Psychoeducation for bipolar disorder

Summary

Bipolar disorder is a complex disorder of mood and behaviour that requires a multimodal treatment approach. In the past 10 years there has been growing interest in psychoeducational interventions delivered as adjuncts to conventional management. Several studies have tested the effectiveness of psychoeducational treatments delivered in a variety of formats. In this article we assess the evidence for the efficacy of these interventions and consider the likely future role of structured psychoeducational treatments in clinical practice.

What is psychoeducation?

Most clinicians would agree that patients with any chronic disorder – such as diabetes, epilepsy or ischaemic heart disease – should, as part of their routine care, be given accurate information about their diagnosis, treatment and prognosis and about how they can help themselves to stay well. In broad terms, this kind of information can be considered ‘psychoeducation’. Similarly, psychoeducation as applied to bipolar disorder can be defined as any intervention that educates patients and their families about their illness with a view to improving their long-term outcome.

Although the principal goal of psychoeducation for bipolar disorder is to provide accurate and reliable information, additional objectives include teaching patients self-management skills and making them more able to make informed decisions about their own management within the context of a collaborative working relationship with their clinical team. Where possible, psychoeducational interventions should also be personalised, for example, by taking account of an individual's unique pattern of illness, their risk factors for relapse and their current social circumstances. Although most patients with bipolar disorder who are currently managed within community mental health teams in the UK will have access to good psychoeducational material, a formal structured psychoeducational intervention is only rarely available.

How can psychoeducation for bipolar disorder be delivered?

Psychoeducational interventions can be delivered in a number of ways. They can be provided to individual patients (either one-to-one or within a group setting), to their families and/or carers and to groups of professionals (for example, training in psychoeducational methods for community psychiatric nurses). A range of media are used, including written, audio, video and interactive or online delivery and many interventions blend more than one delivery method to maximise the impact and retention of material.

How effective are psychoeducational interventions for bipolar disorder?

Although the first published reports of psychoeducation for bipolar disorder date back to the 1980s and early 1990s (Cochran 1984; Harvey 1991; Peet 1991; Van Gent 1993), many of these studies had significant methodological shortcomings (including small sample sizes, brief follow-up periods and no control group comparisons). However, over the past decade a number of higher-quality randomised clinical trials of psychoeducational interventions have been published. These studies are summarised in Table 1 and are discussed in more detail below.

Individual psychoeducation

There is a paucity of studies examining individual or ‘one-to-one’ psychoeducation. Only one randomised trial to date has compared one-to-one sessions of psychoeducation plus routine care against routine care alone (Perry 1999). In this study of 69 people with bipolar disorder, the intervention was focused on teaching them to identify prodromal symptoms of depressive and manic relapse and on producing and rehearsing an action plan once these prodromes had been recognised. Most participants received between 7 and 12 sessions, which were delivered by a research psychologist. The intervention is reported as having been helpful in terms of longer times to manic relapse, fewer manic relapses and better social and occupational functioning in the treatment group, but there was no effect for time to first depressive relapse or number of depressive relapses throughout the 18-month follow-up period. This finding for depression is of note when we consider that most of the morbidity associated with bipolar disorder is caused by chronic depressive rather than relapsing manic features (Judd 2002, 2003).

Family-focused psychoeducation

Miklowitz and colleagues have pioneered family-focused psychoeducational treatments for bipolar disorder (Miklowitz 2000, 2003, 2008; Rea 2003). Their family-focused treatment involves all available immediate family members in 21 one-hour sessions delivered over 9 months (12 weekly, then 6 fortnightly, then 3 monthly). It comprises three consecutive modules:

1. 1 psychoeducation: 7 sessions focusing on the signs and symptoms of bipolar disorder, the aetiology of bipolar episodes according to a stress-vulnerability model and the development of strategies to prevent relapses;

2. 2 communication enhancement training: 7–10 sessions using role-play and between-session rehearsal to teach skills for active listening, ways to deliver positive and negative feedback and constructive ways to request changes in behaviour;

3. 3 problem-solving skills training: 4–5 sessions in which participants learn to identify specific family problems that might contribute to relapse and develop skills for finding acceptable solutions to these problems.

