Sulaiman Alshaks, MD
Psychiatrist and Clinical Associate of Psychiatry
On World Mental Health Day, I reflect on the experiences of women who report their somatic symptoms to clinicians. By drawing on historical accounts of their interactions with healthcare and highlighting modern research, I conclude that clinicians still have work to do to validate women’s experiences and improve their outcomes.
When the diagnosis does not fit the history
A woman presents to the emergency department with several hours of new-onset continuous dizziness, repeated vomiting, and an unsteady gait. She is assessed by the neurologist and told that her symptoms are most likely benign paroxysmal positional vertigo (BPPV). The diagnosis is insisted upon.
She is discharged. Imaging later showed that she was having a cerebellar stroke.
In this example, the history did not fit the syndrome particularly well. Classic BPPV causes brief, recurrent episodes of vertigo triggered by changes in head position, with individual attacks usually lasting less than one minute. Continuous dizziness lasting hours should prompt consideration of alternative diagnoses.1,2 Surely, the neurologist is presumed to be an expert on the diagnostic criteria of these disorders, so lack of knowledge is less likely. So, what went wrong?
This is the broader question behind Are Clinicians Listening to Women?
Relevant background
In the 1800s, a documented pattern of dismissing women who sought medical care was emerging. Hysteria was a diagnosis used for centuries to explain a wide range of symptoms, mostly reserved for women. Anxiety, fatigue, fainting, pain, or emotional distress were among these symptoms. The term derives from the Greek hysterikos (translating to "of the womb"). Instead of investigating whether a woman's symptoms had a physical cause, doctors would sometimes conclude that her body was inherently unstable or overly emotional.3 This meant real physical illnesses could get written off as "nerves," rather than properly worked up and treated.4
Disclaimer: I do not suggest that clinicians purposefully dismiss women and instead maintain that the issue is more subtle. Symptoms can be normalized, under-recognized, attributed to a less serious explanation, or filtered through assumptions about how a patient communicates distress. In other cases, it may speak to our ability as clinicians to face our emotions in response to our patients’ emotions.
Problem 1: The classic medical model and atypical presentations
There is substantial evidence that women's symptoms have historically and contemporarily been interpreted differently from men's.
Cardiovascular research offers clear examples. A systematic review found that women were less likely to receive certain diagnostic investigations, cardiology referrals, and invasive treatment. Their symptoms were more likely to be misattributed to gastrointestinal problems or anxiety.5 Another analysis found that women with acute coronary syndrome experienced treatment delays and were less likely to receive invasive treatment.6 In many cases, atypical symptoms were described as the reasons for the delays.5,6 When managing cardiac chest pain, women in emergency departments were less likely than men to receive opioid analgesics, with the likelihood further decreasing for older women.7
The medical model itself has historically been less well equipped to recognize women's presentations. Women and men are more likely to locate their angina in different places.8 The location often experienced by women was called "atypical." The associated symptoms of angina that women are more likely to present with are often categorized as atypical as well. This is particularly important because what physicians have traditionally been trained to regard as atypical symptoms has been suggested to be more prone to being labeled as non-cardiovascular.9
Psychiatry is no exception. Women may be overdiagnosed with histrionic and borderline personality disorders, and somatization disorder, with some research linking it to gender bias.10 Conditions like Autism Spectrum Disorder (ASD), Attention Deficit and Hyperactivity Disorder (ADHD), and Posttraumatic Stress Disorder (PTSD) may be underdiagnosed due to less recognized symptom presentation.11 This carries its own implications, as some of these diagnoses may carry stigma, while others, despite also carrying stigma, may provide access to recognized accommodations and legally mandated support.
Problem 2: The psychiatric label can become an anchor
During my psychiatric training with patients who have intellectual disabilities with/without comorbid severe autism, we were repeatedly reminded about the concept of diagnostic overshadowing, where one’s newly emerging symptoms are falsely attributed to the previously established diagnosis.12 When these patients acutely display challenging behaviors such as being irritable or "difficult," the reasons behind the behaviors may be brushed off as a part of their condition. It was often the case that this was their best attempt at communicating pain from a physical condition, such as constipation or, as I once witnessed, an undetected cerebrovascular stroke.
