Copyright Blackwell Munksgaard 2008

Bipolar Disorders 2008: 10: 144152

BIPOLAR DISORDERS

Review Article

Diagnostic guidelines for bipolar depression: a

probabilistic approach
Mitchell PB, Goodwin GM, Johnson GF, Hirschfeld RMA. Diagnostic
guidelines for bipolar depression: a probabilistic approach.
Bipolar Disord 2008: 10: 144152. Blackwell Munksgaard, 2008

Philip B Mitchella, Guy M Goodwinb,

Gordon F Johnsonc and Robert MA
Hirschfeldd
a

Objectives: There are currently no accepted diagnostic criteria for bipolar

depression for either research or clinical purposes. This paper aimed to
develop recommendations for diagnostic criteria for bipolar I depression.

Methods: Studies on the clinical characteristics of bipolar and unipolar

depression were reviewed. To identify relevant papers, literature searches
using PubMed and Medline were undertaken.
Results: There are no pathognomonic characteristics of bipolar I
depression compared to unipolar depressive disorder. There are, however,
replicated ndings of clinical characteristics that are more common in both
bipolar I depression and unipolar depressive disorder, respectively, or which
are observed in unipolar-depressed patients who convert (i.e., who later
develop hypo manic symptoms) to bipolar disorder over time. The following
features are more common in bipolar I depression (or in unipolar converters
to bipolar disorder): atypical depressive features such as hypersomnia,
hyperphagia, and leaden paralysis; psychomotor retardation; psychotic
features, and or pathological guilt; and lability of mood. Furthermore,
bipolar-depressed patients are more likely to have an earlier age of onset of
their rst depressive episode, to have more prior episodes of depression, to
have shorter depressive episodes, and to have a family history of bipolar
disorder. The following features are more common in unipolar depressive
disorder: initial insomnia reduced sleep; appetite, and or weight loss;
normal or increased activity levels; somatic complaints; later age of onset
of rst depressive episode; prolonged episodes; and no family history of
bipolar disorder.
Conclusions: Rather than proposing a categorical diagnostic distinction
between bipolar depression and major depressive disorder, we would
recommend a probabilistic (or likelihood) approach. While there is no
point of rarity between the two presentations, there is, rather, a dierential
likelihood of experiencing the above symptoms and signs of depression.
A table outlining draft proposed operationalized criteria for such an
approach is provided. The specic details of such a probabilistic approach
need to be further explored. For example, to be useful, any diagnostic
innovation should inform treatment choices.

School of Psychiatry, University of New South

Wales, Sydney, Australia, bDepartment of
Psychiatry, Oxford University, Oxford, UK,
c
Discipline of Psychological Medicine, University of
Sydney, Sydney, Australia, dDepartment of
Psychiatry and Behavioral Sciences, University of
Texas Medical Branch, Galveston, TX, USA

Key words: bipolar depression bipolar disorder

diagnostic nosology
Received 23 January 2007, revised and accepted
for publication 3 September 2007
Corresponding author: Professor Philip Mitchell,
School of Psychiatry, University of New South
Wales, Prince of Wales Hospital, Randwick, NSW
2031, Australia. Fax: +61 2 93828151;
e-mail: phil.mitchell@unsw.edu.au

PBM has received remuneration for lectures or advisory board membership from AstraZeneca, Eli Lilly & Co., GlaxoSmithKline, Janssen-Cilag, or
Lundbeck in the last 5 years. GMG (as of September 12, 2007) has an interest in relation to one or more organizations that could be perceived as a
possible conict of interest in the context of the subject of this work. The relationship(s) include: grants: Sano-Aventis, Servier; honoraria: AstraZeneca,
Bristol-Myers Squibb, Eisai, Lundbeck, Sano-Aventis, Servier; paid positions: University of Oxford, Oxfordshire & Buckinghamshire NHS Mental
Healthcare Trust; advisory boards: AstraZeneca, Bristol-Myers Squibb, Eli Lilly & Co., Lundbeck, P1Vital, Sano-Aventis, Servier, and Wyeth. GFJ
has been (or is)* an advisory board member, clinical trial investigator, or consultant to AstraZeneca, Bristol-Myers Squibb, Eli Lilly & Co.,
GlaxoSmithKline, Novartis, Pzer, Janssen-Cilag, Sano Aventis, Servier*, Organon Roche, Wyeth Pharmaceuticals, the Commonwealth Government
Pharmaceutical Benets Advisory Committee, and the Australian Drug Evaluation Committee*. RMAH has served as a consultant advisory board
member for Abbott Laboratories, AstraZeneca, Bristol-Myers Squibb, Cephalon, Forest Laboratories, GlaxoSmithKline, Janssen Pharmaceutica, Jazz
Pharmaceutical, KV Pharmaceutical, Ortho McNeil, Eli Lilly & Co., Novartis, Organon, Pzer, Solvay Pharmaceuticals, and Wyeth-Ayerst.

