Early warning of nursing risk based on patient electronic medical record information

In order to study the nursing risk warning of patients' electronic medical record (EMR) information, data of 200 cases of patient from Eastern Medical District of Sichuan Academy of Medical Sciences & Sichuan Provincial People's Hospital is selected. The Logistic regression model and nursing risk warning information system are used to calculate the death probability of the recovered patients 24h before discharge and 24h before the death of the patients. The predictive ability of the system is verified, and incidence of infusion, blood transfusion, surgical patient handover, blood sample collection nursing risk, and corresponding indirect care hours before and after the use of nursing risk warning are compared. The probability of death of the recovered patients within 24h before discharge is less than 0.5, and the death probability of the dead patients before the death of 24h is greater than 0.5. After the nursing risk warning is used, the infusion, surgical patient handover, and incidence of blood sample collection nursing risk are significantly lower than those before using the nursing risk warning and the difference is statistically significant. The indirect nursing hours of infusion, blood transfusion, surgical patient handover, and blood sample collection are significantly lower than those before using the nursing risk warning, and the difference is statistically significant. Therefore, the predictive ability of the patient's EMR information risk warning system has important reference value in clinical nursing, which can significantly reduce the incidence of infusion, surgical patient handover, and blood sample collection and nursing risk, reduce the infusion, blood transfusion, surgical patient handover, and indirect nursing hours for collecting blood samples, and improve the quality and efficiency of hospital diagnosis and treatment.