Intended for healthcare professionals

Views & Reviews Personal View

The new somatic symptom disorder in DSM-5 risks mislabeling many people as mentally ill

BMJ 2013; 346 doi: (Published 19 March 2013) Cite this as: BMJ 2013;346:f1580
  1. Allen Frances, chair of the DSM-IV task force
  1. 11820 Avenida Del Mundo, Coronado, CAL 92118, USA
  1. allenfrances{at}

This new condition suggested in the bible of mental health diagnoses lacks specificity, says Allen Frances

The fuzzy boundary between psychiatry and general medicine is about to experience a seismic shift. The next edition of the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM) is scheduled for release this May amid controversy about many of its new disorders. Among these, DSM-5 introduces a poorly tested diagnosis—somatic symptom disorder—which risks mislabeling a sizeable proportion of the population as mentally ill.

The relation between psychiatry and the rest of medicine has been difficult to manage both for mental health practitioners and for primary care doctors, and this is even more problematic for patients caught in-between. The boundary has never been clear cut or static but has shifted back and forth depending on new findings and fashions. The realm of psychiatry would shrink, and that of medicine would expand, whenever advancing science discovered a cause for a previously poorly understood presentation. The classic example of this is “general paresis of the insane,” which went from psychiatry to neurology as soon as the spirochete was identified as the causal agent.

In DSM-5, “somatic symptom disorder” appears in a new section, “Somatic symptoms and related disorders,” which replaces the “Somatoform disorders” section found in DSM-IV. This new category will extend the scope of mental disorder classification by eliminating the requirement that somatic symptoms must be “medically unexplained.” In DSM-5, the focus shifts to “excessive” responses to distressing, chronic, somatic symptoms with associated “dysfunctional thoughts, feelings, or behaviors.”

The overinclusiveness of this diagnosis is suggested by the results of the DSM-5 field trial study reported by the somatic symptom disorder work group at the 2012 annual meeting of the American Psychiatric Association. Somatic symptom disorder captured 15% of patients with cancer or heart disease and 26% with irritable bowel syndrome or fibromyalgia, and it had a high false positive rate of 7% among healthy people in the general population.1 The rate of psychiatric disorder among medically ill patients is unknown, but these rates seem high, and the burden of proof before introducing any new diagnosis is that it has a favourable risk to benefit ratio. Yet the proposed diagnosis is unsupported by any substantial evidence on its likely validity and safety and was strongly opposed by patients, families, caregivers, and advocacy organizations.2

The DSM-5 definition of somatic symptom disorder is loose. It requires only one bodily symptom that is distressing or disruptive to daily life, which lasts at least six months. It also requires one of the following psychological or behavioral responses: disproportionate thoughts about the seriousness of symptom(s); persistently high level of anxiety about symptom(s); or excessive time and energy spent on health concerns.3 This is far looser than the (rarely used) definition of somatization disorder in DSM-IV. This required a history of many medically unexplained symptoms before the age of 30 years that occurred over several years and which resulted in treatment being sought or psychosocial impairment. A total of eight or more medically unexplained symptoms were needed from four specified symptom groups, with at least four pain and two gastrointestinal symptoms.4

Previous DSM criteria have always included reminders to clinicians to rule out other explanations before concluding that any mental disorder is present. I suggested to the working group that similar reminders should be included this time and that before somatic symptom disorder is diagnosed clinicians should consider whether the health concerns are completely unrealistic or whether an underlying medical disorder might account for them. I also suggested that clinicians should consider whether symptoms might be caused by one of several mental disorders that often present with physical problems (such as depression, generalized anxiety, or panic disorder). The DSM-5 working group reviewed these suggestions and rejected them.

Misapplication of these catch-all criteria, especially in harried primary care practice, may result in inappropriate diagnoses of mental disorder and inappropriate medical decision making.5 Millions of people could be mislabeled, with the burden falling disproportionately on women, because they are more likely to be casually dismissed as “catastrophizers” when presenting with physical symptoms.

A false positive diagnosis of somatic symptom disorder harms patients because it may result in any underlying medical causes being missed. It also subjects patients to stigma, inappropriate drugs, psychotherapy, and iatrogenic disease; disadvantages them in decisions relating to employment, education, and healthcare entitlements; skews their self perceptions and those of family and friends; and places parents of children with chronic illness at risk of accusation of “overinvolvement” or of maintaining “sick role behavior.”

Every diagnostic decision is a delicate balancing act between definitions that will result in too much versus too little diagnosis—the DSM-5 work group chose a remarkably sensitive definition that is also remarkably non-specific. This reflected a consistent bias throughout DSM-5 to expand the boundaries of psychiatric diagnosis with what I believe was insufficient attention to the risks of the ensuing false positive mislabeling.

The DSM-5 diagnosis of somatic symptom disorder is based on subjective and difficult to measure cognitions that will enable a “bolt-on” diagnosis of mental disorder to be applied to all medical conditions, irrespective of cause. ICD-11 (International Classification of Diseases, 11th revision) is now being prepared by some of the same people who worked on DSM-5.6 7 Unless ICD-11 applies a higher standard of evidence and risk benefit analysis, it may repeat the mistake of casually mislabeling the physically ill as also mentally disordered.8

The late Thomas Szasz once said: “In the days of the Malleus, if the physician could find no evidence of natural illness, he was expected to find evidence of witchcraft: today, if he cannot diagnose organic illness, he is expected to diagnose mental illness.”9 Szasz’s general critique of psychiatry was too broad, but he was correct when it comes to the loosely defined somatic symptom disorder in DSM-5. Clinicians are best advised to ignore this new category. When a psychiatric diagnosis is needed for someone who is overly worried about medical problems the more benign and accurate diagnosis is adjustment disorder.


Cite this as: BMJ 2013;346:f1580


  • I thank Suzy Chapman, patient advocate, Poole, UK, of Dx Revision Watch, for valuable comments and suggestions.

  • Competing interests: I have read and understood the BMJ Group policy on declaration of interests and declare the following interests: I will publish two books that include references to DSM-5 (Saving Normal and Essentials of Psychiatric Diagnosis) and I chaired the DSM-IV task force.

  • Provenance and peer review: Not commissioned; externally peer reviewed.


View Abstract

Log in

Log in through your institution


* For online subscription