In a study that compared 21 sessions of family-focused treatment with a treatment of two family-education sessions plus crisis management for 101 patients with bipolar disorder, Miklowitz and colleagues reported benefits for family-focused treatment at both 1 year (Miklowitz 2000) and 2 years (Miklowitz 2003) (Table 1). Specifically, at 1 year, the family-focused treatment group had longer delays to relapse and fewer relapses overall, but although they had fewer depressive symptoms, there was no effect seen for manic symptoms (Miklowitz 2000). At 2-year follow-up, the family-focused treatment group had fewer relapses, longer periods between relapses, greater reductions in mood symptoms overall and better medication adherence than patients receiving the control intervention (Miklowitz 2003). However, it should be noted that, strictly speaking, family-focused treatment was not significantly different from the control intervention in terms of preventing relapses – 11 out of 31 (35%) patients in the family-focused treatment group relapsed versus 38 out of 70 (54%) in the control group.

Family-focused treatment has also been compared with individual psychoeducational treatment by Miklowitz's group (Rea 2003). In a 2-year follow-up study of patients with bipolar disorder recruited soon after an admission to hospital, there was no statistically significant difference between relapse in the family-focused treatment group and the individual psychoeducation group.

Family-focused treatment for bipolar disorder has also been adapted for use with an adolescent population. In a 2003 trial comparing this with ‘enhanced care’ (three family sessions focused on relapse prevention), there were no differences between groups on rates of recovery from the index episode or in time to recurrence of depression or mania, although the family-focused treatment group spent less time in depressive episodes (Rea 2003).

On balance, although there is some evidence that family-focused psychoeducational interventions may be of benefit for some people with bipolar disorder, the evidence for their effectiveness in terms of preventing relapse is currently limited.

Group psychoeducation

Colom & Vieta in Barcelona have developed a group psychoeducation programme for people with bipolar disorder (Colom 2006). This intervention comprises 21 sessions of 90 minutes delivered weekly by two psychologists to groups of between 8 and 12 patients. Four main areas are targeted: illness awareness; adherence to treatment; early detection of prodromal symptoms and recurrences; and lifestyle regularity. The full list of sessions is presented in Box 1. Each session begins with a 30–40 minute presentation on the topic of the day, followed by a related exercise (for example, drawing a life chart or compiling a list of potential triggers for relapse) and concludes with a group discussion.

BOX 1 The 21 sessions of the Barcelona group psychoeducation programme

1. 1 Introduction

2. 2 What is bipolar illness?

3. 3 Causal and triggering factors

4. 4 Symptoms (I): mania and hypomania

5. 5 Symptoms (II): depression and mixed episodes

6. 6 Course and outcome

7. 7 Treatment (I): mood stabilisers

8. 8 Treatment (II): anti-manic agents

9. 9 Treatment (III): antidepressants

10. 10 Serum levels: lithium, carbamazepine and valproate

11. 11 Pregnancy and genetic counselling

12. 12 Psychopharmacology versus alternative therapies

13. 13 Risks associated with treatment withdrawal

14. 14 Alcohol and street drugs: risks in bipolar illness

15. 15 Early detection of manic and hypomanic episodes

16. 16 Early detection of depressive and mixed episodes

17. 17 What to do when a new phase is detected

18. 18 Regularity

19. 19 Stress-management techniques

20. 20 Problem-solving techniques

21. 21 Final session

(Colom 2006)

This intervention has been assessed in a randomised trial involving 120 patients where the control intervention was 20 weekly group sessions with no specific psychoeducational content (Colom 2003a). The participants, most of whom had bipolar I disorder, had to have been euthymic for at least 6 months before the start of the trial. Follow-up was monthly for 2 years. In the group psychoeducation arm significantly fewer patients experienced a relapse, relapsing patients had fewer recurrences and the time to depressive, manic, hypomanic and mixed relapses was longer compared with patients in the control arm (Table 1).

In a subanalysis of participants in this study who had a comorbid personality disorder, those in the group psychoeducation arm were less likely to relapse, spent a longer period well before relapsing and had fewer relapses overall (Colom 2004). Similarly, for the subgroup of patients who were on lithium, levels were significantly higher and more stable for those who received group psychoeducation (Colom 2005).

In a similar study, Colom and colleagues tested their group intervention in a smaller sample of patients who were known to be fully adherent with medication and found that those in the treatment group had significantly fewer relapses overall and fewer depressive relapses over a 2-year follow-up period (Colom 2003b).

A 5-year follow-up of the 120 patients enrolled in the original randomised controlled trial was published recently (Colom 2009 †). Data were available for 50 of the 60 patients in the treatment arm and 49 of the 60 in the control arm. Overall, the treatment group had a significantly longer time to any mood episode recurrence, had fewer total recurrences, spent fewer days acutely unwell and had a lower median number of days in hospital. These findings suggest that group psychoeducation for bipolar disorder has important long-term prophylactic effects.

