BACKGROUND: Restrictive methods such as physical restraint are utilized in mental health facilities as part of patient care by nurses to prevent inpatients from injuring themselves, other patients, or hospital staff (Whittington, Baskind, & Paterson, 2006) but the use of restraints as a non-pharmacological intervention has long been a controversial practice (Barton, Johnson, & Price, 2009). The aim of the study was to reduce prevalence physical restraint incidents and improve patient safety through early intervention. The study was conduct in a 30-bed capacity acute inpatient psychiatric unit at Prince Sultan Military Medical City, Riyadh, Saudi Arabia. Participants included Inpatient Psychiatric unit nurses including Head Nurses, Charge Nurses, registered nurses and patients admitted under admission criteria. METHODS: The FOCUS-PDCA methodology as a quality improvement tool deployed to undertake this project. A pilot study conducted as a baseline assessment to determine contributing risk factors in the increase of physical restraint incident through documentation review of medical records for 3 months. Restraint Incident Log Book (RILB) used for data extraction on daily number of physically restraint patients against the total occupied bed per days to indicate physical restraint rate. Literature review done to identify the effective monitor tool for data collection. INTERVENTION: A Root Cause Analysis conducted to understand the performance variation to propose effective measure in reduction of physical restraint rate. Through formal discussion, the staff identified early recognition through risk assessment and early intervention of verbal de-escalation for implementation to improve the performance. RESULTS: Upon initiation of Quality Improvement Project on March 2020 and implementation of the action plan, the performance improvement was moving towards the target. After the implementation of training education program, early risk assessment, early intervention, restoration therapeutic room and effective verbal de-escalation techniques, the performance restraint rate reached target benchmark of 15 per 1,000 beds day. In fact, the prevalence of physical restraint rate reduced significantly after two PDSA cycles, from 52 per 1,000 beds day in mid-March 2020 to zero per 1,000 beds day by end of August 2020 and sustained improvement within the target benchmark thereafter. CONCLUSIONS: Identification of contributing factors and implementation of effective communication, education training program, early risk assessment-intervention and renovation of therapeutic room was a hallmark for the improvement in physical restraint rate. Leadership also played an essential role in the success of the project by upholding the departmental goal of zero patient harm and creating the culture of safety within psychiatric settings. Although the road to achieve zero harm far off but it is not impossible, additional work is need in several key areas, specifically in assigning a dedicated therapeutic room and introduce non-crisis intervention to replace current outdate practice.