Labels Matter in Diagnoses

iStock

By Mary Carpenter

FROM Metabolic syndrome to Post-Traumatic Stress Disorder and Ehlers-Danlos Syndrome, labels such as “syndrome” or “disorder” can elicit dismissive responses from medical professionals and even to gaslighting of patients—while bestowing the “disease” label can earn better attention and treatment. Lower-order labels—for both mental and physical symptoms—often result when diagnostic criteria for diseases are too restrictive or complex. But ChatGPT appears to be helping—with responses both to complicated symptoms and communication with the patients involved.

Metabolic syndrome, for example, is the cluster of conditions that increase risk for heart disease, stroke and type 2 diabetes—but that combination of risky symptoms receives less attention than the individual diseases, which include increased blood pressure, high blood sugar and excess body fat around the waist, according to the Mayo Clinic.

A diagnosis of PTSD, a different kind of problem, depends on unusually specific parameters—beginning with exposure to a “potentially life-threatening event…severe injury or sexual violence,” according to behavioral researchers studying “the frequency of clinically undetected PTSD” at Wayne State University and elsewhere. In addition, “individuals must experience intrusive symptoms related to the event, engage in persistent avoidance of stimuli associated with the event, and display [changes] in cognition, mood, arousal and reactivity…”

Because diagnostic requirements are complicated, physicians often miss or ignore associated medical conditions that could help with its diagnosis but are not listed in the Diagnostic and Statistical Manual ). The researchers focused on aboulomania—a “neurosis…characterized by pathological indecisiveness…that severely hampers [patients’] ability to function normally” —to “raise awareness both about the limitations of the DSM-5-TR [and the] importance of maintaining an open-minded approach when treating mental illnesses.”

“Is EDS real?” queries a diagnostic algorithm, according to a STAT post—noting frequent failures to diagnose the hypermobility that is related to EDS but remains in “medicine’s grey area.” Very loose joints may fail to receive the hypermobile EDS (hEDS) label because, unlike the 13 other EDS subtypes that are linked to specific genetic variants, hEDS has no known cause.

“Revenge of the Gaslit Patients,” headlines the STAT post—describing hEDS patients who “might look healthy, their tests might show no signs of disease, but they endure repeated joint dislocations, headaches and pain, as well as gastrointestinal issues, fatigue, and stretchy and fragile skin.” Recently discovered genetic links to the condition could help—but await months or even years of research to confirm.

“Syndromes [defined as] certain groups of symptoms that occur together” are the lowest ranking term in the hierarchy of diagnoses, according to verywellhealth. The next level, “’disorders’ …refers to “a group of symptoms that disrupt normal functions in the body or cause significant impairment;” while the highest level is “disease.” Pain in the joints, the example given, can receive a diagnosis of “arthritis disorder,” while additional testing that reveals immune system dysfunction leads to a label change to “rheumatoid arthritis,” which gets much more serious attention.

The greatest challenges can arise with diseases that are rare or have diagnoses based on recently discovered laboratory testing or genetic links—with endless examples in Lisa Sanders’s New York Times’s  “Diagnosis” column. One woman, for example, waited 30 years for a new test that could diagnose her rare disease—a variety of congenital myasthenic syndrome—which came with the good news of a treatment: the common asthma drug albuterol.

Another woman waited 18 years for the discovery of an antibody that attacks a protein in the brain and spinal cord, the myelin oligodendrocyte glycoprotein (MOG) —and her “relief at having an answer was immediate,” writes Sanders—even though the altered treatment that improved her symptoms cannot cure the disease.

Large language models (LLMs), such as ChatGPT, are increasingly helpful to physicians and to patients—as New York gynecological surgeon Sara Farag told the healthcare news site fiercehealth: “The busier that our days get and the more information we have to sift through, the more useful that LLM’s [sic] can be.” And among the non-medical population, a Kaiser Foundation poll found “about one in six adults… use AI chatbots at least once a month to find health information or advice—while one in three trust” that AI chatbots are reliable.

More surprising, though, is the finding that “ChatGPT’s answers to patient questions were rated as more empathetic (and also of higher quality) than those written by actual doctors,” wrote Jonathan Reisman, internist and author of The Unseen Body: A Doctor’s Journey Through the Hidden Wonders of Human Anatomy, in last week’s New York Times. “The reason…in medicine—as in many other areas of life—being compassionate and considerate involves, to a surprising degree, following a prepared script.”

For me, having symptoms taken seriously—or even better, given a label at any level—has been helpful even in the absence of treatment. For example, it came as a relief to learn about my hypermobility, even if not hEDS— to understand that minor weakness in my back and knees may not be entirely my fault, caused by my failure to “stand up straight” as I was exhorted throughout my childhood, or to my ongoing failure to do much more frequent or strenuous exercise.

—Mary Carpenter regularly reports on topical subjects in health and medicine.