Académique Documents
Professionnel Documents
Culture Documents
Stewart Justman
the nocebo effect
Copyright © Stewart Justman, 2015.
Softcover reprint of the hardcover 1st edition 2015 978-1-137-52328-0
®
First published in 2015 by PALGRAVE MACMILLAN in the United States—a
division of St. Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010.
Where this book is distributed in the UK, Europe and the rest of the world,
this is by Palgrave Macmillan, a division of Macmillan Publishers Limited,
registered in England, company number 785998, of Houndmills, Basingstoke,
Hampshire RG21 6XS.
Palgrave Macmillan is the global academic imprint of the above companies and
has companies and representatives throughout the world.
® ®
Palgrave and Macmillan are registered trademarks in the United States, the
United Kingdom, Europe and other countries.
ISBN 978-1-349-57677-7 ISBN 978-1-137-52329-7 (eBook)
DOI 10.1007/978-1-137-52329-7
Acknowledgments vii
Preamble: Indefinite Ailments and Inflammatory Messages ix
Socrates, Asclepius concerned himself only with persons who had “a defi-
nite ailment.” A lot of the ills that disturb us even in health are indefinite,
though they can acquire a semblance of specificity by taking on clinical
names. It may indeed be unwise, as Socrates suggests, to fixate on one’s
ills—not, however, because there’s nothing there but because there is, and
because symptoms can be magnified by the interpretations imposed on
them by doctors and patients.
In order to figure in the marketplace, many of the ills troubling our
minds and bodies have to lose the ambiguity that seems to belong to them.
“Compared with the reality which comes from being seen and heard,”
wrote Hannah Arendt, “even the greatest forces of intimate life—the pas-
sions of the heart, the thoughts of the mind, the delights of the senses—
lead an uncertain, shadowy kind of existence unless and until they are
transformed, deprivatized and deindividualized, as it were, into a shape
to fit them for public appearance.”1 Still more obscure are complaints so
potentially indeterminate that we might report them in different ways—as
chronic fatigue or fibromyalgia, for example—depending on what diagno-
sis happens to be in favor, or we might not report them at all. Precisely
because such internal events lack the confirmed character of things seen
and heard in public, they are highly subject to interpretation and, in fact,
distortion. The inside of our being is the ultimate Platonic cave of shadows.
When problems of mind and body are brought into the public realm, sub-
jected to the shadowless light of the media, and made the topic of campaign-
ing, controversy, and salesmanship, ill-defined events to which there may be
no witness but oneself are transformed into a cause célébre. In recent decades,
many common problems have been elevated into medical issues in this way,
in the process acquiring both evocative names and large constituencies. Ills as
well as goods can be shaped from raw materials, packaged, and popularized.
But once ambiguous ills inherent in human existence are labeled—branded—
as a consequence of medicalization, the experience of having them in the first
place can change. Waters and winds cohere into storms.
Once having entered the marketplace, ideas about illness are capable
of stirring up illness itself through the power of the nocebo: the neglected
twin of the placebo effect.
I will argue that medicalization promotes harm under the auspices of heal-
ing, both by marketing disorders (“fostering the same pathologies intended
to be healed”) and distorting the calculation of harms and benefits.
The quoted author concludes that in order to avoid triggering nocebo
effects, “We must be cautious in both public health communications and
in clinical medicine.” Ironically, in the same year in which this admoni-
tion was issued, direct-to-consumer advertising became legal, and there
ensued a flood of suggestive messages urging us to consider the possibility
that we might be sick. In addition to such ads, messages that simply vio-
late caution are among the power sources of the medicalization movement.
Announcements that populations of staggering magnitude suffer from
undiagnosed disorders—that one in twenty may have bipolar disorder (a
figure five thousand times the prevalence of what used to be called manic-
depressive illness);3 or that one in eight Americans suffers from social anxi-
ety disorder;4 or that as much as 24 percent of the population suffers from
“depressive symptoms in various combinations”;5 or that a quarter of the
population of New York City was in need of psychological treatment fol-
lowing the 9/11 attacks;6 or that 43 percent of all women suffer from sexual
dysfunction;7 or that half of the US population will at some point meet
the requisites of a psychiatric disorder;8 or that half of American men are
“sexually dysfunctional”;9 or that “as many as 72% of people in the work-
force are depressed”;10 or that “three-quarters of the general public will
experience an event that could cause a traumatic response sometime in
their lifetime”;11 or that 82.5 percent of young people will qualify for a psy-
chiatric diagnosis by the age of 21;12 or that depression will soon be the
second leading cause of disability around the world13—announcements
like these, the very opposite of cautious, provide the themes, banners, and
rallying cries of medicalization. Inevitably quoted again and again, their
exaggerations building on one another, they are symptoms of an informa-
tion epidemic that itself defies all limits.
1
Not until the publication of its third edition in 1980 did the DSM attain
its status as the final arbiter, the bible, of mental disorders; before that it
was a little-known, spiral-bound document reflective of the psychoanalytic
assumptions then in the ascendant. The gulf between DSM-II and DSM-
III might be measured by the difference between “narcissistic personality
disorder” and “Major Depression,” the former a diagnosis popular in the
1970s but not listed in DSM-II,4 the latter a diagnosis popular ever since,
anchored in the chapter and verse of DSM-III and its successors. It was
to abolish obscure theorizing about the origins of psychological problems
and to put diagnosis on a solid foundation that DSM-III introduced the
system of tabulating symptoms that reigns to this day. According to this
scheme, a symptom isn’t the manifestation of a problem deeply rooted in
the patient’s early history, as in Freud, but simply evidence of a disorder—
a disorder, not a neurosis. Streamlined by comparison with the cumber-
some machinery of Freudianism, the DSM system possessed an appealing
straightforwardness and a how-to emphasis that recommends itself to
practical minds as Freudian doctrine never could. The product of a zeal for
renewal, DSM-III was American psychiatry’s Reformation. Said a member
of the DSM-III Task Force, “A lot of icons were being smashed.”5
In the words of its designer, Robert Spitzer, DSM-III gave psychia-
try “a fresh start.”6 Animated by this spirit of reform, the framers of the
document eliminated the anarchy that allowed clinicians using diverse
theories to arrive at conflicting diagnoses, and in its place installed a set
of explicit standards written in clear language, without esoteric suppo-
sitions, all intended to bring different observers to identical diagnostic
conclusions. It bears emphasis that precisely as an exercise of system-
building, DSM-III was inspired by a rejection of the confused state of
existing psychiatric judgments. A few years before work started on DSM-
III, a study was published in which young psychiatrists “were no more
likely to agree with an examiner’s diagnosis of a patient than would be
expected by chance.”7 Reporting to President Carter in 1978, the Com-
mission on Mental Health acknowledged that “opinions vary on how
mental health and mental illness should be defined.”8 For a discipline
either claiming or aspiring to the status of science, such bedlam was
intolerable. At the same time, insurance companies and powerful figures
in Washington let it be known that they had no faith in psychiatry’s abil-
ity to explain and defend its findings. In this state of affairs, DSM-III
served to restore and even enhance the credit of psychiatry by acting as
a manual for the making of diagnostic judgments. Over the years since
1980 the edition of the DSM that happens to be in force has been the last
word on the diagnostic criteria of mental disorders, its influence incal-
culable and its authority cited in the medical literature, courtrooms, and
DSM AND THE SHAPING OF DEPRESSION 3
elsewhere. And so, from the chaos of conflicting opinions rose the reli-
able judgments of the DSM system.
The priority of the need to replace capricious judgments with systematic
ones is confirmed by Spitzer, who in 1999 wrote:
The assumption here, the same assumption that seems to have driven the
construction of DSM-III, is that systematizing the diagnosis of mental dis-
orders would lead to better outcomes in the form of fewer false positives
and negatives. Fewer false positives? Far from reducing false positives, DSM-
III so multiplied them that Spitzer’s successor Allen Frances, chair of the
DSM-IV Task Force, indicts the document he inherited on that very ground:
“Diagnostic inflation has been the worst consequence of DSM-III.”10 Indeed,
Spitzer himself has commended a book that deplores the overdiagnosis of
depression as a direct result of the DSM diagnostic system.11 Evidently it’s
possible to systematize the diagnosis of a disorder like depression (first codi-
fied in DSM-III, and now the most researched of all psychiatric conditions)
while not only failing to rein in, but fueling, bad diagnoses. Even though the
DSM-III diagnostic scheme was framed on a medical model in accordance
with the desire to return psychiatry to its identity as a medical discipline,
somehow the first principle of medicine—avoiding harm—was overshad-
owed by the imperative of reducing “criterion variance.”
What if many doctors arrive at the same inflated diagnosis? (Inflation
is a collective event, after all.) The likelihood of inflation was wired into
DSM-III if only because its categories were used to determine the preva-
lence of disorders in the community at large, and those findings were cited
in turn by the American Psychiatric Association (APA) in lobbying for “the
necessary resources.”12 It’s diagnostic inflation that makes possible vast
markets for psychoactive drugs addressed to DSM-defined disorders, such
as depression, social anxiety disorder, and attention-deficit/hyperactivity
disorder (ADHD), the latter two DSM coinages. If diagnoses like these are
now spreading around the world “with the speed of contagious diseases,”13
4 THE NOCEBO EFFECT
A Diagnostic Catch-All
The makers of the DSM may blame the drug industry for overselling dis-
orders, but without the authority of the DSM the disorders couldn’t have
been diagnosed en masse in the first place.14 And the DSM’s authority
is inseparable from its seemingly precise criteria, its itemized specifica-
tions. The lawyerly nature of the diagnosis of depression in particular was
brought out by one of DSM-III’s framers who remarked, not altogether in
jest, that “the diagnosis of a depressive episode . . . is a sentence full of sub-
ordinate clauses and other grammatical intricacies.”15 In part just because
so many elements go into the making of a diagnosis, the text of the DSM
allows for more depression than its appearance as a checklist of require-
ments might suggest. The criteria of a depression diagnosis have changed
little over successive editions of the DSM; those in the most recent edition
appear in Table 1.1.
Built into the criteria for Major Depressive Disorder are low thresh-
olds, alternative entry conditions, and many symptoms common in a
healthy population, as well as a number of particulars that can easily
be forgotten, bent, or waived in practice, as is only fitting for a bible.
Requiring just two weeks of symptoms, the DSM criteria will catch the
common transient distress that would otherwise resolve spontaneously
within four weeks.16 In the template on which the DSM criteria are based,
and which had already been cited perhaps a thousand times in the medi-
cal literature—the Feighner criteria—the symptom period for depressed
mood is “at least one month.”17 (Note too that depressed mood itself isn’t
necessary for a DSM diagnosis of depression.) Though the Feighner cri-
teria for depression, for their part, are patterned on others published in
1957, they omit constipation because “it lacked specificity—many con-
stipated people are not depressed.”18 No more specific than constipation
are a number of symptoms included in both the Feighner and the DSM
criteria for depression, such as fatigue and faulty concentration. In a
study conducted by Spitzer and others in 1994, fully 58 percent of 1,000
primary-care patients reported fatigue on a questionnaire.19 As suggested
by frequent reminders in the medical literature that depression disguises
itself as common complaints, the diagnostic weight accorded such com-
plaints in the DSM has real implications. In addition to the two-week
DSM AND THE SHAPING OF DEPRESSION 5
A. Five (or more) of the following symptoms have been present during the same two-week
period and represent a change from previous functioning; at least one of the symptoms
is either (1) depressed mood or (2) loss of interest or pleasure.
Note: Do not include symptoms that are clearly attributable to another medical
condition.
1. Depressed mood most of the day, nearly every day, as indicated by either subjective
report (e.g., feels sad, empty, hopeless) or observation made by others (e.g., appears
tearful). (Note: In children and adolescents, can be irritable mood.)
2. Markedly diminished interest or pleasure in all, or almost all, activities most of the
day, nearly every day (as indicated by either subjective account or observation).
3. Significant weight loss when not dieting or weight gain (e.g., a change of more than
5 percent of body weight in a month), or decrease or increase in appetite nearly
every day. (Note: In children, consider failure to make expected weight gain.)
4. Insomnia or hypersomnia nearly every day.
5. Psychomotor agitation or retardation nearly every day (observable by others, not
merely subjective feelings of restlessness or being slowed down).
6. Fatigue or loss of energy nearly every day.
7. Feelings of worthlessness or excessive or inappropriate guilt (which may be
delusional) nearly every day (not merely self-reproach or guilt about being sick).
8. Diminished ability to think or concentrate, or indecisiveness, nearly every day
(either by subjective account or as observed by others).
9. Recurrent thoughts of death (not just fear of dying), recurrent suicidal ideation
without a specific plan, or a suicide attempt or a specific plan for committing
suicide.
B. The symptoms cause clinically significant distress or impairment in social, occupational,
or other important areas of functioning.
C. The episode is not attributable to the physiological effects of a substance or to another
medical condition.
Note: Criteria A–C represent a major depressive episode.
Note: Responses to a significant loss (e.g., bereavement, financial ruin, losses from a
natural disaster, a serious medical illness or disability) may include the feelings of intense
sadness, rumination about the loss, insomnia, poor appetite, and weight loss noted in
Criterion A, which may resemble a depressive episode. Although such symptoms may be
understandable or considered appropriate to the loss, the presence of a major depressive
episode in addition to the normal response to a significant loss should also be carefully
considered. This decision inevitably requires the exercise of clinical judgment based
on the individual’s history and the cultural norms for the expression of distress in the
context of loss.
D. The occurrence of the major depressive episode is not better explained by schizoaffective
disorder, schizophrenia, schizophreniform disorder, delusional disorder, or other
specified and unspecified schizophrenia spectrum and other psychotic disorders.
E. There has never been a manic episode or a hypomanic episode.
Note: This exclusion does not apply if all of the manic-like or hypomanic-like episodes
are substance-induced or are attributable to the physiological effects of another medical
condition.
6 THE NOCEBO EFFECT
Expanded Eligibility
In the depressive states of the disease the emotional attitude is regularly that
of gloominess, despair, doubt, and anxiety. Patients complain particularly
of the loss of interest in things; “everything is the same to them,” “they are
DSM AND THE SHAPING OF DEPRESSION 9
desolate and empty,” “they are dead, because they have no feeling,” “music
does not sound natural,” and “the crying of the children no longer creates
sympathy.” They feel as if they no longer belong to this world . . . The psy-
chomotor field in the depressive form presents a retardation of activity . . .
In the mildest degree this retardation appears as a deficiency in the power of
resolution. Actions may not only be performed slowly, but even after being
started may fail of completion. The simplest movements, such as walking
and talking, are performed very slowly and without energy . . . Some patients
are so taciturn and monosyllabic that it is impossible to engage them in con-
versation, and although they are able to count or read aloud as rapidly as
ever, they will sit for hours with a letter in front of them, unable to finish
writing it.43
To begin with, consider the string of options in the crucial first criterion
of depression in DSM-III:
Dysphoric mood or loss of interest or pleasure in all or almost all usual activ-
ities and pastimes.
The word “or” occurs three times, with the effect of expanding eligibility.
No longer is depressed mood necessary for depression, and for that matter
you can lose either interest or pleasure, and you don’t have to lose one or
the other in all things, but just most. (Recall too that the DSM halves the
symptom period required by the Feighner criteria.) By providing alterna-
tive entry criteria for depression—depressed mood or loss of interest or
loss of pleasure—DSM allows three options instead of one, the result being
a test so loose that as many as half of screened patients will score posi-
tive.48 The entry criteria therefore need to be supplemented. At this point
another change comes into play: The number of secondary symptoms
needed for DSM depression is lowered from six to four. It should be noted
that symptoms in this category, such as poor concentration, insomnia,
and fatigue—and insomnia virtually implies fatigue—are common in the
general population and therefore of questionable diagnostic significance.
(Thus, in a survey of healthy students conducted a dozen years before
the diagnosis of Major Depression was codified in DSM-III, 27 percent
reported “inability to concentrate.”)49 Such generic complaints are now the
most frequent presenting symptoms of depression, or, rather, presumed
depression. Yet while common problems are interpreted as symptoms of
depression, the same problems are discounted when produced by the drugs
called antidepressants.50 (Thus, sleep disturbances and fatigue, both DSM
symptoms of depression, are listed by Eli Lilly as common side effects of
Prozac.)51 It’s as if secondary symptoms lose their significance in the eyes
of medicine once they do their job of boosting the diagnosis of depression,
even though interpreting “unexplained” symptoms as evidence of a mental
disorder conflicts with the principle of avoiding speculation that inspired
the diagnostic revolution in psychiatry. Just as the neo-Kraepelinian DSM
enables millions to have the depressive disorder portrayed by Kraepelin
as far from ordinary, the DSM movement adapts to the entire population
diagnostic criteria that were met by “probably the sickest of the whole
group” of 248 psychiatric patients in the Cohen study.
Beginning by identifying manic-depressive disease as “a serious cause
of disability and even death,” the Cohen study later notes that many with
this disease are “merely patients who are tired and have insomnia, head-
ache, and nervousness.”52 Spanning a range of conditions,53 its causal
mechanism unknown but the subject of learned speculation, established
DSM AND THE SHAPING OF DEPRESSION 11
by shifting criteria and thresholds, depression defies its own label as a spe-
cific disorder. A convincing argument holds that patients who would have
identified themselves as anxious a few decades ago, when anxiety ruled
the day, now call themselves depressed because depression happens to be
in the ascendant for marketing reasons.54 (If this is so, then the intimate
experience of being unwell is much more open to social influences than we
may suppose.) In the Cohen study subjects were given over a dozen terms
for “depressed,” including not just “worried” and “despondent” but “angry”
and “disgusted,” and counted as depressed if they checked any of them. The
only thing all seem to have in common is that they lie off the beaten path
of normative cheerfulness. So diagnostically equivocal is the entity reified
as depression that nothing jumps out at us in the statement, “Symptoms
like anxiety, depression, fatigue, and sleep disturbances seem to be found
in many kinds of patient, whatever their physical or psychiatric status.”55
Nothing jumps out even though depression is listed as a symptom and not
a specific disorder with symptoms of its own, including all the other symp-
toms on the same list.56
For many patients with a collection of generic symptoms, depression
acts as a diagnostic net to gather all together. While DSM-III overthrew
the concept of neurosis dear to the Freudians in the name of scientific diag-
nosis, the fact is that the most common symptoms of neurosis in Freud’s
time—insomnia, fatigue, dyspepsia, and the like57—became the most com-
mon symptoms of depression in the age of the DSM. As this may suggest,
the makers and inheritors of the DSM revolution reserved the right to
interpret physical symptoms as indices of a mental disorder even after the
destruction of the Freudian temple.58 Neurosis as conceived by Freud was
a kind of illness continuous with normality, and the same is true of all but
severe depression; but the DSM enabled the production of diagnoses on a
scale beyond the reach of the Freudians, one as large as the marketplace.
This dramatic expansion of diagnosis couldn’t have taken place unless the
protean neurosis (protean because a symptom can signify anything) had
yielded to the codified disorder.
In an intriguing study of the influence of placebos on insomnia con-
ducted a decade before DSM-III, subjects treated with pills with a sup-
posedly arousing effect fell asleep more readily than subjects told the pills
were sedating, evidently because being able to attribute their own aroused
state to a pill eased their mind. Many people, it seems, are “troubled by
their insomnia, taking it as evidence of a major physical or psychological
disorder,” an inference that sets up “a vicious cycle in which symptoms of
insomnia elicit worrisome thoughts that further aggravate the insomnia.”59
The interpretive system installed by DSM-III invites the doctor to construe
insomnia as, in fact, evidence of disorders including Major Depression,
12 THE NOCEBO EFFECT
Skewed Calculations
A half century ago it was recognized that many of the ills that send patients
to the doctor “cannot be labeled as ‘diseases.’ They are patients’ problems,
concerns, complaints, symptoms, and assorted ‘conditions,’ including a
wide variety of social and psychological problems that are the day-to-day
fare of general physicians.”60 Over the intervening decades, many of these
ills have been promoted to the category of disorders and given labels that
have been taken up by patients themselves.
Not just the DSM but what has come to be known as medicalization—
the framing of normal conditions as medical issues—now reaches into all
corners of private and public life, and those who execute its mandate can be
confident that their hunt for ills to treat will be successful, simply because
what they’re searching for is everywhere. If you’re looking for symptoms of
postconcussive syndrome, for example, you’ll find them even in a popula-
tion of healthy college students who have never suffered a head injury.61
In the pages to come I document the crusade against normality, arguing
that human life abounds with raw material for medicalization to mold
into disorders, and that the shaping process produces harm in violation
of medicine’s first principle. Much as the DSM system has defeated its own
goal of curtailing bad diagnostic practices, medicalization in general has
disordered the calculation of risks and benefits. To begin with, the ben-
efits of drugs decline as they are prescribed to populations less unwell than
those on whom they are first tested, which is exactly the pattern seen in
practice over recent years. “New medical interventions tend to be studied
in severely ill patients where significant benefits can be expected. After a
therapy is established, physicians tend to broaden its use and prescribe it
to a wide range of patients, including a high number of less sick patients.”62
Medicalization targets the vast market of the less sick, straining the ratio
of risks and benefits precisely by treating those who don’t stand to enjoy
“significant benefits.”
However, medicine has extended its writ not only over the less sick but
many not sick at all, such as patients suffering from normal distress. Con-
sider the case of a woman whose diagnosis of depression appears not to be
based on DSM-IV, which was in effect at the time.
earlier. She had hoped that her physician would either offer her time in the
consultation to process her sense of shock and validate her feelings or refer
her to a therapist attached to the practice. Instead, her physician suggested
that she should start a course of antidepressants. She accepted the antide-
pressant prescription without expressing her preference for psychological
therapy, but left the consulting room having already decided that she would
not redeem the prescription. She did not understand how antidepressants
could help her to come to terms with her unexpected bereavement. From her
observation of clients on long-term treatment for severe depression, she per-
ceived that one of the adverse effects of antidepressants was reduced energy
and motivation to deal with problems.63
One has the impression that the doctor in this instance didn’t weigh pros
and cons but came to an automatic conclusion, leaving the drawbacks of
antidepressants to the patient. Yet the patient wasn’t depressed but “in
shock,” and for a good reason. Maybe the doctor thought the dulling effect
of an antidepressant would be a balm for someone in shock, but that isn’t
what the patient was looking for; in fact, if she had presented with “reduced
energy and motivation to deal with problems” she might well have been
diagnosed as depressed. The study from which I’ve taken this case (actually
a composite) never questions the diagnostic fashions that lead to the over-
prescription of antidepressants, though it does comment that giving anti-
depressants to the mildly depressed results in a poor risk/benefit ratio.64
Inasmuch as SSRIs are negligibly superior to placebo in treating depression
in most cases, determining the ratio of risks to benefits for an overdiag-
nosed disorder like depression comes to resemble dividing by zero. As if
the benefits of SSRIs and similar drugs did in fact compute to nothing, a
comparison of large surveys conducted in Britain in 1993 and 2000 showed
that “widespread increased prescribing of psychotropic medication”—led
by SSRIs—“has not improved the mental health of the nation.”65
In addition to treating milder forms of disease and even conditions
indistinguishable from normality, medicalization seeks out disease in its
incipient stages, before it has had time to manifest itself. As a result of the
early-detection imperative, “common problems will be identified in many
individuals who would not be harmed by the disease and, therefore, would
not benefit from treatment.”66
One such problem is prostate cancer. Early in the era of prostate-cancer
screening, proponents “dismissed concerns about harms”67 and argued
that there was no time to wait for evidence of benefits to emerge from
clinical trials, as if the principle of weighing harms and benefits simply
didn’t apply to the practice of screening millions for a highly ambiguous
cancer.68 Around the same time, proponents of mass screening for coro-
nary risk factors made the assumption that such a program could benefit
14 THE NOCEBO EFFECT
but could not harm.69 The drive to screen for depression shows the same
one-sidedness; thus, the US Preventive Services Task Force found no data
on the harms of screening adults despite the patent risk of labeling nor-
mal distress as a mental disorder,70 among the many pitfalls and shortfalls
of depression screening. (I will propose an analogy between the crusade
against prostate cancer and that against the “psychological cancer,” depres-
sion.) Clearly, unless the costs of medicalization were somehow effaced,
it would be difficult to get people to buy into the broadening of medical
definitions and boundaries. Yet those costs may also escape the notice of
those who impose them. “Downstream harms from overtesting and over-
treatment may be completely invisible to clinicians,” because of their belief
in their own beneficence, the delayed onset of the harms, or both.71
A measure of the inattention to harm in a climate of medicalization is
the scant information about adverse effects of drugs in published papers.
In 2001, four years after DTC advertising revolutionized the marketing of
drugs, an analysis of safety-reporting in a broad array of drug trials assessed
the provided information as “largely inadequate,” with approximately the
same amount of print space devoted to the authors’ names and affiliations
as to the adverse effects of the drugs under study.72 Trials of drugs used to
treat mental illness follow the same pattern. “Even with lenient criteria,
very few . . . have adequate reporting of clinical adverse events.”73 How
to balance risks and benefits without good information about the risks?
So it is that doctors routinely strain risk/benefit calculation by prescrib-
ing highly promoted drugs whose utility profile remains unknown in the
absence of good information (and prescribing them in decreasingly severe
cases). “Only when more adequate types and numbers of patients are stud-
ied for sufficiently long periods can a more accurate profile of their risks
and benefits emerge.”74 Given, too, that the prescription of psychoactive
drugs is “always a fine balancing act”75 owing to individual differences, the
sometimes paradoxical character of drug effects, and our ignorance of the
actual (in contrast to presumed) causes of mental disorders, the mass pre-
scription of these drugs in and of itself violates the balancing of risks and
benefits. A searching critique of the use of psychoactive drugs as putatively
specific agents concludes that belief in them “has led to their indiscriminate
prescription to millions of people often for decades on end. It is likely that
many people exposed to the harmful physical and psychological effects of
these drugs derive no benefit from them.”76 However, if their prescription
really is indiscriminate—as the evidence suggests—then it’s not just likely
but very likely that for most, not just many, of those who take psychoactive
drugs there are no benefits to set against harms.
In that the overprescription of psychoactive drugs has taken place
under the auspices of the DSM, the ascendancy of the DSM system has
DSM AND THE SHAPING OF DEPRESSION 15
distorted the calculation of risks and benefits. Consider the case of the
DSM-authorized disorder ADHD (originally ADD). In that the safety and
efficacy of the medications used to treat ADHD are established in brief
clinical trials while the drugs are actually used for years on end, the mass
prescription of stimulants represents an experiment in itself. Of course, it
could be said of many other drugs that their long-term use is underwrit-
ten by short-term trials. In this instance, though, the drugs are adminis-
tered to children as young as two, unlikely to be represented in any trial
population.77 An investigator of the prescription of psychoactive drugs
to preschoolers reports that while 55 percent more of these children were
diagnosed with behavioral disorders in 2009 than in 1994, only 29 percent
of those diagnosed were given drugs (as against 43 percent in 1994). “We
were very pleased to see that the rate of psychotropic drug use in this age
group isn’t going steadily up each year,” said the investigator. “But we are
still giving these drugs to young children, so we need more research into if
and how they influence the developing brain.”78 Virtually by definition, it
defies medical prudence to prescribe psychoactive drugs to two-year-olds
even while their effect on the brain remains unknown.79 Medical prudence
involves not just totaling pros and cons in some fashion but weighing risks
and benefits in the light of the particular duty to avoid harm incumbent on
a doctor as a doctor.80
And in the case of ADHD, the balance of the evidence is particu-
larly disturbing. Though the DSM-II precursor of ADHD (Hyperkinetic
Reaction of Childhood) was thought to subside by adolescence,81 and
though nearly 90 percent of ADHD cases appear to be mild to moderate,
long-term outcomes of ADHD are dismal, with diagnosees at markedly
higher risk of dropping out of school, early pregnancy, drug abuse, auto
accidents, being fired, incarceration, even death.82 (In a study that fol-
lowed an ADHD population for 33 years, diagnosees died at a rate 2.5
times higher than controls. Subjects exhibiting aggression or antisocial
behavior were excluded from the study.)83 If a drug showed results like
these it would be pulled from the market; if a trial yielded such results, it
would be stopped. How do outcomes so alarming follow from a disorder
whose symptoms can be as trivial as fidgeting84 and whose cases fall over-
whelmingly into the mild-to-moderate range?85 Either a disorder ques-
tionably distinct from normality86 is inherently fraught with great risk or
it becomes so as a result of a process set in motion by diagnosis itself. The
latter possibility—that diagnosing ADHD can harm, if only by cuing the
behavior of the child and the expectations of others87—doesn’t seem to
enter into the judgments that support the diagnosis. As a result of what
has been called “diagnosis threat,” people whose attention is called to
cognitive deficits supposedly associated with mild head injury perform
16 THE NOCEBO EFFECT
worse on various tests than people with the same history not cued to do
poorly.88 The ADHD label has all the automatic associations and evoca-
tive power of a stereotype, with the authority of medicine to boot, and it
would be strange if such a potent influence had, in fact, no influence on
the labeled child.89 Children are not notably invulnerable to the power
of suggestive messages. A diagnosis that tells children there’s something
radically wrong with their wiring, such that they need a drug to regulate
themselves, can leave them “less well-equipped to draw upon their own
resources to solve their problems.”90 That the neurological cause of the
child’s theorized defect of self-regulation is also theoretical doesn’t keep
the notion of such a cause from being powerful. People caught up in mass
psychogenic outbreaks experience illnesses whose causes, such as a toxic
gas, they are fully convinced of even if they can’t be found.
In laying out the hazards of the DSM system, and DSM-V in particu-
lar, the architect of DSM-IV (which broadened the criteria for ADHD)
reminds both primary-care doctors and psychiatrists to “conduct a risk-
benefit analysis” before making a diagnosis. “In toss-up situations, weigh
the pluses and minuses of giving the diagnosis,” writes Allen Frances. “The
basic question boils down to, ‘Is this diagnosis more likely to help or more
likely to hurt?’”91 Only if clinicians under the influence of the DSM sys-
tem had somehow forgotten such elementary principles would they need
to be reminded of them. Frances concludes that the best that can be said of
DSM-IV is that it didn’t make the DSM system even worse than it already
was.92
Writing in 2009, the chair of the DSM-V Task Force held out the hope
that “Mental disorder syndromes will eventually be redefined to reflect
more useful diagnostic categories (‘to carve nature at its joints’) as well
as . . . clear thresholds between pathology and normality.”93 It’s a sobering
thought that after 30 years of DSM hegemony and the writing of billions
of prescriptions, the boundary between normality and disease remains so
nebulous to psychiatry that its clarification is postponed into the indefinite
future, as something that will “eventually” come about. If medicalization
means the definition and treatment of normal conditions as medical prob-
lems, this statement of hope amounts to a confession that the DSM has
contributed in a large way to exactly that. But in addition to the overpre-
scription of drugs with questionable harm/benefit profiles, medicalization
poses a subtler risk: the risk that the suggestiveness of the diagnoses given
to common problems will color people’s understanding of themselves and
even mold their experience.
Suppose a team of psychologists wants to test the theory that people
receiving a high blood-pressure reading will notice more of the symptoms
they associate with high blood pressure. Doing what experimenters do,
DSM AND THE SHAPING OF DEPRESSION 17
the team devises a study where half the subjects are deliberately given false
high readings and then asked to report the occurrence of symptoms like a
fast heartbeat, headache, and shakiness. Such a study was in fact conducted
(using college students as the guinea pigs), and it confirmed the theory in
question, but for our purposes the point is that the experimenters made
sure the hoax didn’t last long and the subjects were fully debriefed after-
ward.94 The experimenters seem to have been keenly aware of the risks
of feeding false medical information even to a population proverbially
invulnerable to anxiety over health. They point out that they didn’t lead
the subjects to believe they had hypertension and didn’t allow them to
“carry . . . false information with them for several hours or days,” as “such
a procedure would be unethical.”95 If they had given the false information
time to really sink in, it could have changed the subjects’ understanding of
themselves; it could also have abetted the stress that the subjects attributed
their supposedly elevated blood pressure to in the first place.
My point, of course, is that being diagnosed with a mental disorder
when you’re normal is like receiving false information about your blood
pressure, except that in this case the false information comes as a result of
“diagnostic exuberance,”96 it isn’t retracted, it has all the time it needs to
sink in, it necessarily changes your self-understanding, and for all these
reasons it’s more insidious.
Noxious Diagnoses
so it is, for example, that Americans report chronic neck pain after minor
car accidents not because the collisions necessarily cause such pain—in
other societies comparable collisions may cause none—but because we
generally expect it. (The authenticity of the experience bears note: While
some may fake their pain, malingerers are the exception, not the rule.)99
But if our very experience is subject to the power of expectations, those
packed into a suggestive label like “depression” could change the experi-
ence of patients who receive an inflated diagnosis—all the more because
the diagnosis has both medical legitimacy and social currency. Many of
those mislabeled with medical, including psychiatric, disorders can and
will discover corresponding symptoms. According to Spitzer, it’s not just
possible but common for members of the general population to have
symptoms of certain popular disorders without having the disorder.100 In
cases like these an inflated diagnosis could have the perverse effect of rein-
forcing the symptoms.
The speed with which new DSM diagnoses acquire constituencies and
pass into common speech points all too clearly to the document’s power
to produce belief. In addition, “a large proportion of the population” given
an identical diagnosis may generate or be swept up in a trend whose social
power alone makes the diagnosis credible. Tagged with a diagnosis persua-
sive in all these ways and more, patients falsely found or presumed to be
depressed may conclude that something is wrong with their brain, or that
their ability to weather distress is lacking, or that their state is destined to
recur or even worsen unless treated, all of which are depressing thoughts.
(“Don’t worry, it’s clinical depression,” a GP told one of his patients.)101
Diagnostic labels, even false ones, have power. They not only excite belief
but shape understanding and guide the search for confirming signs.102 So
it is that normal people diagnosed as depressed because they are going
through the adversity incident to human life “might . . . be more likely to
see themselves not as normally stressed but as ill . . . and in general play
out the role of a limited person assigned to them by a medical authority.”103
Particularly where the pursuit of happiness is confused with a kind of right
to happiness, those who find themselves unhappy may well assume it’s the
fault of a biochemical problem.
There are patients who suffer from symptoms that can’t be pinned down
to any specific cause but which they ascribe to a disorder that’s a hot topic
at the moment (say chronic fatigue). Like the attribution of insomnia to
an underlying mental disorder, such an ascription can heighten symp-
toms by giving them a potent, suggestive explanation. In cases of this kind
medical commentators speak of normal symptoms mutating into some-
thing unusually distressing as a result of a process of interpretation gone
awry. Yet because of the inherent ambiguity of internal signals, anyone and
DSM AND THE SHAPING OF DEPRESSION 19
there were some sixteen thousand articles on PTSD in the medical litera-
ture,111 and the term had long since entered popular lore.) A number of
official symptoms of PTSD, including disturbed sleep, faulty concentra-
tion, diminished interest, and guilt, are in fact also official symptoms of
depression; other symptoms overlap other disorders, and even core symp-
toms appear in healthy populations.112 In a word, the components of PTSD
appear to be nonspecific. By organizing such symptoms of distress into
a highly connotative disorder and playing down the natural capacity for
recovery, it’s possible to aggravate the symptoms themselves. As noted in
a duly critical account of the rise of the PTSD diagnosis, several studies
have shown that patients who underwent intensive treatment for PTSD
deteriorated as a result.
Cued Expectations
Those who had the sham procedure fared as well as their counterparts.
It would be a mistake, though, to suppose the effects of suggestion are
confined to the operating theater or the laboratory. A review of studies
of negative suggestion concludes fittingly that both “health warnings in
western societies” and diagnoses themselves can engender effects simi-
lar to laboratory manipulations.116 We have every reason to believe that
a diagnosis at once popular and highly suggestive, like depression or
ADHD, can have an effect of its own whether or not its evidentiary basis
is solid. While the sights and sounds of the operating room make for a
uniquely convincing kind of drama, a child diagnosed with ADHD and
medicated accordingly is immersed in social theater indistinguishable
from life itself.
If the DSM diagnostic system is influential enough to be recognized
by courts, governments, and institutions of medicine, it can also color a
patient’s self-understanding. Over the years the DSM system has been in
place, it has become common for patients entering psychotherapy to bring
a diagnostic label with them (“I have bipolar”), suggesting it has seeped
into their way of thinking just as it entered common speech in the first
place.117 And with the population at large now considered as psychiatry’s
constituency, the influence of diagnostic labels extends far beyond those
patients interested in a psychological alternative or complement to phar-
macotherapy. The founder of the DSM system concludes that it “serves to
define . . . how individuals themselves interpret their emotional experi-
ences,”118 which leads inescapably to the conclusion that someone who
receives an inflated DSM diagnosis can come to believe it.119 In an inter-
view a few months before his death, Mandel Cohen observed that once the
diagnosis of ADHD was codified in the DSM, it acquired an illusory reality
and attracted an inordinate number of cases. “Once you get the initials, the
condition solidifies”120—a comment that points to the power of disorders,
especially those with catchy labels, to captivate both the professional and
public imagination. ADD (Attention-Deficit Disorder) made its appear-
ance in DSM-III and solidified even though the 1987 revision of DSM-III
conceded that ADD without hyperactivity was a construct so question-
able that “research is necessary to determine if this is a valid diagnostic
category and, if so, how it should be defined.”121 Thus, from 1980 to 1987
it was possible to be diagnosed with a disorder whose validity—that is,
22 THE NOCEBO EFFECT
Mrs. B’s story shows what happens when physicians repeatedly get it wrong.
Hypercalcemia, not parasites, would eventually prove to explain Mrs. B’s
pruritus. Like a “good” patient, however, she deferred to physicians’ per-
ceived expertise. They told her she had bugs; she believed them. When they
said she was delusional, she also believed that, providing her own delusional
elaboration.129
That diagnosis engages our tendency to interpret our state in the corre-
sponding light is a strong argument against attaching diagnostic labels to
ordinary conditions.
So potent is the act of diagnosis in and of itself that being labeled with a
medical condition “appears to have a significant psychological impact that
is independent of actual risk factor or disease status.”130 Thus, people with
normal blood pressure who have been mislabeled as hypertensive report
more symptoms of depression than normal controls.131 (For that matter,
people who think of hypertension as a condition caused by stress tend to
report the symptoms associated with the folk disorder Hyper-Tension; of
which more in due course.) So too can people in normal health be led to
believe they suffer from depression—all the more because normal troubles
factor into a diagnosis of depression in the first place, and because a disor-
der that has become as paradoxically popular as depression acquires social
currency in addition to the validity conferred by the DSM. Spitzer and
coauthors caution against stretching the diagnostic boundaries of another
disorder with social currency, PTSD, in part because labeling common
modes of stress or distress as PTSD can spawn “iatrogenic misapprehen-
sions.”132 That is, even a spurious diagnosis of PTSD can convince a patient
that he or she has been genuinely traumatized. While pointing out that
PTSD overlaps depression, the authors do not note that the latter diagnosis
too is boundaryless and can cause misapprehensions. As in these instances,
24 THE NOCEBO EFFECT
with a severely troubled family life. Yet while the diagnosis of bipolar dis-
order was so incongruous that Adam rejected it, he found himself coming
to believe it anyway. As he explains,
I never really believed the label myself like on an intellectual level, because
like most young people, I always felt there was a reason for my behavior. I
started to put some odd pieces of the puzzle together, like: I have this “dis-
ease” and it only manifests itself at home in the presence of 2–3 people that
happen to be a part of my life. Then I began to wonder why I had never
had another “manic” episode after a few years and realized that adults with
the disorder don’t always go years on end without a relapse of that kind of
“episode.”
