Diagnostic accuracy is the cornerstone of effective medical care, yet in obstetrics and gynecology—a field that touches some of the most intimate and complex aspects of human health—diagnostic errors remain an underrecognized threat. From missed ectopic pregnancies to delayed or misinterpreted Pap smears, the consequences of diagnostic missteps can be devastating, leading to delayed treatment, unnecessary interventions, and even loss of life.  

This article explores the multifaceted nature of diagnostic error in obstetrics and gynecology, examining its root causes, real-world impact, and the systemic changes needed to improve outcomes. By shedding light on this critical issue, MICA aims to empower physicians and advanced practice providers to recognize vulnerabilities and champion safer, more precise care. 

Diagnostic Error in Medicine 

For many physicians, the topic of diagnostic error is not just academic—it carries a profound emotional weight, shaped by firsthand experiences of its potentially life-altering consequences. Accurate diagnosis is the foundation of effective and timely treatment, and when it falters, diagnostic error emerges as one of the leading causes of preventable harm in obstetrics and gynecology. 

At least 20 studies have shown that accurate diagnoses improve survival rates, reduce mortality rates, decrease the number of unnecessary tests, and assure timely and correct treatment.¹ National rates of diagnostic error are mostly based on extrapolations of data from studies involving emergency department (ED) patients, inpatients, and primary care outpatients. 

A 2019 study of closed medical professional liability (MPL) files showed that three disease categories, now called the Big Three (vascular events, infections, and cancer) accounted for 75% of the files.² Key results of the 2019 study for OBGYNs included the following: 

• • Diagnostic errors represented 21% of all the files considered,
    
    
  • 51.7% of the diagnostic error files involved female patients,
    
    
  • Cancer was the most common diagnosis in the files of patients aged 21 to 80, and
    
    
  • Clinical judgment was by far the most common causal factor in the diagnostic error files. 

The clinical judgment factor contains these subfactors or “steps” in the diagnostic process: failure or delay in considering a diagnostic test; narrow diagnostic focus with failure to establish a differential diagnosis; failure to appreciate and reconcile relevant symptoms, signs, or test results; failure or delay in obtaining consultation or referral; and misinterpretation of diagnostic studies.³ The researchers concluded that problems in these “steps” in the diagnostic process require cognitive solutions.⁴

A 2024 study considered these key results but used larger clinical studies, not malpractice claims, to confirm that diagnostic errors mostly involve the same Big Three diseases.⁵ Based on the larger study size, the researchers concluded that “[D]iagnostic error is probably the single largest source of deaths across all care settings” and “constitutes an urgent public health imperative.⁶

Studies of Pregnancy-related Deaths 

Although modern imaging technologies, laboratory diagnostics, and clinical protocols have significantly advanced, the intricate physiological and emotional aspects of women's health continue to pose challenges to accurate diagnosis. These complexities can obscure clinical presentations, making timely and precise identification of conditions more difficult. 

While maternal death is a rare and extreme negative outcome in MICA’s closed OBGYN claim and lawsuit files, nine state and local maternity mortality review committees agree that “[e]ach maternal death is one too many”⁷ and their study of 237 pregnancy-related deaths resulted in useful conclusions. 

• Most pregnancy-related deaths (45%) occurred within 42 days of the end of the pregnancy followed by deaths while pregnant (38%).⁸
  
  
• 63% of the pregnancy-related deaths were preventable and 63% of the preventable deaths occurred during pregnancy.⁹
  
  

• Patient/family factors were the biggest contributors (38%) to all the pregnancy-related deaths followed by “providers” (34%).¹⁰
  
  
• The top causes of pregnancy-related death were hemorrhage, cardiovascular and coronary conditions, cardiomyopathy, infection, embolism, preeclampsia and eclampsia, and mental health conditions.¹¹

The committees identified themes among the pregnancy-related deaths including recency and adequacy of training, unenforced policies and procedures, inappropriate level of care determinations, patient-physician or -advanced practice provider communication, and mismanagement of mental health conditions.¹²

In the committees’ most recent report, which included files from 2020, preventability increased from 63% to 87%¹³ and the committees considered new and revised factors.  