144

Diagnostic guidelines for bipolar depression

While pathologically elevated mood is the hallmark characteristic of bipolar disorder, syndromal
or subsyndromal depression is now recognized to
comprise the predominant cumulative aect over
time, and to make a major contribution to the
disability related to this condition (18).
There are currently no accepted diagnostic
criteria for bipolar depression for either research
or clinical purposes. This paper aimed to develop
recommendations for diagnostic criteria for bipolar I depression. We examine the literature on
bipolar depression particularly focusing on its
clinical aspects then propose guidelines for its
diagnosis.
Depressive episodes in the course of bipolar disorder

Bipolar disorder presents initially with a depressive

episode in at least 50% of patients, with most
studies reporting that a majority of subjects rst
present in this phase of the illness (9). The age of
onset of the rst depressive episode in bipolar
disorder has been found in many studies to be
earlier than that found in unipolar depression (10,
11).
Patients with bipolar disorder tend to experience
more depressive episodes over a lifetime than those
with unipolar depression (1012). There do not,
however, appear to be any signicant dierences in
the severity of depression between bipolar- and
unipolar-depressed subjects (13, 14).
Phenomenology of bipolar depression

There have now been a relatively large number of

studies comparing the clinical characteristics of
bipolar and unipolar depression. Table 1 details
the clinical dierences which have been reported in
these comparisons of bipolar and unipolar depressive subjects.
Most studies have described samples that would
now be recognized as fullling criteria for bipolar I
disorder. These studies have been the focus of a
series of recent review articles (1519). In one of
these reviews, Cuellar et al. (16) undertook a
detailed methodological critique, noting that many
of the primary data investigations have not controlled for potentially critical confounding variables such as medications or demographic factors
such as age or gender. Most studies have not
corrected for multiple testing. As the older studies
have been discussed in detail in previous reviews,
we will rst overview some of the more recent
studies, before focusing on potential distinguishing
characteristics that may assist in the diagnosis of
bipolar I depression.

Recent studies of the phenomenology of bipolar

depression

We review here studies of bipolar I (or predominantly bipolar I) data sets. In 2000, Parker et al.
(20) examined the relative prevalence of melancholic syndromes in bipolar and unipolar depression. This paper used three large depressive data
sets recruited over a 15-year period. These data sets
comprised 83 bipolar disorder and 904 unipolar
patients assessed in detail whilst depressed. Melancholia was dened using several diagnostic
systems [DSM-III, III-R, or DSM-IV; CORE-I
or II (21, 22); and operationalized clinical diagnoses]. Parker et al. found that bipolar disorder
subjects were signicantly more likely than unipolar patients to be diagnosed as melancholic by each
of the DSM (69% versus 37%), CORE (59%
versus 33%), and clinical (70% versus 29%)
systems. Furthermore, after logistic regression,
bipolar-depressed patients were signicantly more
likely to demonstrate psychomotor disturbance
(particularly retardation) and pathological guilt.
In a second paper from the same group, Mitchell
et al. (23) undertook a ne grained analysis of the
more recent (third) of these data sets. The aim of
this study was to identify particular symptomatic
dierences between bipolar- and unipolar-depressed patients. The sample comprised 270 inpatients and outpatients with a current DSM-IV
major depressive episode (MDE). Thirty-nine of
the patients had DSM-IV bipolar I disorder. These
bipolar-depressed patients were initially compared
with 39 unipolar-depressed subjects matched for
age, sex and the presence or absence of DSM-IV
melancholia. Although there was no dierence in
severity of depression (assessed by the Hamilton
Rating Scale for Depression), those with bipolar
depression were signicantly more likely to have
had a past psychotic depressive episode. With
regard to current symptomatology, they were
more likely to report worthlessness, anhedonia,
restlessness, leaden paralysis, and hypersomnia.
Conversely, they were less likely to manifest
tearfulness, anxiety, and the tendency to blame
others. The most striking dierences, however,
were observed in the objective features of depressive psychomotor disturbance as rated by the
CORE scale (20). (This is an 18-item scale with
subscales for retardation, agitation, and noninteractiveness; the latter being a proxy for
cognitive dysfunction.) Bipolar patients were less
emotionally reactive, had a greater delay in
responding verbally, and demonstrated more slowed
movement. There were no dierences in agitation
items. Bipolar patients scored higher on both the

145

Mitchell et al.
Table 1. Studies of illness characteristics of bipolar (BP) and unipolar (UP) depression
Variable
Course of illness
Age of onset
Duration of episode

Number of prior depressive episodes

Symptomatology
Severity of depression

Unvarying depression
Worthlessness
Low self-esteem
Pessimistic thoughts
Anticipatory anhedonia
Irritability
Subjective restlessness
Social withdrawal
Suicidal

Tearfulness
Initial insomnia

Reduced sleep
Increased sleep
Weight loss
Weight gain
Appetite loss
Appetite disturbance
Increased eating
Atypical features

Muddled thoughts thinking difficulties

Difficulty concentrating
Distractibility
Depression worse in morning
Mania score
Lability of mood
Racing thoughts
More talkative
Increased risky and goal-directed behaviors
Anxiety

Comorbid panic disorder or GAD

Derealization

146

Study

Finding

Andreasen et al. (10)

Solomon et al. (11)
Abrams and Taylor (46)
Coryell et al. (47)
Mitchell et al. (23)
Andreasen et al. (10)
Solomon et al. (11)

UP
UP
UP
UP
UP
UP
UP

Katz et al. (48)

