Culturally adapted psychoeducation cuts bipolar relapse risk in Pakistani trial

A twelve-session, one-to-one psychoeducation programme tailored to Pakistani culture and language substantially delayed the recurrence of mood episodes in adults with bipolar disorder, according to a multicentre randomised controlled trial conducted across seven urban centres in Pakistan. Among 303 adults in remission who were followed for 48 weeks, those who received the culturally adapted psychoeducation (CaPE) alongside treatment as usual had an estimated 46% lower risk of recurrence of any mood episode than those receiving treatment as usual alone, with an adjusted hazard ratio of 0.54 (95% confidence interval 0.42–0.69; p < 0.001). The trial, registered as NCT05223959 and funded by the Canadian Institutes of Health Research, recruited participants between June 20, 2022, and March 31, 2025, from teaching and non-teaching hospitals in Karachi, Lahore, Multan, Rawalpindi, Peshawar, Hyderabad, and Quetta.

Bipolar disorder is a chronic, highly recurrent mood condition in which preventing new episodes, preserving functioning, and restoring quality of life are the central goals of maintenance treatment. Although medication remains the cornerstone of care, many patients continue to experience residual symptoms, repeated mood episodes, impaired social and occupational functioning, and elevated all-cause mortality despite pharmacotherapy. Adjunctive psychosocial interventions such as psychoeducation aim to fill this gap by helping patients understand their illness, recognise early warning signs, maintain treatment adherence, regularise sleep and daily routines, and respond to stressors that might otherwise precipitate a relapse. Systematic reviews and a component network meta-analysis of 39 randomised trials support psychoeducation for reducing relapse risk, but that evidence base comes almost entirely from high-income countries and specialist services, and none of the included psychoeducation trials were conducted in low-resource settings or low- and lower-middle-income countries.

The new trial was designed to test whether a culturally and linguistically adapted version of psychoeducation could work at scale in a setting where structured psychological care is rarely part of routine practice. The intervention was adapted from the Barcelona Bipolar Disorders psychoeducation programme, preserving its core clinical components while changing the mode of delivery and therapeutic engagement. Sessions were conducted in Urdu, used culturally familiar idioms rather than technical jargon, attended to local explanatory models of mental illness, drew on locally meaningful examples and folk narratives, and incorporated examples from religious teachings where relevant and acceptable. Family members or carers could join sessions, with the participant’s agreement, particularly those covering medication interruption and relapse recognition. The investigators chose individual rather than group delivery because stigma, privacy concerns, and discomfort with public discussion of mental illness can substantially reduce engagement with group formats in Pakistan.

Eligible participants were adults aged 18 or older with DSM-5 bipolar I or II disorder confirmed by the Structured Clinical Interview for DSM-5, who were in remission and euthymic at entry, defined as a 17-item Hamilton Depression Rating Scale score below 8 and a Young Mania Rating Scale score below 8. All had been stable on a psychiatric medication regimen for at least three months and lived within the trial catchment area. Exclusions covered active substance use disorder, current manic, hypomanic, depressive, or mixed episodes, active suicidal ideation, and unstable residential arrangements. Of 1228 adults assessed for eligibility, 925 were excluded, leaving 303 randomised in a 1:1 ratio to CaPE plus treatment as usual (155 participants) or treatment as usual alone (148 participants), with allocation stratified by site, bipolar disorder type, and sex assigned at birth. The mean age was 36.1 years, 31.0% were female, 61.1% had bipolar I disorder, and the median education was eight years, with 20.8% reporting no formal education.

CaPE consisted of twelve weekly individual sessions of approximately 60 minutes each, delivered by trained master’s-level clinical psychologists who received two months of training and weekly supervision. Each session combined a brief structured presentation with a practical exercise, such as constructing a life chart, listing relapse triggers, or planning actions for early warning signs. The curriculum covered the concept and causes of illness, symptoms of mania, hypomania, depression, and mixed states, illness course and prognosis including substance misuse, pharmacological treatment and the risks of interrupting it, alternative therapies, recognition of early relapse signs, regularity of habits and routines, stress control, problem-solving, and a final integrative session. Fidelity checks by two independent raters showed 88% adherence to session content and 84% adherence to the prescribed structure, with strong inter-rater reliability. Participants in the control group received routine psychiatric care, which in Pakistan generally consists of medication delivered through medical, psychiatric, and family medicine services with very limited access to structured psychotherapy.

The primary outcome was time from randomisation to recurrence of any mood episode over 48 weeks, assessed with the Longitudinal Interval Follow-up Evaluation, a semi-structured interview that generates weekly psychopathology ratings aligned to DSM-5 definitions. Outcome assessors were masked throughout follow-up, worked separately from therapists where possible, and reminded participants not to disclose their allocation. Recurrence was recorded in 127 of 155 participants (81.9%) in the CaPE group and 138 of 148 (93.2%) in the treatment-as-usual group. Because the proportional-hazards assumption was violated on the Schoenfeld residual test, the prespecified primary estimate used robust standard errors, yielding the adjusted hazard ratio of 0.54; the standard Cox model gave an identical point estimate, indicating the result was not sensitive to the choice of variance estimator. The adjusted survival curves separated within the first 12 weeks and remained separated throughout follow-up.

