. 20
( 68 .)


Hypom+cyclo Hypom
Hypom 9.4%
Minor Minor
Dyst Min BP 0.7%
1.3% MinBP
Mild D Mild D 13.2%

Total 24.7% 48.8% 48.8%
dep/BP 10.8 3.0 1.0

Fig. 5.1 Cumulative prevalence rates of the bipolar spectrum. BP-1, bipolar disorder I; MDD, major
depressive disorder; Hypom, hypomania; Dyst, dystonia; D,/Dep, depression; BP, bipolar;
DSM-IV, Diagnostic and Statistical Manual of Mental Disorders. Reprinted with slight
modifications from Angst, J. and Gamma, A. (2002). Prevalence of bipolar disorders:
traditional and novel approaches. Clin. Approaches Bipolar Disord., 1, 10“14, by permission
of Cambridge Medical Publications.

BP-II, minor bipolar disorders, and hypomania (Angst et al., 2003a; Angst and
Gamma, 2002). Figure 5.1 illustrates the cumulative prevalence rates of the bipolar
spectrum, with its subgroups, as diagnosed in the Zurich study across six interviews
from the age of 20 to 40 (Angst et al., 2003a; Angst and Gamma, 2002). The tip of
the iceberg consists of 0.55% BP-I disorder, below which we found 5.3% BP-II
disorders defined by hard Zurich criteria and a further 5.7% suspected BP-II cases
defined as major depressive episodes (MDE) with hypomanic symptoms. In the
same paper, we also proposed a new group of minor bipolar disorders (MinBP)
with a prevalence rate of 3.2% (hard definition) and 6.2% (soft definition). This
important MinBP group consists of depressives (dysthymia, minor depression,
and recurrent brief depression (RBD)) who, in addition, manifested hypomanic
syndromes (hard definition) or hypomanic symptoms (soft definition).
Figure 5.1 shows that a hard definition of bipolarity identifies about one-quarter
of all mood-disorder cases as bipolars, whereas a soft definition, including suspect
cases, identifies about half of such cases as bipolars. This finding is especially relevant
in regard of MDD, which was found to be no more prevalent than BP-II disorders. A
soft definition of BP-II disorders enables us to identify most hidden bipolar cases,
which are usually misdiagnosed as MDD. For more details, see Angst et al. (2003a).
111 Recurrent brief depression

Recurrent brief psychiatric syndromes
Although the relevant concepts were established a decade ago, recurrent brief
psychiatric syndromes are still not fully researched. They are characterized by a
special course pattern, with frequently recurring (at least monthly) brief episodes,
lasting between a few days and less than 2 weeks. Recurrent brief mood and anxiety
disorders include recurrent brief hypomania (RBM) (Angst, 1992), RBD (Angst,
1988), and recurrent brief anxiety (RBA) (Angst and Wicki, 1992). Recurrent brief
psychiatric syndromes were not conceptualized as new disorders but as elements of
the manic, depressive, and anxiety spectra. Diagnoses of RBD and MDD are not
mutually exclusive in our studies. In the same year a patient may manifest an
episode of MDD and many brief episodes of depression, qualifying for RBD. The
association may also be found during the course of an illness over years. The
combination of recurrent brief psychiatric syndromes with the corresponding
major disorders, for instance major depression, bipolar disorder, panic disorder,
or generalized anxiety disorder, has great clinical relevance in terms of impairment
and treatment.
Associations of RBD with MDDs were termed early on ˜˜combined depression™™ by
analogy with ˜˜double depression™™ (MDD þ dysthymia; Montgomery et al., 1989,
Merikangas et al., 1990). Combined depression (CD) has been shown to be more
severe than pure MDD in the community (Angst et al., 1990) and in clinical samples
(Pezawas et al., 2002b). The increased risk of suicide attempts and the severe clinical
condition observed in CD are intriguing, since CD does not differ from either MDD
or RBD with respect to the required psychopathological depression criteria. It is
important for the understanding of the concept to bear in mind also that the
hierarchy of the DSM-IV system, for example, determines that patients with either
˜˜pure™™ MDD or CD are both diagnosed as MDD. We may therefore assume that this
subgroup of depressive patients may have contributed more than ˜˜pure™™ MDD to
clinical impairment in DSM-IV-based studies on MDD.
This chapter begins by analyzing the associations between RBD (Angst, 1988),
RBM (Angst, 1992), and RBA (Angst and Wicki, 1992). All three recurrent brief
psychiatric syndromes share an ultrarapid cycling pattern of mood symptoms. In
association with MDEs, they clearly increase impairment and worsen treatment
outcomes. Given this greater clinical severity of CD, it is reasonable to hypothesize
that BP-II disorders combined with RBD also represent more severe clinical
conditions than pure BP-II forms. The main goal of this chapter, then, is to test
this hypothesis by comparing diagnostic subgroups of mood disorders with and
without RBD in a large number of validating clinical variables, including family
history, course, personality, and comorbidity. Cases associated with RBD will
hereafter be termed ˜˜combined™™ (e.g., combined BP II disorder, combined
MDD, combined MinBP).
112 J. Angst et al.

