Otenkiya
2006-09-18 17:11:44 UTC
In the course of my normal job duties, I came across this article on
Science Direct. It's a little long, but hopefully, it will be of
interest to some of you.
The moody one,
-Otenkiya ^.^
I live for breaking the mold.
Efficacy of psychoeducational approaches on bipolar disorders: A review
of the literature
Journal of Affective Disorders, Available online 1 September 2006,
Béatrice Weber Rouget and Jean-Michel Aubry
Abstract
Background
To evaluate the efficacy of psychoeducation in the treatment of previous
bipolar disorder according to specific therapeutic targets such as
treatment compliance, patients' and families' knowledge of the illness
and its treatments, relapse prevention, symptomatic (depressive or
(hypo)manic) phases of the illness or social and occupational functioning.
Methods
A systematic review of the literature published on psychoeducation up to
July 2006 was carried out using the main electronic data bases (Medline,
PubMed). The key words employed included previous bipolar disorder,
psychoeducation, depression, mania, relapse prevention and treatment
compliance.
Results
Although the methodological shortcomings of the early studies must be
taken into account, most data accumulated to date suggest that
psychoeducation, used alone or as a component of more complex
interventions, makes it possible to improve the course of the illness,
notably by increasing the patients' and their families' knowledge of the
disorder and of treatment options, by decreasing the risk of (hypo)manic
or depressive relapse and of hospitalization and by improving treatment
compliance.
Limitations
More studies based solely on psychoeducation, rather than
psychoeducation as part of a multicomponent approach, are needed to
confirm the efficacy of PE reported to date.
Conclusions
Given the results published to date, psychoeducation should be part of
the integrated treatment of bipolar disorder. As a complement to
pharmacotherapy, psychoeducation delivered individually or in a group
setting constitutes a first-line psychological intervention. Applicable
to a majority of patients and their families, it can be delivered by a
wide range of health professionals trained in this approach.
1. Introduction
Bipolar disorder (BD) is characterized by substantial chronicity as well
as an elevated rate of relapse and hospitalization sometimes entailing
dramatic consequences on social and occupational functioning and quality
of life. It is estimated that 75% of individuals suffering from bipolar
illness present at least one relapse over the course of the 5 years
following a manic episode (Gitlin et al., 1995). Moreover, the risk of
suicide is 15 times higher among bipolar patients than in the general
population (Harris and Barraclough, 1997). Disruptions in social
functioning seem to remain between the different episodes of the illness
and several studies show that BD is associated with considerable
residual functional deficits that persist after symptomatic recovery
(Ball et al., 2003, Blairy et al., 2004, Dion et al., 1988, Michalak et
al., 2004, Michalak et al., 2005 and Strakowski et al., 1998). In
parallel to the progress achieved at the level of pharmacological
treatments, psychosocial interventions have seen a remarkable
development, playing an ever more important role in the care of bipolar
disorders (Michalak et al., 2004). In fact, psychosocial factors have a
considerable impact (Bloch et al., 1994) on the variance observed in the
evolution of the course of bipolar illness. Among these factors,
negative life events may precipitate the occurrence of an episode
(Ellicott et al., 1990) or delay time to recovery (Johnson and Miller,
1997). The quality of familial relationships and social support also
plays a role in the evolution of the illness (Johnson et al., 2003,
Miklowitz et al., 1988 and O'Connell et al., 1985). In the field of
psychological approaches to bipolar disorder, psychoeducation (PE)
occupies a key position and appears in recent expert recommendations
such as the Canadian Network for Mood and Anxiety Treatments (Yatham et
al., 2005). Reduced to its most basic level, PE consists in providing
information to the patient about his illness. In a broader sense,
Callahan and Bauer (1999) suggest that “psychoeducation should also help
to foster an alliance whereby the patient becomes an active collaborator
in treatment” and that it is “a mutual process that attempts to improve
a patient's illness management skills through the bidirectional sharing
of relevant information”. Jones (2002) further adds that in order to be
effective, PE must be delivered in a collaborative context and on the
basis of patients understanding of their personal history. This applies
not only to patients but also for their family, significant others and
caregivers. For that matter, we can mention the publication of different
manuals (De Hert et al., 2004 and Jones et al., 2002) that provide not
only patients but also their families with personal accounts and
fundamental information on the illness, its treatments and the best ways
of living with this disease.
Several studies have evaluated the efficacy of PE, either delivered
alone or in combination with other therapeutic approaches in the
framework of more complex interventions such as the Family-Focused
Treatment of Miklowitz and Goldstein (1997), the Interpersonal and
Social Rhythm Therapy of Frank et al. (1994), the Cognitive–Behavioral
Therapy of Lam et al. (1999) or the Life Goals Program of Bauer and
McBride, 1996 and Bauer and McBride, 2003. To date, PE has also been
evaluated in the context of a multicomponent intervention program (Bauer
et al., 2006a, Bauer et al., 2006b and Simon et al., 2006). Among the
pioneer works examining the role of PE in bipolar disorder, those of
Cochran (1984), Peet and Harvey (1991), Harvey and Peet (1991), van Gent
and Zwart (1993) are frequently cited. However, solid evidence in favor
of the efficacy of PE as an adjunctive treatment to pharmacotherapy is
fairly recent since methodological limits, such as non-controlled
studies, small sample sizes and short follow-up periods were present in
frequently mentioned prior studies. When a control group was included,
the latter often received a more limited number of consultations and
less attention than the experimental group. Since then, the studies of
Colom et al., 2003a, Colom et al., 2003b and Colom et al., 2005 are good
illustrations of the efforts that have been made to overcome these
shortcomings. For example, in order to limit the variability induced by
some possible supportive effect of the group sessions themselves, the
control group also met for unstructured peer group meetings (Colom et
al., 2003a). Moreover, in this study, the same psychologists led the PE
experimental group and the more informal support sessions offered in the
control condition.
Ideally, PE is delivered during an euthymic phase (Vieta, 2005) in order
to ensure a better assimilation of the information dispensed, notably as
a prophylactic treatment for relapse. However, while many studies have
taken this approach, some have not (Bauer et al., 1998, Bauer et al.,
2006b, Clarkin et al., 1990, de Andres et al., 2006 and Simon et al.,
2006). In fact, patients suffering from a mild depressive episode can
usually gain benefit from PE as well, but the presence of (hypo)manic
symptoms limits its impact. PE can be proposed in individual or group
format by various health professionals such as physicians, clinical
psychologists, nurses or social workers. It is important that the
caregivers delivering PE have received training that is specific to the
domain but PE should also be accessible in order to guarantee its
widespread use in daily clinical practice (Michalak et al., 2004).
Our review goes back to the very first randomized, controlled study
(Cochran, 1984) carried out with bipolar patients who had recently been
addressed to a lithium clinic. While this study was primarily based on
cognitive–behavioral techniques, certain authors (Michalak et al., 2004
and Parikh et al., 1997) consider it to be a PE intervention based on a
cognitive therapy paradigm. The study concludes that adding a
psychological treatment to medication with the goal of improving patient
adherence was beneficial since this made it possible to reduce relapses
and improve treatment compliance. Notably, the patients in the
experimental group displayed less discontinuity in taking lithium (21%)
than those in the control group (57%). These encouraging results cleared
the way for additional research on the application of protocols for the
psychoeducational treatment of bipolar disorder.