Mitchell et al. (23)
Goel et al. (26; SAD sample)
Perlis et al. (25)
Brockington et al. (49)
Mitchell et al. (23)
Mitchell et al. (23)
Benazzi (50)
Olfson et al. (24; primary care sample)
Coryell et al. (33)
Perlis et al. (25)
Mitchell et al. (23)
Benazzi (50)
Mitchell et al. (23)
Goel et al. (26; SAD sample)
Coryell et al. (33)
Brockington et al. (49)
Olfson et al. (24; primary care sample)
Perlis et al. (25)
Mitchell et al. (23)
Brockington et al. (49)
Mitchell et al. (23)
Papadimitriou et al. (27)
Perlis et al. (25)
Andreasen et al. (10)
Benazzi (50)
Abrams and Taylor (46)
Benazzi (50)
Papadimitriou et al. (27)
Olfson et al. (24; primary care sample)
Andreasen et al. (10)
Benazzi (50)
Mitchell et al. (23; leaden paralysis; hypersomnia)
Serretti et al. (28)
Benazzi (51)
Brockington et al. (49)
Coryell et al. (33)
Benazzi (51)
Perlis et al. (25)
Benazzi (50)
Brockington et al. (49)
Parker et al. (52)
Abrams and Taylor (46)
Brockington et al. (49)
Akiskal et al. (34)
Benazzi (50)
Benazzi (50)
Benazzi (50)
Beigel and Murphy (53)
Coryell et al. (33)
Mitchell et al. (23)
Perlis et al. (25)
Simon et al. (54)
Brockington et al. (49)

UP > BPI
n.s.
BP > UP
UP > BPI
UP > BPI
BPI > UP
BPI > UP
BPII > UP
BP > UP
UPs who converted to
BPI > UP
BPI > UP
BPII > UP
BPI > UP
BP > UP
UPs who converted to
UP > BPI
BP > UP
BPI > UP
UP > BPI
UP > BPI
UP > BPI
n.s.
UP > BPI
BP > UP
BPII > UP
UP > BPI
BPII > UP
UP > BPI
n.s.
BP > UP
BPII > UP
BPI > UP
BPI > BPII > UP
BPII > UP
UP > BP
UPs who converted to
BPII > UP
UP > BPI
BPII > UP
BPI > UP
n.s.
BPI > UP
BPI > UP
UPs who converted to
BPII > UP
BPII > UP
BPII > UP
UP > BPI
UPs who converted to
UP > BPI
Inconsistent findings
BP > UP
BPI > UP

>
>
>
>
>
>
>

BP
BP
BPI
BPI
BPI
BP
BP

BPI

BPI

BPI

BPII

BPI

Diagnostic guidelines for bipolar depression

Table 1. Continued
Variable
Current alcohol use disorder
Somatic complaints

Something wrong with body

Pathologic guilt
Tendency to blame others
Psychotic features

Ideas of reference
Auditory or visual hallucinations
Auditory hallucinations
Loss of insight
Mental state signs
Activity
Agitation

Observed anxiety
Retardation

Anger aggression
Haughty attitude
Family history of bipolar disorder

Study

Finding

Olfson et al. (24; primary care sample)

Beigel and Murphy (53)
Katz et al. (48)
Abrams and Taylor (55)
Dunner et al. (56)
Perlis et al. (25)
Brockington et al. (49)
Parker et al. (20)
Mitchell et al. (23)
Guze et al. (57)
Andreasen et al. (10)
Coryell et al. (33)
Strober and Carlsson (58)
Akiskal et al. (32)
Goldberg et al. (59)
Geller et al. (60)
Endicott et al. (61)
Beigel and Murphy (53)
Black and Nasrallah (62)
Mitchell et al. (23)
Brockington et al. (49)
Olfson et al. (24; primary care sample)
Brockington et al. (49)
Brockington et al. (49)

BP > UP
UP > BPI
UP > BPI
n.s.
n.s.
UP > BPI
BPI > UP
BPI > UP
UP > BPI
BPI > UP
BP > UP
UPs who converted
UPs who converted
UPs who converted
UPs who converted
UPs who converted
BPI > UP
n.s.
n.s.
BPI > UP
UP > BP
BP > UP
UP > BP
UP > BP

Beigel and Murphy (53)

Kupfer et al. (63)
Katz et al. (48)
Abrams and Taylor (46)a
Abrams and Taylor (55)
Mitchell et al. (23)
Goel et al. (26; SAD sample)
Benazzi (50)
Brockington et al. (49)
Dunner et al. (56)
Andreasen et al. (10)
Perris (64)
Katz et al. (48)
Popescu et al. (65)
Kuhs and Reschke (66)
Mitchell et al. (67)
Parker et al. (20)
Mitchell et al. (23)
Goel et al. (26; SAD sample)
Benazzi (29)
Beigel and Murphy (53)
Abrams and Taylor (46)a
Brockington et al. (49)
Dunner et al. (56)
Andreasen et al. (10)
Solomon et al. (11)
Coryell et al. (33)
Strober and Carlson (58)
Akiskal et al. (32)

UP > BPI
UP > BPI
UP > BPI
BPI > UP
n.s.
n.s.
BP > UP
BPII > UP
UP > BPI
BPI > UP
BP > UP
n.s.
n.s.
n.s.
n.s.
n.s.
BPI > UP
BPI > UP
BPI > BPII
n.s. (BP II v UP)
UP > BPI
BPI > UP
BPI > UP
n.s.
BP > UP
BP > UP
UPs who converted to BPI
UPs who converted to BP
UPs who converted to BP

to
to
to
to
to

BPI
BP
BP
BP
BP

GAD generalized anxiety disorder; SAD seasonal affective disorder.

a
Correlation with bipolar disorder.
Updated from Mitchell et al. (23, 67).

retardation subscale and total CORE score.