• • Obesity, discrimination, and mental health conditions were new contributing factors to death. 
    
    
    • The committees identified obesity as a circumstance contributing to pregnancy-related death in 29% of the files reviewed and a probable circumstance in 7%.¹⁴
      
      
    • Discrimination contributed to 16% of the pregnancy-related deaths and probably contributed to 17%.¹⁵
      
      
    • Of the pregnancy-related death files with mental health condition as the underlying cause, the manner of death in 31% of the files was suicide.¹⁶
      
      
    • The means of death in files with mental health condition as the underlying cause was unintentional or unknown intent poisoning/overdose (65%).¹⁷
      
      
  • The top contributing factors in preventable pregnancy-related deaths were knowledge, clinical skill/quality of care, lack of access, and substance abuse disorder.¹⁸
    
    
  • The greatest contributors to pregnancy-related deaths were patient/family factors (38%) followed by “providers” (28%, down from 34%).¹⁹

This report did not include themes or other recommendations. Critics of the committees’ reports suggest that changes in the committees’ membership, data pool, and judgment of preventability may have affected the rate of preventability and identification of contributing factors.²⁰  

Cognitive Biases in OBGYN²¹ 

For at least four decades, graduate and continuing medical education programs have addressed cognitive biases but beating the biases requires critical self-analysis, role playing, analyzing hypothetical clinical situations, and practice. While the clinical data behind diagnostic errors in OBGYN develops, physicians and advanced practice providers should approach preventing diagnostic error by identifying and eliminating cognitive biases. 

A few studies are specific to OBGYN and support several MICA risk strategies for preventing cognitive biases: 

Availability Bias 

• • Definition – The tendency to consider the same diagnosis most recently made on another patient.²²
    
    
  • Example - A patient presenting with similar signs and symptoms as a patient recently diagnosed with pulmonary embolism and the physician is now working up the current patient for pulmonary embolism. 

Anchoring Bias 

• • Definition – When the receiving or subsequent treating physician relies on the diagnosis of the transferring or original treating physician. 
    
    
  • Example – A patient is admitted in preterm labor. The patient’s pain becomes localized and acute. The subsequent treating physician does not consider appendicitis or degenerating leiomyoma. 

Base Rate Neglect Bias²³ 

• • Definition – The physician does not integrate the true rate of a disease into the diagnostic process.  
    
    
  • Example – A physician sees a patient without cardiovascular risk factors at 34 weeks. The physical assessment is normal, but the patient complains of mild dyspnea. The physician begins extensive work ups to rule out pulmonary embolism or cardiomyopathy but not dyspnea of pregnancy. 

Confirmation Bias^{24, 25} 

• • Definition – The physician gives more credence to evidence that supports current beliefs and feels reluctant to change her/his/their mind. This is a common cognitive bias. 
    
    
  • Example – After a vaginal delivery with Category I FHR patterns researchers showed physicians blood in a collection container and asked them to estimate blood loss (EBL).
    
    
    • Physicians shown 500 cc of blood in container without volume markings and told the patient is hypotensive substantially underestimated EBL
      
      
    • Physicians shown 1500 cc and told the patient is normotensive underestimated EBL 
      
      
    • Physicians shown 1500 cc of blood in container and told the patient is hypotensive substantially underestimated EBL 
      
      
    • Physicians shown 500 cc and told the patient is normotensive underestimated EBL 
       
  • Example – Researchers told the physicians that the patient is 36 weeks, the previous treating physician ruled out labor and there is a Category I FHR pattern. Researchers showed the physicians an ultrasound image with amniotic fluid pockets free of the umbilical cord and body parts but without caliper measurements. They asked the physicians to estimate the amount of amniotic fluid.
     
