NATURAL HISTORY

Historical Perspectives and Natural History of Bipolar

Disorder
Jules Angst and Robert Sellaro

A review of two centuries’ literature on the natural history from our lifelong Zurich follow-up study (Angst and
of bipolar disorder, including modern naturalistic studies Preisig 1995a, 1995b). The article focuses mainly on
and new data from a lifelong follow-up study of 220 episodes and recurrence and, to a lesser extent, outcome; it
bipolar patients, reaches the following conclusions: the does not deal with rapid cycling and seasonal depression.
findings of modern follow-up studies are closely compat- Recent reviews of the course of bipolar disorder have been
ible with those of studies conducted before the introduc-
published by Lavori et al (1984), Keller (1987), Goodwin
tion of modern antidepressant and mood-stabilizing treat-
ments. Bipolar disorder has always been highly recurrent and Jamison (1990), Coryell and Winokur (1992), Ver-
and considered to have a poor prognosis. doux and Bourgeois (1995), Kessing et al (1998), Gold-
berg and Harrow (1999), Marneros (1999), and Bourgeois
Bipolar patients who have been hospitalized spend about
20% of their lifetime from the onset of their disorder in and Marneros (in press).
episodes. Fifty percent of bipolar episodes last between 2
and 7 months (median 3 months). The intervals between The Concept of Bipolar Disorder
the first few episodes tend to shorten; later the episodes
return at an irregular rhythm of about 0.4 episodes per We owe the categorization of bipolar disorder as an illness
year with high interindividual variability. Switches from to Falret, who in 1851 and 1854 on the basis of longitu-
mania into mild depression and from depression into dinal observations developed the entity of “folie circu-
hypomania were frequently reported in the 19th century laire” (circular madness), defined by manic and melan-
and the first half of the 20th. cholic episodes separated by symptom-free intervals. In
Antidepressant and antimanic drugs have to be given as 1854 Baillarger used the term folie à double forme to
long as the natural episode lasts. Given the poor outcome describe cyclic (manic–melancholic) episodes (Pichot
of bipolar disorders found in naturalistic follow-up studies 1995; Ritti 1879). Kraepelin called such cyclic episodes
and our lifelong investigation, intensive antidepressant, “double attacks.” In both French diagnoses the prognosis
antimanic, and mood-stabilizing treatments are required was considered to be “desperate, terrible and incurable”
in most cases. Despite modern treatments the outcome into
(Bourgeois and Marneros, in press). Circular illness was
old age is still poor, full recovery without further episodes
rare, recurrence of episodes with incomplete remission the described by most authors as a recurrent condition; it
rule, and the development of chronicity and suicide still became the prototype of the larger group of periodic
frequent. Biol Psychiatry 2000;48:445– 457 © 2000 So- psychoses embracing periodic mania, periodic melancho-
ciety of Biological Psychiatry lia, and periodic cyclic disorders (Ballet 1903; Mendel
1881; Pilcz 1901; Ziehen 1902, 1907).
Key Words: Bipolar disorder, natural history, course,
recurrence, outcome
The Concept of Mixed States
The history of the concept of mixed states has been
Introduction extensively studied by Marneros (in press): what we today
call “mixed states” were probably already known at the
T his article briefly reviews the natural history of bipolar
disorder, giving special weight to historical studies
before the era of antidepressants; integrates the results of
beginning of the 19th century and named “mixtures”
(Mischungen) by Heinroth in 1818 and “middle forms”
(Mittelformen) by Griesinger (1845). Guislain (1852) gave
modern naturalistic follow-up studies; and from our own
clear descriptions of different syndromes of mixed states.
findings 1) reanalyzes data from an early multicenter study
The history of bipolar disorder by Haustgen (1995) traces
(Angst et al 1968b, 1973) and 2) includes some new data
the term mixed states to J.P. Falret’s son Jules Falret
(1861).
From the Zurich University Psychiatric Hospital, Zurich, Switzerland. Very influential in this field was Weygandt (1899), who
Address reprint requests to Jules Angst, M.D., Zurich University Psychiatric
Hospital, Box 68, Lenggstreet 31, Zurich 8029, Switzerland.
worked with Kraepelin and whose monograph distin-
Received January 13, 2000; revised April 13, 2000; accepted April 20, 2000. guished three forms of mixed states: manic stupor, agitated

© 2000 Society of Biological Psychiatry 0006-3223/00/$20.00

PII S0006-3223(00)00909-4
446 BIOL PSYCHIATRY J. Angst and R. Sellaro
2000;48:445– 457

melancholia (depression with flight of ideas and agitation), frequently in adolescence (Weissman et al 1988) and that
and unproductive mania (elated mood, increased motor manic episodes manifest usually in the early 20s (Fogarty
activity, and inhibition of thinking). Kraepelin’s (1899) et al 1994).
textbook descriptions of mixed states were founded on
Weygandt’s monograph. Further progress was made by
Rehm’s monograph (1919, 113), which classified mixed Periodic Mania and Switches of Polarity
states systematically on the basis of the permutations of There is considerable interest today in data on the course
the three elements that had been defined by Kraepelin: of single and multiple episodes, which can answer ques-
thought disorder, mood, and psychomotor activity (iden- tions about the psychopathology, duration, and frequency
tified as a, b, and c for mania and as A, B, and C for of episodes; the syndromal stability over lifetime; and the
depression). frequency with which initial major depression develops
into bipolar disorder. Today the switch of an episode from
depression to hypomania is often assumed to be drug
Kraepelin’s Manic–Depressive Insanity induced, but the phenomenon was already very common
At the turn of the 19th century Kraepelin’s unifying as “reactive hyperthymia” before the introduction of
approach to the classification of mood disorders (1899) antidepressants.
resulted in bipolar disorders being subsumed within man- A century ago the concept of periodic mania was well
ic– depressive insanity (MDI), a broad group that included known and the diagnosis much more frequent, quite
single-episode and recurrent depression. Kraepelin (1913, simply because it was applied to cases that today would be
1183) was later himself to raise the possibility of the considered bipolar disorders. For instance, an initial de-
heterogeneity of MDI. Unlike the French concepts, Kra- pressive syndrome cycling into a manic episode, although
epelin’s MDI had a good prognosis and did not develop frequently observed, was not considered an indication of
into severe dementia, although Kraepelin conceded the bipolarity (Mendel 1881; Ziehen 1902). Similarly, “post-
existence of mild residual states after recovery from the melancholic reactive hyperthymia” with clear hypomanic
episodes themselves (Schwächezustände; Kraepelin 1913, symptoms was compatible with the diagnosis of pure
1349) and of mild fluctuations between episodes. Kraepe- periodic melancholia (Ziehen 1907, 26).
lin considered periodicity to be unimportant for the diag- Mania switching into depression was likewise very
nosis (Pilcz 1901). As a consequence of Kraepelin’s commonly reported as “reactive depression” (Ziehen
unification of affective disorders, research on their course 1902, 546, 554; Wernicke 1906, 355). Postmanic depres-
frequently failed to distinguish between depression, ma- sion lasted a few days or a few weeks according to
nia, and bipolar disorder (Bratfos and Haug 1968; Fuller Wernicke (1906, 355). Mild depression was observed
1935; Paskind 1930; Pollock 1931a, 1931b, 1931c; Poort preceding or terminating manic attacks in most of the 128
1945; Rennie 1942; Tomasson 1947). manic patients studied by MacDonald (1918).
Notable contemporary authors nevertheless disagreed Modern naturalistic and treatment studies have also
with Kraepelin’s unitarian approach, and their studies of found that mania frequently cycles into depression: the
the natural history of affective disorders maintained the rates of cycling observed in follow-up studies over 8
distinction between mania, depression, and bipolar disor- weeks vary from 17% (Tohen et al 1990) to 30% (Keller
der (Ballet 1903; Pilcz 1901; Ziehen 1902, 1907). This et al 1986a). Our earlier retrospective record study of 300
data on the course of bipolar disorder collected in the 19th manic patients (admitted between 1920 and 1970) found
century and the first half of the 20th, before the introduc- that 21% of manic episodes cycled into depression, a rate
tion of modern antidepressants and mood stabilizers, is of that did not change significantly during the intervening
special value in that it represents the disorder’s untreated decades (Angst 1987).
natural history. In a retrospective record study, depression switching
into hypomania was found in 29% of bipolar patients
hospitalised between 1920 and 1959 (Angst 1987).
Onset of Bipolar Disorder
The dating of the age of onset is to a certain extent
Diagnostic Change from Depression to
unreliable because it is usually retrospective and depen-
Bipolar Illness
dent on insecure recall. Bipolar disorder begins about 10
years earlier than recurrent depression, as shown by a The syndromal course over lifetime has been little inves-
review of the literature (Angst 1988). Earlier studies tigated. It was frequently assumed that mania was predom-
indicated a mean age of 28 to 33 years; epidemiologic and inant in earlier years and depression in the second half of
newer clinical studies show that bipolar symptoms start life. Kinkelin (1954) followed-up 146 hospital first admis-
Bipolar Disorder: Natural History BIOL PSYCHIATRY 447
2000;48:445– 457

