- Most real world bipolar depression outpatients are RCT-ineligible; only 43.4% met a TEI cutoff of 75.
- RCT-eligible patients reported markedly greater patient-perceived improvement after pharmacotherapy (OR 9.83, 95% CI 4.17-23.18; p < 0.001).
- Evidence gaps require targeted RCTs, broader eligibility, subgroup analyses and observational cohorts to improve care for underrepresented bipolar depression patients.
J Affect Disord. 2026 Aug 10:122363. doi: 10.1016/j.jad.2026.122363. Online ahead of print.
ABSTRACT
BACKGROUND: The present study aimed to address the gap between treatment guidelines and real-world management of acute bipolar depression by examining the prevalence, clinical features, and patient-perceived improvement after pharmacotherapy among bipolar outpatients who would be ineligible for an average randomized controlled trial (RCT).
METHODS: This cross-sectional study classified bipolar depressed outpatients as RCT-eligible or RCT-ineligible using the Trial Eligibility Index (TEI), while retaining the TEI as the primary continuous measure of RCT representativeness. Patient-perceived improvement after pharmacotherapy was assessed during routine clinical evaluation. Descriptive analyses were followed by multivariable regression and exploratory random-forest variable-importance analyses.
RESULTS: At the prespecified TEI cutoff of 75, 53/122 patients (43.4%) were RCT-eligible. RCT-eligible patients had greater odds of reporting perceived improvement than RCT-ineligible patients (OR = 9.83, 95%CI 4.17-23.18; p < 0.001). Sensitivity analyses produced the same classification at a cutoff of 70 and preserved the direction of association at a cutoff of 80 (OR = 9.33, 95%CI 1.14-76.09; p = 0.013). The continuous TEI was associated with several psychiatric and medical characteristics in multivariable analysis.
CONCLUSIONS: Targeted RCTs, broader eligibility criteria, subgroup analyses, and observational cohorts are needed to improve the evidence base for underrepresented patients with bipolar depression.
PMID:42575197 | DOI:10.1016/j.jad.2026.122363
Share Evidence Blueprint
Save to Google Notes

Search Google Scholar
Save as PDF
⭐ My Revision List