    • Physicians who were told the patient is normotensive estimated a normal amount of amniotic fluid 
       
    • Physicians who were told the patient has chronic hypertension reported less than a normal amount of amniotic fluid 
      
      
  • Possible Solution to Bias - The study authors suggested cognitive debiasing of visual estimations, perhaps objectively measuring anything that can be measured. 

Tolerance of Ambiguity²⁶  

• • Definition – The ability to openly and neutrally perceive information that may be uncertain, vague, contrary, or have many meanings.  
    
    
  • Example - After completing five psychometric scales, researchers told physicians that the patient is primiparas for 1 prior low transverse cesarean delivery, now has a term cephalic singleton with Category II fetal heart rate (FHR) pattern, and the patient is asking for trial of labor after cesarean delivery (TOLAC) and vaginal birth after cesarean delivery (VBAC). The physicians with the highest proactive coping, ambiguity tolerance, reflective coping, and need for cognition scores and lowest anxiety scores were more likely to proceed with TOLAC and had greater VBAC rates. 
    
    
  • Possible Solution to Bias – The study authors recommended that physicians learn to embrace uncertainty and remain open to other possible diagnoses. 

Ambiguity, Anxiety, and Coping Skills Bias²⁷ 

• • Definition – A measure of coping skills, tolerance of ambiguity, and anxiety scores. 
    
    
  • Example - Physicians completed five psychometric scales. Researchers told them the patient is nulliparas and now term cephalic non anomalous singleton is a candidate for spontaneous vaginal, operative vaginal, and cesarean delivery.
    
    
    • The physicians with the highest reflective coping scores were significantly less likely to perform operative vaginal delivery.
      
      
    • Those with lower anxiety and higher ambiguity tolerance scores associated with increased risk of chorioamnionitis and postpartum hemorrhage respectively. 
      
      
  • Possible Solution to Bias – The study authors recommended education and training to improve adaptive decision-making. 

Zebra Retreat Bias²⁸ 

• • Definition – A hesitation or unwillingness to consider a rare diagnosis when there is strong evidence in favor of it. 
    
    
  • Example – The patient is 30 weeks, complains of headache and confusion, BP 150/100, urinary protein/creatinine ratio 0.7, platelet count 28,000, and has normal transaminases and elevated lactate dehydrogenase. The diagnoses include severe preeclampsia and hemolysis/elevated liver enzymes/low platelets syndrome. Three hours later the patient’s temperature is 102º. The physician does not consider thrombotic thrombocytopenic purpura. 

Racial and Cultural Biases 

• • Definition – Holding inflexible general beliefs or attitudes about all individuals in a racial or cultural group.
    
    
  • Examples – Believing a patient will not comply with medication and lab work instructions because of the patient’s race or limited English proficiency and choosing a less effective medication.
    
    
  • Possible Solution to Bias – Identifying or recognizing these biases and seeking out opportunities to eliminate them.   

Understanding cognitive biases can significantly and positively influence clinical decision-making and patient outcomes.  

Risk Strategies  

The MICA Risk Team recommends the following strategies for identifying and preventing cognitive biases. 

• • Debias critical thinking²⁹ by 
    • Learning more about 2 types of thinking:
      
      
      • Quick and automatic, which is most common in health care and is more susceptible to biases, and
        
        
      • Slow and systematic, which is less common; and 

• • Studying stereotypes and learning to overcome them by getting to know people in different social groups and thinking about people from a stereotyped group who do not fit the stereotype.³⁰ 
    
    
  • Improve quick and automatic thinking by participating in obstetric- or labor-and-delivery-related team training and drills.
    
    
  • Build on slow and systematic thinking by performing root cause analyses or “cognitive autopsies”³¹ under the protections of an appropriately qualified peer review or quality improvement committee and process.
    
    
  • Log into the MICA website, go to the CME course library, and take specialty-specific CME courses. If you do not know your user identification and password, please contact your insurance broker or the MICA Underwriting Customer Service Team at help@mica-insurance.com.
    