Figure 1. Proportion of syndromes across

20 episodes in male subjects. BP, bipolar;
Man, mania; Dep, depression.

sions suffering from MDI (1929 –1947) until 1948 in a In their monograph, Marneros et al (1991a) found that
study covering an average of 21.8 years of the total course mania frequently developed into schizomania or mixed
of the disorder. Of the 146 cases, 125 began with depres- states.
sion and 21 with mania. During the follow-up period 36
(28.8%) of the 125 depressive patients developed manic
Syndromal Stability of the Course
episodes, a figure that would correspond to a diagnostic
change from depression to bipolar illness of 1.3% per year New data from our Zurich follow-up study confirmed a
of observation. Marneros et al (1991b) reported that the major gender difference in the psychopathology of bipolar
initial diagnosis of depression remained stable in 79% of patients over the first 20 episodes (Angst 1978): female
cases over 27 years. In our preponderantly prospective subjects manifested significantly more depressed episodes
study we found a rate of diagnostic change from depres- and male subjects more cyclic episodes (mania and de-
sion to hypomania/mania of about 1% per year (Angst and pression), whereas pure manic and mixed episodes were
Preisig 1995a). Coryell and colleagues’ (1995a) intensive equally frequent in both genders (Figures 1 and 2). The
prospective follow-up study of 381 depressive subjects syndromal proportions were found to remain remarkably
over 10 years found that 10.2% developed into mania stable over 20 episodes (Angst and Weis 1967), which also
(5.2%) and hypomania (5.0%), which also corresponds to means, for instance, that aging brings no increase in the
a 1% change per year of observation. depressive component of bipolar illness.

Figure 2. Proportion of syndromes across

20 episodes in female subjects. BP, bipolar;
Man, mania; Dep, depression.
448 BIOL PSYCHIATRY J. Angst and R. Sellaro
2000;48:445– 457

Table 1. Length of Episodes (Months) of Patient Samples vival analysis demonstrated that mixed or cycling episodes
Mean Median Q1 Q3
were slower to recover than pure depressed or pure manic
episodes. Perugi et al (1997) also reported that episodes
Mendel (1881) 5– 6 3– 4 6 –7
involving mixed states lasted significantly longer (mean
Kraepelin (1913) 6–8
Panse (1924) 7 13.4 months) than manic episodes (8.8 months).
Wertham (1929) 4–6 2– 4 8 –10 The computation of the length of episodes has to take
Rennie (1942) 3.5a–5.8b into account their lognormal distribution (Angst and Weis
Kinkelin (1954) 3.5– 8.4 1967) and to control for the individual number of episodes
Angst and Preisig (1995a) 4.3c 3d 2 5
(Slater 1938). In patients experiencing multiple cyclic
a
b
First episode. (bipolar) episodes, the episodes tend to be slightly shorter,
Third episode.
c
Mean after logarithmic transformation. whereas in general the latest episode tends to be longer
d
Median of Table 2. because of the development of chronicity in some cases, as
found in our multicenter study on the course of mood
disorders (Angst et al 1973).
Natural Length of Episodes In the Baltimore Epidemiologic Catchment Area Fol-
Valuable data on the natural length of episodes were low-Up Eaton et al (1997) found a median length of
published before the introduction of effective treatments episodes of approximately 8 to 12 weeks; this is shorter
(Table 1). than that reported in treated populations, but similar to the
Mendel (1881, 155) reported data on the length of median episode length of 8 weeks reported in a commu-
manic episodes (N ⫽ 43), from which a median duration nity study of adolescents in Oregon (Lewinsohn et al
of 5– 6 months (Q 1 ⫽ 3– 4 months, Q 3 ⫽ 6 –7 months) 1994).
can be computed; Mendel found only one episode which The most recent data from the Zurich follow-up study
lasted 10 –12 months and one over 1 year. Kraepelin (Angst and Preisig 1995a) showed a mean episode length
(1913) stressed the great variability in episode length but of 4.3 months (s ⫽ 5.44) calculated on the basis of
estimated that most episodes lasted between 6 and 8 intraindividual means. The median length of episodes is
months. not obviously dependent on the total number of epi-
Panse’s follow-up study (1924) of 205 hospitalized sodes—see Table 2, where subgroups with two to 10
bipolar patients found an identical mean length of 7 episodes are computed separately (following Slater’s
months for manic and depressive episodes. [1938] suggestion). On the whole, the median length of
Wertham (1929) provided the most conclusive data on episodes in bipolar illness (on the basis of individual
episode length with his investigation of 1000 male and medians) was 3 months (Q 1 ⫽ 2 months, Q 3 ⫽ 5
1000 female first admissions for mania. From his pub- months). There is a difference in length dependent on
lished histograms we can today estimate the length of psychopathology. Pure manic and pure depressive epi-
episodes as lognormally distributed, with a mode between sodes lasted 3 months (Q 1 ⫽ 2, Q 3 ⫽ 5), as did mixed
2– 4 months and a median duration of 4 – 6 months (Q 1 ⫽ episodes (M ⫽ 3; Q 1 ⫽ 2, Q 3 ⫽ 7); in contrast, cyclic
2– 4 months, Q 3 ⫽ 8 –10 months). There was no episodes lasted almost 50% longer (M ⫽ 4.19; Q 1 ⫽
follow-up; most of these manic cases may therefore have 2.5, Q 3 ⫽ 7.75). These figures are based on intraindi-
been bipolars. vidual medians. We found no gender differences in me-
Rennie (1942) found that there was a lengthening of dian episode length.
repeated manic episodes, with the first episode lasting 3.5 The shorter episode length reported by modern studies
months, the second 5.2 months, the third 5.8 months, and may be a result of their including milder cases or, what is
the fourth and fifth even longer. Rennie ascribed this to a more probable, a consequence of antidepressant therapy:
prolonging effect of aging, since before the age of 45 medication has usually to be maintained for a further 6
episode length remained constant. months after recovery to avoid relapse into the still
Unlike Rennie, Kinkelin (1954), in a longitudinal study, persistent latent episode, which represents the natural
found no systematic change in episode length between the history.
first and seventh episodes. Among 347 bipolar patients the
mean length of depressive episodes varied impressively
from 3.5 to 8.4 months, and of manic episodes from 4 to
Recurrence of Bipolar Disorder
11.6 months. Kinkelin, too, concluded that later episodes
tended to be longer. The total number of episodes experienced by patients is
In the NIMH Collaborative Study on the Psychobiology amply reported. Findings regarding the occurrence of
of Depression Clinical Studies (Keller et al 1986a) sur- single episode cases vary widely, ranging from 0% to 55%
Bipolar Disorder: Natural History BIOL PSYCHIATRY 449
2000;48:445– 457

