| Title: | Long-Term Cardiovascular Risks of Anticholinergic Versus Non-Anticholinergic Antidepressants: A Target Trial Emulation With Negative Control Correction |
| Journal: | Pharmacoepidemiology and Drug Safety |
| Published: | 14 Jul 2026 |
| Pubmed: | https://pubmed.ncbi.nlm.nih.gov/42449494/ |
| DOI: | https://doi.org/10.1002/pds.70434 |
Publication 20282
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
Abstract
PURPOSE: Antidepressants remain widely prescribed worldwide, and the potential cardiovascular risks associated with their anticholinergic properties are poorly understood.</p>
METHODS: We employed an active-comparator new-user design to emulate a target trial spanning 2006-2021 in UK Biobank. Participants aged ≥ 40 years with linked primary care records who newly initiated an anticholinergic vs. a non-anticholinergic antidepressant were included. The primary outcome was hospitalization or death from cardiovascular events. To adjust for measured and unmeasured confounding, we applied propensity score matching (PSM) with proximal causal inference (PCI) and negative control outcome calibration (NCOC) to estimate the intention-to-treat hazard difference (HD). Subgroup analyses were conducted by age, sex, socioeconomic status, lifestyle factors, apolipoprotein E genotype, and major comorbidities. Sensitivity analyses included data-driven negative control selection, alternative anticholinergic burden scale, comparisons across different anticholinergic burden levels, complete-case analysis, a 6-month induction period, and a 2-year washout period.</p>
RESULTS: The study included 24 547 anticholinergic antidepressant users and 20 519 non-anticholinergic users. Measured covariates were balanced among 32 588 PS-matched participants with a median follow-up of 9.1 years. Negative control correction revealed anticholinergic antidepressant use was associated with increased risk of cardiovascular events (HDPSM + PCI: 59.96 per 10 000 person-years [95% CI: 43.32, 76.60]; HDPSM + NCOC: 31.35 [17.95, 45.26]). Elevated risks were observed for ischemic stroke, acute coronary syndrome, heart failure, arrhythmias, arterial disease, and venous thromboembolism. Results were consistent across subgroup and sensitivity analyses.</p>
CONCLUSIONS: Anticholinergic antidepressants may increase the risk of cardiovascular events compared with non-anticholinergic antidepressants. These findings underscore prioritizing therapeutic options with a lower anticholinergic burden in clinical practice.</p>
13 Keywords
- Adult
- Aged
- Antidepressive Agents
- Cardiovascular Diseases
- Cholinergic Antagonists
- Female
- Hospitalization
- Humans
- Male
- Middle Aged
- Propensity Score
- UK Biobank
- United Kingdom
9 Authors
- Yang Xu
- Hong Xu
- Jinxin Guo
- Qoua Liang Her
- Ruyu Li
- Yongjie Lai
- Yue Zhang
- Edwin C. K. Tan
- Siyan Zhan
1 Application
| Application ID | Title |
|---|---|
| 89514 | Integrating analyses of multidimensional exposures on complex diseases |
Enabling scientific discoveries that improve human health