    
  • Stay tuned for new Risk Team OBGYN resources in 2026. 

The unique nature of women’s health care—spanning preventive screening, pregnancy, and reproductive health—creates numerous opportunities for diagnostic errors and diagnostic errors remains understudied. As the research continues, physicians should try to recognize and address cognitive biases to improve diagnostic accuracy and maternal and reproductive health outcomes.

 

<sup>[1] Alharbi, TA, Rababa, M, et al. (2025) Diagnostic Challenges and Patient Safety: The Critical Role of Accuracy – A Systematic Review. J Multidisip Healthc 18:3051-03064. Last accessed 10-06-2025 at Diagnostic Challenges and Patient Safety: The Critical Role of Accuracy – A Systematic Review - PMC.

[2] Newman-Toker DE, Schaffer AC, Yu-Moe CW, et al. Serious misdiagnosis-related harms in malpractice claims: the "big three" - vascular events, infections, and cancers.
Diagnosis (Berl) 2019;6:227–40. Last accessed 10-05-2025 at 10.1515_dx-2019-0019.pdf.

[3] Alharbi, Rababa, et al. at pp 234-234.

[4] Alharbi, Rababa, et al. at p22-23.
 
[5] Newman-Toker DE, Nassery N, Schaffer AC, et al. BMJ Qual Saf 2024;33:109–120. Last accessed 10-06-2025 at Burden of serious harms from diagnostic error in the USA.

[6] Id. at p116

[7] Building U.S. Capacity to Review and Prevent Maternal Deaths. (2018). Report from nine maternal mortality review committees. Last accessed 10-05-2025 at ReportfromNineMMRCs.pdf at p55.
 
[8] Id. at p14

[9] Id. at p22

[10] Id. at p24

[11] Id.

[12] Id. at p29

[13] Pregnancy-Related Deaths: Data from Maternal Mortality Review Committees | Maternal Mortality Prevention | CDC. Subscription required. Last accessed 10-05-2025.

[14] Id.

[15] Id.</sup>

<sup>[16] Id.

[17] Id.

[18] Id.

[19] Qian, J, et al. (2025). Insights from Preventability Assessments Across 42 State and City Maternal Mortality Review in the United States. Am J Obstet & Gynecology, 232(4):P394.E1-394.E10. Last accessed 10-05-2025.

[20] Atallah, F et al. (2022). Society for Maternal-Fetal Medicine Special Statement: Cognitive bias and medical error in obstetrics—challenges and opportunities. Am J Obstet & Gynecology, 227(2)B2 - B10. Last accessed 10-05-2025 at Society for Maternal-Fetal Medicine Special Statement: Cognitive bias and medical error in obstetrics—challenges and opportunities - American Journal of Obstetrics & Gynecology. Last accessed 10-05-2025.

[21] Id.

[22] Id.

[23] Id.

[24] Id.

[25] Atallah, F, Moreno-Jackson, R, McLaren, R, et al. (2021). Confirmation Bias Affects Estimation of Blood Loss and Amniotic Fluid Volume: A Randomized Simulation-based Trial. Am J Pernatol 38:1277-1280. Subscription required. Last accessed 10-05-2025.

[26] Yee, LM, Lieu, LY, Grabman, WA. (2015). Relationship Between Obstetricians’ Cognitive and Affective Traits and Delivery Outcomes Among Women with Prior Cesarean. Am J Obstet & Gynecology Subscription required. Last accessed 10-05-2025.

[27] Yee, LM, Liu, LY, Grabman, WA. (2014). Relationship Between Obstetricians’ Cognitive and Affective Traits and Their Patients’ Delivery Outcomes. Am J Obstet & Gynecology, 211(6):P692. E1-692. Subscription required. Last accessed 10-05-2025.

[28] Yee, et al.

[29] Atallah, et al.

[30] Id.

[31] Id.</sup>