2. Methods
A systematic review of the literature published on psychoeducation up to
July 2006 was carried out using the main electronic data bases (Medline,
PubMed). The key words employed included bipolar disorder,
psychoeducation, depression, mania, relapse prevention and treatment
compliance. Studies evaluating psychoeducation delivered alone or in
combination with other therapeutic approaches were included in our
review. However, when the psychoeducational component of a more complex
therapeutic approach was only minimal and/or not described in a clinical
trial, we did not include this type of study in our review.
Studies in Table 1 are presented chronologically and in all of them,
psychosocial interventions were used in adjunction, rather than in
comparison, to pharmacological treatments. We decided to report as (+) a
positive outcome on a variable (e.g. increased treatment compliance), as
(−) a negative (e.g. increased level of anxiety) and as (empty set) the
absence of effect on a variable (e.g. no difference in hospitalization
rate). Outcomes regarding the efficacy of PE have been reexamined
according to specific therapeutic targets as well as the different
phases of bipolar illness, and described under process measures or
outcome measures. When statistical analyses were lacking, the studies
were still mentioned in our evidence table and indicated with an
asterisk but they were not taken into account in our critical review. It
should also be specified that articles included in our review did not
differentiate between terms of “recurrence” and “relapse”. For that
reason, we decided to only use “relapse”.
2.1. Process measures
2.1.1. Treatment compliance
PE's effect on lithium compliance has been the object of many studies,
starting about 20 years ago. A majority of studies included in our
review suggest that PE or interventions including psychoeducational
components improve compliance, both to lithium (Bauer et al., 1998,
Clarkin et al., 1998, Cochran, 1984, Colom et al., 2003a, Colom et al.,
2005, Harvey and Peet, 1991 and Miklowitz et al., 2003a) and to other
mood stabilizing treatments (Bauer et al., 1998, Clarkin et al., 1998,
Miklowitz et al., 2003a and van Gent and Zwart, 1993). In contrast, a
few studies with patients (Cerbone et al., 1992) or their spouses (van
Gent and Zwart, 1991) concluded that PE or treatments including PE
components did not seem to significantly modify lithium or other mood
stabilizers treatment adherence. Bearing on a large sample (N = 120),
Colom et al. (2003a) showed a difference for lithium but not for the
other mood stabilizers. Doing a post hoc analysis, they further
confirmed that serum levels of lithium may be increased and stabilized
by PE (Colom et al., 2005). These authors suggest that PE not only
enhances treatment adherence but may be considered as a “mood
stabilizer”, mentioning a possible triple and synergistic action of
group PE added to standard pharmacological treatment (helps avoid
relapses, improves adherence and decreases variability in serum levels).
2.1.2. Patient's knowledge and attitude towards the illness and its
treatments
Demographic variables such as sex and age are related to beliefs towards
mental illness (notably about causes of depression) as shown by
Srinivasan et al. (2003) on a sample of subjects with depressive
disorders. These findings need to be taken into account when refining
patient's psychoeducation. Several studies have suggested that PE makes
it possible to improve the patient's knowledge and attitude towards the
illness and its treatments (Bauer et al., 1998, Bauer et al., 2006b,
Brennan, 1995, Colom et al., 1998, Peet and Harvey, 1991 and van Gent
and Zwart, 1991).
2.2. Outcome measures
2.2.1. Symptom reduction and relapse prevention
Although conducted on a limited number of patients (N = 14), the study
of Kripke and Robinson (1985) suggested that PE may improve clinical
status. This was later confirmed by others (Clarkin et al., 1990). In an
open trial, Cerbone et al. (1992) showed that an open-ended group
treatment combining PE with support and counseling decreased duration
and severity of affective episodes during the intervention period.
Regarding PE in the context of broader interventions such as FFT,
Miklowitz et al., 2000 and Miklowitz et al., 2003b have shown that this
approach may improve depressive, but not manic symptoms. Nevertheless,
other studies by the same group of authors have shown improvement of
both manic and depressive symptomatology (Miklowitz et al., 2003a and
Miklowitz et al., 2004). More recently, Simon et al. (2006) also found a
clinical status improvement, but mainly regarding manic symptomatology,
with lower mean of mania ratings as well as less time in hypomanic or
manic episode. They also reported no overall effect on depression
symptomatology across the first 12 months of follow-up (Simon et al.,
2005), which was confirmed by analysis of the complete data (Simon et
al., 2006). According to Simon et al., early recognition and early
intervention for mood episodes might have a greater effect on preventing
episodic mania than on relieving more chronic depression. In a study
evaluating cognitive therapy with a psychoeducational component, Ball et
al. (2006) found less severe depression scores and less dysfunctional
attitudes after 6 months treatment. But at 12 months follow-up, the
cognitive therapy group showed only a trend toward lower hypomanic or
manic scores than the treatment as usual group, suggesting that booster
sessions may be crucial for maintaining the beneficial effects of the
therapy. In a 3 year multisite randomized study, Bauer et al., 2006b and
Bauer et al., 2006a addressed whether a collaborative model for chronic
care would have a positive impact on clinical outcome, function and
cost. With group psychoeducation as part of their approach, they found
that the intervention significantly reduced weeks in affective episode,
primarily mania. Finally, an open study on the impact of the structured
Life Goals Program (Bauer and McBride, 1996 and Bauer and McBride,
2003), showed that a psychoeducational approach allows to improve mood
stability as measured by the MADRS, as well as relapse prevention
according to participants self report (de Andres et al., 2006).
2.2.2. Prevention of depressive episodes
Numerous studies have evaluated relapse prevention, by teaching patients
to better identify prodromal symptoms and to develop an action plan for
response when the prodromes first appear.
In a well designed randomized-controlled study, Colom et al. (2003a)
proposed 21 sessions in group format of either PE or non-structured
meetings (control group) as an add-on treatment to standard psychiatric
care (plus non-structured meetings to avoid the placebo effect of
meeting with the group) to 120 bipolar patients. At the end of the
6-month treatment phase and at the 2-year follow-up, PE proved to be a
significantly more effective prophylactic treatment than the control
intervention, for both manic and depressive relapses (at 6 months, 60%
of the control subjects met the relapse criteria compared to 38% of
subjects in the experimental group, these rates evolving into 92% and
67% respectively at the 2-year follow-up) and the average length of
hospitalizations per patient at the 2-year follow-up (48.66 days for the
controls versus 24.18 days for the subjects of the experimental group).
The same year, this research team published another study (Colom et al.,
2003b) involving a more limited number of bipolar subjects in remission
(N = 50) receiving standard pharmacological care. This study showed that
even subjects already presenting a good treatment compliance could
benefit from PE. In fact, out of this patient sample already
demonstrating good treatment adherence, 77% of the control subjects,
compared to 46.7% of the patients benefiting from PE met the criteria
for relapse (hypomanic, manic, mixed or depressive) during the course of
treatment. These figures evolve into 92% and 60% respectively for
controls versus subjects in the PE group at the end of the 2 years of
follow-up.
