Objective Patients with bipolar disorder and schizophrenia are at high cardiovascular risk; yet, the risk of out-of-hospital cardiac arrest (OHCA) compared with the general population remains scarcely investigated. Methods We conducted a nested case-control study using Cox regression to assess the association of bipolar disorder and schizophrenia with the HRs of OHCA of presumed cardiac cause (2001-2015). Reported are the HRs with 95% CIs overall and in subgroups defined by established cardiac disease, cardiovascular risk factors and psychotropic drugs. Results We included 35 017 OHCA cases and 175 085 age-matched and sex-matched controls (median age 72 years and 66.9% male). Patients with bipolar disorder or schizophrenia had overall higher rates of OHCA compared with the general population: HR 2.74 (95% CI 2.41 to 3.13) and 4.49 (95% CI 4.00 to 5.10), respectively. The association persisted in patients with both cardiac disease and cardiovascular risk factors at baseline (bipolar disorder HR 2.14 (95% CI 1.72 to 2.66), schizophrenia 2.84 (95% CI 2.20 to 3.67)) and among patients without known risk factors (bipolar disorder HR 2.14 (95% CI 1.09 to 4.21), schizophrenia HR 5.16 (95% CI 3.17 to 8.39)). The results were confirmed in subanalyses only including OHCAs presenting with shockable rhythm or receiving an autopsy. Antipsychotics-but not antidepressants, lithium or antiepileptics (the last two only tested in bipolar disorder)-increased OHCA hazard compared with no use in both disorders. Conclusions Patients with bipolar disorder or schizophrenia have a higher rate of OHCA compared with the general population. Cardiac disease, cardiovascular risk factors and antipsychotics represent important underlying mechanisms.