Caregiver group psychoeducation

Reinares and colleagues have compared caregiver group psychoeducation (a relative or partner living with the patient) using the content of the Barcelona group psychoeducation programme versus no specific intervention (Reinares 2008). At 1-year follow-up, patients whose caregivers had been randomised to the group intervention had lower rates of relapse and longer relapse-free intervals.

Psychoeducation delivered as part of a comprehensive management programme

Simon and colleagues (2005) in the USA have conducted a large population-based study of 451 patients with bipolar disorder randomised to either a comprehensive treatment programme or treatment as usual (Table 1). Psychoeducation in this programme was delivered by a nurse care manager as a part of a comprehensive ‘systematic care-management’ programme, which included initial assessment and care-planning, monthly telephone monitoring, coordination with the treating team and a ‘structured group psychoeducational programme’. This group programme was adapted from Bauer & McBride's Life Goals Program (Bauer 2003) and comprised five weekly sessions (phase 1) followed by twice-monthly sessions for the remaining time (phase 2; up to 24 months total). Phase 1 addressed the nature of bipolar disorder, triggers and early symptoms of relapse, and self-management strategies for relapse. Phase 2 used problem-solving techniques to focus on the accomplishment of specific life goals.

Simon and colleagues have reported outcome analyses for this trial at both 12-month and 24-month follow-up that show an effect of the overall intervention on the experience of manic symptoms but not for depressive symptoms (Simon 2005, 2006).

Psychoeducation v. other psychosocial interventions

Very few high-quality studies have directly compared psychoeducational interventions with other psychosocial treatments for bipolar disorder. However, a systematic review of psychosocial interventions for preventing relapse in bipolar disorder concluded that cognitive–behavioural therapy, group psychoeducation and possibly family-focused psychoeducation may be beneficial adjuncts to medication maintenance treatments (Beynon 2008). It found no evidence for the effectiveness of either care management or integrated group therapy in preventing relapse.

Why might psychoeducation be effective?

There are several mechanisms by which psycho-educational interventions might exert their therapeutic effect. At present it is not known whether the active ingredient is the ‘group experience’, the educational material itself, some other non-specific psychotherapeutic effect or, as seems likely, some combination of all three.

It is notable that, on current evidence, group interventions appear to be more effective than both family and individual interventions. This suggests that patients with bipolar disorder might benefit most from sharing experiences, expertise and insights about their illness within a supportive group setting. It is also plausible that the personality, charisma and expertise of a group facilitator will be important for maximising the benefit gained from group sessions (this is as yet an underresearched question). Group psychoeducation also operates within the framework of what might be considered a more ‘medical’ view of bipolar illness, where the biological as well as social and psychological aspects of the disorder are acknowledged and where the synergistic benefits of both medication and psychosocial interventions are emphasised. It seems possible that many patients with bipolar disorder find this kind of formulation of their illness helpful.

Although there is some evidence that group psychoeducation enhances medication adherence (Colom 2003b), the finding that patients who are already good at adhering to medication also benefit suggests that other, non-medication-related aspects of the intervention are important. These might include improved insight into personal triggers, better recognition of prodromal symptoms, lifestyle changes, reduced drug and alcohol use, regularity of routines, increased competence in self-management techniques and enhanced relationships with family members, caregivers and professionals. The identification of these ‘active ingredients’ within psychoeducational interventions is an important area for future research (Box 2).

BOX 2 Questions for future research

1. • What is the ‘active ingredient’ of psychoeducation?

2. • How do individual interventions compare with group and online interventions?

3. • Are psychoeducational interventions cost-effective and, if so, which is best value for money?

4. • What level of skills or training is required of those who deliver psychoeducation in the NHS?

5. • Is there a dosage effect: does more information or input necessarily mean better outcome?

6. • Which patients are most likely to respond to psychoeducation (should patients with certain clinical characteristics be excluded, e.g. drug and alcohol use disorder, comorbid personality disorder)?

7. • Should resources be targeted at certain ‘high-risk’ groups (e.g. adolescents or young adults with first-episode mania)?

8. • Is it necessary for patients to be in clinical remission for effective psychoeducation to be given?

9. • Is there a role for ‘expert patients’ in the delivery of psychoeducation?

10. • What elements of psychoeducation for bipolar disorder could be delivered effectively to other patient groups, e.g. those with schizophrenia?

Should formal psychoeducation for bipolar disorder be part of routine NHS care?

