Objective:To explore the effect of three-level delirium nursing management process on the treatment outcome and quality of care in the pediatric intensive care unit (PICU).Methods:Since January 2019, three-level delirium management process for PICU were developed. 75 children were randomly selected as the experimental group for 6 months, and 75 patients who received routine care between July and December 2018 were selected as the experimental group. The control group received routine nursing care, and compared the differences in the incidence, duration, and length of PICU stay in the intensive care unit (DICU) between the two groups of children. The family members' satisfaction with the quality of care and the evidence-based knowledge of PICU nurses were compared before and after implementation.Results:The incidence of DICU in the experimental group was 21.33% (16/75), the incidence of unsafe events related to DICU was 2.67% (2/75), and the control group was 40.00% (30/75), 12.00% (9/75). The difference was statistically significant ( χ2 values were 6.145, 4.807, P<0.05); the duration of the DICU in the test group (14.16±4.25) hours, and the length of ICU hospitalization (48.62 ± 8.85) hours were less than (17.37±5.36) hours and (52.03±8.10) hours in the control group, the difference was statistically significant ( t values were 2.070, 2.462, P<0.05); The score of Nursing Delirium Screening Scale (Nu-DESC) for nursing care of DICU patients in the two groups was not statistically significant ( P>0.05). The score of the test group after nursing was (1.72 ± 0.75) points, which was lower than (2.17±0.63) points of the control group. The difference was statistically significant ( t value was 2.159, P<0.05); The results of the nursing quality evaluation questionnaire showed that the satisfaction of family members in the test group with respect to nursing services was 93.33% (60/75), which was higher than 82.67% (52/75) in the control group, with a significant difference ( χ2 value was 4.040, P<0.05). After the implementation, the correct answer rate of the "risk factors for delirium" questionnaire in the deliberate evidence-based knowledge questionnaire was significantly higher than before the implementation ( P < 0.05). After 6 months of implementation, the score of evidence-based knowledge of delirium was (6.82 ± 1.74) points higher than (4.26 ± 0.85) points before implementation, the difference was statistically significant ( t value was 3.739, P<0.01). Conclusions:The three-level nursing management process in the PICU ward can better prevent the occurrence of paralysis, strengthen the quality of nursing services, and exercise the comprehensive quality of nursing staff.