the night. The first masks were designed against the patient population the clinical literature had identified as having sleep apnea, which was, by the field's own count at the time, predominantly male. The masks were engineered for male facial anatomy.

The female-specific mask category, marketed with names like AirFit For Her and GoLife for Women, did not arrive at scale until the 2010s. Roughly thirty years passed between the introduction of CPAP therapy and the engineering of the hardware that would fit women's faces. During those thirty years, women diagnosed with OSA were prescribed masks designed for facial structures that did not match theirs. The mask leaked. The seal failed. The therapy was ineffective. The patient reported non-adherence. The clinical record interpreted the non-adherence as the patient's failure rather than the hardware's.

The non-adherence literature on women with OSA is, in this light, an unusual document. It documents that women diagnosed with OSA discontinue CPAP at higher rates than men. It cites patient factors: comfort, claustrophobia, body image concerns, partner intimacy. It rarely centers the engineering question. The masks did not fit. The patient discontinued because the equipment she was prescribed was the wrong size and shape for her face. The clinical literature framed her discontinuation as a behavioral and psychological matter.

The arrival of the For Her mask category is, structurally, a manufacturer's tacit acknowledgment of the thirty-year design failure. The acknowledgment is delivered as a product line rather than as a paper. The gallery's curatorial position is that the engineering artifact and the non-adherence literature are the same artifact viewed from different angles. The institution built equipment that did not fit half its diagnosed patients and explained the resulting non-use as a behavioral characteristic of the patients.

Narcolepsy is a neurological disorder. It is, in its type 1 form, an autoimmune condition that destroys the hypothalamic neurons producing orexin. It is not a mood disorder. It is not a personality trait. It is not a stress response. The diagnostic criteria are biochemical and electrophysiological: low or absent orexin in cerebrospinal fluid, characteristic findings on the multiple sleep latency test. The diagnosis is, in principle, available.

In practice, the diagnosis is reached after an average delay of ten to fifteen years from symptom onset. The delay is documented in cohort studies across countries. In some cases the delay exceeds twenty years. The delay is not because the symptoms are subtle. The symptoms include falling asleep during conversations, falling asleep while driving, sudden muscle weakness in response to emotion, and visual or auditory hallucinations at sleep onset. The symptoms are dramatic. They are dramatic enough to come to medical attention. They come to medical attention and are misidentified.

The misidentification follows a recognizable pattern. More than half of patients with narcolepsy are diagnosed with depression before they are diagnosed with narcolepsy. A 2022 Frontiers in Psychiatry case report described three patients misdiagnosed as depression, dissociative disorder, and schizophrenia respectively, each receiving years of psychiatric treatment for narcolepsy. The case report's authors framed the misdiagnoses as a problem of clinical knowledge: many psychiatrists are not trained to recognize narcolepsy and many sleep specialists are not the first physicians these patients encounter.

This exhibit sits in the women's wing because women are roughly twice as likely as men to receive a diagnosis of depression in the general clinical encounter. A woman with undiagnosed narcolepsy entering primary care with excessive daytime sleepiness, low energy, and concentration difficulties is, statistically, headed toward a depression diagnosis before she is headed toward a polysomnograph. Female hormones complicate the routing further: clinicians attribute her fatigue to menstrual irregularities, to postpartum recovery, to perimenopause, to menopause. Each attribution defers the sleep workup the symptoms warranted from the onset. The cumulative effect is the diagnostic delay the literature now treats as a fact about narcolepsy. The delay is not a fact about narcolepsy. It is a fact about who is allowed to be tired and what their tiredness is allowed to mean.

A woman presents to primary care with a sleep complaint. The complaint is, depending on her age and circumstances, attributed to anxiety, to depression, to her menstrual cycle, to her children, to her recent loss, to perimenopause, to menopause, to stress. The clinical encounter takes twelve minutes. The clinician has the prescription pad. The pad contains, among other options, alprazolam, lorazepam, temazepam, zolpidem. The prescription is written. The woman leaves with a sleeping pill.