Table 2. Total Number of Episodes and Length (Zurich Study)

Episodes Patients Median length of episodes (months)
2 4 1.5 5.8
3 8 6.0 2.8 9.0
4 9 3.0 5.0 6.0 9.0
5 15 5.5 4.5 4.6 4.0 5.8
6 14 2.5 5.5 5.0 3.8 2.8 2.9
7 21 3.0 5.0 4.0 4.3 3.5 3.0 4.0
8 12 2.9 3.0 2.0 3.0 2.4 2.5 4.0 4.0
9 13 3.5 4.5 4.0 4.0 3.0 2.5 2.5 2.0 1.0
10 15 4.5 3.5 3.0 3.5 4.5 3.0 3.0 3.0 3.0 6.0

Episodes 1 2 3 4 5 6 7 8 9 10

(Goodwin and Jamison 1990), with five of 11 studies number of two episodes in the previous 2 years.) The
reporting rates between 0% and 8%. It is clear that any loss 4-year follow-up of a naturalistic study of mania showed
of information will inflate nonrecurrence rates; further- recurrence in 72% of patients, with a mere 28% remaining
more, the length of follow-up and development into in remission (Tohen et al 1990). Patients who were
chronicity have to be taken into account in assessing symptomatic at the 6-month follow-up had a 45% greater
findings. chance of a recurrence within the next 3.5 years.
The earlier literature assumed that periodic mania had a
better prognosis with fewer episodes and better outcome
Course of Subtypes of Bipolar Disorder
than periodic melancholia or circular disorder, the latter
having the worst outcome (Rehm 1919, 107); however, Important findings from a naturalistic study, the National
Rehm also found twice as many manic cases (53%) than Institute of Mental Health (NIMH) Collaborative Program
circular cases (26%) with short free intervals between on the Psychobiology of Depression, Clinical Studies
episodes (up to 1 year), reflecting a higher periodicity of (Keller et al 1993), showed a high rate of recurrence for
mania (it is unclear whether the length of observation in pure mania (48% by 1 year and 81% by 5 years) and even
these cases was the same). higher rates for the mixed cycling group (57% by 1 year
Certain methodological advances contributed to the and 91% by 5 years). Over 7 years the rate of recurrence
investigation of the question of recurrence. One was the was 81%. The length of sustained recovery was associated
introduction of a more precise terminology and clear with a lower risk for recurrence over the subsequent 4
definitions of remission, recovery, relapse, and recurrence years, but over a period of 10 years this predictive power
(Frank et al 1991). Another was the introduction of life decreased considerably (Coryell et al 1995b); the authors
table analyses into psychiatry, a method first used for showed that even under sustained lithium prophylaxis
length of hospitalization (Kramer 1969) and for follow-up recurrences were present in more than 70% of cases within
data after recovery from depression (Klerman et al 1974) 5 years of recovery. This finding is consistent with the
and soon also applied in longitudinal studies of bipolar outpatient study of Gitlin et al (1995), in which 73% of 82
disorder (Dunner et al 1976, 1979; Fleiss et al 1976, 1978). bipolar patients had relapses/recurrences over an average
Lavori et al (1984) applied life table methods to reanalyze of 4.3 years despite maintenance pharmacotherapy (two
40 earlier studies on the relapse/recurrence of affective thirds of patients who relapsed experienced multiple re-
disorders; the results varied considerably between the lapses). Among the 26 patients who suffered no relapse
studies, and the authors formulated the hypothesis of the 46% continued to show significant symptoms of mania or
heterogeneity of patients’ courses in terms of low or high depression.
hazard with a low or high risk of relapse. More recently, The NIMH Collaborative Program on the Psychobiol-
survival analysis was applied to prospective data on ogy of Depression, Clinical Studies (Coryell et al 1989)
recurrence after remission (Fleiss et al 1978; Gitlin et al provided no evidence that the course of bipolar II disorder
1995; Keller et al 1993; Lavori et al 1984). differed from that of bipolar I, which confirms our own
In a 2-year placebo-controlled lithium study Fleiss et al findings (Angst 1986).
(1978) found that under placebo 80% of bipolar I patients A recent study comparing patients suffering from bipo-
experienced recurrences within about 70 weeks. (Admit- lar II disorder with major depressive subjects under
tedly a treatment study of this type dealt with a selected fluoxetine showed similar relapse/recurrence rates of 36%
sample of patients, who, for instance, had had a mean versus 35% after 50 weeks and 44% versus 49% after 62
450 BIOL PSYCHIATRY J. Angst and R. Sellaro
2000;48:445– 457

weeks (Amsterdam et al 1998). This study provides no cycles on the basis of the Cologne naturalistic follow-up
indication of a deterioration in the course of bipolar II study of 30 bipolar and 56 schizo-bipolar patients. Zis et al
disorder due to antidepressant medication; on the contrary, (1980) and Zis and Goodwin (1979) arrived at similar
the disorder had a better course than under placebos. conclusions. This seemed to confirm earlier findings that
Marneros et al (1991a) reported that bipolar disorder had suggested decreasing cycle length (e.g., those of
and schizo-bipolar disorder had very similar course Kraepelin [1913, 1325] or Kinkelin [1954], although not
characteristics. controlled for number of cycles—see Figure 6-3 in Good-
win and Jamison [1990]). All these results, together with
the finding that precipitation rates decrease (Angst 1966,
Is Recurrence of Bipolar Illness 41) with increasing recurrence, led Roy-Byrne et al (1985)
Progressive? to speak of sensitization and Post et al (Post 1992; Post et
Describing the intervals between episodes, Kraepelin al 1984, 1986) to develop the theory that vulnerability
(1913, 1365) found a progressive shortening of the first grows with the number of episodes and the theory of
three intervals (first interval of male/female subjects 4.6/ conditioning, sensitization, and kindling (by analogy with
4.3 years, second interval 2.8/2.0 years, third interval electrophysiologic kindling).
1.2/1.4 years). His data, like that of some other authors, are National data on hospital admissions and readmissions
not controlled for the total number of episodes: clearly the can also provide a rough estimate of the natural history of
intervals between episodes in patients having few episodes severe cases. Kessing et al (1998a, 1998b) recently de-
over a lifetime will be longer than in patients who scribed the course of Danish hospital admissions on the
experience many. We owe this particular methodological basis of a nationwide register of ICD-8 diagnoses. Rehos-
breakthrough to Eliot Slater (1938), who, in his paper on pitalizations were taken as a measure of recurrence. The
the periodicity of manic– depressive insanity, investigated authors found, on the basis of 2903 bipolar cases, a
116 patients of the Forschungsanstalt für Psychiatrie in progressive shortening of the interval between discharge
Munich who had been personally diagnosed by Kraepelin. from hospital and the next rehospitalization and, therefore,
Studying the length of intervals between episodes, Slater a deteriorating course. Selection bias did not completely
made his major methodological contribution by control- explain the shortening intervals between hospital admis-
ling for the total number of episodes and analyzing sions, and control for gender and age did not alter the
separately the subgroups with 1, 2, 3, . . . , n intervals. He conclusions. So far, then, it would seem to have been
showed that there was indeed a shortening of the intervals, established that the course of bipolar disorder is recurrent
but only between the first few manifestations of the illness. and progressive, but this aspect is still surrounded by
Investigating the individual periodicity, he concluded that considerable controversy, which may partially be due to
every patient had his or her own rhythm. the possibility “that recent studies deal with different,
In 1968 Bratfos and Haug applied Slater’s method to the more broadly diagnosed populations than the seminal,
follow-up data on 215 cases of manic– depressive disorder earlier studies” (Grof et al 1995).
(including depression) in an analysis of the length of the
intervals between episodes. Correcting for the number of
episodes, the authors found that the first interval length
The Lithium Controversy: Does Recurrence
was 2.1 years, the second was 4.8 years, and the third was
Improve Spontaneously?
2.2 years; thus no clear tendency emerged from these data. One controversy about the natural history of bipolar
Again following Slater’s method of correcting for the disorder dates back to 1968 and the criticism by two
number of episodes, Angst et al (1973) published the reputed British authors, Blackwell and Shepherd of lith-
results of a multinational retrospective and partially pro- ium trials. They assumed not only the inefficacy of lithium
spective hospital record study on the course of 393 bipolar but also that bipolar disorder had a good prognosis,
and 634 unipolar depressive patients. Analyzing cycle making long-term prophylaxis unnecessary. This view was
lengths, they indirectly confirmed Slater’s finding that the supported theoretically by Lader (1968) and empirically
intervals shortened as the number of recurrences in- by Saran (1969), who found no evidence for high recur-
creased; however, the median cycle lengths gave clear rence in his follow-up data and who concluded that past
evidence only for continuous shortening of the first three recurrence was not predictive for recurrence in the future.
cycles; the pattern of later recurrences seemed to be Saran’s findings were of critical importance because the
unpredictable. On the other hand, nonparametric tests early work on lithium had been based statistically on the
between successive cycle lengths showed a significant assumption that high recurrence in a patient’s previous
shortening from cycle one to cycle 11 (Angst et al 1973, history should be expected to repeat itself in the future.
499). Marneros et al (1991a) confirmed the shortening of Saran concluded that his findings on the spontaneous
Bipolar Disorder: Natural History BIOL PSYCHIATRY 451
2000;48:445– 457

