Electronic Health Record Tools to Identify Child Maltreatment: Scoping Literature Review and Key Informant Interviews

Laura Stilwell; Megan Golonka; Kristin Ankoma-Sey; Madeleine Yancy; Samantha Kaplan; Lindsay Terrell; Elizabeth J Gifford

doi:10.1016/j.acap.2022.01.017

Electronic Health Record Tools to Identify Child Maltreatment: Scoping Literature Review and Key Informant Interviews

Acad Pediatr. 2022 Jul;22(5):718-728. doi: 10.1016/j.acap.2022.01.017. Epub 2022 Feb 4.

Authors

Laura Stilwell¹, Megan Golonka², Kristin Ankoma-Sey³, Madeleine Yancy³, Samantha Kaplan⁴, Lindsay Terrell⁵, Elizabeth J Gifford⁶

Affiliations

¹ Duke University School of Medicine (L Stilwell, S Kaplan), Durham, NC; Sanford School of Public Policy (L Stilwell, E Gifford), Durham, NC.
² Center for Child and Family Policy (M Golonka, E Gifford), Durham, NC; Department of Psychology and Neuroscience (M Golonka), Center for the Study of Adolescent Risk and Resilience, Duke University, Durham, NC. Electronic address: megan.golonka@duke.edu.
³ Duke University (K Ankoma-Sey, M Yancy), Durham, NC.
⁴ Duke University School of Medicine (L Stilwell, S Kaplan), Durham, NC.
⁵ Duke Department of Pediatrics (L Terrell), Durham, NC.
⁶ Sanford School of Public Policy (L Stilwell, E Gifford), Durham, NC; Center for Child and Family Policy (M Golonka, E Gifford), Durham, NC; Duke Department of Pediatrics (L Terrell), Durham, NC.

Abstract

Objective: To prevent missed cases and standardize care, health systems are beginning to implement EHR-based screens (EHR-CA-S) and clinical decision supports systems (EHR-CA-CDSS) for the identification and management of child maltreatment. This study aimed to 1) document the existing research evidence on the performance of EHR-CA-S and EHR-CA-CDSS and 2) examine clinical perspectives regarding the use of such tools and factors that affect uptake.

Methods: We searched MEDLINE, Embase, EBSCO, Scopus, and CINAHL databases for English language articles published prior to November 2021 that describe and/or evaluated an EHR-CA-S and/or EHR-CA-CDSS involving 0 to 18-year olds. We performed semistructured interviews with 20 individuals who have experience in identifying, evaluating, and/or treating child maltreatment and/or conducting research on these topics.

Results: Our search identified 574 articles; 16 met inclusion criteria. Studies examined screening, alerts and triggers, and quality improvement. None evaluated long-term clinical outcomes. Sensitivity ranged from 0.14 to 1.00, specificity from 0.865 to 1.00, positive predictive value from 0.03 to 1.00 and negative predictive value from 0.55 to 1.00. A variety of EHR-CA-S and/or EHR-CA-CDSS have been implemented at institutions in our sample. Interviewees cited missed cases, policy requirements, and the lack of standardization of care as impetuses for adopting these tools, yet expressed concerns regarding insufficient evidence, bias, and time-intensiveness of implementation.

Conclusions: Interviewees and the literature agree that current evidence does not support adoption of a particular CA-S or CA-CDSS. Further refinement and research on EHR-CA-S and EHR-CA-CDSS is necessary for these tools to be feasibly implemented and sustained, reliable for clinical practice, and not cause any unintentional harms.

Keywords: child abuse; child maltreatment; clinical decision support systems; electronic health records.

Publication types

Review
Research Support, N.I.H., Extramural

MeSH terms

Child
Decision Support Systems, Clinical*
Electronic Health Records*
Humans
Mass Screening
Quality Improvement

Abstract

Publication types

MeSH terms

Grants and funding