These positive effects were previously not found in a very small sized
pilot study (N = 10) and quoted in a review from 1998 (Colom et al.,
1998). No effect for PE on treatment compliance or on the rate of
relapse was reported. This may be due both to the very limited number of
patients included and to the fact that PE sessions were only delivered
on a monthly basis. Moreover, the authors mention that when excluding
patients with comorbid personality disorders, results almost reached
statistical significance. Among other studies conducted on this topic,
the one from Perry et al. (1999) mentions a positive effect on time to
manic relapse and overall days manic, but this was not the case for
depressive relapse and overall days depressed. Given this observation of
a higher efficacy of PE in the presence of manic prodromes, Perry et al.
suggest that warnings signs of manic relapses are easier to identify
than depressive ones. Moreover, depression has proved more resistant to
pharmacological treatment than mania. Our impression based on clinical
experience support this conclusion of poorer efficacy of treatments,
whether psychoeducational or pharmacological, in case of depression.
2.2.3. Prevention of (hypo)manic episodes
The study of Perry et al. (1999) was one of the first to demonstrate the
positive effect of PE on both the time interval before the next relapse
and on the frequency of these relapses. As previously mentioned, the
studies of Colom et al., 2003a and Colom et al., 2003b also showed that
patients benefiting from PE present a lower rate of relapse (hypomanic,
manic, mixed or depressive) than the control group, both at the end of
the treatment phase and at the 24-month follow-up.
2.2.4. Hospitalizations
This topic was already addressed in the first mentioned study of Cochran
(1984), showing lower rates of hospitalization at 6 months follow-up.
The following years, further data suggested that PE had a positive
effect on the length and/or number of hospitalizations (Cerbone et al.,
1992, Colom et al., 2003a, Colom and Vieta, 2004, Honig et al., 1997,
Rea et al., 2003 and van Gent and Zwart, 1993).
2.2.5. Functional outcome and quality of life
Among the studies cited in our review, several report functional
improvement with PE or treatments involving psychoeducational components
(Bauer et al., 1998, Bauer et al., 2006b, Cerbone et al., 1992, Clarkin
et al., 1990, Clarkin et al., 1998, Foelker et al., 1986, Kripke and
Robinson, 1985 and Perry et al., 1999). Moreover, a recent study
(Michalak et al., 2005) examining the impact of PE dispensed to patients
in euthymic phase or with very few symptoms, with the goal of improving
their subjective quality of life, showed that although quality of life
remains unsatisfactory, it nevertheless displays some improvement. This
result confirms prior conclusions of van Gent et al. (1988). In a study
including a mixed group of inpatients with mood or schizophrenic
spectrum disorders, Clarkin et al. (1990) showed that manic depressive
patients improved their global ratings, symptoms, social role functions
and family attitudes. Interestingly, beneficial effects in bipolar
patients were solely due to improvement in women and the psychosocial
intervention seemed to worsen patients with major depression.
2.2.6. Bipolar disorder with psychiatric comorbidity (borderline
personality disorder, substance abuse)
Comorbid disorders are associated with higher complexity in treatment
and poorer therapeutic outcomes. They must therefore be carefully
considered when assessing bipolar patients for treatment (Jones et al.,
2005). As previously seen, PE may also prove efficacious in the case of
patients commonly considered as “difficult to treat,” such as those
presenting comorbid personality disorders (Colom et al., 2004). In their
study, these authors have shown that the group that benefited from PE
had a lower rate of relapse (67%) than the control group (100%) at the
end of 24 months of follow-up. What is more, PE also increased the
symptom-free interval between episodes and reduced the total number of
relapses (manic or depressive). The average length of hospitalization
was also longer in the control group, although the study did not
demonstrate a significant difference between the 2 groups for the
percentage of patients requiring hospitalization.
Knowing that among axis I disorders, bipolar illness has one of the
highest risk for coexisting substance use, a combined approach including
elements of PE for bipolar substance abusers (Weiss et al., 1999, Weiss
et al., 2004 and Weiss et al., 2005) has also been developed. This
integrated group therapy, emphasizing primarily cognitive–behavioral
strategies, consists of 20 weekly one hour group sessions, each of which
is dedicated to a specific and relevant common disorder topic. In a open
study bearing on a limited sample of patients (N = 45) among which 21
received this intervention focusing simultaneously on bipolar disorder
and substance dependence, Weiss et al. (2000) report better outcomes in
terms of lower scores on a measure of drug-related problem severity and
more months of abstinence from both drug and alcohol for the
experimental group. Moreover, although differences between the
experimental and the control group were found in this trial in terms of
greater improvement in manic symptoms and in severity of alcohol-related
problems, this was no longer true after adjustment for age
(significantly older integrated group therapy cohort). In a later
article concerning the same larger study pilot-testing this integrated
group therapy but addressing new research questions, Weiss et al. (2004)
show that the experimental group reported fewer days of drug use over
the 6 month study period compared to the control group. This difference
was not significant among patients without perceived substance-induced
improvement in bipolar disorder symptoms, which suggests that integrated
group treatment only improves patients who report that substances use
improves their bipolar illness (self-medication hypothesis). Potential
inaccuracy of self-reports regarding the impact of substance use on mood
may be considered as a limitation of this study. Further research needs
to be conducted on this topic, in order to seriously take into account
this frequent comorbid condition and to test whether psychological
interventions addressed to bipolar patients with comorbid substance use
are only efficacious in the case of patients using drugs of abuse in
response to bipolar disorder symptoms.
2.2.7. Interventions for the family (and caregivers)
When dispensed to relatives (family, spouses), PE increases their
knowledge of the illness, its treatments and social support strategies
(Brennan, 1995, Honig et al., 1997 and van Gent and Zwart, 1991). PE for
family members and caregivers of bipolar patients makes it possible to
alleviate the burden of the illness. The more information on this
psychic illness caregivers and family members disposed of, the less they
were inclined to hold the patient responsible for the genuine burden
tied to this pathology (financial and family problems) (Reinares et al.,
2002). What is more, the improvement obtained through PE in the support
provided by the family to the patient resulted in a more positive
clinical outcome (Rea et al., 2003). In fact, when family members had
benefited from PE, patients presented a lower rate of relapse, longer
time intervals between episodes, a better treatment adherence and a
reduction in symptoms.
2.2.8. Suicide prevention
Although suicide is of major concern when treating bipolar patients,
this topic has rarely been directly addressed through PE. This lack of
data on suicide prevention may possibly be explained by the sample
needed to find results on this topic. In bipolar patients at high risk
of suicide, augmentation of pharmacotherapy with PE as a component of
wider psychological interventions, including interpersonal therapy plus
daily routines regulation, made it possible to reduce suicidal behavior
(Rucci et al., 2002). Enhancing medication adherence because of the
protective effects of pharmacotherapy (namely lithium) against suicide
(Goodwin et al., 2003) should be a major focus of PE.
