Background The purpose of this study is to determine whether new-onset postoperative atrial fibrillation (NOPAF) among patients after hip arthroplasty can predict 1-year mortality. Methods All patients over 65 years who underwent hip arthroplasty from January 2013 to December 2017 in a Chinese tertiary hospital were retrospectively analyzed. Patients with paroxysmal and persistent atrial fibrillation were ruled out. 2438 patients were identified to be eligible. The primary endpoint was 1-year mortality after the arthroplasty. Results Among the 2438 patients, 101 (4.1%) had NOPAF and 2337 (95.9%) had not. Only the current use of beta blocker could predict the occurrence of NOPAF after hip arthroplasty. 1-year mortality for patients with NOPAF was significantly higher than that for patients without NOPAF (70.3% vs 19.0%; p < 0.001). Anti-arrhythmic and anticoagulant treatments were related to 1-year mortality, respectively. With multivariate analysis, NOPA was the most significant variable related to 1-year mortality (hazard ratio 7.8, 95% CI 2.9-24.6). Conclusions Among elderly patients after hip arthroplasty, 1-year mortality is increased significantly for patients with NOPAF.