LAUSR

The study by Saunders and colleagues1 in JAMA Network Open is an important contribution to the field of what I’ll call medically unexplained symptoms (MUSs). MUSs in their most severe form encompass what the authors call somatic symptom and related disorders, as well as numerous other similar terms, such as somatic symptom disorder from psychiatry and chronic pain, irritable bowel syndrome, chronic fatigue, and fibromyalgia from medicine. From the administrative database in Ontario, the authors identified 33 272 individuals as a study cohort: inpatients with MUS-type codes from the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision–Canada (ICD-10-CA) and outpatients with psychosomatic codes for incident visits from 2008 to 2015. Approximately one-third of the patients were aged 4 to 17 years; the remaining two-thirds of the patients were aged 18 to 24 years. The authors calculated the health care costs and use in the year before and after the incident MUS diagnosis. Health care costs and use were high and remained so at follow-up, and many patients with MUSs did not receive physician-delivered mental health care for the high rates of mental disorders that were found. About three-fourths of these data reflect on young adults aged 18 to 24 years. It would be valuable to know whether these patients differed from the younger patients in this study and from older adults reported in the literature. This would help inform the question of a spectrum of severity based on age of incident diagnosis of MUSs or a history of suffering from severe childhood disadvantages, such as abuse and poverty. There are data indicating that chronic childhood abuse (physical, sexual, or psychological) is highly associated with MUSs in adults.2 The authors note correctly a major problem for the entire MUS field: that obtaining valid definitions of MUSs is problematic. This observation merits every clinician’s attention, because validity has not been demonstrated for any MUS diagnosis (whether from medicine or psychiatry). Particularly problematic has been that the various symptom criteria often overlap; for example, pain occurs in irritable bowel syndrome, chronic pain, and fibromyalgia. The problem becomes worse for clinicians. Although useful for research purposes, formal MUS diagnostic criteria have too many false positives (ie, an organic disease falsely identified as an MUS) for clinical use.3,4 Further limiting is the considerable complexity of various MUS criteria. There is another clinical pitfall for using MUS diagnoses in medical settings. Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition; DSM-IV) somatoform criteria, presumably identifying patients with the most severe MUSs, miss over three-fourths of patients with severe MUSs who frequently use health care services in primary care.5 More useful for clinical purposes, the same study5 found a 60.2% prevalence of nonsomatoform DSM-IV psychiatric disorders, mostly anxiety and depression. Although the Saunders et al1 study did not indicate the specific disorders, they also reported a high prevalence of mental disorders in their younger population. My conclusion and proposal is that, for clinical purposes, we should stop using and ignore the many unvalidated MUS diagnoses from psychiatry and medicine. They add nothing and can distract from the real problem of underlying depressive and anxiety disorders. Although we will still see many patients with unexplained symptoms, we should understand them simply as symptoms—not a diagnosis themselves. When are they significant to the clinician? The unexplained symptoms become + Related article