While a woman’s psychiatric history is clinically meaningful, it is worthwhile examining whether it creates a form of diagnostic overshadowing. In a study of older women who had reported symptoms to their healthcare providers, those with a history of anxiety and depression were more likely to have their breast cancer diagnoses delayed by 3 months compared with women without these mental health conditions.13
Anxiety, depression, somatic symptom disorder, psychosis, and other psychiatric conditions may influence how patients experience and communicate symptoms. A pre-existing mental health diagnosis should inform the assessment rather than determine it. Otherwise, it risks becoming the explanation behind the patient’s symptoms before the underlying organic condition is recognized. A woman can have anxiety with an arrhythmia, or depression with endometriosis. She can have a psychiatric diagnosis and subsequently develop a medical illness.
Problem 3: Emotional distress as sole diagnostic evidence
The same principle applies even when there is no psychiatric history, but patients express certain emotions. Take distress, for example, which can be expressed in different ways. Some patients become withdrawn, while others may cry, ask the same questions, or become frustrated.
A patient who is experiencing severe physical symptoms may understandably become emotionally distressed. A patient who has repeatedly sought help without receiving an explanation may become even more distressed. And sometimes, that presentation may be difficult for the clinician.
Patients of all genders express emotions. Why specifically mention them when discussing women? Gender stereotypes about women generally regard them as being more expressive of their emotions.14 Stereotypes about female patients’ emotionality have been suggested to increase distrust in their reported pain.15 When physicians were given case vignettes of women and men having identical symptoms, women were more likely to be diagnosed with psychosomatic rather than organic illnesses compared with men. The mere stereotype about women’s emotional expression, regardless of a patient’s actual expression of emotions, may have a role in the care plan she receives.16, 17
Another point relates to how the clinician manages their internal response to the patient’s emotions and whether it influences their clinical judgment. For example, when faced with a frustrated patient, one could ask themself, "Am I feeling overwhelmed by this patient? Am I likely to want her to leave the office soon and hence am I less likely to take a comprehensive history or perform appropriate tests?"
That is precisely when we need to become more aware of our own reactions.
Acknowledging what one is feeling, no matter how uncomfortable the content of those thoughts may be, is the first step in ensuring the clinical decision is not compromised. One could follow with "I understand I am upset, and when people are upset they may do things in a rush to escape the feeling. But I am a clinician, and I will not allow my emotions to make me perform any less for this patient."
A further astute follow-up could be, "Why could this patient be so upset? Am I doing anything that could be making it worse that I should stop doing? Is there something that I can do to ease things for her? Perhaps I should reassure her, show her how I am listening by applying methods of active listening, or maybe offer her a glass of water."
If I find myself becoming impatient with a patient's repeated questions on a busy clinic day, frustrated by her anger, or inclined to think her emotional response is disproportionate, I face two options. I can recognize my feelings and the patient’s, consider what might be causing both, and think about ways to resolve them. The other option is to ignore them and let them possibly influence my clinical judgment.
What can we do differently?
1. Listen carefully
Listen, and believe unless there is a good reason not to. Duration, onset, triggers, progression, and associated symptoms are not minor details. In the dizziness example, "continuous for several hours" is clinically different from "brief episodes whenever I turn in bed." The diagnosis should follow the history, not the other way around.1,2
2. Reassess whether the differential diagnosis fits the whole presentation
Ask whether the proposed differentials are comprehensive for the presentation. If an apparently benign diagnosis requires us to ignore certain findings, reconsider it. Also, it is worthwhile remembering hidden biases and the limitations of research in some conditions when it comes to studying women, emphasizing certain symptoms while downplaying others.
3. Consider the organic causes
Begin by separating the patient's physical symptoms from her emotional response and psychiatric history and allow enough focus to consider relevant organic causes. Only after you have ruled out the physical causes with reasonable confidence should you be diagnosing a mental illness as the sole cause of all the patient’s symptoms. If you assess that