The first draft of DSM-VBMJ 2010; 340 doi: http://dx.doi.org/10.1136/bmj.c1168 (Published 02 March 2010) Cite this as: BMJ 2010;340:c1168
- Allen Frances, chair, DSM-1V task force
The Diagnostic and Statistical Manual of Mental Disorders (DSM), published by the American Psychiatric Association, is the official method for making psychiatric diagnoses in the United States. It is also widely used around the world, particularly for research purposes. The current edition, DSM-IV, was published in 1994. The first draft of the next revision, DSM-V, was recently posted on the internet (www.dsm5.org) and was accompanied by considerable press fanfare and professional controversy.1
DSM-V has been in preparation for three years and is scheduled to appear in 2013. The work on DSM-V began with the unrealistic ambition of producing a paradigm shift in psychiatric diagnosis. The working groups preparing the various sections were encouraged to be innovative and to think “out of the box.”2 The criteria for making changes and the requirements were specified only recently and are fairly fluid.3 The whole process has also been criticised for being secretive, closed to external influences, and disorganised.4 5 6
The experience with DSM-IV should offer a painful lesson and caution. Efforts were made to be conservative and rigorous.7 Nonetheless, DSM-IV was an unwitting contributor to three false positive “epidemics.” Its publication coincided with high rates of attention deficit hyperactivity disorder, autistic disorder, and childhood bipolar disorders.8 Other factors contributed to these epidemics, particularly the ubiquitous marketing efforts of drug companies directed at doctors and the general public.
The lesson is clear: once the diagnostic system is in general use, even small changes can be amplified and twisted, with harmful and unintended consequences. The proposals contained in the first draft of DSM-V could potentially set off at least eight new false positive epidemics of psychiatric disorder. In their efforts to innovate, the working groups could expand the territory of mental disorder and thin the ranks of the normal. Five proposed new diagnoses are defined by non-specific symptoms that are common in the general population—binge eating, mixed anxiety depression, minor neurocognitive problem risk of psychosis, and temper dysregulation. Three existing disorders would have a major lowering of their already overinclusive diagnostic thresholds: attention deficit hyperactivity disorder, bipolar disorder, and major depressive disorder.
The changes suggested for DSM-V are well meaning. They are intended to promote the early identification and treatment of mental disorders and reduce resistance to treatment. The problem is that every increase in the sensitivity of a psychiatric diagnosis is accompanied by a concomitant drop in its specificity. False negatives can be reduced only at the cost of producing many more false positives. Because the suggested changes all occur at the boundary between mental disorder and normality, they could create vast numbers of misdiagnosed new “patients.”
The consequences are grave. For individuals, these include unnecessary treatment with drugs that have unproved benefit but known harm (particularly weight gain); stigma; difficulties getting life insurance and disability insurance; and a reduced sense of personal responsibility and control. For society there is the expense of unnecessary treatment; the diversion of scarce resources away from people who need it to those who essentially don’t; and a reduction in morale and resiliency, as the usual problems of everyday life become medicalised into mental disorders.
Early identification and intervention require that a specific diagnostic test (with a low false positive rate) and an effective and safe treatment is available. The suggestions for DSM-V would offer the worst of all possible worlds. They would cast so wide a net as to guarantee a harvest of false positives. The efficacy of the corresponding treatments has not been proved, and their safety—especially the atypical antipsychotics, which can cause weight gain—is in doubt.9
Two hundred years ago, Pinel defined the domain of psychiatry with the first systematic classification of its disorders. Every new system since has progressively expanded its boundaries. Old disorders are almost never discarded; yet new disorders and lowered thresholds have taken ever bigger bites out of normality. Who is responsible? Psychiatric classifications are inevitably created by experts. Experts dread false negatives yet can be blind to the problems of false positives. It must be appreciated that disorders originating and studied in highly selected research environments take on a life of their own when transplanted to primary care and nurtured by drug company promotions. The enthusiasms of experts should always be contained by a careful risk-benefit analysis that includes a critical review of the scientific literature, field testing in primary care settings, and a consideration of all the potential unintended consequences. Such analyses would probably sink the prospects of all the false positive diagnoses that are suggested for DSM-V and thus spare us from having yet another round of costly and dangerous iatrogenic epidemics.
Cite this as: BMJ 2010;340:c1168
Competing interests: The author is chair of the DSM-IV task force.
Provenance and peer review: Commissioned; not externally peer reviewed.