The use of brain-penetrant calcium channel blockers (BP-CCBs) and risk of suicidality and hospitalization among patients with neuropsychiatric disorders: a retrospective comparative cohort study

Background: Multiple studies have suggested a possible positive therapeutic effect of brain-penetrant calcium channel blockers (BP-CCBs) on multiple neuropsychiatric disorders.

Objectives: To assess the risk of suicidality and disease-related inpatient hospitalizations or ER visits in those previously diagnosed with major depression (MDD), bipolar disorder (BD), schizoaffective disorder, and schizophrenia and initiating BP-CCBs versus those who initiated non-brain penetrant CCBs (NP-CCBs).

Methods: We followed a retrospective active comparator cohort design. New users of BP-CCBs and NP-CCBs among patients with MDD, BD, schizoaffective disorder, or schizophrenia were separately identified in 9 claims and EHR databases from across the globe (e.g., Merative MarketScan, Japan Medical Data Center). We separately assessed the occurrence of suicidality and disease-related inpatient hospitalizations or ER visits during the 365 days after treatment initiation. We matched treatment groups on propensity-scores (PSs) to control for observed confounding; including all diagnosed comorbid conditions, all prior and concomitant medication use, demographics, and other variables. To estimate the average treatment effect among the treated (BP-CCBs), we used conditional Cox proportional hazards models. To account for residual bias, we calibrated our estimates against a distribution of effect estimates from 180 negative-control outcomes. Study diagnostics were used to assess the validity of the analysis.

Results: Over 9 databases, 98,427-160,117 BP-CCB new users with MDD were matched to NP-CCB new users, depending on outcome. Among patients with BD, schizoaffective disorder, and schizophrenia the number of matched patients per group was 7,665-17,547, 916-3,039, and 2,122-6,038, respectively. For the outcome of hospitalization among patients with MDD, HRs (95% CIs) ranged from 0.88 (0.68, 1.14) to 1.10 (0.97, 1.26) with a meta-analysis HR of 1.02 (0.95, 1.10). Similarly, HRs (95% CIs) for suicidality ranged from 0.67 (0.44,1.01) to 1.46 (0.96,2.23) with a meta-analysis HR of 1.01 (0.85,1.20). Meta-analytic HRs among patients with BD were 0.97 (0.81, 1.15) and 1.04 (0.83, 1.30) for hospitalization and suicidality, respectively. Among patients with schizophrenia the HR was 0.96 (0.62,1.48) for suicidality, while no database passed diagnostics for hospitalization. Results for patients with schizoaffective disorder [HR 1.06 (0.56,2.01) for suicidality] showed similar null effects.

Conclusions: Across outcomes and subgroups, incidence was similar between initiators of BP-CCBs and NP-CCBs, ruling out all but very modest benefits.