course corresponded to the course as observed under the Zurich sample was excluded, to avoid any
lithium treatment by Baastrup and Schou (1968). Saran’s overlap with the Zurich follow-up study. The re-
findings have not been confirmed by other studies: a maining sample consisted of 329 bipolar patients.
methodological investigation by Isaaksson et al (1969) The data collection was mainly of the retrospective
demonstrated the persistence of recurrence for bipolar and type based on case histories and verbal information.
unipolar depression, and the same conclusion was reached The documentation of the course was carried out by
by Laurell and Ottosson (1968). means of a standardized form of protocol in which
all data were entered separately for each episode and
subsequent interval (Angst and Weis 1968). In
Stability of Recurrence after Initial retrospect the onset (date) and length (months,
Deterioration weeks) of previous episodes and aspects of treat-
Another debated question is whether the course of bipolar ment (none, ambulatory, hospital) were assessed.
disorder is really progressive, characterized by unlimited The degree of remission was coded as full, partial,
shortening of cycles throughout (Angst et al 1973). This or unknown. Psychopathology was coded with a list
model is disconfirmed by a number of new studies. The of 10 syndromes. The data were reanalyzed for this
recent summary of the literature by Kessing et al (1998b) study.
shows that Fukuda et al (1983), who investigated not the 2. The Zurich follow-up study consisted of 406 con-
early but the later course of the illness, could find no secutive hospital admissions for severe depression
shortening of cycles (a finding that would still be compat- or mania from 1959 to 1963. Regular follow-up
ible with the hypothesis that shortening is a feature of the investigations (by telephone, interviews, and record
first few cycles only). This aspect of the systematic collection) were conducted in 1963, 1965, 1979,
shortening of cycle length was also very seriously ques- 1975, 1980, and 1985. In 1991 and 1997 mortality
tioned in the reviews by Coryell and Winokur (1992) and data were available from the Swiss federal office of
Solomon et al (1995), which drew mainly on Winokur and statistics. Seventy-six percent of patients had died
colleagues’ findings (1994) from the prospective natural- by the end of 1997. The sample and the methodol-
istic NIMH study covering 10 years (Turvey et al 1999). ogy of assessments were described in detail by
This study found that the second cycle was clearly shorter Angst and Preisig (1995a). The principal data col-
than the first, the third a little longer than the second, but lected were comparable to the multicenter study, but
the fourth and fifth cycles were again shorter. On the other psychopathology and treatment were assessed in
hand, Winokur et al (1993) stressed that bipolar illness more detail. The 220 bipolar patients were reana-
was highly recurrent, with an “inexorable continuation of lyzed for this article.
episodes and hospitalisations,” and could find no data
suggesting that the illness burned out at a later stage. Results
On the basis of prospective data Coryell and Winokur Table 3 presents the data from the mainly retrospective
(1992) found that rapid cycling, which is observed in 20% multicenter study and demonstrates a systematic shorten-
to 25% of bipolar patients, is usually a transient manifes- ing of the first four cycles. The predominantly prospective
tation and not therefore a characteristic of the long-term data from the Zurich sample (Table 4) shows a significant
course. shortening between cycles 1 and 2 only, with no system-
atic change thereafter. In both studies the conclusions were
confirmed by t tests for dependent measures.
Meta-Analysis of Two Studies on the Course In a survival analysis of the Zurich follow-up data
Samples and Methodology significant differences were found between cycles 1–5 but
were difficult to interpret. The first cycle was longer and
Here we reanalyze some data from our two studies, the
the second cycle shorter than all the others. Otherwise the
mainly retrospective multicenter study published by Angst
survival curves were very similar (Figure 3). The mean
et al (1973) and the Zurich follow-up study described by
cycle length was 28.7 months (s ⫽ 30.93) but the
Angst and Preisig (1995a).
median length, which is much more representative, only
1. The multicenter study (Angst et al 1968a, 1973) 18 months (Q 1 ⫽ 3, Q 3 ⫽ 18); exclusion of the first
consisted of consecutive hospital admissions of longer cycle does not change this median.
bipolar and unipolar depressive patients from Basel In conclusion, we found a shortening of cycle length at
(103); Berlin (104); Hamilton, Canada (69); Prague the beginning of the disorder only; later episodes were
(132); Zurich (392); Glostrup, Denmark (100); and persistently recurrent but occurred at irregular intervals
Landeck, Germany (140). For the present analysis without any systematic deterioration or amelioration, thus
452 BIOL PSYCHIATRY J. Angst and R. Sellaro
2000;48:445– 457

Table 3. Total Number of Cycles and Length (Multicenter Study)

Cycles Patients Median length of cycles (months)
1 18 23
2 35 26 15
3 38 38 20 15
4 45 21 27 13 14
5 39 35 18 19 13 11
6 31 42 31 19 15 15 9
7 26 28 18 16 12 13 15 12
8 27 42 17 14 14 9 11 9 10
9 17 37 20 18 9 8 12 15 15 12

Cycles 1 2 3 4 5 6 7 8 9
This table does not include Zurich data.