Although this review has focused on bipolar disorder, clearly several other psychiatric conditions, such as schizophrenia, depression and anxiety disorders, may benefit from a structured approach to delivering psychoeducational material. Many psychiatrists and community psychiatric nurses in the National Health Service (NHS) will feel that they already deliver high-quality psychoeducation to their patients without the need to formalise this

within a new treatment. Although there are clearly several areas (summarised in Box 2) where further research questions need to be explored, it is our view that psychoeducation for bipolar disorder, particularly group psychoeducation, should be available within the NHS to patients with bipolar disorder. Box 3 lists some useful resources for clinicians and patients.

BOX 3 Psychoeducation resources for bipolar disorder

1. • PsychEducation.org: US website run by Jim Phelps, MD at www.psycheducation.org

2. • ‘Bipolar disorder’: Royal College of Psychiatrists at www.rcpsych.ac.uk/mentalhealthinformation/mentalhealthproblems/bipolarmanicdepression.aspx

3. • Miklowitz D (2002) The Bipolar Disorder Survival Guide. Guilford Press

4. • Owen S, Saunders A (2008) Bipolar Disorder. The Ultimate Guide. Oneworld Publications.

5. • Phelps J (2006) Why am I Still Depressed? Recognizing and Managing the Ups and Downs of Bipolar II and Soft Bipolar Disorder. McGraw-Hill.

Several groups across the UK are developing and testing psychoeducational interventions for bipolar disorder. These include programmes of work in Oxford (www.bipolar-foundation.org) and in Lancaster (www.spectrumcentre.org). In Cardiff, we are developing and testing two bipolar psychoeducation interventions: ‘BeatingBipolar. org’ and a ten-session group psychoeducation intervention called Bipolar Education Programme Cymru (BEP-Cymru).

BeatingBipolar.org

BeatingBipolar.org is a web-based interactive psychoeducational intervention for bipolar disorder that has been developed by the Mood Disorders Research Team at Cardiff University. This intervention was developed using an iterative process involving focus groups with patients, their carers, families and health professionals. The purpose of these groups was to advise on the design and content of the programme, as well as to revise and refine the intervention. The primary focus of the intervention is the recognition and self-management of chronic depressive symptoms, depressive relapse and associated functional impairments, although manic features are also covered to some degree. The intervention involves a blending of different delivery mechanisms, with initial face-to-face delivery, followed by written and web-based interactive delivery of factual content and ongoing support via a web forum. The web-based content requires the reader to be engaged in a number of interactive exercises to maximise long-term retention of the material. The key areas covered in the package are listed in Box 4. BeatingBipolar. org is currently being evaluated in an exploratory clinical trial (Simpson 2009).

BOX 4 BeatingBipolar.org modules

1. 1 The accurate diagnosis of bipolar disorder

2. 2 The causes of bipolar disorder

3. 3 The role of medication

4. 4 The role of lifestyle changes

5. 5 Relapse prevention and early intervention

6. 6 Psychological approaches

7. 7 Bipolar disorder and women

8. 8 Advice for family and carers

Bipolar Education Programme Cymru (BEP-Cymru)

Bipolar Education Programme Cymru (BEP-Cymru) is a group psychoeducation treatment for bipolar disorder that has been funded for patients in Wales by the Big Lottery Fund (www.bep-c.org). It is being delivered by the Department of Psychological Medicine at Cardiff University and comprises ten sessions, each lasting 90 minutes, that are facilitated by two senior community psychiatric nurses. The first session is introductory and orients participants to the programme, the next eight sessions follow the content of the eight Beating Bipolar modules listed in Box 4, and the final session is for feedback and conclusions. Sessions begin with a 30 minute presentation on the topic of the day (for example, lifestyle changes or medication) followed by a 30 minute interactive exercise, such as completing a life chart or creating an early warning signature. They conclude with a 30 minute open discussion. Participants are encouraged to share their views and experiences in an open and supportive group environment.

Conclusions

Bipolar disorder is a complex illness that requires a comprehensive programme of treatment. There is some evidence that psychoeducational interventions, particularly group psychoeducation (when delivered alongside standard medical care), help patients in the long-term management of their condition. Group psychoeducation is not currently widely available under the NHS but work is underway to develop and assess the effectiveness of several different psychoeducational interventions for bipolar disorder and, if clinically effective and cost-effective, these treatments may become part of a standard package of care offered to patients and their families.

