Heart attack misdiagnosis in women: how often the first call is wrong

She was 48, a nonsmoker, with a cholesterol panel her internist had called "fine for her age." At 3 a.m. she woke with pressure under her sternum and nausea, plus an exhaustion she described as "flu without a fever." Urgent care labeled it anxiety and reflux and sent her home. Twelve hours later she was in a catheterization lab with a mid-left anterior descending occlusion. Nobody had ordered a troponin.

I see versions of that story more often than textbooks prepare you for. The first clinical impression when a woman presents with ischemia is still too often anxiety, gastritis, or muscle strain. That first wrong call costs time. Time is myocardium.

According to the CDC, heart disease killed 304,970 U.S. women in 2023 (about 1 in every 5 female deaths), and more than 60 million women, or 44%, live with some form of heart disease. When women do have acute myocardial infarction, they are more likely than men to be misdiagnosed at first contact and to wait longer for reperfusion. Short-term death rates after MI are also higher for women in several datasets. This article gathers the numbers a reporter can lift, with the primary source named beside each one.

Measure	Women	Men	Source
MI presentation without chest pain	42.0%	30.7%	Canto et al., JAMA 2012
Young AMI patients with chest pain	87.0%	89.5%	VIRGO / Lichtman, Circulation 2018
Perceived symptoms as stress/anxiety	20.9%	11.8%	VIRGO / Lichtman, Circulation 2018
Provider said prior symptoms not heart-related	53%	37%	VIRGO / Lichtman, Circulation 2018
STEMI 30-day mortality after PCI	9.3%	6.5%	Stehli et al., JAHA 2019
In-hospital STEMI mortality	9.4%	4.5%	Ezekowitz et al., Circulation 2020

Source: Canto et al., JAMA 2012; Lichtman et al., Circulation 2018 (VIRGO); Stehli et al., JAHA 2019; Ezekowitz et al., Circulation 2020.

Survey year	Awareness rate
1997	30%
2009	54%
2012	56%
2019	44%

Source: Mosca et al., Circulation 2013 telephone series (1997: 30%; 2009: 54%; 2012: 56%); Cushman et al., Circulation 2021 / PMC11181805 online survey (2019: 44%). A separate online comparison in Cushman put 2009 awareness at 65% versus 44% in 2019. Rounded to whole percentages as reported.

How often the first diagnosis is wrong

The cleanest large-scale measure of first-contact error comes from England and Wales. Wu and colleagues analyzed 564,412 patients discharged with a final diagnosis of STEMI or NSTEMI from 243 acute hospitals between 2004 and 2013, using the Myocardial Ischaemia National Audit Project. Nearly one in three (29.9%, or 168,534 patients) had an initial diagnosis that did not match the final diagnosis.

That mismatch was not cosmetic. Patients who arrived with "other" initial diagnoses (non-cardiac labels such as pancreatitis, or non-infarct cardiac labels) received markedly less guideline-directed care: lower rates of pre-hospital ECG, aspirin on admission, cardiologist care, invasive coronary procedures, and discharge medications. For STEMI, one-year mortality after discharge was 5.6% when the initial diagnosis was STEMI, 8.4% when it was NSTEMI, 8.3% when it was chest pain of uncertain cause, and 21.3% when it was another initial diagnosis. For NSTEMI, the parallel figures were 10.7% with a correct initial NSTEMI label and 25.5% with other initial diagnoses. The authors estimated that correct initial classification might have prevented about 33 STEMI deaths and 218 NSTEMI deaths per year in that system.

Sex-specific risk within that same MINAP work is what made the numbers travel. According to the British Heart Foundation's report on the University of Leeds analysis, women finally diagnosed with STEMI had a 59% greater chance of an initial misdiagnosis than men, and women finally diagnosed with NSTEMI had a 41% greater chance. The BHF noted that additional, unpublished analysis by the same team put the combined figure near a 50% higher chance overall for women. The published 59% and 41% breakdowns are the ones I prefer to hand a reporter.

The pattern is not a UK curiosity. It matches what we document in medical gaslighting and diagnostic delay statistics: symptoms that do not match the male textbook case get filed under something else until laboratory proof forces a re-label.

What women actually feel

The misdiagnosis story is often told as if women "don't have chest pain." The better data is more precise.

Canto and colleagues examined 1,143,513 MI hospitalizations in the National Registry of Myocardial Infarction from 1994 to 2006 (481,581 women and 661,932 men). Presentation without chest pain was 42.0% among women and 30.7% among men. The sex gap was largest at younger ages. In-hospital mortality was 14.6% for women and 10.3% for men. Presentation without chest pain carried a mortality penalty that was especially steep for younger women.