Mitchell et al. then compared the 20 bipolardepressed patients with DSM-IV-dened melan-

cholia against age- and sex-matched melancholic

unipolar patients. First, with respect to symptoms,
bipolar disorder patients reported more anhedonia,

147

Mitchell et al.

persistence of depressed mood, and hypersomnia;

but less initial insomnia and anxiety. Secondly,
bipolar disorder patients were rated on the CORE
scale as demonstrating more facial immobility,
slower movement, and a greater delay in initiating
motor activity. Again, they had signicantly higher
scores on the retardation subscale of this instrument.
These ndings suggested features with potential
clinical utility for diagnosing bipolar depression
which were described by the authors as a bipolar
depression signature. These were: signs of psychomotor retardation; melancholic symptoms such as
worthlessness, unvarying mood, and marked anhedonia; atypical depressive symptoms such as
hypersomnia and leaden paralysis; a past history
of psychotic depression; and the absence of anxiety, initial insomnia, the tendency to blame others,
and tearfulness. Mitchell et al. acknowledged that
these features reected diering group means and
were not pathognomonic of bipolar depression.
The studies of Mitchell and Parker suggest that
bipolar I depression is characterized by an admixture of melancholic, atypical, and (less commonly)
psychotic features. This pattern suggests that the
clinical presentation of bipolar I depression is
distinct from the pure atypical depression of, for
example, seasonal aective disorder or the pure
melancholic presentation of some patients with
severe unipolar depression.
Olfson et al. (24) studied characteristics of
depression in a low-income primary care clinic,
with bipolar patients identied using the Mood
Disorder Questionnaire (MDQ). All assessments
were performed using a variety of patient selfreport measures. Twenty-ve percent of depressed
patients were assessed as suering from bipolar
disorder according to the MDQ. Those with
bipolar depression were signicantly more likely
than those with unipolar depression to report
hallucinations, current suicidal ideation, and low
self-esteem, but were less likely to acknowledge
disturbed appetite (decreased or increased).
Perlis et al. (25) used three large data sets (two
unipolar and one bipolar depressed) that had been
recruited for pharmaceutical studies. The paper
reported on items from two measures common to
these studies, i.e., the Montgomery-Asberg Depression Rating Scale and the Hamilton Anxiety
Rating Scale. The strength of this report is the
large numbers involved (n = 477 in bipolar
depression trial; n = 1,074 for the major depressive disorder trials). The limitations were that the
original studies were distinct, with diering raters
and sites, and lacked other measures of depressive
phenomenology. Perlis et al. found bipolar depression to be associated with a family history of

148

bipolar disorder, earlier age of onset, and more

prior depressive episodes. With regard to specic
items from the rating scales, bipolar patients
reported more fears, whereas unipolar patients
more commonly experienced sadness, insomnia,
cognitive diculties, and somatic symptoms. A
logistic regression including these various symptoms correctly classied 87% of subjects.
Solomon et al. (11) recently used data from the
NIMH Collaborative Program on the Psychobiology of Depression cohort to develop a clinical
screening instrument to screen for bipolar disorder
in patients presenting with a current MDE. Using a
split-sample data analytic procedure, the authors
rst investigated the most discriminating of 59
socio-demographic and clinical variables in a
comparison of an index bipolar I sample compared
to major depressive disorder. They selected the
three most distinguishing variables, then tested
them in a second and separate bipolar I disorder
sample (the bipolar I disorder cross-validation
sample), and a bipolar II disorder cross-validation
sample. These three questions comprised what the
authors termed the Screening Instrument of
Depression Polarity (SAD-P). The three variables
were: (i) presence of delusions at baseline assessment; (ii) number of prior episodes of major
depression; and (iii) family history of major
depression or mania. The sensitivity of the instrument was 0.82 in the index bipolar I disorder
sample, 0.72 in the bipolar I disorder crossvalidation sample, and 0.58 in the bipolar II
disorder cross-validation sample. The positive
predictive value of the screening instrument was
low, however, with ndings of 0.36, 0.29 and 0.27,
respectively.
Reports from other recent studies include: more
psychomotor agitation and social withdrawal in
bipolar I depression than major depressive disorder
in a group with seasonal aective disorder (26); a
greater frequency of appetite loss in unipolar
depression (27); and more atypical but less melancholic symptoms in bipolar disorder when assessed
retrospectively using the operationalized criteria
checklist system (28).
Bipolar II depression

Benazzi has undertaken similar studies with bipolar II-depressed patients, predominantly in a private practice outpatient context. As in a number of
the studies of bipolar I-depressed patients described above, he has reported an increased prevalence of atypical features in bipolar II compared
with unipolar-depressed patients, but in contrast to
the bipolar I studies, has found no dierence in