3. Discussion
Although to date few reviews (Bauer and McBride, 1996, Bauer and
McBride, 2003, Colom et al., 2003b and Michalak et al., 2004) have been
published concerning the role of PE in bipolar disorders, to our
knowledge, this review is the first to examine its efficacy by clearly
distinguishing between different therapeutic targets. In fact, most
reviews (Bauer, 2002, Huxley et al., 2000, Michalak et al., 2004, Otto
et al., 2003, Scott, 2004 and Swartz and Frank, 2001) summarize the
conclusions of studies related to psychological interventions by
classifying them according to the type of approach (psychoeducation,
cognitive–behavioral therapy, interpersonal therapy…). Regarding our
review, it should be underlined that in several cited studies, PE is one
of the multiple components integrated in psychosocial approaches. And
even when clearly stated as PE, treatments delivered with a large number
of sessions probably also imply other therapeutic elements. Therefore,
it is often difficult to identify clearly which therapeutic effects are
due solely to PE and which ones are due to other psychosocial approaches
or to both. This issue cannot be really sorted out since the conceptual
bases of the various interventions overlap, at least partially.
Although the efficacy of PE approaches in the context of bipolar
disorder has been demonstrated, certain key questions remain to be
answered. Among them, the phase of the illness during which PE might be
efficacious is still to be determined. Data are insufficient or absent
for PE delivered in the presence of mixed, hypomanic symptoms, or rapid
cycling. Some authors (Swartz and Frank, 2001) have suggested adapting
the intervention in function of the patient's mood state. With this
strategy of treatment that is specific to a particular phase of the
illness, different psychotherapeutic approaches are proposed
sequentially according to the patient's mood. However, no evidence
exists to this date to support this type of intervention.
We have found no trial determining the impact of PE on the different
subtypes of bipolar disorder, many studies including both bipolar type I
and II patients. There also is a lack of information concerning PE
efficacy according to the age of the patients (adult, older adults,
adolescents). In fact, our review includes only one study of a geriatric
population (Foelker et al., 1986) and one of adolescents (Miklowitz et
al., 2004).
Concerning the ideal context of care for delivering PE, a majority of
authors recommend outpatient settings, as psychoeducation should mainly
be performed when patients in euthymic state. Research trials to date
have primarily been carried out in ambulatory follow-up settings. Some
studies have involved hospitalized patients. Sometimes, although PE was
carried out only after discharge, patients were recruited during
hospitalization. The question of whether PE produces benefits when
dispensed after a first episode also remains unresolved (Zaretsky,
2003). Patients who have gone through multiple episodes of depression
and mania are less susceptible to respond to treatment and they often
develop complications in the form of medical comorbidity, cognitive
limitations, as well as psychosocial deficits. Introducing interventions
with specific targets earlier when the disorder first appears would
probably enhance the acquisition of illness management coping skills in
order to avoid long-lasting consequences. Similarly, the impact of PE on
recently diagnosed patients, as compared to those who have already been
recognized as bipolar for several years, should be examined. This issue
is of major interest as patients often complain that they only received
information about their disorder long after the occurrence of the first
symptoms and after having finally been correctly diagnosed (Hirschfeld
and Vornik, 2004).
Regarding treatment of acute depressive, (hypo)manic or mixed phases, if
several psychosocial interventions have been attempted, none of them
have been retained as efficacious treatments for manic, hypomanic or
mixed phases of bipolar disorder (Colom and Vieta, 2004 and Swartz and
Frank, 2001). Parikh et al. (1997) suggest that support, combined with
more behavioral techniques aiming to reduce stimuli and daily
activities, is more advisable during manic phases. Brief interventions
based on this type of specific technique, proposed for isolated manic
symptoms, are thought to constitute an efficacious method to prevent the
evolution towards an actual manic episode. These authors emphasize that
maintaining a therapeutic alliance must remain the primary goal of the
follow-up of bipolar patients over the course of the various phases of
the illness, with resort to the principles of supportive therapy in
periods of more severe symptomatic phases.
Concerning the goal of improving functional outcome in bipolar disorder
through PE, only little attention has been devoted to treatments
addressing directly the issue of social and occupational status, of
subjective quality of life in this affection, in comparison to
schizophrenia, as mentioned by Bauer and McBride, 1996 and Bauer and
McBride, 2003. Although PE may have positive effects on global
functioning and quality of life, data regarding this issue are too
sparse and inconclusive to date.
Insufficient clues are available to clarify the mechanisms of action of
psychological interventions or the specific approach that should be
preferred in the context of bipolar disorder (Scott and Todd, 2002 and
Scott, 2004). Given the sparse empirical data, the choice of one
technique over another should be based on clinical judgment at this
stage, notably depending on the difficulties or specificities of the
patient (Scott and Gutierrez, 2004). All bipolar patients do not have
the same capacity to receive psychological “input” due to differences in
cognition, personality or even family context (Jones et al., 2005). As a
result, each individual should be evaluated to determine specific
treatment needs. Jones et al. (2005) suggest that when the patient
presents good levels of introspection and commitment as well as a
relatively mild symptomatology, PE probably constitutes the first-line
psychological intervention. As such, it is strongly recommended to
integrate PE in the basic clinical practice of psychiatrists or any
other member of the healthcare team, in order to allow as many patients
as possible to benefit from this treatment. However, in presence of a
patient with limited levels of commitment and introspection or when high
number of relapse risk factors are present, these authors believe that
more structured psychoeducational approaches, or even group
cognitive–behavioral interventions, are worth considering. They can be
dispensed by clinicians with specific but limited psychotherapeutic
training who conform to structured treatment protocols. Intensive
psychological interventions are all the more pertinent as the subject's
clinical picture becomes more complex or the history of relapse
lengthens. These interventions require clinicians who have been well
trained in the cognitive–behavioral approach, in individual or group format.
Finally, the optimal protocol for applying PE remains to be determined,
notably regarding the length of treatment (Jones, 2002) or the setting
(individual or group). Easier to dispense, PE constitutes a serious
alternative to more traditional psychotherapeutic approaches such as
cognitive–behavioral therapy and family focused therapy (FFT).
In fact, the common sense and apparent simplicity of PE seem to indicate
that it has no counter-indications other than that of the patient being
in an acute phase or having cognitive deficits. However, as Vieta (2005)
points out, it may not prove beneficial for certain individuals and may
even produce unwanted effects. For example, in patients with
obsessive–compulsive personality, PE may lead to an excessive
preoccupation with the detection of early relapse signs, unreasonably
increasing the number of medical consultations or the prescription of
medication. One study in our review also showed that PE increased
patients level of anxiety (van Gent and Zwart, 1991). Finally, as
suggested in a recent study of cognitive–behavioral therapy (Scott,
2004), patients who are still symptomatic and who have previously
presented an elevated number of episodes may not be helped by this type
of psychological intervention and even see their mood state worsen.
In summary, resorting to PE in the treatment of bipolar patients is
highly recommended. As Vieta points out, PE functions as an adjunctive
treatment for bipolar disorder and will not work as a monotherapy
without medications (Vieta, 2005). It is also important to point out
that psychoeducation is part of most other psychosocial interventions.
Patient preference, as well as the importance of including persons close
to the patients in the treatment process and the availability of trained
therapists (Colom et al., 2005 and Scott and Colom, 2005), can dictate
the choice of the intervention technique. Several multicenter studies
related to the evaluation of PE for bipolar disorder are currently
underway. These studies should make it possible to reach a conclusion on
the true impact, underestimated for many years, of this psychological
intervention in the treatment of bipolar patients.