I did however sort of believe it, only because if you tell a kid something
long enough, they’ll start to believe it. And of course, if I question my crazi-
ness, that’s part of the “illness.” So I got put in a double bind that really did
make me feel like I was trapped or going crazy.146
While Adam was certainly in distress, it’s not clear that he suffered from
any actual disorder despite being put on a slew of psychoactive drugs.
To the Australian psychiatrist who came to know him his case stands as
a warning example of “the iatrogenic harm of erroneous labeling.”147 The
remarkable thing about Adam’s account is not the tale of being torn by
conflicting beliefs as a result of a wrongheaded but wildly popular diagno-
sis, but the fact that it appears in a medical paper at all. Patients’ voices are
conspicuously absent from the medical literature.
Over its course, this study will offer diverse examples of the power of
medical labels, even false ones, over a patient’s experience. I see no reason
to suppose that DSM labels like “bipolar disorder” have less influence over
mind and body. In estimating the power of misplaced belief in diagnostic
fashions driven by the DSM system, it’s important to bear in mind that
DSM-III was the solution to an acute crisis of confidence in psychiatry, a
crisis that saw psychiatry accused not only of masking whim and preju-
dice as objective judgments but of being unable to tell the sane from the
insane, the normal from the crazy.148 The very legitimacy of psychiatry was
sharply impugned by critics who have been known ever after as the school
of “antipsychiatry.” Unlike other institutions, notably the government of
the United States, caught in a credibility crisis at the same time, psychiatry
was in a position to repair its damaged reputation by projecting its identity
as a science and a discipline of medicine. In the end, the means chosen by
the makers of DSM-III to accomplish this task strengthened the credit of
psychiatry immeasurably. It was to standardize diagnosis, thereby creating
the impression that diagnostic judgments simply had to be valid because
they were arrived at by different clinicians following the same rules.149 The
DSM AND THE SHAPING OF DEPRESSION 27
Of the harms that may follow diagnosis, arguably the worst is that it “may
lead you to regard yourself as forever flawed and incapable of ‘rising above’
your problem.”150 That this message is depressing in itself suggests that a
diagnosis of depression can deliver a double hit. It might also induce the
sort of compliance shown by a certain patient who didn’t like the idea of
taking antidepressants but found herself embracing them:
I think I’ll always regret having started medication even for any of the good
that it did in my life. To not know what I could do on my own without it is
very difficult. I mean, there was a sort of deterioration of resolve in me to get
through anything when I started taking drugs. I sort of felt like, “Well, gee,
now that I’m sick, it’s okay to be sick.”151
antidepressant because of its noxious side effects looks like one who drops
out of treatment because he or she is simply depressed. To take a drug for
an ambiguous ailment like depression is to enter a reality where misappre-
hension is always at hand.
Patients on drugs like antidepressants “must come to distinguish side
effects from intended real effects . . . They must get a sense of what it’s
like to be one’s real self while on the medications, and so on. These are all
interpretive problems introduced by the medications, from the inside out,
as it were.”155 All such judgments are slippery. Interpretation itself is an
equivocal activity inseparable from the possibility of misinterpretation—
an activity all the more uncertain in the case of someone who is the object
of his or her interpretation, and who may have received an inflated diag-
nosis to begin with. In that the humanities center on the interpretation of
ambiguous evidence and are predicated on the uncertainty that attends it,
problems like those associated with diagnostic inflation and drug effects
invite humanistic reflection.
the point that it was estimated in 2005 that half the American population
stands to be diagnosed with a DSM mental disorder at some point in their
lifetime.160
Such pitfalls are built into the DSM diagnostic system that the chair of
DSM-IV Task Force, now an outspoken critic of DSM-V, has written a book
detailing them, in the course of which he points out that “the DSM men-
tal disorders are no more than descriptive syndromes; they are not neces-
sarily discrete diseases.”161 Some doctors, it seems, have qualms about the
DSM labels they employ, even explaining to the patient that they’re nothing
but instruments of convenience or tags for insurance purposes.162 But if
labels weren’t powerful, they wouldn’t have to be disclaimed; and a discrete
disease—a distinct pathology—is exactly what any patient diagnosed with
a DSM disorder would assume the disorder to be. If the DSM diagnostic
system can muddle professionals to the extent that it calls for a book of
caveats, what can it do to the minds of patients? The medical literature’s
failure even to pose such a question points to a lack of imagination, which
in turn suggests that works of the imagination may have something to say
to our era even if they don’t share current medical assumptions. A world
in which patients suffer the effects of an inflated diagnosis and doctors are
misled by the ambiguity of normality—the world we inhabit—shares much
after all with the world of fiction, where nothing is more common than
entanglement in illusion. Moreover, while medical papers use strictly neu-
tral language—shunning connotation and tonal effects as offenses against
objectivity—the humanities well know that words are surrounded by a sug-
gestive field. From this point of view it makes sense that a label as potent
as “depression” or “ADHD” can exert a suggestive effect whether or not the
diagnosis is valid.
A critic of medicine’s tendency to exceed its own authority could hardly
do better than the humanist Montaigne when he mocks “all the warn-
ing threats, inferences, and complications which Medicine stuffs into our
heads.”163 The other side of this comment on the false certainties of an
arrogant institution is an appreciation of uncertain knowledge, an appre-
ciation that went into the making of the humanities164 and serves us well
even today. Without a due sense of the uncertain, we might take a three-
week trial of an ADHD medication as probative, forgetting that such exer-
cises, even if well conducted, tell us nothing about the long-term effects of
medication or the eventual outcomes of an ADHD diagnosis. The fixation
on clinical trials as the gold standard of medical knowledge can constrict
thinking and has led to such abuses as poor reporting, as noted above. No
trial can be run to determine if a false diagnosis of progressive dementia
can lead to suicide, but a thought-experiment tells us that if a valid diagno-
sis can precipitate that outcome an invalid one can also, if only because the
30 THE NOCEBO EFFECT
effects of various waters on his kidney stones during his journey to Italy
in 1580–81.170
While we have long since passed Montaigne in the conduct of experi-
ments, in other ways we’re still catching up. Montaigne recounts the tale,
which he had on good authority, of a certain merchant of Toulouse who
graduated from real to imaginary enemas; researchers have shown that
we respond not only to acupuncture per se but to the same procedure
covertly performed with nonpiercing needles—imaginary acupuncture,
in effect. In questioning the attribution of common symptoms to a theo-
rized underlying depression, I too follow Montaigne, who subscribed to
no fixed distinction between mind and body and scoffed at the idea of
dividing a person in two171 but rejected the inference that in his own case
certain somatic signs pointed to mental distress. “My looking-glass never
strikes me with terror,” he writes, “because even in my youth I would often
take on a turbid complexion and a look which boded ill without much
happening, with the result that the doctors, who could find no cause in
my body which produced that outward deterioration, attributed it to my
mind and to some secret passion gnawing away within me. They were in
error.”172
The follies and obsessions leading us to sacrifice health to medicine
are subjected by Montaigne to the same skepticism with which he views
a secret passion declared to exist because its existence seems called for. By
implication, Montaigne also throws into question the promotion of “aware-
ness” in health campaigns today, such as those staged to inform the public
of the signs of depression or social anxiety, or the lack of signs of prostate
cancer. “We are far less aware of perfect health than of the slightest illness,”
he notes, detailing our habit of inventing illnesses and worrying ourselves
with fantasies and prognostications. Far from celebrating an awareness
that’s really a euphemism for anxiety, Montaigne affirms the example of
those who don’t enslave themselves to worry—those like the ploughman
who “only feels ill when he really is ill,” whereas someone more in the know
“often has stone in the mind before stone in the kidney.”173 Montaigne was
on to something. For statistical reasons, if the blood-chemistry offered by
your Wellness Program has 12 tests, the chances are better than even that
one of your values will be abnormal;174 but at least now you’re aware. A
convention of health crusades and driver of the medicalization of ordi-
nary problems, the rhetoric of awareness plants stones in the mind. A study
of the effect of simply informing men of the risks of sexual dysfunction
associated with finasteride (an approved treatment for benign enlargement
of the prostate) found triple the rate of dysfunction “among the patients
aware of the drug side effects compared to those blinded to the drug side
effects.”175 The result is consistent with the entire nocebo literature.
32 THE NOCEBO EFFECT
the first indispensable step is to assemble a rich enough “case history.” Until
that has been done, the wise physician will suspend judgment. If he is too
quick to let theoretical considerations influence his clinical analysis, they
may prejudice the collection of a full and accurate case record and so distract
him from what later turn out to have been crucial clues.182
The doctor who sees a patient just long enough to prescribe an antidepres-
sant transforms the process of diagnosis into a cursory confirmation of a
preconceived idea. The age of the DSM seems to have forgotten how rich
cases can be. That’s one reason for my use at times of literary sources. Pre-
cisely in depicting people as agents (often of their own misfortune), the lit-
erature of the imagination reminds us as little else can of the unworthiness
of the ideal of a life under medical or psychiatric management. Besides, if
medicalization exploits “the gray area”183 of our experience—the realm of
the ambiguously normal—the ambiguities of human existence are investi-
gated with incomparable subtlety in the literature of the imagination. Let
one or two examples suffice for now.
Her soul “overflowing with despair, shame, and humiliation,” Natasha
in War and Peace falls ill after nearly eloping with the worthless Anatole
Kuragin. The description of her “grief ” is stamped with the author’s con-
tempt of the all-knowing doctors who pretend to treat her:
The symptoms of Natasha’s illness were that she ate little, slept little, coughed,
and was always low-spirited. The doctors said that she could not get on with-
out medical treatment, so they kept her in the stifling atmosphere of the
town, and the Rostovs did not move to the country that summer of 1812.184
34 THE NOCEBO EFFECT
Whatever they made of the cough, today’s doctors would go down the
inevitable DSM checklist, ticking depressed mood, insomnia, and loss
of appetite, and if they knew what lies behind Natasha’s grief they would
probably also tick “feelings of worthlessness or excessive or inappropriate
guilt,” thus arriving at the doorstep of diagnosis. In the original, not only
does Tolstoy editorialize against the theory that disease acts the same way
in different people, he makes it clear that regardless of symptoms Natasha’s
illness isn’t a medical matter at all—its source is entirely moral, residing
in a series of events the doctors know nothing of, just as its cure comes
with the passage of time and not as a consequence of treatment. Yet while
Tolstoy’s proclaimed philosophy leads to the dead end of medical nihilism,
his account of the ritualistic administration of pills and powders is remark-
ably subtle and nuanced, suggesting that treatment is variously beneficial,
harmful, and simply irrelevant. In the antidepressant era, all three possi-
bilities remain in play.
It was during Tolstoy’s lifetime that medicine first acquired the status of
a science, though to Tolstoy himself that status was nothing but an auda-
cious sham. The image of doctors as imposing pretenders is fixed unfor-
gettably in the reader’s mind by Tolstoy, especially perhaps in his medical
novella, The Death of Ivan Ilych (whose hero dies in 1882, the year of Koch’s
revolutionary discovery of the bacillus responsible for tuberculosis). But
as concerned as it is with the vanity of medicine, the novella also exposes
the nothingness of the hero’s own work as a bureaucratic enforcer of a new
legal code, as if the legal make-work of sorting cases and the medical cha-
rade of handling cases were reflections of one another. Where does the
similarity lie? Evidently in a drive for classification, the same imperative
that saw the compilation of international statistics on the causes of death
in the later nineteenth century, around the time of The Death of Ivan Ilych.
(Significantly, the exact cause of the hero’s death in Tolstoy’s tale remains
a complete mystery.) From these statistical manuals evolved the Interna-
tional Classification of Diseases, now in its tenth edition, and often cited
in psychiatric literature alongside the DSM. I attempt to view the DSM not
as a settled document whose existence and even authority are to be taken
for granted—how could it be settled when it’s frequently revised?—but a
document as new to our age as the international organization of medical
data was to Tolstoy’s.
Tolstoy himself passed through a period of suicidal misery for which
“depression” seems too light a word, but which he firmly maintains was not
a medical issue. (“Far from being insane or mentally diseased, I enjoyed on
the contrary a strength of mind and body such as I have seldom met with
among men of my kind.”)185 Had he been somehow cured of his misery by
the intervention of medicine, he would have been denied the experience
DSM AND THE SHAPING OF DEPRESSION 35
that set the course for the rest of his life. A philosopher on whom Tolstoy
was a strong influence also wrestled with suicide. On his return to civilian
life following the First World War, Wittgenstein sank into despair:
The defeat and impoverishment of his home country, the death of his most
beloved friend, the frustration at not being able to re-establish old friend-
ships and the strain of putting his whole life on a new footing might them-
selves be sufficient to account for Wittgenstein’s suicidal state during the
autumn of 1919. But perhaps the most important cause of his depression
was his failure to find a publisher for the Tractatus—or even a single person
who understood it.186
If Hotspur suffers from post-traumatic stress rather than the more amor-
phous and less clinical melancholy named by Lady Hotspur, he had to be
traumatized in the first place. (According to the DSM criteria mentioned
but not cited in the BJP article, the response to the causal event had to be
marked by “intense fear, helplessness, or horror.”) But as anyone who has
ever read or seen 1 Henry IV knows, Hotspur, far from being traumatized
by war, can’t wait to be on the battlefield, rejoices in combat, lives for it.
“Come, let me taste my horse.” Let it be noted too that the quoted speech is
embedded in a scene of loving banter between husband and wife.
The PTSD diagnosis uniquely supports my claim that the credibil-
ity of DSM-III profited by the loss of credibility suffered by American
institutions—in this case the government and military above all, but also
the secondary institutions collectively known as society. According to the
original proponents of the diagnosis that came to be PTSD, what made
the experience of the Vietnam veteran so wounding wasn’t the savagery
of the conflict per se, but the savagery of the conflict compounded by the
betrayal of those who fought it at the hands of their own government,
commanders, and fellow citizens. Though PTSD came to be framed more
generally, the inspiration for the diagnosis was the feeling of many a vet-
eran that they were deceived and abused both by the powers that be and
“society at large.”189 It goes without saying that Hotspur’s conviction that
he and his clan have been betrayed is nothing like this. In contrast to the
veteran who confronts a hypocrisy hard to pin down because so many
people are party to it, Hotspur’s indignation focuses on a single offender,
DSM AND THE SHAPING OF DEPRESSION 37
A case of the chronic fatigue syndrome must fulfill major criteria 1 and 2,
and the following minor criteria: 6 or more of the 11 symptom criteria and 2
or more of the 3 physical criteria; or 8 or more of the 11 symptom criteria.14
Reinforcing the analogy is that depression figures among the criteria for
CFS, fatigue figures among the criteria for depression, and other symp-
toms (sleep disturbance and inability to concentrate) are common to both.
Recall too that depression is often diagnosed in primary care on the basis
of somatic complaints that are medically unexplained, like those of CFS.
Yet syndromes like CFS are seen in the medical literature as both cause and
effect of a noxious process of magnifying ordinary symptoms. What the
proposed analogy suggests, then, is that both patients and medicine itself
can exacerbate ailments that are part and parcel of human life by molding
them into disturbing syndromes that somehow catch the spirit of the times
and the imagination of many.
Part of the common lexicon not long ago were rough-and-ready medi-
cal explanations like “nerves” that denoted nothing specific but did serve
to cover some of the ailments of normality. Arguably, what distinguishes
postmodern patients isn’t that they experience more or worse ills, but that
they’re so tuned to the problems inseparable from normality, “the eternal
twitches and twinges to which humankind is heir.”15 A New York Times
article a few years ago focused on young people who choose their own
prescription drugs and acquire them through trade or purchase, the bet-
ter to manage their stress, fatigue, anxiety, and defects of concentration.16
Complaints like these are part of everyday existence and occur widely in
the general population, which helps explain why some would be reported
by placebo groups in clinical trials. Events that normally blend into the
42 THE NOCEBO EFFECT
Medicalization
Not only do the DSM-III criteria for depression tap into the ills of
health, but in one instance they define a moral trait—possibly a commend-
able one—as a symptom. Among the secondary symptoms of depression
according to DSM-III are “feelings of worthlessness, self-reproach, or
excessive or inappropriate guilt (either may be delusional),” which is to
say that in the eyes of the makers of the DSM revolution any degree of
self-reproach was pathological.43 (The Feighner criteria likewise list “feel-
ings of self-reproach or guilt,” whether well-founded or delusional, among
the symptoms of depression. Note that while self-reproach and guilt are
equivalent according to the Feighner criteria, DSM-III deems any feeling of
self-reproach, but only inordinate guilt, a symptom of depression.) On this
showing, when Allen Frances, the architect of DSM-IV, confesses that he
and his colleagues “should have been far more active in educating the field
and prospective patients about the risks of overdiagnosis,”44 he displays a
symptom of depression. Why should the entirely warranted self-reproach
he voices be discounted as a mark of a mental disorder? In defense of self-
reproach, the historian of ideas Isaiah Berlin warns that altering our ways
of thinking to make it impossible to say that we shouldn’t have done some-
thing “would do great violence to our present concepts and usages.”45 Only
if human beings lived in a state of innocence would there be no reason,
ever, for us to reproach ourselves. The assumption that any and all self-
reproach is pathological is plainly utopian.
The dream of a new beginning that inspired the making of DSM-III
was a utopian one in itself, implying as it did that it was possible for psy-
chiatry to break free from history, including its own, and attain a kind of
conceptual innocence; at last it would rely on “data rather than opinion or
tradition,” in the words of the Feighner criteria.46 No less utopian is the ideal
of capturing every possible mental ill in managerial language, an ambition
responsible for a document as complex and mutable as a tax code.47 In
place of the laws dictating every aspect of existence in Plato’s happy com-
monwealth of Magnesia, we now have a psychiatric manual of almost a
thousand pages whose purview extends to every facet of private life down
to Spiritual Problems (V62.89) and Problems Related to Lifestyle (V69.9).
It turns out that students who write badly actually suffer from Specific
Learning Disorder, one of whose symptoms is
Nothing, not even punctuation, escapes DSM cognizance. And where the
laws in Magnesia speak not as hostile powers but friends, the ordinances of
48 THE NOCEBO EFFECT
the DSM are written in the language of benevolence. In that utopias from
Magnesia forward—worlds where the causes of discord and disease have
been ruled out—are places too good to be true, if not too regimented to be
good, this is a cautionary analogy.48 Even in an imaginary commonwealth,
a price is paid for the establishment of health and happiness. Few of us
would actually care to inhabit the sanitized worlds of the utopian tradi-
tion. There is much to be said for tolerating the imperfections of ordinary
existence.49
81.6% reported at least one symptom [over the preceding week] causing
at least mild impairment and 22.1% at least one symptom causing severe
impairment . . . This population survey demonstrates that medically unclear
complaints are an everyday phenomenon. About three out of every four
cases are below clinical relevance.51
But that doesn’t mean such complaints can’t be promoted to clinical rel-
evance. From a marketing point of view, the prevalence of ambiguous
symptoms is a Potosi of possibilities waiting to be shaped into medical
problems. Moreover, the promotion of symptoms into evidence of a disor-
der or disease can set up a reinforcing process, whereby the symptoms are
perceived more vividly just because they now have a ready-made patho-
logical interpretation.
How is clinical relevance to be determined? According to the DSM-
IV criteria for somatization disorder, the act of taking medication for a
complaint gives the complaint significance (which is presumably the same
as relevance): “A somatic complaint is considered to be clinically signifi-
cant if it results in medical treatment (e.g., the taking of medication) or
causes significant impairment.” One would have thought you took medica-
tion because a problem was significant, not that a problem was significant
because you take medication for it. If this test of significance is extended
beyond somatization disorder, then any condition can become of medical
THE ILLS OF HEALTH 49
concern if people can be induced to take a pill for it—in other words, if
enough persuasive power can be brought to bear. Advertising urges me to
take medication for Condition X, and if I do, I myself become an advertise-
ment for the significance of Condition X, reinforcing the original message.
Given the somatic ills inseparable from health, the ambiguity and preva-
lence of conditions like anxiety and depression, and the might of the drug
companies, we can therefore imagine any number of ailments or discon-
tents being elevated to issues of medical concern and treatments of these
conditions being sold to a large, open-ended market.
Before proceeding to the harms introduced by the treatment of nor-
mal conditions, I want to emphasize the blurriness that distinguishes them
despite the seeming-specificity of diagnostic criteria (in the case of DSM
disorders) and the painstaking quantification of medical literature. Untold
millions wouldn’t be eligible for such diseases and disorders if they were
not, in practice, defined liberally enough to include symptoms that blend
into the normal.
While the use of medication to treat common conditions gives them a
veneer of specificity, trials of medications yield their own ambiguities. The
story of the medicalization of shyness turns on the marketing of Paxil, or
paroxetine; and though a review article concluded that three studies appar-
ently found a formulation of paroxetine effective in treating panic disor-
der (a cousin of social phobia, for which Paxil is commonly prescribed),
“according to the corresponding FDA statistical review, only one study was
strongly positive, the second study was non-significant regarding the pri-
mary outcome (and marginally significant for a secondary outcome), and
the third study was clearly negative.”52 The problem of slanted reporting
appears to be endemic in the medical literature, with industry-sponsored
trials yielding positive findings more frequently but also yielding fewer
published reports.53 It’s common practice for industry to publish favor-
able findings while withholding inconclusive or negative ones, possibly in
trials of the same agents. Just that took place in studies of antidepressants,
studies even more unimpressive than the paroxetine trio.54 The dubious
showing of antidepressants in clinical trials may have something to do with
the ambiguity of depression itself—a disorder that includes self-limiting
episodes as well as appropriate responses to life’s sorrows—because it’s the
most severe forms of depression, those most remote from normality, that
most clearly benefit from medication. A number of conditions I consider
in this study, including IBS and ADHD in addition to depression, have
been the subject of questionable reporting in the medical literature.
However, not only psychiatric or psychogenic conditions lend them-
selves to mixed findings. Prostate cancer medicine represents an active
frontier in the expanding empire of medicine, and when the FDA reviewed
50 THE NOCEBO EFFECT
the findings of two trials of drugs to prevent the disease (the PCPT and the
REDUCE Trial), it downgraded the drugs’ alleged benefits on the grounds
that the cancer prevented was in most cases of no clinical significance; it
represented a background condition of normality that had been brought
to the fore only because of the investigative demands of the trials them-
selves.55 Prostate cancer, it seems, spans a blurry continuum, most of which
never manifests itself clinically. The trials in question searched for latent
cancer and found it, but at the expense of promoting a normal condition to
something of alarming import. Thus the more favorable of the two results
was that 39 men would have to take the drug in question in order to spare
one “possibly clinically relevant” case of prostate cancer.56 But the tendency
to focus on and find what you are looking for is by no means limited to
researchers. As I will argue, patients are subject to a similar tendency—and
medicalization plays to it.
Given, then, that normal conditions can be molded into medical prob-
lems such that broad swaths of the population become candidates for treat-
ment, what’s the wrong with this? Not, as some allege, that medicalization
casts a pall of uniformity over the world at large,57 for you could just as
well argue that the creation of scores of new disorders in successive edi-
tions of the DSM has made the world as diverse as a botanical garden.
Nor has the expansion of medical jurisdiction been imposed on an unwill-
ing population. In many cases (CFS, for example), patients themselves
have campaigned for the recognition of new disorders, as indeed millions
prove eager consumers of drugs of questionable medical value. If the trend
toward the creation of medical problems couldn’t proceed without the ills
of normality to fuel it, neither could it have become a trend in the first place
without the consent of the consumer. If men by the thousand are treated
for prostate cancer of dubious significance every year, it isn’t because sur-
gery or radiation is pressed on them against their will, though by the same
token we can fairly question the quality of the information they receive
when screened for the disease to begin with. What opens medicalization to
objection is that by defining normal conditions as pathologies it introduces
harms that weren’t in play until then.
As suggested by the image of the balance scale—an instrument so tra-
ditional it appears in the zodiac sign, Libra—the procedure of weighing
the risks and benefits of medical treatment was never intended for a state
of affairs in which millions of people inspired by campaigns and crusades
seek medical treatments, such as surgery or radiation, that can be harmful
in themselves. The mass marketing of drugs, too, yields outcomes that sim-
ply defy this model of medical decision-making. Of the first blockbuster
drugs, Valium and Librium, Allen Frances concludes that “it is an open
question whether they contributed more harm or good to patients—they
THE ILLS OF HEALTH 51
calmed people down but often addicted them and caused all sorts of with-
drawal problems.”58 Common prudence would argue against prescribing a
drug in a single case, let alone millions of cases, where benefits don’t clearly
outweigh harms. Evidently there was something in the rage for Valium and
Librium—the first market sensations of their kind—that overwhelmed
common prudence. (The allusion to the balance scale in the name “Lib-
rium” is an especially subversive touch.) And that was before the era of
direct-to-consumer advertising.
A potent accelerant of medicalization, the practice of DTC advertising
of drugs was made possible by the Food and Drug Modernization Act of
1997, which lifted the requirement to cite all risks and side effects provided
only that an ad referred the consumer to some other ad that did cite them.
But what does “all risks and side effects” mean? If it means, first of all,
the adverse events tied to drugs in randomized clinical trials, these were
routinely underreported in the medical literature at the time DTC drug
advertising began. Because trial data about adverse events is needed to
estimate benefit/harm ratios,59 the inadequacy of the provided data could
easily compromise calculation of the pros and cons of prescribed drugs.
In cases where the data was reported cursorily or simply buried in com-
pany archives, the literature itself could be as misleading as advertising.
In any case, the upshot of DTC advertising—an inundation of ads pro-
moting a given drug while revealing as little as legally possible about its
drawbacks—illustrates the power of medicalization to drive the question
of harms offstage even as it thrusts new products and disorders into the
public spotlight. Without accepting the claim of Allen Frances that his cau-
tious and well-intentioned work in DSM-IV was abused by Big Pharma
and DTC advertising, it’s hard to deny that the mass marketing of drugs
for ambiguously normal conditions distorts medical reasoning. Clearly,
something has gone wrong with the calculation of risks and benefits when
psychoactive drugs are prescribed for children before good information
about their effects is in hand. Yet just this has occurred. A review of the
practice of pediatric psychopharmacology in 2001 concluded that the pre-
scription of psychoactive drugs for children “has outpaced the empirically
based data on their safety, efficacy, and effectiveness.”60 Among the targeted
pediatric disorders is depression. Considering that a careful meta-analysis
in 2011 found that “consistent and clinically meaningful benefits of antide-
pressant medications, relative to placebo, have not been demonstrated in
depressed children,” it becomes difficult to reconcile pediatric antidepres-
sants with risk/benefit analysis at all.61
Under the combined auspices of the DSM and DTC advertising, many
a medical consultation has been reduced to a hasty ritual of prescription
writing. In a study of doctors’ prescribing practices where actors played
52 THE NOCEBO EFFECT
For the majority of those who use psychiatric drugs the process involves try-
ing many different medications over time, each one beginning with the hope
of success, but then too often ending in disappointment. The search for the
“right” drug is analogous to the search for Mr. or Ms. Right. Just as people
hold our hope of finding the perfect mate despite repeated disappointments,
those suffering from depression often remain convinced that there’s an ideal
drug for them.64
“Hope of success” is of course a nutrient of the placebo effect; and the pat-
tern of discarded medications echoes the narrative line of medical history,
with one treatment after another falling into disuse as its efficacy vanishes,
for the simple reason that it was a placebo in the first place. But though the
benefits of SSRIs trivially exceed placebo in most cases, their side effects go
well beyond placebo, such that members of the treatment arm in antide-
pressant trials guess their group with a frequency that defies randomness.
THE ILLS OF HEALTH 53
Paxil class look good by comparison, their various effects are still such that
an overwhelming majority of blinded subjects in trials of SSRIs can figure
out that they have been randomized to drug. Moreover, a number of SSRI
side effects, including both somnolence and insomnia, both loss of appetite
and weight gain, as well as loss of libido, cunningly resemble symptoms of
depression itself. The medical literature doesn’t discuss the irony of treating
depression with drugs that mimic many of its statutory symptoms. Con-
cerning sexual side effects in particular, so suggestive is their resemblance
to a DSM-III symptom of depression that some caution against the whole-
sale attribution of adverse sexual events to antidepressants;75 they might be
due to the disorder, not the treatment. The least we can say is that it’s no
easy matter to distinguish the sexually depressing effects of SSRIs from the
condition they are supposed to relieve. The effects continue as long as one
takes the drug.76
Depression lent itself to mass marketing because of the arrival of a new
generation of more tolerable drugs not long after the installation of new
criteria of depression in the DSM: criteria that made the disorder appear
definite and specific, as befits a medical condition, even while extending
eligibility for it to an indeterminate population whose ills are as common
as fatigue, faulty concentration, and sadness itself. Yet sadness by no means
denotes the same thing as depression. How would we view a campaign
to increase Sadness Awareness? Not only does depression seem to war-
rant treatment while sadness doesn’t, but depression caused by a “chemical
imbalance” will presumably remain in place unless and until the problem
is corrected, while sadness may wax and wane with circumstances. And as
the millions of prescriptions written annually for antidepressants testify,
the calculation of harms and benefits changes in a climate of medicaliza-
tion. The chair of the DSM-IV Task Force concedes that this authorita-
tive, internationally cited document—the arbiter of psychiatric diagnosis,
insurance coverage, and much else—“probably resulted in more harm
than good,” the very opposite of what we would expect if harms were duly
weighed against benefits. He maintains, however, that “DSM-IV didn’t by
itself do much harm,”77 implying that if it “resulted” in harm, that was the
fault of those who abused the document’s good intentions. He is less kind
to the latest iteration of the bible of psychiatry, charging that DSM-V “was
prepared without adequate consideration of risk-benefit ratios” and is in
itself, therefore, a harmful document.78
Yet it’s doubtful that DSM-V began the practice of not giving full weight
to obvious possibilities of harm. The former director of the National Insti-
tute of Mental Health describes the diagnostic scheme of DSM-IV as
a sort of artificial system, noting in particular that “the substantial need
for NOS [Not Otherwise Specified] criteria, excessive comorbidity [that
THE ILLS OF HEALTH 55
If the principle of weighing risks and benefits weren’t itself at risk, it couldn’t
be said that as a result of medicalization “Medicine’s much hailed ability to
help the sick is being challenged by its propensity to harm the healthy.”83
It’s the aggressive expansion of medicine, such that large regions of nor-
mality become subject to its cognizance, that makes such harm possible.
By the same token, the harms authorized by medicalization expose the fal-
lacy of the claim that normality is nothing but a mechanism of repression.
According to that way of thinking, which has taken root in academia owing
in large part to the influence of the French theorist Michel Foucault, nor-
mality is the insidious force that both stigmatizes deviance and imposes the
chilling uniformity that makes modern society what it is. Writes Foucault,
The authority of Foucault has reached the point that he’s cited not only in
the medical literature but by critics of medicalization, ironically enough.85
THE ILLS OF HEALTH 57
Magnification of Symptoms
The problem is that feeling and believing that one is a victim and hence
incompetent can inhibit action. That results in a positive feedback effect,
because acting to improve one’s life—or failing to take such action—can
have a profound influence on belief and feeling. An erroneous belief that
past abuse necessarily has lasting effects can thus lead to drift and a worsen-
ing of the current lives of those who believe they were abused in the past.97
Maybe awareness isn’t the pure good it’s thought to be. Though hyperten-
sion is generally symptomless, people may report symptoms when told they
have it, even if the numbers they are given are false.101 In many cases, in
fact, the experience of symptoms varies with our expectations as well as our
understanding of what the symptoms are, what causes them, and what they
portend. Though a DSM diagnosis is based on counting symptoms almost
as if they were a unit of measurement, a symptom is after all “a subjective
indication” (as the Oxford English Dictionary puts it), a signal liable to inter-
pretation. I may experience symptom A (say, a sleep problem) differently
when I learn it’s diagnostically connected with a disorder (say, depression)
in a way I never supposed; I may experience symptom B (say, a dislike of
public bathrooms; “shy bladder syndrome” in DSM-V) differently when it’s
lumped with other, seemingly unrelated symptoms (say, an aversion to pub-
lic speaking) in a way that constitutes a disorder. In both cases I confront
a diagnostic whole framing a number of components—a whole, moreover,
that has a clinical definition and status that an everyday problem doesn’t.
Moreover, once I know what a disorder consists of, one symptom may sug-
gest another by association. “Relating symptoms to a specific set may lead
to a false perception and reporting of other symptoms associated with the
set.”102 Whether or not a given symptom is collected with others under the
rubric of a disorder, the awareness acclaimed by the medicalization move-
ment can change the experience that a symptom is. With the spotlight of
medical attention now shining on it, and my attention sharpening accord-
ingly, my symptoms themselves seem to become more strongly defined. Like
imagination in Montaigne, heightened attention animates its objects.
Such attention is itself a sort of adverse effect of participation in a clinical
trial. Subjects in these trials who simply read consent forms specifying the
possible side effects of a test drug commonly report such symptoms—even
from placebos—at a higher rate than controls who aren’t so informed. Two
critics of unwittingly suggestive consent documents therefore urge those
who frame them to give thought to “the harm that may be caused by the
[consent] ritual itself.”103 Our very sensations can be shaped by what we’re
cued to expect, as shown by an ingenious experiment where subjects who
were led by a consent form to expect pain reported pain, and those led to
expect pleasure reported pleasure, from the same stimulus.104 What does
the cuing of such seemingly primary sensations as pleasure and pain have
to do with medical expansionism and the accompanying drive to raise pub-
lic medical awareness? Guided by the finding that subjects whose attention
is drawn to possible side effects tend to report those very events at a higher
rate, I argue that the awareness campaign is a formula for the nocebo effect.
Precisely by intensified awareness of their own symptoms—what the lit-
erature calls hypervigilance—patients without abnormal indications can
THE ILLS OF HEALTH 61
amplify their distress to the point that they’re diagnosed as suffering from a
disorder. For patients in this category, and they are many, raised awareness
is more like the problem than the solution.
A process well known to medicine, symptom amplification can take
place whenever we direct anxious attention toward our own sensations;
and given that even in a state of health we are subject to many adverse
sensations, from headache to abdominal uneasiness, it follows that magni-
fication is a standing possibility. The anxiety that serves to fuel the process
is itself, similarly, a common occurrence; that as many as 25 percent of
subjects treated with placebo may report reactions of anxiety (approxi-
mately the same percentage reporting fatigue)105 suggests that anxiety is a
background condition brought to light by the power of the placebo, much
as symptoms themselves become more pronounced under the influence
of anxious attention. So too, because amplification isn’t a process specifi-
cally linked with certain disorders or syndromes and no others, in any
condition brought to our attention as a medical problem, symptoms—if
it has symptoms—can presumably be amplified. Just as it can’t be said that
consent forms can cue certain sensations but not others (they can even
cue contradictory sensations), so symptom amplification—the probable
mechanism of such induced sensations—can heighten any distress and is
to that extent in play whenever a hitherto normal ailment is promoted to
the level of a medical problem.
We can easily overinterpret internal signals, and if the process is trig-
gered by conditions as dissimilar as whiplash and fibromyalgia, then its
application is likely to be open-ended. The editorial that first pointed out
that consent documents plant suggestions likened the suggestions to pla-
cebos in drug trials, which indeed elicit all kinds of pharmacologically
improbable effects.106 Placebo responses occur at a high though varying
rate across the entire range of clinical trials, whose painstaking design
is dictated in part by the necessity of controlling for them. A process of
symptom amplification that worked even roughly as broadly as the placebo
effect—and human suggestibility didn’t come into the world in the middle
of the twentieth century with the advent of the randomized clinical trial—
would implicate countless millions of people. But while symptom ampli-
fication resembles a nocebo effect, note that the process of amplification
doesn’t call on us to invent sensations out of thin air but simply to work
up what’s already there (as you need only work up common symptoms to
simulate ADHD). In that symptom amplification isn’t limited to a particu-
lar set of disorders and has a rich supply of ore to work with in the general
population, its field of operation is, in fact, indefinite.
In an effort to build the market for a pill for a recently codified disor-
der, a drug company launches a promotional campaign, both direct and
62 THE NOCEBO EFFECT
Most of the papers cited in the course of this study report directly or indi-
rectly on carefully designed trials. How can one draw conclusions about
the stretching of medical boundaries in the world at large from studies
THE ILLS OF HEALTH 63
ourselves for symptoms of a disorder could also bias our experience of our
own states, constituting as they do selective search instructions. If such ads,
which virtually target a weakness of our perceptual system, were unable
to boost the perception of symptoms, it seems unlikely that royal sums
would be wasted on making and airing them. Besides, laboratory manip-
ulations are more like commercial ones than might appear. In the small
study just referred to where subjects were tricked into thinking they were
taking an actual drug, the proceedings were given an air of verisimilitude
by theatrical touches like the taking of a medical history. Pharmaceutical
ads feign verisimilitude in their own way—using actors to play doctors,
for example—and in the notable case of antidepressants, inspire public
belief in medications trivially superior to placebo. The principle of balanc-
ing risks and benefits is itself jeopardized when consumers are exhorted
to search for and interpret signals of ambiguous disorders like depression,
and to seek medications whose therapeutic effects may be minimal but
whose side effects aren’t.
Beyond the specific content of an ad, the very nature of a publicity cam-
paign echoes the process by which common symptoms are magnified into
chronic distress. Symptom amplification is a repetitive, self-reinforcing
process, with patients sometimes being asked over and over again to rate
their pain—a process known to worsen it—and adaptations aggravating
the distress they were meant to relieve.114 Publicity campaigns, which are
nothing if not redundant, are also self-reinforcing. Building on themselves,
they are a sort of public version or vehicle of the dynamic of directed
attention. After all, medicalization takes place not just here or there but
across society in general, and therefore can’t occur without some means to
propagate its messages. Not only advertising, awareness drives, and medi-
cal trends (as potent as such sources are) but the uptake and echoing of all
these in the media, online, and in common speech provide the context of
“overwhelming information”115 that powers the medicalization of everyday
experience. Thus, in the process of symptom amplification by which ordi-
nary ills grow into CFS, a crucial element is the information or folklore in
general circulation about the syndrome. Patients with CFS may “expect,
from what they have read, that chronic pain could eventually arise as part
of the syndrome,”116 and arise it does. Or consider the following case.
