DOI: 10.1111/bdi.

12642

ORIGINAL ARTICLE

Psychoeducation for psychiatric inpatients following remission

of a manic episode in bipolar I disorder: A randomized
controlled trial

Runsen Chen1,2  | Xuequan Zhu1 | Liliana P Capitão2 | Huijun Zhang1 | 

Jiong Luo1  | Xue Wang1 | Yingjun Xi1 | Xiuping Song1 | Yancun Feng1 | 
Liuzhong Cao1 | Gin S Malhi3,4,5

1
The National Clinical Research Center for
Mental Disorders & Beijing Key Laboratory
of Mental Disorders, Beijing Anding
Hospital, Capital Medical University, Beijing,
China
2
Department of Psychiatry, University of
Oxford, Oxford, UK
3
Academic Department of
Psychiatry, Northern Sydney Health District,
St Leonards, NSW, Australia
4
Sydney Medical School
Northern, University of Sydney, Sydney,
NSW, Australia
5
CADE Clinic, Royal North Shore
Hospital, Northern Sydney Local Health
District, St Leonards, NSW, Australia
Correspondence
Jiong Luo, The National Clinical Research
Center for Mental Disorders & Beijing key
Laboratory of Mental Disorders, Beijing
Anding Hospital, Capital Medical University,
Beijing, China.
Email: luojiong2004@163.com
Funding information
Capital Characteristic Clinical Application
Research, Grant/Award Number:
Z141107002514032
Objective: The aim of the present study was to evaluate the effectiveness of psych-
oeducation for bipolar I inpatients following remission of a manic episode in a Chinese
population.
Method: The study recruited currently medicated bipolar I patients, aged 18–60  years,
who were in remission from a manic episode, as determined using the Diagnostic and
Statistical Manual of Mental Disorders Fifth Edition (DSM-­5). Patients were rand-
omized (1:1) to either eight sessions of group-­based psychoeducation (active treat-
ment group) or regular free discussions (control group). The primary outcomes were
the rates of any type of recurrence and rehospitalization following treatment. The
secondary outcomes were changes in mood symptoms, medication adherence, global
functioning, as well as treatment response (as measured using the Clinical Global
Impression scale). Subjects were assessed at baseline and then at 2 weeks, and 1, 2,
3, 5, 7, 9, and 12 months following treatment.
Results: At 1 year, patients receiving the psychoeducation treatment demonstrated
significantly less recurrence. Those in the treatment group also showed a significant
reduction in mania recurrence but not depressive recurrence, and psychoeducation
increased time to remission. Notably, lower rates of rehospitalization were found in
the active treatment group. Those receiving the psychoeducation treatment also re-
vealed higher change from baseline on measures of depression (17-­item Hamilton
Rating Scale for Depression), mania (Young Mania Rating Scale), global functioning
(Clinical Global Impression–severity scale and World Health Organization Disability
Assessment Schedule) (P<.05). However, there were no significant group differences
for the medication adherence scores.
Conclusion: This preliminary evidence suggests that short, group-­based psychoedu-
cation benefits currently medicated inpatients following the remission of mania in
bipolar I disorder. This intervention warrants further investigation, especially in other
Chinese populations. If future studies confirm its benefits, group-­based psychoedu-
cation could be incorporated into routine psychiatric inpatient care for bipolar pa-
tients in China.

Runsen Chen and Xuequan Zhu contributed equally to this study.

