Don’t stop at the label
I have been reflecting on the risks of turning every problem into a diagnosis. It is critical that patients are seen, heard, recognised and, in some cases, diagnosed. But there is a risk when we view the person in front of us through a single diagnostic lens, whether mental health or hormonal. Labels carry weight, and that weight can narrow our view of the person and the many factors that may be driving their symptoms or distress.
I also want to highlight the counter to this. Women have been underdiagnosed, dismissed and disbelieved for a very long time. Research on gender bias in pain care, shows women's pain is more likely to be attributed to emotional or psychological causes. Conditions like endometriosis, EDS and autoimmune disease are often associated with years of delay before recognition. For many women, a diagnosis is validation, access and the first time they have felt believed.
The question is really about what happens after the label. A diagnosis can shape identity, access to services, disability status and treatment options. But not every symptom is a disorder. Sadness, fear, grief and trauma responses can be normal reactions to life, not pathology. Feeling overwhelmed, sleeping poorly and being tired can arise from many inputs, not only hormonal ones.
When the measure becomes the target
Goodhart's law says that when a measure becomes a target, it ceases to be a good measure. We see this when diagnostic criteria become the gateway to support. Meeting the criteria becomes a strategic skill in its own right, and people who fall just outside them can be left without help.
Consider someone who narrowly misses the criteria for hEDS. Their hypermobility, and its real consequences for function and quality of life, may be dismissed, even though the burden is often similar whatever the label. Or consider a woman whose fatigue and poor sleep are attributed to perimenopause, with HRT offered as the answer, when stress load, nutrition, sleep disorders, thyroid function, iron status or pain were not explored.
The reverse also happens. A person can meet the criteria, receive the corresponding treatment and still live with significant distress, because the broader picture wasn’t addressed.
The social media effect
Social media amplifies all of this. Algorithms reward short, certain, relatable messages: "your fatigue is your hormones", "it's your nervous system", "this symptom means you have X". These posts can help people feel recognised and may prompt them to seek care. But they also flatten complexity. The idea of "concept creep", used by psychologist Nick Haslam, describes where terms like trauma and disorder expand to cover ever more ordinary experience.
The risk runs in both directions:
Over identification: normal variation or understandable distress gets pathologised, and the label becomes a shortcut that closes down curiosity.
Under investigation: a confident online explanation stops people, and sometimes clinicians, from looking for other contributors.
Identity capture: a self applied label shapes how someone sees themselves before any proper assessment has happened.
In an article by Farid Zaid, the suggestion is that we treat diagnostic criteria as tools that point to patterns, probabilities and available interventions, not as master keys to personal identity. I think that is a useful reframe. The article by Farid is a very interesting read.
The strongest version of this is not "diagnose less" but diagnose well, and don’t stop at the label:
Take symptoms seriously and investigate them thoroughly, especially in groups with a history of being dismissed.
Use diagnosis to open access and understanding, not to close the inquiry.
Ask what else is contributing: sleep, stress, nutrition, pain, hormones, life circumstances, relationships, environment.
Treat people who sit just outside the criteria as people with real needs, not as non cases.
Help patients evaluate what they read online without dismissing their lived experience.
And maybe, the diagnosis doesn’t have to define the person. A diagnosis should serve the person, not the other way around.