Science Direct. It's a little long, but hopefully, it will be of
interest to some of you.
The moody one,
-Otenkiya ^.^
I live for breaking the mold.
Efficacy of psychoeducational approaches on bipolar disorders: A review
of the literature
Journal of Affective Disorders, Available online 1 September 2006,
Béatrice Weber Rouget and Jean-Michel Aubry
Abstract
Background
To evaluate the efficacy of psychoeducation in the treatment of previous
bipolar disorder according to specific therapeutic targets such as
treatment compliance, patients' and families' knowledge of the illness
and its treatments, relapse prevention, symptomatic (depressive or
(hypo)manic) phases of the illness or social and occupational functioning.
Methods
A systematic review of the literature published on psychoeducation up to
July 2006 was carried out using the main electronic data bases (Medline,
PubMed). The key words employed included previous bipolar disorder,
psychoeducation, depression, mania, relapse prevention and treatment
compliance.
Results
Although the methodological shortcomings of the early studies must be
taken into account, most data accumulated to date suggest that
psychoeducation, used alone or as a component of more complex
interventions, makes it possible to improve the course of the illness,
notably by increasing the patients' and their families' knowledge of the
disorder and of treatment options, by decreasing the risk of (hypo)manic
or depressive relapse and of hospitalization and by improving treatment
compliance.
Limitations
More studies based solely on psychoeducation, rather than
psychoeducation as part of a multicomponent approach, are needed to
confirm the efficacy of PE reported to date.
Conclusions
Given the results published to date, psychoeducation should be part of
the integrated treatment of bipolar disorder. As a complement to
pharmacotherapy, psychoeducation delivered individually or in a group
setting constitutes a first-line psychological intervention. Applicable
to a majority of patients and their families, it can be delivered by a
wide range of health professionals trained in this approach.
1. Introduction
Bipolar disorder (BD) is characterized by substantial chronicity as well
as an elevated rate of relapse and hospitalization sometimes entailing
dramatic consequences on social and occupational functioning and quality
of life. It is estimated that 75% of individuals suffering from bipolar
illness present at least one relapse over the course of the 5 years
following a manic episode (Gitlin et al., 1995). Moreover, the risk of
suicide is 15 times higher among bipolar patients than in the general
population (Harris and Barraclough, 1997). Disruptions in social
functioning seem to remain between the different episodes of the illness
and several studies show that BD is associated with considerable
residual functional deficits that persist after symptomatic recovery
(Ball et al., 2003, Blairy et al., 2004, Dion et al., 1988, Michalak et
al., 2004, Michalak et al., 2005 and Strakowski et al., 1998). In
parallel to the progress achieved at the level of pharmacological
treatments, psychosocial interventions have seen a remarkable
development, playing an ever more important role in the care of bipolar
disorders (Michalak et al., 2004). In fact, psychosocial factors have a
considerable impact (Bloch et al., 1994) on the variance observed in the
evolution of the course of bipolar illness. Among these factors,
negative life events may precipitate the occurrence of an episode
(Ellicott et al., 1990) or delay time to recovery (Johnson and Miller,
1997). The quality of familial relationships and social support also
plays a role in the evolution of the illness (Johnson et al., 2003,
Miklowitz et al., 1988 and O'Connell et al., 1985). In the field of
psychological approaches to bipolar disorder, psychoeducation (PE)
occupies a key position and appears in recent expert recommendations
such as the Canadian Network for Mood and Anxiety Treatments (Yatham et
al., 2005). Reduced to its most basic level, PE consists in providing
information to the patient about his illness. In a broader sense,
Callahan and Bauer (1999) suggest that “psychoeducation should also help
to foster an alliance whereby the patient becomes an active collaborator
in treatment” and that it is “a mutual process that attempts to improve
a patient's illness management skills through the bidirectional sharing
of relevant information”. Jones (2002) further adds that in order to be
effective, PE must be delivered in a collaborative context and on the
basis of patients understanding of their personal history. This applies
not only to patients but also for their family, significant others and
caregivers. For that matter, we can mention the publication of different
manuals (De Hert et al., 2004 and Jones et al., 2002) that provide not
only patients but also their families with personal accounts and
fundamental information on the illness, its treatments and the best ways
of living with this disease.
Several studies have evaluated the efficacy of PE, either delivered
alone or in combination with other therapeutic approaches in the
framework of more complex interventions such as the Family-Focused
Treatment of Miklowitz and Goldstein (1997), the Interpersonal and
Social Rhythm Therapy of Frank et al. (1994), the Cognitive–Behavioral
Therapy of Lam et al. (1999) or the Life Goals Program of Bauer and
McBride, 1996 and Bauer and McBride, 2003. To date, PE has also been
evaluated in the context of a multicomponent intervention program (Bauer
et al., 2006a, Bauer et al., 2006b and Simon et al., 2006). Among the
pioneer works examining the role of PE in bipolar disorder, those of
Cochran (1984), Peet and Harvey (1991), Harvey and Peet (1991), van Gent
and Zwart (1993) are frequently cited. However, solid evidence in favor
of the efficacy of PE as an adjunctive treatment to pharmacotherapy is
fairly recent since methodological limits, such as non-controlled
studies, small sample sizes and short follow-up periods were present in
frequently mentioned prior studies. When a control group was included,
the latter often received a more limited number of consultations and
less attention than the experimental group. Since then, the studies of
Colom et al., 2003a, Colom et al., 2003b and Colom et al., 2005 are good
illustrations of the efforts that have been made to overcome these
shortcomings. For example, in order to limit the variability induced by
some possible supportive effect of the group sessions themselves, the
control group also met for unstructured peer group meetings (Colom et
al., 2003a). Moreover, in this study, the same psychologists led the PE
experimental group and the more informal support sessions offered in the
control condition.
Ideally, PE is delivered during an euthymic phase (Vieta, 2005) in order
to ensure a better assimilation of the information dispensed, notably as
a prophylactic treatment for relapse. However, while many studies have
taken this approach, some have not (Bauer et al., 1998, Bauer et al.,
2006b, Clarkin et al., 1990, de Andres et al., 2006 and Simon et al.,
2006). In fact, patients suffering from a mild depressive episode can
usually gain benefit from PE as well, but the presence of (hypo)manic
symptoms limits its impact. PE can be proposed in individual or group
format by various health professionals such as physicians, clinical
psychologists, nurses or social workers. It is important that the
caregivers delivering PE have received training that is specific to the
domain but PE should also be accessible in order to guarantee its
widespread use in daily clinical practice (Michalak et al., 2004).
Our review goes back to the very first randomized, controlled study
(Cochran, 1984) carried out with bipolar patients who had recently been
addressed to a lithium clinic. While this study was primarily based on
cognitive–behavioral techniques, certain authors (Michalak et al., 2004
and Parikh et al., 1997) consider it to be a PE intervention based on a
cognitive therapy paradigm. The study concludes that adding a
psychological treatment to medication with the goal of improving patient
adherence was beneficial since this made it possible to reduce relapses
and improve treatment compliance. Notably, the patients in the
experimental group displayed less discontinuity in taking lithium (21%)
than those in the control group (57%). These encouraging results cleared
the way for additional research on the application of protocols for the
psychoeducational treatment of bipolar disorder.
