Author:
Hanson Rochelle F.,Zhu Vivienne,Are Funlola,Espeleta Hannah,Wallis Elizabeth,Heider Paul,Kautz Marin,Lenert Leslie
Abstract
Abstract
Background
Child abuse and neglect (CAN) is prevalent, associated with long-term adversities, and often undetected. Primary care settings offer a unique opportunity to identify CAN and facilitate referrals, when warranted. Electronic health records (EHR) contain extensive information to support healthcare decisions, yet time constraints preclude most providers from thorough EHR reviews that could indicate CAN. Strategies that summarize EHR data to identify CAN and convey this to providers has potential to mitigate CAN-related sequelae. This study used expert review/consensus and Natural Language Processing (NLP) to develop and test a lexicon to characterize children who have experienced or are at risk for CAN and compared machine learning methods to the lexicon + NLP approach to determine the algorithm’s performance for identifying CAN.
Methods
Study investigators identified 90 CAN terms and invited an interdisciplinary group of child abuse experts for review and validation. We then used NLP to develop pipelines to finalize the CAN lexicon. Data for pipeline development and refinement were drawn from a randomly selected sample of EHR from patients seen at pediatric primary care clinics within a U.S. academic health center. To explore a machine learning approach for CAN identification, we used Support Vector Machine algorithms.
Results
The investigator-generated list of 90 CAN terms were reviewed and validated by 25 invited experts, resulting in a final pool of 133 terms. NLP utilized a randomly selected sample of 14,393 clinical notes from 153 patients to test the lexicon, and .03% of notes were identified as CAN positive. CAN identification varied by clinical note type, with few differences found by provider type (physicians versus nurses, social workers, etc.). An evaluation of the final NLP pipelines indicated 93.8% positive CAN rate for the training set and 71.4% for the test set, with decreased precision attributed primarily to false positives. For the machine learning approach, SVM pipeline performance was 92% for CAN + and 100% for non-CAN, indicating higher sensitivity than specificity.
Conclusions
The NLP algorithm’s development and refinement suggest that innovative tools can identify youth at risk for CAN. The next key step is to refine the NLP algorithm to eventually funnel this information to care providers to guide clinical decision making.
Funder
South Carolina Clinical and Translational Research Institute
Publisher
Springer Science and Business Media LLC
Subject
Health Informatics,Health Policy,Computer Science Applications
Reference46 articles.
1. Saunders BE, Adams ZW. Epidemiology of traumatic experiences in childhood. Child Adolesc Psychiatr Clin. 2014;23(2):167–84.
2. Finkelhor D, Turner HA, Shattuck A, Hamby SL. Prevalence of childhood exposure to violence, crime, and abuse: results from the national survey of children’s exposure to violence. JAMA Pediatr. 2015;169(8):746–54.
3. APA Presidential Task Force on Posttraumatic Stress Disorder and Trauma in children and adolescents. Children and trauma: update for mental health professionals. Washington DC: American Psychological Association; 2008. Retrieved January 7, 2019 from http://www.apa.org/pi/families/resources/children-trauma-update.aspx.
4. De Bellis MD, Zisk A. The biological effects of childhood trauma. Child Adolesc Psychiatr Clin. 2014;23(2):185–222.
5. McLaughlin KA, Sheridan MA, Lambert HK. Childhood adversity and neural development: deprivation and threat as distinct dimensions of early experience. Neurosci Biobehav Rev. 2014;47:578–91.