Diagnostic guidelines for bipolar depression

rates of psychomotor retardation (51). In a more
recent paper (it is not clear if the sample overlaps
with prior reports from this researcher), Benazzi
has reported on a comparison of the symptoms of
depression in 379 bipolar II-depressed and 271
major depressive disorder outpatients (29). The
features that were more common in the bipolar IIdepressed patients were: weight gain, increased
eating, hypersomnia, psychomotor agitation,
worthlessness, and a diminished ability to concentrate. Inside-MDE hypomanic symptoms that
were more common in BP-II were distractibility,
racing thoughts, irritability, talkativeness, and
increased risky and goal-directed activities. Multiple logistic regression showed that hypersomnia,
racing thoughts, irritability, and psychomotor
agitation were independent predictors of BP-II
disorder. These reports by Benazzi suggest both
commonalities and distinctions between bipolar I
and II depression. They are limited in deriving
from only one researcher, without any independent
ratings. For more detailed discussion of bipolar II
depression, please refer to the paper by Vieta and
Suppes in this special issue (30).
Studies of clinical characteristics of converters
from unipolar depression to bipolar disorder

Another means of dening the characteristics of

bipolar depression is identifying apparent unipolar patients who on long-term follow-up convert
to bipolar I or II disorder. Angst and Preisig (31)
have reported that about 1% of patients with
unipolar depression initially identied while hospitalized convert from depression to bipolar I
disorder each year.
In 1983, Akiskal et al. (32) reported that the
predictors of a later emergence of bipolar disorder
in unipolar patients were: onset of depression prior
to 25, hypersomnia and motor retardation, a
family history of bipolar disorder, medicationprecipitated manic episodes, and postpartum
depression. The NIMH Collaborative Programme
on the Psychobiology of Depression 11-year prospective follow-up found that 4% converted to
bipolar I disorder and 9% to bipolar II disorder.
Those who switched to bipolar I disorder were
more likely to have been psychotic or hospitalized
at the index depressive assessment, compared with
those who continued to have a diagnosis of major
depressive disorder (33). Bipolar I converters also
rated worse on psychic anxiety, inability to concentrate, social withdrawal, and feelings of inadequacy and discouragement. Mood lability in the
depressive state was the most specic predictor of
switching to bipolar II disorder (34). In a follow-up

study of inpatient adolescents with major depression, Strober et al. (35) found that those who later
developed bipolar disorder were more likely to
have evidenced psychosis and psychomotor retardation at the index assessment.
Neuropsychological differences between bipolar
depression and unipolar depression

The neuropsychological proles of bipolar and

unipolar depression overlap, though there is some
evidence of greater impairment in bipolar disorder.
Both groups of patients have been found to be
predominantly impaired in the domains of memory
and executive functioning, with some evidence of
worse executive functioning in bipolar-depressed
subjects (36). Patients with bipolar depression have
impaired sustained attention and poor immediate
and delayed verbal recall which is greater than that
found in unipolar-depressed subjects and controls
(3739). Burt et al. (40) found bipolar depression
to be associated with worse memory as indicated
by the California Verbal Learning Test. The
pattern of executive function is broadly similar to
that found in unipolar depression, with bipolardepressed subjects performing poorly on tests of
concept formation, problem solving, and decision
making (4143). In unipolar depression both
semantic and phonemic decits contribute to a
reduction in verbal uency (44), whereas in bipolar
depression the decit appears to be more selective
with only semantic categories being aected (45).
Overall, as with the phenomenological studies
described in previous sections, there are no apparent pathognomonic neuropsychological characteristics of bipolar depression, but rather some
provisional evidence of dimensional dierences.
A proposed probabilistic approach to the
diagnosis of bipolar I disorder

It is apparent that there are no pathognomonic

characteristics of bipolar depression compared to
unipolar depression (major depressive disorder).
There are, however, replicated ndings of clinical
characteristics that are more common in both
bipolar depression and major depressive disorder,
respectively, or are observed in unipolar-depressed
patients who convert to bipolar disorder over
time.
Drawing on the various studies outlined in this
paper, those clinical features that have been most
commonly and consistently reported to be more
common in those with bipolar depression (or
unipolar-depressed converters to bipolar disorder)
are:

149

Mitchell et al.
Table 2. A proposed probabilistic approach to the diagnosis of bipolar I depression in a person experiencing a major depressive episode with no clear prior
episodes of mania

The greater likelihood of the diagnosis of BIPOLAR I DEPRESSION

should be considered if 5 of the following features are presenta
Symptomatology and mental state signs
Hypersomnia and or increased daytime napping
Hyperphagia and or increased weight
Other atypical depressive symptoms such as leaden paralysis
Psychomotor retardation
Psychotic features and or pathological guilt
Lability of mood manic symptoms
Course of illness
Early onset of first depression (< 25 years)a
Multiple prior episodes of depression ( 5 episodes)a
Family history
Positive family history of bipolar disorder

The greater likelihood of the diagnosis of

UNIPOLAR DEPRESSION should be considered
if 4 of the following features are presenta

Initial insomnia reduced sleep

Appetite and or weight loss
Normal or increased activity levels
Somatic complaints

Later onset of first depression (> 25 years)a

Long duration of current episode (> 6 months)a
Negative family history of bipolar disorder

See original study data in Table 1.

a
Confirmation of the specific numbers to be used requires further study and consideration.