The same woman, presenting with the same complaint at the same severity, has substantially lower odds of leaving with a referral for a sleep study. The referral takes longer to write than the prescription does. The referral commits the system to time it does not have. The referral does not generate the same documentation of clinical action that the prescription does. The prescription pad wins, statistically, in the encounters the literature has measured.

The mechanism is not malice. The mechanism is what the literature on time-pressured clinical decision-making calls a satisficing response: the clinician selects an action that adequately addresses the presenting complaint within the time available, and moves on. The action is suppression of the symptom rather than investigation of the cause. The benzodiazepine reliably reduces the patient's reports of poor sleep. The reduction is not because the underlying disorder has been addressed. It is because the patient is sedated.

The downstream cost is the same as the wing's other galleries document: the underlying condition, which may be obstructive sleep apnea, restless legs syndrome, periodic limb movement disorder, circadian rhythm disorder, or any of several diagnosable neurological conditions, continues unmeasured and untreated. The patient becomes dependent on the medication. The dependence becomes its own clinical problem. The cycle is structurally indistinguishable from the substitution pattern Gallery II documented in the Calderone postoperative data, where women received sedatives where men received analgesia. The gallery's curatorial position is that the substitution is the artifact, in 1990 and in 2026 alike. The pharmacological options have changed. The substitution has not.

A 2022 case report describes three patients. The first was diagnosed with depression. The second with dissociative disorder. The third with schizophrenia. Each received psychiatric treatment for years before the actual diagnosis, narcolepsy, was reached. Each had received accurate symptom reports from the patient that, in retrospect, met the diagnostic criteria for narcolepsy from the first clinical encounter.

The case report's authors describe the misdiagnoses as a problem of clinical knowledge. Many psychiatrists are not trained to recognize narcolepsy. The hypnagogic hallucinations that occur at sleep onset in narcolepsy can resemble psychotic symptoms. The cataplexy that occurs in response to emotion can resemble dissociation. The fatigue and concentration difficulties can resemble depression. The clinical literature on narcolepsy uses these resemblances as cautionary notes. The gallery uses them differently.

The resemblance is the point. Narcolepsy resembles the conditions the clinical encounter is structured to detect: psychiatric conditions presenting in primary care or psychiatric care. The conditions it does not resemble, neurological sleep disorders requiring polysomnography, are the conditions it actually is. The clinical encounter is structured around the resemblances rather than the underlying mechanism. A patient who comes in describing the underlying mechanism, that she falls asleep involuntarily during the day, that her body collapses when she laughs, that she sees things at sleep onset, is heard as describing the psychiatric resemblance. The narcological diagnosis would require the clinician to set aside the resemblance and order the test.

The gallery's curatorial position is that the three patients in the 2022 case report are not exceptions. They are samples from a population whose distribution the wing's curator essay described: women with neurological sleep disorders routed through psychiatric explanations before the neurological workup. The case report frames this as a clinical-knowledge problem to be solved by training. The gallery frames it as a structural pattern in which women's symptoms are interpreted through their emotional resonance rather than their physiological mechanism, regardless of the training the clinician has received.

The wing's previous exhibits have drawn from the institutional record: published research, diagnostic criteria, engineering history, prescribing-pattern literature. This closing exhibit draws from the patient-facing record being published in 2026 by sleep medicine educators trying to help women patients recognize what the clinical literature has historically refused to recognize for them.

The patient-facing literature now openly states what the gallery's earlier exhibits had to be argued for. Women describe sleep symptoms differently than men. The descriptions are not less accurate. They are, in many ways, more accurate, because they include the lived consequences the clinical questionnaire does not ask about: the children who are not being kept up with, the work that is foggy, the exhaustion that is constant. The clinical encounter, structured around questions designed for the male presentation ("Do you fall asleep at the wheel?" "Has anyone told you that you stop breathing in your sleep?"), filters out the descriptions women give. The descriptions are routed elsewhere: to the obstetrician, to the psychiatrist, to the well-meaning friend who suggests yoga.