Secondary time-to-event analyses pointed in the same direction. CaPE was associated with a reduced risk of recurrent major depressive episodes (adjusted hazard ratio 0.58, 95% CI 0.45–0.75) and of recurrent manic or hypomanic episodes (adjusted hazard ratio 0.65, 95% CI 0.50–0.85). On the continuous secondary outcomes, measured at baseline and at 3, 6, and 12 months, recovery scores on the Bipolar Recovery Questionnaire were higher in the intervention group at every post-baseline assessment, with adjusted between-group differences of 161.8 points at 3 months, 179.6 points at 6 months, and 144.8 points at 12 months. Knowledge and attitudes measured with the Bipolar Knowledge and Attitudes Questionnaire improved consistently, with differences of 71.7, 81.2, and 78.2 points at the three timepoints. Among illness-perception domains, illness coherence was the most responsive, while timeline, consequences, personal control, treatment control, cyclical timeline, and emotional representation showed little consistent evidence of sustained between-group differences.

Medication adherence, measured with the 4-item Morisky scale where lower scores indicate better adherence, improved by an adjusted −0.4 points at 3, 6, and 12 months. Depressive symptoms on the Hamilton scale showed robust improvement, with adjusted differences of −5.0, −5.8, and −5.3 points at 3, 6, and 12 months respectively, all statistically significant. Manic symptoms improved early, with differences of −2.4 points at 3 months and −3.2 points at 6 months, but the 12-month difference of −1.8 points did not reach statistical significance (p = 0.062). Health-related quality of life showed a mixed pattern: EQ-5D visual analogue scores favoured CaPE at all timepoints by roughly 4.6 to 4.7 points, while the EQ-5D index difference was significant at 3 and 6 months but not at 12 months. Follow-up completion exceeded 93% at every post-baseline timepoint, and six adverse events were recorded, four of them serious, none judged related to the intervention.

The investigators frame these results against the landmark UK PARADES trial, a pragmatic multicentre study that compared structured group psychoeducation with an active peer-support intervention in remitted bipolar disorder and found no overall superiority, though psychoeducation was more acceptable and appeared to benefit participants earlier in illness course. The authors note that their findings extend rather than contradict PARADES: in Pakistan, structured peer support of the kind used as the PARADES comparator is not routinely available, so usual care represents the realistic alternative. A previous pilot of CaPE in 34 adults in Karachi had suggested benefits for adherence, knowledge, quality of life, and relapse prevention, and the new trial confirms those signals in a larger, geographically broader sample. In an exploratory analysis, bipolar II disorder was associated with a lower recurrence risk than bipolar I disorder (hazard ratio 0.62, 95% CI 0.48–0.79), a non-randomised comparison the authors say warrants cautious interpretation but may help identify patients at greatest risk.

The trial has limitations that the authors acknowledge. Participants, therapists, and treating clinicians could not be masked to allocation, and the early separation of the survival curves, including a steep early rise in recurrence in the control arm, could partly reflect expectancy effects among participants who knew they had not received the active intervention. The control condition was heterogeneous by design, the euthymic and stably medicated sample may not generalise to acutely unwell patients, people with active suicidality or substance use disorder, or those without stable housing, and recruitment from urban outpatient services leaves rural populations and people not engaged with care unrepresented. Secondary outcomes were exploratory, with no adjustment for multiplicity, and race and ethnicity data were not collected. The authors call for health economic evaluation, process evaluation to identify which intervention elements drove benefit, and longer follow-up to test whether gains persist and whether booster sessions are needed. They argue that the task-sharing model, delivered by master’s-level psychologists rather than scarce specialists, offers a scalable template for health systems serving South Asian and other underserved communities where stigma, linguistic mismatch, and limited psychotherapy infrastructure constrain access to maintenance care.

Subject of Research: Culturally adapted psychoeducation for relapse prevention in bipolar disorder in Pakistan

Article Title: Culturally adapted psychoeducation for bipolar disorder: a multicentre, assessor-masked, randomised controlled trial

Article References: Husain, M. I., Asif, M., Khoso, A. B., Umer, M., Shakoor, S., Kiran, T., Tahir, M. R., Rajput, A. H., Saqib, M., Husain, M. O., Masood, S., Mufti, A. A., Naeem, F., Bhatia, M. R., Siddiqui, K. M. S., Tamizuddin, A., de Oliveira, C., Wang, W., Xue, S., … Chaudhry, I. B. (2026). Culturally adapted psychoeducation for bipolar disorder: a multicentre, assessor-masked, randomised controlled trial. eClinicalMedicine, 101, Article 104271. https://doi.org/10.1016/j.eclinm.2026.104271

Image Credits: AI Generated

DOI: 10.1016/j.eclinm.2026.104271

Keywords: bipolar disorder, psychoeducation, randomised controlled trial, cultural adaptation, relapse prevention, Pakistan, medication adherence, low- and middle-income countries, mental health, depressive symptoms, recovery, task-sharing