The anxious bipolar phenotype: clinical complexity and treatment response

Abstract Background Anxiety disorders (ANX) affect 30–60% of individuals with bipolar disorder (BD), yet limited research has systematically examined clinical characteristics and treatment patterns in this comorbid population. This study investigated demographic, clinical, and pharmacotherapeutic differences between individuals with BD with and without comorbid ANX. Methods Cross-sectional data from 2,225 adults with BD enrolled in the Mayo Clinic Bipolar Disorder Biobank were analyzed. Participants were assessed for comorbid ANX, demographics, clinical characteristics, medication use, and treatment response using the Alda-A scale. Results Overall, 61% (n = 1,366) had comorbid ANX. Individuals with BD + ANX were younger (40.4 vs. 43.6 years, p < 0.001), more likely female (66.6% vs. 54.8%, p < 0.001), and exhibited higher rates of rapid cycling (64.2% vs. 45.2%, p < 0.001), suicide attempts (40.4% vs. 24.8%, p < 0.001), substance use disorders (63.5% vs. 54.8%, p < 0.001), and somatic comorbidities (MCIRS: 6.68 vs. 5.42, p < 0.001). Pharmacotherapeutically, individuals with BD + ANX were less likely to be currently prescribed lithium, a trend‑level difference (37.1% vs. 47.8%, p = 0.005) and showed a trend towards lower valproic acid use (21.7% vs. 29.6%, p = 0.047), but more likely to receive antidepressants (53.8% vs. 39.5%, p < 0.001), benzodiazepines (39.9% vs. 26.6%, p < 0.001), and gabapentinoids (8.5% vs. 4.5%, p < 0.001). Notably, 17.3% of individuals with BD + ANX received antidepressants without mood stabilizer coverage. Treatment response (Alda-A) scores were significantly lower in BD + ANX group for lithium (4.91 vs. 6.05, p < 0.001) and second-generation antipsychotics (4.67 vs. 5.73, p < 0.001), with a trend‑level reduction observed for mood-stabilizing anticonvulsants (5.16 vs. 6.01, p = 0.005). Similar patterns were observed in both BD-I and BD-II subtypes. Conclusions Individuals with BD + ANX represent a more severely affected subgroup with distinct prescribing patterns favoring antidepressants over mood stabilizers and attenuated response to mood stabilizers. These findings highlight the need for anxiety-informed treatment algorithms recognizing anxiety comorbidity as a negative prognostic factor.