2. Methods
A systematic review of the literature published on psychoeducation up to
July 2006 was carried out using the main electronic data bases (Medline,
PubMed). The key words employed included bipolar disorder,
psychoeducation, depression, mania, relapse prevention and treatment
compliance. Studies evaluating psychoeducation delivered alone or in
combination with other therapeutic approaches were included in our
review. However, when the psychoeducational component of a more complex
therapeutic approach was only minimal and/or not described in a clinical
trial, we did not include this type of study in our review.
Studies in Table 1 are presented chronologically and in all of them,
psychosocial interventions were used in adjunction, rather than in
comparison, to pharmacological treatments. We decided to report as (+) a
positive outcome on a variable (e.g. increased treatment compliance), as
(−) a negative (e.g. increased level of anxiety) and as (empty set) the
absence of effect on a variable (e.g. no difference in hospitalization
rate). Outcomes regarding the efficacy of PE have been reexamined
according to specific therapeutic targets as well as the different
phases of bipolar illness, and described under process measures or
outcome measures. When statistical analyses were lacking, the studies
were still mentioned in our evidence table and indicated with an
asterisk but they were not taken into account in our critical review. It
should also be specified that articles included in our review did not
differentiate between terms of “recurrence” and “relapse”. For that
reason, we decided to only use “relapse”.
2.1. Process measures
2.1.1. Treatment compliance
PE's effect on lithium compliance has been the object of many studies,
starting about 20 years ago. A majority of studies included in our
review suggest that PE or interventions including psychoeducational
components improve compliance, both to lithium (Bauer et al., 1998,
Clarkin et al., 1998, Cochran, 1984, Colom et al., 2003a, Colom et al.,
2005, Harvey and Peet, 1991 and Miklowitz et al., 2003a) and to other
mood stabilizing treatments (Bauer et al., 1998, Clarkin et al., 1998,
Miklowitz et al., 2003a and van Gent and Zwart, 1993). In contrast, a
few studies with patients (Cerbone et al., 1992) or their spouses (van
Gent and Zwart, 1991) concluded that PE or treatments including PE
components did not seem to significantly modify lithium or other mood
stabilizers treatment adherence. Bearing on a large sample (N = 120),
Colom et al. (2003a) showed a difference for lithium but not for the
other mood stabilizers. Doing a post hoc analysis, they further
confirmed that serum levels of lithium may be increased and stabilized
by PE (Colom et al., 2005). These authors suggest that PE not only
enhances treatment adherence but may be considered as a “mood
stabilizer”, mentioning a possible triple and synergistic action of
group PE added to standard pharmacological treatment (helps avoid
relapses, improves adherence and decreases variability in serum levels).
2.1.2. Patient's knowledge and attitude towards the illness and its
treatments
Demographic variables such as sex and age are related to beliefs towards
mental illness (notably about causes of depression) as shown by
Srinivasan et al. (2003) on a sample of subjects with depressive
disorders. These findings need to be taken into account when refining
patient's psychoeducation. Several studies have suggested that PE makes
it possible to improve the patient's knowledge and attitude towards the
illness and its treatments (Bauer et al., 1998, Bauer et al., 2006b,
Brennan, 1995, Colom et al., 1998, Peet and Harvey, 1991 and van Gent
and Zwart, 1991).
2.2. Outcome measures
2.2.1. Symptom reduction and relapse prevention
Although conducted on a limited number of patients (N = 14), the study
of Kripke and Robinson (1985) suggested that PE may improve clinical
status. This was later confirmed by others (Clarkin et al., 1990). In an
open trial, Cerbone et al. (1992) showed that an open-ended group
treatment combining PE with support and counseling decreased duration
and severity of affective episodes during the intervention period.
Regarding PE in the context of broader interventions such as FFT,
Miklowitz et al., 2000 and Miklowitz et al., 2003b have shown that this
approach may improve depressive, but not manic symptoms. Nevertheless,
other studies by the same group of authors have shown improvement of
both manic and depressive symptomatology (Miklowitz et al., 2003a and
Miklowitz et al., 2004). More recently, Simon et al. (2006) also found a
clinical status improvement, but mainly regarding manic symptomatology,
with lower mean of mania ratings as well as less time in hypomanic or
manic episode. They also reported no overall effect on depression
symptomatology across the first 12 months of follow-up (Simon et al.,
2005), which was confirmed by analysis of the complete data (Simon et
al., 2006). According to Simon et al., early recognition and early
intervention for mood episodes might have a greater effect on preventing
episodic mania than on relieving more chronic depression. In a study
evaluating cognitive therapy with a psychoeducational component, Ball et
al. (2006) found less severe depression scores and less dysfunctional
attitudes after 6 months treatment. But at 12 months follow-up, the
cognitive therapy group showed only a trend toward lower hypomanic or
manic scores than the treatment as usual group, suggesting that booster
sessions may be crucial for maintaining the beneficial effects of the
therapy. In a 3 year multisite randomized study, Bauer et al., 2006b and
Bauer et al., 2006a addressed whether a collaborative model for chronic
care would have a positive impact on clinical outcome, function and
cost. With group psychoeducation as part of their approach, they found
that the intervention significantly reduced weeks in affective episode,
primarily mania. Finally, an open study on the impact of the structured
Life Goals Program (Bauer and McBride, 1996 and Bauer and McBride,
2003), showed that a psychoeducational approach allows to improve mood
stability as measured by the MADRS, as well as relapse prevention
according to participants self report (de Andres et al., 2006).
2.2.2. Prevention of depressive episodes
Numerous studies have evaluated relapse prevention, by teaching patients
to better identify prodromal symptoms and to develop an action plan for
response when the prodromes first appear.
In a well designed randomized-controlled study, Colom et al. (2003a)
proposed 21 sessions in group format of either PE or non-structured
meetings (control group) as an add-on treatment to standard psychiatric
care (plus non-structured meetings to avoid the placebo effect of
meeting with the group) to 120 bipolar patients. At the end of the
6-month treatment phase and at the 2-year follow-up, PE proved to be a
significantly more effective prophylactic treatment than the control
intervention, for both manic and depressive relapses (at 6 months, 60%
of the control subjects met the relapse criteria compared to 38% of
subjects in the experimental group, these rates evolving into 92% and
67% respectively at the 2-year follow-up) and the average length of
hospitalizations per patient at the 2-year follow-up (48.66 days for the
controls versus 24.18 days for the subjects of the experimental group).
The same year, this research team published another study (Colom et al.,
2003b) involving a more limited number of bipolar subjects in remission
(N = 50) receiving standard pharmacological care. This study showed that
even subjects already presenting a good treatment compliance could
benefit from PE. In fact, out of this patient sample already
demonstrating good treatment adherence, 77% of the control subjects,
compared to 46.7% of the patients benefiting from PE met the criteria
for relapse (hypomanic, manic, mixed or depressive) during the course of
treatment. These figures evolve into 92% and 60% respectively for
controls versus subjects in the PE group at the end of the 2 years of
follow-up.
