Introduction <p>New-onset atrial fibrillation (NOAF) in intensive care units (ICU) carries a high risk of recurrence and atrial fibrillation (AF) related complications. However, evidence on the safety of anticoagulation (AC) in critically ill patients with NOAF is limited. This study aimed to evaluate the impact of AC initiation on clinical outcomes in this population.</p> Methods <p>A multicentre, retrospective cohort study was conducted across four hospitals in Saudi Arabia from January 1, 2018, to February 28, 2024. Adult ICU patients (≥ 18 years) with documented NOAF were included and categorized into two groups: those who received anticoagulation (AC group) and those who did not (no AC group). The study outcomes were the incidence of bleeding or thrombotic events during a 1-year follow-up.</p> Results <p>A total of 179 patients were included, with 52.5% in the AC group and 47.4% in the no AC group. The overall bleeding incidence was 21.2% and did not differ between AC and non-AC groups (21.3% vs. 21.2%). Thrombotic events were infrequent (3.9%) with no significant between-group difference. Atrial fibrillation recurrence within one year occurred in 20.1% of patients and was similar across groups. After competing-risk analysis, therapeutic anticoagulation was not associated with increased bleeding risk (SHR 0.97; 95% CI 0.45–2.05). Prior bleeding history independently predicted bleeding. In time-varying Cox analysis, anticoagulation was not significantly associated with 30-day mortality (HR 0.81; 95% CI 0.38–1.70), whereas vasopressor use was associated with higher mortality risk.</p> Conclusions <p>In critically ill patients with NOAF, anticoagulation did not significantly alter the risk of bleeding or thrombotic events within 1 year of follow-up. These findings highlight the need for further prospective studies to guide anticoagulation decisions in this high-risk population.</p>