Course of illness: earlier age of onset, shorter

duration of episodes, and more prior episodes
Symptomatology: worthlessness, low selfesteem, social withdrawal, hypersomnia,
hyperphagia weight gain, atypical features
(such as leaden paralysis), lability of mood,
and psychotic features
Mental state signs: psychomotor retardation lower activity levels
Family history: positive for bipolar disorder

Conversely, those clinical features that have been

most commonly and consistently reported to be
more common in those with unipolar depression
are:

Course of illness: later age of onset, longer

duration of episodes, and less prior episodes
Symptomatology: initial insomnia, appetite
loss weight loss, and somatic complaints
Mental state signs: higher activity levels
Family history: negative for bipolar disorder

Rather than proposing a categorical diagnostic

distinction between bipolar depression and major
depressive disorder, we would recommend a probabilistic (or likelihood) approach. While there is
no point of rarity between the two presentations,
there is, rather, a dierential likelihood of experiencing the above symptoms and signs of depression.
As some of these features (such as worthlessness,
low self-esteem, and social withdrawal) are common in depression, we have considered these to be
of low diagnostic specicity, and have therefore
not included these in the nal probabilistic
approach to the diagnosis of bipolar depression.

150

The draft operationalized criteria for such an

approach are detailed in Table 2. We have utilized
specic depressive clinical signs or symptoms (as
well as characteristics of illness course and family
history), rather than proposing a requirement for
particular syndromal presentations of depression
such as atypical or melancholic depression. This
table is premised on a current MDE in patients
with no clear prior episodes of hypo mania. We
specify particular number of features required to
indicate the greater likelihood of bipolar depression or unipolar depression, respectively. The
actual number of features specied have been
determined on the basis of both the literature
reviewed in this paper and clinical reasonability.
Conrmation of the specic numbers to be used
requires further study and consideration. We
acknowledge the limitation of this somewhat
arbitrary specication and would recommend that
the specic details of such a probabilistic approach
need to be further explored, for example in existing
or future prospective longitudinal data sets. Furthermore, to be useful, any diagnostic innovation
should inform treatment choices either with respect
to ecacy or adverse eects such as switch to
hypo mania. Thus, there is a pressing need to
understand treatment response to dierent classes
of treatment in relation to diagnostic dimensional
extremes and along the bipolar unipolar continuum.
Acknowledgement
Some of the material included in this paper has been adapted
from Mitchell and Malhi (15).

Diagnostic guidelines for bipolar depression

References
1. Judd LL, Akiskal HS, Schettler PJ et al. The long-term
natural history of the weekly symptomatic status of bipolar
I disorder. Arch Gen Psychiatry 2002; 59: 530537.
2. Judd LL, Akiskal HS, Schettler PJ et al. A prospective
investigation of the natural history of the long-term weekly
symptomatic status of bipolar II disorder. Arch Gen Psychiatry 2003; 60: 261269.
3. Kupka RW, Altshuler LL, Nolen WA et al. Three times
more days depressed than manic or hypomanic in both
bipolar I and bipolar II disorder. Bipolar Disord 2007; 9:
531535.
4. Leverich GS, Altshuler LL, Frye MA et al. Risk of switch
in mood polarity to hypomania or mania in patients with
bipolar depression during acute and continuation trials of
venlafaxine, sertraline, and bupropion as adjuncts to mood
stabilizers. Am J Psychiatry 2006; 163: 232239.
5. Bauer MS, Kirk GF, Gavin C, Williford WO. Determinants of functional outcome and healthcare costs in
bipolar disorder: a high-intensity follow-up study. J Affect
Disord 2001; 65: 231241.
6. Bryant-Comstock L, Stender M, Devercelli G. Health care
utilization and costs among privately insured patients with
bipolar I disorder. Bipolar Disord 2002; 4: 398405.
7. Altshuler LL, Gitlin MJ, Mintz J, Leight KL, Frye MA.
Subsyndromal depression is associated with functional
impairment inpatients with bipolar disorder. J Clin Psychiatry 2002; 63: 807811.
8. Judd LL, Akiskal HS, Schettler PJ et al. Psychosocial
disability in the course of bipolar I and II disorders: a
prospective, comparative, longitudinal study. Arch Gen
Psychiatry 2005; 62: 13221330.
9. Kinkelin M. Course and prognosis in manic-depressive
psychosis. Schweiz Arch Neurol Psychiatr 1954; 73: 100
146.
10. Andreasen NC, Grove WM, Endicott J et al. The phenomenology of depression. Psychiatr Psychobiol 1988; 3:
110.
11. Solomon DA, Leon AC, Maser JD et al. Distinguishing
bipolar major depression from unipolar major depression
with the screening assessment of depression-polarity
(SAD-P). J Clin Psychiatry 2006; 67: 434442.
12. Roy-Byrne PR, Post RM, Uhde TW et al. The longitudinal course of recurrent aective illness: life chart data from
research patients at the NIMH. Acta Psychiatr Scand 1985;
71: S1S32.
13. Ahearn EP, Carroll BJ. Short-term variability of mood
ratings in unipolar and bipolar depressed patients. J Affect
Disord 1996; 36: 107115.
14. Dorz S, Borgherini G, Conforti D, Scarso C, Magni G.
Depression in inpatients: bipolar vs. unipolar. Psychol Rep
2003; 92: 10311039.
15. Mitchell PB, Malhi GS. Bipolar depression: phenomenological overview and clinical characteristics. Bipolar Disord 2004; 6: 530539.
16. Cuellar AK, Johnson SL, Winters R. Distinctions between
bipolar and unipolar depression. Clin Psychol Rev 2005;
25: 307339.
17. Mansell W, Colom F, Scott J. The nature and treatment of
depression in bipolar disorder: a review and implications
for future psychological investigation. Clin Psychol Rev
2005; 25: 10761100.
18. Schweitzer I, Maguire K, Ng CH. Should bipolar disorder
be viewed as manic disorder? Implications for bipolar
depression. Bipolar Disord 2005; 7: 418423.
19. Bowden CL. A dierent depression: clinical distinctions
between bipolar and unipolar depression. J Affect Disord
2005; 84: 117125.