Already with his share of aches and pains, Jones suffers a minor jolt in a
motor vehicle collision. True to the common belief in our part of the world
that collisions cause whiplash, he begins to experience neck pain and head-
ache, unaware that the symptoms of whiplash are all but indistinguishable
from ordinary background events and that minor collisions don’t beget
whiplash in countries where the concept of whiplash is unknown. Hyper-
vigilant,117 he finds to his dismay that his symptoms persist and decides to
THE ILLS OF HEALTH 65
do something about them. Like one who learns that what he thought of as
unrelated events are actually symptoms of a single disorder—symptoms
that seem to reinforce one another as a result—Jones’s neck pain and head-
ache are now mutually confirming. And just as those caught up in a case of
imaginary toxic exposure are likely to have definite beliefs about the cause
of toxicity, Jones knows the cause of his symptoms.
In a reversal of the usual procedure, he therefore brings a ready-made
diagnosis of whiplash to a doctor. The pain medications he receives cause
a certain dizziness, another sign of whiplash. Refraining from usual activi-
ties, he begins to record his symptoms, which become more stubborn the
more immobile he becomes and the more anxiously he follows them. In
fear of aggravating the injury he adopts awkward postures that make things
worse. He sees a lawyer, at which point his pain becomes truly chronic. In
another society the aftereffects of the collision might have cleared up in a
matter of days, but in this case the cocktail of implicit expectation, anxious
attention, maladaptation, drugs, and legal action works to produce more
harm than the original injury. Through it all, Jones’s symptoms have only
become “more intense, noxious, and worrisome.”118
As this example suggests, information that isn’t managed at all, but sim-
ply circulates in our cultural water system, is as capable of shaping our
experience as well-engineered advertising campaigns. The medicalization
of a normal condition has been achieved not when psychiatrists declare
it a disorder with a given constellation of symptoms, but when its exis-
tence as a disorder seems as intuitive to people at large as whiplash. In this
respect, the power of DSM diagnoses is perhaps best shown by their ability
to become part of our vernacular, which is also to say our way of thinking.
Open-Ended Possibilities
Our propensity to find the symptoms we’re looking for ought to be well
known to doctors, because they share it. The phenomenon of medical stu-
dents’ disease refers to the documented tradition among such students,
especially perhaps in their first year, of discerning in themselves the symp-
toms of studied diseases. “The reported symptoms are often common
ambiguous somatic symptoms which in other situations may be inter-
preted as normal bodily sensations,”119 but which in this charged setting
are heightened even as conflicting signals are ignored. Medical students’
disease refers not to a single malady but a capacity to pick up any malady
under study—at least its presumed symptoms. And precisely because they
are interpreted as evidence of disease, somatic signals that might other-
wise be registered as normal sensations, if registered at all, now arouse dis-
tress. (“Now I understand—I have heart disease!”) As reported in a study a
66 THE NOCEBO EFFECT
half century ago, the severity of distress associated with medical students’
disease “ranged from moderate concern to levels that were disruptive of
daily functioning.”120 Here then is a case in which common symptoms are
sharply heightened once attached to specific maladies.
In its ability to mimic all sorts of conditions using common symptoms
as its raw material, medical students’ disease points to the boundless pos-
sibilities of overinterpreting bodily states. Medicalization exploits those
possibilities in the general population. Medical students aren’t selected
for their suggestibility, after all; if these candidates for induction into the
supposed temple of mind-body dualism can discover in themselves all
manner of disturbing conditions in a state of health, so can everyone else.
Indeed, it’s said that medical students’ disease represents nothing more
than “a normal response to learning about new disease information.”121
It follows that the potential to experience the symptoms of imputed dis-
orders occurs broadly, and that anyone immersed in information about
diseases is subject to the possibility of distorted perceptions. “People tend
to look for additional symptoms to fit [diagnostic labels]. Thus, the per-
son who attributes his chest pain to a heart attack is more likely to notice
other sensations consistent with heart disease.”122 By the same token, if
diagnosed with depression mistakenly, I can still pick up confirming evi-
dence of depression, all the more if I learn that an everyday problem like
sleep trouble is a symptom.
If nonspecific ills are part of normality and the process of construing
them as of medical import is itself a common one, then multitudes are
capable of being swept up in the process of medicalization at any time.
Medicalization after all implies not just the professional interpretation of
normal variations as medical problems, but the uptake of this practice by
the public. Thus, like medical students, members of the general population
can and do pick up thematic diseases or disorders by interpreting signals
that might not otherwise pass the threshold of awareness. The reports sur-
rounding such conditions, and giving them currency, act like the informa-
tion on which medical students model their symptoms.
But while medical students presumably outgrow medical students’
disease, others may grow into their disorder, reinforcing it by the atten-
tion devoted to it and the adjustments made in its name. Is the distress
resulting from the magnification of ordinary ills nothing but an optical
illusion? If that were so, there would be no reason to caution the authors
of consent forms about the risk of enumerated side effects springing to life
in the minds and bodies of subjects in clinical trials. Among the reactions
induced by consent procedures is nausea.123 It would be a mockery to tell
people their nausea is imaginary. Doctors, too, see many patients whose
distress is to some degree self-authored but no less distressing for that.
THE ILLS OF HEALTH 67
If people with the ills common in a healthy population are to seek treat-
ment, their problems must somehow be elevated to a higher category, and
the evidence suggests this is very possible. The theme of awareness, sug-
gesting as it does something entirely free of bias—something as pure as
the flame of consciousness itself—serves as an ideal cover for this kind of
inflationary operation. Or consider the propagation of “functional somatic
syndromes” (FSS’s) such as IBS, in which common symptoms are inter-
preted as markers of abnormality. Often these ailments are denied the
imprimatur of medical legitimacy, and the patients who claim them are left
to seek relief for a condition without a known treatment. It’s otherwise in
the case of ills like depression or anxiety, which in one shade or another are
also prevalent in the population (and feed into the FSS’s),124 but are autho-
rized by medicine and said to be treatable with a pill. Where the FSS’s are
engulfed in a polemical folklore portraying these conditions as disabling,
the publicity depicting depression and anxiety as potentially catastrophic
and disabling emanates from accredited sources, including the drug indus-
try and its surrogates, and may be all the more influential for that reason.
By casting common problems as medical disorders and surrounding them
with exaggeration (including the use of “Major Depression” as the official
title of a condition that often abates without treatment), the promoters of
medicalization virtually model the process of magnifying symptoms.
Proof of Principle
pain was traced to other sources. “In each of the 55 patients the absence of
any cardiac abnormality and the positive non-cardiac diagnosis of the chest
pain [were] communicated clearly to the patient and to his or her relatives
and to the general practitioner, with a conscious attempt to reassure them
about its benign prognosis.” Nevertheless, almost all the patients continued
to believe they suffered from chest pain, and most believed that too much
exertion was dangerous for them. “It is clear . . . that it is difficult to rescind
a diagnosis of angina once it has been made.”128
What of psychiatric misdiagnosis? Is it possible that a normal person
diagnosed for some reason with a mental disorder could come to believe
he or she really had it? Under suitably oppressive conditions, it might be
hard not to believe it. Thus, during the Chinese Cultural Revolution many
of those detained on ostensibly psychiatric, actually political charges seem
to have found themselves succumbing to the fictions they were accused
of. “The police pressure on those arrested for alleged political offenses
was often so great that many people began to believe that they actually
had committed ‘towering crimes against the people’ . . . and in the course
of their daily forced-confession writing sessions in prison, they began to
reinterpret large sections of their own pasts in lurid and entirely fabulous
terms.”129 But can people buy into false diagnoses without police power
bearing down on them, merely as a result of the prestige, social currency,
and intuitive appeal of medical labels?
Earlier we met “Adam,” diagnosed with bipolar disorder at the age of 12,
at a time when the diagnosis had acquired a kind of sweeping popularity
both inside and outside the halls of psychiatry. Even while he recognized
that the label didn’t fit his case, Adam came to believe it in some core way.
Precisely because an inflated or merely trendy diagnosis can take root in a
patient despite its weak basis, some psychiatrists have made it their work to
“undiagnose.”130 The American Psychiatric Association itself undiagnosed
an entire population as part of the same reform movement that eventually
led to DSM-III. The group was homosexuals. And much as retracted medi-
cal diagnoses can live on in the minds and bodies of patients, the damage
done by the classification of homosexuality as a mental disorder was not
easy to undo even after the APA’s official revocation of the diagnosis.
In accordance with prevailing psychiatric opinion, homosexuality was
listed as a mental disorder in the 1952 edition of the DSM, a definition that
remained in place until rescinded in 1973. The delisting of homosexuality
had the unintended effect of leaving homosexuals who happened to be in
treatment without a reimbursable condition.131 Why were they in treatment?
To claim they now suffered from “ego-dystonic homosexuality”—a successor
disorder arising not from their own pathology but hostility toward homo-
sexuality in the world around them—is equivalent to saying they suffered
from the antagonism codified in the original APA diagnosis. The diagnosis
THE ILLS OF HEALTH 69
was clearly iatrogenic. If the expectation of sickness can cause just that,132
so can exposure to the medically authorized message that you’re sick con-
vince you that sick you are. It’s said that “the gay population’s insistence that
homosexuality be removed from DSM-II is an illustration of the enormous
authority and significance given to the APA’s diagnostic system.”133
We have every reason to believe that a system so powerful—a system,
moreover, that now reaches more deeply into everyday life than in the
1970s—can convince us that we’re suffering from depression, PTSD, or
some other disorder, all the more if the diagnosis is in favor and its symp-
toms blend into normality. After all, our ailments don’t identify themselves
to us; we learn to identify them. Not that long ago, patients didn’t recognize
as “depression” symptoms like those aggregated in DSM. Said an English
GP in 1972,
There has to be a first time when the patient has not been conditioned to
say “I am depressed, doctor.” This is a “doctor phrase” rather than a “patient
phrase” and results from the constant reiterated question, “do you feel
depressed?” The new patient will present with one or more of a great variety
of physical and mental symptoms.134
In that mix of symptoms are sure to be some common ones. Thus, high on
the list of common symptoms published in 1968, before the popularization
of depression, are a number deemed indicators of depression from DSM-III
(1980) to the present. While the questionnaire the list was based on didn’t
ask about depression or its then more popular sibling anxiety, “irritability”
scored almost as high as “excessive sleepiness” and “fatigue,” both now sig-
nifiers of depression. (Also as reported in 1968, a community survey found
“irritability” clustered with mild depression, insomnia, and “nerves” in as
much as 30 percent of the population.)135 Maybe the irritable were patients
who hadn’t yet learned to say, “I am depressed.” But to acquire that practice
is to do more than engage in a form of words. That people told suggestively
enough that they show signs of depression can come to think of themselves
as depressed is shown by the great numbers—far more than ever before—
now treated for the disorder. Common symptoms collected under a label at
once authoritative and connotative can assume a new power.
Many would argue that people still need to be taught to say, “I am
depressed.” Robert Spitzer and coauthor propose the following case of a
patient who may be depressed with knowing it:
An individual who has marked loss of interest and pleasure (not depressed
mood) with five of the nine symptoms [of depression] (e.g., hypersomnia,
weight gain) present to a marked degree nearly every day for 2 weeks might
be neither significantly distressed nor role-impaired. Such an individual may
still have a depressive disorder.136
70 THE NOCEBO EFFECT
Red Flags
When introducing consumers to the idea that they might have a medi-
cal problem, the agents of medicalization like to say they’re promoting
awareness. What is it about the word “awareness” that recommends it? In
that being aware seems better than not being aware, the word is definitely
loaded. In liberation rhetoric, the oppressed who think of themselves as
oppressed—who have achieved awareness—are more politically advanced
and more capable of action than those at a lower stage of consciousness.
Arguably, the trope of coming-to-awareness would apply best to a case
where someone who doesn’t perceive a looming danger comes to appreci-
ate that it exists: registers the red flags. Rarely, though, does danger present
itself with the clarity demanded by this scenario.
Soon after Maj. Nidal Malik Hasan shot to death 13 and injured many
more at Fort Hood, Texas, on Nov. 5, 2009, reporters and commentators
began to wonder aloud whether warning signs of the homicidal outburst
had been ignored. “Officials may not have heeded warning signs,” declared
a headline in the Washington Post of Nov. 7. According to an article in
the New York Times on Nov. 9, the FBI and the Army may be guilty of
“missing possible warning signs that might have stopped a mass killing.”
Whether or not such a massacre was predictable, the retrospective invoca-
tion of warning signs seems to take place regularly—predictably—in the
aftermath of mass murder. Within a day of the massacre at Virginia Tech
in April 2007, CBS News already had an article on its website headlined,
“Warning Signs from Student Gunman.” Appended to the report of the
Virginia Tech Review Panel, likewise, is a list of “red flags, warning signs
and indicators.” It’s as if the ritual repetition of phrases served to soften the
shock of events. However, the notion that such atrocities are preceded by
legible warnings, and could therefore have been prevented if only the cues
72 THE NOCEBO EFFECT
were properly heeded, obscures the reality that it’s easier to predict events
after they occur.
Regardless of the language that sprang up seemingly automatically in the
press in the immediate aftermath of Fort Hood, those events, too, yielded
warning signs only in retrospect. After the fact it came out that Hasan was
known to be in communication with an anti-American cleric in Yemen,
yet “there was no indication that Major Hasan was planning an imminent
attack.”1 The ominous import his exchanges with the cleric seemed to pos-
sess in the light of later events escaped intelligence analysts in real time.
Even those who think Hasan’s actions might have been prevented would
have been reluctant at the time to charge him with conspiracy to commit
mass murder on the strength of messages that gave no evidence of any such
thing.
A number of diagnoses proposed for inclusion in DSM-V, such as
“psychotic risk syndrome,” were inspired by the dream of preventing a
catastrophe by reading its signs in advance. The trouble with the inher-
ently attractive idea of a warning sign lies, in part, in the concept of
advance notice of an impending disaster. (The word “disaster” itself, liter-
ally meaning an “ill-starred” event, has traces of astrology.) Compound-
ing the uncertainty of behavioral signals as such with reference to an
as-yet nonexistent event, such a sign seems doubly uncertain. By defini-
tion, we can easily miss the import of an ambiguous signal directing our
gaze to something that hasn’t yet taken place. However, it’s also easy to
make something of nothing—to convert a datum into a warning sign by
reading into it ominous import it doesn’t possess or warrant. The notion
of a warning sign is rife with false positives. In point of fact, “there are
no accurate methods of discriminating those who will go on to develop a
bona fide mental disorder from those who do not,”2 the use of retrospec-
tive judgment to make it appear that “I knew it would happen all along” is
a patent example of recall bias, and psychological experts have a notably
poor prediction record.3 In an amicus brief filed in a capital case some 30
years ago, the APA itself declared that “even under the best of conditions,
psychiatric predictions of future dangerousness are wrong in at least two
out of every three cases.”4
Say a youth who turns a gun on his fellow students is discovered to
have liked a song that celebrates killing. In retrospect the association
seems highly significant, though in real time no one may have read any-
thing into it (after all, millions of others also liked it), and in any case
it would have been impossible to predict so terrible an outcome on the
strength of such slender evidence, even supported by other evidence of
the same kind. Was his liking for the song a sign? Why didn’t anyone rec-
ognize it as one? How, on the basis of evidence as tenuous as this, would it
SEARCHING FOR SIGNS 73
Guided Self-Examination
It’s as if a spell had been broken. However, the witchdoctor too lifts a spell
of suffering, telling a sick man he violated a family taboo and prescribing
the sacrifices necessary for restitution. In both cases recovery is speedy for
the simple reason that both therapists identify the cause of the illness, and
“the very act of naming has a therapeutic effect.” Such is “the magic of the
right word.”7
Whether or not words are magical, everything about the vignette of the
psychiatrist is a little too perfect to be true. No sooner are certain formulaic
phrases spoken than a patient locked into a history of suffering she barely
understood begins to see daylight. It seems strange that the client’s his-
tory, though forgotten, comes out instantly, fully formed, the moment it’s
called for. Could it be that the client has been cued by the therapist to bring
forth a particular kind of story about her father? If you bear in mind that
memories are subject to the power of suggestion8 and that psychotherapy is
highly conducive to suggestion,9 the skit certainly reads differently. Suspect
too is the very neatness—the psychological precision—of the causal mech-
anism said to lie behind the patient’s problems. Concerning “cause-and-
effect stories” told in the course of psychotherapy, David Jopling observes
in a probing study, “What clients take to be deep and accurate insights into
their psychology and behavior may really be stock psychodynamic causal
generalizations and confabulations they have learned during therapy.”10 In
the case of the angry-looking client, the magic of the therapeutic word may
lie not in any curative power but the ability to summon up a story so con-
ventional that it inspires the person who fashions it to believe it.
What if a diagnostic label is applied not to a single patient come in search
of it but millions of potential patients at once, all of whom are advised to
scan themselves for signs of the condition in question? In 1979, on the eve
of DSM-III, Christopher Lasch noted with a prophet’s eye that “medicine
and psychiatry . . . reinforce the pattern created by other cultural influences,
in which the individual endlessly examines himself for signs of aging and
SEARCHING FOR SIGNS 75
ill health, for tell-tale symptoms of psychic stress.”11 Such anxious search-
ing is not only a folly but a trap, in that looking for evidence of trouble
means trouble is going to be found; you need only interpret ambiguous
indicators in the light of what you’re looking for. By the same token, medi-
cal and psychiatric labels with enough suggestive power can elicit confirm-
ing signs much as a therapeutic convention can elicit a confirming story.
“Research on selective monitoring has shown that providing persons with
a diagnosis can guide interpretation of sensations by setting up expecta-
tions regarding what symptoms they should be experiencing.”12 Especially
in a climate of emotive publicity, merely providing patients with the label,
TBI (Traumatic Brain Injury), can aggravate symptoms and degrade cog-
nitive performance.13 Once Mrs. B, a patient mentioned in Chapter 1, was
diagnosed with lice, she started making suppositions about the source of
the infestation, scratched herself bloody, and even quarantined herself for
fear of infecting others, even though, as it turned out, she had no parasite
at all.14
Advertising and awareness campaigns urging members of the gen-
eral public to scan for signs of this or that disorder constitute primers in
misinterpretation and are especially suggestive when the featured disor-
der is a public theme (like depression or bipolar disorder in recent years)
and broadly overlaps normality. Once the hitherto obscure diagnosis of
Asperger’s syndrome acquired a kind of celebrity, people began to notice
the signs of Asperger’s in themselves. They did so all the more readily in
that the diagnostic markers are commonplace and the syndrome acquired
a paradoxical cachet. A popular self-administered quiz for Asperger’s syn-
drome asks you to agree or disagree with such innocuous propositions as
the following:
Given (again) the triviality of the indicators, it’s no wonder that Asperger’s
is on the rise, to the point that this once-esoteric term has something like
the popular currency of a brand name.
The process of detecting or overdetecting cued symptoms takes diverse
forms. Under the influence of expectation a depressed person can inter-
pret a mood variation as a confirming symptom, a finding depressing in
itself.15 Yet noticing certain adverse signs or signals that have been flagged
for attention is a normal process of wide occurrence. Many side effects of
76 THE NOCEBO EFFECT
to conform with the diagnostic conventions of the day) coupled with our
perceptual biases helps explain why it is that patients and doctors alike see
clear evidence of anxiety when anxiety is in favor, depression when depres-
sion is in favor, attention deficits as soon as they become diagnostically
popular, and all the rest.
The case of the woman inspired by the magic of the right word has been
cited as an example of the placebo effect in action,20 and some say the act
of diagnosis in and of itself possesses a similar therapeutic magic. Outside
the clinic (and even inside, though that matter will be taken up later) the
message that there just might be something wrong with you, the message
of medicalization, may not be so benign. Given that the process of guided
search is “prone to certain biases and distortion in perception”—all the
more, you would think, when the search is an anxious one—it can read-
ily elicit magnified sensations and ready-made interpretations. Just as
placebos can “redirect attention” from pain (which is an argument against
using them to determine whether or not pain exists),21 so, conversely, can
focusing attention on symptoms exacerbate the symptoms themselves.22
Considering, then, that in a state of good health we can and do experience
varying symptoms and ambiguous ills and discontents, and that height-
ened attention to these events can heighten their intensity, considering in
addition that interpreting a symptom as evidence of a disorder or disease
can sharpen it, it follows that the trend toward medicalization can result in
keener symptoms and a keener sense of being unwell. Everything known
about placebo and nocebo effects favors this possibility.
An example of a particularly anxious guided search will illustrate the
possibility of overreading somatic signs.
A randomized trial of breast self-examination (BSE) involving over a
quarter million Chinese women found a marked increase in biopsy rates
but no decrease in breast-cancer mortality over 11 years, suggesting that
BSE resulted in the overdetection of suspicious tissue.23 Anxious atten-
tion makes for the misinterpretation of signs. And attention is all the more
likely to be anxious if the woman performing BSE can’t really be sure she
hasn’t missed something, like an American woman who said,
It’s a no-win situation. I don’t feel reassured after doing [BSE], mainly
because I’m never really sure that I didn’t find anything. It’s funny; I’m pretty
sure I would find a lump if I had one, but I’m not really sure about finding
nothing.24
In the privacy of this woman’s experience the diagnostic maxim that a false
negative is worse than a false positive seems to have been transmuted into
a painful bind.
78 THE NOCEBO EFFECT
Now assume that women urged to perform BSE have been thoroughly
exposed to the sort of campaign in favor of mammography waged in
Europe and the United States in the name of breast-cancer awareness.
Whereas mammography appears to spare approximately one breast-cancer
death per thousand women screened (albeit without reducing overall mor-
tality), a study in 2009 found that over 90 percent of respondents in Europe
overestimated the figure by at least a factor of ten. In France, the Nether-
lands, and the United Kingdom, over 40 percent of respondents believed
the reduction of mortality to be 100 or 200 per thousand screened. In the
UK, over 25 percent of respondents exaggerated the mortality benefit by a
factor of 200.25 Such conditions would strongly favor the overdetection of
concerning signs through BSE. Also in 2009, the US Preventive Services
Task Force concluded that the harms of BSE exceeded the benefits. While
breast cancer is a special case owing to its potential lethality, its uniquely
emotive valence, and the ambiguous value of preventive efforts, the risks of
searching oneself for signs of highly publicized conditions apply elsewhere
as well.
Distressing symptoms are omnipresent in daily life . . . Under the influence
of medical scrutiny, public health concern, and media attention, a process
of symptom amplification that alters the perception of these endemic symp-
toms can be set in motion . . . The assumption that one is seriously ill also
heightens self-scrutiny and prompts a confirmatory search for other symp-
toms to corroborate one’s suspicions. Ambiguous sensations that were previ-
ously ignored, dismissed as innocuous, or never consciously noticed are now
interpreted as further evidence of the presence of the suspected disease.30
Medical scrutiny, public health concern, and media attention all drive med-
icalization. Without the first, normal ills couldn’t be framed and treated as
medical issues; without the second, medicine couldn’t go out into the com-
munity to foment awareness—it would have to wait for the community to
walk in the door; without the third, the message of concern couldn’t be
broadcast to the world. But to say that a certain process of distortion alters
the perception of a symptom is to say it alters the symptom, for a symptom
is as perceived. If medicalization triggers the magnification of a symptom,
it does not, therefore, do something to the interpretation of the symptom
while leaving the thing itself untouched.
We may have good reason for not paying too much attention to every-
day ills. A study conducted in London a few years ago found that while
38 percent of women reported indications of one or another sexual
dysfunction—a figure that gives dysfunction itself a kind of ambiguous
80 THE NOCEBO EFFECT
During the past month, how frequently have you experienced the following?
(An answer is required for each, even if it may not apply to you.)
Sexual interest (0) Not applicable
Sexual arousal (1) Much lower level
Sexual activity (2) Somewhat lower level
Sexual satisfaction (3) No change
Sexual anxiety (4) Somewhat higher level
(5) Much higher level
Anxiogenic Ads
Our predisposition to find what we’re cued to look for is borne out in stud-
ies where subjects are asked to rate the accuracy of allegedly specific, but
actually completely generic, descriptions of their own personality. Often, it
seems, such spurious generalities are accepted as valid descriptors. A study
in 1949 served up the following personality sketch, loaded with vacuous
particulars, to each of 39 subjects:
1. You have a great need for other people to like and admire you.
2. You have a tendency to be critical of yourself.
82 THE NOCEBO EFFECT
3. You have a great deal of unused capacity which you have not turned
to your advantage.
4. While you have some personality weaknesses, you are generally able
to compensate for them.
5. Your sexual adjustment has presented problems for you.
6. Disciplined and self-controlled outside, you tend to be worrisome
and insecure inside.
7. At times you have serious doubts as to whether you have made the
right decision or done the right thing.
8. You prefer a certain amount of change and variety and become dis-
satisfied when hemmed in by restrictions and limitations.
9. You pride yourself as an independent thinker and do not accept
others’ statements without independent proof.
10. You have found it unwise to be too frank in revealing yourself to
others.
11. At times you are extroverted, affable, sociable, while at other times
you are introverted, wary, reserved.
12. Some of your aspirations tend to be pretty unrealistic.
13. Security is one of your major goals in life.
Cued Sensations
the name of its inventor, Perkins, and described as a tractor for its alleged
power to draw pains from the body, this cult object of brass was exposed
as a sham by English medical investigators in a series of trials that were no
less probing than the inquest into Mesmerism, though less remembered.
Amid enthusiasm for the healing power of the Perkins tractor, the phy-
sician John Haygarth, assisted by William Falconer (author of a notable
study of The Influence of the Passions upon Disorders of the Body [1788]),
put its powers to the test in 1799 by treating chronic-rheumatism patients
in the Bath General Hospital first with wooden rods made to look identical
to the Perkins instrument, then with the thing itself.
On the 7th of January, 1799, the wooden Tractors were employed. All the five
patients, except one, assured us that their pain was relieved, and three much
benefited by the first application of this remedy . . . If any person would
repeat these experiments, it should be done with due solemnity. During the
process, the wonderful cures which this remedy is said to have performed
ought to be particularly related. Without these indispensible [sic] aids, other
trials will not prove as successful as those which are reported above. The
whole effect undoubtedly depends upon the impression which can be made
upon the patient’s imagination.3
begged the operation might be discontinued. His request was complied with,
and he immediately went to bed, with a pulse at 120. Three quarters of an
hour after (being still in bed) I asked him how he felt himself? He replied,
that “he was in more pain than when the surgeon took five pieces of bone
from his leg in a compound fracture, which he unfortunately met with in
Wales.” It may, perhaps, be thought that he feigned all this. I cannot assert
that he did not; but he could have no point to gain by such a conduct, and he
certainly must have been a very excellent mimick to deceive so many people.
A month later, a patient suffering from “great stiffness and occasional pain
in the shoulder” received the same treatment, with similar results.
In a few minutes he was seized with a tremor in his limb, and so violent a
shivering fit, that it was judged prudent to desist for the present. The next
day I wished to repeat the trial, but he positively refused; alleging that his
arm was quite well, and therefore it would be putting him to useless pain . . .
This case bears a striking resemblance to Peter Seward’s; both wonderfully
depicting the powers of the Imagination. Had these effects followed the use
of the Patent Tractors, it would have been termed metallick influence, and
would have been considered as resulting from their mysterious combination
of metals.
Placebo/Nocebo
The placebo effect refers to a therapeutic benefit arising as a rule from the
expectation of benefit itself, not from the actual composition of a treat-
ment. Thus patients who are told they are receiving a painkiller but actually
get saline solution can still enjoy an analgesic effect owing to the evocative
power of the medical ritual. Note that this model doesn’t entail that the
benefit of the treatment is illusory. On the contrary, placebos can stimulate
the release of the body’s opioids, an effect that can be blocked by the opioid
antagonist, naloxone. Correspondingly, the nocebo effect refers to a harm
arising from the expectation of harm, as when people led to believe they
have suffered a toxic exposure fall ill. While the nocebo effect is ethically
difficult to study and much of the knowledge about it derives from obser-
vation and accidental findings rather than experiment, both the placebo
and nocebo effect appear to operate largely through the power of sugges-
tive messages, enhanced in many cases by theatrical effects. Among the
messages capable of springing to life in our minds and bodies are those
that propel the process of medicalization. “Telling people they are sick
undoubtedly has a strong negative placebo effect.”1
Whether because of the bias in favor of the positive that makes for the
underreporting of harms and adverse events in clinical trials2 or because
the placebo effect represents a potential resource while its twin doesn’t,
the medical literature takes a disproportionate interest in the power of the
placebo.3 In 2012 it was estimated that there are 15 times more articles
on placebo than nocebo in the PubMed database, and well over 15 times
more studies.4 While no law of nature dictates that for every placebo event
there has to be a nocebo counterpart, the nocebo effect is clearly of greater
import than the one-sidedness of the literature suggests. Indeed, the same
substances—nitrous oxide, for example—can produce either painkilling or
painful effects depending on their presentation, and the same experimental
94 THE NOCEBO EFFECT
they were not better.”11 By treating the minor illnesses of Group B as if they
weren’t exactly minor, Thomas could have fed the nocebo effect. In a related
study he got surprisingly good results by giving no diagnosis at all to over a
thousand patients suffering only from “the ordinary ups-and-downs of life,”
patients who “are not ill in the accepted sense of the word . . . and want only
reassurance and support from their doctor.”12
Studies that run treatment against placebo but lack a no-treatment
group don’t allow firm conclusions about the placebo effect. In an unusual
study where the treatment itself was a placebo, and was run against no-
treatment, men with chest pain were given superfluous tests for the sake
of reassurance.13 Yet despite the study’s positive findings, it would be
a bad idea to recommend tests purely for their presumed psychological
benefits. To say nothing of the resources wasted on these exercises, such a
practice would be in defiance of the qualified conclusions and cautionary
language of the study itself, including the reminder that “these findings
may not apply to other patients, chief symptoms, or diagnostic tests.”14 A
study in which men with chest pain have unnecessary tests isn’t intended
to furnish a model for general application. Its application even to its own
subset of patients is limited. As the authors acknowledge, after four months
(the second of two checkpoints) there were no significant differences in
reported pain or activity level between the tested and untested groups,
which suggests that the benefits of the tests were temporary and that those
who had them could have arrived at the same outcome with the passage of
time. Meanwhile, though, their notion of good medicine could have been
recalibrated to include excessive testing, as noted in the study itself. “The
use of diagnostic tests may affect the patient’s future expectations.”15 One
test leads to another, and an anxious patient undergoing successive tests
that reveal nothing may derive from the exercise not relief but frustration
and, indeed, anxiety. Even if they don’t yield incidental findings that are
concerning but go nowhere,16 tests recommended for their placebo value
could actually, like the Thomas study, feed the nocebo effect.
At times, indeed, the placebo effect seems shadowed by its opposite.
It’s known that in antidepressant trials response to placebo runs highest
in those with mild forms of depression—that is, those whose diagnosis as
clinically depressed is most questionable.17 In order to enjoy the placebo
effect, these subjects had to believe themselves depressed in the first place,
which in turn means (as I’ve suggested) that they were led or misled to
believe that their brain chemistry was awry, or their disorder was destined
to continue or even progress unless treated, or their ability to ride out
adversity was compromised. If they hadn’t received a noxious diagnosis
they wouldn’t have qualified as placebo responders. Especially in an era
of rampant overdiagnosis, we err to overlook the harmful potential of the
96 THE NOCEBO EFFECT
diagnostic label: its potential not so much to stigmatize the patient in the
eyes of society (about this the literature has plenty to say) as to plant sug-
gestions in the mind of the patient.
In the fictitious case of the woman released from a repressed history of
suffering by “the magic of the right word”—an act of healing likened to the
ministrations of a witch doctor—diagnosis has therapeutic value in and of
itself. In the same spirit, a paper in the Journal of Family Practice in 1980
argued that diagnosis per se constitutes treatment, first because it enables
patients to make sense of their suffering, second because “diagnosis is often
a crucial factor in encouraging the expression of caring and support from
family and friends,” lastly because “the ability to give something a name
implies the ability to gain control over it.”18 Just as the paper never men-
tions the nocebo effect but discusses its opposite, it never considers the
possibility that a diagnostic label can intensify symptoms. In the case of
chronic fatigue syndrome, for example, some conclude that the label
In keeping with the assumption that nocebo effects have as much claim on
our concern as the placebo effects preferentially discussed in the medical
literature, this chapter looks into the possible harms of diagnosis—in par-
ticular, those that follow from the elevation of common conditions into the
disorders we read and hear about daily in an era of medicalization.
OVERDIAGNOSIS AND ITS HARMS 97
Depression
She worked full time and had no chronic physical or psychological prob-
lems, and no family history of depression. She had been feeling “kind of
OVERDIAGNOSIS AND ITS HARMS 101
down” for 1 month, worse over the past 2 weeks. She complained of loss of
interest and involvement in usual activities, low energy and fatigue, sensitiv-
ity to criticism, poor appetite on some days only, and poor sleep with early
morning awakening. She had occasional trouble concentrating at work but
no excessive crying, confusion, slowing, agitation, distorted thinking or sui-
cidal thoughts.37
With the patient’s diagnosis up in the air, the paper goes on to state,
1990 and will soon be second only to heart disease as a cause of disabil-
ity. Antidepressants and depression-specific psychological treatments are
clearly effective for depression, improving both depressive symptoms and
functional status.
You follow up on Mr P’s “frazzled” comment and learn that he has been
under intense work-related stress. Knowing that recent stress increases
the likelihood of clinical depression, you ask, “Have you been feeling sad
or depressed much of the time?” Mr P has been feeling down nearly every
day for several weeks and on further questioning meets criteria for major
depression.43
The information provided gives no idea of how Mr. P. meets at least three
of the DSM criteria, featured in a table in the paper itself.
In all probability, a patient who reported feeling depressed “nearly every
day for several weeks” would be diagnosed as depressed then and there,
whether meeting all the other criteria or not—and whether he or she felt
depressed “most of the day,” as DSM language specifies. In fact, DSM-V
allows clinicians to waive diagnostic criteria at their own discretion. States
the introductory essay on the concept of a mental disorder, “Clinicians
may . . . encounter individuals whose symptoms do not meet full criteria
for a mental disorder but who demonstrate a clear need for treatment or
care. The fact that some individuals do not show all symptoms indicative of
a diagnosis should not be used to justify limiting their access to appropriate
care”—a grant of interpretive liberty that moots the painstaking specifica-
tion of DSM language at one stroke.
In sum, then, the textual criteria for depression may or may not be
observed in practice—and at that, these criteria include common com-
plaints, classify normal sadness as a clinical symptom, and don’t even
require sadness for a finding of depression. At once scholastically pre-
cise and expansive, such criteria make for inflated estimates, which feed
the assumption that depression is insidiously common (the assumption
104 THE NOCEBO EFFECT
framing the sketchy case of Mr. P.), which leads to the all-too-common
prescription of favored drugs.
Being diagnosed with depression when you aren’t depressed, but dis-
play symptoms misleadingly similar to depression, is a standing possibility
for pregnant women. Some 35 percent of pregnant women are reported to
show depressive symptoms while perhaps 10 percent suffer from depres-
sion per se, a disparity that suggests that symptom-counting—the basis of
the DSM diagnostic system—is better at producing diagnoses than at dis-
tinguishing facially similar conditions. Say a pregnant woman who doesn’t
seem to herself depressed, because she isn’t, learns from a doctor that her
answers to a questionnaire indicate depression. Whether or not informed
of the medical literature alleging the risks of untreated depression in preg-
nancy, such a patient could be shocked into taking antidepressants despite
the questionable wisdom of doing so.44
The most common treatment for diagnosed depression, antidepressants
carry their own risks to the patient’s well-being by way of their diverse
side effects including intestinal ills, weight gain, drowsiness, insomnia, and
reduced libido. The tendency to play down such events is illustrated by an
incidental comment in a generally fair-minded investigation of the theory
and practice of American psychiatry. Discussing the superiority of Prozac
to earlier drugs, the author notes that
with Prozac, people actually lose weight (at least for a while), and the major
side effect for most people seems to be jitteriness and, for a significant per-
centage of men, impotence. Prozac has meant that taking psychiatric medi-
cation for common anxiety and depression has become, practically speaking,
risk free.45
How a drug can be practically risk-free when a lot of the men taking it expe-
rience impotence is simply baffling. This isn’t to diminish the importance
of nonsexual side effects. A 2005 study found that treatment with antide-
pressants “inadvertently reduced the self-worth in patients non-receptive
to antidepressant medication,”46 while a related analysis points out that
many patients in the study were treated with antidepressants against their
wishes and saw their “physical role functioning” deteriorate.47 A medica-
tion whose common side effects include both drowsiness and intestinal
problems might well impair the ability to perform everyday tasks. A study
by Spitzer and others found “substantial” deterioration of “role function”
associated with insomnia, fatigue, abdominal pain, and dizziness.48 Con-
sidering that any one of these symptoms can degrade the ability to work
and that SSRIs have more than one side effect, it stands to reason that the
drugs can damage self-respect.
OVERDIAGNOSIS AND ITS HARMS 105
that, more and more, seems natural—not made by human hands.”53 If pro-
fessionals can be convinced of the independent reality of their own con-
structs, patients can surely be persuaded that their diagnosed disorder is
real. Convinced doctors make for convinced patients. As we know, DSM-
III’s master-builder, Robert Spitzer, grants that DSM language influences
not only how third parties make decisions, but how patients see them-
selves. “It serves to define how researchers collect their samples, what con-
ditions insurance companies will reimburse, what conditions courts and
social agencies treat as illnesses, and how individuals themselves interpret
their emotional experiences.”54
This comment appears in a laudatory introduction by Spitzer to a book
that makes a compelling case that depression is massively overdiagnosed.
The unavoidable inference, that those wrongly diagnosed are led to misin-
terpret their own experience, is not drawn by Spitzer.
Prostate Cancer
Why better treatment would reduce the numbers who undergo treatment
in the first place is far from clear. For now, in any case, the most common
treatment of prostate cancer, surgical removal of the prostate, entails a high
risk of such woeful outcomes as incontinence and impotence. It’s hard to
imagine that many would volunteer with eyes open for a test where 48 or
indeed 33—the revised figure given in 201257—ran the risk of impotence
for every prostate-cancer death spared, which suggests why informed con-
sent for PSA testing has been scanted in practice while affirmed on paper.
OVERDIAGNOSIS AND ITS HARMS 107
ADHD: Adults
the “general symptom pool” of normal persons, along with such other pro-
saic ailments as headaches and disturbed sleep. So too, a review done in
Germany in 1999 rated “clouding of consciousness” as the most common
subjective adverse reaction to placebo,64 which probably means the inac-
tive medication picked up and amplified symptoms that were already there,
not that it somehow caused them. Evidently symptoms of hyperactivity,
too, are not particularly rare. The ordinariness of the elements of ADHD
gives the faker material to work with and makes the work easy, in that he
or she doesn’t have to go to the trouble of inventing symptoms from zero.