MCQs

1. 1 Psychoeducation for bipolar disorder is:

   1. a a form of cognitive–behavioural therapy

   2. b a form of dynamic psychotherapy

   3. c a way to improve patients' knowledge about the causes of bipolar disorder and how to manage it better

   4. d contraindicated in bipolar disorder with comorbid personality disorder

   5. e only deliverable by psychiatrists.

2. 2 To date, the best evidence for psycho-educational interventions in bipolar disorder exists for:

   1. a internet-based delivery

   2. b groups

   3. c one-to-one

   4. d groups of caregivers

   5. e groups of community psychiatric nurses.

3. 3 Regarding family-focused psycho-education:

   1. a families are excluded

   2. b patients are excluded

   3. c all immediate family members must be involved

   4. d only willing family members should be involved

   5. e family members with any psychiatric diagnosis should be excluded.

4. 4 Group psychoeducation for bipolar disorder:

   1. a has been pioneered in Barcelona

   2. b refers to caregivers only

   3. c is ineffective

   4. d comprises 20–25 sessions

   5. e is only effective with charismatic group leaders.

5. 5 Mechanisms by which psychoeducation is effective are unlikely to include:

   1. a improved knowledge about the disorder

   2. b experimenting with stopping long-term medications

   3. c learning to recognise relapses earlier

   4. d improved adherence to medication

   5. e moderation of drug and alcohol use.

MCQ answers

1		2		3		4		5
a	f	a	f	a	f	a	t	a	f
b	f	b	t	b	f	b	f	b	t
c	t	c	f	c	f	c	f	c	f
d	f	d	f	d	t	d	f	d	f
e	f	e	f	e	f	e	f	e	f