Males Females
Age Age
19 2201 2346 20 1978 Screening

20 292 299 21 1979 Interview
21 234 270 22 1980 Questionnaire
22 220 236 23 1981 Interview

27 225 232 28 1986 Interview
29 200 224 30 1988 Interview
34 192 215 35 1993 Interview
40 162 205 41 1999 Interview

Fig. 5.2 Design of the Zurich cohort study.

The Zurich study is based on the population of the canton of Zurich, which
accounts for approximately one-sixth of the population of Switzerland. In a first
stage of the study (1978) 4547 subjects (male, 2201; female, 2346) were screened
using the Symptom Check List 90-R (SCL-90R). Derogatis™s (1977) subsample
selected for interview consisted of 591 subjects, two-thirds of whom were high-
scorers on the SCL-90R (85th percentile or more) and one-third were randomly
selected from those with lower scores. By weighting the two strata, it is possible to
extrapolate to a representative group of 2599 persons of the general population of
the same age.
The male subsample (n ¼ 292) was screened at the age of 19 and the female
subsample (n ¼ 299) at the age of 20; to date they have been interviewed six times
between the ages of 20 and 40 (male) and 21 and 41 (female). Figure 5.2 shows the
design of the Zurich study.

The six interviews were conducted in 1979, 1981, 1986, 1988, 1993, and 1999. After
20 years, 60% of the sample remained in the study and there had been no
significant distortion of the sample by attrition; the proportion of the two strata
of high- and low-scorers on the SCL-90 remained stable (Eich et al., 2003). The
interviews were carried out mainly by trained clinical psychologists and psychia-
trists with the Structured Psychopathological Interview and Rating of the Social
Consequences for Epidemiology (SPIKE: Angst et al., 1984). The interview has
113 Recurrent brief depression

been shown to have good diagnostic validity in depressed hospitalized and ambu-
latory patients (Illes, 1981; Pfortmuller, 1983; Busslinger, 1984; Meier, 1985); an
interrater reliability study showed good specificity and sensitivity (Angst et al.,
2003c) for the syndromes of depression and anxiety.

Bipolar spectrum
A spectrum of bipolar disorders was defined as follows (for a detailed description
of the bipolar spectrum, see Angst et al. (2003a)):
(1) BP-I disorder: DSM-IV criteria
(2) BP-II disorders: Zurich criteria using two definitions:
(a) a hard syndromal definition required the presence of hyperactive or mood
symptoms plus three of seven symptoms of mania plus personal or social
(b) a soft definition required only the presence of manic symptoms. No
episode-length criterion was applied, as we could find no empirical
evidence for its validity
(3) MinBP: the definition of MinBP required the diagnosis of dysthymia, minor
depression (MinD), or RBD plus a hypomanic syndrome (a) or hypomanic
symptoms (b).
(4) Cyclothymia was diagnosed in cases where dysthymia was combined with
hypomania or hypomanic symptoms and where more days than not had been
spent in affective states during the 12 months prior to an interview.

Depressive spectrum
The depressive spectrum was as follows:
(2) Dysthymia: DSM-IIIR
(3) MinD: the diagnosis of MinD is often defined by any depressive symptoms
present over at least 2 weeks (Judd et al., 1994); in the Zurich study we
restricted the definition to syndromes with three to four of the nine DSM-
IIIR criterial symptoms, because under this threshold there was almost no
treatment-seeking or impairment
(4) RBD.