In the case of a person without dishonest intent, the preexistence of
symptoms like poor concentration or jumpiness means that the folklore
and publicity surrounding ADHD (also ubiquitous) act like the context of
“overwhelming information” that makes it possible for someone in a minor
accident to get caught up in chronic whiplash.65 Investigators of functional
somatic syndromes emphasize the importance of charged reports and
public controversy in disseminating not only information about the disor-
ders but the disorders themselves. Tapping into the appeal of folklore and
the authority of medicine at once, public talk about ADHD can heighten
ordinary symptoms precisely by giving them such an evocative label and
such a powerful validation. In this sense, the capital letters and bold cari-
cature on the cover of Time model the process of highlighting symptoms
and elevating their significance. A suggestive circumstance strengthens the
analogy between FSS’s and ADHD: Patients in the former category some-
times come to the doctor armed with their own diagnosis. (An important
element in the syndromes in question is the belief that you have them,
a belief supported by a kind of literature and confirmed by the evidence
of your symptoms.) Correspondingly, many patients are said to diagnose
themselves with ADHD and see a doctor to have the diagnosis made offi-
cial. “Frequently, adults who encounter a description of the disorder sense
that ‘this is me’ and go on to seek professional confirmation of their new
identity.”66 That one’s identity should be bound up in a diagnosis suggests
an overinvestment in the disorder at issue, and the sort of attention to
symptoms implied in the act of self-diagnosis is but a step away from the
anxious vigilance that fuels the promotion of ordinary events into clinical
symptoms.
If a key ingredient of a syndrome like ADHD is the belief that you have
it, what happens if you’re falsely diagnosed with ADHD as a result of loose
criteria and slippery diagnostic thresholds? While DSM-V was in prepara-
tion, the architect of DSM-IV and a coauthor argued that overdiagnosis
would certainly ensue if proposed changes in ADHD language were put
in place.67 (Rates of childhood ADHD had already shot up as a result of
textual changes introduced into DSM-IV.) Yet when detailing the harms
110 THE NOCEBO EFFECT
ADHD: Children
The social underpinnings of ADHD may bear on the results of the nota-
ble MTA trial (Multimodal Treatment Study of Children with ADHD), in
which ADHD treatments including an intensive medication regimen and
ordinary care were run against each other but not against placebo, so that
the placebo effect was built into the results. Though most in the ordinary-
care group were also taking stimulants, children receiving medication as
part of the study protocol showed better results at 14 months, and the text
of the study suggests why. Those randomized to the medication group had
a degree and kind of medical attention that transcended routine care. For
example, “during half-hour monthly medication visits, pharmacotherapists
provided support, encouragement and practical advice,” as well as adjusting
the child’s medication if necessary.81 So closely was the medication group
monitored that their saliva was tested. Whereas children and families who
refused treatment were “encouraged to reconsider their decision throughout
the study,”82 those who did take part were under careful, continuous observa-
tion that heightened compliance with the protocol and in all likelihood was
therapeutic in itself. When the protocol came to an end after 14 months and
the study became “naturalistic,” the superior results of the medication group
began to decay; by two years they had halved and by three years vanished.83
While difficult to interpret, the narrative of lost efficacy suggests that as the
elaborate ritual structure supporting the medication collapsed, the value of
the medication sank with it. The subjects took their stimulant less religiously
and their outcomes ceased to show any relation, except possibly a nega-
tive one, with pill intake at all. The MTA story reminds us that ADHD is as
much a social as a biomedical construct, and that offstage factors including
the character of care may have much to do with the efficacy of the pills that
engross public attention. The authors of a three-year follow-up on the MTA
trial offer the Tolstoyan conclusion that the improvement over baseline seen
in all treatment groups despite the waning differences between them may be
due to nothing more or less than the passage of time.84
Some consider that the placebo effect operates by something like a self-
fulfilling prophecy, as when a subject led to expect saline to act like a pain-
killer enjoys that very outcome.85 If a message can activate a neutral solution,
what of a message delivered to a child every day with his ADHD medication
and reinforced by one and all? That this could contribute to badly regulated
behavior is no remote possibility. A paper in American Psychologist by no
means critical of ADHD as a diagnostic category concedes that
the messages that attend stimulant therapy may be beneficial, enhancing the
child’s sense of accomplishment and self-perceived efficacy and diminish-
ing parental feelings of guilt and futility. Or, the effects may be unsalutary,
confirming preexisting assumptions that the child is and will remain unable
OVERDIAGNOSIS AND ITS HARMS 115
to take control of his own behavior without external aids and constraints . . .
Until more information is available about such effects, as well as about long-
term outcomes and individual differences, the best clinical strategy would
appear to be one of judiciousness and caution.86
that you can’t regulate yourself without methylphenidate does not promote
and could undermine the practice of self-control. If screening programs
“potentially influence our self-concepts . . . and offer a new form of social
penetration of our private emotions,”92 how much more is this true of a
diagnosis received in childhood and reinforced daily for years on end.
Medical thinking duly cautious of harm wouldn’t fail to factor this risk into
the evaluation of the ADHD diagnosis.
According to DSM-III-R (1987), ADHD “is common; it may occur in as
many as 3% of children.” The commonness of ADHD has since shot up to
the point that it’s now estimated that 11% of American children, and nearly
20% of American boys, will be diagnosed with ADHD by the age of 17.93
That’s what medicalization comes to.
Erectile Dysfunction
Patients who expect distressing side effects before taking a medication are
more likely to develop them. Such negative expectations make the individual
more likely to notice and attend to new or unwelcome sensations; interpret
preexisting, ambiguous, and vague sensations unfavorably and attribute
them to the medication.96
a post hoc approach, selection bias, and recall bias; admits that subjects
were recruited from one of the authors’ practices and from a Propecia help
forum; and notes that subjects attributed to the drug not only their sexual
complaints but miscellaneous problems from fatigue to “cognitive difficul-
ties.”97 The reason blinded, placebo-controlled trials are run in the first
place is to correct for the biases and loose causal attributions built into this
study.98 Information about Propecia is thus falling into the same pattern of
repetition and exaggeration seen in other syndromes, with the difference
that in this case the cause of the alleged epidemic has already been identi-
fied. Given that noninflammatory consent language alone can increase the
side effects of finasteride, it’s entirely possible that the scare stories now
surrounding Propecia can do the same or worse.
In this matter we might take advice from Montaigne’s essay “On the
Power of the Imagination,” much of which is devoted to the topic of impo-
tence. According to Montaigne, a fear that has become exceptionally
vivid, that takes command of the imagination, is enough to bring on this
affliction.
The horror stories about Propecia now in circulation echo the tales of
magic impotence that coursed through Montaigne’s world, at once base-
less and potentially effectual. The deep but skeptical interest in mind-body
transactions at work in “On the Power of the Imagination”—an interest
that leads Montaigne to reflect on the grounds of his own knowledge—is
akin to the attitude inspiring the study of placebo and nocebo effects today,
just as Montaigne’s fascinated but not credulous reports of exotic practices
resemble the reserved judgments of the nocebo literature regarding such
practices as voodoo death.
Adjustment Disorder
Given the multitude of studies where patients serve as the subjects, it seems
poetically just that researchers turn the tables now and then, putting doctors
themselves under investigation. Notable among these studies-in-reverse is
OVERDIAGNOSIS AND ITS HARMS 119
Though Paxil is among the more poorly tolerated drugs of its class,102
nowhere in the study in question do the doctors mention or appear to con-
sider the known effects of the drugs they prescribe for adjustment disorder.
Still more disturbing is the assumption that adversity constitutes a clini-
cal problem in the first place and that life itself is appropriately subject to
medical management.
5
Name Games
“Just like oatmeal.” This statement will result in good physicians using infe-
rior drugs that might not even work to save your patient’s lives [sic]. Oatmeal
indeed—and for your information, I always buy “Quaker” oats and I bet you
do too. If not you deserve the mush you might get.18
126 THE NOCEBO EFFECT
The power of labels is brought out by the chair of the DSM-IV Task Force
in a discussion of three late-nineteenth century medical fashions, one of
which enlisted William James. Of hysteria, multiple personality, and neur-
asthenia, Allen Frances writes,
convinces the patient that it’s present in full force and its symptoms somehow
predetermined. Much as the information that a drug is going to be adminis-
tered can induce the drug’s side effects even if it’s not in fact given, and much
as the side effects of placebos in clinical trials often track those of the drug
under study as listed in consent forms—that is, anticipated side effects—so the
expectation that our symptoms will conform to a certain official pattern can
strongly color our experience, whether or not we have an investment in the
sick role. A named disorder, underwritten by medicine and validated by com-
mon practice, has a way of becoming real. Even though the diagnoses codified
in the DSM were wrangled over before being set down in print in 1980, over
the intervening decades most have come to be accepted as genuine entities.
Writes the former director of the National Institute of Mental Health, “Dur-
ing the development of DSM-III, the process of splitting larger clusters into
more specific disorders elicited substantial debate and disagreement, but these
disagreements have largely disappeared, not because of scientific progress, but
rather, I would argue, because the named disorders became passively accepted
and reified.”32 If the mystique of a name can so enchant professionals that they
forget that it was ever instituted by human beings, a name can also influence
patients to fill out a disorder by producing the sort of symptoms this strangely
reified agent seems to call for. This is all the more likely if patients tend to pres-
ent “symptoms that their physicians will find convincing.”33
As we know, even in a state of health there are plenty of symptom-like
events available as raw material for a diagnostic label to organize. In that
many of these are ambiguous and diffuse, more like uneasy sensations than
determinate entities, a label that summons them into prominence and
gives them definition can act as a signal-booster. A labeled disorder enjoys
the prestige of specificity, and under its influence symptoms themselves
can become more specific.
In order to be eligible for amplification, an event has to be identifiable
as a signal in the first place. Of the many healthy volunteers, including
medical students, who in one study checked a box indicating that they
recently experienced “irritability,” some might have interpreted the term
to mean short temper or crankiness and some an ill-defined state of over-
excitement akin to distraction or loss of concentration.34 Irritability in the
latter, medical sense is so nebulous that you might not identify it at all
unless presented with a ready-made label for it. Today, someone diagnosed
or self-diagnosed with “irritable” bowel may have symptoms little different
from someone who walks the streets without the diagnosis, while faulty
concentration may be associated with ADHD. The instant recognizability
of the ADHD label, a feature it shares with a brand name or popular logo,
serves to sharpen the outlines of symptoms or quasi-symptoms so prosaic
as to merge with normality. That in itself is signal amplification.
NAME GAMES 131
An Iatrogenic Label
Such is the evocative power of a label that it can persuade us of the reality
of a nonexistent deficiency or the specificity of a syndrome with no labora-
tory findings and many generic symptoms, like fibromyalgia. As with the
patient who is convinced he suffers from whiplash and carries that diagno-
sis with him to the doctor, the label itself appears to be crucial to a number
of syndromes—the element that binds the entire experience together and
gives it definition and life.
In the controversial instance of Multiple Personality—a term used
both in DSM-III and popular lore—a special set of labels seems to be of
the essence. The various inmates inhabiting a patient with a plural iden-
tity don’t appear fully present and haven’t convincingly established their
existence until each one has a proper name. As noted by Allen Frances,
most cases of this disorder “were induced by the efforts of well-meaning
but misguided therapists who were as clueless about what was going on
as were the patients. It is not hard for a suggestible therapist treating a
suggestible patient to turn any run-of-the-mill psychiatric problem into
MPD [Multiple Personality Disorder].”50 Central to the rite of conjuring,
eliciting, or producing multiple personalities is the act of naming. Thus,
people diagnosed as multiples call themselves by one name or another
depending on which personality happens to be on stage at the moment.51
But while names serve as the keystone solidifying the entire fictitious
arch of multiple personality, they’re unlikely to spring from the patient
spontaneously.
NAME GAMES 135
Stencil Work
While it’s often said that patients want a label for their condition, the fact
is that a label can also mobilize symptoms, provide a suggestive script for
a disorder’s performance, and convey an impression of specificity where
none exists. For the doctor, too, a label has risks, in that once the patient’s
condition has acquired a name, it’s hard to revoke it and reopen the case.
Say amorphous ills are organized into symptoms and the symptoms are
referred to a disorder presumed to lend itself to medical management.
A disorder created in this way “becomes difficult to unmake.”61 Names
reify—not only in medical records (where a diagnostic label may live on
for years even if erroneous)62 but in the minds of doctor and patient. One
reason for the counterintuitive popularity of depression may be just this,
that the diagnosis packs so many potentially shapeless distresses and dis-
contents into a single all-too-convincing package.
A curious study of misdiagnosis brings out the power of depression to
impose a ready-made form on ambiguous cases. In 1994 some four hun-
dred primary-care doctors were asked in a survey if during the preceding
two weeks they had recorded an alternative diagnosis “when they recog-
nized that a patient met DSM-III-R criteria for major depression”—or so a
paper in Archives of Family Medicine reports.63 Actually the doctors were
asked more ambiguously, “In the last 2 weeks, how many . . . patients do you
estimate met DSM-III-R criteria for major depression? (Criteria listed on
back),” suggesting that both the number of patients and the degree of their
match with DSM criteria were estimates. Fully half of the respondents said
they had given an alternative diagnosis, the reason most frequently cited
being that they were uncertain about the diagnosis of depression. (Elimi-
nated from the study were 43 doctors who reported not seeing any patients
who met the DSM language.) However, the authors insist that the real rea-
son for the alternative diagnosis is that the doctors wanted to ensure that
the patient was reimbursed or spared the stigma that attaches to depres-
sion, or the like. The authors are so certain that anyone who exhibits some-
thing approximating DSM depression is depressed that they simply write
off their own respondents’ claim that they were uncertain.
How could one be uncertain about whether a patient really qualified
as depressed? The alternative diagnoses provided by the doctors suggest a
good reason: The presenting symptoms may be largely generic. The most
frequent non-DSM diagnoses were “fatigue/malaise” (60 percent) and
“insomnia” (44 percent)—complaints that in and of themselves will take a
patient almost half way to a DSM diagnosis of Major Depression, though
even in these cases the symptoms may not have been present “most of
the day, nearly every day” over the same two-week period as required by
138 THE NOCEBO EFFECT
An individual who has marked loss of interest and pleasure (not depressed
mood) with five of the nine symptoms (e.g., hypersomnia, weight gain) pres-
ent to a marked degree nearly every day for 2 weeks might be neither sig-
nificantly distressed nor role-impaired. Such an individual may still have a
depressive disorder.64
Precisely because the patient doesn’t seem any the worse for meeting the
paper criteria for depression, a doctor might well decide against apply-
ing such a potentially depressing label. Spitzer considers not diagnosing
depression in this case as a false negative. However, another of the movers
of the DSM-III revolution takes the position that in some cases a respon-
sible clinician may not diagnose a disorder even though it meets DSM
criteria.
When the Manual prescribes a diagnosis the clinician does not feel is pres-
ent, he may simply not follow the rules. It is formally possible for a person
to meet DSM-III criteria for alcohol dependence with only two symptoms,
one indicating tolerance, another indicating social problems. Anyone who
drinks heavily and regularly will probably develop tolerance, and if he or she
has a relative or friend who regularly tells him that they do not like his drink-
ing so much, he can qualify as alcohol dependent, because criticism from
family members over a one-month period plus tolerance equals dependence.
If there are no other symptoms, and no more serious impairments than fam-
ily complaints, many clinicians would be reluctant to make a diagnosis of
alcohol dependence.65
Another study in which Spitzer took part finds that the presence of a
common physical symptom like fatigue, insomnia, or headache sharply
raises the odds of a diagnosis of mood or anxiety disorder (which in the
case of the first two symptoms is unsurprising, both being on the DSM list
of secondary criteria for depression). But what elevates the risk of diagno-
sis even more is a somatoform—that is, psychogenic—symptom, defined
as being “physically unexplained.”66 However, many ills, including those
just named, don’t necessarily have detectable causes. That a symptom is
medically unclear67 doesn’t establish the person who has it as a somatizer.
(The word suggests someone oversensitive to bodily problems and prone
to complain about them, but in the study in question we aren’t told why the
patients happened to be visiting their doctor when they were asked to fill out
a questionnaire inquiring about 15 common symptoms.) The strong link
between somatoform symptoms and the diagnosis of depression may say
more about the doctor than the patient, as it’s the doctor who decides that
because the symptom lacks a physical explanation it must have a psychic
one. While the framers of DSM-III broke with psychoanalytic tradition in
refusing to speculate about causes—indeed, a shared dislike of etiological
theorizing gave them a strong sense of group identity—particularly in the
case of depression the theory of covert causation remains in effect.
There’s much to be said for the ability to preserve uncertainty, to keep
an unclear case open by not filing it under a ready-made category or con-
cluding that a mental disorder has produced a symptom because no physi-
cal cause presents itself. A trenchant medical paper contrasts the work
and style of the primary-care doctor, who commonly sees problems that
don’t really lend themselves to diagnosis, with those of the specialist whose
patients are referred just because they’re probable candidates for diagnosis.
It’s a case of “the family physician as the simple humanist who prefers action
to words” against “the specialist as a single-minded biomedical scientist”—
roughly, Montaigne vs. Descartes.68 In accordance with this contrast of
methods, the primary-care doctor may take practical steps before resolv-
ing the niceties of diagnosis. So it is that the author approves prescribing
antidepressants to determine if the patient is depressed, reasoning that “‘If
it gets better with X, then it must have been Y’ is a diagnostic strategy that
makes sense at the primary care level.”69 But the patient who gets better
with an antidepressant might get better without one. In a study reported
by K. B. Thomas in 1974, 82 percent of patients with common, medically
unexplained complaints like those now packaged as depression improved
despite being treated with nothing but reassurance and, in most cases,
placebo.70 Antidepressants have a high placebo index of their own, and
patients treated with such drugs experimentally, to ascertain if they were
depressed in the first place, could have their understanding of themselves
140 THE NOCEBO EFFECT
altered by the trickery of the placebo and the mystique of the diagnostic
explanation.
The temptation to interpret events in the light of pseudo-explanations is
a strong one. A disciple of Wittgenstein who became a psychiatrist cautions
against false explanations and conceptual foreclosure, noting “the danger
of my case history taking to become stereotyped.”71 Wittgenstein fought
against even his own tendency to fall into traps of explanation and was in
turn influenced by Tolstoy, who underscores the attraction of stereotype
in a famous simile in War and Peace likening the neat explanation to a
stencil. We assume the campaign of 1812 was dictated by Napoleon’s will,
blotting out anything and everything that doesn’t fit this narrative, “just as
in stencil work this or that figure comes out, not because the colour was
laid on from this side or in that way, but because it was laid on from all
sides over the figure cut in the stencil.”72 An indeterminate distress or ill
is like an amorphous color-field that takes on a legible shape as a result of
a figure laid over it. (Thus, if a doctor goes down a checklist of symptoms
for IBS, with the patient endorsing them one by one, the patient becomes
an irritable bowel case even though she might well have other symptoms
belonging to kindred syndromes, including, say, noncardiac chest pain.)73
A diagnostic label, on this showing, is the figure that provides a dramatic
illusion of definition because all that doesn’t fit remains outside the field of
vision. A stenciled figure makes sense to our meaning-seeking minds in a
way an irregular shape doesn’t.
In medical scenes in The Death of Ivan Ilych, Anna Karenina, and War
and Peace itself, doctors act like human stencils, producing or not pro-
ducing an illusion of mastery by imposing their ready-made ideas on
the cases before them. Ivan Ilych’s doctors perform diagnostic charades,
labeling his ailment “floating kidney” or “vermiform appendix” even as he
deteriorates under their eyes, ravaged by a disease Tolstoy leaves name-
less. In fact, the doctors in this tale behave much like those described
by Montaigne three hundred years before, contradicting one another
and putting on imperious airs as if a behavioral pretense could make up
for their ignorance. (“The art of medicine,” writes Montaigne, “has not
reached such certainty that, no matter what we do, we cannot find some
authority for doing it . . . If your own doctor does not find it good for
you to sleep, to use wine or any particular food, do not worry: I will find
you another who does not agree with his advice.”)74 In portraying medi-
cal men who might as well have been practicing in the sixteenth century,
Tolstoy contests the ideal of medical progress that each of them implicitly
professes to represent.
While satirists had long derided the pretensions of medicine, Tolstoy’s
critique of medical arrogance is especially strong, perhaps because it was
NAME GAMES 141
during his lifetime that medicine first won the status of a science; it there-
fore seemed a particularly audacious imposture. Predicated on a strong
analogy between the patient and the doctors who treat him, The Death of
Ivan Ilych suggests an analogy, too, between the protagonist as one of the
“new men” called into being by the legal reforms of 1860s and the doctors
intensely proud of the newly scientific status of medicine. Not only are Tol-
stoyan doctors like these puffed up with vanity as their tribe had long been
in the eyes of literature, but they profess to be at the forefront of knowl-
edge itself. Tolstoy was particularly critical of the mystique of specificity,
as in the caustic sketch in Anna Karenina of a certain medical celebrity
who puts on airs of “special knowledge.”75 Though Kitty, the patient, shows
signs of consumption, what is consuming her is a matter outside the pur-
view of medicine. If medicine emboldened us to think of a human being as
“a possible object of positive knowledge—a living individual whose body
and mind could be understood by scientific reason,”76 Tolstoy denies this
reduction of a person to a knowable object with every bit of the consider-
able power at his command.
So it is that Natasha’s grief in War and Peace mocks the efforts of the
learned doctors called in to treat her. “Doctors came to see her singly and
in consultation, talked much in French, German, and Latin, blamed one
another, and prescribed a great variety of medicines for all the diseases
known to them,” which is a sure sign they know nothing about what’s
wrong with her. It’s at this point that Tolstoy delivers the astonishing obser-
vation that Natasha’s disease eluded medical knowledge “as no disease suf-
fered by a live man can be known, for every living person has his own . . .
disease, unknown to medicine.”77 Only in the latter third of the nineteenth
century did diseases come to be pictured generally as entities that exist
in independence of particular persons.78 In the story of Natasha’s private
disease Tolstoy challenges this notion explicitly, even polemically, just as it
begins to be established.
By letting the reader but not the doctors into the story behind Natasha’s
ailment, Tolstoy reminds us of the narrowness of an exclusive preoccupa-
tion with presented symptoms. That the doctors have no idea what really
ails Natasha doesn’t stop them from dosing her with pills and powders. She
is given the impression that her pain
would soon pass if only the coachman went to the chemist’s in the Arbat and
got a powder and some pills in a pretty box for a ruble and seventy kopecks,
and if she took those powders in boiled water at intervals of precisely two
hours, neither more nor less . . . Even to Natasha herself it was pleasant to see
that so many sacrifices were being made for her sake, and to know that she
had to take medicine at certain hours, though she declared that no medicine
would cure her and that it was all nonsense.79
142 THE NOCEBO EFFECT
While the pills and powders are nonsense, they serve to rally the family and
provide a focal point for its endeavors, and in any case stand as something
definite in a menacing sea of the unknown. How unlike her condition itself
is the box the pills come in, the first unspecific and impossible to slot into
medical categories, the second as specific as can be and appealingly stan-
dard in shape. While the pills within are of no medical value, the box itself
is perfectly adapted to its job of mobilizing the family on Natasha’s behalf—
organizing their concern and “sacrifices.” As Tolstoy makes plain, attend-
ing to Natasha in this way solaces the family as well as providing her some
comfort. In effect, then, the rituals of care buy time for the patient to return
to herself, as she eventually does. “Natasha’s grief began to be overlaid by
the impressions of daily life, it ceased to press so painfully on her heart,
and she began to recover physically.” While the pills and powders engross
everyone’s attention, the most unobservable, the least salient of forces, time
itself, heals Natasha.
6
With suicidal thoughts or attempts listed in the DSM among the symptoms
of depression itself (arguably the gravest of the lot), one would not expect
an antidepressant to increase this risk in any group, let alone children.
As it happened, my health plan also recommended prostate-cancer
screening for all men over age 50 and offered it free. At the time I was not
only unaware of any argument against early detection of cancer, I couldn’t
have conceived of one. The principle that early detection saves lives seemed
as intuitive as an axiom, or rather it would have seemed so if I’d thought
about it at all. Like millions of others, I therefore signed up for PSA testing.
144 THE NOCEBO EFFECT
Only later did I come to learn that the wisdom of prostate-cancer screening
was hotly debated in the medical literature right from the beginning of the
PSA era 25 years ago and that the harms of PSA testing were and are more
certain than its benefits. I had no idea a screening test could harm. Once
that fallacy was dispelled, I asked my Wellness Program to provide infor-
mation about PSA testing, both pro and con, to eligible men. The admin-
istrator flatly refused.
This, it seemed to me, showed a closed mind. The claim that antidepres-
sants saved lives closed debate. But even as I came to see that the campaigns
against depression and prostate cancer both used inflamed arguments and
appealed to fixed ideas, I failed to appreciate how strong the family resem-
blance between the two really is.
That a number of syndromes share generic symptoms like fatigue, sleep
difficulties, headache, joint pain, and problems with concentration sug-
gests that they grow out of the same large set of ills inherent in normality.
This chapter focuses on two conditions that, by contrast, have no common
substrate whatever but have been subject to similar patterns of overdiag-
nosis. Except that a prostate cancer patient might become depressed, the
only thing depression and prostate cancer have in common is the crusad-
ing dynamic that has made an epidemic of both.
Blurry Targets
In October 2011, days after the US Preventive Services Task Force con-
troversially recommended against routine screening for prostate cancer,
the National Center for Health Statistics reported that the use of antide-
pressants quadrupled over the last two decades. Both stories were widely
reported, though no one to my knowledge connected them or drew the
analogy, which was there for the making, between soaring treatment of
prostate cancer and soaring use of antidepressants over the same period.
While the launch of SSRIs and the launch of PSA testing both date to 1987,
more is at work in the analogy than temporal concurrence. In both cases
the condition in question is fundamentally ambiguous, in both cases exist-
ing means of detection are insufficiently discerning, in both cases large
populations are screened anyway. Indeed, in both cases screeners have gone
out into the community rather than waiting for the community to present
itself. In both cases people have been diagnosed by the millions. Just as PSA
testing assumes that cancer uncovered as a result of screening behaves like
cancer that manifests itself clinically, those who screen the general popula-
tion for depression take for granted “the myth of the equivalence of com-
munity and clinical diagnoses.”2 While PSA testing “can tap into a very
large reservoir of clinically silent prostate cancers,”3 testing for depression
BEWARE WHAT YOU LOOK FOR 145
that “it will inevitably happen from time to time that we will be asked and
expected to remedy the normal discontents and disappointments that are
part of our common human life.”12
As a result of mass screening with crude instruments (the analogy with
PSA testing is striking) as well as overzealous diagnosis, all abetted by a
strong trend toward the medicalization of life, the doctrine that clinical
depression is a common affliction of the American population has estab-
lished itself and gone into general circulation. Defining common ills as
symptoms and ignoring the sources of sadness in life itself, the DSM crite-
ria for Major Depression authorize the interpretation of distress as a men-
tal disorder. As we know, you don’t even have to show depressed mood
to qualify as depressed under the DSM; “markedly diminished interest or
pleasure” will do. In fact, a screening questionnaire designed by Spitzer and
others to cut through the complexities of the DSM criteria omits depressed
mood altogether.13 While depression without feeling depressed verges on
paradox,14 such a category serves to open up the diagnosis of depression
to more cases. It appears that patients with depression-like symptoms are
considerably more common than those with depression per se—two to
three times more common, by one estimate.15 If there were a way to elevate
these ambiguous cases to clinical status, the depressed population could be
greatly enlarged. The DSM criteria provide a way. That they certify so many
as depressed helps explain why they prove useful to drug makers interested
in exploiting the market for antidepressants to the limits of the possible, as
well as why they have remained essentially unchanged for 30 years. In turn,
antidepressants have been approved on the basis of studies that generated
statistically but not clinically significant findings—while prostate cancer is
commonly treated with surgery or radiation whether or not it has clinical
import.
Like the idea that prostate cancer might be a normal condition of life,
the possibility that much Major Depression might not be a disorder at all
has been drowned out by medical activism.
Imperceptible Danger
In order to qualify for Major Depression, you have to meet either (1) or (2)
along with four secondary criteria out of a possible seven. The DSM-III criteria
made no provision for depressed mood or diminished interest based on the
judgment of third parties, though the text did note that a child might not com-
plain of depressed mood and that one who suffers a loss of interest or pleasure
“may not complain of this or even be aware of the loss.” The alternative sources
of evidence written into the DSM-IV and DSM-V depression language are in
the tradition of the alternative possibilities originally installed in DSM-III to
allow for the diagnosis of depression in the absence of depressed mood.18
In common parlance a medical symptom refers to a “subjective indica-
tion,” which complicates the notion of a symptom that reveals itself to a
third party but not its owner. The subjectivity of symptoms survives in the
medical distinction between a symptom and an objective sign, usually a
laboratory finding. In the case of the secondary symptoms of depression,
it’s clear, anyway, that you can’t think about suicide unknowingly, or at the
other end of the scale, suffer from insomnia unknowingly.19 Certainly I
can feel empty and hopeless and not label my condition as Major Depres-
sive Disorder, but that isn’t to say I’m unaware of feeling as I do. Cogni-
tive research has identified the sort of thoughts that stream through the
mind of a profoundly depressed person, thoughts such as “I’m worthless,”
“I don’t deserve to be loved,” and “Life isn’t worth living.”20 That you can
BEWARE WHAT YOU LOOK FOR 149
be under siege by thoughts like these and not know it seems improbable.
Imputing depression to patients who don’t claim to be depressed stands
in contrast to the ordinary practice of basing diagnosis on a patient’s own
words, a practice so essential that history-taking is accounted “the most
powerful diagnostic tool available to the internist.”21
Those who maintain, even so, that the depressed can be unaware of
their own state have an argument at the ready: that I fail to recognize my
depression because I fear the stigma of mental illness. This claim not only
presumes the point at issue (presumes I am depressed) but attributes a fear
that may or may not exist. Some 20 years ago a prominent psychiatrist
protested that
the troubled person is led to believe that he can’t help himself and must seek
out a professional healer when confronted with distress related to every-
day problems of living. His confidence . . . is eroded because he accepts the
view that emotional disturbance arises from forces beyond his grasp. He
can’t hope to understand himself through his own efforts, because his own
notions are dismissed as shallow and insubstantial.22
providing the statutory loss of interest or pleasure. Clearly, it’s possible to ele-
vate common problems into matters of psychiatric significance; clearly too, the
availability of popular pills for depression increases the diagnostic temptation.
Given that many patients seen in primary care appear to meet some but not
all of the criteria for DSM depression,25 a doctor could also inflate symptoms
with no great difficulty to make up the difference. (As for the clause stipulating
that the symptoms must cause “clinically significant distress or impairment” to
warrant a diagnosis, you could simply say that by definition the symptoms of
depression do one or the other, or both.) Yet patients who complain of fatigue
or insomnia only to learn that their real problem is depression have received
an unsettling message they are unlikely ever to forget. A normal patient given
a diagnosis “may never be quite the same again.”26
In some cases other than depression it would surely be regarded as pre-
sumptuous for doctors to decide that something not deemed a problem by the
patient is nevertheless exactly that. It appears many women have what could
be defined as a sexual dysfunction but don’t judge it a problem.27 If doctors
were to insist on regarding these women as candidates for treatment anyway,
we wouldn’t hesitate to fault them for overstepping their authority and distort-
ing a common condition into a disorder.28 Why is it any less presumptuous for
medicine to diagnose patients who don’t complain of depression? Nor can it be
said that it does no harm for medicine to simply define the reluctant as candi-
dates for treatment. Whether owing to the prestige of medicine, an instinct of
compliance, or some other reason, some patients who don’t want to be treated
may accede when treatment is pressed on them.29 In a study of treatments of
depression in which one quarter of the eligible population refused to partici-
pate,30 half of those who did participate were “unreceptive” to antidepressants
but treated with them all the same, in many cases with adverse results.
The medicalization of sadness began with efforts to screen the commu-
nity at large for depression, and the rationale for mass screening crumbles
unless the condition screened for is potentially or actually invisible to the
screened population. Mammography can detect incipient forms of cancer
that would otherwise escape observation, and the Depression Awareness,
Recognition, and Treatment Program launched in 1988 virtually alludes
to the Breast Cancer Detection Awareness Program launched two years
before. The notion that you can be depressed without knowing it represents
an assumption that simply has to be in place in order to capture as many
candidates for treatment as possible, the upshot of it all being that many
classified as depressed “do not believe that they have signs of mental disor-
der, do not raise such questions with their physician, do not seek help.”31
To antidepression crusaders, that depression can be imperceptible to its
victim is one more proof of its insidiousness. In reality, far from being a life-
saving measure, screening for depression seems to be of little benefit. “In
BEWARE WHAT YOU LOOK FOR 151
The sketch reads like a pastiche of generic ideas, a printout, virtually a par-
ody. The study itself, which portrayed psychiatry as a cuckoo’s nest, was
demolished in detail by Robert Spitzer.
One detail rang true to Spitzer, however: the high-handed interpreta-
tion of the 39-year-old’s history as a tale of “ambivalence.” Wrote Spitzer,
“Here, for the first time, I believe Rosenhan [author of the Science paper]
has hit the mark. What he described happens all the time and often makes
attendance at clinical case conferences extremely painful, especially for
those with a logical mind and a research orientation”—the last comment
a slap at the Freudians.34 Not that Spitzer exempted researchers from the
tendency to construct the world in the image of their theories and prefer-
ences. He once observed in an interview,
Researchers always give maximal preference for the disorders that they have
a particular interest in. In other words, if you’re really interested in panic
disorder, you’re going to tend to say it’s very common. You never hear an
expert say, “My disorder is very rare.” Never. They always tend to see it as
more common.35
152 THE NOCEBO EFFECT
This comes close to saying that the investigator with an interest in a dis-
order will tend to discover evidence of that disorder—enough to make it
seem “very common.”
Because we find what we look for, gastroenterologists peering through
the lens of their specialty may diagnose fibromyalgia as irritable bowel syn-
drome, while rheumatologists may diagnose an IBS patient with fibromy-
algia. Sometimes a preferred disorder seems common enough to sustain a
movement, as when a 1993 editorial in favor of PSA testing declared over
90 percent of screening-detected prostate cancers to be “serious as judged
by their volume and histologic grade.”36 By contrast, two investigators
estimated that in the highly screened population of the landmark Pros-
tate Cancer Prevention Trial, which ran from 1993 to 2003, seven detected
cancers out of eight were without significance.37 According to the FDA
analysis, 80 percent of the tumors discovered in the companion REDUCE
trial met the pathology criteria for “very low-risk disease, which indicates
that a reduction in their incidence is unlikely to be clinically significant.”38
I’ve referred to a 1968 study of the incidence of common symptoms in
healthy volunteers. As high as the frequency of reported symptoms ran, it
would have run even higher if the search had been conducted differently—if,
for example, a third of one group hadn’t been excluded from the study simply
because they took aspirin. The frequency of symptoms thus depends “on the
intensity with which the examiner searches for the symptom.”39 In the case of
both prostate cancer and depression, prevalence depends on search. Thus “the
likelihood of being diagnosed with [prostate cancer] is directly related to the
rigor with which one looks for it.”40 Employing as it did an artificially intensive
biopsy regimen, the PCPT found the disease in 24.3 percent of the placebo
group, men admitted into the study in the first place only if they were at low
risk of prostate cancer. What if men were sufficiently committed to early detec-
tion to skip PSA testing, which is after all nothing but a gateway to biopsy, and
proceed directly to biopsy itself, say annually or biannually? Even more cancer
would be found than at present; indeed, more would be found if the detection
system remained as is but more biopsy samples were taken.
Although the official checklist of symptoms of depression is so cumber-
some that doctors have trouble remembering it, it has been retained since
1980, through all the turmoil and acrimony of revision of the DSM, no
doubt because it enables the catching of cases. Such generic symptoms as
fatigue, sleep problems, and trouble with concentration alone will take you
60 percent of the way to a diagnosis of Major Depression. Add that insomnia
virtually implies fatigue and it’s easy to see why searches for depression so
often prove successful. Just as 89 percent of a large population was found to
have suffered exposure to a traumatic event according to DSM-IV criteria
for PTSD,41 it’s possible to generate high rates of Major Depression with the
BEWARE WHAT YOU LOOK FOR 153
right tests. If you want to produce the finding that a third of Americans are
depressed in a given month, simply ask people if they feel “really depressed”
once a month or more, as an ABC News poll did in 2002;42 but don’t ask
them if they feel that way all the time for a month. As it is, we’re told that one
in five Americans will be visited with Major Depression at some point, just as
one in six American men now stands to be diagnosed with prostate cancer.
The combination of a potentially serious condition, undiscriminating tools,
loose language, diagnostic fervor, and emotive publicity produces the sort of
inflation that shows up in the numbers who take antidepressants and who
have been treated for prostate cancer without clinical significance.
An example of a blunt diagnostic instrument is the questionnaire pri-
mary care physicians in New York City have been asked to administer to
patients who might be depressed. How often have they been bothered
over the past two weeks by various symptoms, including such common-
place problems as feeling tired, overeating, and a lack of “pleasure in doing
things”? According to this checklist, I would qualify for a diagnosis of
Minor Depressive Disorder if I simply had too little energy and felt too
little pleasure. Other symptoms on the list, such as “feeling down,” may
well be fitting responses to reality rather than signs of mental disorder,
though that possibility is ignored. The questionnaire casts a wide net, as if
its designers were troubled by the possibility of catching too few cases but
not too many. As simplistic as it is, however, the questionnaire is modeled
squarely on the checklist of symptoms of depression in the DSM.
The last absurdity of using checklists to fish for cases is represented by a ten-
statement test funded by Eli Lilly and posted on WebMD for a time in 2010.
The supposedly diagnostic statements read, “I feel tired almost every day”; “I
feel sad or down most of the time”; “I find myself thinking a lot about dying”;
and so on. Even if you answered No to each and every one of these proposi-
tions, you got a message advising that you “may be at risk for major depres-
sion,” for the simple reason that not everyone with depression experiences the
same symptoms. “Every individual is unique.”43 According to this idiosyncratic
understanding of risk requiring only that you be a specific person, every single
member of the population “may be at risk” of Major Depression.
Like the DSM criteria for depression, the Hamilton Rating Scale used
to grade depression on a scale from 0 to 51 is a less fine instrument than
it appears. If it really were a fine instrument, a difference of two points
might mean something. In clinical trials the benefit of antidepressants over
placebo comes to just that—two Hamilton points—the same quantum of
improvement that would be tallied if the patient fidgeted less while being
interviewed.54 This is a comment on the properties of the antidepressants
now being taken by 10 percent of the American population, but also on the
illusory precision of a scale calibrated in increments of one. The same vial
of blood divided in two may give different PSA readings, and a two-point
difference on the Hamilton scale may possess no more real significance
than that.
PSA isn’t specific to prostate cancer, and sadness isn’t specific to depres-
sion. A critical flaw of the Hamilton Rating Scale is that it ignores the
sources from which sadness arises, leaving open the possibility that some-
one gloomy for a good reason may qualify as a case of depression and a
candidate for treatment, along with the side effects of treatment. As crit-
ics have charged, the Hamilton scale and the practices that have grown
up around it pathologize normal sadness. An example of such a normal
response to events themselves is the depression that sometimes descends,
if only for a while, on those diagnosed with prostate cancer—a “risk of
diagnosis” cited in the literature. Until recently, the clinicians of sadness
did allow an exception for grief for a lost loved one, though even here the
arbitrary enters in, for if mourning extended beyond two months it was
counted as pathological. (Coincidentally, this seems about the same length
of time Hamlet has been in mourning—“But two months dead, nay, not so
much, not two!”—when he is reprimanded in front of the court for griev-
ing overlong.) How the figure of two months was arrived at by the APA I
don’t know, but it reads like a compromise between not allowing a grief
exception and allowing a more liberal period of mourning, which might
limit the pool of candidates. So bold is psychiatry’s claim to cognizance
over human life that in DSM-V grief is declared a dangerous “psychoso-
cial stressor”55 warranting careful medical surveillance and, if necessary,
intervention.