Bipolar Disorders. 2018;1–10. wileyonlinelibrary.com/journal/bdi   © 2018 John Wiley & Sons A/S. |  1
Published by John Wiley & Sons Ltd
 |
2       CHEN et al.
KEYWORDS
bipolar disorder, Chinese population, group-based psychoeducation, psychiatric inpatients
1 |  I NTRO D U C TI O N
Bipolar disorder (BD) is a serious mental illness, characterized by
periods of mania and depression that eventuate in chronic inter-
personal and occupational difficulties over a patient’s lifetime.1
BD type I (BD-­I) is the sixth leading cause of disability globally.
Manic episodes are a common feature of BD-­I, and the lifetime
prevalence is about 1.5% in European countries. 3 Acute manic
episodes usually require emergency admission to a psychiatric
4
hospital to facilitate rapid recovery. Most patients experience
depressive or manic episodes during the periods of remission after
discharge from hospital. Previous studies have demonstrated that
the relapse rates are high, ranging between 28% and 44% within
5
1 year. It is therefore crucial to evaluate and implement strate-
gies that can reduce the rates of rehospitalization and extend the
period between recurrences. One promising way to achieve this is
to implement psychoeducation techniques, in which patients learn
how to recognize and manage the early warning signs of mania and
depression. 6 clinical patients accommodated by a hospital or in the community.16
Psychoeducation techniques are usually implemented in com-
bination with pharmacological treatment. This joint treatment
approach is, in fact, recommended by several regulatory agen-
cies across the world, such as the Canadian Network for Mood
7
and Anxiety Treatments for bipolar disorder, Australian and New
Zealand practice guidelines for mood disorders 8 and Japanese
9
Society of Mood disorders guidelines. Previous studies have also
suggested that psychoeducation interventions are effective in
preventing relapse and in reducing hospital admission rates. For
instance, Kessing et al.10 conducted a study to assess the effects
of treatment in a specialized outpatient mood disorder clinic (con-
sisting of evidence-­b ased pharmacological intervention plus group
psychoeducation) vs standard care (i.e., treatment by a general
practitioner, a private psychiatrist, or from the local community
mental health center). The authors found that the combined treat-
ment including psychoeducation resulted in a 40% reduction in
the risk of readmission to hospital. Similarly, Morriss et al.11 con-
ducted a multicenter, parallel group study to evaluate the efficacy
of structured group psychoeducation vs optimized unstructured
peer support in remitted BD patients. Psychoeducation delayed
the time of occurrence of the next bipolar episode compared to
peer support in participants with 1–7 previous episodes. There
was no difference between groups for participants with 8–19
previous bipolar episodes or 20 or more episodes. There was also
evidence showing that the proportion of participants with a fur-
ther mania-type episode was lower in the psychoeducation group.
Finally, the benefits of group psychoeducation were also con-
12
firmed in a meta-­a nalysis. The authors found that group-­b ased,
but not individual-­b ased, psychoeducation was effective against
both poles of relapse. Psychoeducation also improved medication
adherence and short-­term knowledge about medication. However,
no consistent effects were seen on mood symptoms, quality of
life, or functioning.
Few psychoeducational studies have been conducted in psy-
2 chiatric inpatient settings.13 Inpatients are more severely impaired
than outpatients, and most patients experiencing manic episodes
require hospital admission. It is unclear whether psychoeducational
programs benefit psychiatric routine care in severely manic patients.
Dropout rates are relatively high in outpatient settings, perhaps due
to the long duration of psychoeducation programs.14 Therefore, the
present study adjusted the length of treatment, to examine whether
a short duration psychoeducational program could reduce dropout
rates while resulting in similar benefits.
Furthermore, research indicates a need to examine psychoed-
ucation for BD across diverse ethnic backgrounds.15 Owing to the
shortage of mental healthcare services and staff throughout main-
land China, it is not possible to conduct psychoeducation with all
Therefore, pharmacotherapy is still the main treatment approach for
patients with BD in China’s psychiatric services, despite this treat-
ment leading to higher rates of relapse than from using concurrent
psychotherapeutic interventions.17 To our knowledge, the present
study is the first to evaluate a psychoeducational program for BD in
China.
The aim of the present randomized controlled trial was to assess
the effectiveness of a short psychoeducation program in BD-­I in-
patients who recently remitted from a manic episode in a Chinese
population. The primary outcome was the prevention of recurrences
and rehospitalization over a period of 12 months, while the second-
ary outcomes referred to changes from baseline to month 12 in de-
pression, mania, global functioning, and adherence to medication.
2 | M E TH O DS
2.1 | Trial design
This was a randomized, rater-­blind, controlled trial with a 12-­month
follow-­up. The study was conducted by the Beijing Anding Hospital-­
affiliated Capital Medical University research team. The Beijing
Anding Hospital is one of the largest psychiatric hospitals in China,
comprising 25 960 square meters, with more than 1000 inpatient
beds for mental health patients, receiving up to 5000 inpatient ad-
missions per year and over 1200 outpatient visits per day. Nearly
half of the patients have mood disorders, therefore making it pos-
sible to recruit large numbers of patients for research studies. This