This routing is now visible to the patient-facing educators writing in 2026 because the patient communities have been writing it down for thirty years. The accumulated patient testimony has produced a body of writing that the sleep medicine field is now beginning to draw on. The Ubie article, the Healthline expert columns, the Mayo Clinic Press piece on narcolepsy and depression: these are not patient testimony themselves. They are clinical educators incorporating what the patient population has been saying. The incorporation is recent. The saying is not.

The gallery closes here, in 2026, with the institutional literature beginning to use vocabulary patient communities developed in the 1990s and 2000s to describe what was happening to them. The vocabulary describes a routing the institution has performed and continues to perform. The recognition of the routing is the museum's structural answer, which Wing V documents at greater length. The Sleep Gallery's contribution to that answer is to show what the routing looks like in one quieter clinical domain, where the consequences are measured not in deaths in the emergency department, as in Gallery III, but in years of exhaustion, missed careers, missed parenting, and the slow erosion of a life by a treatable condition the institution declined to detect because it was looking for a different patient.

In September 2023, the Journal of Medical Ethics published a symposium of papers on consent during labor. The symposium's existence is the artifact this exhibit displays. A serious medical-ethics journal devoting an issue to whether and how consent should be obtained from women in labor is evidence that, after a century of consent doctrine in every other branch of medicine, the question remained sufficiently unsettled in obstetrics to require its own scholarly intervention.

The symposium documented what the patient communities had documented for decades: that unconsented episiotomies, unconsented vaginal exams, unconsented cervical sweeps, and unconsented artificial rupture of membranes occur regularly across many jurisdictions. The 2024 Birth Trauma Association survey found that 42.4 percent of women in the United Kingdom reported procedures performed without consent during labor, with vaginal examination (17.9 percent) and episiotomy (16.6 percent) the most frequent. The Swedish Oneplus trial measured the same pattern in Sweden.

The medical-ethics literature treats this as a problem requiring careful philosophical analysis: whether labor is a setting in which standard consent doctrine applies, whether implied or opt-out consent can substitute for informed consent, whether the patient's capacity to consent during contractions is sufficient. The gallery's curatorial position is that the symposium's question is the artifact. The same questions are not asked, with similar gravity, about consent for a colonoscopy, for a dental procedure, for a hip replacement. The patient's pre-procedure consent for those interventions is taken as a baseline professional standard. The patient's consent for procedures performed inside her vagina during labor is, in 2023, a topic the medical-ethics literature was still working out.

The wing's curatorial position is that this is the operation the wing has been describing. The patient is presumed to have ceded interpretive authority over what happens to her body, by the act of going into labor in a hospital. The interpretive authority belongs, by default, to the clinician. The clinician's professional judgment supersedes the patient's pre-procedure refusal. The 2023 symposium exists because the patient communities, and a small number of clinicians and ethicists, have refused to accept the default. The default has not, in many jurisdictions, been retired.

On April 1, 2024, the Department of Health and Human Services issued a memorandum requiring hospitals to obtain explicit informed consent for pelvic, rectal, and other sensitive examinations, including examinations performed on anesthetized or unconscious patients. The memorandum's existence is the artifact this exhibit displays. The federal government, in 2024, found it necessary to clarify that pelvic exams performed on unconscious patients require the patient's prior informed consent.

The practice the memorandum addressed was the longstanding teaching custom in many United States hospitals of medical students and residents performing pelvic exams on anesthetized patients during unrelated surgeries, without the patient's specific knowledge or consent. The practice was not a misunderstanding or an isolated abuse. It was the documented teaching method in obstetrics and gynecology residencies across the country for decades. The patients were anesthetized for other procedures, often hysterectomies or other gynecological surgeries. Their unconscious bodies were used as training material for trainees who were not part of the surgical team.

State legislative action began in 2003. By 2024, twenty-one states had passed statutes prohibiting the practice. The federal memorandum followed the state movement. The Annals of Surgery Open article that triggered the federal action documented the persistence of the practice in states without specific statutes, and the legal ambiguity in states that had passed statutes but not yet enforced them. The federal memorandum's existence is the institutional acknowledgment, in 2024, that the consent doctrine the United States had operated under for a century had not been applied to women's pelvises in teaching hospitals.

