Noor F. Shaik, MD, PhD
When I was younger, one of my favorite TV shows was Mystery Diagnosis. I was captivated by the unfolding drama: a string of baffling symptoms, test after test negative, treatments failing — until, finally, a breakthrough. I breathed a sigh of relief when the patient appeared beside the clinician who had solved the puzzle.
Noor F. Shaik, MD, PhDIt wasn’t until my final year of medical school that I realized how uncommon those tidy endings really are.
Uncertainty is woven into the fabric of medicine. From diagnosis to treatment decisions, to prognostication, clinicians navigate ambiguity daily. And our patients vary in whether they want uncertainty disclosed. What is clear, though, is that intolerance of uncertainty contributes to psychological distress in both medical students and physicians. Fortunately, more seasoned clinicians tend to be more comfortable with the unknown.
As one of the most experienced neurologists in our department puts it, “A neurologist must learn to tolerate ambiguity.”
As I’ve progressed in training, I’ve come to recognize how uncertainty permeates the practice of neurology.
Stroke Mimics
Stroke is often perceived as a binary diagnosis: Either a patient has a stroke, or they don’t. If only it were that straightforward.
Emergency physicians can miss up to 15% of cases, particularly when patients present with nontraditional symptoms such as confusion or dizziness. Meanwhile, only about half of patients initially labeled as having “possible stroke” are ultimately confirmed to have a stroke by a neurologist. Many turn out to have stroke mimics such as migraine, vertigo, or seizures. Strikingly, more than one third of these patients have symptoms without a clearly identifiable cause.
Given that most strokes are ischemic and require timely initiation of thrombolytic therapy, misdiagnosis cuts both ways: overtreatment when someone has a stroke mimic, and dangerous undertreatment when it is missed.
Anti‑NMDAR Encephalitis, or Brain on Fire
Another well ‑ known example of neurologic misdiagnosis comes from Susannah Cahalan’s memoir Brain on Fire, which was also adapted into a film. She describes how her personality shifts, paranoia, and hallucinations were initially interpreted as psychiatric in origin. She was even treated with antipsychotic medications.
It wasn’t until she had seizures that she was diagnosed with anti-N-methyl-D-aspartate receptor (anti-NMDAR) encephalitis, a disease first described just 2 years earlier. With immunotherapy, she made a full recovery. But the illness’s highly variable blend of psychiatric and neurologic symptoms continues to delay diagnosis for many patients.
The confusing nature of the disease is a dual-edged sword. In one study, more than 25% of people initially labeled with anti‑NMDAR encephalitis were ultimately found to have different conditions, exposing them to unnecessary, and sometimes harmful, treatments.
MS and FND: Diagnostic Doppelgängers
Diagnostic ambiguity becomes even more pronounced with conditions such as multiple sclerosis (MS) and functional neurologic disorder (FND). Both MS and FND share several epidemiologic and clinical features: They are more common in women; often begin in early adulthood; and can present with highly variable symptoms such as numbness, tingling, or disequilibrium.
Yet their underlying mechanisms and treatments are profoundly different.
A meta‑analysis revealed that MS is often both misdiagnosed and underdiagnosed. Correcting the diagnostic error took anywhere from a few months to more than 15 years. Another study found that at least 10% of people initially diagnosed with MS actually had FND. Meanwhile, although the misdiagnosis rate of FND is lower, about 10% of these individuals may truly have MS.
These diagnostic swings are not benign. Over 70% of patients misdiagnosed with MS received unnecessary immune‑suppressing therapies, and more than one third experienced complications. Conversely, patients with FND who are mistakenly labeled as having MS may miss out on rehabilitative therapies that could significantly improve their function.
The Desire to Label
Amid these complexities, a lesson from another neurologist echoes in my mind: “It is easier to give a diagnosis than to take it away.”
As clinicians, we naturally want to give patients answers — something to name their suffering and help partner with them. But when a diagnosis is wrong, it can entrench misunderstanding, misdirect treatment, and have long-term psychological consequences.
So as we strive to solve the “mystery diagnoses” our patients bring to us, we should be humble enough to say “I don’t know yet,” to revisit our assumptions, and to recognize when uncertainty is not failure but a vital part of the diagnostic process.
Noor Shaik, MD, PhD, is a neurology resident physician at the Hospital of the University of Pennsylvania, Philadelphia. She has published academic research and patient-facing articles on the prevention and recovery after neurologic injuries, such as stroke and traumatic brain injury. The opinions expressed in this article do not necessarily represent those of the University of Pennsylvania Health System or the Perelman School of Medicine.
