Alcohol use across trials of psychological interventions for bipolar: A systematic review and meta-analysis

Clinical guidelines recommend psychological interventions for bipolar disorder (BD). However, although BD commonly co-occurs with alcohol use disorder (AUD), it is not established how frequently individuals with AUD are excluded from BD psychological intervention trials, or whether this exclusion impacts trial efficacy. Consequently, it is unclear whether evidenced-based treatment decisions can be made for this population.

A systematic review of RCTs of psychological interventions for BD (PROSPERO ref: CRD42023474548) was conducted to explore the proportion of trials i) reporting alcohol-related exclusion criterion, and/or ii) measuring alcohol use at follow-up. Random effects meta-analyses with sub-group comparisons were also conducted, to investigate whether intervention efficacy differed between trials including and excluding individuals with an AUD.

Across the 92 trials identified, 31 (33.7%) reported alcohol-related exclusion, and only 3 (3.3%) measured alcohol use at follow-up. The pooled effect sizes for depressive or manic symptoms did not significantly differ between trials including or excluding individuals with co-occurring AUD across the main analyses, although these may be sensitive to methodological changes. Certainty of evidence was moderate.

Overall, findings indicate that alcohol-related exclusion is common across trials of psychological interventions, but that intervention efficacy does not significantly differ between trials including and excluding individuals with AUD. Consequently, it is recommended that intervention studies do not exclude people with AUD, in order for evidenced-based treatment to be evaluated in those with co-occurring problems.