rated into the definition of atypical

depression resembles a personality

trait rather than a mood.
The tendency to anticipate failure
and humiliation and give up easily
when frustrated is a characteristic of
avoidant personality (passive, timid,
submissive, easily hurt). Borderline
personalities are emotionally unstable,
self-destructively impulsive, chronically bored or angry, with constantly
shifting moods. Histrionic personalities are flamboyant, self-dramatizing,
self-centered, and emotionally shallow. These personality descriptions
have much in common with the rejection sensitivity and mood reactivity of
atypical depression.
Bipolar depression, like atypical
depression, is characterized by mood
reactivity and often by the atypical
physical symptoms of oversleeping
and overeating. Especially in their less
intense variants, bipolar symptoms
can be difficult to distinguish from
depression with atypical features. Some
research suggests that a family history
of bipolar disorder is more likely in
depressed people with the atypical
symptoms of leaden paralysis and
oversleeping.
Some believe atypical depression
is related to a whole group of disorders that involve emotional insta-

bility resulting from defective mood

regulation. This category extends
beyond mood disturbances in the narrow sense. It may include not only
borderline personality and avoidant
personality but panic disorder, obsessive-compulsive disorder, and bulimia.
Many of these disorders are now routinely treated with antidepressants and
mood stabilizers.
Today the diagnosis of depression
with atypical features does not dictate
any particular treatment. MAO inhibitors are rarely a first choice for depression because of their many side effects,
which include drowsiness, dizziness,
insomnia, and a sudden, potentially
dangerous rise in blood pressure when

combined with certain foods (pickles, cheese, red wine). The preferred
antidepressants are selective serotonin
reuptake inhibitors like fluoxetine
(Prozac). When depression with atypical features turns out to be a form of
bipolar disorder, lithium or anticonvulsants can be prescribed. Patients
who also have panic disorder, phobias, or obsessive-compulsive rituals
may need both behavioral treatment
and antidepressants. And long-term
psychotherapy (along with antidepressants) may be especially useful
for patients with atypical depression
because of their changeable moods
and symptoms that resemble a personality disorder.

References
Agosti V, et al. Atypical and NonAtypical Subtypes of Depression:
Comparison of Social Functioning,
Symptoms, Course of Illness,
Comorbidity and Demographic
Features, Journal of Affective
Disorders (January 2001): Vol. 65, No 1,
pp. 7579.

Depression: A Prospective Study,

American Journal of Psychiatry
(February 2004): Vol. 161, No. 2,
pp. 25561.
Posternak MA. Biological Markers of
Atypical Depression, Harvard Review
of Psychiatry (JanuaryFebruary 2003):
Vol. 11, No. 1, pp. 17.

Matza LS, et al. Depression with

Atypical Features in the National
Comorbidity Survey, Archives of
General Psychiatry (August 2003): Vol.
60, No. 8, pp. 81726.

Quitkin FM, et al. Atypical

Depression: Current Status, Current
Opinion in Psychiatry (2004): Vol. 17,
pp. 3741.

Oquendo MA, et al. Instability

of Symptoms in Recurrent Major

For more references, please see

www.health.harvard.edu/mentalextra.

Dropping out of psychotherapy

ts a big problem. Surveys show

that nearly half of people who
begin psychotherapyindividual,
group, or couplesquit, dissatisfied,
against the therapists recommendation. An article in the Harvard Medical
School Psychiatry Departments journal, the Harvard Review of Psychiatry,
discusses why this happens so often
and suggests some ways to prevent it.
The authors note some reasons why
patients drop out: They are unwilling
to open up about themselves; they

september 2005

cannot agree with the therapist about

what the problem is; they just dont
get along with or feel confidence in
the therapist; they believe they are not
improving quickly enough; they have
unrealistic expectations.
The result, often, is that the patient
feels like a failure. His or her problems
are likely to get worse, and the symptoms are more likely to become chronic. When a patient drops out of group
therapy, other group members may
feel abandoned and group cohesion

www.health.harvard.edu

may be damaged. Psychotherapists

may be demoralized because they feel
rejected, and this feeling may interfere
with the treatment of other patients.
What can be done about it? To
find answers, the authors reviewed 35
years of scientific literature. They base
their recommendations on the several
dozen research studies and clinical
descriptions they found.
Patient selection. Before starting
therapy, it may help to screen patients