hospital houses the first and largest mood disorder treatment center
in China. All patients with BD-­I admitted to the inpatient unit who
CHEN et al.
received standard hospitalized treatment between September 2015
and April 2016 were identified as potential participants and invited
to take part in the current study. Participants were referred to the
program as soon as their clinical symptoms remitted.
The study was approved by the Human Research and Ethics
Committee of Beijing Anding Hospital, Capital Medical University.
Informed consent was obtained from all participants. The par-
ticipants were also informed that they could withdraw from the
study at any time, without giving a reason and without pen-
alty. The trial is registered with the Chinese Clinical Trial Registry
(ChiCTR-­IOR-­17012449).
2.2 | Participants
The study recruited 140 patients with BD-­I. The inclusion criteria
comprised: (i) acute inpatients aged 16–60 years, inclusive; (ii) a cur-
rent diagnosis of BD-­I and current manic episode, according to the
Diagnostic and Statistical Manual of Mental Disorders Fifth Edition
(DSM-­V )18; (iii) achieved remission, determined by a score of less
than 8 in the Young Mania Rating Scale (YMRS) after 1–2 weeks’
standard hospitalization treatment; (iv) at least 9 years of education.
Exclusion criteria comprised: (i) current drug or substance abuse or
dependence; (ii) a diagnosis of any other psychotic disorder; (iii) the
presence of neurological conditions; (iv) the presence of serious or
unstable medical disease; (v) pregnant or lactating women; (vi) intel-
lectual disability.
2.3 | Assessment
Potential patients were given information about the study and, if
interested, were subsequently screened for eligibility by the psy-
chiatrists. Assessment of the intervention and control groups was
performed by a psychiatrist who was blind to the treatment condi-
tion. At each session, the type of recurrence, pharmacotherapy ad-
herence, disability levels, and number of hospital admissions were
assessed, based on the following guidelines by the psychiatrist:
1. Recurrence was assessed using the 17-item Hamilton Rating
Scale for Depression (HRSD17)19 and YMRS.20 A depressive
episode was defined as a total score ≥17 on the HRSD17,
persisting for more than 2 weeks. A manic episode was defined
as a total score ≥20 on the YMRS, persisting for more than
1 week. A hypomanic episode was defined as scores ≥12 on
the YMRS, persisting for more than 4 days. The time to re-
currence was recorded. Subthreshold manic or depressive ep-
isodes were defined as HRSD17 scores of 7–17 or YMRS scores
of 8–12, where participants did not meet recurrence criteria.
2. Assessment of hospitalization due to BD-I was derived from med-
ical records.
3. Pharmacotherapy adherence was assessed by the research doc-
tor, using a four-point visual analogue scale. Patients or their car-
egivers were asked by the nurse or research doctor if they had
forgotten to take their medicine on time. This was scored from 1
|
      3
to 4, corresponding to “never”, “occasionally”, “around half of the
time”, and “most of time” (i.e., full adherence; partial adherence;
poor adherence; and non-adherence); higher scores reflected
poorer adherence.
4. Global functioning was assessed using the 36-item World Health
Organization Disability Assessment Schedule (WHODAS 2.0),
consisting of six dimensions, as well as the Clinical Global
Impression (CGI) scale.
2.4 | Procedures
2.4.1 | Randomization
The randomization code was drawn up by a statistician not involved
in the study, using a block method. Sequentially enrolled patients
were assigned the randomization number available in blocks of six,
using a 1:1 assignment ratio. The number table was kept separately
by a nurse who was not involved in the study.
The outcome measures were assessed by blind raters (psychia-
trists) who were not involved in the treatment and who were also
blind to the study protocol and treatment assignment.
2.4.2 | Treatment group
Patients in the treatment group received standard pharmacother-
apy for BD-­I and were simultaneously enrolled in a group-­b ased
psychoeducation program. The program was delivered in eight
sessions of 40–60 minutes each, in groups consisting of 8–12 pa-
tients. These sessions were delivered in a period of over 2 weeks,
and were conducted during the remission phase of BD-­I, before
the patients were discharged from the hospital. A few patients
who were discharged from the hospital early (N=4) returned to
the hospital and finished the treatment sessions. These sessions
were delivered by three psychiatrists and two clinical psycholo-
gists, and the facilitators were trained by an academic expert in
psychoeducation (Professor Yingjun Xi). We modified and opti-
mized the original contents of the handbook “Psychoeducation for
Bipolar Disorder”, 21 to better suit the Chinese population. Group
sessions included the following topics: Session 1: introduction of
BD disease knowledge, such as biological etiology, epidemiology,
and concepts; Session 2: definition of mania and hypomania, de-
pression (mixed state), and psychotic symptoms; Session 3: biolog-
ical rhythms and manic/depressive episode; Session 4: the role of
pharmacological treatment, different types of medication; Session
5: medication adherence and monitoring, electroconvulsive ther-
apy, and psychotherapy; Session 6: stress management, problem
coping strategies, and interpersonal relationships; Session 7: re-
current symptoms, early detection of episodes, and how to seek
help; Session 8: review and assessment, creating a management
plan, and how to monitor daily mood. The psychoeducation ma-
terials (in Chinese) are available on request by email from the cor-
responding author.
 |
4       CHEN et al.

