Objective:To explore the application effect of the modified early warning score (MEWS) in the allocation of nursing human resources per shift in ICU.Methods:A total of 470 critically ill patients admitted to the ICU of Xuancheng People′s Hospital from January to December 2019 were selected as the experimental group. In each shift, the nursing leader conducted MEWS score for patients in the ward, and assigned nurses at the corresponding level according to the MEWS score. A total of 346 critically ill patients admitted to ICU from January 2018 to December 2018 were selected as the control group. The head nurse assigned nurses daily according to clinical experience.The incidence of accidental events, nurse error events and medical satisfaction were compared between the two groups.Results:The incidences of accident and nurse error in ICU patients in the experimental group were 5.96% (28/470) and 12.13% (57/470), respectively, which were lower than those in the control group [11.56% (40/346) and 17.92% (62/346)]. The difference between the two groups was statistically significant ( χ2=8.19, 5.37, both P<0.05). The satisfaction degree of the experimental group was 98.63 ± 1.06 and 95.96 ± 1.84, respectively, which were higher than that of the control group 92.13 ± 2.17 and 90.43 ± 2.91, and the difference between the two groups was statistically significant ( t=7.62, 8.56, both P<0.05). Conclusions:Using MEWS score to guide the allocation of nursing human resources in each shift of ICU is helpful to reduce the incidence of accidental events in ICU patients and the incidence of nurses′ errors, ensure patient safety and improve medical satisfaction.