20. Parker G, Roy K, Wilhelm K, Mitchell P, Hadzi-Pavlovic

D. The nature of bipolar depression: implications for
the denition of melancholia. J Affect Disord 2000; 59:
217224.
21. Parker G, Hadzi-Pavlovic D, Boyce P et al. Classifying
depression by mental state signs. Br J Psychiatry 1990; 157:
5564.
22. Parker G, Hadzi-Pavlovic D, Wilhelm K et al. Dening
melancholia: properties of a rened sign-based measure. Br
J Psychiatry 1994; 164: 316326.
23. Mitchell PB, Wilhelm K, Parker G, Austin M-P,
Rutgers P, Malhi GS. The clinical features of bipolar
depression: a comparison with matched major depressive disorder patients. J Clin Psychiatry 2001; 62: 212
216.
24. Olfson M, Das AK, Gamero MJ et al. Bipolar depression
in a low-income primary care clinic. Am J Psychiatry 2005;
162: 21462151.
25. Perlis RH, Brown E, Baker RW, Nierenberg AA. Clinical
features of bipolar depression versus major depressive
disorder in large multicentre trials. Am J Psychiatry 2006;
163: 225231.
26. Goel N, Terman M, Terman JS. Depressive symptomatology dierentiates subgroups of patients with seasonal
aective disorder. Depress Anxiety 2002; 15: 3441.
27. Papadimitriou GN, Dikeos DG, Daskalopoulou EG,
Soldatos CR. Co-occurrence of disturbed sleep and appetite loss dierentiates between unipolar and bipolar
depressive episodes. Prog Neuropsychopharmacol Biol
Psychiatry 2002; 26: 10411045.
28. Serretti A, Mandelli L, Lattuada E, Cusin C, Smeraldi E.
Clinical and demographic features of mood disorder subtypes. Psychiatry Res 2002; 112: 195210.
29. Benazzi F. Psychomotor changes in melancholic and
atypical depression: unipolar and bipolar II subtypes.
Psychiatry Res 2002; 112: 211220.
30. Vieta E, Suppes T. Bipolar II disorder: arguments for and
against a distinct diagnostic entity. Bipolar Disord 2008;
10: 163178.
31. Angst J, Preisig M. Course of a clinical cohort of unipolar,
bipolar and schizoaective patients. Results of a prospective study from 1959 to 1985. Arch Neurol Psychiatr 1995;
146: 516.
32. Akiskal HS, Walker P, Puzantian VR et al. Bipolar outcome in the course of depressive illness: phenomenologic,
familial, and pharmacologic predictors. J Affect Disord
1983; 5: 115128.
33. Coryell W, Endicott J, Maser JD. Long term stability of
polarity distinctions in the aective disorders. Am J
Psychiatry 1995; 152: 385390.
34. Akiskal HS, Maser JD, Zeller PJ. Switching from unipolar to bipolar II: an 11-year prospective study of clinical
and temperamental predictors in 559 patients. Arch Gen
Psychiatry 1995; 52: 114123.
35. Strober M, Lampert C, Schmidt S, Morrell W. The course
of major depressive disorder in adolescents: 1. Recovery
and risk of manic switching in a follow-up of psychotic and
nonpsychotic subtypes. J Am Acad Child Adolesc Psychiatry 1993; 32: 3242.
36. Borkowska A, Rybakowski JK. Neuropsychological
frontal lobe tests indicate that bipolar depressed patients
are more impaired than unipolar. Bipolar Disord 2001; 3:
8894.
37. Brand N, Jolles J. Information processing in depression
and anxiety. Psychol Med 1987; 17: 145153.
38. Wolfe J, Granholm E, Butters N, Saunders E, Janowsky
D. Verbal memory decits associated with major aective
disorders: a comparison of unipolar and bipolar patients.
J Affect Disord 1987; 13: 8392.