confirming Winokur and colleagues’ (1993, 1994) a recurrence rate of 0.66 episodes per year. On the basis of
findings. an episode length of 3 months (median), bipolar patients
spent about 2 months/year in episodes. We could not find
any gender differences (Mann–Whitney U test) in episode
Episode Frequency per Year frequency measured by cycle length. Recently Gottschalk
Periodicity can also be expressed by the episode frequency et al (1995) found some evidence for a nonlinear deter-
per year. In a follow-up of 140 bipolar I patients over 11.4 ministic structure in long-term daily mood records of six
years Dunner et al (1979) found 0.54 episodes per year; we out of seven bipolar patients, something not found in
found 0.37 (Angst and Preisig 1995a). Marneros’ group normal control subjects. Further research on a chaotic
found 0.41 episodes per year (Marneros 1999; Marneros et course is certainly desirable.
al 1991a) for bipolar disorder and about half as many in
the case of recurrent depression. These figures included
Correlates of Recurrence
the first cycle, which is considerably longer than later
ones. Dunner et al (1979) found no relationship between For the analysis of correlates with long-term recurrence
episode frequency and age of onset (which was not we excluded the first two cycles and entered the median of
consistent with our early results; Angst and Weis 1967). all intraindividual medians of cycle length as the depen-
The question of a gender difference in episode frequency dent variable into a multiple regression analysis. Cycle
per year remained open. After the onset of their disorder, length did not correlate with gender, retrospective versus
bipolar patients spent on the average about 19% of their prospective data collection, or long-term administration of
lives in affective episodes over an observation period of 27 tricyclic antidepressants. This finding does not confirm the
years (Angst and Preisig 1995a). assumed deterioration of the course of bipolar illness
In the present analysis the median cycle length (on the under tricyclics (Arnold and Kryspin-Exner 1965; Kouko-
basis of individual medians) was 18 months (Q 1 ⫽ 12 poulos et al 1980; Till and Vuckovik 1970; Tondo et al
months, Q 3 ⫽ 33.5 months), a length that corresponds to 1981). Unsurprisingly, long-term lithium and antipsy-

Table 4. Total Number of Cycles and Length (Zurich Study)

Cycles Patients Median length of cycles (months)
1 4 64
2 8 105 12
3 9 55 21 34
4 15 78 25 33 45
5 14 30 22 37 19 20
6 21 44 35 36 24 24 17
7 12 42 23 34 16 17 30 17
8 13 34 22 23 30 23 12 17 28
9⫹ 124 26 19 21 15 15 12 12 12 12

Cycles 1 2 3 4 5 6 7 8 9⫹
Bipolar Disorder: Natural History BIOL PSYCHIATRY 453
2000;48:445– 457

Figure 3. Survival analysis of cycle lengths (first to

fifth).

chotic medication correlated with shorter cycles; this observed 0.19 episodes of depression and 0.29 episodes of
result is explicable by the selection of highly recurrent mania per year followed up. Only 2% of all patients had
cases for prophylaxis. The syndromal characteristics of manic episodes with no lifetime depressions; this small
episodes had some impact on cycle length: it was signif- group had the best outcome. Bipolar patients (N ⫽ 301)
icantly shorter in cases with cyclic versus pure (manic or had the poorest outcome: 7% were recovered, 50% im-
depressive) episodes and in those with more manic than proved, and 43% unimproved at the follow-up rating. The
depressive episodes. rates of recovered, improved, and unimproved patients
were 19%, 56%, and 25% for pure mania (N ⫽ 16) and
Poor Outcome of Bipolar Disorder in Early 25%, 44%, and 32% for pure depression (N ⫽ 700).
Follow-Up Studies In a follow-up of 86 manic patients over 30 to 40 years,
outcome (measured by psychiatric symptoms) was good in
In the predrug era most bipolar cases were described as 43% of cases, fair in 18%, and poor in 25% (Tsuang et al
manifesting residual symptoms after recovery from epi- 1979).
sodes. Their state was described as “unsteady, moody,
irritable, indolent, egocentric” (Pilcz 1901, 61– 62), a
symptom that often preceded the episodes or had even Poor Outcome in Recent Follow-Up Studies
been present since childhood. Kraepelin (1913, 1185), too, Modern studies (e.g., a 4.5-year follow-up study by
although assuming a good prognosis, admitted the exis- Goldberg et al [1995]) showed that only 41% of bipolar
tence of mild residual “debility states.” Ziehen (1902) patients had a good overall outcome, the remainder being
estimated complete recovery in periodic mania to occur at moderately impaired (37%) or showing poor functioning
most in 20% of cases. (22%). Gitlin et al (1995) found a poor outcome even
In a follow-up of first admissions from 1920 to 1947 under prophylactic medication (mainly lithium); poor
over a mean of 22 years Kinkelin (1954) found chronicity outcome was more closely associated with the number of
(including severe residual states) in 14.6% of depressive depressive episodes than the number of manic episodes. In
subjects (N ⫽ 89) and in 41% of circular cases (N ⫽ their above-mentioned NIMH Collaborative Program on
51). the Psychobiology of Depression with a 15-year intensive
An important follow-up study of 297 bipolar and 945 follow-up, Coryell et al (1998) found that 56.6% of 113
unipolar patients in the Phipps Clinic (admissions from bipolar I patients had had no symptoms over the past 12
1913 to 1940) was published by Stephens and McHugh months, 22.1% had had symptoms for fewer than 52
(1991). Compared with depressive subjects, bipolar pa- weeks, and 20.4% had had symptoms in all 52 weeks,
tients had an earlier age of onset on admission and were representing a poor outcome (defined as having had
more likely to be psychotic; they also had more serious symptoms of major depressive disorder, mania, or schizo-
premorbid characteristics, more sudden onsets (44%), affective disorder in all 52 weeks of year 15).
more previous admissions (62% for depression and 67% The short-term outcome, 12 months after discharge
for mania), and more problems with alcohol, but more from hospital, was similar in patients with manic episodes
bipolar than depressed patients were discharged in a or mixed episodes (Keck et al 1998), whereas in Keller
recovered state. In bipolar disorders they found a lifetime and colleagues’ (1986b) study after 18 months of fol-
average of 3.6 hospitalizations, and in their follow-up they low-up mixed episodes also developed more frequently
454 BIOL PSYCHIATRY J. Angst and R. Sellaro
2000;48:445– 457