TABLE 1 Randomised trials of psychoeducation for bipolar disorder

Study, country	Details of intervention, control arm and setting	Delivery	Diagnostic group	Treatment arm (control arm), n	Mood state at entry	Follow-up period and assessments	Significant findings
Perry et al (1999), UK	Between 7 and 12 individual sessions of ‘teaching patients to identify early symptoms of relapse and obtain treatment’, plus routine care. Delivered by a research psychologist in out-patient secondary care. Control arm was routine care only	Individual	DSM–III–R BP–I (n = 63) and BP–II (n =6)	34 (35)	Not stated	Assessments at baseline, 6, 12 and 18 months	Longer time to manic relapse, fewer manic relapses and better social and occupational functioning in the treatment group. No effect for time to first depressive relapse or number of depressive relapses
Miklowitz et al (2000), USA	Family-focused treatment delivered in 21 sessions over 9 months versus a comparison treatment of 2 family education sessions and follow-up crisis management. All patients were maintained on mood-stabilising medication	Family	DSM–III–R BP–I	31 (70)	Recruited ‘shortly after an illness episode’ (about 80% were admitted to hospital at entry)	Three-monthly assessments during treatment phase for 9 months and at 12 months after start of treatments	Family-focused treatment group had fewer relapses, longer delays to relapses and improved depressive (but not manic) symptoms. Improvements were most marked in family-focused treatment families who had high expressed emotion
Colom et al (2003a), Spain	Group psychoeducation (8–12 participants in each group), 21 sessions lasting 90 minutes, each delivered by two psychologists. Content was focused on illness awareness, adherence to treatment, early detection of prodromal symptoms and recurrences, and lifestyle regularity. Control arm was 20 weekly group meetings of 8–12 patients with the same two psychologists but with no specific psychoeducational content. Both arms received standard psychiatric care, including pharmacotherapy	Group	DSM–IV BP–I (n = 100) and BP–II (n =20)	60 (60)	Euthymic for previous 6 months	Assessed monthly during 20 weeks of treatment and for 2 years of follow-up	Treatment group had fewer relapsed patients, fewer recurrences per patient and increased time to depressive, manic, hypomanic and mixed relapses. Number and length of admissions to hospital for each patient were also lower in the treatment group
Colom et al (2003b), Spain	This study only recruited patients who were known to be fully adherent to their medication. The intervention was as per Colom et al (2003a) above	Group	DSM–IV BP–I	25 (25)	Euthymic for previous 6 months	Monthly assessment for 2 years of follow-up	Treatment group had fewer relapses overall and fewer depressive episodes
Rea et al (2003), USA	Family-focused psychoeducational treatment versus individual treatment. In-patient setting initially and on-going treatment with mood-stabilising medication	Family	DSM–III–R BP–I	28 (25)	Manic, recently admitted to hospital	Assessments at 3-month intervals for 1 year during active treatment and for 1 year after treatment	Family-focused treatment group were less likely to be readmitted during the 2-year study period and experienced fewer mood disorder relapses
Miklowitz et al (2003), USA	Same study as Miklowitz et al (2000) above but with a 2-year (rather than 1-year) follow-up period	Family	DSM–IV BP–I (n = 99) and BP–II (n =2)	31 (70)	Recruited ‘shortly after an illness episode’ (about 80% were in-patients at study entry)	Three-monthly follow-up during treatments for 9 months and at 24 months after start of treatments	Family-focused treatment group had fewer relapses and longer survival intervals
Colom et al (2004), Spain	Subanalysis of the Colom et al (2003a) study above analysing a subgroup of patients with a comorbid personality disorder.	Group	DSM–IV BP–I plus any DSM–IV personality disorder	15 (22)	Euthymic for previous 6 months	Assessed monthly during 20 weeks of treatment and during 2 years of follow-up	Treatment group were less likely to relapse, had a longer time to relapse and had fewer relapses overall
Simon et al (2005), USA	Psychoeducation delivered by a nurse care manager as a part of a comprehensive ‘systematic care management’ programme, which included initial assessment and care-planning, monthly telephone monitoring, coordination with the treating team and structured group psychoeducation (adapted from the Life Goals Program) comprising five weekly group sessions followed by twice-monthly sessions for the remaining time, up to 24 months total. The control arm was ‘usual care’. Results analysed at 12 months	Comprehensive care programme	DSM–IV BP–I (n = 336) and BP–II (n =115)	212 (229)	At entry, most patients had manic or depressive symptoms of varying severity	Three-monthly assessment for 24 months but this paper presents findings at 12 months	Treatment group had lower mean mania ratings over the 12 months and less time in hypomanic or manic episodes, but no difference found between groups for mean depression ratings
Colom et al (2005), Spain	Subanalysis of the Colom et al (2003a) study above analysing the subgroup of patients who had been taking lithium during the original trial	Group	DSM–IV BP–I (n = 80) and BP–II (n =13)	49 (44)	Euthymic for previous 6 months	Assessed monthly during 20 weeks of treatment and 2 years of follow-up. Adherence assessed by interviews with patients and relatives and plasma levels of lithium obtained before randomisation and at 6, 12, 18 and 24 months	Lithium levels were higher and more stable in the psychoeducation group
Simon et al (2006), USA	Findings at 24 months of the same study group as in Simon et al (2005) above	Comprehensive care programme	DSM–IV BP–I (n = 336) and BP–II (n =115)	212 (229)	At entry, most patients had manic or depressive symptoms of varying severity	Three-monthly assessment for 24 months	Treatment group had lower mean mania ratings over the 24 months but no difference found between groups for mean depression ratings
Miklowitz et al (2008), USA	Family-focused treatment for adolescents plus protocol pharmacotherapy versus enhanced care plus protocol pharmacotherapy. Family-focused treatment was 21 sessions delivered over 9 months of psychoeducation, communication training and problem-skills training. Enhanced care was three family sessions focused on relapse prevention	Family	58 (mean age 14.5 years) with BP–I (n = 38), BP–II (n = 6) or BP–NOS (n = 14).	30 (28)	Mood episode within the previous 3 months	Assessments every 3 months in year 1 and then every 6 months in year 2	No differences between intervention and control groups for rates of recovery from index episode or in time to recurrence of depression or mania but family-focused treatment group spent fewer weeks in depressive episodes
Reinares et al (2008), Spain	Caregiver group psychoeducation versus no specific intervention. The intervention was twelve 90 minute group sessions focused on knowledge of bipolar disorder and training in coping skills. Patients did not attend the groups	Caregiver group	Patients: DSM–IV BP–I (n = 94) or BP–II (n = 19)	57 (56)	Patients euthymic for at least 3 months prior to study entry	Assessed monthly during the intervention and during the 12 month follow-up period	Intervention group had fewer patients with mood recurrence and longer relapse-free periods. There was an effect for hypomanic and manic relapses but not for depressive or mixed episode relapses
Colom et al (2009), Spain	Five-year follow-up of Colom et al (2003a) study above: 99 out of 120 patients completed follow-up	Group	DSM–IV BP–I (n = 100) and BP–II (n =20)	50 (49)	Euthymic for previous 6 months	Five years of follow-up. Two-monthly assessments.	Treatment group had a longer time to any recurrence, had fewer recurrences overall, spent less time acutely unwell and had a lower median number of days in hospital