VIRGO complicates the "atypical only" narrative. Lichtman and colleagues interviewed 2,009 women and 976 men aged 18 to 55 years hospitalized for AMI at 103 U.S. hospitals. Chest pain (pain, pressure, tightness, or discomfort) was present in 87.0% of women and 89.5% of men. Most young women with heart attacks still had chest symptoms. What differed was the surrounding picture: women more often reported three or more associated symptoms (61.9% versus 54.8%), including epigastric symptoms and pain in the jaw, neck, arms, or between the shoulder blades. Women with STEMI were more likely than men to present without chest pain (adjusted odds ratio 1.51).

Perception mattered as much as physiology. In VIRGO, 20.9% of women attributed symptoms to stress or anxiety versus 11.8% of men. About 29.5% of women and 22.1% of men had sought care for similar symptoms before the index hospitalization; of those who did, 53% of women said their provider did not think the symptoms were heart-related, versus 37% of men. Patient and clinician both under-attribute, and the clock keeps running.

Crushing substernal pain is classic. Classic is a teaching device, not a gate. New jaw pressure, unexplained dyspnea, or sudden profound fatigue in a midlife woman has earned an ECG and a troponin.

The ER discharge gap

Missed hospitalization is the sharp end of misdiagnosis. Pope and colleagues studied 10,689 emergency-department patients evaluated for possible acute cardiac ischemia across 10 U.S. hospitals. Among 889 patients with acute MI, 2.1% were mistakenly discharged; among 966 with unstable angina, 2.3% were discharged. Multivariable analysis found that patients with acute cardiac ischemia were more likely not to be hospitalized if they were women under 55 (odds ratio 6.7; 95% CI 1.4-32.5), were nonwhite (OR 2.2), reported shortness of breath as the chief symptom (OR 2.7), or had a normal or nondiagnostic ECG (OR 3.3).

The confidence interval on that 6.7 odds ratio is wide (thinner data than a national registry), but the direction has held up. Young women are the group clinicians least expect to infarct, and expectation drives disposition. Nonwhite patients carried an independent discharge risk in the same model, so sex bias and racial bias can stack. That is one reason our page on Black maternal health disparities keeps pointing back to cardiovascular risk: pregnancy and the postpartum year already raise cardiac risk for young Black women, and a low-suspicion ER visit is a dangerous place to land.

Pope's study is from 2000, and high-sensitivity troponin has changed emergency medicine since then. It has not erased the problem. The Wu MINAP data, through 2013, still found nearly a third of infarctions mislabeled at first contact. Better assays help only when someone orders them and believes the result.

Time to treatment

Even after the diagnosis is on the board, women wait longer.

Stehli and colleagues analyzed 13,451 patients (22.5% female) who underwent PCI for STEMI or NSTEMI in the Victorian Cardiac Outcomes Registry between 2013 and 2016. Among STEMI patients, women had longer symptom-to-door times (adjusted geometric mean ratio 1.20) and longer door-to-balloon times (adjusted geometric mean ratio 1.12). Absolute adjusted delays: 28.8 minutes longer symptom-to-door and 7.7 minutes longer door-to-balloon. Prehospital delay was about four times the in-hospital delay. Thirty-day STEMI mortality was 9.3% in women and 6.5% in men; female sex independently predicted higher 30-day mortality (odds ratio 1.67). For NSTEMI in the same cohort, adjusted times and early mortality did not differ by sex. The gap is not uniform across infarct types.

That four-to-one prehospital-to-in-hospital ratio is the part public-health campaigns should own. Door-to-balloon has been a quality obsession for years and has improved. Symptom-to-door for women has not closed the same way, because it depends on whether she recognizes the symptom as cardiac and whether bystanders take it seriously enough to call EMS. Those are awareness problems as much as catheter-lab problems.

Our heart disease in women statistics page shows the same architecture: higher mortality and delayed recognition. Treatment patterns still lag after risk adjustment in several datasets. Misdiagnosis is one mechanism producing the outcome gap.

SCAD and younger women

Spontaneous coronary artery dissection is the diagnosis that most often blindsides clinicians trained on atherosclerotic plaque rupture. The 2018 American Heart Association scientific statement on SCAD (Hayes et al.) estimates that SCAD accounts for 1% to 4% of all acute coronary syndromes and up to 35% of myocardial infarctions in women 50 years of age and under. It is a leading cause of pregnancy-associated MI. Patients are often younger women with few conventional risk factors. Management differs from atherosclerotic MI, including a stronger preference for conservative therapy when the patient is stable, because intervention in dissected vessels carries higher technical failure rates.