Early Detection
When the mixed evidence of clinical trials of PSA finally came in long after
mass screening for prostate cancer had become self-perpetuating, the mor-
tality benefit of PSA testing remained unclear, although the human costs
of the overtreatment of prostate cancer had long since become undeni-
able. Indeed, the European Randomized Study of Screening for Prostate
Cancer (ERSPC), which found a reduction of prostate-cancer deaths as a
result of screening (though no reduction in overall mortality), also found
that for every death averted, 48 men would have to be treated over nine
years.56 Using the language of risk/benefit calculation, urological medicine
had come to rely on a mode of screening whose benefits were unclear but
whose harms certainly weren’t. Both PSA testing, which appeals to evi-
dence presumed to exist, and testing for depression, which appeals to diag-
nostic criteria conceded by the designer of the DSM-III to be seriously
defective (just as Horwitz and Wakefield allege),57 strain the very process
of weighing evidence.
At the same time as the PCPT tested the possibility of preventing a
common cancer with a drug already in use for other purposes (finaste-
ride),58 it was proposed that people who have recovered from an episode
of depression take an antidepressant to prevent another onset.59 Also in
parallel with cancer rhetoric, advocates argued that screening and only
screening could catch depression in its early stages and prevent it from
progressing. “Unless brought into treatment,” it was said, “people who
meet symptomatic criteria [for depression] may develop chronic, recur-
rent, and deteriorating conditions.” So it is that a concerted effort was
made to identify members of the population who were depressed but
didn’t seek treatment, possibly because they failed to understand that
they were depressed.
Fearing that there was an unmet need for mental health services among peo-
ple who might not even recognize that they were suffering from a disorder,
policy makers . . . placed a high priority on finding ways to identify people
with untreated disorders in the community and bring them into treatment.
And it was decided that if people would not come in for diagnosis, then
diagnosis must go to the people.60
Like those who hunt for given patterns in literary works and inevitably
find them (because the ambiguity of literature gives their preoccupation
something to feed on), those looking for depression are predisposed to find
it, and do find it.
In the case of both prostate cancer and depression, the logic of early
detection leads to a push for earlier detection and lowered thresholds. In
the latter case this means looking not only for Major but Minor Depres-
sion, in which only a few of the listed criteria must be satisfied—an
158 THE NOCEBO EFFECT
It appears many doctors feel the same way. “Studies have shown that up to
one third of men screened for prostate cancer were unaware that they were
being tested,”63 contrary to the promotion of what’s called awareness by
advocates of PSA testing.
Awareness
the test.) Organizations that originally promoted mass screening for pros-
tate cancer, such as the American Cancer Society and the AUA, have since
taken note of the harms that followed from such an indiscriminate practice
and have belatedly scaled back their recommendations.
Just as advocates of mass PSA testing point to the legions diagnosed with
and treated for prostate cancer as evidence that it saves lives, psychiatrists
defend the categories of the DSM “by pointing to the vast numbers of North
Americans they have identified as suffering from the afflictions catalogued.”66
In the case of depression, inflated estimates alone have served rhetorically as
proof of urgency. Going back decades, such figures eventually realized them-
selves in the inordinate use of antidepressants. But the translation of inflated
numbers into cases began with the estimates themselves, because as soon as
community-based screening for depression is introduced, “the trend is for
the costly second-stage diagnostic screens to be eliminated or reduced and
for the minimal prescreen itself to become the basis for diagnosis and treat-
ment procedures.”67 (Somewhat analogously, blood-pressure readings given
in supermarkets and drug stores are often not followed up even though a
solitary reading can be misleading.)68 While prostate-cancer medicine can’t
correspondingly omit the second stage (biopsy)—because PSA alone doesn’t
really indicate anything—a great many cases of treated prostate cancer
constitute false positives in their own right, in that they would never have
progressed if left untreated. At this point there is some evidence that mass
screening for prostate cancer has reduced disease-specific mortality (though
improved treatments have also contributed), little evidence that it has led
to reduced mortality overall, and questionable evidence that screening for
depression improves outcomes.
Precisely because depression and prostate cancer shade off into nor-
mal conditions, it’s possible to diagnose more and more cases; hence, for
example, the hubristic estimate of the World Health Organization that
depression is soon to become the second leading cause of disability across
the globe, if it isn’t already at the top of the list.69 However, practices that
aggressively inflate the prevalence of depression not only don’t do good,
but may harm. As Horwitz and Wakefield reasonably contend, even to be
preliminarily diagnosed as depressed only to be cleared by further testing
“creates a space of pathological possibilities that did not previously exist.”70
Newly aware, I wonder if I might be carrying the cancer of depression
without realizing it, which makes it that much more insidious. Ill-founded
diagnoses of depression can also come back to haunt in applications for
health insurance, custody hearings, and other settings.
Among the noxious possibilities introduced by the overdiagnosis of
depression is a lingering worry over the label after depression has resolved,
as it may do within weeks after detection. Overdiagnosed prostate cancer
BEWARE WHAT YOU LOOK FOR 161
Emotive Advertising
But while screening for prostate cancer and screening for depression
mirror one another in some detail, this isn’t to say that urology and psy-
chiatry are engaged in mutual imitation like next-door neighbors. Psychia-
trists and urologists work at different ends of the human animal. Urological
activism is in fact patterned on its breast-cancer equivalent, which supplied
the model and precedent for mass screening, just as mammography pre-
dated PSA testing.73
To enforce the urgency of timely treatment, a poster during World War II
reminded Americans that more of them died every two weeks of cancer diag-
nosed belatedly than died at Pearl Harbor.74 Playing to the intuitive notion
that cancer cells are destined to progress inexorably, messages like this made
the value of early detection seem incontestable and self-evident. Mammog-
raphy couldn’t have established itself in American life without decades of
such publicity behind it, publicity appealing powerfully to hope and fear.
In the event, though, mammography has proven more effective at detect-
ing cancer of doubtful significance than at preventing cases of advanced dis-
ease, whose incidence remained virtually constant over the mammography
decades. According to epidemiological data, the rising incidence of breast
cancer since the 1970s is “driven almost exclusively by changes in the pro-
portion of early-stage disease, with almost no observed differences in the
incidence of late-stage disease over a period of 30 years . . . The inference
is that there are now more new cases of early-stage disease being identified
[through mammography] than cases of late-stage disease being prevented:
this represents overdiagnosis.”75 A meticulous evaluation of mammography
found its benefits “modest” and its harms “significant,”76 not the ratio one
would expect of a medical practice that has become an institution.
In the climate of overcharged publicity surrounding mass screening,
detection must intercept the advance of disease, and the dramatic rise of
incidence as a result of screening reads like a powerful argument in favor
of screening. Screening propaganda maintains silence on the definitional
issues, biological ambiguities, borderline cases, and conventions of diagno-
sis that make breast and prostate cancer, or indeed depression, as common
as they now are. It’s also silent on the harms of screening itself, harms that
may account for the failure of either breast- or prostate-cancer screening
to reduce all-cause mortality in most studies, regardless of their effect on
disease-specific mortality.77 Even if cancer screening saves lives, it may
also be costing lives—in which event all-cause mortality could remain
unchanged. In the Malmö (Sweden) mammography study, “evidence for
an excess cardiovascular mortality in the screened group was found . . .
which more or less offset the mortality gain from screening. The possibility
that the stress of screening and of positive diagnoses contributed to this
cannot be ruled out.”78 Less conjectural is the case of a man who died as an
BEWARE WHAT YOU LOOK FOR 163
While men getting screened for prostate cancer are strongly motivated and
patients now ask for antidepressants without the formalities of diagno-
sis, the fact remains that neither of the pseudo-epidemics discussed here
would be possible without medical activism. The ambiguity of the target
condition, the unreliability of diagnostic instruments, and emotive public-
ity represent the ways and means of such activism: ambiguity supplying a
field of opportunity, imprecise instruments lending themselves to indis-
criminate detection, and promotional campaigns generating uptake. To the
spirit of activism the commonness of prostate cancer on the one hand and
sadness on the other represents not a caution against mass screening but
a call to arms. Without activism, screeners wouldn’t have gone out into
the community searching for depression, and without the inflated esti-
mates of depression that followed, the market for antidepressants would
lack medical legitimacy. Without activism, PSA testing wouldn’t have been
offered to millions of men with the promise of a mortality benefit that is
still—25 years later—unclear, nor would the American Urological Associa-
tion have found itself in the paradoxical position of endorsing PSA testing
even though, by its own estimate, it may actually yield a net harm. The
same spirit of activism screens in churches for high cholesterol and pros-
tate cancer itself, and now advocates general screening for autism despite
the amorphousness of the condition, the variety of diagnostic instruments,
the openness of diagnostic criteria such as a lack of “social imagination,”
and the absence of proven treatments.
While the likelihood of finding prostate cancer depends on how hard
you look for it, medical activism implies more than an energetic search
for evidence: It implies an expectation of finding it and a disposition to
interpret equivocal data accordingly. Medical activism is the leavening that
allows the evidence of a pseudo-epidemic to rise and expand. Much as crit-
ics look for their favorite patterns in works of literature in the assurance
of finding them, medical activism anticipates its own conclusions. In the
immediate aftermath of the September 11 attacks, a number of therapists
“planned to approach citizens on the street [in New York City] and conduct
quick sidewalk assessments and brief interventions for those presumably
164 THE NOCEBO EFFECT
too avoidant to seek help on their own.”80 (The innocence of the interven-
tions is taken for granted.) Given that these counselors assumed a desper-
ate need for psychological first aid, such all-out screening for indications
of PTSD would probably have found what it was searching for—provided
passersby would submit to screening. Like these activists, the members
of the depression-screening movement who took diagnosis to the people
did so not without preconceptions, but, on the contrary, with a definite
expectation of finding high rates of depression. As Horwitz and Wakefield
put it, they “feared that there was an enormous unmet need for mental
health services among people who might not even recognize that they were
suffering from a disorder.” Their findings confirmed their fears. Perhaps a
corresponding fear of leaving anyone out has fueled successive editions of
the DSM.
The chair of the DSM-IV Task Force has detailed, one by one, the diag-
nostic disasters that followed from its provisions, often blaming the drug
companies for exploiting possibilities the authors of the document hardly
knew they were creating. If we ask how intelligence and good will could
possibly have fashioned a document that gave rise to so much harm, the
answer can only be that the writers were misled by their own zeal—by the
felt imperative of specifying each and every malady of the mind so that all
of it could be brought under enlightened management.
Signs of Hope
protocol, subjects were biopsied if their PSA rose above 4.0, but addition-
ally, at the completion of the trial the men were asked to undergo an end-
of-study biopsy (that is, a research biopsy) even if no previous biopsy had
been performed, their PSA hadn’t risen, and other signs were normal. The
lion’s share of the thousands of biopsies performed in the PCPT were in
fact performed thus, not for cause. Predictably (given everything known
about the prevalence of prostate cancer), the research biopsies detected
hundreds of cancers. What to make of these malignancies that would have
remained latent but for the investigative demands of a clinical trial?
Examining the PCPT data, FDA reviewers threw out the results of the
research biopsies as essentially meaningless. “FDA regards the informa-
tion from ‘end of study biopsies’ to be related to study conduct and not
reflective of clinical practice.” Furthermore, “the primary efficacy results
for the PCPT are driven by the ‘End of Study’ biopsies and therefore raise
questions regarding the applicability of the results to the current standard
of care for management of healthy middle-aged and elderly men with no
clinical evidence of prostate cancer.”82 But years before the FDA reviewers
came to this cogent conclusion, some thirty-nine hundred men, almost a
quarter of those enrolled in the PCPT, decided there was no good reason to
subject themselves to a research biopsy and refused to have one. While the
PCPT hit a predicted 25 percent reduction in cancer incidence almost dead
center (at least until the FDA discarded some of the trial’s data), the study
designers also predicted that only 5 percent of participants would refuse to
be biopsied without cause. As they note, “The rate of refusal of biopsy was
higher than had originally been estimated.”83
Following the lead of the original paper on the PCPT, the medical lit-
erature simply reads the refusers out of the study, but like many others
who figure only parenthetically or as ignored subtexts in medical papers
(or like the 50 percent of male internists who don’t have PSA testing),84
they serve as examples of people who had the spirit to decline the ques-
tionable overtures of activist medicine. As in the PCPT, the numbers of
refusers are far from negligible. Contrary to the zeal of therapists offering
psychological first aid, it appears that only about 10 percent of trauma sur-
vivors seek help; considering that a common form of aid, Critical Incident
Stress Debriefing, has yielded questionable results in controlled trials, such
widespread refusal is probably a good thing. In 2006 it was reported that
“30–60% of patients in primary care decline to participate in clinic screen-
ing interviews [for depression] offered by researchers or clinic nurses dur-
ing routine attendance.”85 In a study that looked for evidence that a fear of
being stigmatized would keep people from seeking treatment for depres-
sion, fully two-thirds of those found eligible chose not to participate.86 As
for those who have simply “thrown [their antidepressants] down the toilet
166 THE NOCEBO EFFECT
when they felt better and experienced no subsequent ill effects,”87 no one
knows how many they are. (Recall, though, that as many as 70 percent of
those prescribed antidepressants prove noncompliant.) In diverse ways, all
such refusers remind us that the trend toward medicalization is neither
inevitable nor incontestable.88
In a study looking into adherence to prescribed antidepressants, 49
percent of those eligible refused to participate, and, of those who did,
another 17 percent were lost in the course of the study. Among the 147
completers, only 19 percent took antidepressants as prescribed over a
period of six months. It was this study that introduced us to Ms. A, the
patient in shock over the sudden death of her mother who was prescribed
antidepressants against her wishes and never filled the prescription.
According to the authors’ calculations, patients with misgivings about
antidepressants like Ms. A’s have a 2 percent probability of adherence
over six months.89
Refusers send a message. In a large British trial of drug treatments for
mild hypertension, 44 percent of the men withdrew, presumably because
of a high rate of side effects including impotence. Almost as many women
withdrew, and in doing so they may have benefited more than they knew—
among women in the treatment group, the death rate rose.90 At times cir-
cumstances themselves counsel refusal. Patients diagnosed with dementia
may lose not only their insurance and driver’s license but their indepen-
dence, even their sense of self; little wonder that as many as half of those
screening positive for cognitive impairment—a common condition that by
no means sentences a patient to Alzheimer’s disease—may refuse any fur-
ther evaluation.91
Maybe we can count as implicit refusers those who don’t seek medical
care for ordinary problems at all. In 1991 two investigators of the way lay
people think of illness concluded, “It is apparent from the high frequency
of self-recognized illness episodes and the correspondingly low frequency
of medical consultation that contemporary sufferers tend to perceive their
afflictions as private and individualistic occurrences, which are readily
managed outside the realms of the formal medical-care system.”92 In this
context the tag-line of drug ads, “Ask your doctor,” reads like an effort to
nudge people into the formal medical-care system who might otherwise
just look after themselves. As the cited discrepancy suggests, such people
may be many. Wrote a British psychiatrist in 2008:
In 1996, just before the Royal Colleges of Psychiatrists and General Practi-
tioners began their “Defeat Depression” campaign, they surveyed lay peo-
ple’s attitudes to depression and its treatment. What they found was broadly
subversive to the basic assumptions driving the campaign (though they
BEWARE WHAT YOU LOOK FOR 167
didn’t let this put them off): most people did not subscribe to a depression-
as-disease model, and saw things in terms of situational problems which
were not something to take to the doctor.93
Patients may believe or hope that something physical is wrong, the cause
will be found, and they will be healed. Because of stigma issues, there may
be a potent investment on the part of the patient in keeping the focus on the
somatic complaints. The physician, too, may feel more comfortable in the
physical realm, and the assessment will emphasize these complaints.97
On this showing, the doctor who fails to diagnose depression when the
patient doesn’t complain of it ignores the obvious because she is trapped
in her comfort zone. However, it may be that the doctor refrains from
making tendentious inferences about the patient or hesitates to impose a
diagnosis that can’t be retracted for a condition that often wanes, or both.
168 THE NOCEBO EFFECT
our needs . . . But perhaps we may also familiarize ourselves with the idea
that there are difficulties attaching to the nature of civilization which will
not yield to any attempt at reform.”4 That the world is in some way beyond
reform isn’t a message utopianism wants to hear.
What happens when the dust of war settles and a new paradigm rules, as
the biomedical concept of mental disorder ruled once the psychodynamic
model was officially deposed? As Kuhn explains, there ensues a return to
normality. The revolution is consolidated and routine work begins on sets
of problems brought to the fore by the new conceptual regime.
Few people who are not actually practitioners of a mature science realize
how much mop-up work . . . a paradigm leaves to be done or quite how
fascinating such work can prove in the execution. And these points need
to be understood. Mopping-up operations are what engage most scientists
throughout their careers. They constitute what I am here calling normal sci-
ence. Closely examined, whether historically or in the contemporary labora-
tory, that enterprise seems an attempt to force nature into the preformed and
relatively inflexible box that the paradigm supplies.5
Consider . . . the case of a boat captain whose fishing vessel is lost at sea,
resulting in the death of several crew. Though not physically injured, the cap-
tain starts feeling “on edge,” suffers from insomnia and begins to withdraw
from usual activities. Most alien to the fisherman’s self-concept, he becomes
anxious when considering a return to his usual occupation. Consequently,
he turns down offers to work on other vessels, and he becomes isolated
from the fishing industry. Without income, this man becomes increasingly
anxious and depressed. Prior to the introduction of PTSD in 1980, a psy-
chiatrist would have conceptualised this fisherman’s problems, first, as nor-
mal bereavement over lost friends who died in the incident, and second,
as a phobic disorder caused by the traumatic event. A third concern would
have addressed the development of situational depression as a consequence
of adjustment issues and the fisherman’s inability to return to sea. Now, in
our post–DSM-III era, we can ask whether the introduction of PTSD has
furthered our understanding of this patient’s reactions to a life-threatening
event.8
But we can also ask exactly what’s gained by checking boxes that say “pho-
bic disorder,” “adjustment issues,” or whatnot, other than the illusion that
official labels bring the captain’s catastrophe under professional cognizance
and management. In short, what’s the captain doing in a psychiatrist’s office
at all?
Artificial Lines
In fine, the DSM allows professionals to speak the same language and
check the same boxes. A member of the DSM-III Task Force agrees but
goes perhaps even further:
Disorder four, and Conduct Disorder three? A member of the DSM-V Task
Force was quoted in the press as saying, “We’re basically drawing artificial
lines, and the body and mind do not work like that.”17
Arbitrary distinctions haven’t only recently crept into the DSM, how-
ever: They’ve been there for decades. Almost a quarter century ago the
Lancet noted the capriciousness of temporal requirements in the DSM sys-
tem, pointing out that
Mr. Jones is a fifty-one-year-old divorced white male with the chief com-
plaint “I don’t want to live anymore.” He presented in the ER last night with
intense suicidal ideation. He described feelings of hopelessness and guilt
and reported a weight loss of ten pounds in the last three weeks. He reports
extremely poor sleep with early morning awakening.23
Criteria designed to generate diagnoses yield the same diagnosis when the
patient is frazzled as when he is intensely suicidal. The upshot of this kind
of inflation is that a tenth of the American population now take antide-
pressants for a condition the builder of DSM-IV concedes is “not really
‘major,’ is not really ‘depressive,’ and is not really a ‘disorder.’”24 Evidently
the thought of exempting millions from psychiatric diagnosis seemed to
successive authors of the DSM like a flaw in their system.
The use of symptom checklists for diagnostic purposes was the hall-
mark of DSM-III, which, as noted, repudiates the Freudian or quasi-
Freudian model of neurosis as insufficiently scientific and installs in its
place a system capable of bringing different clinical observers to reliably
similar conclusions. Yet DSM-III rejects more than DSM-II. As Horwitz
and Wakefield establish in The Loss of Sadness, its simplistic criteria for
depression zero out the entire Western tradition defining melancholy as
uncaused, inordinate, or habitual sadness. In a profound dissent from the
view that the past has nothing to teach the present, Horwitz and Wakefield
consider the traditional understanding of melancholy a repository of wis-
dom and expose the costs of discarding the concept of normal sadness—
written into the tradition from its beginning in antiquity—in the name of
theoretically scientific diagnosis.
Under the DSM system, if someone shows enough presumptive signs
of depression, then that person is depressed and becomes a candidate for
medical management, no matter whether the sadness (if present at all)
arose from the breakdown of a marriage, the loss of a job, the failure of
an endeavor, transient distress, or no identifiable cause. Without bundling
the sadness incident to human life—in Hamlet’s words, “the heartache . . .
that flesh is heir to”—with the pathology of depression, it would be difficult
to discover depression at the epidemic level it has attained. Significantly,
the designer of the DSM-III diagnostic system grants the validity of the
Horwitz–Wakefield criticism of its context-free criteria, particularly in
THE FOLLY OF SYSTEMS 175
the case of depression.25 DSM-V adds language to the Criteria for Major
Depressive Disorder that reluctantly accommodates their argument that
what is called depression may be a warranted response to life itself—
reluctantly, because while the qualifying symptoms “may be . . . considered
appropriate” to a crisis, “the presence of a major depressive episode in addi-
tion to the normal response to a significant loss should also be carefully
considered.” Normality is under suspicion.
fusion of transient and habitual sadness as in DSM, but the very principle
of codifying or systematically specifying mental disorders. How fitting that
the great figure of melancholy in Shakespeare is a character of the most
bewildering indeterminacy and many-sidedness: Hamlet, whose complex-
ity is such that we can’t really say where his simulation of madness ends and
something like actual madness begins.
Though the Anatomy of Melancholy looks into the causes, divisions,
manifestations, and cures of melancholy—the most indeterminate of
human afflictions—it’s less a systematic treatise than a flowing or overflow-
ing meditation on an illimitable subject. Burton in fact denies the possibil-
ity of pinning down an ill as variable as melancholy:
Fear and sorrow, the principal ingredients, are symptoms (the ‘inseparable
companions’ of most melancholy) and causes (in the case of sudden frights
or bereavements); jealousy, too, is both a symptom and a cause . . . in fact,
all the passions of the mind can be seen in this way. The outline is simply
inadequate to the material . . . Burton states simply that his categories are a
convenience, not part of the nature of things, and even at times ludicrously
misleading.30
between normal and abnormal woe in the DSM than by the shrinkage of
melancholy to a few items—the reduction of an “infinite” phenomenon to a
handful of markers, an “irregular” one to the logic of a checklist. The DSM
not only wrestles with Proteus but straightjackets him, though this isn’t
to say it has banished the sort of instability that Burton discovers in mel-
ancholy. For the DSM itself has changed shape. Indeed, it has become an
illustration of mutability. Not only has it jettisoned the Freudian assump-
tions of editions one and two (1952 and 1968, respectively), it has swelled
to some nine hundred pages—including apparatus—as if the immensity it
sought to master couldn’t be repressed after all. DSM-I contained 106 diag-
noses, DSM-II 182, the revolutionary third edition (1980) 265, DSM-IV
297, the 2000 revision of DSM-IV no less than 365,35 and the newest DSM
somewhere over 500, with many disorders subdivided into mild, moder-
ate, and severe variants, and the table of contents alone running 27 pages.
It’s as if the DSM made up for the absence of tests of mental disorder with
an abundance of text. With such a rain forest of available pathologies, who
doesn’t qualify for one or another authorized disorder? How many par-
ents are exempt from Parent-Child Relational Problem (V61.20)? If you
have a dispute with your landlord, DSM-V has a code for you: V60.89.
So encompassing is the DSM system that it has a diagnosis for the mag-
nification of ordinary ills into medical problems promoted by some of its
own categories. The diagnosis is Somatic Symptom Disorder—if not Illness
Anxiety Disorder, which for some reason isn’t classified among the Anxiety
Disorders.
A showcase of clinical fashion, of new coats, the DSM has introduced
such celebrated diagnoses as chronic fatigue syndrome and social phobia.
From DSM-II to DSM-III, Hyperkinetic Reaction of Childhood became
Attention Deficit Disorder, which was subdivided into three branches—
inattentiveness, impulsivity, and overactivity—which in the next revision
of the DSM reunited in a single disorder (ADHD) for which a child needed
to show eight of fourteen possible symptoms. In DSM-V inattention and
impulsivity (the latter now assimilated with hyperactivity) have their own
symptom lists, each requiring six matches. As to social anxiety, it may never
have occurred to the sufferer that dislike of public bathrooms and dislike
of being criticized are components of a single pathology. While many of
the changes introduced into the DSM by fashion and scholastic ingenuity
are impossible to take seriously, the shifting shape of the manual over the
decades does suggest that Burton, in emphasizing the uncertainty of men-
tal disorders, was on to something.
Indeed, the chair of the DSM-IV Task Force uses language reminiscent
of Burton to bring out the indeterminacy of mental disorders. “‘Mental
disorder’ and ‘normality’ are both extremely protean concepts—each so
THE FOLLY OF SYSTEMS 179
The result is an even more elaborate system that incorporates mild versions
of many putative disorders, thus claiming still more of the territory of nor-
mality for medical purposes.
Don’t let anyone expect that, like the common lot of speech-makers, I’m
going to begin with a definition and then go on to divide up my topic—
that least of all. Either procedure would be inappropriate; circumscribing a
THE FOLLY OF SYSTEMS 181
There’s no checklist of symptoms for being a whale. Even the colloquial def-
inition of a whale as “a spouting fish with a horizontal tail,” which has the
virtues of clarity and common usage, leaves out the matter of magnitude
(and misclassifies a mammal). Who shall prescribe rules to Leviathan?
Melville’s brilliant catalogue of whales as so many species of books, with
sperm whales as folios, mocks the principle of an authoritative summa-
tion or ordering—a bible—of knowledge. It’s offered in a spirit of the serio-
comic and stands as a monument to incompletion, with Melville writing
at its end, “I now leave my cetological System standing thus unfinished,
even as the great Cathedral of Cologne was left, with the crane still stand-
ing upon the top of the uncompleted tower.”50 We won’t find system in a
182 THE NOCEBO EFFECT
Mockery of Medicine
Amorphous Distress
Cultures provide publicly available and shared meanings that facilitate cer-
tain kinds of symptom interpretations while discouraging others . . . Naming
THE FOLLY OF SYSTEMS 187
the West. But we don’t have to look back a century to witness an overnight
shift of diagnostic practice. The Feighner criteria of 1972 made history by
codifying the diagnostic markers of some dozen disorders, the first few
on which the data were said to be in hand. In their euphoria the authors
concluded their paper by proclaiming it a “synthesis based on data rather
than opinion or tradition.”84 Yet among the disorders covered in the paper
was homosexuality, a diagnosis requiring only “persistent homosexual
experiences beyond age 18,” with two technical exclusions. No more than a
year separated the triumphalism of the Feighner criteria from the deletion
of homosexuality from the ranks of mental disorders by vote of the APA.
Looking back on this sequence of events, it’s impossible to credit the claim
that the inclusion of homosexuality on the Feighner list in the first place had
nothing to do with opinion or tradition.
So too, what might have been diagnosed as anxiety in the 1960s, when
antianxiety pills were all the rage, could well be classified as depression in the
era of depression.85 Indeed, a patient who carried over from the 1980s to the
1990s could have been diagnosed first with anxiety, then with depression,
for the same set of problems.86 Depression became a diagnosis of choice just
when the formerly booming market for antianxiety drugs ran into trouble.
Once this presumption was in place, it was just one more step to claim that
depression hid behind common somatic symptoms. So arose the popu-
larity of antidepressants, which from this point of view look like second-
generation antianxiety drugs. Side-by-side references to depression and
anxiety are in fact everywhere in the medical literature, a pairing sug-
gesting that the two disorders represent variant configurations of distress.
For perhaps half of the twentieth century mixed anxiety-depression was
the most common psychiatric illness in the United States, and the hybrid
remains common to this day.88 That both anxiety and depression have a
self-referential component, so that you can be anxious about being anxious
and depressed about being depressed, suggests that the two disorders have
more in common than their official specificity denotes.
The drugs used to treat social anxiety are in fact antidepressants and
advertised as such—Paxil being trumpeted to psychiatrists as “the anxiolytic
THE FOLLY OF SYSTEMS 189
A Hero in Distress
The satiric tradition ran strong in Dickens, his longest and most ambitious
novel, Little Dorrit (1857), exposing a society of pretenders through which
bankruptcy races with the speed and virulence of a “moral infection.”92
Dickens’s original conception called for a novel titled Nobody’s Fault, its
protagonist a man “who should bring about all the mischief in it, lay it all
on Providence, and say at every fresh calamity, ‘Well, it’s a mercy, however,
nobody was to blame you know!’” In the medical literature growing out of
190 THE NOCEBO EFFECT
sat down before the dying fire, as he had stood at the window of his old room
looking out upon the blackened forest of chimneys, and turned his gaze back
upon the gloomy vista by which he had come to that stage in his existence.
So long, so bare, so blank.94
The same Arthur Clennam is also beset with anxiety. “I am in a very anx-
ious and uncertain state,” he says at one point; “a state that even leads me
to doubt whether anything now seeming to belong to me, may really be
mine,”95 because he may have inherited tainted money. A stranger to the
notion that depression and anxiety are specific conditions—each with its
theoretical cause, symptom set, and treatment—and not bound by the
artificial distinctions and illusory certainties of a clinical lexicon, Dickens
reveals in Clennam a nonspecific sense of lack appropriate to someone who
thinks of himself as Nobody. (Similarly, the guilt Clennam carries around
may or may not conform to the DSM standard of “excessive or inappropri-
ate guilt.”)96 All the more ironic, then, that Dickens’s fiction is now being
used in medical education, mainly, it seems, as an illustrated catalogue of
psychiatric conditions the author had the prescience to identify before they
were codified by medicine.97 Among the diagnoses said to be prefigured in
Dickens are multiple personality disorder, PTSD, and narcissistic person-
ality disorder, all coinages of the late twentieth century. Presumably, when
the next iteration of the DSM appears, Dickens will have predicted that too.
Dickens didn’t have his eye on the clinical discoveries or fashions of
later generations but on his own time and place, and Little Dorrit was writ-
ten under the inspiration of satire’s intense topicality. Within a few years of
its appearance the argument was made that regular medical examinations
are a necessity and physiological abnormalities must be dealt with in their
THE FOLLY OF SYSTEMS 191
Instead of the sort of system that hypertrophies and grows more inventively
specific with each iteration, Nortin Hadler classifies general distresses into
four main groups, the first consisting of:
The theory seems to be that symptoms in this group open the way for those
in the others, such as gastrointestinal complaints. One of the merits of the
model is that it accounts for the sort of inflammation of common ills that
propels the process of medicalization. That vigilance as to unusual symp-
toms, which is itself a mode of distress, should heighten the perception of
other distresses points to the vicious dynamic that makes many syndromes
resistant to treatment. The same model allows us to speak of depression
without the baggage of its codified symptoms, thresholds, and qualifiers,
and without reference to a presumed cause. It also captures the kinship of
depression and anxiety.
192 THE NOCEBO EFFECT
Homer wasn’t an allegorist. (“Is it possible that Homer really wanted to say
all that people have made him say?” asks Montaigne.)105 The fact is, a psy-
chological disorder born in the year 1980 under particular circumstances
doesn’t represent a biomedical entity immune to historical fashions, like
smallpox.106 Members of the DSM-III Task Force themselves hesitated to
codify a diagnosis “tied specifically to a historical event,”107 that is, PTSD.
Nor did the historicity of PTSD stop there. In the next iteration of the
DSM, you qualified for PTSD if you were merely “confronted with” the
actual or threatened death of “others,” language so vague it included those
who watched the events of 9/11 on television. Once again, the line between
clinical disorder and normal distress was effaced.108 The PTSD diagnosis
has certainly allowed for inflated findings. It appears rear-end collisions
less violent than those of bumper-cars at amusement parks may lead to
“post-traumatic stress.”109 And a population tested positive for PTSD when
asked whether they were bothered by thoughts of the worst movie they had
seen in recent months—in fact, “51.7% of participants in the movie sample
scored higher than plane crash victims [in a PTSD study] two months after
the crash.”110
3. If many disorders represent ways of channeling distress into socially
recognized forms, then the DSM, by recognizing one new disorder after
another, has actually contributed to their occurrence—has fed the nocebo
effect. Once multiple personality disorder was installed in DSM-III as a
result of persistent lobbying, it came to be diagnosed much more often.111
Strong evidence suggests that MPD was an artifact of treatment by ther-
apists who believed in MPD. Ironically, the progenitor of the diagnostic
revolution in psychiatry thought the same of PTSD. Said Mandel Cohen in
an interview in 2000,
The Underground Man qualifies for many of these epithets, and knows it.
If diagnosis means subsuming a case under a clinical model, maybe he
qualifies for SAD as well.
As the authors of The Loss of Sadness argue, a diagnosis that looks at symp-
toms only, ignoring the story surrounding them, will interpret many mood
variations as depression that lie outside the jurisdiction of medicine. In the
case of the Underground Man, in order to arrive at a diagnosis of SAD you
have to ignore a good half of his story, for example his way of torturing oth-
ers because others are torture to him. Representing the Underground Man
as a sufferer from social anxiety not only makes him more of an innocent
than his story shows him to be, and not only tags him with a disorder that
was codified a century later and half a world away (and with controversy at
that), but attaches a clinical definition to one whose first concern seems to be
to resist all such attempts at summary or objectification. As M. M. Bakhtin
writes of a number of Dostoevsky’s heroes including the Underground Man,
in order to make fun of men of action . . . but really, gentlemen, who can take
pride in his illnesses, still more boast about them?3
“I’ll wager you think”: Here then is our hero’s attempt to keep one step
ahead of everyone who has him figured out, as if the curse of heightened
consciousness rendered others shallow by comparison. Such consciousness
isn’t an incidental trait of the Underground Man but one that makes itself
felt through the entirety of his narration as he gives play to a noxious habit
of self-reference. Unforgettably, he likens himself to “an intensely aware
mouse.”4
In its professed rationality and transparency and its international reach, the
DSM diagnostic system introduced in 1980 adapted to psychiatry the basic
principles that steel-and-glass modernism brought to architecture decades
before. As suggested by Robert Spitzer’s comment that psychiatry was in
need of a new beginning when DSM-III was born, the diagnostic revolution
had a utopian strain, and the same is arguably true of architectural modern-
ism.5 Notes from the Underground was written in reply to a utopian novel
that spoke in its own way for new principles of construction, a novel that
proved immensely influential, feeding into the most world-transforming
utopian ideology of the twentieth century: Marxism–Leninism.
As sharp as his mockery of himself is, the Underground Man’s worst
taunts are reserved for those who would like to make the world a happy
and rational place by rebuilding it in the image of a Crystal Palace, the
immense hall of glass constructed with stunning speed and efficiency for
the Great Exhibition of 1851 in London. The image of this wonder came
to Notes from the Underground by way of Chernyshevsky’s What Is to Be
Done? (published only a year before), a kind of novelistic allegory of rev-
olution that proved sacred to generations of Russian radicals. In one of
the work’s dream sequences, the heroine, Verochka, glimpses the beautiful
meadows and orchards of a new earth, in the midst of which stands a “pal-
ace” of “cast iron and crystal, crystal and cast iron,” which can be none
other than the Crystal Palace transported from Victorian London into the
engineered world of the future.6 When the Underground Man mocks as a
Crystal Palace fantasy the ambition of constructing a world so rational that
nothing will happen, he alludes all but explicitly to Chernyshevsky’s utopia.
According to Chernyshevksy’s ideas, the revolution inaugurating a new
world was to take place within two years of his work’s publication in 1863,
and the progressives at the center of What Is to Be Done?—young doctors
among them—seem inspired with a feeling of impending transformation.
198 THE NOCEBO EFFECT
You can become an equal of the people described here in full, if only you
wish to work a bit on your own development. Anyone who is beneath them is
very low indeed. Come up out of your godforsaken underworld, my friends,
come up. It’s not so difficult. Come out into the light of day, where life is
good; the path is easy and inviting . . . No sacrifices are needed, no depriva-
tions required—they’re all unnecessary. Desire to be happy—that’s all.8
who was it who first announced, who first proclaimed that man only does
vile things when he does not know where his real interests lie? and that if
he were enlightened, if his eyes were opened to his real, normal interests, he
would at once cease doing vile things and would immediately become good
and honourable.9
stands to reason that the men and women of the future—the inheritors
of enlightenment—are envisioned in What Is to Be Done? as “burst[ing]
with health.”10 The Underground Man not only seems to cross his own
real interests, not only proclaims himself in his degradation an intensely
aware mouse, but shows in everything he says and does that his heightened
consciousness is a source not of health and happiness but the opposite.
By cherishing his sickness and challenging the principle that we pursue
pleasure and shun pain (the axiom of the utilitarian thinking that powers
Chernyshevsky’s novel), the Underground Man also suggests that one can
adopt the sick role not because its benefits outweigh its disadvantages, but
for more paradoxical reasons entirely.
the depressed person . . . urged her terminally ill friend to go on, to not hold
back, to let her have it: What terms might be used to describe and assess such
a solipsistic, self-consumed, bottomless emotional vacuum and sponge as
she now appeared to herself to be?12
200 THE NOCEBO EFFECT
A few years ago, cars of a certain brand started lurching forward dangerously.
The manufacturer contended that the car would not lurch if the driver only
paid constant attention, and isn’t that what drivers should do anyway? Not at
all, and to demand as much is to demand the impossible. The manufacturer
might just as well have insisted that drivers float six inches above the seat.
The author continues, “Our minds wander even during the most intense
and engaging activities, and when they do, we operate by habit.”13 Of course
our language tends to deprecate habit, in turn, as a rote performance, a
sort of enacted unawareness. Yet habit is highly operative even in feats of
excellence, such as musical or athletic performances. The performer of a
complicated skill like dribbling a basketball through defenders hasn’t the
time or luxury to attend to each and every movement of the body, even
assuming such a thing were possible, which of course isn’t to say that ath-
letes don’t know what they’re doing. Athletes know what they’re doing pre-
cisely in that their skills are so well honed that they don’t have to concern
themselves with the details in the way someone less trained would. Said
William James, “We walk a beam the better the less we think of the position
of our feet upon it.”14 In the famous mowing scene in Anna Karenina Levin
rejoices in such mastery:
The longer Levin mowed, the oftener he felt the moments of unconscious-
ness in which it seemed that the scythe was mowing by itself . . . and as
though by magic, without thinking of it, the work turned out regular and
precise by itself. These were the most blissful moments.15
Health itself, when undisturbed by pain, gives pleasure, without any exter-
nal excitement at all. Even though it appeals less directly to the senses than
the gross gratifications of eating and drinking, many consider this to be the
greatest pleasure of all . . . [The Utopians] rejected long ago the opinion of
those who doubted whether a stable and tranquil state of health was really
a pleasure, on the grounds that pleasure made itself felt only when aroused
from without . . . The idea that health cannot be felt they consider completely
wrong. Every man who’s awake, they say, feels that he’s in good health—
unless he isn’t.17
It seems the good health of the Utopians just wouldn’t be good enough
if it weren’t also conscious of itself. Whether health automatically entails
awareness (as with the Utopians) or awareness automatically entails lib-
eration (as in Chernyshevsky), awareness seems to be the crux of human
happiness.