151

Mitchell et al.
39. Ilsley JE, Mooot APR, OCarroll RE. An analysis of
memory dysfunction in major depression. J Affect Disord
1995; 35: 19.
40. Burt T, Prudic J, Peyser S, Clark J, Sackheim HA.
Learning and memory in bipolar and unipolar major
depression: eects of aging. Neuropsychiatry Neuropsychol Behav Neurol 2000; 13: 246253.
41. Sweeney JA, Kmiec JA, Kupfer DA. Neuropsychologic
impairments in bipolar and unipolar mood disorders on
the CANTAB neurocognitive battery. Biol Psychiatry
2000; 48: 674685.
42. Murphy FC, Rubinstein JS, Michael A, Rogers RD,
Robbins TW, Paykel ES. Decision-making cognition in
mania and depression. Psychol Med 2001; 31: 679693.
43. Murphy FC, Sahakian BJ. Neuropsychology of bipolar
disorder. Br J Psychiatry 2001; 178: 120127.
44. Porter RJ, Gallagher P, Thompson JM, Young AH.
Neurocognitive impairment in drug-free major depressive
disorder. Br J Psychiatry 2003; 182: 214220.
45. Calev A, Nigal D, Chazan C. Retrieval from semantic
memory using meaningful and meaningless constructs by
depressed, stable bipolar and manic patients. Br J Clin
Psychol 1989; 28: 6773.
46. Abrams R, Taylor MA. A comparison of unipolar and
bipolar depressive illness. Am J Psychiatry 1980; 137:
10841087.
47. Coryell W, Keller M, Endicott J, Andreasen N, Clayton P,
Hirschfeld R. Bipolar II illness: course and outcome over a
ve-year period. Psychol Med 1989; 19: 129141.
48. Katz MM, Robins E, Croughan J, Secunda S, Swann A.
Behavioural measurement and drug response characteristics of unipolar and bipolar depression. Psychol Med 1982;
12: 2536.
49. Brockington IF, Altman E, Hillier V, Meltzer HY, Nand
S. The clinical picture of bipolar aective disorder in its
depressed phase: a report from London and Chicago. Br J
Psychiatry 1982; 141: 558562.
50. Benazzi F. Symptoms of depression as possible markers of
bipolar II disorder. Prog Neuropsychopharmacol Biol
Psychiatry 2006; 30: 471477.
51. Benazzi F. Clinical dierences between bipolar II depression and unipolar major depressive disorder: lack of an
eect of age. J Affect Disord 2003; 75: 191195.
52. Parker G, Malhi G, Hadzi-Pavlovic D, Parker K. Sleeping
in? The impact of age and depressive subtype on hypersomnia J Affect Disord 2006; 90: 7376.

152

53. Beigel A, Murphy DL. Unipolar and bipolar aective illness: dierences in clinical characteristics accompanying
depression. Arch Gen Psychiatry 1971; 24: 215220.
54. Simon NM, Smoller JW, Fava M et al. Comparing anxiety
disorders and anxiety-related traits in bipolar disorder and
unipolar depression. J Psychiatr Res 2003; 37: 187192.
55. Abrams R, Taylor MA. Unipolar and bipolar depressive
illness. Arch Gen Psychiatry 1974; 30: 320321.
56. Dunner DL, Dwyer T, Fieve RR. Depressive symptoms in
patients with unipolar and bipolar aective disorder.
Compr Psychiatry 1976; 17: 447451.
57. Guze SB, Woodru RA, Clayton PJ. The signicance of
psychotic aective disorders. Arch Gen Psychiatry 1975;
32: 11471150.
58. Strober M, Carlson G. Bipolar illness in adolescents with
major depression: clinical, genetic, and psychopharmacologic predictors in a three-to four-year prospective followup investigation. Arch Gen Psychiatry 1982; 39: 549555.
59. Goldberg JF, Harrow M, Whiteside JE. Risk for bipolar
illness in patients initially hospitalized for unipolar
depression. Am J Psychiatry 2001; 158: 12651270.
60. Geller B, Fox LW, Clark KA. Rate and predictors of
prepubertal bipolarity during follow-up of 6- to 12-yearold depressed children. J Am Acad Child Adolesc Psychiatry 1994; 33: 461468.
61. Endicott J, Nee J, Andreasen N, Clayton P, Keller M,
Coryell W. Bipolar II: combine or keep separate? J Affect
Disord 1985; 8: 1728.
62. Black DW, Nasrallah A. Hallucinations and delusions in
1,715 patients with unipolar and bipolar aective disorders. Psychopathology 1989; 22: 2834.
63. Kupfer DJ, Weiss BL, Foster FG, Detre TP, Delgado J,
McPartland R. Psychomotor activity in aective states.
Arch Gen Psychiatry 1974; 30: 765768.
64. Perris C. A study of bipolar (manic-depressive) and unipolar recurrent depressive psychoses. Acta Psychiatr Scand
1966; 42: 1189.
65. Popescu C, Ionescu R, Jipescu I. Psychomotor functioning
in unipolar and bipolar aective disorders. Rom J Neurol
Psychiatry 1991; 29: 1733.
66. Kuhs H, Reschke D. Psychomotor activity in unipolar and
bipolar depressive patients. Psychopathology 1992; 25:
109116.
67. Mitchell P, Parker G, Jamieson K et al. Are there any
dierences between bipolar and unipolar depression?
J Affect Disord 1992; 25: 97106.