Recurrent brief psychiatric syndromes
(1) RBD required the presence of the major depressive syndrome according to
DSM-IIIR symptom criteria (five of nine symptoms) with brief duration (less
114 J. Angst et al.

than 2 weeks, usually 1“3 days) and frequent recurrence (at least monthly over
the previous 12 months) plus work impairment
(2) The definition of RBM required the presence of a hypomanic syndrome
(Angst et al., 2003a) of brief duration (1“3 days) and frequent recurrence
(at least monthly over the previous 12 months). In addition, we tentatively
defined a diagnosis of recurrent brief hypomanic symptoms (RBMS) with
brief duration and high recurrence
(3) RBA was diagnosed in the presence of a DSM-III syndrome of generalized
anxiety disorder of brief duration (less than 2 weeks) and frequent recurrence
(at least monthly over the previous 12 months; Angst and Wicki, 1992).

Combined syndromes
This paper will concentrate on combined major depression and combined bipolar
As stated earlier, we defined combined major depression as the association of
MDD with RBD; combined BP-II disorders as the association of BP-II with RBD;
and combined MinBP as the association of MinBP with RBD. The combined
groups will be compared with the pure groups in order to test the hypothesis that
the combined are more severe than the pure syndromes.

Rapid cycling and seasonality
Rapid cycling was diagnosed if the subjects reported having suffered from four or
more episodes of depression or hypomania per year; brief episodes with a length
under the diagnostic thresholds of DSM-IV were included.
Seasonality was assessed by questions on the presence of the episodes in spring,
summer, fall, and winter.

Treatment and distress
Treatment was defined as consultation of doctors or psychologists for mood
syndromes. Distress as a consequence of manic and depressive symptoms was
estimated on an analogue scale from 0 to 100 (thermometer) over the 12 months
prior to the interview.

A personality trait of mood instability (ups and downs) was assessed by the
question: ˜˜Would you say you were one of those people who have frequent ups
and downs?™™
A diagnosis of depressive personality disorder (DPD) was derived from the
General Behavior Inventory of Depue et al. (1981; items 16, 20, 36, 47, 56, 62, 71
115 Recurrent brief depression

corresponding to the seven items of DPD in DSM“IV); unlike DSM“IV we
required not just five but all seven items to be present at least sometimes.
Anxious personality features were assumed to be present if the subject answered
yes to both of the following questions:
1. As a child or adolescent were you more anxious than your peers?
2. Do you feel your anxieties impaired your development?
Neuroticism (N), extroversion (E) and aggression (A) were measured by the
Freiburg Personality Inventory (FPI) of Fahrenberg and Selg (1970) using a
questionnaire with 212 items given at the fourth and fifth interviews (ages:
males, 29/30; females, 34/35). These three factors “ aggression, extroversion, and
neuroticism “ were derived from large Swiss population samples (6315 males, 1381
Impulsivity/irritability was represented by subfactor 4 of the FPI.

SAS for Windows version 8.01 was used. For group comparisons, chi-squared
tests, Fisher™s exact tests, and Kruskal“Wallis tests were applied. Prevalence rates
and standard errors were computed by Stata 7.0 with adjustment for sample
stratification. Cumulative prevalence rates refer to the sum of 1-year prevalence
rates across all interviews. For certain analyses, logistic regression and biserial
correlations were computed.

Recurrent brief psychiatric syndromes and their overlap
The main focus of this chapter is the relationship of the two mood spectra with
recurrent brief psychiatric syndromes (RBD, RBM, and RBA), with particular
emphasis on RBD. An association is very frequent: Table 5.1 demonstrates
that 44% of BP-II cases and 39% of MDD cases received an additional diagnosis
of RBD; in addition 71% of MinBP cases were associated with RBD. Furthermore,
over one-third of mood-disorder cases also received a diagnosis of RBA.
This raises the question of the interrelationship between RBD, RBM, and RBA.
Figure 5.3 shows their overlap computed as odds ratios: it is intriguing that the
highest associations were found between RBA and RBD (OR ¼ 5.2) and RBA and
RBM (OR ¼ 3.8). However, RBM was more closely associated with RBA than with
RBD (OR ¼ 2.9), which underlines the important role of anxiety in brief hypo-
mania and bipolar disorders in general, as shown in Table 5.1. The associations are
very similar in the case of RBMS.
Figure 5.3 shows the overlap on the basis of prevalence rates.
116 J. Angst et al.


. 20
( 68 .)