Oral injuries in children less than 24 months of age in a pediatric emergency department

Development and Validation of a Natural Language Processing Tool to Identify Injuries in Infants Associated With Abuse

OBJECTIVES

Medically minor but clinically important findings associated with physical child abuse, such as bruises in pre-mobile infants, may be identified by frontline clinicians yet the association of these injuries with child abuse is often not recognized, potentially allowing the abuse to continue and even to escalate. An accurate natural language processing (NLP) algorithm to identify high-risk injuries in electronic health record notes could improve detection and awareness of abuse. The objectives were to: 1) develop an NLP algorithm that accurately identifies injuries in infants associated with abuse and 2) determine the accuracy of this algorithm.

METHODS

An NLP algorithm was designed to identify ten specific injuries known to be associated with physical abuse in infants. Iterative cycles of review identified inaccurate triggers, and coding of the algorithm was adjusted. The optimized NLP algorithm was applied to emergency department (ED) providers' notes on 1344 consecutive sample of infants seen in 9 EDs over 3.5 months. Results were compared with review of the same notes conducted by a trained reviewer blind to the NLP results with discrepancies adjudicated by a child abuse expert.

RESULTS

Among the 1344 encounters, 41 (3.1%) had one of the high-risk injuries. The NLP algorithm had a sensitivity and specificity of 92.7% (95% confidence interval [CI]: 79.0%-98.1%) and 98.1% (95% CI: 97.1%-98.7%), respectively, and positive and negative predictive values were 60.3% and 99.8%, respectively, for identifying high-risk injuries.

CONCLUSIONS

An NLP algorithm to identify infants with high-risk injuries in EDs has good accuracy and may be useful to aid clinicians in the identification of infants with injuries associated with child abuse.

The impact of a child abuse guideline on differences between pediatric and community emergency departments in the evaluation of injuries

BACKGROUND

Although child physical abuse is missed more frequently in community (CEDs) vs. pediatric emergency departments (PEDs), little information exists describing how evaluations of high-risk injuries differ between these settings.

OBJECTIVES

To determine differences in evaluations of infants for abuse between a PED and CEDs and whether a child abuse guideline reduced these differences.

PARTICIPANTS AND SETTING

Infants presenting to one PED (n = 162) and three CEDs (n = 159) with 3 injury categories: 1) Injuries for which the American Academy of Pediatrics recommends skeletal survey (SS) testing (infants <5-months with an oral injury or bruising, <9-months with a non-skull fracture, and < 12-months with an intracranial hemorrhage); 2) an oral injury or high-risk bruising in older infants; and 3) multiple types of high-risk injuries.

METHODS

We assessed differences in SS testing and child protective services (CPS) reporting between the PED and CEDs before and after implementation of a child abuse guideline.

RESULTS

The median (IQR) age was 4 months (2-7). Before guideline implementation, infants with injuries in categories 1 and 2 had an increased odds of SS testing in the PED vs. the CEDs (Category 1: aOR 2.83, 95% CI: 1.01-8.10; Category 2: aOR 10.1, CI: 1.2-88.0) and CPS reporting (Category 1: aOR 7.96, CI: 2.3-26.7; Category 2: aOR 12.0, CI: 1.4-103.5). After guideline implementation, there were no statistically significant differences in testing and reporting for any injury category.

CONCLUSIONS

Implementation of a child abuse guideline minimized differences between a PED and CEDs in the evaluation of infants with injuries concerning for abuse.

The Use of Experts to Evaluate a Child Abuse Guideline in Community Emergency Departments

OBJECTIVE

Guidelines and pathways exist to help frontline providers evaluate injured children for suspected child abuse. Little, however, is known about whether the decision-making resulting from these interventions is correct. Therefore, in the absence of an available gold-standard test, we used experts' judgments to examine the appropriateness of these clinical decisions. We evaluated community emergency department (ED) providers' adherence to a guideline recommending a child protection team (CPT) consultation for infants with injuries associated with abuse. We then compared providers' decision-making to experts' recommendations before and after guideline implementation.

METHODS

Two experts conducted a blinded, retrospective review of injured infants from 3 community EDs (N = 175). Experts rated the likelihood that an injury was abusive, indeterminate, or accidental, and made recommendations that were compared with skeletal survey (SS) testing and child protective services (CPS) reporting by providers before and after guideline implementation.

RESULTS

Postguideline implementation, there was a significant increase in CPT consultations in indeterminate cases (14.3% vs 72.2%, P < .001) and in SS testing when experts recommended SS (20.6% vs 56.8%, P = .002). In contrast, a higher percentage of cases for whom the experts did not recommend reporting were reported to CPS (1.8% vs 14.6%, P = .02).

CONCLUSIONS

Providers consulted the CPT most often for indeterminate cases. SS testing was in line with expert recommendations, but CPS reporting diverged from expert recommendations. Interventions linking community ED providers with a CPT may improve the evaluation of infants with injuries concerning for abuse.

Child Protection Team Consultation for Injuries Potentially Due to Child Abuse in Community Emergency Departments

OBJECTIVES

Emergency care for children is provided predominantly in community emergency departments (CEDs), where abusive injuries frequently go unrecognized. Increasing access to regional child abuse experts may improve detection of abuse in CEDs. In three CEDs, we intervened to increase involvement of a regional hospital child protection team (CPT) for injuries associated with abuse in children < 12 months old. We aimed to increase CPT consultations about these infants from the 3% baseline to an average of 50% over 12 months.

METHODS

We interviewed CED providers to identify barriers and facilitators to recognizing and reporting abuse. Providers described difficulties differentiating abusive from nonabusive injuries and felt that a second opinion would help. Using a plan-do-study-act approach, beginning in April 2018, we tested, refined, and implemented interventions to increase the frequency of CPT consultation, including leadership and champion engagement, provider training, clinical pathway implementation, and an audit and feedback process. Data were collected for 15 months before and 17 months after initiation of interventions. We used a statistical process control chart to track CPT consultations about children < 1 year old with high-risk injuries, use of skeletal surveys (SSs), and reports to child protective services (CPS).

RESULTS

Evidence of special cause was identified beginning in June 2018, with a shift of 8 points to one side of the center line. For the subsequent 8-month period, the CPT was consulted for a mean of 47.5% of children with high-risk injuries; this was sustained for an additional 7 months. The average percentage of infants with high-risk injuries who received a SS increased from 6.7% to 18.9% and who were reported to CPS increased from 10.7% to 32.6%.

CONCLUSION

Targeted interventions in CEDs increased the frequency of CPT consultation, SS use, and reports to CPS for infants with high-risk injuries. Such interventions may improve recognition of physical abuse.