Enrollment Assessed for eligibility (n = 153)

Excluded (n = 13)
Not meeng inclusion criteria
(n = 7)
Declined to parcipate (n = 6)

Randomized (n = 140)
Allocaon

Treatment group (n = 71) Control group (n = 69)

1st month: 1 mania recurrence 1st month 2 mania recurrence

2nd month: 1 lost to follow up; 1 lost to follow-up

Follow-up
3rd month: 1 lost to follow-u; 2nd month 4 mania recurrence

5th month: 2 mania recurrence 3rd month 1 mania recurrence

1 depressive recurrence;
1 depressive recurrence
7th month: 1 lost to follow-up
5th month 3 mania recurrence
9th month: 1 depressive recurrence
1depressive recurrence
1 lost to follow-up
7th month 3 mania recurrence
12th month: 3 mania recurrence
1 depressive recurrence
2 lost to follow-up

9th month 4 lost to follow-up F I G U R E   1   Trial profile:

psychoeducation for psychiatric inpatients
following remission of a manic episode
in bipolar I disorder. FAS, full analysis
Analysis Analyzed (FAS) (n = 71) Analyzed (FAS)(n = 69) set

insomnia?” The self-­

2.4.3 | Control group
Patients randomized to the control group continued on standard
pharmacotherapy treatment (with mood stabilizers and antipsychotic
agents, among others) as prescribed and adjusted by their psychia-
trist. They were also given the same handbook of self-­management
that was given to the treatment group, and were enrolled for regular
free discussions with research nurses at corresponding time points.
Of note, the self-­management book only contained questions, with-
out answers, pertaining to each of the sessions received by the ac-
tive treatment group. These included questions such as: “Do you
think that taking the medication is important for treatment?”; “Have
you received modern electroconvulsive therapy (MECT)?”; “Have
you previously experienced a manic episode?”; “Have you previ-
ously experienced a depressive episode?”; and “Do you suffer from
management materials (in Chinese) are also
available on request by email from the corresponding author. The
nurses avoided giving any psychoeducational feedback, except for
the normal feedback necessary for nurse–patient interactions.
Both groups were then scheduled for follow-­up assessments, at
2 weeks and then 1, 2, 3, 5, 7, 9, and 12 months following treatment.
2.5 | Statistical analysis
All collected data were analyzed using the SAS Statistical Package
(version 9.4; Beijing Anding Hospital, Capital Medical University).
The sample characteristics were compared using the chi-­square test
for categorical variables, using Fisher’s Z when needed; t tests for
normal quantitative variables; and Mann-­Whitney’s U test for non-­
normal quantitative variables.
CHEN et al.
TA B L E   1   Number of recurrences at each time point
<1 Mo 1–2 Mo 2–3 Mo 3–5 Mo
M D L M D L M D L M D L
TG 1 1 1 2 1 1
CG 2 1 4 1 1 3 1 3
CG, control group; D, depression recurrence; L, lost to follow-­up; M, mania recurrence; Mo, month; TG, treatment group.
Efficacy analyses were based on an intent-­to-­treat approach,
which corresponds to the full analysis set (FAS), comprising all pa-
tients who were assigned to any treatment group and who had a
valid baseline assessment of the primary efficacy outcome variables
(HRSD17 and YMRS total scores). Kaplan-­Meier survival analyses
were used to calculate the estimated time from baseline to recur-
rence. The cox proportional hazards regression model was used to
compare the estimated time to recurrence between the two groups
while controlling for covariates such as age, gender, and duration of
illness. Secondary analyses were performed to assess changes from
baseline to month 12 on HRSD17, YMRS, Positive and Negative
Syndrome Scale, CGI–severity scale (CGI-­S), and WHODAS 2.0 total
score, using the mixed model for repeated measures (FAS, MMRM),
with treatment as the fixed factor and baseline score, age, and
course of the episode as covariates. A P-­value lower than .05 was
considered statistically significant.
3 | R E S U LT S
3.1 | Recruitment, retention, and sample
characteristics
The method for recruiting patients into the study is outlined in
Figure 1 and Table 1. Of the 153 potential patients who signed the
informed consent form and were screened, 140 met all the study
criteria and were randomized. Seventy-­one medicated patients were
assigned to the group-­based psychoeducation and 69 to free regular
discussions. The trial profile is shown in Figure 1.
During the study, a total of 11 subjects withdrew. In the active
treatment group, four patients withdrew (5.63%), of whom two
were uncooperative, and two did not participate in the 9-­m onth
follow-­
up. In the control group, seven patients (10.14%) were
withdrawn from the study owing to lack of participation in the
follow-­up assessments. The proportion of dropouts was distrib-
uted equally across the active treatment and control groups (χ2
(1) = 0.984, P=.321).
Of the 11 subjects who withdrew from the study, the mean
(standard deviation [SD]) age was 28.09 (8.53) years, mean number
of years of education was 13.55 (2.62), the HRSD17 total score was
2.45 (4.41), and YMRS total score was 4.82 (5.25). The CGI-­S mean
score was 3.0 (1.10). These measures did not differ statistically be-
tween groups (P>.05 for all).
The FAS consisted of 140 patients (n = 71 in the active treat-
ment group and n = 69 in the control group). The mean age was
|
      5
5–7 Mo 7–9 Mo 9–12 Mo Total
M D L M D L M D L M D L
1 1 3 6 2 4
1 2 4 13 3 7
33.3 years for the treatment group and 34.5 for control group, and
approximately 88.73% were women in the treatment group and
91.30% in the control group. There were no significant differences
at baseline between groups in demographic or baseline clinical char-
acteristics (Table 2) (P>.05 for all). The FAS included 140 patients,
with the exclusion of 0 patient from the groups. The mean base-
line HRDS17 total score was 2.73 (SD = 2.98) for the treatment
group and 2.12 (SD = 2.21) for the control group; the mean base-
line YMRS total score was 3.24 (SD = 3.99) for the treatment group
and 3.57 (SD = 3.60) for the control group, indicating remission from
prior treatment, as also reflected in the mean CGI-­S score of 2.27
(SD = 0.48) for the treatment group and 2.42 (SD = 0.98) for the con-
trol group (see Table 2).
3.2 | Outcomes
3.2.1 | Primary outcomes
A total of 24 subjects met the criteria for any recurrence; eight of
them recurred over the entire course of the study in the treatment
group, and 16 in the control group (see Table 3). The estimated in-
terval for any relapse in the two groups in the Kaplan-­Meier anal-
ysis is illustrated in Figure 2. The difference between the survival
curves was significant for any relapse (log rank chi-­square = 4.0282,
P=.0447). The adjusted hazard ratio of time to any recurrence was
0.40 (95% confidence interval [CI]: 0.17–0.97). The age at the first
episode was significantly associated with time to any recurrence
(P=.0307), giving an adjusted hazard ratio of 0.9 (95% CI: 0.87–0.99).
Over the course of the study, a total of 15 patients went back
into hospital owing to recurrence, four of whom were in the treat-
ment group. This group showed significantly less rehospitalization
compared to the control group (χ2 (1) = 3.887, P=.049).
3.2.2 | Secondary outcomes
There were significant differences in any type of subthreshold recur-
rence (χ2 (1) = 13.3825, P = .000), which meant that depressive or
hyper/manic symptoms were apparent but did not meet the criteria
for a full diagnosis of recurrence. However, there were no statistical
differences between groups when considering depression or mania
subthreshold recurrence specifically (P>.05) (See Table 3). Analyses
of secondary efficacy measures yielded results consistent with the
primary endpoint analysis. There were significant differences be-
tween the treatment groups on change from baseline to end visit for
 |
6       CHEN et al.

TA B L E   2   Demographic and clinical data from the treatment and

control groups

Treatment group Control group

(n = 71) (n = 69)

Female (n [%]) 63 (88.73%) 63 (91.30%)

Age (y)
Mean (SD) 33.34 (10.34) 34.52 (11.95)
Marital status (n [%])
Single 31 (43.66) 23 (33.33)
Married 34 (47.89) 35 (50.72)
Divorced/widowed 6 (8.45) 11 (15.94)
Years of education (y)
Mean (SD) 13.68 (3.42) 12.70 (3.09)
Family history (n [%]) 18 (25.35) 25 (36.23)
F I G U R E   2   Kaplan-­Meier estimates of time to any recurrence
Age of onset (y)
Mean (SD) 23.98 (7.46) 25.04 (8.95)
Duration of illness (mo)
the HRSD17, YMRS, CGI-­S, and WHODAS scales, as revealed in the
Median (range) 8.0 (52.0) 7.0 (42.0)
MMRM analysis (see Table 4). Regarding the MMRM analysis of the
Type of first episode (n [%]) HRSD17 and YMRS total score, the confidence internal for the dif-
Manic 24 (33.80) 35 (50.72) ference between the two groups did not include zero (indicating sta-
Hypomanic 1 (1.41) 1 (1.45) tistically significant superiority of the treatment group). Moreover,
Depressive 44 (61.97) 29 (42.03) the difference on the WHODAS between the two groups using the
Mixed 2 (2.82) 4 (5.80) MMRM estimates of change to end visit was significant (mean 0.41,
Times of episode 95% CI: 0.43–3.58).
Median (range) Medication adherence was rated by a research doctor, who was

