Integration of physical abuse clinical decision support at 2 general emergency departments.

OBJECTIVE: The study sought to develop and evaluate an electronic health record-based child abuse clinical decision support system in 2 general emergency departments. MATERIALS AND METHODS: A combination of a child abuse screen, natural language processing, physician orders, and discharge diagnoses were used to identify children <2 years of age with injuries suspicious for physical abuse. Providers received an alert and were referred to a physical abuse order set whenever a child triggered the system. Physician compliance with clinical guidelines was compared before and during the intervention. RESULTS: A total of 242 children triggered the system, 86 during the preintervention and 156 during the intervention. The number of children identified with suspicious injuries increased 4-fold during the intervention (P < .001). Compliance was 70% (7 of 10) in the preintervention period vs 50% (22 of 44) in the intervention, a change that was not statistically different (P = .55). Fifty-two percent of providers said that receiving the alert changed their clinical decision making. There was no relationship between compliance and provider or patient demographics. CONCLUSIONS: A multifaceted child abuse clinical decision support system resulted in a marked increase in the number of young children identified as having injuries suspicious for physical abuse in 2 general emergency departments. Compliance with published guidelines did not change; we hypothesize that this is related to the increased number of children identified with suspicious, but less serious injuries. These injuries were likely missed preintervention. Tracking compliance with guidelines over time will be important to assess whether compliance increases as physician comfort with evaluation of suspected physical abuse in young children improves.

Implementation of routine electronic health record-based child abuse screening in General Emergency Departments.

BACKGROUND: Routine child abuse screening is an approach to early identification of abuse. Previous studies evaluated paper-based screens; the widespread use of electronic health records suggests that screening is more likely to succeed if integrated into the electronic record. OBJECTIVE: To implement an electronic health record-based child abuse screen in a diverse hospital system and to evaluate the screening rate, rate of positive screens, and number of reports to Child Protective Services and assess whether hospital and patient characteristics are associated with these rates. PARTICIPANTS AND SETTING: Children <13 years of age evaluated at one of 13 Emergency Departments within University of Pittsburgh Medical Center Health System. METHODS: A previously validated child abuse screen was slightly modified and integrated into Cerner. Multivariable logistic regression models were used to estimate the odds of the outcomes of interest, controlling for key covariates. RESULTS: Of 17,163 eligible children: 68% received the screen of which 1.9% were positive. The rate of reports to Child Protective Services was higher among children who were screened (p < 0.0001). Younger children were more likely to be screened, have a positive screen, and have a report filed. There was no difference in the odds of being screened according to hospital teaching status, size or urban vs rural location. CONCLUSIONS: A child abuse screening tool can be integrated into the electronic health record in a large health-care network. The increased number of reports among children who were screened suggests that screening facilitates detection of suspected maltreatment.