FIG1 A remarkable quote recently
appeared in the Boston Globe regarding a study of breast masses that
had been published in the New England Journal of Medicine. The study
found, in effect, that the prognosis for patients with noncancerous breast
masses depends on the particular cytology of the mass—the more clearly"
abnormal" or atypical, the worse the prognosis.
This study prompted an astonishingly candid response from a renowned breast
cancer surgeon, as quoted in the Globe: "Dr. Susan Love, a Los
Angeles breast cancer surgeon, author, and researcher, said a big problem is
that a pathologist's judgment that cells look `abnormal' is extremely
subjective. `It's in the eye of the beholder,' she said."
This quote appeared at roughly the same time that actor Tom Cruise
infamously insisted that psychiatry is a "pseudoscience" and not
long after Rep. Ron Paul of Texas—an obstetrician by
training—cited a pediatrician to the effect that "[p]sychiatric
diagnoses are inherently subjective and based on `social
constructions.'" APA President Steven Sharfstein, M.D., has taken Mr.
Cruise and Rep. Paul to task for espousing these views.
It should not surprise us that those who attack psychiatry for its supposed"
subjectivity" do not understand much about our field. What is
more surprising is that these individuals also misunderstand general medicine,
which has always struggled with finding the right mix of"
objectivity" and "subjectivity." And as Dr. Love's
comment suggests, even the pathologist's diagnosis is heavily influenced by
personal judgment and clinical experience.
Critics of psychiatry seem to believe that objectivity resides solely in
the detection of abnormal lumps or bumps—or in some infallible blood
test. These critics endlessly lambaste psychiatry for making diagnoses on the
basis of subjective criteria. But what do philosophers of science understand
the term "objective" to mean? And how does psychiatry compare with
other medical specialties on accepted measures of objectivity?
The philosopher Amartya Sen described two essential features of
objectivity: observation dependence and impersonality. "Objectivity
demands taking observations seriously," Sen argued. Objectivity also
requires that there be "some invariance" with respect to the
person carrying out the observation. By "invariance," Sen meant
that the observer's conclusions should be more or less reproducible by other
observers, within the natural limits of human perception. To put it in the
more technical language of the philosopher Thomas Nagel (in his book The
View From Nowhere), "The wider the range of subjective types to
which a form of understanding is accessible.. .the more objective it
is." In short, if one observer sees a quacking, waddling bird and says,"
It's a duck," that's a relatively subjective statement. If a
hundred observers can agree it's a duck, we have a more objective basis for
hypothesizing that "it's a duck."
Psychiatry more than meets the two tests proposed by Prof. Sen. First,
psychiatrists take observation very seriously—indeed, it is the art and
science we live or die by in our profession.
From the moment the patient walks into our office, we are compiling a
staggering array of empirical observations: The patient appears disheveled, he
walks with a slightly ataxic gait, he appears agitated and confused, there is
an odor of alcohol on his breath, his thought processes are difficult to
follow, he is unable to subtract 7s serially from 100, he whispers to himself
and glances over his shoulder frequently, his speech is loud and pressured, he
picks constantly at his clothing, and so on. Depending on the case, many of us
will supplement our own observations with neuropsychological testing,
laboratory studies to rule out underlying medical disorders, and brain imaging
studies to detect tumors, strokes, and dementia.
So far, so good—but do psychiatric observations meet Sen's second
test—what scientists would term "interrater
There is a huge literature addressing this issue, but a comparison of two
recent studies is instructive. The first one (by van Jaarsveld et al. in the
December 1999 Journal of Hypertension) examined the degree to which
three "experienced radiologists" could agree on the interpretation
of 312 renal angiograms; for example, whether and where renal artery stenosis
was present. The second study (by Majet et al. in the January-March 2000
Journal of Affective Disorders) assessed the degree to which two
psychiatrists could agree on whether 150 patients met DSM-IV criteria
for three conditions: schizoaffective disorder, mania, and major
Suffice to say that for two of the three psychiatric diagnoses—mania
and major depression—interrater reliability was better between the
psychiatrists than it was among the radiologists in the first study.
All of clinical medicine involves a subtle interplay between reason and
intuition, objectivity and subjectivity, strict diagnostic criteria and
clinical wisdom. Yes, psychiatry has much work to do in refining its methods
of diagnosis, but this is a challenge facing the entire medical field. In the
meantime, psychiatry's clinical "eye" remains pretty sharp.▪