In the inflamed consciousness of the Underground Man, Dostoevsky
satirizes this principle.
I am a sick man . . . I’m a spiteful man . . . I think there is something wrong
with my liver. But I cannot make head or tail of my illness and I’m not
absolutely sure which part of me is sick. I’m not receiving treatment, nor
have I ever done, although I do respect medicine and doctors. Besides, I’m
extremely superstitious, if only in that I respect medicine . . . I know per-
fectly well that I cannot in any way “sully” the doctors by not consulting
them. I know better than anyone that in doing this I shall harm no one but
myself. Anyway, if I’m not receiving medical treatment it’s out of spite. If my
liver is hurting, then let it hurt all the more!19
intensifying a condition that might otherwise have been too diffuse to con-
stitute an actual disorder—much less to serve as a banner—it was decided
that the cardinal symptom was fatigue. Now with a theme, the disorder
was branded. It was a fatigue disorder. As if the focal symptom served to
rally other ills around it, patients from far and wide began pouring “like
pilgrims” into Reno, the point of origin of the mystery syndrome. The press
picked up the idea that the agent behind this epidemic in the making was
Epstein-Barr virus, an entity whose impressively particular name seemed
to guarantee the specificity of the disorder. Rolling Stone proclaimed
Epstein-Barr a “growing nightmare.”23 With information itself acting as a
virus, cases seemed to increase in tandem with press reports, and support
groups for chronic fatigue patients sprang up around the world. In the UK
the syndrome acquired the formidable name, myalgic encephalomyelitis,
ME for short. Chronic fatigue took on the character of a movement, with
all the familiar trappings of advocacy. Where the Underground Man chal-
lenges the rule of reason with his tonal and other paradoxes, a disorder
whose thematic symptom was debilitating lethargy challenged medicine
with the paradox of vigorous activism.
Candidates for the cause of the disorder multiplied as each proved
groundless, and medical breakthroughs were proclaimed by the press one
after another, all likewise coming to nothing. Uncritical reporting about
the disorder, which overwhelmingly presumed a biological cause, merely
“increased public awareness of symptoms and fueled the epidemic.”24 That
heightened awareness served only to power the nocebo effect poses a com-
ment on the cult of awareness then and now, and on the use of “awareness”
as a crusading banner, incantation, and code-word all at once. (It appears
that as long as people in Hong Kong were unaware of the Western form of
anorexia—anorexia as defined by the DSM—they didn’t manifest that dis-
order, but once it entered the newspapers and public consciousness in the
1990s it exploded.)25 The label that gives a syndrome like chronic fatigue
salience and attracts the notice of the public in the first place may also
make things worse, both by recruiting new sufferers and by “promising a
scientific answer some time in the future.”26
Also said to be caused by some subtle agent that awaited discovery was
the fatigue disorder known as neurasthenia which flourished among the
well-to-do until its disappearance early in the twentieth century. Identified
by the American neurologist George Beard within a few years of Notes from
the Underground, neurasthenia was an encompassing malady whose cardi-
nal symptom was fatigue both mental and muscular. In addition to fatigue,
however, it included an open-ended list of possible symptoms—over 70,
by Beard’s count. “Sufferers from neurasthenia often time wonder and
complain that they have so many symptoms; that their pain and distress
THE MALADY OF AWARENESS 205
attack so many parts and organs.”27 Clearly, without a name to hold them
together, the miscellaneous symptoms of neurasthenia would never cohere.
In this sense, then, the transiently spectacular existence of neurasthenia
testifies to the power of a label to organize and mobilize symptoms. In that
the career of neurasthenia coincided temporally with the discovery, one
after another, of pathogens responsible for infectious disease, this malady
for which a cause couldn’t be found delivered an implied challenge, even
an affront, to the positivism of medicine. So it is with the Underground
Man, who holds himself up as a case of sickness that defies medicine and
rebuts the ideal of a society transformed by the new medical science, as in
Chernyshevsky’s vision of young doctors bringing a rational world to birth.
Syndromes in the tradition of neurasthenia continue to inflame common
symptoms and challenge medicine. The best analysis on record of neur-
asthenia’s successor—CFS itself—characterizes many patients as resentful
and hostile toward medicine and especially psychiatry,28 which approxi-
mates the Underground Man’s attitude toward the champions of reason
and progress.
According to some, “placebo remedies are destined to fill the gap in
knowledge as long as there are illnesses that cannot be treated adequately
and the causes of illness are unknown,”29 but just as new placebos arise on
demand (taking pill form in societies enchanted with pills), so illnesses
that elude understanding, treatment, or both emerge as if to exploit the
knowledge gap and challenge medicine. Among these are many syndromes
that lack laboratory signs—such as neurasthenia or its stepchild chronic
fatigue—and therefore pose a special enigma for medical science. Rather
than using a psychiatric codebook as the key to understanding the Under-
ground Man, we might consult the Underground Man to make better sense
of conditions that contest medicine itself.30
Epilogue
Return to Sources
Gadamer on Medicine
have a sense of their own limits because, above all, they understand that
they can’t make a patient well.
Doctors can never completely entertain the illusion that health is something
they simply “make” or which they can fully control . . . [Medicine] must
participate in the wonderful capacity of life to renew itself, to set itself aright.
The real task which confronts the doctor is that of assisting in this process of
restoration or recovery.
I know only too well how illness can make us insistently aware of our bodily
nature by creating a disturbance in something which normally, in its very
freedom from disturbance, almost completely escapes our attention . . .
What is well-being if it is not precisely this condition of not noticing?
Thus, while it makes sense to ask if someone feels ill, it “would border on
the absurd to ask someone, ‘Do you feel healthy?’ Health is not a condition
RETURN TO SOURCES 209
sleep, its power to release from trouble and restore strength, is so ancient an
idea that it seems fully presumed in the Odyssey, built into the poem’s very
structure. In a poem where the giving and receiving of gifts are ligatures
of social life, sleep itself is often given and received, as when Athena sheds
a sweet sleep over Penelope’s eyes—at least for a while (19.604; 21.358).12
Earlier, Telemachus, Odysseus and the swineherd Eumaeus, after eating,
also accept the gift of sleep (16.481). So too, as the hero is conveyed back
to Ithaca at last by the Phaeacians, there descends on him a profound sleep,
deathlike but without the bitterness of death (13.79–81), as if nothing less
than this draught of Lethe could release Odysseus from the travail of his
adventures and prepare him for the severe tests awaiting in Ithaca. Such
episodes of sleep punctuate the Odyssey, contributing to its ornate, highly
patterned texture.
A Hero Asleep
At one point in the Odyssey a medicine produces the release from care
elsewhere associated with the healing power of sleep. In search of informa-
tion about Odysseus, Telemachus and a son of Nestor are received royally
by Menelaus and Helen, the latter of whom concludes the supper by mixing
a subtle Egyptian drug into the wine. If sleep in the Odyssey releases the
mind’s distress, the properties of this anodyne are such (we are told) that
if you drink it you would not feel distress even if your own brother or son
were murdered in front of your eyes. Helen doesn’t mix the mysterious
compound into the wine surreptitiously, as Circe poisons Odysseus’s men,
but in full view, as part of a richly social proceeding. Presumably she tells
the men of the drug’s properties, thereby priming their expectations, just
as she tells them to pour the wine. Not only, then are the giving and receiv-
ing of gifts central to social life in the Odyssey, but the giving and taking of
a drug are shown as social acts in themselves—as social as the telling and
hearing of stories with their own enchanting effect. Indeed, once the wine
is drunk, what follows are stories; nothing more is said about the mystery
drug, as if it had dissolved not just into the wine but the occasion.
Does medicine still resemble a social transaction? It has been shown
that even as potent a drug as morphine produces a greater effect when
administered openly, as part of a medical ritual, than when given to
patients without their knowledge. This isn’t an anomalous or stand-alone
finding but one of many pointing to the conclusion that medicine is much
more of a social proceeding than commonly supposed. If the very ordi-
nariness of sleep may obscure its enigmatic power, the social background
underwriting the power of the placebo may escape notice precisely because
it is background. Factors that play into therapeutic outcomes
room layout. Thus, the [treatment] context is made up of anything that sur-
rounds the patient under treatment, including doctors, nurses, hospitals,
syringes, pills, machines, and such like, but certainly doctors and nurses are
a very important component of the context, as they can transmit a lot of
information to the patient through their words, attitudes and behaviours.18
The therapeutic power of words is idealized in the story of the boy Odys-
seus being healed with incantations after a gashing by a boar on Mount
Parnassus. For that matter, we might consider Menelaus’s gleaming, well-
ordered palace, with its attentive staff and exemplary hospitality, as a sort
of ideal hospital or clinic in which a heartsick Telemachus finds strength
and solace, as well as reassurance that he is his father’s son—something he
professed in Book 1 to doubt.19
But the value and interest of literature to the student of medicine go
well beyond direct depictions of healing. If the prosaic context of treat-
ment, including everything from the layout of rooms to the garb of nurses,
is powerful enough to transform inert substances into treatments (as the
placebo literature documents), then novelists are right to believe that
rooms, costumes, colors, and all the other props, cues, and furnishings
of social life, right down to the boxes pills arrive in, merit description. It
seems Montaigne was correct: The most ordinary things, the things that
easily escape specific notice, may be sources of wonder.20
Precisely because the power of the placebo is rooted in circumstance
and social cues, the author of a seminal study of the placebo effect in
psychotherapy concludes that there can be nothing like an exhaustive
list of the factors that enable therapy to occur. “Aspects of the context
that determine the meanings of utterances include the words themselves,
facial expressions, gestures, tone of voice, as well as background features
of the interaction . . . The list of potential determinants of the mean-
ings of all human interactions is endless.”21 Henry James said it better:
“Really, universally, relations stop nowhere, and the exquisite problem
of the artist is eternally but to draw, by a geometry of his own, the circle
within which they shall happily appear to do so.”22 The finding that words
feed the power of the placebo—that patients pick up on a doctor’s tone
of voice,23 for example, and frame treatment expectations accordingly—
comes as no surprise to readers who continually witness the effect of
words on the page.
Medicalization turns normal conditions into health problems by engag-
ing the power of the imagination, the same power sufficient to transform
the social packaging and ritual context of treatment into a species of treat-
ment per se. Works of the imagination thus take us to the power source
of the medicalization movement.24 In opening I suggested that an era of
214 THE NOCEBO EFFECT
aggressive medical expansion has much to gain from the spirit of skep-
ticism that went into the making of the humanities. It now appears that
perhaps only the literature of the imagination can enlighten our under-
standing of a process that plays daily on the imagination itself.
Two implications of the power of context follow. First, because context
is so inclusive and open-ended (“relations stop nowhere”), and because the
placebo effect is so context-dependent that it can be understood as a con-
text effect,25 it’s impossible to pinpoint a source of the power of the placebo.
We can’t isolate and test a single factor responsible for the placebo effect
because an entire ensemble of confounders goes into its making. Whether
or not the placebo effect operates nonspecifically on a patient, it’s nonspe-
cific in origin. Second, if the placebo effect can’t be reduced to the expecta-
tion elicited by a white coat or a confident tone or the look of a pill (for all
and much more are at work), neither can the power of suggestion itself be
reduced to the placebo effect. A double-edged sword, suggestion also takes
a noxious form; and the diagnostic trends that sweep through society at
large in an era of medicalization—trends that could never be captured and
calculated in a laboratory—are powerful conductors of this negative force.
Notes
Preamble
Chapter 1
38. John Helzer and Lee Robins, “The Diagnostic Interview Schedule: Its Develop-
ment, Evolution, and Use,” Social Psychiatry and Psychiatric Epidemiology 23
(1988): 6–16.
39. Robert Spitzer and Jerome Wakefield, “DSM-IV Diagnostic Criterion for Clin-
ical Significance: Does It Help Solve the False Positive Problem?” American
Journal of Psychiatry 156 (1999): 1863.
40. On this dogma as well as the overdiagnosis of depression, see Summerfield,
“Depression: Epidemic or Pseudo-Epidemic.”
41. Christopher Dowrick and Allen Frances, “Medicalising Unhappiness: New
Classification of Depression Risks More Patients Being Put on Drug Treatment
from Which They Will Not Benefit,” BMJ 347 (2013): f7410.
42. Michael Klinkman, James Coyne, Susan Gallo et al., “False Positives, False Neg-
atives, and the Validity of the Diagnosis of Major Depression in Primary Care,”
Archives of Family Medicine 7 (1998): 451–61.
43. Clinical Psychiatry: A Textbook for Students and Physicians, abstracted and
adapted from the 7th German edition of Kraepelin’s Lehrbuch der Psychiatrie, ed.
A. Ross Diefendorf (New York: Macmillan, 1923), pp. 386–390 passim.
44. Cf. Katharina Trede, Paola Salvatore, Christopher Baethge et al., “Manic-
Depressive Illness: Evolution in Kraepelin’s Textbook, 1883–1926,” Harvard
Review of Psychiatry 13 (2005): 162: “It . . . appears that his case sampling was
strongly biased toward limited contacts with the relatively poor patients with
severe disorders found in public institutions of the era.”
45. Cited in David Healy, Pharmageddon (Berkeley: University of California Press,
2012), p. 177.
46. Walter Cassidy, Norris Flanagan, Marie Spellman, and Mandel Cohen, “Clinical
Observations in [sic] Manic-Depressive Disease,” JAMA 164 (1957): 1535–46.
47. Walter Cassidy, quoted in Kendler, Muñoz, Murphy, et al., “The Development
of the Feighner Criteria”: 136.
48. David Brody, Steven Hahn, and Robert Spitzer, “Identifying Patients with
Depression in the Primary Care Setting: A More Efficient Method,” Archives
of Internal Medicine 158 (1998): 2469. For an example of a crude test for
depression that yielded a positive rate of over 50 percent, see David Kessler,
Keith Lloyd, Glyn Lewis et al., “Cross Sectional Study of Symptom Attribu-
tion and Recognition of Depression and Anxiety in Primary Care,” BMJ 318;
7181 (1999): 436–40. Also see the commentary immediately following by Iona
Heath, “There Must Be Limits to the Medicalisation of Human Distress.” In
a 1978 survey of 242 teachers that defined depression as “a significant lower-
ing of mood, with or without feelings of guilt, hopelessness and helplessness,
or a drop in one’s self-esteem or self-regard,” fully 95 percent reported such
episodes, “with a mean of six . . . per year.” See Gordon Parker, “Is Depression
Overdiagnosed? Yes,” BMJ 335 (Aug. 18, 2007): 328.
49. Marcus Reidenberg and David Lowenthal, “Adverse Nondrug Reactions,” New
England Journal of Medicine 279 (1968): 678–79.
50. Considering that placebo groups in clinical trials regularly report adverse
effects, some might say the adverse effects leading patients to discontinue
220 NOTES
SSRIs are not necessarily owing to the drugs themselves. See, e.g., Winfried
Rief, Stefan Hofmann, and Yvonne Nestoriuc, “The Power of Expectation—
Understanding the Placebo and Nocebo Phenomenon,” Social and Personality
Psychology Compass 2/4 (2008): 1624–37. Nevertheless, the most cogent expla-
nation of the results of antidepressant clinical trials, including those buried
by their sponsors, is that members of the treatment group break the blind by
identifying the unpleasant effects of the drug. See Kirsch, The Emperor’s New
Drugs, esp. p. 128: “If placebos can induce side effects, antidepressants produce
significantly more of them.” Moreover, if placebo takers report side effects,
it’s probably (as Rief et al. propose) because they ascribe to the studied drug
the sort of ills common in a healthy population—ills made more salient by the
conditions of the study. If we err to take such reports as information about the
drug, why is it correct to interpret similar ills as evidence of clinical depression
in a primary-care patient who doesn’t claim to be depressed?
51. On the paradoxical effects of antidepressants, see Joanna Moncrieff and David
Cohen, “Do Antidepressants Cure or Create Abnormal Brain States?” PLOS
Medicine, June 6, 2006; DOI: 10.1371/journal.pmed.0030240. Commonly
reported side effects of the ADHD drug Ritalin, including weight loss, insom-
nia, and loss of interest, read like a depression checklist. Indeed, they include
depression itself—understood as a symptom of the same order as the others.
52. Cassidy, Flanagan, Spellman, and Cohen, “Clinical Observations in [sic]
Manic-Depressive Disease”: 1535, 1543.
53. Cf. van Praag, “Nosologomania”: 154.
54. David Healy, “Shaping the Intimate: Influences on the Experience of Everyday
Nerves,” Social Studies of Science 34 (2009): 219–45.
55. C. V. R. Blacker and A. W. Clare, “The Prevalence and Treatment of Depression
in General Practice,” Psychopharmacology 95 (1988): S16.
56. On depression as both symptom and diagnosis, see M. Drury, The Danger
of Words (New York: Humanities Press, 1973), p. 3. The author advises, “Let
us beware lest from this unsystematic nomenclature suppositions are drawn,
which then become presumptions and only too easily pass over into established
truths” (ibid.). DSM-IV lists fatigue, disordered sleep, and depression as symp-
toms of Postconcussion Syndrome. On the nonspecific nature of the diagnosis,
see Erica Wang and Felise Zollman, “Postconcussion Syndrome: Diagnostic
Characteristics and Clinical Manifestations,” in Manual of Traumatic Brain
Injury, ed. Felise Zollman (New York: Demos Medical, 2011), pp. 113–18.
57. Allan Horwitz, Creating Mental Illness (Chicago: University of Chicago Press,
2002), p. 45. Built into the history of the term “neurosis” was a presumption that
presenting symptoms were of psychic origin. See Drury, The Danger of Words, p. 4.
58. On the quest for underlying explanations, see Ludwig Wittgenstein, Philosophi-
cal Investigations, tr. G. E. M. Anscombe (New York: Macmillan, 1953), para.
471: “It often happens that we only become aware of the important facts, if we
suppress the question ‘why?’; and then in the course of our investigations these
facts lead us to an answer.” Cf. para. 109: “We must do away with all explana-
tion, and description must take its place.” On the overemphasis on psychological
explanations in American medicine, see Kawanishi, “Somatization of Asians.”
NOTES 221
59. Michael Ross and James Olson, “An Expectancy-Attribution Model of the
Effects of Placebos,” Psychological Review 88 (1981): 409–10. Cf. Carol Lowery,
Douglas Denney, and Michael Storms, “Insomnia: A Comparison of the Effects
of Pill Attributions and Nonpejorative Self-Attributions,” Cognitive Therapy
and Research 3 (1979): 161–64.
60. Kerr White, “The Ecology of Medical Care: Origins and Implications for
Population-Based Healthcare Research,” HSR: Health Services Research 32
(1997): 18. White cites a UK report from 1966.
61. Julie Suhr and John Gunstad, “Postconcussive Symptoms Report: The Relative
Influence of Head Injury and Depression,” Journal of Clinical and Experimen-
tal Neuropsychology 24 (2002): 981–93. On the similarity between symptoms
of concussion and those common in the general population, see Rodney
Vanderploeg and Heather Belanger, “Screening for a Remote History of Mild
Traumatic Brain Injury: When a Good Idea Is Bad,” Journal of Head Trauma
and Rehabilitation 28 (2013): 214.
62. Bernhard Gehr, Christel Weiss, and Franz Porzsolt, “The Fading of Reported
Effectiveness. A Meta-analysis of Randomised Controlled Trials,” BMC Medi-
cal Research Methodology 6 (2006); doi:10.1186/1471-2288-6-25.
63. Hunot, Horne, Leese et al., “A Cohort Study of Adherence to Antidepressants
in Primary Care”: 98.
64. Hunot, Horne, Leese et al., “A Cohort Study of Adherence to Antidepressants in
Primary Care”: 97. Cf. Richard Kravitz, Ronald Epstein, Mitchell Feldman et al.,
“Influence of Patients’ Requests for Direct-to-Consumer Advertised Antidepres-
sants,” JAMA 293 (2005): 1996: “Patients with minor symptoms of short dura-
tion who are prescribed antidepressants at initial presentation would be subject
to short-term adverse effects (e.g., sexual dysfunction) and potential hazards
(including suicidality) that would have to be weighed against marginal gains.”
65. Traolach Brugha, Paul Bebbington, Nicola Singleton et al., “Trends in Service
Use and Treatment for Mental Disorders in Adults Throughout Great Britain,”
British Journal of Psychiatry 185 (2004): 383.
66. Robert Kaplan, Theodore Gamiats, and Dominick Frosch, “Diagnostic and
Treatment Decisions in US Healthcare,” Journal of Health Psychology 9 (2004):
32. Cf. Jerome Hoffman and Richelle Cooper, “Overdiagnosis of Disease: A
Modern Epidemic,” Archives of Internal Medicine 172 (2012): 1123–24.
67. Steven Woolf and Russell Harris, “The Harms of Screening: New Attention to
an Old Concern,” JAMA 307 (2012): 565.
68. In Britain, leaflets advertising breast-cancer screening often don’t mention
harms, either. Fiona Godlee, “Less Medicine Is More,” BMJ 338 (2009): b2561.
Cf. Stewart Justman, “How Did the PSA System Arise?” Journal of the Royal
Society of Medicine 103 (2010): 309–12.
69. Howard Stoate, “Can Health Screening Damage Your Health?” Journal of the
Royal College of General Practitioners 39 (1989): 193–95.
70. Brett Thombs, James Coyne, Pim Cuijpers et al., “Rethinking Recommenda-
tions for Screening for Depression in Primary Care,” CMAJ 2011; doi: 10.1503/
cmaj.111035, p. 2: “Potential harms from routine screening for depression are
rarely made explicit.”
222 NOTES
71. Tanner Caverly, Brandon Combs, Christopher Moriates et al., “Too Much
Medicine Happens Too Often,” JAMA Internal Medicine 174 (2014): 8.
72. John Ioannidis and Joseph Lau, “Completeness of Safety Reporting in Rand-
omized Trials,” JAMA 285 (2001): 437–43.
73. Panagiotis Papanikolaou, Rachel Churchill, Kristian Wahlbeck et al., “Safety
Reporting in Randomized Trials of Mental Health Interventions,” American
Journal of Psychiatry 161 (2004): 1695.
74. Gordon Schiff and William Galanter, “Promoting More Conservative Prescrib-
ing,” JAMA 301 (2009): 866.
75. Moncrieff and Cohen, “Do Antidepressants Cure or Create Abnormal Brain
States?”
76. Joanna Moncrieff, The Myth of the Chemical Cure, p. 239.
77. Remarkably, though, drug trials with children aged four have been run. David Healy,
“The Latest Mania: Selling Bipolar Disorder,” PLoS Medicine 3 (4) (2006): e185.
78. Dr. Tanya Froehlich as cited in MedPage Today, Sept. 30, 2013.
79. See “Children and Psychiatric Drugs,” editorial in The Lancet 372 (2008): 1632.
On the slender benefits of antidepressants for children, see, e.g., Matthew
Miller, Sonja Swanson, Deborah Azrael et al., “Antidepressant Dose, Age, and
the Risk of Deliberate Self-harm,” JAMA Internal Medicine published online
April 28, 2014; doi:10.1001/jamainternmed.2014.1053. On the prescription
of psychoactive drugs to children in advance of safety data, see Mark Riddle,
Elizabeth Kastelic, and Emily Frosch, “Pediatric Psychopharmacology,” Journal
of Child Psychology and Psychiatry 42 (2001): 73–90.
80. On professional membership and special duties, see Stephen Toulmin, “How
Medicine Saved the Life of Ethics,” Perspectives in Biology and Medicine 25
(1982): 736–50.
81. Russell Barkley, Attention-Deficit Hyperactivity Disorder: A Handbook for Diag-
nosis and Treatment, Third Edition (New York: Guilford Press, 2006), p. 9.
82. On nearly 90 percent of ADHD cases in the mild to moderate range, see the
Centers for Disease Control and Prevention Morbidity and Mortality Weekly
Report, Nov. 12, 2010, pp. 1440–43. Official UK estimates of the prevalence of
“severe” ADHD in children appear consistent with the American numbers. See
Iain McClure, “Prescribing Methylphenidate for Moderate ADHD,” BMJ 347
(2013): f6216. On ADHD outcomes, see Rachel Klein, Salvatore Mannuzza,
María Ramos Olazagasti et al., “Clinical and Functional Outcomes of Child-
hood Attention-Deficit/Hyperactivity Disorder 33 Years Later,” Archives of
General Psychiatry 69 (2012): 1295–1303; also Russell Barkley, Mariellen
Fischer, Lori Smallish et al., “Young Adult Follow-Up of Hyperactive Children:
Antisocial Activities and Drug Use,” Journal of Child Psychology and Psychiatry
45 (2004): 195–211. Measured against long-term data, the results of ADHD-
drug trials running 10 or 21 days recede into triviality.
83. Klein, Mannuzza, Ramos Olazagasti et al., “Clinical and Functional Outcomes
of Childhood Attention-Deficit/Hyperactivity Disorder 33 Years Later.”
84. The one ADHD symptom given in DSM-III-R that seems genuinely concerning—
“often engages in physically dangerous activities without considering possible
NOTES 223
consequences . . . e.g., runs into the street without looking”—is ranked lowest in
discriminating power: number 14 of 14. Fidgeting is ranked highest.
85. On the preponderance on mild ADHD, see Rick Mayes, Catherine Bagwell,
and Jennifer Erkulwater, “ADHD and the Rise in Stimulant Use Among Chil-
dren,” Harvard Review of Psychiatry 16 (2008): 161.
86. Ilina Singh, “Beyond Polemics: Science and Ethics of ADHD,” Nature Reviews:
Neuroscience 9 (2008): 960.
87. “Expectancies can directly alter our subjective experience of internal states.
In addition, when we expect a particular outcome, we sometimes unwittingly
behave so as to produce that outcome.” Robert Michael, Maryanne Garry, and
Irving Kirsch, “Suggestion, Cognition, and Behavior,” Current Directions in
Psychological Science 2 (2012): 153.
88. Julie Suhr and John Gunstad, “‘Diagnosis Threat’: The Effect of Negative Expec-
tations on Cognitive Performance in Head Injury,” Journal of Clinical and
Experimental Neuropsychology 24 (2002): 448–57.
89. There is some evidence that children with ADHD overestimate their abilities.
See Julie Sarno Owens, Matthew Goldfine, Nicole Evangelista et al., “A Criti-
cal Review of Self-perception and the Positive Illusory Bias in Children with
ADHD,” Clinical Child and Family Psychology Review 10 (2007): 335–51. I am
unaware of evidence that children with ADHD have a higher-than-warranted
opinion of their ability to control themselves.
90. Chris Travell and John Visser, “‘ADHD Does Bad Stuff to You’: Young People’s
and Parents’ Experiences and Perceptions of Attention Deficit Hyperactivity
Disorder (ADHD),” Emotional and Behavioural Difficulties 11 (2006): 212. On
pill-taking and self-regulation, cf. Duncan Double, “The Limits of Psychiatry,”
BMJ 324 (2002): 900–4.
91. Allen Frances, Essentials of Psychiatric Diagnosis: Responding to the Challenge of
DSM-V (New York: Guilford Press, 2013), p. 9.
92. Frances, Saving Normal, p. 169.
93. Regier, Narrow, Kuhl et al., “Conceptual Development of DSM-V”: 648. The
chair of the DSM-V is the fourth author, David Kupfer.
94. Linda Baumann, Linda Cameron, Rich Zimmerman et al., “Illness Representa-
tions and Matching Labels with Symptoms,” Health Psychology 8 (1989): 453.
95. Baumann, Cameron, Zimmerman et al., “Illness Representations and Match-
ing Labels with Symptoms”: 467.
96. Laura Batstra and Allen Frances, “Diagnostic Inflation: Causes and a Suggested
Cure,” Journal of Nervous and Mental Disease 200 (2012): 474.
97. Horwitz and Wakefield, Loss of Sadness, p. 154.
98. Jerome Frank and Julia Frank, Persuasion as Healing: A Comparative Study of
Psychotherapy (Baltimore: Johns Hopkins University Press, 1991; orig. pub.
1961). Wittgenstein saw psychoanalysis as a mode of persuasion and called
attention to its power to mislead patients into thinking they have actually
made discoveries about themselves. L. Wittgenstein, Lectures & Conversa-
tions on Aesthetics, Psychology and Religious Belief, compiled from notes
taken by Yorick Smythies, Rush Rhees, and James Taylor; ed. Cyril Barrett
224 NOTES
(Berkeley: University of California Press, 1967), pp. 27, 51–52. The same is
arguably true of other psychotherapies.
99. Robert Hahn, “Expectations of Sickness: Concept and Evidence of the
Nocebo Phenomenon,” in How Expectancies Shape Experience, ed. Irving
Kirsch (Washington, DC: American Psychological Association, 1999),
p. 340.
100. Spitzer and Wakefield, “DSM-IV Diagnostic Criterion for Clinical Signifi-
cance”: 1857.
101. Sian Lewis, “A Search for Meaning: Making Sense of Depression,” Journal of
Mental Health 4 (1995): 369–82; unpaginated.
102. The possibility that a patient might actually believe a baseless diagnosis is
strangely missing from psychiatric discussions of the risks of overdiagnosis.
Thus, among the “dangers of diagnostic inflation” Allen Frances and coauthor
cite treatment with unnecessary, possibly dangerous drugs; the “inevitable
stigma” that attaches to someone diagnosed with a mental disorder; and the
shame and ridicule to which psychiatry exposes itself by its own folly. See
Batstra and Frances, “Diagnostic Inflation”: 476. Ignored is the damage done
by implanting false ideas in the patient’s mind.
103. Derek Summerfield, “Proposals for Massive Expansion of Psychological Ther-
apies Would Be Counterproductive Across Society,” British Journal of Psychia-
try 192 (2008): 327.
104. James Pennebaker, The Psychology of Physical Symptoms (New York: Springer-
Verlag, 1982), p. 38.
105. Robert Ferrari, “The Biopsychosocial Model—A Tool for Rheumatologists,”
Best Practice and Research Clinical Rheumatology 14 (2000): 790.
106. In the happy phrase of Robert Hahn, such syndromes might be thought of as
communicable diseases. See Hahn, Sickness and Healing: An Anthropological
Perspective (New Haven: Yale University Press, 1995), p. 89.
107. See, e.g., David Goldberg, Martin Privett, Bedirhan Ustun et al., “The Effects
of Detection and Treatment on the Outcome of Major Depression in Primary
Care: A Naturalistic Study in 15 Cities,” British Journal of General Practice 48
(1998): 1840–44.
108. Thombs, Coyne, Cuijpers et al., “Rethinking Recommendations for Screening
for Depression in Primary Care”: 3.
109. On the risks of diagnosis and treatment of depression, see, e.g., Dowrick,
Beyond Depression, pp. 85–86. On drugs that work some fraction of the
time, see Nicholas Christakis, “Does This Work for You?” BMJ 337 (2008):
1025.
110. “Medical practice expropriates the potential of people to deal with their
human condition in an autonomous way and becomes the source of a new
kind of un-health.” Ivan Illich, “Medical Nemesis,” Journal of Epidemiology
and Community Health 57 (2003): 919. Originally published in 1974.
111. Derek Summerfield, “The Invention of Post-Traumatic Stress Disorder and
the Social Usefulness of a Psychiatric Category,” BMJ 322 (2001): 95.
112. Robert Spitzer, Michael First, and Jerome Wakefield, “Saving PTSD from Itself
in DSM-V,” Journal of Anxiety Disorders 21 (2007): 233–41.
NOTES 225
113. Paul McHugh and Glenn Treisman, “PTSD: A Problematic Diagnostic Cat-
egory,” Journal of Anxiety Disorders 21 (2007): 219.
114. On unrecognized risk, cf. Lawrence Tierney, “Iatrogenic Illness,” Western
Journal of Medicine 151 (1989): 538, 540.
115. Raine Sihvonen, Mika Paavola, Antti Malmivaara et al., “Arthroscopic Partial
Meniscectomy versus Sham Surgery for a Degenerative Miniscal Tear,” New
England Journal of Medicine 369 (2013): 2517.
116. F. Benedetti, M. Lanotte, L. Lopiano et al., “When Words Are Painful: Unrave-
ling the Mechanisms of the Nocebo Effect,” Neuroscience 147 (2007): 269. Cf.
Fabrizio Benedetti, Placebo Effects: Understanding the Mechanisms in Health
and Disease (Oxford: Oxford University Press, 2009), p. 83: “Negative warn-
ings sent out by the mass media may have an important impact on the per-
ceived symptoms of many people.”
117. Eugene Epstein, Manfred Wiesner, and Lothar Duda, “DSM and the
Diagnosis-MacGuffin: Implications for the Self and Society,” Australian and
New Zealand Journal of Family Therapy 34 (2013): 156–57.
118. See Spitzer’s foreword to Horwitz and Wakefield, Loss of Sadness, p. viii. The
first principle of Hahn’s theory of medicine holds, in part, that “conceptions,
theories, and experiences of sickness are elements of socially transmitted cul-
tural systems” (Sickness and Healing, p. 267), implying that our very experi-
ence of illness is colored by the messages we receive about it.
119. Certainly those who receive misleading diagnoses of cancer and then con-
sider themselves survivors have come to believe them. “Overdiagnosis gives
the screenee a false diagnosis of cancer for life.” William Black, “Overdiagno-
sis: An Underrecognized Cause of Confusion and Harm in Cancer Screen-
ing,” Journal of the National Cancer Institute 92 (2000): 1281.
120. David Healy, “Mandel Cohen and Origins of the Diagnostic and Statistical
Manual of Mental Disorders, Third Edition: DSM-III,” History of Psychiatry
13 (2002): 219.
121. DSM-III-R 314.00.
122. Howard Brody and David Waters, “Diagnosis Is Treatment,” Journal of Fam-
ily Practice 10 (1980): 447. Cf. Howard Spiro, The Power of Hope: A Doctor’s
Perspective (New Haven: Yale University Press, 1998), p. 45.
123. Herbert Spiegel and David Spiegel, Trance and Treatment: Clinical Uses of
Hypnosis (Arlington: American Psychiatric Publishing, 2004), p. 11.
124. In this connection cf. a patient quoted in David Karp, Is It Me or My Meds?
Living with Antidepressants (Cambridge, MA: Harvard University Press,
2006), p. 85: “The medication only works . . . because of the doctor who is giv-
ing it to me and the way we work together.” Elsewhere the same patient says,
“I don’t know, I think that it’s important to think about the way society . . . is
constructing these diseases” (p. 79).
125. The literature is vast, but see Arthur Shapiro and Elaine Shapiro, The Powerful
Placebo: From Ancient Priest to Modern Physician (Baltimore: Johns Hopkins
University Press, 1997).
126. K. B. Thomas, “General Practice Consultations: Is There Any Point in Being
Positive?” BMJ 294 (1987): 1200–2.
226 NOTES
127. Note, however, that patients can be harmed even by being told nothing is
wrong with them. See Ray Fitzpatrick, “Telling Patients There Is Nothing
Wrong,” BMJ 313 (1996): 311–12.
128. Pennebaker, Psychology of Physical Symptoms, pp. 153–54.
129. Jessica Bury and J. Michael Bostwick, “Iatrogenic Delusional Parasitosis:
A Case of Physician-Patient Folie a Deux,” General Hospital Psychiatry 32
(2010): 212. On the case of a patient who so invests in the diagnosis of lupus
that she resists giving it up even though it’s false, see Lisa Rosenbaum, “Living
Unlabeled—Diagnosis and Disorder,” New England Journal of Medicine 359
(2008): 1652.
130. Robert Croyle and John Jemmott III, “Psychological Reactions to Risk Factor
Testing” in Mental Representation in Health and Illness, eds. J. A. Skelton and
Robert Croyle (New York: Springer-Verlag, 1991), p. 88.
131. Joan Bloom and Susan Monterossa, “Hypertension Labeling and Sense of
Well-Being,” American Journal of Public Health 71 (1981): 1228–32.
132. Gerald Rosen, Robert Spitzer, and Paul McHugh, “Problems with the Post-
Traumatic Stress Disorder Diagnosis and Its Future in DSM-V,” British Journal
of Psychiatry 192 (2008): 4.
133. Benedetti, Lanotte, Lopiano et al., “When Words Are Painful”: 269.
134. Wang and Zollman, “Postconcussion Syndrome”: 115.
135. Luana Colloca, “The Influence of the Nocebo Effect in Clinical Trials,” Open
Access Journal of Clinical Trials 4 (2012): 65.
136. “All practitioners who evaluate and care for children with ADHD have
encountered children given that label who do not in fact have ADHD.” Ruth
Stein, “Measurement of ADHD Outcomes: Implications for the Future,” Jour-
nal of Pediatric Psychology 32 (2007): 728.
137. “People, including children, are agents, they act, as the philosophers say, under
descriptions. The courses of action they choose, and indeed their ways of
being, are by no means independent of the available descriptions under which
they may act . . . We are especially concerned with classifications that, when
known by people or by those around them, and put to work in institutions,
change the ways in which individuals experience themselves—and may even
lead people to evolve their feelings and behavior in part because they are so
classified.” Ian Hacking, The Social Construction of What? (Cambridge: Har-
vard University Press, 1999), pp. 103–4. The context is a discussion of ADHD.
138. Healy, Pharmageddon, p. 172.
139. D. L. Rosenhan, “On Being Sane in Insane Places,” Science 179 (1973): 254.
140. David LeCouteur, Jenny Doust, Helen Creasey et al., “Political Drive to Screen
for Pre-Dementia: Not Evidence Based and Ignores the Harms of Diagnosis,”
BMJ 347 (2013): f5125; doi 10.1136/bmj.f5125.
141. Brian Draper, Carmelle Peisah, John Snowdon et al., “Early Dementia Diag-
nosis and the Risk of Suicide and Euthansia,” Alzheimer’s and Dementia 6
(2010): 75–82. How can Smith express his anger and confusion over such an
act of diagnostic hubris without seeming demented, or reject it without seem-
ing to be “in denial”?
142. See, e.g., Karp, Is It Me or My Meds?
NOTES 227
143. Howard Leventhal and Michael Diefenbach, “The Active Side of Illness Cog-
nition” in Mental Representation in Health and Illness, eds. Skelton and Croyle,
p. 260.
144. Note, however, the fleeting comment in Frances, Saving Normal, p. 109: “If
you are told you are sick, you feel and act sick, and others treat you as if you
are sick.” Cf. Moncrieff, The Myth of the Chemical Cure, p. 221: Children diag-
nosed with ADHD “may absorb the unhelpful message that they cannot con-
trol their own behaviour without the use of drugs.”