Table 5. Outcome of 219 Patients in the Zurich Follow-Up historical fact has to be taken into account in today’s
Study (Median Age at Follow-Up or Death, 68 Years) hypothesis of drug-induced hypomania or drug-
Recovered (GAS score ⬎ 60, no episodes over the past 16.0% induced depression; these have to be proven statis-
5 years) tically by placebo-controlled trials.
Remitted (GAS score ⬎ 60) but still recurrent (⬍5 25.5% 2. Our decades-long prospective study shows that over
years since last episode) lifetime the proportions of mania and depression in
Incomplete remission (GAS score 1– 60) over more 7.8%
than 5 years
bipolar disorder remain stable into old age. Bipolar
Incomplete remission, course still recurrent 27.0% female subjects manifest more depression than bi-
Chronic (last episode without remission, minimum 15.9% polar male subjects.
length 2 years) 3. Mixed states have been described since the early
Suicide 7.8% 19th century. Modern studies have demonstrated
GAS, Global Assessment Scale. that they have a poorer prognosis than other bipolar
conditions, with slower remissions and higher risk
for chronicity.
into chronicity (32%) than pure manic episodes (7%). In 4. The natural length of affective episodes has proba-
bipolar I patients poor outcome did not correlate clearly bly not changed over the past 120 years. Patients
with early onset of the disorder or cycling per se (Coryell responding to antidepressants still require a mainte-
et al 1998), but persistence of depressive symptoms in nance treatment throughout the underlying episode.
years 1 and 2 was correlated with impairment after 15 In clinical studies the median length of episodes is 3
years (household duties, recreational activities, overall to 6 months; in epidemiological studies it is 2 to 3
satisfaction, and global social adjustment). Such early months.
persistence of depressive symptoms predicted a poor 5. The recurrence of bipolar disorder was always the
prognosis of a bipolar subtype, whereas this was not the rule; it now seems to be established that there is
case for early persistence of manic symptoms. some initial shortening of intervals/cycles, followed
The lifetime outcome of bipolar disorder in our Zurich by an irregular persistent recurrence, with a median
follow-up study is given in Table 5 and demonstrates a cycling of 18 months. In contrast to earlier reports,
poor prognosis despite modern treatments. Up to a median the new studies show that there is no unlimited
age of 68 years only 16% of patients had recovered; 52% shortening of cycle length—not therefore supporting
still suffered from recurrent episodes and the remaining the kindling model. In several studies rapid cycling
patients had become chronically ill or had committed has been found to be relatively frequent, but usually
suicide. These data underline the poor outcome into old transient.
age and the need for intensive treatment. This table shows 6. Lifelong outcome has rarely been studied, and
the outcome of bipolar disorder before the occurrence of precise data on the natural outcome are scarce. Some
an organic brain syndrome in the elderly, which was found new prospective studies demonstrate that most pa-
in 14.5% of cases. The previous number of affective tients continue to suffer from residual depressive or
episodes was not correlated with the development of an hypomanic symptoms between episodes, and many
organic brain syndrome (Angst and Preisig 1995b). are functionally impaired.
Future studies on outcome should clearly distinguish 7. Overall research into the natural history of bipolar
between different outcome measures, since Tohen et al illness shows that it has a poor prognosis, as
(2000) have shown that functional recovery can be much reflected by high recurrence, chronicity of episodes
worse than syndromal recovery. or residual symptoms, and premature death by
suicide and somatic disorders; however, unlike
Conclusions schizophrenia, which is characterized by much
higher chronicity and the predominance of negative
Several important conclusions regarding episodes, recur- and psychotic symptoms, chronicity in bipolar dis-
rence, and outcome have emerged from this review and orders is rarer (10 –20%) and the more frequent
data analysis of the natural history of bipolar disorder: residual states are limited to characteristic depres-
1. Before the introduction of modern drugs, spontane- sive and hypomanic symptoms.
ous mild depression following a manic episode and
spontaneous hypomania following a melancholic
episode were very common; they were interpreted as The authors thank Professor Andreas Marneros (University of Halle) and
“reactive” and had no effect on the principal diag- Dr. Mauricio Tohen (Lilly Company) for their contribution to this article
nosis of pure mania or pure melancholia. This in the form of comments and suggestions.
Bipolar Disorder: Natural History BIOL PSYCHIATRY 455
2000;48:445– 457

Aspects of this work were presented at the conference “Bipolar sung des Verlaufes des manisch-depressiven Krankheitsge-
Disorder: From Pre-Clinical to Clinical, Facing the New Millennium,” schehens durch Antidepressiva. Wien Med Wochenschr 115:
January 19 –21, 2000, Scottsdale, Arizona. The conference was spon- 929 –934.
sored by the Society of Biological Psychiatry through an unrestricted
Baastrup PC, Schou M (1968): Prophylactic lithium. Lancet
educational grant provided by Eli Lilly and Company.
I:1419 –1422.
Baillarger J (1854): De la folie à double forme. Ann Med Psychol
6:369 –384.
References
Ballet G (1903): Traité de Pathologie Mentale. Paris: Doin.
Amsterdam JD, Garcia-Espana F, Fawcett J, Quitkin FM, Reim-
Blackwell B, Shepherd M (1968): Prophylactic lithium: Another
herr FW, Rosenbaum JF, et al (1998): Efficacy and safety of
therapeutic myth? An examination of the evidence to date.
fluoxetine in treating bipolar II major depressive episode.
Lancet 1(7549):968 –971.
J Clin Psychopharmacol 18:435– 440.
Bourgeois ML, Marneros A (in press): Course and prognosis of
Angst J (1966): Zur Aetiologie und Nosologie Endogener De-
bipolar disorders. In: Marneros A, Angst J, editors. Bipolar
pressiver Psychosen. Eine Genetische, Soziologische und
Affective Disorders. Dordrecht, The Netherlands: Kluwer.
Klinische Studie. Berlin: Springer.
Bratfos O, Haug JO (1968): Course of manic-depressive psycho-
Angst J (1978): The course of affective disorders. II. Typology of
sis. A follow-up investigation of 215 patients. Acta Psychiatr
bipolar manic-depressive illness. Arch Psychiatr Nervenkr
Scand 44:89 –112.
226:65–73.
Angst J (1986): The course of major depression, atypical bipolar Coryell W, Endicott J, Maser JD, Keller MB, Leon AC, Akiskal
disorder, and bipolar disorder. In: Hippius H, Klerman GL, HS (1995a): Long-term stability of polarity disinctions in the
Matussek N, editors. New Results in Depression Research. affective disorders. Am J Psychiatry 152:385–390.
Berlin: Springer, 26 –35. Coryell W, Endicott J, Maser JD, Mueller T, Lavori P, Keller M
Angst J (1987): Switch from depression to mania, or from mania (1995b): The likelihood of recurrence in bipolar affective
to depression. J Psychopharmacol 1:13–19. disorder: The importance of episode recency. J Affect Disord
33:201–206.
Angst J (1988): Clinical course of affective disorders. In:
Helgason T, Daly RJ, editors. Depressive Illness: Prediction Coryell W, Keller M, Endicott J, Andreasen N, Clayton P,
of Course and Outcome. Berlin: Springer, 1– 48. Hirschfeld R (1989): Bipolar II illness: Course and outcome
over a five-year period. Psychol Med 19:129 –141.
Angst J, Baastrup PC, Grof P, Hippius H, Poeldinger W, Weis P
(1973): The course of monopolar depression and bipolar Coryell W, Turvey C, Endicott J, Leon AC, Mueller T, Solomon
psychoses. Psychiatr Neurol Neurochir 76:489 –500. D, Keller M (1998): Bipolar I affective disorder: Predictors of
outcome after 15 years. J Affect Disord 50:109 –116.
Angst J, Battegay R, Bente D, Berner P, Broeren W, Cornu F, et
al (1968a): Das Dokumentationssystem der Arbeitsgemein- Coryell W, Winokur G (1992): Course and outcome. In: Paykel
schaft für Methodik und Dokumentation in der Psychiatrie ES, editor. Handbook of Affective Disorders, 2nd ed. London:
(AMP). Arzneimittelforschung 18:3– 8. Churchill Livingstone, 89 –108.
Angst J, Grof P, Hippius H, Poeldinger W, Weis P (1968b): La Dunner DL, Fleiss JL, Fieve RR (1976): The course of develop-
psychose maniaco-depressive est-elle periodique ou intermit- ment of mania in patients with recurrent depression. Am J
tente? Dans: Cycles biologiques et psychiatrie. In: de Aju- Psychiatry 133:905–908.
riaguerra J, editor. Symposium Bel-Air III, Geneve 1967. Dunner DL, Murphy D, Stallone F, Fieve RR (1979): Episode
Geneva: Georg et Cie, 339 –351. frequency prior to lithium treatment in bipolar manic-depres-
Angst J, Preisig M (1995a): Course of a clinical cohort of sive patients. Compr Psychiatry 20:511–515.
unipolar, bipolar and schizoaffective patients. Results of a Eaton WW, Anthony JC, Gallo J, Cai G, Tien A, Romanoski A,
prospective study from 1959 to 1985. Schweiz Arch Neurol et al (1997): Natural history of Diagnostic Interview Schedule
Psychiatr 146:5–16. DSM-IV major depression. The Baltimore Epidemiologic
Angst J, Preisig M (1995b): Outcome of a clinical cohort of Catchment area follow-up. Arch Gen Psychiatry 54:993–999.
unipolar, bipolar and schizoaffective patients. Results of a Falret J (1861): Principes à suivre dans la classification des
prospective study from 1959 to 1985. Schweiz Arch Neurol maladies mentales. Ann Med Psychol 19:145.
Psychiatr 146:17–23.
Falret JP (1851): Marche de la folie. Gazette Hopitaux 24:18 –19.
Angst J, Weis P (1967): Periodicity of depressive psychoses. In:
Brill H, Cole JO, Deniker P, Hippius H, Bradley PB, editors. Falret JP (1854): De la folie circulaire ou forme de maladie
Neuro-Psychopharmacology. Proceedings of the Fifth Inter- mentale caractérisée par l’alternative régulière de la manie et
national Congress of the Collegium Internationale Neuro- de la mélancholie. Bull Acad Med (Paris) 19:382.
Psychopharmacologicum, Washington, D.C. 1966, Interna- Fleiss JL, Dunner DL, Stallone F, Fieve RR (1976): The life
tional Congress Series No 129. Amsterdam: Excerpta table: A method for analyzing longitudinal studies. Arch Gen
Medica, 703–710. Psychiatry 33:107–112.
Angst J, Weis P (1968): Aetiologie und Verlauf endogener Fleiss JL, Prien RF, Dunner DL, Fieve RR (1978): Actuarial
Depressionen. In: Rémy M, editor. Depressionen und ihre studies of the course of manic-depressive illness. Compr
Behandlung. Fortbildungskurse der SGP, Vol 1. Basel, Swit- Psychiatry 19:355–362.
zerland: Karger, 8 –16. Fogarty F, Russell JM, Newman SC, Bland RC (1994): Mania.
Arnold OH, Kryspin-Exner K (1965): Zur Frage der Beeinflus- Acta Psychiatr Scand Suppl 376:16 –23.
456 BIOL PSYCHIATRY J. Angst and R. Sellaro
2000;48:445– 457