Total 4 (23) 3 (16) asked to assess the adherence of patients based on the self-­report
of the patient and/or their carer, on a continuum using a four-­point
Manic 2 (13) 2 (16)
visual analogue scale. An integer score between 1 and 4 was derived,
Depressive 1 (23) 1 (4)
with higher scores reflecting worse adherence to medication.
Times of admission
The score for medication adherence in the first visit was 1.20
Median (range) 1 (13) 1 (6)
(SD = 0.47) for the treatment group and 1.25 (SD = 0.47) for the
Lifetime smoking history 5 (7.04) 5 (7.25)
control group; at the last visit, the score was 1.37 (SD = 0.92) for
(n [%])
the treatment group and 1.20 (SD = 1.03) for the control group. The
Lifetime drinking history (n [%]) 5 (7.04) 2 (2.90)
adjusted change of score for compliance was 0.13 (SD = 0.06) for
HRSD score at baseline 2.73 (2.98) 2.12 (2.21)
the treatment group and 0.11 (SD = 0.06) for the control group. We
(mean [SD])
used MMRM to assess group differences, with treatment as a fixed
YMRS score at baseline
factor, and age and years of education as covariates. As the results
Mean (SD) 3.24 (3.99) 3.57 (3.60)
demonstrate, there were no significant differences between groups;
PANSS score at baseline
the mean difference in score between the two groups had a 95% CI
Mean (SD) 33.18 (2.90) 35.03 (4.94) of 0.01 (−0.08 to 0.11).
CGI-­S score at baseline
Mean (SD) 2.27 (0.48) 2.42 (0.98)
WHODAS score at baseline 4 | D I S CU S S I O N
Mean (SD) 46.25 (11.82) 49.25 (11.71)
Valproic acid salt 60 (84.51) 56 (81.16) Although previous research has shown the positive effects of psy-

Lithium 54 (76.06) 47 (68.12) choeducation on BD, few studies have examined whether BD-­
I
inpatients with current manic episodes also benefit from these
Antipsychotic agents 65 (91.55) 63 (91.30)
techniques. The objective of the present study was therefore to ex-
CGI-S, Clinical Global Impression–severity scale; HRSD, Hamilton
amine the effectiveness of psychoeducation in inpatients with BD-­I
Depression Rating Score; mo, month; PANSS, Positive and Negative
Syndrome Scale; SD, standard deviation; WHODAS, WHO Disability following recent remission of a manic episode after receiving treat-
Assessment Schedule; y, year; YMRS, Young Mania Rating Scale ment in a psychiatric hospital. The results demonstrated that a short,
CHEN et al. |
      7

TA B L E   3   Recurrence, subthreshold
Treatment group Control group
recurrence, and admission rates for the
N (%)a (n [missing] = 67 [4]) (n=69) P-­value
treatment group vs control group during
the 12-­month follow-­upa Any recurrences 8 (11.94) 16 (25.81) .0432b
Manic recurrences 6 (8.96) 13 (20.97) .0544b
Depressive recurrences 2 (2.99) 3 (4.84) .6709c
Any subthreshold recurrences 4 (5.97) 19 (30.65) .0003b
Rehospitalization (admissions due 4 (5.63) 11 (15.94) .0487b
to recurrences)
a
Based on nonmissing values.
b
Based on chi-­square test.
c
Based on Fisher’s Z test.

TA B L E   4   Group differences in clinical symptoms at month 12

Treatment group Control group

Mean change
Variables Baseline End visit Baseline End visit P-­valuea (95% CI)

HRSD17 total 2.73 ± 2.98 1.98 ± 4.56 2.12 ± 2.21 2.03 ± 5.25 .0214 0.41 (0.06-­0.76)

scores
YMRS total 3.24 ± 3.99 4.02 ± 9.96 3.57 ± 5.52 5.52 ± 9.37 <.0001 1.55 (0.87-­2.23)
scores
CGI-­S 2.27 ± 0.48 1.79 ± 1.47 2.42 ± 0.98 2.40 ± 1.61 <.0001 0.40 (0.28-­0.52)
WHODAS total 46.25 ± 11.82 40.70 ± 15.93 49.25 ± 11.71 44.82 ± 22.78 .0129 2.01 (0.43-­3.58)
scores

CGI-­I, Clinical Global Impression–improvement scale; CGI-­S, Clinical Global Impression–severity scale; CI, confidence interval; HRSD, Hamilton
Depression Rating Score; WHODAS, World Health Organization Disability Assessment Schedule; YMRS, Young Mania Rating Scale.
a
Models fit via maximum likelihood. Adjusted for clustering by gender.