These positive effects were previously not found in a very small sized
pilot study (N = 10) and quoted in a review from 1998 (Colom et al.,
1998). No effect for PE on treatment compliance or on the rate of
relapse was reported. This may be due both to the very limited number of
patients included and to the fact that PE sessions were only delivered
on a monthly basis. Moreover, the authors mention that when excluding
patients with comorbid personality disorders, results almost reached
statistical significance. Among other studies conducted on this topic,
the one from Perry et al. (1999) mentions a positive effect on time to
manic relapse and overall days manic, but this was not the case for
depressive relapse and overall days depressed. Given this observation of
a higher efficacy of PE in the presence of manic prodromes, Perry et al.
suggest that warnings signs of manic relapses are easier to identify
than depressive ones. Moreover, depression has proved more resistant to
pharmacological treatment than mania. Our impression based on clinical
experience support this conclusion of poorer efficacy of treatments,
whether psychoeducational or pharmacological, in case of depression.
2.2.3. Prevention of (hypo)manic episodes
The study of Perry et al. (1999) was one of the first to demonstrate the
positive effect of PE on both the time interval before the next relapse
and on the frequency of these relapses. As previously mentioned, the
studies of Colom et al., 2003a and Colom et al., 2003b also showed that
patients benefiting from PE present a lower rate of relapse (hypomanic,
manic, mixed or depressive) than the control group, both at the end of
the treatment phase and at the 24-month follow-up.
2.2.4. Hospitalizations
This topic was already addressed in the first mentioned study of Cochran
(1984), showing lower rates of hospitalization at 6 months follow-up.
The following years, further data suggested that PE had a positive
effect on the length and/or number of hospitalizations (Cerbone et al.,
1992, Colom et al., 2003a, Colom and Vieta, 2004, Honig et al., 1997,
Rea et al., 2003 and van Gent and Zwart, 1993).
2.2.5. Functional outcome and quality of life
Among the studies cited in our review, several report functional
improvement with PE or treatments involving psychoeducational components
(Bauer et al., 1998, Bauer et al., 2006b, Cerbone et al., 1992, Clarkin
et al., 1990, Clarkin et al., 1998, Foelker et al., 1986, Kripke and
Robinson, 1985 and Perry et al., 1999). Moreover, a recent study
(Michalak et al., 2005) examining the impact of PE dispensed to patients
in euthymic phase or with very few symptoms, with the goal of improving
their subjective quality of life, showed that although quality of life
remains unsatisfactory, it nevertheless displays some improvement. This
result confirms prior conclusions of van Gent et al. (1988). In a study
including a mixed group of inpatients with mood or schizophrenic
spectrum disorders, Clarkin et al. (1990) showed that manic depressive
patients improved their global ratings, symptoms, social role functions
and family attitudes. Interestingly, beneficial effects in bipolar
patients were solely due to improvement in women and the psychosocial
intervention seemed to worsen patients with major depression.
2.2.6. Bipolar disorder with psychiatric comorbidity (borderline
personality disorder, substance abuse)
Comorbid disorders are associated with higher complexity in treatment
and poorer therapeutic outcomes. They must therefore be carefully
considered when assessing bipolar patients for treatment (Jones et al.,
2005). As previously seen, PE may also prove efficacious in the case of
patients commonly considered as “difficult to treat,” such as those
presenting comorbid personality disorders (Colom et al., 2004). In their
study, these authors have shown that the group that benefited from PE
had a lower rate of relapse (67%) than the control group (100%) at the
end of 24 months of follow-up. What is more, PE also increased the
symptom-free interval between episodes and reduced the total number of
relapses (manic or depressive). The average length of hospitalization
was also longer in the control group, although the study did not
demonstrate a significant difference between the 2 groups for the
percentage of patients requiring hospitalization.
Knowing that among axis I disorders, bipolar illness has one of the
highest risk for coexisting substance use, a combined approach including
elements of PE for bipolar substance abusers (Weiss et al., 1999, Weiss
et al., 2004 and Weiss et al., 2005) has also been developed. This
integrated group therapy, emphasizing primarily cognitive–behavioral
strategies, consists of 20 weekly one hour group sessions, each of which
is dedicated to a specific and relevant common disorder topic. In a open
study bearing on a limited sample of patients (N = 45) among which 21
received this intervention focusing simultaneously on bipolar disorder
and substance dependence, Weiss et al. (2000) report better outcomes in
terms of lower scores on a measure of drug-related problem severity and
more months of abstinence from both drug and alcohol for the
experimental group. Moreover, although differences between the
experimental and the control group were found in this trial in terms of
greater improvement in manic symptoms and in severity of alcohol-related
problems, this was no longer true after adjustment for age
(significantly older integrated group therapy cohort). In a later
article concerning the same larger study pilot-testing this integrated
group therapy but addressing new research questions, Weiss et al. (2004)
show that the experimental group reported fewer days of drug use over
the 6 month study period compared to the control group. This difference
was not significant among patients without perceived substance-induced
improvement in bipolar disorder symptoms, which suggests that integrated
group treatment only improves patients who report that substances use
improves their bipolar illness (self-medication hypothesis). Potential
inaccuracy of self-reports regarding the impact of substance use on mood
may be considered as a limitation of this study. Further research needs
to be conducted on this topic, in order to seriously take into account
this frequent comorbid condition and to test whether psychological
interventions addressed to bipolar patients with comorbid substance use
are only efficacious in the case of patients using drugs of abuse in
response to bipolar disorder symptoms.
2.2.7. Interventions for the family (and caregivers)
When dispensed to relatives (family, spouses), PE increases their
knowledge of the illness, its treatments and social support strategies
(Brennan, 1995, Honig et al., 1997 and van Gent and Zwart, 1991). PE for
family members and caregivers of bipolar patients makes it possible to
alleviate the burden of the illness. The more information on this
psychic illness caregivers and family members disposed of, the less they
were inclined to hold the patient responsible for the genuine burden
tied to this pathology (financial and family problems) (Reinares et al.,
2002). What is more, the improvement obtained through PE in the support
provided by the family to the patient resulted in a more positive
clinical outcome (Rea et al., 2003). In fact, when family members had
benefited from PE, patients presented a lower rate of relapse, longer
time intervals between episodes, a better treatment adherence and a
reduction in symptoms.
2.2.8. Suicide prevention
Although suicide is of major concern when treating bipolar patients,
this topic has rarely been directly addressed through PE. This lack of
data on suicide prevention may possibly be explained by the sample
needed to find results on this topic. In bipolar patients at high risk
of suicide, augmentation of pharmacotherapy with PE as a component of
wider psychological interventions, including interpersonal therapy plus
daily routines regulation, made it possible to reduce suicidal behavior
(Rucci et al., 2002). Enhancing medication adherence because of the
protective effects of pharmacotherapy (namely lithium) against suicide
(Goodwin et al., 2003) should be a major focus of PE.