HARVARD MENTAL HEALTH LETTER

for a good match to the therapist

and the therapy. Plenty of attention
has been devoted to this subject but
not much controlled research, say the
authors. Theres some evidence that
screening questionnaires for psychodynamic and cognitive behavioral
therapy can help to distinguish patients
who will complete therapy from those
who wont. With reliable screening,
patients at high risk for dropping out
might be offered a different treatment,
or specific preparation for treatment.
Preparation. Before beginning psycho
therapy, some patients need to be educated about the process. They can be
given an explanation of the rationale,
the roles and obligations of patient
and therapist, expected difficulties and
realistic hopes. This can be done with
audiotaped or videotaped instructions,
or by simulated therapy sessions (or
videotaped excerpts from actual therapy sessions).
Group therapy, the authors say, re
quires more preparation because of the
greater threat to control, privacy, and
emotional safety. One approach is experiential pretrainingattending actual
group therapy sessions to learn what it
is like.
All these approaches have helped to
improve attendance and lower dropout rates in some controlled studies,
although results are not consistent.

and ends of therapywhat this person needs to accomplish and how it

is to be accomplished. Negotiation is
especially important in group therapy
because otherwise, patients referred
to groups may think that the unique
features of their own situation are
being ignored.
When the treatment is brief or has
a fixed end point, dropout rates
tend to be lowerin some studies,
as much as 50% lower.

formed quickly; some believe that if it

does not develop within the first three
sessions, it never will.
Appointment reminders. Remind
ing patients of their appointments is
routine for many health care professionals but sometimes avoided by psychotherapists because they want to
promote responsibility in patients, or
because they believe its better to
explore the meanings behind cancellations. The authors suggest that encouraging consistent attendance is more
important.

Facilitating expression of feelings.

The therapist must create an atmosphere in which a patient can safely
discuss uncomfortable feelings, doubts,
and questions about the therapy and
the therapist. Otherwise, the patient
may become uneasy and abandon the
therapeutic project.
The authors note that there is far
too little research on this subjectonly 15 studies since 1970, and only 4
since 1985possibly because many
psychotherapists take a casual attitude
toward the problem. They point out
Motivational enhancement. Some that no single strategy will work for all
times the problem is that the patient is patients and in all situations, and they
not yet sufficiently willing or ready to recommend that clinicians try several
change. Motivational enhancement approaches. But only more research
aims to promote confidence in the will make it possible to compare ways
ability to change and create a climate of preventing dropouts and to suggest
in which commitment to change more specific recommendations. They
Short-term or time-limited therapy. becomes possible. It is already com- invite others to contribute ideas.
When the treatment is brief or has a mon in the treatment of alcoholism,
fixed end point, dropout rates tend to drug addiction, and eating disorders.
References
be lowerin some studies, as much (See Harvard Mental Health Letter,
Ogrodniczuk JS, et al. Strategies
as 50% lower. As the authors note, that March 2005).
for Reducing Patient-Initiated
is partly because the less time a patient
Premature Termination of Psycho
spends in therapy, the less opportunity
Establishing the therapeutic allitherapy, Harvard Review of Psych
iatry (MarchApril 2005): Vol. 13,
there is for premature termination. But ance. Many studies have shown that
No. 2, pp. 5770.
knowing when it will end may provide the critical feature of all successful
Harvard Mental Health Letter
a sense of urgency and purpose that psychotherapy is a strong working
subscribers can obtain a special
prevents patients from becoming dis- relationship between the patient and
discounted subscription to the
couraged.
the therapist. There is no formula for
Harvard Review of Psychiatry
at www.tandf.co.uk/journals/
achieving it, although warmth, empatitles/10673229.asp or by calling
Negotiation. Therapist and patient thy, respect, and interest are always
800-354-1420, ext 216.
should agree in advance on the means important. The alliance should be

HARVARD MENTAL HEALTH LETTER

Case management. This is sometimes necessary to solve problems that

make psychotherapy difficult, such as
lack of adequate housing or employment or a disastrous family situation.
Case management today is used mainly for people suffering from severe
mental illness, especially those with
low income and little education. One
study found that case management for
severely depressed patients in group
therapy reduced the rate of quitting
by 50%.

www.health.harvard.edu

september 2005