group-­based psychoeducation program was effective. Specifically,
it prevented recurrence and rehospitalization, and improved mood
symptoms, disability, and clinical global functioning. However, medi-
cation adherence scores were not significantly affected.
Meta-­analysis studies have previously reported that psychoedu-
22–24
cation is effective for the prevention of recurrence. We demon-
strated that psychoeducation significantly reduced mania in the
intervention group compared to the control group, but this finding
did not extend to depression (although this was limited by the fact
that few participants showed symptoms of depression). This is in
25
line with previous research. For instance, D’Souza et al. reported
lower mania recurrence in a group receiving psychoeducation, but
no differences were seen for depression recurrence. Similarly, Perry
26
et al. conducted a psychoeducation intervention on BD patients,
and the results showed that there were fewer recurrences of mania
than of depression. Our study is also consistent with a recent meta-­
analysis, conducted by Bond and Anderson,12 in which psychoed-
ucation was found to be effective in preventing any recurrence, as
well as manic/hypomanic recurrence, but not depression recurrence.
However, these studies included participants with both BD-­I and bi-
polar II disorder, so the present study is the first to show the effects
of a psychoeducation program on reducing the recurrence of manic
episodes in patients with BD-­I. The fact that we did not find signif-
icant differences in the recurrence of depression might have been
due to the timing of our follow-­up (the last visit was 12 months after
the intervention). It may be necessary to increase the length of the
follow-­up in order to detect clinically significant differences in de-
pression, as patients in remission following a manic episode may take
longer to re-­experience depression than mania.
In the present study, we also found that psychoeducation is an
effective way to prevent rehospitalization in BD-­
I patients with
manic episodes, which is in agreement with a range of previous
studies. Rouget and Aubry27 conducted a meta-­analysis demon-
strating that psychoeducation reduced the rates of hospitalization,
and Miklowitz24 found that medication treatment and psychoedu-
cation contributed to a 40% reduction in relapse over a period of
12–30 months. Candini et al. 28 conducted a study to evaluate the
effectiveness of psychoeducation in reducing the number of hos-
pitalizations at the 1-­year follow-­up in Italy, and the results showed
that the mean number of hospitalizations per patients was signifi-
cantly lower for the patients receiving psychoeducation vs controls.
However, in a 5-­year follow-­up study, there appeared to be no differ-
ences between a group receiving psychoeducation vs controls in the
number of hospitalizations for BD, but psychoeducation reduced the
length of hospitalization. 29 Longer follow-­up assessment is therefore
required to see the full impact of psychoeducation interventions.
Medication is the central treatment of BD, and patients
learning the importance of adhering to medication within a psy-
choeducational program could be a key factor for preventing
recurrence and rehospitalization. 25 In our study, there were no
 |
8       CHEN et al.
statistically significant group differences for medication adher-
ence. Interestingly, both groups showed high medication ad-
herence scores, although the group receiving psychoeducation
showed a slightly higher score than the control group. This could
be explained by a number of factors. The control group received
standardized hospitalized treatment, which may have included
informal supportive consulting, and they also regularly saw their
attending psychiatrist in the outpatient unit and interview site
after discharge, possibly receiving repeated explanations of the
importance of their medication treatment. This may also have con-
tributed to the low recurrence and rehospitalization rates within
1 year (17.6% and 11%, respectively). Furthermore, in the present
study, medication adherence was measured by the research doctor
using a four-­p oint visual analogue scale, which was based on the
self-­report given by the patient and/or the caregiver, which was
therefore susceptible to recall bias. Future studies should there-
fore include formal medication adherence questionnaires 30,31 and
innovative electronic tools such as smartphones to measure med-
ication adherence objectively. 32
Although there were significant group differences for depression
(HRSD) and mania (YMRS) scores following the psychoeducation
treatment, the clinical implications of these findings are difficult to
interpret as only five patients experienced depression recurrence.
Both groups demonstrated low mood symptoms at baseline as well
as at the last visit, given that participants had been discharged from
the hospital within 1 year. In addition, we included only inpatient
participants following remission of a manic episode, and future
studies should consider including psychiatric inpatients following re-
mission of a depressive episode in BD, to evaluate the impact of psy-
choeducational intervention on depressive symptoms experienced
within BD.
BD patients may still experience difficulties at different func-
tional levels during periods of symptomatic remission or recovery. It
is important to measure their functioning level, which could help to
predict their treatment outcome. There are various assessment ques-
tionnaires to measure functioning. For instance, Miklowitz et al.33
conducted a study to assess the functional outcomes of patients
with BD over 9 months, using the Range of Impaired Functioning tool
(LIFE-­RIFT), and found that psychoeducation improved patients’ re-
lationship functioning and life satisfaction. Similarly, Lobban et al.34
reported that a group receiving psychoeducation had a higher mean
level of functioning compared to the control group, as assessed using
the Social and Occupational Functioning Assessment Scale (SOFAS).
Our study found a difference in the WHODAS scores, indicating a
significant reduction in disability in the group receiving psychoed-
ucation in different domains of functioning, such as understanding
