Lancet Psychiatry. 2026 Oct;13(10):861-869. doi: 10.1016/S2215-0366(26)00239-7.
ABSTRACT
BACKGROUND: The early stage of bipolar I disorder is considered a crucial period for improving long-term outcomes, yet evidence guiding treatment selection after a first manic episode is scarce. Using data from the BD-CAUSAL Collaboration, we aimed to emulate a pragmatic target trial among individuals aged 16 years or older presenting with a first manic episode with psychotic features.
METHODS: For the target trial, individuals were eligible if they were aged 16 years or older, treated at a first-episode psychosis clinic with a diagnosis of bipolar I disorder between Jan 1, 2006, and Dec 31, 2021, with a manic episode (with or without mixed features) with psychotic symptoms based on DSM-5 criteria; not currently hospitalised; had no previous use of mood stabilisers or second-generation antipsychotics in the past 3 months; and had no history of severe side-effects. Treatment strategies compared were initiating mood stabiliser monotherapy, second-generation antipsychotic monotherapy, or combination therapy. Individuals are followed up for up to 2 years. We emulated the target trial using the BD-CAUSAL Collaboration database that includes participants from North America, Chile, and Spain, and is a subset of a large consortium of observational cohorts. Information on mood stabilisers and antipsychotic prescription was identified via data linkages with outpatient and hospitalisation databases. We assessed eligibility for each person-visit between Jan, 1, 2006, and Dec, 31, 2021, and assigned individuals to the treatment strategy that was compatible with their data. The main outcome was hospitalisation or an emergency department visit due to any psychiatry reason over 2 years. Intention-to-treat and per-protocol effects were estimated using pooled logistic regression adjusted for baseline and time-varying confounders. People with lived experience were not involved in the study design, analysis, or interpretation of findings.
FINDINGS: Among 371 eligible individuals with bipolar I disorder (465 eligible treatment initiations), 141 (38%) were female, 230 (62%) were male, 216 (58%) were White, 155 (42%) were not White, and the median age was 22 years (IQR 19-25). Of 371 patients, 155 (42%) were hospitalised due to psychiatric reasons over a median of 19·8 months (IQR 16·1-24·0). The 2-year hospitalisation or emergency department visit risk was 42·3% (95% CI 32·5 to 61·1) for mood stabiliser monotherapy, 49·5% (42·4 to 60·6) for second-generation antipsychotic monotherapy, and 42·2% (33·3 to 52·5) for combination therapy. Compared with second-generation antipsychotic monotherapy, the risk was lower with mood stabiliser monotherapy (risk difference -7·3% [95% CI -15·8 to -1·4]; risk ratio 0·82 [0·63 to 0·97]) and combination therapy (-7·4% [-16·0 to -2·1]; 0·82 [0·63 to 0·91]). Risks were similar between mood stabiliser monotherapy and combination therapy.
INTERPRETATION: Mood stabiliser-based strategies, alone or in combination, had a lower relapse risk than second-generation antipsychotic monotherapy in the continuation and maintenance treatment after a first episode of mania with psychotic symptoms. These findings would support prioritising mood stabilisers in early treatment and highlight the need for randomised trials for confirmation.
FUNDING: ASISA and National Institute of Mental Health.
PMID:42716056 | DOI:10.1016/S2215-0366(26)00239-7
