Background: Mental health- and neurological disorders constitute 13% of the global burden of disease. Alarmingly this burden has risen by 41% in the last 20 years. In low-and-middle-income countries as few as 10% of people living with bipolar disorder receive care. In western countries, the efficacy of psychoeducation, as an add-on treatment to pharmacotherapy in the treatment of symptoms and in relapse prevention initiatives with respect to bipolar disorder, is well documented. Yet, few studies on psychosocial interventions for bipolar disorder have been conducted in a low-income country.

Aim: To determine the effect, feasibility and acceptability of psychoeducation for patients with bipolar disorder on all three levels of the health care system in Rwanda - at the community health centre, district- and university hospital.

Methods: Patients will be randomized into either group A) group-psychoeducation at a referral hospital; or B) group-psychoeducation for both patients and relatives or C) waiting list. Moreover a district trial will test the impact and feasibility of psychoeducation at the district level.

Outcomes: Reduction in symptom severity and incidence of relapse, improved quality of life, medical adherence and knowledge, as well as reduced self-stigmatization.

Perspectives: If proven successful, this is of importance for closing the huge treatment gap in mental health particularly affecting low- and middle-income countries and may reduce the mortality and increase quality of life in the population suffering from bipolar disorder. Furthermore, potential positive outcomes may be implemented in similar low-resource settings elsewhere.

Background: Mental health and neurological disorders constitute 13% of the global burden of diseases. Alarmingly this burden has risen by 41% in the last 20 years. It is estimated that severe mental disorders (i.e. severe depression, bipolar disorder, schizophrenia and other psychotic disorders) have a two to three times higher average mortality compared to the general population. Treatment rates for these disorders are low in low-and-middle-income countries (LMICs), where treatment gaps of more than 90% have been documented.

In western countries, the efficacy of psychoeducation, as an add-on treatment to pharmacotherapy in the treatment of symptoms and relapse prevention initiatives concerning bipolar disorder (BD), is well documented. Yet, few studies on psychosocial interventions for BD have been conducted in low-income countries.

The overall aim of the study is to determine the effect, feasibility and acceptability of psychoeducation for patients with BD in Rwanda.

Methods: The study is divided into a prospective randomized controlled trial (RCT) and a district trial. Patients with bipolar disease type I or II that meet DSM-V diagnostic criteria given by a trained psychiatrist and age ≥ 18 years will be invited to participate.

For the RCT study participants will be randomised to 1) group-psychoeducation for patients and relatives, 2) a waiting list. The RCT will take place at a referral hospital. The district trial compares the impact of psychoeducation given at the district level by mental health nurses with psychoeducation conducted at referral hospitals.

Intervention: Manual-structured group psychoeducation with eight sessions of 90 minutes over eight weeks (at one session per week). Patients will be offered to invite their relatives for 2-3 psychoeducation-days for relatives.

Outcomes: The primary outcome is a reduction in symptom severity, the incidence of relapse and hospitalization. Secondary outcomes include Improved quality of life and medication adherence and knowledge, as well as reduced self-stigmatization. All outcomes will be assessed at baseline, immediately post-intervention, and at the 3 and 6 months follow-up.

Sample size: In the literature on group-psychoeducation, 13 out of 18 RCT's have a reduction in general psychiatric symptom severity, the incidence of relapse and hospitalization as main outcomes. Eighteen RCT's on group psychoeducation for BD were reviewed before a study of Colom et al. was selected as the base of the power calculation. The incidence of relapse in the study was (92%) (55 subjects) in the control group vs 67% (40 subjects) in the psychoeducation group. For this study, a sample size of 40 patients for each arm is required to achieve a level of 80% power with a 5% level of significance when comparing the mean change in each intervention with the control arm via a two-sample t-test. Adjusting for a drop-out rate of 20%: 40/(1-(20/100))= 50 participants will be needed for each group (50 for intervention, 50 for waiting-list and 50 for the district trial).

Randomization: Study participants at the hospital level who meet the inclusion criteria and sign the informed consent form will be randomized individually into either intervention-arm or waiting list through block-randomization with a ratio of 1:1.

Patients at the district level will not be randomized since the number of patients with BD at these levels is unknown and can result in the sample size will be too small. Instead, all will be offered participation.

Ethical Approval The research protocol and study-related documents have been approved by the College of Medicine and Health Sciences Institutional Review Board, Rwanda and The National Council for Science and Technology (NCST) in Rwanda.

Inclusion Criteria:

• A diagnosis of BD type I or II that meets DSM-V diagnostic criteria given by a trained psychiatrist
• No episode in the preceding 4 weeks. Age ≥ 18 years.

Exclusion Criteria:

• Previous participation in any structured psychological intervention
• Insufficient understanding of Kinyarwanda
• Clinical evidence of substantial cognitive impairments.
• Alcohol or drug-dependence