and communication; self-­
care; physical movement; getting along
with others; physical activity, learning, and studying; and social par-
ticipation. These results are similar to those of George et al.,13 who
reported that psychoeducation improved overall functioning in BD
patients using the WHODAS 12-­item questionnaire.
Our results demonstrated that it is feasible to implement a
2-­
week group-­
based psychoeducation program in a psychiatric
inpatient setting. The dropout rate in our study (only two partici-
pants) was significantly lower than in other group-­based psychoed-
ucation studies.35–37 This might have been due to the fact that our
study implemented a short period of psychoeducation. Participants
may lose interest in the program if the gap between sessions is too
long.38 In addition, inpatient settings are easier to manage than out-
patient settings. Normally, it takes at least 1 month for inpatients
receiving standardized hospitalization treatment to reach symptom
remission before enrolling in group psychoeducation programs.
Most of the patients with mania spend 4–6 weeks in inpatient treat-
ment before being discharged from the hospital. In our study, a few
patients were discharged from the hospital early, then returned to
finish their remaining group sessions. Considering the potential for
this program to be more cost-effective for inpatients than others,
we suggest that future studies evaluate whether shorter psychoed-
ucation interventions have similar benefits for BD inpatient samples,
so that patients can complete all sessions before being discharged
from hospital.
Group-­based programs can help participants to realize that they
are not alone in suffering from BD, and they may benefit from expo-
sure to a wider range of perspectives on their situation, from other
group members. Learning that they are not alone in dealing with
the illness can alleviate feelings of shame and isolation.39 However,
there is a paucity of studies that directly compare the effectiveness
of group-­based vs individual psychoeducation programs for inpa-
tients with BD. Future research should include individual psychoed-
ucation programs in an inpatient setting, in order to determine the
optimal method for achieving effective treatment outcomes.
4.1 | Limitations
The present study had several limitations. For example, it did not
include a formal sample size calculation. Moreover, a large propor-
tion of patients was female, relatively young, and with few previ-
ous bipolar episodes. Future research should explore differences
in group dynamics in both males and females, as well as across a
range of age categories and clinical characteristics (e.g., patients
with a higher number of previous episodes). We also excluded
patients with psychotic disorders but the comorbidity with other
DSM-­V psychiatric disorders was not measured. This will have had
an impact on the generalizability of the findings. In addition, at
12 months, the length of the follow-­up period may not have been
sufficient to observe the full impact of psychoeducation on pa-
tients following a manic episode, particularly regarding any ben-
efits for depressive symptoms, so longer follow-­up periods should
be considered in future research. Further, the present study used
visual analogue scales to evaluate medication adherence; future
studies should include formal medication adherence question-
naires or electronic tools, in order to eliminate recall bias. Finally,
the study was conducted in a single hospital; future research
should consider implementing this intervention across multiple
sites (including in community-­b ased settings), to explore the gen-
eralizability of the present findings.
CHEN et al.
5 | CO N C LU S I O N
The study was the first to assess group-­b ased psychoeducation
programs for BD within a Chinese sample. It demonstrated the ef-
ficacy of a short and low-­cost psychoeducation program for inpa-
tients with BD, which resulted in the prevention of recurrence and
rehospitalization, and in improvements in mood symptoms and
global functioning. Considering that most patients experiencing a
manic episode in the context of BD-­I require psychiatric hospital
admission, based on the benefits of our program, this interven-
tion could be a positive addition to routine psychiatric inpatient
care. Therefore, it is suggested that policy makers consider placing
greater emphasis on and funding of the provision of more accessi-
ble psychoeducation interventions to inpatients with BD in China,
and across the world.
AC K N OW L E D G E M E N T
This research was supported by the Capital Characteristic Clinical
Application Research [grant number Z141107002514032].
D I S C LO S U R E
There are no competing interests to declare.
ORCID
Runsen Chen  http://orcid.org/0000-0002-2145-8630
Jiong Luo  http://orcid.org/0000-0002-7211-2429
REFERENCES
1. American Psychiatric Association. Diagnostic and Statistical Manual
of Mental Disorders. Arlington: American Psychiatric Publishing;
2013.
2. Inder ML, Crowe MT, Luty SE, et al. Randomized, controlled trial
of Interpersonal and Social Rhythm Therapy for young people with
bipolar disorder. Bipolar Disord. 2015;17:128‐138.
3. Pini S, de Queiroz V, Pagnin D, et al. Prevalence and burden of bi-
polar disorders in European countries. Eur Neuropsychopharmacol.
2005;15:425‐434.
4. Karamustafalıoğlu O, Reif A, Atmaca M, et  al. Hospital stay in
patients admitted for acute bipolar manic episodes prescribed
quetiapine immediate or extended release: a retrospective non-­
interventional cohort study (HOME). BMC Psychiatry. 2014;14:
246.
5. Coryell W, Scheftner W, Keller M, Endicott J, Maser J, Klerman GL.
The enduring psychosocial consequences of mania and depression.
Am J Psychiatry. 1993;150:720‐727.
6. Morriss R, Faizal MA, Jones AP, Williamson PR, Bolton CA,
McCarthy JP. Interventions for helping people recognise early signs
of recurrence in bipolar disorder. Cochrane Database Syst Rev.
7. Yatham LN, Kennedy SH, Parikh SV, et al. Canadian Network for
Mood and Anxiety Treatments (CANMAT) and International Society
for Bipolar Disorders (ISBD) collaborative update of CANMAT
guidelines for the management of patients with bipolar disorder:
update 2013. Bipolar Disord. 2013;15:1‐44.
|
      9