The gallery's curatorial position is that this exhibit and Exhibit 5.1 are the same exhibit in two registers. The clinical-ethics literature was working out, in 2023, whether consent during labor was a settled question. The federal government was clarifying, in 2024, that consent for pelvic exams on unconscious patients was also a settled question. Both exhibits document the institutional position that the patient's authority over what happens inside her body was, until very recently, not the default presumption in obstetric and gynecological practice in many parts of the United States.

The CDC released its 2024 maternal mortality data in March 2026. Overall U.S. maternal mortality declined slightly from 18.6 to 17.9 deaths per 100,000 live births, a change the CDC noted was not statistically significant. The decline was concentrated in white, Hispanic, and Asian American populations. The Black maternal mortality rate, 44.8 per 100,000 live births, did not decline.

The numbers, taken in isolation, are difficult to interpret. The numbers, set against the rates in other high-income countries, are not difficult to interpret. The OECD average for high-income countries is approximately 10 deaths per 100,000 live births. The United States rate is nearly twice that. The Black maternal mortality rate is more than four times the OECD average. The American Indian and Alaska Native rate, reported by the March of Dimes for 2019-2023 at 60.8 per 100,000, is six times the OECD average.

The leading causes of Black maternal death in the United States are concentrated in conditions that respond to timely medical intervention: preeclampsia, eclampsia, postpartum hemorrhage, postpartum cardiomyopathy. Black women are approximately 60 percent more likely to develop preeclampsia than white women. Their risk of dying from it is approximately five times higher. The 2023 American Journal of Public Health study found that Black women with preeclampsia faced nearly three times the in-hospital death risk even after adjustment for clinical variables. Postpartum cardiomyopathy kills Black women at five times the rate of white women.

These are not biological inevitabilities. These are downstream cardiovascular and obstetric events occurring in a medical system that, when the patient is Black, is documented across the previous four galleries of this wing as less likely to recognize her pain, less likely to recognize her cardiac presentation, less likely to recognize her sleep symptoms, less likely to administer the diagnosis the lab value supported. The 44.8 figure is what the previous four galleries produce when integrated across the trajectory of a pregnancy. The figure is the cumulative outcome the wing's other artifacts predict. The gallery's curatorial position is that the figure is the wing's most consequential single number, and it is being measured in patients alive at the time the wing opens.

In 2023 and 2024, the Parliament of New South Wales conducted an inquiry into birth trauma. The inquiry received more than four thousand submissions, the largest response in the parliament's history for any inquiry. The testimony, taken together, documented a pattern the wing's lobby statement named: the dismissal of birth trauma as ingratitude.

The pattern works as follows. A woman gives birth. The birth involves procedures she did not consent to, or interventions whose necessity she questioned, or treatment by clinicians whose behavior she experienced as coercive. She tells someone afterward. The someone, often the same clinician but sometimes a nurse, a midwife, a partner, a family member, or a friend, responds with some version of the formula: you have a healthy baby. The formula functions as a closing argument. The woman's experience of the birth is treated as a private grievance about a delivery that produced the outcome the institution had been hired to produce.

The formula has weight in part because it is true. A healthy baby is a serious good. The formula's weight is also the reason it works as a silencing device. To raise concerns about how the birth was conducted, after the formula has been deployed, is to be heard as ungrateful for the outcome. The patient communities have, for decades, developed vocabulary to describe what the formula performs. Obstetric violence. Birth trauma. Procedural consent failure. The vocabulary has, slowly, entered the medical-ethics literature, the parliamentary inquiry record, and a small fraction of the obstetric training curricula.