3. Discussion
Although to date few reviews (Bauer and McBride, 1996, Bauer and
McBride, 2003, Colom et al., 2003b and Michalak et al., 2004) have been
published concerning the role of PE in bipolar disorders, to our
knowledge, this review is the first to examine its efficacy by clearly
distinguishing between different therapeutic targets. In fact, most
reviews (Bauer, 2002, Huxley et al., 2000, Michalak et al., 2004, Otto
et al., 2003, Scott, 2004 and Swartz and Frank, 2001) summarize the
conclusions of studies related to psychological interventions by
classifying them according to the type of approach (psychoeducation,
cognitive–behavioral therapy, interpersonal therapy…). Regarding our
review, it should be underlined that in several cited studies, PE is one
of the multiple components integrated in psychosocial approaches. And
even when clearly stated as PE, treatments delivered with a large number
of sessions probably also imply other therapeutic elements. Therefore,
it is often difficult to identify clearly which therapeutic effects are
due solely to PE and which ones are due to other psychosocial approaches
or to both. This issue cannot be really sorted out since the conceptual
bases of the various interventions overlap, at least partially.
Although the efficacy of PE approaches in the context of bipolar
disorder has been demonstrated, certain key questions remain to be
answered. Among them, the phase of the illness during which PE might be
efficacious is still to be determined. Data are insufficient or absent
for PE delivered in the presence of mixed, hypomanic symptoms, or rapid
cycling. Some authors (Swartz and Frank, 2001) have suggested adapting
the intervention in function of the patient's mood state. With this
strategy of treatment that is specific to a particular phase of the
illness, different psychotherapeutic approaches are proposed
sequentially according to the patient's mood. However, no evidence
exists to this date to support this type of intervention.
We have found no trial determining the impact of PE on the different
subtypes of bipolar disorder, many studies including both bipolar type I
and II patients. There also is a lack of information concerning PE
efficacy according to the age of the patients (adult, older adults,
adolescents). In fact, our review includes only one study of a geriatric
population (Foelker et al., 1986) and one of adolescents (Miklowitz et
al., 2004).
Concerning the ideal context of care for delivering PE, a majority of
authors recommend outpatient settings, as psychoeducation should mainly
be performed when patients in euthymic state. Research trials to date
have primarily been carried out in ambulatory follow-up settings. Some
studies have involved hospitalized patients. Sometimes, although PE was
carried out only after discharge, patients were recruited during
hospitalization. The question of whether PE produces benefits when
dispensed after a first episode also remains unresolved (Zaretsky,
2003). Patients who have gone through multiple episodes of depression
and mania are less susceptible to respond to treatment and they often
develop complications in the form of medical comorbidity, cognitive
limitations, as well as psychosocial deficits. Introducing interventions
with specific targets earlier when the disorder first appears would
probably enhance the acquisition of illness management coping skills in
order to avoid long-lasting consequences. Similarly, the impact of PE on
recently diagnosed patients, as compared to those who have already been
recognized as bipolar for several years, should be examined. This issue
is of major interest as patients often complain that they only received
information about their disorder long after the occurrence of the first
symptoms and after having finally been correctly diagnosed (Hirschfeld
and Vornik, 2004).
Regarding treatment of acute depressive, (hypo)manic or mixed phases, if
several psychosocial interventions have been attempted, none of them
have been retained as efficacious treatments for manic, hypomanic or
mixed phases of bipolar disorder (Colom and Vieta, 2004 and Swartz and
Frank, 2001). Parikh et al. (1997) suggest that support, combined with
more behavioral techniques aiming to reduce stimuli and daily
activities, is more advisable during manic phases. Brief interventions
based on this type of specific technique, proposed for isolated manic
symptoms, are thought to constitute an efficacious method to prevent the
evolution towards an actual manic episode. These authors emphasize that
maintaining a therapeutic alliance must remain the primary goal of the
follow-up of bipolar patients over the course of the various phases of
the illness, with resort to the principles of supportive therapy in
periods of more severe symptomatic phases.
Concerning the goal of improving functional outcome in bipolar disorder
through PE, only little attention has been devoted to treatments
addressing directly the issue of social and occupational status, of
subjective quality of life in this affection, in comparison to
schizophrenia, as mentioned by Bauer and McBride, 1996 and Bauer and
McBride, 2003. Although PE may have positive effects on global
functioning and quality of life, data regarding this issue are too
sparse and inconclusive to date.
Insufficient clues are available to clarify the mechanisms of action of
psychological interventions or the specific approach that should be
preferred in the context of bipolar disorder (Scott and Todd, 2002 and
Scott, 2004). Given the sparse empirical data, the choice of one
technique over another should be based on clinical judgment at this
stage, notably depending on the difficulties or specificities of the
patient (Scott and Gutierrez, 2004). All bipolar patients do not have
the same capacity to receive psychological “input” due to differences in
cognition, personality or even family context (Jones et al., 2005). As a
result, each individual should be evaluated to determine specific
treatment needs. Jones et al. (2005) suggest that when the patient
presents good levels of introspection and commitment as well as a
relatively mild symptomatology, PE probably constitutes the first-line
psychological intervention. As such, it is strongly recommended to
integrate PE in the basic clinical practice of psychiatrists or any
other member of the healthcare team, in order to allow as many patients
as possible to benefit from this treatment. However, in presence of a
patient with limited levels of commitment and introspection or when high
number of relapse risk factors are present, these authors believe that
more structured psychoeducational approaches, or even group
cognitive–behavioral interventions, are worth considering. They can be
dispensed by clinicians with specific but limited psychotherapeutic
training who conform to structured treatment protocols. Intensive
psychological interventions are all the more pertinent as the subject's
clinical picture becomes more complex or the history of relapse
lengthens. These interventions require clinicians who have been well
trained in the cognitive–behavioral approach, in individual or group format.
Finally, the optimal protocol for applying PE remains to be determined,
notably regarding the length of treatment (Jones, 2002) or the setting
(individual or group). Easier to dispense, PE constitutes a serious
alternative to more traditional psychotherapeutic approaches such as
cognitive–behavioral therapy and family focused therapy (FFT).
In fact, the common sense and apparent simplicity of PE seem to indicate
that it has no counter-indications other than that of the patient being
in an acute phase or having cognitive deficits. However, as Vieta (2005)
points out, it may not prove beneficial for certain individuals and may
even produce unwanted effects. For example, in patients with
obsessive–compulsive personality, PE may lead to an excessive
preoccupation with the detection of early relapse signs, unreasonably
increasing the number of medical consultations or the prescription of
medication. One study in our review also showed that PE increased
patients level of anxiety (van Gent and Zwart, 1991). Finally, as
suggested in a recent study of cognitive–behavioral therapy (Scott,
2004), patients who are still symptomatic and who have previously
presented an elevated number of episodes may not be helped by this type
of psychological intervention and even see their mood state worsen.
In summary, resorting to PE in the treatment of bipolar patients is
highly recommended. As Vieta points out, PE functions as an adjunctive
treatment for bipolar disorder and will not work as a monotherapy
without medications (Vieta, 2005). It is also important to point out
that psychoeducation is part of most other psychosocial interventions.
Patient preference, as well as the importance of including persons close
to the patients in the treatment process and the availability of trained
therapists (Colom et al., 2005 and Scott and Colom, 2005), can dictate
the choice of the intervention technique. Several multicenter studies
related to the evaluation of PE for bipolar disorder are currently
underway. These studies should make it possible to reach a conclusion on
the true impact, underestimated for many years, of this psychological
intervention in the treatment of bipolar patients.