8. Malhi GS, Bassett D, Boyce P, et al. Royal Australian and New
Zealand College of Psychiatrists clinical practice guidelines for
mood disorders. Aust N Z J Psychiatry. 2015;49:1087‐1206.
9. Kanba S, Kato T, Terao T, Yamada K. Guideline for treatment of bi-
polar disorder by the Japanese Society of Mood Disorders, 2012.
Psychiatry Clin Neurosci. 2013;67:285‐300.
10. Kessing LV, Hansen HV, Hvenegaard A, et al. Treatment in a special-
ised out-­patient mood disorder clinic v. standard out-­patient treat-
ment in the early course of bipolar disorder: randomised clinical
trial. Br J Psychiatry. 2013;202:212‐219.
11. Morriss R, Lobban F, Riste L, et al. Clinical effectiveness and ac-
ceptability of structured group psychoeducation versus opti-
mised unstructured peer support for patients with remitted
bipolar disorder (PARADES): a pragmatic, multicentre, observer-­
blind, randomised controlled superiority trial. Lancet Psychiatry.
2016;3:1029‐1038.
12. Bond K, Anderson IM. Psychoeducation for relapse prevention in
bipolar disorder: a systematic review of efficacy in randomized con-
trolled trials. Bipolar Disord. 2015;17:349‐362.
13. George LS, Sharma P, Nair S. Effect of psycho-­education on quality
of life and disability among persons with bipolar disorder: random-
ized controlled trial. Int J Educ Manage Stud. 2015;6:461.
14. Gumus F, Buzlu S, Cakir S. Effectiveness of individual psychoedu-
cation on recurrence in bipolar disorder; a controlled study. Arch
Psychiatr Nurs. 2015;29:174‐179.
15. Poole R, Smith D, Simpson S. Patients’ perspectives of the feasi-
bility, acceptability and impact of a group-­ based psychoeduca-
tion programme for bipolar disorder: a qualitative analysis. BMC
Psychiatry. 2015;15:184.
16. Song Y, Liu D, Chen Y, He G. Using focus groups to design a psycho-
education program for patients with schizophrenia and their family
members. Int J Clin Exp Med. 2014;7:177.
17. Michalak E, Yatham L, Lam R. The role of psychoeducation in the
treatment of bipolar disorder: a clinical perspective. Clin Approaches
Bipolar Disord. 2004;3:5‐11.
18. Association AP. Diagnostic and Statistical Manual of Mental Disorders
(DSM-5®). Arlington, VA: American Psychiatric Publishing; 2013.
19. Hamilton M. A rating scale for depression. J Neurol Neurosurg
Psychiatry. 1960;23:5642.
20. Bech P. The Bech-­Rafaelsen Mania Scale in clinical trials of thera-
pies for bipolar disorder. CNS Drugs. 2002;16:47‐63.
21. Colom F, Vieta E. Psychoeducation Manual for Bipolar Disorder.
Cambridge, UK: Cambridge University Press; 2006.
2 2. Soares-Weiser K, Vergel YB, Beynon S, et al. A systematic review
and economic model of the clinical effectiveness and cost-­
effectiveness of interventions for preventing relapse in peo-
ple with bipolar disorder. Health Technol Assess 2007;11:iii‐iv,
ix–206.
23. Beynon S, Soares-Weiser K, Woolacott N, Duffy S, Geddes JR.
Psychosocial interventions for the prevention of relapse in bipo-
lar disorder: systematic review of controlled trials. Br J Psychiatry.
2008;192:5‐11.
24. Miklowitz DJ. Adjunctive psychotherapy for bipolar disorder: state
of the evidence. Am J Psychiatry. 2008;165:1408‐1419.
25. D’Souza R, Piskulic D, Sundram S. A brief dyadic group based psy-
choeducation program improves relapse rates in recently remitted
bipolar disorder: a pilot randomised controlled trial. J Affect Disord.
2010;120:272‐276.
26. Perry A, Tarrier N, Morriss R, McCarthy E, Limb K. Randomised
controlled trial of efficacy of teaching patients with bipolar disorder
to identify early symptoms of relapse and obtain treatment. BMJ.
1999;318:149‐153.
27. Rouget BW, Aubry J-M. Efficacy of psychoeducational approaches
on bipolar disorders: a review of the literature. J Affect Disord.
2007;98:11‐27.
 |
10       CHEN et al.
28. Candini V, Buizza C, de Girolamo G, et al. Structured group psy-
choeducation for bipolar patients: a study of effectiveness in Italy.
Compr Psychiatry. 2013;54:e18.
29. Colom F, Vieta E, Sanchez-Moreno J, et al. Group psychoeducation
for stabilised bipolar disorders: 5-­year outcome of a randomised
clinical trial. Br J Psychiatry. 2009;194:260‐265.
3 0. Fialko L, Garety PA, Kuipers E, et al. A large-­scale validation study
of the Medication Adherence Rating Scale (MARS). Schizophr Res.
2008;100:53‐59.
31. Thompson K, Kulkarni J, Sergejew A. Reliability and validity of a
new Medication Adherence Rating Scale (MARS) for the psychoses.
Schizophr Res. 2000;42:241‐247.
32. Faurholt-Jepsen M, Frost M, Vinberg M, Christensen EM, Bardram
JE, Kessing LV. Smartphone data as objective measures of bipolar
disorder symptoms. Psychiatry Res. 2014;217:124‐127.
33. Miklowitz DJ, Otto MW, Frank E, et al. Intensive psychosocial inter-
vention enhances functioning in patients with bipolar depression:
results from a 9-­month randomized controlled trial. Am J Psychiatry.
2007;164:1340‐1347.
3 4. Lobban F, Taylor L, Chandler C, et al. Enhanced relapse preven-
tion for bipolar disorder by community mental health teams:
cluster feasibility randomised trial. Br J Psychiatry. 2010;196:
59‐63.
35. Parikh SV, Zaretsky A, Beaulieu S, et al. A randomized controlled
trial of psychoeducation or cognitive-­behavioral therapy in bipolar
disorder: a Canadian Network for Mood and Anxiety treatments
(CANMAT) study [CME]. J Clin Psychiatry. 2012;73:803‐810.
36. Castle D, White C, Chamberlain J, et al. Group-­based psychosocial
intervention for bipolar disorder: randomised controlled trial. Br J
Psychiatry. 2010;196:383‐388.
37. Colom F, Vieta E, Martínez-Arán A, et al. A randomized trial on the
efficacy of group psychoeducation in the prophylaxis of recur-
rences in bipolar patients whose disease is in remission. Arch Gen
Psychiatry. 2003;60:402‐407.
38. Chen R, Xi Y, Wang X, Li Y, He Y, Luo J. Perception of inpatients
following remission of a manic episode in bipolar I disorder on a
group-based Psychoeducation program: a qualitative study. BMC
Psychiatry. 2018;18:26.
39. Yalom ID, Leszcz M. Theory and Practice of Group Psychotherapy.
New York City, NY: Basic books; 2005.
How to cite this article: Chen R, Zhu X, Capitão LP, et al.
Psychoeducation for psychiatric inpatients following
remission of a manic episode in bipolar I disorder: A
randomized controlled trial. Bipolar Disord. 2018;00:1–10.
https://doi.org/10.1111/bdi.12642