This online-first article will have minor typographical differences from the final, printed version.
Annals of Internal Medicine
The New Crisis in Confidence in Psychiatric Diagnosis
Allen Frances, MD
I n the early 1970s, 2 highly publicized studies showed
that psychiatric diagnosis, as it was then conducted, was
unreliable and inaccurate. The first found that British and
U.S. psychiatrists came to different diagnostic conclusions
when viewing the same patients on videotape (1). The sec-
ond found that healthy volunteers claiming to hear voices
were admitted to psychiatric hospitals for extended stays
despite subsequently acting normally (2). Was psychiatry
entitled to a place among the other medical specialties
when its diagnoses were so random? The response was
quick and effective. The Diagnostic and Statistical Manual
of Mental Disorders, Third Edition (DSM-III), published in
1980, featured definitions of mental disorders that, when
properly used, achieved reliability equivalent to that of
most medical diagnosis. The DSM-III stimulated an out-
pouring of psychiatric research. In most medical schools,
mental health research now ranks behind only internal
medicine in National Institutes of Health funding.
Unfortunately, the extensive research has had no effect
on psychiatric diagnosis, which still relies exclusively on
fallible subjective judgments rather than objective biologi-
cal tests. Brain complexity makes the translational step
from basic science to clinical practice more difficult in psy-
chiatry than in other fields of medicine. Biological find-
ings, however exciting, are never robust enough to become
test-worthy because within-group variability cancels out
between-group differences. We will be stuck with descrip-
tive psychiatry for the foreseeable future.
Psychiatric diagnosis is facing a renewed crisis of con-
fidence caused by diagnostic inflation. The boundaries of
psychiatry are easily expanded because no bright line sep-
arates patients who are simply worried from those with
mild mental disorders. The DSM-III opened the door to
loose diagnosis by defining conditions that were no more
than slightly more severe versions of such everyday prob-
lems as mild depression, generalized anxiety, social anxiety,
simple phobias, sexual dysfunctions, and sleep disorders.
The fourth edition of the DSM (DSM-IV), published
in 1994, tried to hold the line against further diagnostic
inflation by taking the conservative stance of discouraging
all changes and requiring substantial scientific evidence for
them (3). Of 94 suggested new diagnoses, the DSM-IV
added only 2, but this caution did not prevent the unex-
pected occurrence of 3 market-driven diagnostic fads. In
the past 20 years, the rate of attention-deficit disorder tri-
pled, the rate of bipolar disorder doubled, and the rate of
autism had a more than 20-fold increase (4). The lesson
should be clear that every change in the diagnostic system
can lead to unpredictable overdiagnosis.
This article was published at www.annals.org on 17 May 2013.
Downloaded From: http://annals.org/ on 07/14/2013
Ideas and Opinions
The DSM-5, the recently published fifth edition of the
diagnostic manual, ignored this risk and introduced several
high-prevalence diagnoses at the fuzzy boundary with nor-
mality. With the DSM-5, patients worried about having a
medical illness will often be diagnosed with somatic symp-
tom disorder (5), normal grief will be misidentified as ma-
jor depressive disorder, the forgetfulness of old age will be
confused with mild neurocognitive disorder, temper tan-
trums will be labeled disruptive mood dysregulation disor-
der, overeating will become binge eating disorder, and the
already overused diagnosis of attention-deficit disorder will
be even easier to apply to adults thanks to criteria that have
been loosened further.
These changes will probably lead to substantial false-
positive rates and unnecessary treatment. Drug companies
take marketing advantage of the loose DSM definitions by
promoting the misleading idea that everyday life problems
are actually undiagnosed psychiatric illness caused by a
chemical imbalance and requiring a solution in pill form.
This results in misallocation of resources, with excessive
diagnosis and treatment for essentially healthy persons
(who may be harmed by it) and relative neglect of those
with clear psychiatric illness (whose access to care has been
sharply reduced by slashed state mental health budgets)
(6). Only one third of persons with severe depression re-
ceive mental health care, and a large percentage of our
swollen prison population consists of true psychiatric pa-
tients with no other place to go. Meta-analysis shows that
the results of psychiatric treatment equal or surpass those
of most medical specialties (7), but the treatments must be
delivered to patients who really need them instead of being
squandered on those likely to do well on their own.
The DSM-5 did not address professional, public, and
press charges that its changes lacked sufficient scientific
support and defied clinical common sense. It was prepared
without adequate consideration of risk– benefit ratios and
the economic cost of expanding the reach of psychiatry just
