Background: Physical restraint (PR) has to be performed in a well-organized, structured, and careful manner. The safety and effectiveness of the PR procedure in clinical practice are crucial to patient care. Purposes: The aim of this study was to examine the effect in neurological intensive care units of a multidisciplinary PR reduction program on the overall incidence rate of PR, the monthly rate of PR, and themonthly rate of PR ofmore than and less than 24 hours, respectively, after controlling for associated factors. Methods: A before-and-after study design was conducted in three neurological intensive care units (total of 45 beds) of a medical center in northern Taiwan. A multidisciplinary PR reduction program was implemented, with four phases of data collection. Data on the number and duration of PR were extracted from a database. In the first preintervention phase, data on the rate (number or frequency) and duration of PR were collected for 12 months before program implementation. The next three data collection phases were 1 month (introduction), 3 months (intermediate), and 6 months (maintenance) after programimplementation. The chi-square test, one-way analysis of variance with Scheffe’s post hoc tests, and logistic regression method were used for data analysis. Results: The results showed that PR was used 72.8 mean times per month before program implementation. After implementation, the mean times per month fell to 40 after the first postimplementation month (introduction), 53.3 after 3 months (intermediate), and 36.2 after 6 months (maintenance). Implementation of the PR reduction program resulted in a decreased incidence rate of PR from 6.1% to 3.2% (event/total patient day) and a decrease in the percentage of patients who were restrained for more than 24 hours from 69.6% to 34.1%. The results of the logistic regressionmodel showed that PR of more than 24 hours was significantly less in the postintervention phases: introduction phase (odds ratio [OR] = 0.31), intermediate phase (OR = 0.16), and maintenance phase (OR = 0.18). Moreover, PR of more than 24 hours was significantly higher in patients with endotracheal intubation (OR = 1.51) and nasogastric intubation (OR = 2.16) and with Glasgow Coma Scale scores of 6Y8 (OR = 2.36), 9Y12 (OR = 2.40), and 13Y15 (OR = 2.15).Conclusions/Implications for Practice: The results support that a multidisciplinary collaborative program has the potential to decrease the incidence of PR in neurological intensive care units. A standardized protocol and policies are essential for healthcare professionals to effectively use PR as well as to effectively improve the quality of care that is provided to intensive care patients.