SCAD is easy to miss on angiography if the operator is not looking for it, and easy to misclassify if the patient is a 38-year-old runner with normal lipids who presents with neck pain and nausea. Pregnancy, the postpartum period, extreme emotional stress, and intense physical effort are recognized triggers. Our coverage of pregnancy complications including preeclampsia and gestational diabetes statistics already flags pregnancy as a cardiovascular stress test; SCAD is the acute extreme for a smaller group of women.

If you leave one SCAD number with a reader, make it the 35%. In women 50 and under with MI, SCAD is not a footnote. It is a leading etiology, and it requires a different mental model from "clogged arteries."

Outcome gaps after MI

Misdiagnosis and delay would matter less if outcomes equalized once care started. They do not.

Ezekowitz and colleagues studied 45,064 patients with first-time MI in Alberta between 2002 and 2016 (30.8% women). Women were older (median age 72 versus 61), had more comorbidities, and underwent diagnostic angiography less often (74% versus 87%). When angiography was performed, women had less multivessel obstructive disease and more single-vessel or nonobstructive disease. Unadjusted in-hospital mortality was 9.4% for women with STEMI versus 4.5% for men, and 4.7% versus 2.9% for NSTEMI. After adjustment, the STEMI mortality gap remained significant (adjusted odds ratio 1.42); the NSTEMI gap did not. After discharge, heart failure developed in 22.5% of women versus 14.9% of men after STEMI, and in 23.2% versus 15.7% after NSTEMI. Over five years, women remained at higher risk of death or heart failure after both infarct types even after accounting for angiography and revascularization.

Stehli's PCI cohort showed a similar early signal: 9.3% versus 6.5% 30-day STEMI mortality. Canto's NRMI analysis put unadjusted in-hospital mortality at 14.6% for women and 10.3% for men. The AHA scientific statement on acute MI in women (Mehta et al., 2016) remains the clinical synthesis of sex-specific presentation, treatment gaps, and outcomes.

Some excess risk is age and comorbidity. Some is less aggressive care. Some is biology: microvascular disease, plaque erosion, SCAD, MINOCA. We cannot pin the entire gap on misdiagnosis. We can say misdiagnosis and delay are measurable, modifiable pieces of it, and that they hit women harder at the front end of care. Underrepresentation in cardiovascular trials still distorts therapy evidence; see our review of the women's clinical trials and research funding gap.

Awareness is sliding backward

Public recognition is supposed to shrink prehospital delay. For a while it did. Then it slipped.

American Heart Association telephone surveys found that awareness among U.S. women that heart disease is the leading cause of death rose from 30% in 1997 to 54% in 2009 and 56% in 2012 (Mosca et al.). Matched online AHA surveys later documented a reverse: awareness fell from 65% in 2009 to 44% in 2019 (Cushman et al.). The 2019 survey found the greatest declines among Hispanic women, non-Hispanic Black women, and women ages 25 to 34. Those groups need prevention most, and they are overrepresented in SCAD and early-onset MI relative to older stereotypes. In 2019, women were more than twice as likely as in 2009 to incorrectly name breast cancer as the leading cause of death for women. Unaided recognition of several heart-attack warning signs also fell versus 2009.

That reverse is not trivia. If a woman believes breast cancer is her primary lethal risk, she is less likely to treat jaw pain and dyspnea as emergencies. If her clinician shares that mental model, the ER discharge odds shift.

For women tracking reproductive health (tools such as our period calculator, ovulation calculator, and cycle length tracker), the cardiovascular message is adjacent. Hypertensive disorders of pregnancy and gestational diabetes mark higher long-term heart-disease risk, as do early menopause and polycystic ovary syndrome. A normal lipid panel at 42 does not cancel a history of preeclampsia. The misdiagnosis gap starts when pregnancy complications are treated as obstetrical events only, not as lifelong cardiovascular risk factors.

What would close the gap? Faster ECG and troponin use for non-classic symptoms. Sex-specific troponin thresholds where validated. Angiographers trained to recognize SCAD. Quality metrics that track symptom-to-door by sex, not only door-to-balloon. A cultural reset that stops treating midlife women's nausea and fatigue as anxiety until proven otherwise. The numbers have been consistent long enough. What remains is operational.

Frequently asked questions