145. Frances, Saving Normal, p. 144.
146. Peter Parry, “Biologism in Psychiatry: A Young Man’s Experience of Being
Diagnosed with ‘Pediatric Bipolar Disorder,’” Journal of Clinical Medicine 3
(2014): 337–38.
147. Parry, “Biologism in Psychiatry”: 341.
148. On the last point, see Temerlin, “Suggestion Effects in Psychiatric Diagnosis.”
149. In actuality, “validity has been sacrificed to achieve reliability.” Nancy
Andreasen, “DSM and the Death of Phenomenology in America: An Exam-
ple of Unintended Consequences,” Schizophrenia Bulletin 33 (2007): 111. The
author was a key player in the making of DSM-III.
150. Richard Smith, “In Search of Non-Disease,” BMJ 324 (2002): 884.
151. Karp, Is It Me or My Meds? p. 89.
152. Michael Thase, Harald Murck, and Anke Post, “Clinical Relevance of Dis-
turbances of Sleep and Vigilance in Major Depressive Disorder: A Review,”
Primary Care Companion to the Journal of Clinical Psychiatry 12 (2010):
PCC.08m00676.
153. Bruce Rollman, Marian Block, and Herbert Schulberg, “Symptoms of Major
Depression and Tricyclic Side Effects in Primary Care Patients,” Journal of
General Internal Medicine 12 (1997): 284–91.
154. Cf. Kelly McKinney and Brian Greenfield, “Self-compliance at ‘Prozac Cam-
pus,’” Anthropology and Medicine 17 (2010): 181. See also Kaitlin Bell Barnett,
Dosed: The Medication Generation Grows Up (Boston: Beacon, 2012), p. 66:
“At the time, she and her parents assumed the drowsiness was a residual
symptom of depression; it wasn’t until many years later that they realized that
the medication [Prozac] might have caused or exacerbated her sleepiness.”
155. Jeffrey Stepnisky, “The Biomedical Self: Hermeneutic Considerations,” Social
Theory and Health 5 (2007): 201.
156. Frederik Feys, Geertruida Bekkering, Kavita Singh et al., “Do Randomized
Clinical Trials with Inadequate Blinding Report Enhanced Placebo Effects
for Intervention Groups and Nocebo Effects for Placebo Groups?” Systematic
Reviews 3 (2014): 54.
157. The MTA Cooperative Group, “A 14-Month Randomized Clinical Trial of
Treatment Strategies for Attention-Deficit/Hyperactivity Disorder,” Archives
of General Psychiatry 56 (1999): 1073–86.
158. See the discussion in Chapter 2.
159. Robert Croyle and Gerald Sande, “Denial and Confirmatory Search: Paradox-
ical Consequences of Medical Diagnosis,” Journal of Applied Social Psychology
18 (1988): 473–90.
228 NOTES
160. Ronald Kessler, Patricia Berglund, Olga Demler et al., “Lifetime Preva-
lence and Age-of-Onset Distributions of DSM-IV Disorders in the National
Comorbidity Survey Replication,” Archives of General Psychiatry 62 (2005):
593–602.
161. Frances, Essentials of Psychiatric Diagnosis, p. 9. Cf. John Mirowsky and Cath-
erine Ross, “Psychiatric Diagnosis as Reified Measurement,” Journal of Health
and Social Behavior 30 (1989): 1–25. “Major Depression” and similar catego-
ries “are linguistic pigeonholes into which some people are placed and other
are not” (22).
162. Cf. Bell Barnett, Dosed, p. 105.
163. Montaigne, “On the Resemblance of Children to Their Fathers,” Complete
Essays, tr. M. A. Screech (London: Penguin, 2003), p. 860.
164. Toulmin, Cosmopolis.
165. On the artificiality of clinical trials, see Russell Glasgow, Thomas Vogt, and
Shawn Boles, “Evaluating the Public Health Impact of Health Promotion
Interventions: The RE-AIM Framework,” American Journal of Public Health
89 (1999): 1322–27.
166. Cf. Horwitz, Creating Mental Illness, pp. 192–93.
167. Montaigne, “On the Power of the Imagination,” p. 109. On latter-day miracle
cures, see, e.g., Sami Timimi, Pathological Child Psychiatry and the Medicali-
zation of Childhood (Hove: Brunner-Routledge, 2002), p. 85: “Some years
ago I attended a conference organized by a pro-Ritalin® and pro-ADHD
group . . . I heard about Ritalin® and other psychostimulants spoken of [sic]
in terms of miracles and complete transformations . . . Non-believers were
spoken about in a derogatory manner as if their concerns amounted to
blasphemy.”
168. Montaigne, “On the Power of the Imagination,” p. 117. It has been dem-
onstrated that subjects who see a drug being administered tend to enjoy a
stronger effect that those who receive it without knowing.
169. Montaigne, “An Apology for Raymond Sebond,” p. 547. Hamlet, in his solilo-
quy at the end of Act II, exclaims on the power of the player’s acting—“his
whole function suiting / With forms to his conceit,” that is, his very body
expressing his imagination.
170. See on this Raymond La Charité, “The Relationship of Judgment and Experi-
ence in the ‘Essais’ of Montaigne,” Studies in Philology 67 (1970): 31–40.
171. Montaigne, “On the Education of Children,” p. 185: “We are not bringing up
a soul; we are not bringing up a body: we are bringing up a man. We must not
split him into two.”
172. Montaigne, “On Experience,” p. 1246.
173. Montaigne, “Apology for Raymond Sebond,” pp. 547–49. On the undermin-
ing of people’s confidence in their own health as a result of medicalization, see
Marcel Verweij, “Medicalization as a Moral Problem for Preventive Medicine,”
Bioethics 13 (1999): 89–113. Cf. Joseph Dumit, Drugs for Life: How Phar-
maceutical Companies Define Our Health (Durham: Duke University Press,
2012), p. 2: “The more we know, the more we fear.” The principle that health is
NOTES 229
Chapter 2
47. On the “utopian vision” guiding the framers of DSM-III, and on the document
itself as a product of “collective fantasy,” see Alix Spiegel, “The Dictionary of
Disorder: How One Man Revolutionized Psychiatry,” New Yorker, Jan. 3, 2005.
48. The bereavement exclusion has been deleted from the DSM. In the equine uto-
pia of Gulliver’s Travels the death of a spouse is not mourned. “I remember my
master having made an appointment with a friend and his family to come to his
house upon some affair of importance; on the day fixed, the mistress and her
two children came very late . . . Her excuse for not coming sooner was that her
husband dying late in the morning, she was a good while consulting her serv-
ants about a convenient place where his body should be laid; and I observed she
behaved herself at our house, as cheerfully as the rest.”
49. On the utopianism of the medicalization movement, see Arthur Barsky, Wor-
ried Sick: Our Troubled Quest for Wellness (Boston: Little, Brown, 1988); Paul
Chodoff, “The Medicalization of the Human Condition,” Psychiatric Services 53
(2002): 627–28.
50. K. L. White, T. F. Williams, and B. G. Greenberg, “The Ecology of Medical
Care,” New England Journal of Medicine 265 (1961): 885–92.
51. Wolfgang Hiller, Winfried Rief, and Elmar Brähler, “Somatization in the Pop-
ulation: From Mild Bodily Misperceptions to Disabling Symptoms,” Social
Psychiatry and Psychiatric Epidemiology 41 (2006): 704.
52. Natalie McGauran, Beate Wieseler, Julia Kreis et al., “Reporting Bias in Medical
Research—A Narrative Review,” Trials 11 (2010): 37.
53. Florence Bourgeois, Srinivas Murthy, and Kenneth Mandl, “Outcome Report-
ing Among Drug Trials Registered in ClinicalTrials.gov,” Annals of Internal
Medicine 153 (2010): 158–66; Joel Lexchin, Lisa Bero, Benjamin Djulbego-
vic et al., “Pharmaceutical Industry Sponsorship and Research Outcome and
Quality: Systematic Review,” BMJ 326 (2003): 1167–70. However, analogous
problems of bias compounded by inattention to harms plague the literature
on social interventions; see Joan McCord, “Cures That Harm: Unanticipated
Outcomes of Crime Prevention Programs,” Annals of the American Academy of
Political and Social Science 587 (May 2003): 16–17.
54. Irving Kirsch, The Emperor’s New Drugs: Exploding the Antidepressant Myth
(New York: Basic, 2010). On suppressed studies of paroxetine, see p. 40.
55. See December 1, 2010 Briefing Document prepared for the FDA’s Oncologic
Drugs Advisory Committee (ODAC).
56. See, again, “Risk-Benefits of 5ARIs for PRCA Prevention” (2013), available at
www.fda.gov.
57. Peter Conrad, The Medicalization of Society: On the Transformation of Human
Conditions into Treatable Disorders (Baltimore: Johns Hopkins University
Press, 2007), p. 148.
58. Frances, Saving Normal, p. 92.
59. John Ioannidis and Joseph Lau, “Completeness of Safety Reporting in Rand-
omized Trials,” JAMA 285 (2001): 437. Cf. Panagiotis Papanikolaou, Rachel
Churchill, Kristian Wahlbeck et al., “Safety Reporting in Randomized Trials of
Mental Health Interventions,” American Journal of Psychiatry 161 (2004): 1695.
234 NOTES
60. Mark Riddle, Elizabeth Kastelic, and Emily Frosch, “Pediatric Psychopharma-
cology,” Journal of Child Psychology and Psychiatry 42 (2001): 73. “Efficacy”
here refers to a drug’s performance in a controlled setting, “effectiveness” to its
performance in clinical practice.
61. Amanda Drews, David Antonuccio, and Irving Kirsch, “A Meta-Analysis of
Randomized Placebo Controlled Trials of Antidepressant Medication in
Depressed Children: Do the Benefits Justify the Risks?” Journal of Mind-Body
Regulation 1 (2011): 85–95.
62. Richard Kravitz, Ronald Epstein, Mitchell Feldman et al., “Influence of Patients’
Requests for Direct-to-Consumer Advertised Antidepressants,” JAMA 293
(2005): 2000.
63. Klinkman et al., “False Positives, False Negatives, and the Validity of the Diag-
nosis of Major Depression in Primary Care”: 454. Cf. Ramin Mojtabai and
Mark Olfson, “Proportion of Antidepressants Prescribed without a Psychiatric
Diagnosis Is Growing,” Health Affairs 30 (2011): 1434–42.
64. David Karp, Is It Me or My Meds? Living with Antidepressants (Cambridge, MA:
Harvard University Press, 2006), pp. 75–76.
65. Kirsch, The Emperor’s New Drugs, p. 44; on side effects of SSRIs vs. placebo, see
p. 128.
66. S. Rose, J. Bisson, R. Churchill et al., “Psychological Debriefing for Preventing
Post Traumatic Stress Disorder (PTSD),” Cochrane Library 2009, Issue 1: 10.
67. Ferguson, “SSRI Antidepressant Medications: Adverse Effects and Tolerabil-
ity”: 24.
68. Arthur Barksy, Ralph Saintfort, Malcolm Rogers et al., “Nonspecific Medica-
tion Side Effects,” JAMA 287 (2002): 625. In a review of the treatment outcomes
of 1,131 depressed patients, only 46 percent completed a minimum course of
antidepressant medication, defined as two months or more. See Kimberley
Hepner, Melissa Rowe, Kathryn Rost et al., “The Outcome of Adherence to
Practice Guidelines on Depression Outcomes,” Annals of Internal Medicine 147
(2007): 320–29.
69. John Goethe, Stephen Woolley, Alex Cardoni et al., “Selective Serotonin Reup-
take Inhibitor Discontinuation: Side Effects and Other Factors That Influence
Medication Adherence,” Journal of Clinical Psychopharmacology 27 (2007):
451–58.
70. David Healy, “The Dilemmas Posed by New and Fashionable Treatments,”
Advances in Psychiatric Treatment 7 (2001): 323.
71. Elias Khawam, Georgia Laurencic, and Donald Malone, “Side Effects of Anti-
depressants: An Overview,” Cleveland Clinic Journal of Medicine 73 (2006): 351;
see also Christopher Ron Cantrell, Michael Eaddy, Manan Shah et al., “Meth-
ods for Evaluating Patient Adherence to Antidepressant Therapy,” Medical Care
44 (2006): 300–3.
72. Kirsch, The Emperor’s New Drugs. Cf. Montaigne, “On Experience,” Complete
Essays, tr. M. A. Screech (London: Penguin, 2003), p. 1233: “Since we must risk
being wrong, let us risk what gives us pleasure. The world does the reverse,
thinking that nothing does you good unless it hurts.”
NOTES 235
73. Randolph Evans and Malcolm Rogers, “Headaches and the Nocebo Effect,”
Headache 43 (2003): 1114.
74. Ferguson, “SSRI Antidepressant Medications: Adverse Effects and Tolerabil-
ity”: 22; cf. W. F. Boyer and John Feighner, “Side-Effects of the Selective Sero-
tonin Re-uptake Inhibitors,” in Selective Serotonin Re-uptake Inhibitors: The
Clinical Use of Citalopram, Fluoxetine, Fluvoxamine, Paroxetine, and Sertraline
(New York: John Wiley & Sons, 1991), p. 146: “Most of the published compara-
tive studies [of SSRIs] have been against amitriptyline and imipramine, which
may have biased the side-effect outcome in favour of the SSRIs.”
75. Ferguson, “SSRI Antidepressant Medications.”
76. Khawam, Laurencic, Malone, “Side Effects of Antidepressants”: 353.
77. Frances, Saving Normal, pp. xiv, 169.
78. Allen Frances, “The New Crisis of Confidence in Psychiatric Diagnosis,”
Annals of Internal Medicine online, May 17, 2013; doi:10.7326/0003-4819-159-
3-201308060-00655; cf. Saving Normal, p. 174.
79. Steven Hyman, “The Diagnosis of Mental Disorders: The Problem of Reifica-
tion,” Annual Review of Clinical Psychology 6 (2010): 171.
80. American Urological Association, Prostate-Specific Antigen Best Practice State-
ment: 2009 Update, p. 9.
81. Marc Theoret, Yang-Min Ning, Jenny Zhang et al., “The Risks and Benefits of
5α-Reductase Inhibitors for Prostate-Cancer Prevention,” New England Journal
of Medicine 365 (2011): 99.
82. Ian Thompson, Phyllis Goodman, Catherine Tangen et al., “Long-Term Sur-
vival of Participants in the Prostate Cancer Prevention Trial,” New England
Journal of Medicine 369 (2013): 603.
83. Ray Moynihan, Jenny Doust, and David Henry, “Preventing Overdiagnosis: How
to Stop Harming the Healthy,” BMJ 2012;344:e3502 doi: 10.1136/bmj.e3502.
84. Michel Foucault, Discipline and Punish: The Birth of the Prison, tr. Alan Sheri-
dan (NY: Vintage, 1995), p. 296.
85. Christopher Dowrick, Beyond Depression: A New Approach to Understanding
and Management (Oxford: Oxford University Press, 2009), p. 101.
86. Cf. Annemarie Goldstein Jutel, Putting a Name to It: Diagnosis in Contempo-
rary Society (Baltimore: Johns Hopkins University Press, 2011), p. 48: “Normal-
ity, the dreaded bugaboo to which we aspire in cultural practices is . . . deeply
ingrained in medicine.”
87. Horwitz and Wakefield, Loss of Sadness; Christopher Lane, Shyness: How Nor-
mal Behavior Became a Sickness (New Haven: Yale University Press, 2007).
88. Jeffrey Tuttle, Neil Scheurich, and John Ranseen, “Prevalence of ADHD
Diagnosis and Nonmedical Prescription Stimulant Use in Medical Students,”
Academic Psychiatry 34 (2010): 220–23.
89. Frances, Saving Normal, p. xiv.
90. Ronald Kessler, Patricia Berglund, Olga Demler et al., “Lifetime Prevalence and
Age-of-Onset Distributions of DSM-IV Disorders in the National Comorbidity
Survey Replication,” Archives of General Psychiatry 62 (2005): 593–602. On the
requirement of “clinically significant” distress or impairment, see 593.
236 NOTES
127. Caryn Lerman, Bruce Trock, Barbara Rimer et al., “Psychological Side Effects
of Breast Cancer Screening,” Health Psychology 10 (1991): 259–67.
128. A. M. Dart, H. Alban Davies, T. Griffith et al., “Does It Help to Undiagnose
Angina?” European Heart Journal 4 (1983): 462. Cf. Ira Ockene, Marilyn Shay,
Joseph Alpert et al., “Unexplained Chest Pain in Patient with Normal Coro-
nary Arteriograms—A Follow-up Study of Functional Status,” New England
Journal of Medicine 303 (1980): 1249–52.
129. Robin Munro, “Judicial Psychiatry in China and Its Political Abuses,” Colum-
bia Journal of Asian Law 14 (2000): 42.
130. Martyn Patfield, “Undiagnosis: An Important New Role for Psychiatry,” Aus-
tralasian Psychiatry 19 (2011): 107–9.
131. Nathaniel Pallone and James Hennessy, “Benevolent Misdiagnosis,” Society 31
(1994): 11–17.
132. Robert Hahn, “Expectations of Sickness: Concept and Evidence of the Nocebo
Phenomenon” in How Expectancies Shape Experience, ed. Irving Kirsch
(Washington, DC: American Psychological Association, 1999), p. 333.
133. Hannah Decker, The Making of DSM-III: A Diagnostic Manual’s Conquest of
American Psychiatry (Oxford: Oxford University Press, 2013), p. 330.
134. Shorter, How Everyone Became Depressed, p. 127.
135. Shorter, How Everyone Became Depressed, p. 16.
136. Robert Spitzer and Jerome Wakefield, “DSM-IV Diagnostic Criterion for
Clinical Significance: Does It Help Solve the False Positive Problem?” Ameri-
can Journal of Psychiatry 156 (1999): 1862.
Chapter 3
10. David Jopling, Talking Cures and Placebo Effects (Oxford: Oxford University
Press, 2008), pp. 92–93.
11. Christopher Lasch, The Culture of Narcissism: American Life in an Age of
Diminishing Expectations (New York: Warner, 1979), p. 99. Recall that “nar-
cissistic personality disorder,” a popular diagnosis in its day, was not listed in
DSM-II.
12. Christopher Williams and Paul Lees-Haley, “Perceived Toxic Exposure: A
Review of Four Cognitive Influences on Perception of Illness,” Journal of Social
Behavior and Personality 8 (1993): 500.
13. Rodney Vanderploeg and Heather Belanger, “Screening for a Remote History
of Mild Traumatic Brain Injury: When a Good Idea Is Bad,” Journal of Head
Trauma and Rehabilitation 28 (2013): 215.
14. Jessica Bury and J. Michael Bostwick, “Iatrogenic Delusional Parasitosis: A
Case of Physician-Patient Folie a Deux,” General Hospital Psychiatry 32 (2010):
210–12.
15. Irving Kirsch, The Emperor’s New Drugs: Exploding the Antidepressant Myth
(New York: Basic, 2010), p. 146.
16. Yvonne Nestoriuc, E. John Orav, Matthew Liang et al., “Prediction of Nonspe-
cific Side Effects in Rheumatoid Arthritis Patients by Beliefs about Medicines,”
Arthritis Care and Research 62 (2010): 791.
17. James Pennebaker, The Psychology of Physical Symptoms (New York: Springer-
Verlag, 1982), p. 42.
18. Martin Myers, John Cairns, and Joel Singer, “The Consent Form as a Possi-
ble Cause of Side Effects,” Clinical Pharmacology and Therapeutics 42 (1987):
250–53.
19. Pennebaker, Psychology of Physical Symptoms, p. 53.
20. Jopling, Talking Cures and Placebo Effects, p. 111.
21. David Spiegel, “Placebos in Practice,” BMJ 329 (2004): 927.
22. W. Häuser, E. Hansen, and P. Enck, “Nocebo Phenomena in Medicine: Their
Relevance in Everyday Clinical Practice,” Deutsches Artzeblatt International
109 (2012): 461.
23. Maire Brid Mac Bride, Sandhya Pruthi, Therese Bevers et al., “The Evolution
of Breast Self-Examination to Breast Awareness,” The Breast Journal 18 (2012):
641–43.
24. Delia Cioffi, “Asymmetry of Doubt in Medical Self-Diagnosis: The Ambiguity of
‘Uncertain Wellness,’” Journal of Personality and Social Psychology 61 (1991): 970.
25. Gerd Gigerenzer, Jutta Mata, and Ronald Frank, “Public Knowledge of Breast
and Prostate Cancer Screening in Europe,” JNCI 101 (2009): 1216–20. On
the loss of confidence in one’s health as a result of medicalization, see Marcel
Verweij, “Medicalization as a Moral Problem for Preventive Medicine,”
Bioethics 13 (1999): 113.
26. Ben Goldacre, Bad Pharma: How Drug Companies Mislead Doctors and Harm
the Public (New York: Faber and Faber, 2012), p. 260.
27. Gerald Rosen, Robert Spitzer, and Paul McHugh, “Problems with the Post-
Traumatic Stress Disorder Diagnosis and Its Future in DSM-V,” British Journal
of Psychiatry 192 (2008): 4.
240 NOTES
28. Wiley Mittenberg, Diane DiGiulio, Sean Perrin et al., “Symptoms Following
Mild Head Injury: Expectation as Aetiology,” Journal of Neurology, Neurosur-
gery, and Psychiatry 55 (1992): 200–4; cf. Lars-Gunnar Lundh, “Placebo, Belief,
and Health. A Cognitive-Emotional Model,” Scandinavian Journal of Psychol-
ogy 28 (1987): 134.
29. James Ferguson, “SSRI Antidepressant Medications: Adverse Effects and Toler-
ability,” Primary Care Companion to the Journal of Clinical Psychiatry 3 (2001):
22–27.
30. Arthur Barsky and Jonathan Borus, “Functional Somatic Syndromes,” Annals
of Internal Medicine 130 (1999): 913–14.
31. Michael King, Victoria Holt, and Irwin Nazareth, “Women’s Views of Their
Sexual Difficulties: Agreement and Disagreement with Clinical Diagnoses,”
Archives of Sexual Behavior 36 (2007): 281–88.
32. Hannah Decker, The Making of DSM-III: A Diagnostic Manual’s Conquest of
American Psychiatry (Oxford: Oxford University Press, 2013), p. 160.
33. Robert Spitzer and Jerome Wakefield, “DSM-IV Diagnostic Criterion for Clin-
ical Significance: Does It Help Solve the False Positive Problem?” American
Journal of Psychiatry 156 (1999): 1861.
34. Allan Horwitz and Jerome Wakefield, The Loss of Sadness: How Psychiatry
Transformed Normal Sorrow into Depressive Disorder (Oxford: Oxford Univer-
sity Press, 2007), p. 146.
35. Bertram Forer, “The Fallacy of Personal Validation: A Classroom Demon-
stration of Gullibility,” Journal of Abnormal and Social Psychology 44 (1949):
119–20.
36. E.g., Kirsch, The Emperor’s New Drugs, p. 129.
37. Christopher Lane, Shyness: How Normal Behavior Became a Sickness (New
Haven: Yale University Press, 2007), p. 109.
38. R. Ferrari and H. Schrader, “The Late Whiplash Syndrome: A Biopsychoso-
cial Approach,” Journal of Neurology, Neurosurgery, and Psychiatry 70 (2001):
725.
39. Robert Hahn, “The Nocebo Phenomenon: Scope and Foundations” in The Pla-
cebo Effect: An Interdisciplinary Exploration, ed. Anne Harrington (Cambridge:
Harvard University Press, 1997), p. 61.
40. Hans-Georg Gadamer, The Enigma of Health, tr. Jason Gaiger and Nicholas
Walker (Stanford: Stanford University Press, 1996), p. 96.
41. Arthur Barksy, Ralph Saintfort, Malcolm Rogers et al., “Nonspecific Medica-
tion Side Effects,” JAMA 287 (2002): 623.
42. Phil Edwards, “Questionnaires in Clinical Trials: Guidelines for Optimal
Design and Administration,” http://www.trialsjournal.com/content/11/1/2.
43. M. Sibille, N. Deigat, A. Janin et al., “Adverse Events in Phase-1 Studies: A
Report in 1015 Healthy Volunteers,” European Journal of Clinical Pharmacology
54 (1998): 13–20; Pierre Rosenzweig, Sylvie Brohier, and Andreas Zipfel, “The
Placebo Effect in Healthy Volunteers: Influence of Experimental Conditions
on the Adverse Events Profile during Phase I Studies,” Clinical Pharmacy and
Therapeutics 54 (1993): 578–83.
44. Ferguson, “SSRI Antidepressant Medications”: 24.
NOTES 241
45. Grant Iverson, Brian Brooks, Lynn Ashton et al., “Interview Versus Question-
naire Symptom Reporting in People with the Postconcussion Syndrome,” Jour-
nal of Head Trauma and Rehabilitation 25 (2010): 26–27.
46. Allyson Harrison, Melanie Edwards, and Kevin Parker, “Identifying Students
Faking ADHD: Preliminary Findings and Strategies for Detection,” Archives of
Clinical Neuropsychology 22 (2007): 577–88.
47. Winfried Rief, Yvonne Nestoriuc, Anna von Lilienfeld-Toal et al., “Differences in
Adverse Effect Reporting in Placebo Groups in SSRI and Tricyclic Antidepres-
sant Trials: Systematic Review and Meta-Analysis,” Drug Safety 32 (2009): 1050.
48. Iverson, Brooks, Ashton et al., “Interview Versus Questionnaire Symptom
Reporting”: 29.
49. Dawes, House of Cards, p. 262.
50. S. Rose, J. Bisson, R. Churchill et al., “Psychological Debriefing for Preventing
Post Traumatic Stress Disorder (PTSD),” Cochrane Library 2009, Issue 1: 2. On
CISD and RCTs, see also Simon Wessely, “A Defence of the Randomized Con-
trolled Trial in Mental Health,” BioSocieties 2 (2007): 115–27.
51. Richard McNally, Richard Bryant, and Anke Ehlers, “Does Early Psychologi-
cal Intervention Promote Recovery from Posttraumatic Stress?” Psychological
Science in the Public Interest 4 (2003): 46.
52. Scott Lilienfeld, “Psychological Treatments That Cause Harm,” Perspectives on
Psychological Science 2 (2007): 59.
53. Cf. Richard Bootzin and Elaine Bailey, “Understanding Placebo, Nocebo, and
Iatrogenic Treatment Effects,” Journal of Clinical Psychology 61 (2005): 874.
54. Ethan Watters, Crazy Like Us: The Globalization of the American Psyche (New
York: Free Press, 2010), p. 119.
Interlude
1. The phrases used in various studies appear in Lene Vase, Joseph Riley, and
Donald Price, “A Comparison of Placebo Effects in Clinical Analgesic Trials
Versus Studies of Placebo Analgesia,” Pain 99 (2002): 447.
2. Stewart Justman, To Feel What Others Feel: Social Sources of the Placebo Effect
(San Francisco: University of California Medical Humanities Press, 2012),
ch. 11.
3. John Haygarth, Of the Imagination as a Cause and a Cure of Disorders of the
Body (Bath: Cruttwell, 1800), pp. 3–4.
4. Franklin Miller and Ted Kaptchuk, “Deception of Subjects in Neuroscience
Research,” Journal of Neuroscience 28 (2008): 4841–43.
5. Don Long, Sumio Uematsu, and Richard Kouba, “Placebo Responses to Medi-
cal Device Therapy for Pain,” Stereotactic and Functional Neurosurgery 53
(1989): 151–52. On the surprising efficacy of sham electroconvulsive therapy,
see Pat Bracken, Philip Thomas, Sami Timimi et al., “Psychiatry Beyond the
Current Paradigm,” British Journal of Psychiatry 201 (2012): 430–34.
6. Barbara Henker and Carol Whalen, “Hyperactivity and Attention Deficits,”
American Psychologist 44 (1989): 221.
242 NOTES
7. Gary Saul Morson, “Anna Karenina” in Our Time: Seeing More Wisely (New
Haven: Yale University Press, 2007), p. 45.
Chapter 4
1. Lynn Payer, Disease-Mongers: How Doctors, Drug Companies, and Insurers Are
Making You Feel Sick (New York: John Wiley, 1992), p. 16.
2. Richard Rosenfield, “Harm,” Otolaryngology—Head and Neck Surgery 146
(2012): 687–89.
3. Experiments designed to induce harm also incur ethical objections, though
placebo experiments aren’t free of such problems either.
4. Winfried Häuser, Emil Hansen, and Paul Enck, “Nocebo Phenomena in Medi-
cine,” Deutsches Ärtzteblatt International 109 (2012): 460.
5. Irving Kirsch, The Emperor’s New Drugs: Exploding the Antidepressant Myth
(New York: Basic, 2010), p. 137.
6. F. Benedetti, M. Lanotte, L. Lopiano et al., “When Words Are Painful: Unrave-
ling the Mechanisms of the Nocebo Effect,” Neuroscience 147 (2007): 268.
Cf. Martina Amanzio, Luca Latini Corazzini, Lena Vase et al., “A Systematic
Review of Adverse Events in Placebo Groups of Anti-Migraine Clinical Trials,”
Pain 146 (2009): 261.
7. Thomas Weihrauch and Thomas Gauler, “Placebo-Efficacy and Adverse Effects
in Controlled Clinical Trials,” Arzneimittel-Forschung 49 (1999): 390.
8. A. F. Cooper, “Whose Illness Is It Anyway? Why Patient Perceptions Matter,”
International Journal of Clinical Practice 52 (1998): 555.
9. K. B. Thomas, “General Practice Consultations: Is There Any Point in Being
Positive?” BMJ 294 (1987): 1200–1202.
10. Derek Summerfield, “Cross-Cultural Perspectives on the Medicalization of
Human Suffering,” in Posttraumatic Stress Disorder: Issues and Controversies,
ed. G. M. Rosen (New York: John Wiley, 2004), pp. 233–45, esp. p. 237.
11. Thomas, “General Practice Consultations”: 1201. Häuser, Hansen, and Enck,
“Nocebo Phenomena in Medicine,” identify “causing uncertainty” as a mecha-
nism of the nocebo effect (461). Exactly what Thomas did when he told patients
suffering from transient illnesses that he didn’t know what was wrong with them
(and that he couldn’t be sure a pill would help them, if a pill were prescribed).
12. K. B. Thomas, “Temporarily Dependent Patient in General Practice,” British
Medical Journal, Vol. 1, No. 5908 (March 30, 1974): 626.
13. Harold Sox, Jr., Iris Margulies, and Carol Hill Sox, “Psychologically Mediated
Effects of Diagnostic Tests,” Annals of Internal Medicine 95 (1981): 680–85.
14. Sox, Jr., Margulies, and Hill Sox, “Psychologically Mediated Effects of Diagnos-
tic Tests”: 684.
15. Sox, Jr., Margulies, and Hill Sox, “Psychologically Mediated Effects of Diagnos-
tic Tests”: 685.
16. On “incidentalomas,” see, e.g., Lisa Rosenbaum, “Living Unlabeled—Diagnosis
and Disorder,” New England Journal of Medicine 359 (2008): 1650–53.
17. See, e.g., Laura Batstra and Allen Frances, “Diagnostic Inflation: Causes and a
Suggested Cure,” Journal of Nervous and Mental Disease 200 (2012): 474–79.
NOTES 243
18. Howard Brody and David Waters, “Diagnosis Is Treatment,” Journal of Family
Practice 10 (1980): 448–49.
19. Marcus Huibers and Simon Wessely, “The Act of Diagnosis: Pros and Cons of
Labelling Chronic Fatigue Syndrome,” Psychological Medicine 36 (2006): 898.
A thoughtful critique of “diagnosis for the sake of giving a diagnosis” is John
Bedson, Rob McCarney and Peter Croft, “Labelling Chronic Illness in Primary
Care: A Good or a Bad Thing?” British Journal of General Practice 54 (2004):
936.
20. Payer, Disease-Mongers, p. 259.
21. Allen Frances, “The New Crisis of Confidence in Psychiatric Diagnosis,”
Annals of Internal Medicine online, May 17, 2013, doi:10.7326/0003-4819-
159-3-201308060-00655.
22. On the abuse potential of stimulants, see, e.g., Nora Volkow, “Stimulant Medi-
cations: How to Minimize Their Reinforcing Effects?” American Journal of
Psychiatry 163 (2006): 359–61.
23. Fabrizio Benedetti, “Drugs and Placebos: What’s the Difference?” EMBO
Reports (2014): 2. doi:10.1002/embr.201338399.
24. Steven Miles, The Hippocratic Oath and the Ethics of Medicine (Oxford: Oxford
University Press, 2004), p. 129.
25. Simon Gilbody, Trevor Sheldon, and Simon Wessely, “Should We Screen for
Depression?” BMJ 332 (2006): 1027.
26. Allan Horwitz and Jerome Wakefield, The Loss of Sadness: How Psychia-
try Transformed Normal Sorrow into Depressive Disorder (Oxford: Oxford
University Press, 2007), p. 164.
27. Michael Klinkman, James Coyne, Susan Gallo et al., “False Positives, False Neg-
atives, and the Validity of the Diagnosis of Major Depression in Primary Care,”
Archives of Family Medicine 7 (1998): 451–61.
28. Harold Pincus, Wendy Wakefield Davis, and Laurie McQueen, “‘Subthresh-
old’ Mental Disorders: A Review and Synthesis of Studies on [sic] Minor
Depression and Other ‘Brand Names,’” British Journal of Psychiatry 174
(1999): 294.
29. Brett Thombs, James Coyne, Pim Cuijpers et al., “Rethinking Recommen-
dations for Screening for Depression in Primary Care,” Canadian Medical
Association Journal 2011. doi:10.1503 /cmaj.111035, p. 4.
30. Horwitz and Wakefield, Loss of Sadness, p. 156.
31. Sian Lewis, “A Search for Meaning: Making Sense of Depression,” Journal of
Mental Health 4 (1995): 369–82; unpaginated.
32. Joanna Moncrieff, The Myth of the Chemical Cure: A Critique of Psychiatric
Drug Treatment (New York: Palgrave Macmillan, 2008), pp. 164–68.
33. Larry Goldman, Nancy Nielsen, and Hunter Champion, “Awareness, Diag-
nosis, and Treatment of Depression,” Journal of General Internal Medicine 14
(1999): 569–80.
34. If stigma were really such a formidable deterrent, antidepressants wouldn’t be
enjoying a booming market.
35. Richard Mayou, “Medically Unexplained Physical Symptoms,” BMJ 303 (1991):
534–35.
244 NOTES
36. Kurt Kroenke, Robert Spitzer, Janet Williams et al., “Physical Symptoms in Pri-
mary Care: Predictors of Psychiatric Disorders and Functional Impairment,”
Archives of Family Medicine 3 (1994): 774–79.
37. Richard Kravitz, Ronald Epstein, and Mitchell Feldman, “Influence of Patients’
Requests for Direct-to-Consumer Advertised Antidepressants,” JAMA 293
(2005): 1996.
38. In a diagnostic questionnaire for primary-care patients, Spitzer—the author of the
DSM criteria—has actually substituted “more than half the days” for the “nearly every
day” criterion. See Robert Spitzer, Kurt Kroenke, Janet Williams et al., “Validation
and Utility of a Self-Report Version of PRIME-MD,” JAMA 282 (1999): 1737–44.
39. G. Andrews, T. M. Anderson, T. Slade et al., “Classification of Anxiety and
Depressive Disorders: Problems and Solutions,” Depression and Anxiety 25
(2008): 274–81. Robert Spitzer, the designer of DSM-III, has argued that “reduc-
ing the number of symptoms required to make a diagnosis of depression . . .
would make it easier for primary care physicians to remember and evaluate
them.” David Brody, Steven Hahn, and Robert Spitzer, “Identifying Patients with
Depression in the Primary Care Setting: A More Efficient Method,” Archives of
Internal Medicine 158 (1998): 2470.
40. Like the two-question test proposed by the USPSTF, the user-friendly version
of the DSM-IV criteria for depression given in T. M. Luhrmann’s study of the
culture of American psychiatry leaves out all temporal requirements except
the period of two weeks, a critical omission that may well reflect the way the
criteria are actually used. T. M. Luhrmann, Of Two Minds: An Anthropologist
Looks at American Psychiatry (New York: Vintage, 2001), pp. 34–35.
41. Robert Spitzer and Jerome Wakefield, “DSM-IV Diagnostic Criterion for Clin-
ical Significance: Does It Help Solve the False Positive Problem?” American
Journal of Psychiatry 156 (1999): 1857.
42. Gregory Simon, Michael VonKorff, Marco Piccinelli et al., “An International
Study of the Relation between Somatic Symptoms and Depression,” New
England Journal of Medicine 341 (1999): 1332.
43. John Williams, Polly Hitchcock Noël, Jeffrey Cordes et al., “Is This Patient Clin-
ically Depressed?” JAMA 287 (2002): 1160-1, 1169.
44. David Healy, Derelie Mangin, and Barbara Mintzes, “The Ethics of Rand-
omized Placebo Controlled Trials of Antidepressants with Pregnant Women,”
International Journal of Risk and Safety in Medicine 22 (2010): 7–16.
45. Luhrmann, Of Two Minds, p. 236.
46. Jeffrey Pyne, Kathryn Rost, Farah Farahati et al., “One Size Fits Some: The
Impact of Patient Treatment Attitudes on the Cost-Effectiveness of a Depres-
sion Primary-Care Intervention,” Psychological Medicine 35 (2005): 850.
47. Kathryn Rost, Paul Nutting, Jeffrey Smith et al., “Improving Depression Out-
comes in Community Primary Care Practice,” Journal of General Internal Med-
icine 16 (2001): 147.
48. Kroenke, Spitzer, Williams et al., “Physical Symptoms in Primary Care”: 777.
49. On distress caused by a diagnosis of low sexual desire—not by the lack of desire
itself—see Annemarie Goldstein Jutel, Putting a Name to It: Diagnosis in Contem-
porary Society (Baltimore: Johns Hopkins University Press, 2011), pp. 105–6.
NOTES 245
50. J. Alexander Bodkin, Harrison Pope, Michael Detke et al., “Is PTSD Caused by
Traumatic Stress?” Journal of Anxiety Disorders 21 (2007): 176–82.
51. Allan Horwitz, Creating Mental Illness (Chicago: University of Chicago Press,
2002), p. 80.
52. On a popular treatment for PTSD, see my To Feel What Others Feel: Social
Sources of the Placebo Effect (San Francisco: University of California Medical
Humanities Press, 2012), ch. 11.
53. Mitchell Wilson, “DSM-III and the Transformation of American Psychiatry: A
History,” American Journal of Psychiatry 150 (1993): 408.
54. Foreword to Horwitz and Wakefield, Loss of Sadness, p. viii.
55. Joanna Moncrieff and Irving Kirsch, “Efficacy of Antidepressants in Adults,”
BMJ 331 (2005): 155–57.
56. Fritz Schröder, Jonas Hugosson, Monique Roobol et al., “Screening and
Prostate-Cancer Mortality in a Randomized European Study,” New England
Journal of Medicine 360 (2009): 1320, 1326.
57. Fritz Schröder, Jonas Hugosson, Monique Roobol et al., “Prostate Cancer Mor-
tality at 11 Years of Follow-up,” New England Journal of Medicine 366 (2012):
981–90.