when the field is about to achieve parity within an ex-
panded national insurance system (8). I found the DSM-5
process secretive, closed, and disorganized. Deadlines were
consistently missed. Field trials produced reliability results
that did not meet historical standards. I believe that the
American Psychiatric Association (APA)’s financial conflict
of interest, generated by DSM publishing profits needed to
fill its budget deficit, led to premature publication of an
incompletely tested and poorly edited product. The APA
refused a petition for an independent scientific review of
the DSM-5 that was endorsed by more than 50 mental
health associations (9). Publishing profits trumped public
interest.
© 2013 American College of Physicians 1
 This online-first article will have minor typographical differences from the final, printed version.
Ideas and Opinions The New Crisis in Confidence in Psychiatric Diagnosis
The APA has been responsible for the diagnostic sys-
tem for 100 years, having initially accepted the task when
it was too unimportant for anyone else to care. However,
the DSM has since acquired perhaps too much real-world
influence as the arbiter of who gets what treatment and
whether it will be reimbursed; who is eligible for disability
benefits, Veterans Affairs benefits, and school and mental
health services; and who qualifies to receive life insurance,
adopt a child, fly an airplane, or buy a gun.
New psychiatric diagnoses are now potentially more
dangerous than new psychiatric drugs. Diagnostic expan-
sions lead to drug company promotions that dramatically
increase the use of unnecessary medications, with high cost
and potentially harmful side effects. In the United States,
we carefully monitor new drug development but do not
have an effective system to vet the safety and efficacy of
new psychiatric diagnoses. The problems associated with
the DSM-5 prove that the APA should no longer hold a
monopoly on psychiatric diagnosis. Another mechanism
for revising the diagnostic system must be developed.
My advice to physicians is to use the DSM-5 cau-
tiously, if at all. It is not an official manual; no one is
compelled to use it unless they work in an institutional
setting that requires it. The codes needed for reimburse-
ment are available for free on the Internet (10).
From Duke University, Durham, North Carolina.
Potential Conflicts of Interest: Disclosures can be viewed at www
.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum⫽M13
-0997.
2 Annals of Internal Medicine
Downloaded From: http://annals.org/ on 07/14/2013
Requests for Single Reprints: Allen Frances, MD, Duke University,
PO Box 39950, Durham, NC 27710.
Author contributions are available at www.annals.org.
References
1. Kendell RE, Cooper JE, Gourlay AJ, Copeland JR, Sharpe L, Gurland BJ.
Diagnostic criteria of American and British psychiatrists. Arch Gen Psychiatry.
1971;25:123-30. [PMID: 5569450]
2. Rosenhan DL. On being sane in insane places. Science. 1973;179:250-8.
[PMID: 4683124]
3. Frances AJ, Widiger T. Psychiatric diagnosis: lessons from the DSM-IV past
and cautions for the DSM-5 future. Annu Rev Clin Psychol. 2012;8:109-30.
[PMID: 22035240]
4. Batstra L, Hadders-Algra M, Nieweg E, Van Tol D, Pijl SJ, Frances A.
Childhood emotional and behavioral problems: reducing overdiagnosis without
risking undertreatment. Dev Med Child Neurol. 2012;54:492-4. [PMID:
22571729]
5. Frances A. The new somatic symptom disorder in DSM-5 risks mislabeling
many people as mentally ill. BMJ. 2013;346:f1580.
6. Wang PS, Aguilar-Gaxiola S, Alonso J, Angermeyer MC, Borges G, Bromet
EJ, et al. Use of mental health services for anxiety, mood, and substance disorders
in 17 countries in the WHO world mental health surveys. Lancet. 2007;370:841-
50. [PMID: 17826169]
7. Leucht S, Hierl S, Kissling W, Dold M, Davis JM. Putting the efficacy of
psychiatric and general medicine medication into perspective: review of meta-
analyses. Br J Psychiatry. 2012;200:97-106. [PMID: 22297588]
8. Frances A. How Many Billions a Year Will the DSM-5 Cost? Bloomberg Web
site. 20 December 2012. Accessed at http://mobile.bloomberg.com/news/2012
-12-20/how-many-billions-a-year-will-the-dsm-5-cost-.html on 6 May 2013.
9. Division 32 Committee on DSM-5. The Open Letter to DSM 5 Task Force.
Coalition for DSM-5 Reform Web site. Accessed at http://dsm5-reform.com/the
-open-letter-to-dsm-5-task-force on 6 May 2013.
10. Centers for Disease Control and Prevention. International Classification of
Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). Atlanta, GA:
Centers for Disease Control and Prevention; 2012. Accessed at www.cdc.gov
/nchs/icd/icd9cm.htm on 6 May 2013.
www.annals.org
 This online-first article will have minor typographical differences from the final, printed version.

Author Contributions: Conception and design: A. Frances.
Analysis and interpretation of the data: A. Frances.
Drafting of the article: A. Frances.
Critical revision of the article for important intellectual content: A.
Frances.
Final approval of the article: A. Frances.
Administrative, technical, or logistic support: A. Frances.
Collection and assembly of data: A. Frances.

www.annals.org Annals of Internal Medicine W-1

Downloaded From: http://annals.org/ on 07/14/2013