Development and evaluation of the automated risk assessment system for multidrug-resistant organisms (autoRAS-MDRO).

摘要: Summary Background A high proportion of infections acquired in hospitals are caused by multidrug-resistant organisms (MDROs). The priority MDRO prevention is to detect high-risk patients and implement preventive intervention as soon possible. Aim To develop an automated risk assessment system for MDROs (autoRAS-MDRO) screen at infection evaluate the predictive validity autoRAS-MDRO. Methods Data 4200 variables were extracted from electronic health records (EHRs) constructing risk-scoring algorithm, which was based on a logistic regression model. autoRAS-MDRO designed such that classification (high, moderate, low risk) could be automatically displayed nursing Kardex EHRs system. For development 1000 with 4000 without selected; similarly, evaluation, 2173 8692 MDROs, respectively, selected. Findings follows: (i) 6-month evaluation: sensitivity, 81%; specificity, 79%; positive value (PPV), 49%; negative (NPV), 94%; Youden index, 0.60; (ii) 12-month sensitivity 79%, specificity 78%, PPV 47%, NPV 94%, 0.57. Conclusion had moderate validity. It useful redirecting nurses’ time efforts required implementation control measures, reducing incidence hospitals, thereby contributing patient safety.