The artifact in this exhibit, a single submission from the NSW inquiry, is representative of thousands. The submitter's statement is precise. Stating you are going to do something is not consent. The precision is not a coincidence. The patient communities have spent thirty years developing the precision because the clinical literature did not supply it for them. The wing's coda will note that the museum's structural answer to the pattern this gallery documents is in Wing V. This exhibit gestures toward that answer by displaying a patient who, after four births, four cases of unconsented procedures, and a parliamentary inquiry's solicitation, articulated the operation the institution had performed on her in language the institution was finally listening to.

In June 2022, the U.S. Supreme Court ruled in Dobbs v Jackson Women's Health Organization, overturning the federal constitutional protection of abortion. Eighteen states implemented abortion bans between 2022 and 2024. The bans were drafted by legislators who, the laws' supporters argued, were balancing the interests of the pregnant patient and the fetus. The wing displays one document from the subsequent clinical literature.

The 2024 study in AJOG Maternal-Fetal Medicine interviewed thirty-six health care professionals across multiple states between November 2022 and October 2024. The interviewees described categories of patient scenarios in which pregnant patients did not receive the standard of care due to post-Dobbs laws. The categories were not abortion in the political sense the bans had addressed. They were management of ectopic pregnancy, management of preeclampsia approaching viability, management of intrauterine fetal demise, management of preterm premature rupture of membranes, management of cancer diagnoses during pregnancy. In each category, the clinician's professional judgment about what would prevent future health harms to the pregnant patient was, by the operation of the law, constrained.

The study's authors named this an "exclusively reactive practice." The clinician was permitted to intervene once the patient was already deteriorating, hemorrhaging, septic, or in active organ failure. The clinician was not permitted to intervene to prevent the deterioration. The legal apparatus had inserted itself between the clinician's professional judgment and the patient, with the effect that the patient's care was structured around when she could legally be helped rather than when she clinically needed help.

The wing's curatorial position is that this exhibit documents the legal completion of the operation the previous exhibits documented in the medical practice itself. The previous galleries showed clinicians dismissing women's pain, women's hearts, women's sleep, women's symptoms. This exhibit shows the state, in 2022-2024, building a legal architecture that constrains clinicians who do recognize the patient's condition. The clinician who understands what is happening is, under the law, prevented from acting on her understanding. The interpretive authority over the pregnant patient's body that this wing has been tracking across two thousand years now resides, in eighteen states, partly in the legislature rather than in the patient and the clinician together. The lobby's statement that the pattern is not yet finished applies most concretely here.

The wing's closing exhibit is not a new artifact. It is the titles of the 2023 medical-ethics symposium read as a sequence. Read together, the titles describe the operation this wing has been documenting in language that, in 2023, the institutional literature was finally adopting from the patient communities that had been using it for thirty years.

The first paper title names the kind of injustice that occurs when a patient is denied interpretive authority over her own body by a clinical system that holds the credentials to override her account. Trauma-informed remedies. The recognition that the system the patient encounters at the most vulnerable moment of her life is, often, traumatizing. Thinking beyond the existing consent framework. The recognition that the framework medicine has used was not built for, and does not fit, the obstetric encounter. No place for opt-out consent. The recognition that the institutional practice of taking the patient's silence as agreement is not consent and never was.

Each of these phrases was developed, in some form, by patient activists, by feminist scholars, by survivors of obstetric procedures, in the decades before 2023. The phrases arrived in the medical-ethics literature in 2023 because the medical-ethics literature was, finally, listening. The arrival is the closing artifact of this wing. The institutional literature is now naming, in its own terminology, the operation the institutional literature was performing.

The naming is not the same as the retiring. The 44.8 figure in Exhibit 5.3 was measured in 2024, after the medical-ethics symposium and after the HHS memorandum and after the NSW parliamentary inquiry. The legislative apparatus described in Exhibit 5.5 was being constructed in the same year the consent symposium was published. The naming and the operation coexist. The naming has not, by itself, retired the operation. The retiring is the work that remains. The wing's coda turns to that work directly. The exhibits in this gallery, taken together with the exhibits in the previous four, supply the case that has been made against the institution by the institution itself for two hundred years, finally arriving in the institutional literature in the institution's own vocabulary, two hundred years late.