Optimal parameters for evaluating the effectiveness of prone positioning in acute respiratory distress syndrome (ARDS) remain undefined. This study aims to investigate the relationship between dynamic change in mechanical power during prone positioning and mortality in patients with ARDS.This was a single-center retrospective cohort study.The Center of Critical Care Medicine of Peking Union Medical College Hospital.ARDS patients who underwent prone positioning while receiving invasive mechanical ventilation were enrolled.None.A total of 1078 patients (720 [66.8%] male; median age, 63 yr [interquartile range (IQR), 51-72 yr]) were enrolled. The median duration of selected prone position sessions was 15.0 hours (IQR, 9.0-17.0 hr). ICU mortality was 20.9% (225/1078). Mechanical power during prone positioning increased in nonsurvivors compared with survivors (0.8 × 10-2 J/min/kg [IQR, -3.3 to 5.6 × 10-2 J/min/kg] vs. -0.6 × 10-2 J/min/kg [IQR, -4.9 to 3.3 × 10-2 J/min/kg]; p = 0.001). Patients with increased mechanical power during prone positioning had higher ICU mortality (23.9% vs. 17.8%; p = 0.011), in-hospital mortality (25.2% vs. 19.5%; p = 0.018), and 28-day mortality (33.2% vs. 25.4%; p = 0.002). Multivariable time-dependent Cox proportional hazards model confirmed that increased mechanical power was independently associated with higher ICU mortality risk (hazard ratio for each 10-U increase in 10-2 J/kg/min 1.071; 95% CI, 1.020-1.125; p = 0.007). Additionally, increased mechanical power during prone positioning was also independently associated with higher in-hospital mortality risk, 28-day mortality risk, and fewer ventilator-free days.Dynamic increases in mechanical power during prone positioning are linked to higher ICU mortality in ARDS patients. Continuous monitoring of mechanical power may guide patient selection for prone positioning.Copyright © 2025 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.