Frank E, Prien RF, Jarrett RB, Keller MB, Kupfer DJ, Lavori Keller MB, Lavori PW, Rice J, Coryell W, Hirschfeld RMA
PW, et al (1991): Conceptualisation and rationale for consen- (1986b): The persistent risk of chronicity in recurrent epi-
sus definitions of terms in major depressive disorder. Remis- sodes of nonbipolar major depressive disorder: A prospective
sion, recovery, relapse, and recurrence. Arch Gen Psychiatr follow-up. Am J Psychiatry 143:24 –28.
48:851– 855. Kessing LV, Andersen PK, Mortensen PB (1998a): Predictors of
Fukuda K, Etoh T, Iwadate T, Ishii A (1983): The course and recurrence in affective disorder. A case register study. J Affect
prognosis of manic-depressive psychosis: A quantitative anal- Disord 49:101–108.
ysis of episodes and intervals. Tohoku J Exp Med 139:299 –
Kessing LV, Andersen PK, Mortensen PB, Bolwig TG (1998b):
307.
Recurrence in affective disorder. I. Case register study. Br J
Fuller RG (1935): What happens to mental patients after dis- Psychiatry 172:23–28.
charge from hospital. Psychiatr Q 9:95–104.
Kinkelin M (1954): Verlauf und Prognose des Manisch-Depres-
Gitlin MJ, Swendsen J, Heller TL, Hammen C (1995): Relapse siven Irreseins. Schweiz Arch Neurol Neurochir Psychiatr
and impairment in bipolar disorder. Am J Psychiatry 152: 73:100 –146.
1635–1640.
Klerman GL, DiMascio A, Weissman MM, Prusoff B, Paykel ES
Goldberg JF, Harrow M (1999): Poor-outcome bipolar disorders.
(1974): Treatment of depression by drugs and psychotherapy.
In: Goldberg JF, Harrow M, editors. Bipolar Disorders.
Am J Psychiatry 131:186 –191.
Clinical Course and Outcome. Washington, DC: American
Psychiatric Press, 1–20. Koukopoulos A, Reginaldi D, Laddomada P, Floris G, Serra G,
Goldberg JF, Harrow M, Grossman LS (1995): Course and Tondo L (1980): Course of manic-depressive cycle and
outcome in bipolar affective disorder: A longitudinal fol- changes caused by treatments. Pharmacopsychiatry 13:156 –
low-up study. Am J Psychiatry 152:379 –384. 167.
Goodwin FK, Jamison KR (1990): Manic-Depressive Illness. Kraepelin E (1899): Psychiatrie. Ein Lehrbuch für Studierende
New York: Oxford University Press. und Ärzte, 6th ed, Vol 2. Leipzig, Germany: Barth.
Gottschalk A, Bauer MS, Whybrow PC (1995): Evidence of Kraepelin E (1913): Psychiatrie. Ein Lehrbuch für Studierende
chaotic mood variation in bipolar disorder. Arch Gen Psychi- und Aerzte, 8th ed, Vol III. Leipzig, Germany: Barth.
atry 52:947–959. Kramer M (1969): Application of Mental Health Statistics.
Griesinger W (1845): Pathologie und Therapie der Psychischen Geneva: World Health Organization.
Krankheiten für Aerzte und Studierende, 1st ed. Stuttgart: A. Lader MH (1968): Prophylactic lithium? Lancet II:103.
Krabbe.
Laurell B, Ottosson JO (1968): Prophylactic lithium? Lancet
Grof P, Alda M, Ahrens B (1995): Clinical course of affective 2:1245–1246.
disorders: Were Emil Kraepelin and Jules Angst wrong?
Psychopathology 28(suppl 1):73– 80. Lavori PW, Keller MB, Klerman GL (1984): Relapse in affective
disorders: A reanalysis of the literature using life table
Guislain J (1852): Leçons Orales sur les Phrénopathies, ou
methods. J Psychiatr Res 18:13–25.
Traité Théorique et Pratique des Maladies Mentales. Ghent,
Belgium: Hebbelynck. Lewinsohn PM, Clarke GN, Seeley JR, Rohde P (1994): Major
Haustgen T (1995): Aspects historiques des troubles bipolaires depression in community adolescents: Age at onset, episode
dans la psychiatrie Française. Encephale 21(suppl 6):13–20. duration, and time to recurrence. J Am Acad Child Adolesc
Psychiatry 33:809 – 818.
Heinroth JCA (1918): Lehrbuch der Störungen des Seelenlebens
Oder der Seelenstörungen und Ihre Behandlung, vom Ra- MacDonald JB (1918): Prognosis in manic-depressive insanity.
tionalen Standpunkt aus Entworfen. Leipzig, Germany: Vo- J Nerv Ment Dis 47:20 –30.
gel. Marneros A (1999): Handbuch der Unipolaren und Bipolaren
Isaaksson A, Ottosson J-O, Perris C (1969): Methodologische Erkrankungen. Stuttgart: Thieme.
Aspekte der Forschung über prophylaktische Behandlung bei Marneros A (in press): Origin and development of the concept of
affektiven Psychosen. In: Hippius H, Selbach H, editors. Das “mixed states”. J Affect Disord.
Depressive Syndrom. Internationales Symposion Berlin am
Marneros A, Deister A, Rohde A (1991a): Affektive, Schizoaf-
16. und 17. Februar 1968. Munich: Urban & Schwarzenberg,
fektive und Schizophrene Psychosen. Eine Vergleichende
561–574.
Langzeitstudie. Berlin: Springer.
Keck PE Jr, McElroy SL, Strakowski SM, West SA, Sax KW,
Hawkins JM, et al (1998): 12-Month outcome of patients with Marneros A, Deister A, Rohde A (1991b): Stability of diagnoses
bipolar disorder following hospitalization for a manic or in affective, schizoaffective and schizophrenic disorders.
mixed episode. Am J Psychiatry 155:646 – 652. Cross-sectional versus longitudinal diagnosis. Eur Arch Psy-
chiatry Clin Neurosci 241:187–192.
Keller MB (1987): Differential diagnosis, natural course, and
epidemiology of bipolar disorders. Annu Rev 6:10 –31. Mendel E (1881): Die Manie. Eine Monographie. Vienna: Urban
& Schwarzenberg.
Keller MB, Lavori PW, Coryell W, Andreasen NC, Endicott J,
Clayton PJ, et al (1986a): Differential outcome of pure manic, Panse F (1924): Untersuchungen über Verlauf und Prognose
mixed/cycling, and pure depressive episodes in patients with beim Manisch-Depressiven Irresein. Monatsschr Psychiatr
bipolar illness. JAMA 255:3138 –3142. Neurol 56:15– 82.
Keller MB, Lavori PW, Coryell W, Endicott J, Mueller TI Paskind HA (1930): Manic-depressive psychosis as seen in
(1993): Bipolar I: A five-year prospective follow-up. J Nerv private praxis. Length of the attack and length of the interval.
Ment Dis 181:238 –245. Arch Neurol 23:789 –794.
Bipolar Disorder: Natural History BIOL PSYCHIATRY 457
2000;48:445– 457