58. Otis Brawley, “Trends in Prostate Cancer in the United States,” Journal of the
National Cancer Institute Monographs 45 (2012): 154.
59. R. Brian Haynes, David Sackett, D. Wayne Taylor et al., “Increased Absentee-
ism from Work after Detection and Labeling of Hypertensive Patients,” New
England Journal of Medicine 299 (1978): 741–44.
60. Haynes, Sackett, Taylor et al., “Increased Absenteeism”: 742.
61. August 25, 2013.
62. Allyson Harrison, Melanie Edwards, and Kevin Parker, “Identifying Students
Faking ADHD: Preliminary Findings and Strategies for Detection,” Archives of
Clinical Neuropsychology 22 (2007): 577–88.
63. Harrison, Edwards, Parker, “Identifying Students Faking ADHD”: 579.
64. Weihrauch, and Gauler, “Placebo-Efficacy and Adverse Effects in Controlled
Clinical Trials”: 388.
65. Robert Ferrari, “The Biopsychosocial Model—A Tool for Rheumatologists,”
Best Practice and Research Clinical Rheumatology 14 (2000): 790.
66. Peter Conrad, The Medicalization of Society: On the Transformation of Human
Conditions into Treatable Disorders (Baltimore: Johns Hopkins University
Press, 2007), p. 59.
67. Laura Batstra and Allen Frances, “DSM-V Further Inflates Attention Deficit
Hyperactivity Disorder,” Journal of Nervous and Mental Disease 200 (2012):
486–88.
68. Herbert Spiegel, “Nocebo: The Power of Suggestibility,” Preventive Medicine 26
(1997): 620.
69. Arthur Barsky and Jonathan Borus, “Functional Somatic Syndromes,” Annals
of Internal Medicine 130 (1999): 915.
70. Julie Suhr and John Gunstad, “Further Exploration of the Effect of ‘Diagno-
sis Threat’ on Cognitive Performance in Individuals with Mild Head Injury,”
Journal of the International Neuropsychological Society 11 (2005): 23.
246 NOTES
71. Richard Morrow, E. Jane Garland, and James Wright, “Influence of Relative
Age and Diagnosis on Treatment of Attention-Deficit/Hyperactivity Disorder
in Children,” Canadian Medical Association Journal 184 (2012): 755–62. The
grade cutoff in Canada is Dec. 31.
72. Jean-Luc Mommaerts, Beerens Gerlinde, Van den Block Lieve et al., “Influence
of Methylphenidate Treatment Assumptions on Cognitive Function in Healthy
Young Adults in a Double-Blind, Placebo-Controlled Trial,” Psychological
Research and Behavior Management 6 (2013): 65–74.
73. Fabrizio Benedetti, Placebo Effects: Understanding the Mechanisms in Health
and Disease (Oxford: Oxford University Press, 2009), p. 136.
74. Cf. Jennifer Davis-Berman and Frances Pestello, “Medicating for ADD/ADHD:
Personal and Social Issues,” International Journal of Mental Health and Addic-
tion 8 (2010): 487.
75. Ilina Singh, “Clinical Implications of Ethical Concepts: Moral Self-
Understandings in Children Taking Methylphenidate for ADHD,” Clinical
Child Psychology and Psychiatry 112 (2007): 172.
76. A. Sandler, C. Glesne, and G. Geller, “Children’s and Parents’ Perspectives on
Open-Label Use of Placebos in the Treatment of ADHD,” Child: Care, Health
and Development 34 (2008): 111–20.
77. Sandler, Glesne, and Geller, “Children’s and Parents’ Perspectives on Open-
Label Use of Placebos in the Treatment of ADHD”: 111.
78. Ibid.
79. Adrian Sandler, “Placebo Effects in Developmental Disabilities: Implications
for Research and Practice,” Mental Retardation and Developmental Disabilities
Research Reviews 11 (2005): 164–70.
80. Glen Cayler, David Lynn, and Edwina Stein, “Effect of Cardiac ‘Nondisease’
on Intellectual and Perceptual Motor Development,” British Heart Journal 35
(1973): 543–47.
81. The MTA Cooperative Group, “A 14-Month Randomized Clinical Trial of
Treatment Strategies for Attention-Deficit/Hyperactivity Disorder,” Archives of
General Psychiatry 56 (1999): 1075.
82. Ibid.
83. Peter Jensen, L. Eugene Arnold, James Swanson et al., “3-Year Follow-up of the
NIMH MTA Study,” Journal of the American Academy of Child and Adolescent
Psychiatry 46 (2007): 989–1002.
84. Jensen, Arnold, Swanson et al., “3-Year Follow-up of the NIMH MTA Study”:
999.
85. On medical self-fulfilling prophecies, see Nicholas Christakis, Death Foretold:
Prophecy and Diagnosis in Medical Care (Chicago: University of Chicago Press,
1999), ch. 6.
86. Barbara Henker and Carol Whalen, “Hyperactivity and Attention Deficits,”
American Psychologist 44 (1989): 221.
87. Moncrieff, The Myth of the Chemical Cure, p. 221.
88. Russell Barkley, “Major Life Activity and Health Outcomes Associated with
Attention-Deficit/Hyperactivity Disorder,” Journal of Clinical Psychiatry 63
(2002), supp. 12: 10–15.
NOTES 247
Chapter 5
4. David Healy, “The Latest Mania: Selling Bipolar Disorder,” PLoS Medicine 3 (4)
(2006): e185. Cf. Joanna Moncrieff, “The Medicalisation of ‘Ups and Downs’:
The Marketing of the New Bipolar Disorder,” Transcultural Psychiatry 51
(2014): 581–98.
5. Roy Porter, The Greatest Benefit to Mankind (Norton, 1999), p. 442.
6. Kenneth Kendler, Rodrigo Muñoz, and George Murphy, “The Development of
the Feighner Criteria,” American Journal of Psychiatry 167 (2010): 134–42.
7. Walter Cassidy, Norris Flanagan, Marie Spellman, and Mandel Cohen, “Clini-
cal Observations in [sic] Manic-Depressive Disease,” JAMA 164 (1957): 1535.
8. Joanna Moncrieff and David Cohen, “How Do Psychiatric Drugs Work?” BMJ,
published online May 29, 2009; doi:10.1136/bmj.b1963
9. Winfried Häuser, Emil Hansen, and Paul Enck, “Nocebo Phenomena in Medi-
cine: Their Relevance in Everyday Clinical Practice,” Deutsches Artzeblatt
International 109 (2012): 463. Cf. Kate Faasse and Keith Petrie, “The Nocebo
Effect: Patient Expectations and Medication Side Effects,” Postgraduate Medi-
cal Journal, doi:10.1136/postgradmedj-2012-131730; Kate Faasse, Tim Cundy,
Greg Gamble et al., “The Effect of an Apparent Change to a Branded or Generic
Medication on Drug Effectiveness and Side Effects,” Psychosomatic Medicine 75
(2013): 90–96.
10. A. Branthwaite and P. Cooper, “Analgesic Effects of Branding in Treatment of
Headaches,” British Medical Journal 202 (1981): 1576–78.
11. E.g., Stewart Justman, “Placebos in the Clinic,” Journal of the Royal Society of
Medicine 106 (2013): 208–9.
12. Luana Colloca and Franklin Miller, “Harnessing the Placebo Effect: The Need
for Translational Research,” Philosophical Transactions of the Royal Society B
366 (2011): 1922–30.
13. Benedetti, Placebo Effects, p. 34.
14. Winfried Rief, Jerry Avorn, and Arthur Barsky, “Medication-Attributed Adverse
Effects in Placebo Groups,” Archives of Internal Medicine 166 (2006):158.
15. Kurt Kroenke, Robert Spitzer, Janet Williams et al., “Physical Symptoms in Pri-
mary Care: Predictors of Psychiatric Disorders and Functional Impairment,”
Archives of Family Medicine 3 (1994): 777.
16. Marcus Reidenberg and David Lowenthal, “Adverse Nondrug Reactions,” New
England Journal of Medicine 279 (1968): 678–79.
17. Henry Beecher, “The Powerful Placebo,” JAMA, Dec. 24, 1955; 1602–6.
18. Jeremy Greene, Generic: The Unbranding of Modern Medicine (Baltimore: Johns
Hopkins University Press, 2104), p. 63.
19. See Stewart Justman, The Psychological Mystique (Evanston: Northwestern Uni-
versity Press, 1998), ch. 2.
20. On the use of KOLs, see, e.g., Melissa Raven and Peter Parry, “Psychotropic
Marketing Practices and Problems: Implications for DSM-V,” Journal of Nerv-
ous and Mental Disease 200 (2012): 512–16.
21. Edward Bernays, Propaganda (Brooklyn, New York: Ig, 2005; orig. pub. 1928), p. 98.
22. Zinelabidine Abouelfadel and E. David Crawford, “Experience of Prostate
Cancer Awareness Week,” in Prostate Cancer Screening, eds. Ian Thompson,
Martin Resnick, and Eric Klein (Totowa, NJ: Humana Press, 2001), p. 241. On
NOTES 249
Chapter 6
13. David Brody, Steven Hahn, and Robert Spitzer, “Identifying Patients with
Depression in the Primary Care Setting: A More Efficient Method,” Archives of
Internal Medicine 158 (1998): 2469–75.
14. See Joel Smith, “Depression: Darker Than Darkness,” American Scholar 66
(1997): 495–99 for a strong argument that actual depression entails a radical
reduction of energy, an also radical reduction in enthusiasm for living, a loss
of hope and corresponding preoccupation with death, and self-loathing. What
DSM refers to as “depressed mood” is an almost euphemistic description of
the third.
15. Wayne Katon and Herbert Schulberg, “Epidemiology of Depression in Primary
Care,” General Hospital Psychiatry 14 (1992): 244.
16. On incipient risk, see Joseph Dumit, Drugs for Life: How Pharmaceutical Com-
panies Define Our Health (Durham: Duke University Press, 2012).
17. Robert Spitzer and Jerome Wakefield, “DSM-IV Diagnostic Criterion for Clin-
ical Significance: Does It Help Solve the False Positive Problem?” American
Journal of Psychiatry 156 (1999): 1859.
18. See Robert Spitzer, Jean Endicott, and Eli Robins, “Research Diagnostic Cri-
teria: Rationale and Reliability,” Archives of General Psychiatry 35 (1978): 777:
“Whereas the Feighner criteria required an acknowledged dysphoric mood,
the criteria for major depressive disorder accept as an alternative the pres-
ence of a pervasive loss of interest or pleasure. This was done with the clinical
recognition that many patients with an obvious depressive syndrome do not
acknowledge feeling depressed.”
19. For an argument that you can commit suicide more or less unknowingly, see A.
Alvarez, The Savage God: A Study of Suicide (New York: Norton, 1990), p. 155.
20. Aaron Beck, Gary Brown, Robert Steer et al., “Differentiating Anxiety and
Depression: A Test of the Cognitive Content-Specificity Hypothesis,” Journal
of Abnormal Psychology 96 (1987): 179–83.
21. Michael Peterson, John Holbrook, De Von Hales et al., “Contributions of the
History, Physical Examination, and Laboratory Investigation in Making Medi-
cal Diagnoses,” Western Journal of Medicine 156 (1992): 164.
22. Aaron Beck as cited in Robyn Dawes, A House of Cards: Psychology and Psycho-
therapy Built on Myth (New York: Free Press, 1994), p. 284.
23. Edward Shorter, How Everyone Became Depressed: The Rise and Fall of the
Nervous Breakdown (Oxford: Oxford University Press, 2013), p. 189.
24. Kurt Kroenke, Robert Spitzer, Janet Williams et al., “Physical Symptoms in Pri-
mary Care: Predictors of Psychiatric Disorders and Functional Impairment,”
Archives of Family Medicine 3 (1994): 778.
25. Larry Goldman, Nancy Nielsen, and Hunter Champion, “Awareness, Diag-
nosis, and Treatment of Depression,” Journal of General Internal Medicine 14
(1999): 575.
26. F. Dudley Hart, “The Importance of Non-Disease,” The Practitioner 211 (1973):
194.
27. Michael King, Victoria Holt, and Irwin Nazareth, “Women’s Views of Their
Sexual Difficulties: Agreement and Disagreement with Clinical Diagnoses,”
Archives of Sexual Behavior 36 (2007): 281–88.
254 NOTES
28. Cf. Trisha Greenhalgh, How to Read a Paper: The Basics of Evidence-Based
Medicine (Oxford: BMJ Books, 2010), p. 140.
29. Cf. Joanna Moncrieff, The Myth of the Chemical Cure: A Critique of Psychiatric
Drug Treatment (New York: Palgrave Macmillan, 2008), p. 235.
30. Jeffrey Pyne, Kathryn Rost, Farah Farahati et al., “One Size Fits Some: The
Impact of Patient Treatment Attitudes on the Cost-Effectiveness of a Depres-
sion Primary-Care Intervention,” Psychological Medicine 35 (2005): 842.
31. Horwitz and Wakefield, Loss of Sadness, p. 146.
32. Gilbody, Sheldon, Wessely, “Should We Screen for Depression?”: 1028.
33. D. L. Rosenhan, “On Being Sane in Insane Places,” Science 179 (1973): 253.
34. Robert Spitzer, “On Pseudoscience in Science, Logic in Remission, and Psy-
chiatric Diagnosis: A Critique of Rosenhan’s ‘On Being Sane in Insane Places,’”
Journal of Abnormal Psychology 84 (1975): 448. On the tendentiousness of
Freudian interpretation, see M. Drury, The Danger of Words (New York:
Humanities Press, 1973), p. 17.
35. Interview with the author in Christopher Lane, Shyness: How Normal Behavior
Became a Sickness (New Haven: Yale University Press, 2007), p. 77. Cf. Wor-
thington Hooker, Lessons from the History of Medical Delusions (New York:
Baker and Scribner, 1850), p. 36.
36. Catalona, “Screening for Prostate Cancer: Enthusiasm”: 114.
37. Otis Brawley and Barnett Kramer, “Cancer Screening in Theory and Practice,”
Journal of Clinical Oncology 23 (2005): 298.
38. Marc Theoret, Yang-Min Ning, Jenny Zhang et al., “The Risks and Benefits of
5α-Reductase Inhibitors for Prostate-Cancer Prevention,” New England Journal
of Medicine 365:2 (2011): 99.
39. Marcus Reidenberg and David Lowenthal, “Adverse Nondrug Reactions,” New
England Journal of Medicine 279 (1968): 679.
40. Howard Parnes, Ian Thompson, and Leslie Ford, “Prevention of Hormone-
Related Cancers: Prostate Cancer,” Journal of Clinical Oncology 23 (2005): 374.
41. Naomi Breslau and Ronald Kessler, “The Stressor Criterion in DSM-IV Post-
traumatic Stress Disorder: An Empirical Investigation,” Biological Psychiatry 50
(2001): 699–704.
42. Shorter, How Everyone Became Depressed, p. 193.
43. Cited in Ben Goldacre, Bad Pharma: How Drug Companies Mislead Doctors
and Harm the Public (New York: Faber and Faber, 2012), p. 259.
44. Peter Albertsen, “What Is the Risk Posed by Prostate Cancer?” JNCI Mono-
graphs 45 (2012): 169–74.
45. Allison Glass, Matthew Cooperberg, Maxwell Meng et al., “Role of Active Sur-
veillance in the Management of Localized Prostate Cancer,” JNCI Monographs
45 (2012): 204.
46. Alexandre Zlotta, Shin Egawa, Dmitry Pushkar et al., “Prevalence of Prostate
Cancer on Autopsy: Cross-Sectional Study on Unscreened Caucasian and
Asian Men,” JNCI 105 (2013): 1050–58.
47. Dr. William Catalona as cited in The Economist, March 8, 2014.
48. Laura Esserman and Ian Thompson, “Solving the Overdiagnosis Dilemma,”
JNCI 102 (2010): 582.
NOTES 255
49. Frederick Crews, “Talking Back to Prozac,” New York Review of Books, Dec,
6, 2007.
50. Horwitz and Wakefield, Loss of Sadness, p. 135.
51. Kenneth Kendler, Rodrigo Muñoz, and George Murphy, “The Development of
the Feighner Criteria,” American Journal of Psychiatry 167 (2010): 136.
52. Kendler, Muñoz, Murphy, “The Development of the Feighner Criteria”: 140.
53. Spitzer, Endicott, Robins, “Research Diagnostic Criteria: Rationale and Reli-
ability”: 777.
54. Irving Kirsch, The Emperor’s New Drugs: Exploding the Antidepressant Myth
(New York: Basic, 2010), pp. 29, 48, 50.
55. See DSM-V, p. 811.
56. According to later calculations, “To prevent one death from prostate cancer . . .
1055 men would need to be invited for screening and 37 cancers would need to
be detected.” Fritz Schröder, Jonas Hugosson, Monique Roobol et al., “Prostate-
Cancer Mortality at 11 Years of Follow-up,” New England Journal of Medicine
366 (2012): 981. A 2014 update of ERSPC concluded that “further quantifica-
tion of harms and their reduction are still considered a prerequisite for the
introduction of populated-based screening.” At this point mass screening had
already been in effect for 20 years in the United States. The Lancet, Early Online
Publication, August 7, 2014; doi:10.1016/S0140-6736(14)60525-0.
57. See the Foreword by Robert Spitzer, MD, to The Loss of Sadness.
58. Ian Thompson, Phyllis Goodman, Catherine Tangen et al., “The Influence of
Finasteride on the Development of Prostate Cancer,” New England Journal of
Medicine 349 (2003): 215–24.
59. Goldman, Nielsen, and Champion, “Awareness, Diagnosis, and Treatment of
Depression,”: 572.
60. Horwitz and Wakefield, Loss of Sadness, p. 145. On taking diagnosis to the peo-
ple, see also David Healy, Pharmageddon (Berkeley: University of California
Press, 2012), p. 6.
61. See http://suicide.com/suicidecrisiscenter/whylive.html.
62. William Catalona, “Informed Consent for Prostate-Specific Antigen Screen-
ing,” Urology 61 (2003): 18.
63. Amir Qaseem, Michael Barry, Thomas Denberg et al., “Screening for Prostate
Cancer: A Guidance Statement from the Clinical Guidelines Committee of the
American College of Physicians,” Annals of Internal Medicine online publica-
tion 04/08/13.
64. David Ransohoff, Mary McNaughton Collins, and Floyd Fowler, “Why Is Pros-
tate Cancer Screening So Common When the Evidence Is So Uncertain? A
System without Negative Feedback,” American Journal of Medicine 113 (2002):
663–67.
65. Steven Woloshin and Lisa Schwartz, “The U. S. Postal Service and Cancer Screen-
ing: Stamps of Approval?” New England Journal of Medicine 340 (1999): 886.
66. Lane, Shyness, p. 70.
67. Horwitz and Wakefield, Loss of Sadness, p. 150.
68. Joan Bloom and Susan Monterossa, “Hypertension Labeling and Sense of Well-
Being,” American Journal of Public Health 71 (1981): 1228–32.
256 NOTES
69. See, e.g., Goldman, Nielsen, and Champion, “Awareness, Diagnosis, and
Treatment of Depression,”: 570.
70. Horwitz and Wakefield, Loss of Sadness, p. 149.
71. Stewart Justman, “Uninformed Consent: Mass Screening for Prostate Cancer,”
Bioethics 26 (2012): 143–48.
72. The favored term is now “active surveillance.”
73. Stewart Justman, “How Did the PSA System Arise?” Journal of the Royal Society
of Medicine 103 (2010): 309–12.
74. Barron Lerner, The Breast Cancer Wars (Oxford: Oxford University Press
2001), p. 48.
75. Jennifer Croswell, David Ransohoff, and Barnett Kramer, “Principles of Cancer
Screening: Lessons from History and Study Design Issues,” Seminars in Oncol-
ogy 37 (2010): 208.
76. Lydia Pace and Nancy Keating, “A Systematic Assessment of Benefits and Risks
to Guide Breast Cancer Screening Decisions,” JAMA 311 (2014): 1332.
77. David Newman, “Screening for Breast and Prostate Cancers: Moving Toward
Transparency,” JNCI 102 (2010): 1008–11. For an exception see Anna Bill-
Axelson, Lars Holmberg, Mirja Ruutu et al., “Radical Prostatectomy versus
Watchful Waiting in Early Prostate Cancer,” New England Journal of Medicine
352 (2005): 1977–84; and the follow-up article in NEJM 364 (2011): 1708–17.
78. Jane Wardle and Rosemary Pope, “The Psychological Costs of Screening for
Cancer,” Journal of Psychosomatic Research 36 (1992): 612. Some believe radia-
tion therapy for mammography-detected cancers may contribute to cardiovas-
cular mortality.
79. Otis Brawley, How We Do Harm: A Doctor Breaks Ranks about Being Sick in
America (New York: St. Martin’s, 2011), chapters 20–21.
80. Richard McNally, Richard Bryant, and Anke Ehlers, “Does Early Psychological
Intervention Promote Recovery from Posttraumatic Stress?” Psychological Sci-
ence in the Public Interest 4 (2003): 46.
81. Peter Ditto and Robert Croyle, “Understanding the Impact of Risk Factor
Test Results: Insights from a Basic Research Program,” in Psychosocial Effects
of Screening for Disease Prevention and Detection, ed. Robert Croyle (Oxford:
Oxford University Press, 1995), p. 160.
82. FDA Briefing Document, Oncologic Drugs Advisory Committee Meeting,
December 1, 2010: 28, 33.
83. Thompson, Goodman, Tangen et al., “The Influence of Finasteride on the
Development of Prostate Cancer”: 217.
84. Brawley, How We Do Harm, p. 229.
85. Gilbody, Sheldon, Wessely, “Should We Screen for Depression?”: 1028.
86. Carol Roeloffs, Cathy Sherbourne, Jürgen Unützer et al., “Stigma and Depres-
sion among Primary Care Patients,” General Hospital Psychiatry 25 (2003):
311–15. In a British study that looked for evidence that patients on antidepres-
sants suffered from real or imagined stigma, 70 percent declined to participate.
See Sara Garfield, Felicity Smith, and Sally-Anne Francis, “The Paradoxical
Role of Antidepressant Medication—Returning to Normal Functioning While
Losing the Sense of Being Normal,” Journal of Mental Health 12 (2003): 521–35.
NOTES 257
Chapter 7
1. A version of this chapter appeared as “The Folly of Systems: The Satiric Tradi-
tion and Mental Disorders,” Philosophy and Literature 37 (2013): 472–85.
2. John Helzer and Lee Robins, “The Diagnostic Interview Schedule: Its Develop-
ment, Evolution, and Use,” Social Psychiatry and Psychiatric Epidemiology 23
(1988): 15.
3. On DSM-III as a paradigm shift, see Hannah Decker, The Making of DSM-III:
A Diagnostic Manual’s Conquest of American Psychiatry (Oxford: Oxford Uni-
versity Press, 2013), p. 309f.
4. The Standard Edition of the Complete Psychological Works of Sigmund Freud
(London: Hogarth Press, 1953): 21, 115. See also my “Freud and His Nephew,”
Social Research 61 (1994): 457–76. Cf. Lionel Trilling, Sincerity and Authentic-
ity (Cambridge, MA: Harvard University Press, 1971), p. 151. Trillling’s son has
reported that in his private life Trilling “never hoped for glamorous insights or
a decisive cure” from psychoanalysis, only for some responsible reconciliation
with reality. James Trilling, “My Father and the Weak-Eyed Devils,” American
Scholar, Spring 1999: 17.
5. Thomas Kuhn, The Structure of Scientific Revolutions (Chicago: University of
Chicago Press, 1970), p. 24.
6. The revolution that installed DSM-III might be compared to the revolution that
enshrined the new linguistic philosophy and with it “a specialized set of prob-
lems, methods and techniques. Thanks to Moore, Russell and Wittgenstein, the
authors were saying, we are now ‘real philosophers’ and can look scientists in
the eye . . . We have given ourselves a respectable academic job, on which we
can embark with confidence and industry.” Allan Janik and Stephen Toulmin,
Wittgenstein’s Vienna (Chicago: Ivan R. Dee, 1996), p. 259. DSM-III enabled
psychiatry to look medicine in the eye and gave it plenty to do.
7. Mitchell Wilson, “DSM-III and the Transformation of American Psychiatry: A
History,” American Journal of Psychiatry 150 (1993): 399–410.
8. Gerald Rosen, Robert Spitzer, and Paul McHugh, “Problems with the Post-
Traumatic Stress Disorder Diagnosis and Its Future in DSM-V,” British Journal
NOTES 259
71. Valeria Sobol, “Reading the Invisible: The Mind, the Body, and the Medical
Examiner in Lev Tolstoy’s Anna Karenina,” Studies in Slavic Cultures 2 (2001):
17. Among the presumed causes of tuberculosis around the time Tolstoy
wrote War and Peace were “depressive emotions.” See Susan Sontag, Illness as
Metaphor and AIDS and Its Metaphors (New York: Picador, 1989), p. 54. The
antituberculosis drug iproniazid was in fact the first antidepressant; see Joanna
Moncrieff, The Myth of the Chemical Cure: A Critique of Psychiatric Drug Treat-
ment (New York: Palgrave Macmillan, 2008), pp. 121–24.
72. Leo Tolstoy, War and Peace, tr. Louise and Aylmer Maude (Oxford: Oxford
University Press, 1991), pp. 615, 617.
73. Lee Robins, “Diagnostic Grammar and Assessment: Translating Criteria into
Questions,” Psychological Medicine 19 (1989): 67.
74. Gary Saul Morson, “The Prosaics of Process,” Literary Imagination 2 (2000): 387.
75. Cf. Payer, Disease-Mongers, ch. 9.
76. Leslie Aaron and Dedra Buchwald, “A Review of the Evidence for Overlap
among Unexplained Clinical Conditions,” Annals of Internal Medicine 134
(2001): 877.
77. John Talbott, “An In-Depth Look at DSM-III: An Interview with Robert
Spitzer,” Hospital and Community Psychiatry 31 (1980): 28. Cf. William James,
Writings 1902–1910 (New York: Library of America, 1988), p. 138: “Most cases
are mixed cases, and we should not treat our classifications with too much
respect.” A turgid analysis of the difficulty of fitting things into categories is to
be found in Geoffrey Bowker and Susan Leigh Star, Sorting Things Out: Clas-
sification and Its Consequences (Cambridge, MA: MIT Press, 2000).
78. Allan Horwitz, Creating Mental Illness (Chicago: University of Chicago Press,
2002), p. 116.
79. It thus makes sense to ask whether the results of clinical trials run in one coun-
try necessarily apply to another, and whether “patients diagnosed with depres-
sion in China [are] really the same as patients diagnosed with depression in
California.” Ben Goldacre, Bad Pharma: How Drug Companies Mislead Doctors
and Harm the Public (New York: Faber and Faber, 2012), p. 115.
80. James Ferguson, “SSRI Antidepressant Medications: Adverse Effects and Toler-
ability,” Primary Care Companion to the Journal of Clinical Psychiatry 3 (2001):
22–27.
81. David Baldwin and Thomas Foong, “Antidepressant Drugs and Sexual
Dysfunction,” British Journal of Psychiatry 202 (2013): 397. The authors con-
tinue, “The presence of antidepressant treatment-emergent sexual dysfunction
can significantly reduce quality of life and self-esteem, and impose burdens on
interpersonal relationships, over and above those associated with depression.”
82. Edward Shorter, From Paralysis to Fatigue: A History of Psychosomatic Illness in
the Modern Era (New York: Free Press, 1992).
83. Simon Wessely, “Old Wine in New Bottles: Neurasthenia and ‘ME,’” Psychologi-
cal Medicine 20 (1990): 47.
84. John Feighner, Eli Robins, and Samuel Guze, “Diagnostic Criteria for Use in
Psychiatric Research,” Archives of General Psychiatry 26 (1972): 62.
NOTES 263
85. A brilliant 1983 review of Ann Beattie’s fiction sets its tone as “cool sorrow”
and “listless despair.” “Her parched, exhausted stories themselves seem numb.
She reads them herself in a plain, flat, utterly toneless voice that suggests dead-
ened feelings or, on occasion, a determined effort to fight back tears.” Here
then is a world depressed. The characters take Valium. See Pico Iyer, “The
World According to Beattie,” Partisan Review 50 (1983): 548–53. So popular
was Valium that by the mid-1970s it was used generically for drugs of its kind.
See Nikolas Rose, “Neurochemical Selves,” Society 41 (2003): 49.
86. David Healy, “The Dilemmas Posed by New and Fashionable Treatments,”
Advances in Psychiatric Treatment 7 (2001): 324.
87. Healy, Pharmageddon, pp. 32–33. For a hypothetical case that would have
read as anxiety in the 1960s and depression in the 1990s, see p. 260.
88. Edward Shorter, How Everyone Became Depressed: The Rise and Fall of the
Nervous Breakdown (Oxford: Oxford University Press, 2013), pp. 110, 119; cf.
Christopher Dowrick, Beyond Depression: A New Approach to Understand-
ing and Management (Oxford: Oxford University Press, 2009), pp. 67–68, and
David Pilgrim and Richard Bentall, “The Medicalisation of Misery: A Critical
Realist Analysis of the Concept of Depression,” Journal of Mental Health 8
(1999): 261–74.
89. Ferguson, “SSRI Antidepressant Medications.”
90. Bernd Löwe, Robert Spitzer, Janet Williams et al., “Depression, Anxiety and
Somatization in Primary Care: Syndrome Overlap and Functional Impair-
ment,” General Hospital Psychiatry 30 (2008): 191–92.
91. David Lodge, Therapy (London: Penguin, 1995). See pp. 4, 218, 107, etc.
92. Dickens, Little Dorrit (Oxford: Oxford University Press, 1967), p. 627.
93. Dickens, Little Dorrit, p. 203.
94. Dickens, Little Dorrit, p. 206.
95. Dickens, Little Dorrit, p. 641.
96. See Stewart Justman, “Direct and Indirect Guilt in Little Dorrit,” Soundings,
Spring-Summer 2002: 39–52.
97. B. C. Douglas, “Dickens’ Characters on the Couch: An Example of Teaching
Psychiatry Using Literature,” Medical Humanities 34 (2008): 64–69.
98. Horace Dobell, Lectures on the Germs and Vestiges of Disease, and on the
Prevention of the Invasion and Fatality of Disease by Periodical Examinations
(London: John Churchill, 1861), p. 150.
99. Jennifer Croswell, David Ransohoff, and Barnett Kramer, “Principles of Can-
cer Screening: Lessons from History and Study Design Issues,” Seminars in
Oncology 37 (2010): 202.
100. Nortin Hadler, Worried Sick: A Prescription for Health in an Overtreated
America (Chapel Hill: University of North Carolina Press, 2008), p. 135.
101. Herb Kutchins and Stuart Kirk, Making Us Crazy: DSM: The Psychiatric Bible
and the Creation of Mental Disorders (New York: Free Press, 2003), p. x.
102. Michael King, Victoria Holt, and Irwin Nazareth, “Women’s Views of Their
Sexual Difficulties: Agreement and Disagreement with Clinical Diagnoses,”
Archives of Sexual Behavior 36 (2007): 281–88.
264 NOTES
103. Nancy Andreasen, “Acute and Delayed Posttraumatic Stress Disorders: A His-
tory and Some Issues,” American Journal of Psychiatry 161 (2004): 1321–23.
104. Jonathan Shay, Odysseus in America: Combat Trauma and the Trials of Home-
coming (New York: Scribner, 2002), p. 141. If Odysseus really suffered from
PTSD, the wound he receives in Book 11 of the Iliad would be mentioned
somewhere in the Odyssey. It isn’t.
105. Montaigne, “Apology for Raymond Sebond,” in Complete Essays, p. 662.
106. For the same reason, but also because it pretends cultural omniscience, the
DSM-V claim that “ADHD occurs in most cultures in about 5% of children”
falls to the ground.
107. Richard McNally, “Progress and Controversy in the Study of Posttraumatic
Stress Disorder,” Annual Reviews in Psychology 54 (2003): 230.
108. McNally, “Progress and Controversy in the Study of Posttraumatic Stress Dis-
order”: 232. DSM-V excludes the diagnosis of PTSD in someone who neither
suffered or witnessed trauma nor learned that someone closely connected
suffered trauma. Ruled out is “exposure through electronic media, television,
movies, or pictures, unless this exposure is work related.”
109. W. H. M. Castro, M. Schilgen, S. Meyer et al., “Do ‘Whiplash Injuries’ Occur
in Low-Speed Rear Impacts?” European Spine Journal 6 (1997): 374.
110. Paul Lees-Haley, J. Randall Price, and Christopher Williams, “Use of the Impact
of Events Scale in the Assessment of Emotional Distress and PTSD May Pro-
duce Misleading Results,” Journal of Forensic Neuropsychology 2 (2001): 49.
111. Ian Hacking, Mad Travelers: Reflections on the Reality of Transient Mental Ill-
nesses (Cambridge: Harvard University Press, 1998), p. 83.
112. David Healy, “Mandel Cohen and Origins of the Diagnostic and Statistical
Manual of Mental Disorders, Third Edition: DSM-III,” History of Psychiatry
13 (2002): 210–11.
Chapter 8
1. Dostoevsky, Notes from the Underground, tr. Jane Kentish (Oxford: Oxford Uni-
versity Press, 1991), p. 12. Subsequent quotations from pp. 45, 47, 50, 51, 77.
2. M. M. Bakhtin, Problems of Dostoevsky’s Poetics, tr. Caryl Emerson (Minne-
apolis: University of Minnesota Press, 1984), p. 59.
3. Notes from the Underground, p. 10.
4. Notes from the Underground, p. 13.
5. See e.g., Witold Rybczynski, How Architecture Works: A Humanist’s Toolkit
(New York: Farrar, Straus and Giroux, 2013), p. 224.
6. Nikolai Chernyshevsky, What Is to Be Done? tr. Michael Katz (Ithaca: Cornell
University Press, 1989), p. 370.
7. What Is to Be Done? p. 104.
8. What Is to Be Done? p. 313.
9. Notes from the Underground, p. 22.
10. What Is to Be Done? p. 378.
11. E.g., Ethan Watters, Crazy Like Us: The Globalization of the American Psyche
(New York: Free Press, 2010).
NOTES 265
12. David Foster Wallace, “The Depressed Person,” Harper’s Magazine, Jan.
1998, p. 64.
13. Gary Saul Morson, “Anna Karenina” in Our Time: Seeing More Wisely (New
Haven: Yale University Press, 2007), p. 158.
14. Richard Shusterman, “William James, Somatic Introspection, and Care of the
Self,” Philosophical Forum 36 (2005): 431.
15. Tolstoy, Anna Karenina, tr. Constance Garnett; rev. Leonard Kent and Nina
Berberova (New York: Modern Library, 1993), p. 289.
16. Morson, “Anna Karenina” in Our Time, p. 228.
17. Utopia, tr. Robert M. Adams (New York: Norton, 1975), pp. 59–60.
18. Arthur Barsky and Jonathan Borus, “Functional Somatic Syndromes,” Annals
of Internal Medicine 130 (1999): 910.
19. Notes from the Underground, p. 7.
20. Jon Stone, Wojtek Wojcik, Daniel Durrance et al., “What Should We Say to
Patients with Symptoms Unexplained by Disease? The ‘Number Needed to
Offend,’” BMJ 325 (2002): 1449–50.
21. Michael André Bernstein, Bitter Carnival: Ressentiment and the Abject Hero
(Princeton: Princeton University Press, 1992).
22. Elaine Showalter, Hystories: Hysterical Epidemics and Modern Media (New
York: Columbia University Press, 1997), p. 118.
23. Showalter, Hystories, pp. 122–23.
24. Showalter, Hystories, p. 126.
25. Watters, Crazy Like Us, e.g. p. 48: “As the public became aware that anorexia
was in the symptom pool, local clinicians . . . noticed an increase in cases of
eating disorders.”
26. Nortin Hadler, “Fibromyalgia, Chronic Fatigue, and Other Iatrogenic Diagnos-
tic Algorithms,” Postgraduate Medicine 102 (1997): 162.
27. Simon Wessely, “Old Wine in New Bottles: Neurasthenia and ‘ME,’” Psychologi-
cal Medicine 20 (1990): 36.
28. Wessely, “Old Wine in New Bottles.”
29. Arthur Shapiro and Elaine Shapiro, The Powerful Placebo: From Ancient
Priest to Modern Physician (Baltimore: Johns Hopkins University Press, 1997),
p. 77.
30. On syndromes that challenge medicine, see R. S. Ferguson, “The Doctor-
Patient Relationship and ‘Functional’ Illness,” in Patients, Physicians and Ill-
ness, ed. E. Gartly Jaco (Glencoe, IL: Free Press, 1958), p. 435: patients with
psychosomatic and functional illness “try us at our weakest point . . . Reliable
observers estimate that of all attendees at surgeries and clinics, anything from
25 to 50 percent are not suffering from any demonstrable pathological entity or
nosological syndrome.”
Epilogue
20. “In my opinion the most ordinary things, the most commonplace and best-
known, can constitute, if we know how to present them in the right light, the
greatest of Nature’s miracles and the most amazing of examples, notably on the
subject of human actions.” Montaigne, “On Experience” in Complete Essays, tr.
M. A. Screech (London: Penguin, 2003), pp. 1227–28.
21. Jerome Frank, “Psychotherapy as Rhetoric: Some Implications,” Clinical Psy-
chology: Science and Practice 2 (1995): 90.
22. Preface to the New York edition of Roderick Hudson (London: Penguin, 1986),
p. 37.
23. Kate Faasse and Keith Petrie, “The Nocebo Effect: Patient Expectations
and Medication Side Effects,” Postgraduate Medical Journal, doi:10.1136/
postgradmedj-2012-131730.
24. As noted in Chapter 1, Montaigne’s meditation “On the Power of the Imagina-
tion” investigates the imagination’s rule over mind and body while throwing
doubt on accounts of miracle cures. Montaigne tells of “a woman who believed
she had swallowed a pin in her bread” and was cured when a wise trickster
“made her vomit and secretly tossed a bent pin into what she had brought up”
(p. 117). The subplot of The Brothers Karamazov turns on the story of a dog that
snapped up a piece of bread with a pin stuck in it. The dog doesn’t die, but the
child who played the trick does. We are told that the sight of the dog as it ran
away squealing “haunted” the boy, that is, entered his imagination.
25. E.g., Franklin Miller and Ted Kaptchuk, “The Power of Context: Reconceptual-
izing the Placebo Effect,” Journal of the Royal Society of Medicine 101 (2008):
222–25.
Index
Hadler, Nortin, 132, 191 mammography, 42–43, 67, 78, 150, 162
Hamilton, Max, 9 medicalization, viii-ix, 12–14, 16, 24,
Hamilton scale, 9, 156 30–33, 43, 45–48, 50–51, 56,
harm, ix, 3, 12–17, 19, 22, 26–27, 46, 65–66, 70, 77, 79, 92, 128, 134,
49–56, 58, 60, 65, 87, 93, 96, 107, 145, 163, 191, 213
INDEX 271