Perugi G, Akiskal HS, Micheli C, Musetti L, Paiano A, Quilici C, chen Behandlung auf den Verlauf endogener Depressionen.
et al (1997): Clinical subtypes of bipolar mixed states: Int Pharmacopsychiatry 4:210 –219.
Validating a broader European definition in 143 cases. J Tohen M, Hennen J, Zarate CM Jr, Baldessarini RJ, Strakowski
Affect Disord 43:169 –180. SM, Stoll AL, et al (2000): Two-year syndromal and func-
Pichot P (1995): The birth of the bipolar disorder. Eur Psychiatry tional recovery in 219 cases of first-episode major affective
10:1–10. disorder with psychotic features. Am J Psychiatry 157:220 –
Pilcz A (1901): Die Periodischen Geistesstörungen. Eine Kli- 228.
nische Studie. Jena, Germany: G. Fischer. Tohen M, Waternaux C, Tsuang MT (1990): Outcome in mania.
Pollock HM (1931a): Prevalence of manic-depressive psychoses A 4-year prospective follow-up of 75 patients utilizing
in relation to sex, age, environment, nativity and race. In: survival analysis. Arch Gen Psychiatry 47:1106 –1111.
White WA, Davis TK, Frantz AM, editors. Manic-Depressive Tomasson H (1947): Investigations on manic depressive psycho-
Psychosis. Baltimore: Williams & Wilkins, 655– 667. sis. Acta Psychiatr Scand 47:472– 480.
Pollock HM (1931b): Recurrence of attacks in manic-depressive Tondo L, Laddomada P, Serra G, Minnai G, Kukopoulos A
psychoses. Am J Psychiatry 11:567–574. (1981): Rapid cyclers and antidepressants. Int Pharmacopsy-
chiatry 16:119 –123.
Pollock HM (1931c): Recurrence of attacks in manic-depressive
psychoses. In: White WA, Davis TK, Frantz AM, editors. Tsuang MT, Woolson RF, Fleming JA (1979): Long-term
Manic-Depressive Psychosis. Baltimore: William & Wilkins, outcome of major psychoses. I. Schizophrenia and affective
668 – 675. disorders compared with psychiatrically symptom-free surgi-
cal conditions. Arch Gen Psychiatry 36:1295–1301.
Poort R (1945): Catamnestic investigations on manic-depressive
psychoses with special reference to the prognosis. Acta Turvey CL, Coryell WH, Solomon DA, Leon AC, Endicott J,
Psychiatr Scand 20:59 –74. Keller MB, Akiskal H (1999): Long-term prognosis of bipolar
I disorder. Acta Psychiatr Scand 99:110 –119.
Post RM (1992): Transduction of psychosocial stress into the
neurobiology of recurrent affective disorder. Am J Psychiatry Verdoux H, Bourgeois ML (1995): Evolution pronostic des
149:999 –1010. troubles bipolaires. In: Bourgeois ML, Verdoux H, editors.
Les Troubles Bipolaires de l’Humeur. Paris: Masson, 197–
Post RM, Rubinow DR, Ballenger JC (1984): Conditioning, 207.
sensitization, and kindling: Implications for the course of
affective illness. In: Post RM, Ballenger JC, editors. Neuro- Weissman MM, Leaf PJ, Tischler GL, Blazer DG, Karno M,
biology of Mood Disorders. Frontiers of Clinical Neuro- Bruce ML, Florio LP (1988): Affective disorders in five
sciences, Vol 2. Baltimore: Williams & Wilkins, 432– 466. United States communities. Psychol Med 18:141–153.
Post RM, Rubinow DR, Ballenger JC (1986): Conditioning and Wernicke C (1906): Grundriss der Psychiatrie in Klinischen
sensitisation in the longitudinal course of affective illness. Vorlesungen. Leipzig, Germany: Thieme.
Br J Psychiatry 149:191–201. Wertham FI (1929): A group of benign chronic psychoses:
Rehm O (1919): Das Manisch Melancholische Irresein, Monog- Prolonged manic excitements. With a statistical study of age,
raphien aus dem Gesamtgebiete der Neurologie und Psychia- duration and frequency in 2000 manic attacks. Am J Psychi-
trie, Vol 17. Berlin: Springer. atry 9:17–78.
Weygandt W (1899): Uber die Mischzustände des manisch-
Rennie TAC (1942): Prognosis in manic-depressive psychoses.
depressiven Irreseins. Medizinische Fakultät. Würzburg,
Am J Psychiatry 98:801– 814.
Germany: Königliche Julius-Maximilians-Universität.
Ritti A (1879): Folie avec conscience. In: Dechambre A, editor.
Winokur G, Coryell W, Akiskal HS, Endicott J, Keller M,
Dictionnaire Encyclopédique des Sciences Médicales. Paris:
Mueller T (1994): Manic-depressive (bipolar) disorder: The
P. Asselin, G. Masson, 520 –559.
course in light of a prospective ten-year follow-up of 131
Roy-Byrne P, Post RM, Uhde TW, Porcu T, Davos D (1985): patients. Acta Psychiatr Scand 89:102–110.
The longitudinal course of recurrent affective illness: Life
Winokur G, Coryell W, Keller M, Endicott J, Akiskal H (1993):
chart data from research patients at the NIMH. Acta Psychiatr
A prospective follow-up of patients with bipolar and primary
Scand Suppl 317:1–34.
unipolar affective disorder. Arch Gen Psychiatry 50:457–
Saran BM (1969): Lithium. Lancet I:1208 –1209. 465.
Slater E (1938): Zur Periodik des manisch-depressiven Irreseins. Ziehen T (1902): Psychiatrie für Ärzte und Studirende. Leipzig,
Z Ges Neurol Psychiatr 162:794 – 801. Germany: S. Hirzel.
Solomon DA, Keitner GI, Miller IW, Shea MT, Keller MB Ziehen T (1907): Die Erkennung und Behandlung der Melan-
(1995): Course of illness and maintenance treatments for cholie in der Praxis. Halle, Germany: Carl Marhold.
patients with bipolar disorder. J Clin Psychiatry 56:5–13. Zis AP, Goodwin FK (1979): Major affective disorders as a
Stephens JH, McHugh PR (1991): Characteristics and long-term recurrent illness. Arch Gen Psychiatry 36:835– 839.
follow-up of patients hospitalized for mood disorders in the Zis AP, Grof P, Webster M, Goodwin FK (1980): Prediction of
Phipps Clinic, 1913–1940. J Nerv Ment Dis 179:64 –73. relapse in recurrent affective disorder. Psychopharmacol Bull
Till E, Vuckovik S (1970): Ueber den Einfluss der thymoleptis- 16:47– 49.