Automated detection of substance use information from electronic health records for a pediatric population.

OBJECTIVE: Substance use screening in adolescence is unstandardized and often documented in clinical notes, rather than in structured electronic health records (EHRs). The objective of this study was to integrate logic rules with state-of-the-art natural language processing (NLP) and machine learning technologies to detect substance use information from both structured and unstructured EHR data. MATERIALS AND METHODS: Pediatric patients (10-20 years of age) with any encounter between July 1, 2012, and October 31, 2017, were included (n = 3890 patients; 19 478 encounters). EHR data were extracted at each encounter, manually reviewed for substance use (alcohol, tobacco, marijuana, opiate, any use), and coded as lifetime use, current use, or family use. Logic rules mapped structured EHR indicators to screening results. A knowledge-based NLP system and a deep learning model detected substance use information from unstructured clinical narratives. System performance was evaluated using positive predictive value, sensitivity, negative predictive value, specificity, and area under the receiver-operating characteristic curve (AUC). RESULTS: The dataset included 17 235 structured indicators and 27 141 clinical narratives. Manual review of clinical narratives captured 94.0% of positive screening results, while structured EHR data captured 22.0%. Logic rules detected screening results from structured data with 1.0 and 0.99 for sensitivity and specificity, respectively. The knowledge-based system detected substance use information from clinical narratives with 0.86, 0.79, and 0.88 for AUC, sensitivity, and specificity, respectively. The deep learning model further improved detection capacity, achieving 0.88, 0.81, and 0.85 for AUC, sensitivity, and specificity, respectively. Finally, integrating predictions from structured and unstructured data achieved high detection capacity across all cases (0.96, 0.85, and 0.87 for AUC, sensitivity, and specificity, respectively). CONCLUSIONS: It is feasible to detect substance use screening and results among pediatric patients using logic rules, NLP, and machine learning technologies.

Finding warning markers: Leveraging natural language processing and machine learning technologies to detect risk of school violence.

INTRODUCTION: School violence has a far-reaching effect, impacting the entire school population including staff, students and their families. Among youth attending the most violent schools, studies have reported higher dropout rates, poor school attendance, and poor scholastic achievement. It was noted that the largest crime-prevention results occurred when youth at elevated risk were given an individualized prevention program. However, much work is needed to establish an effective approach to identify at-risk subjects. OBJECTIVE: In our earlier research, we developed a risk assessment program to interview subjects, identify risk and protective factors, and evaluate risk for school violence. This study focused on developing natural language processing (NLP) and machine learning technologies to automate the risk assessment process. MATERIAL AND METHODS: We prospectively recruited 131 students with or without behavioral concerns from 89 schools between 05/01/2015 and 04/30/2018. The subjects were interviewed with two risk assessment scales and a questionnaire, and their risk of violence were determined by pediatric psychiatrists based on clinical judgment. Using NLP technologies, different types of linguistic features were extracted from the interview content. Machine learning classifiers were then applied to predict risk of school violence for individual subjects. A two-stage feature selection was implemented to identify violence-related predictors. The performance was validated on the psychiatrist-generated reference standard of risk levels, where positive predictive value (PPV), sensitivity (SEN), negative predictive value (NPV), specificity (SPEC) and area under the ROC curve (AUC) were assessed. RESULTS: Compared to subjects' sociodemographic information, use of linguistic features significantly improved classifiers' predictive performance (P < 0.01). The best-performing classifier with n-gram features achieved 86.5 %/86.5 %/85.7 %/85.7 %/94.0 % (PPV/SEN/NPV/SPEC/AUC) on the cross-validation set and 83.3 %/93.8 %/91.7 %/78.6 %/94.6 % (PPV/SEN/NPV/SPEC/AUC) on the test data. The feature selection process identified a set of predictors covering the discussion of subjects' thoughts, perspectives, behaviors, individual characteristics, peers and family dynamics, and protective factors. CONCLUSIONS: By analyzing the content from subject interviews, the NLP and machine learning algorithms showed good capacity for detecting risk of school violence. The feature selection uncovered multiple warning markers that could deliver useful clinical insights to assist personalizing intervention. Consequently, the developed approach offered the promise of an accurate and scalable computerized screening service for preventing school violence.

Automated Risk Assessment for School Violence: a Pilot Study.

School violence has increased over the past ten years. This study evaluated students using a more standard and sensitive method to help identify students who are at high risk for school violence. 103 participants were recruited through Cincinnati Children's Hospital Medical Center (CCHMC) from psychiatry outpatient clinics, the inpatient units, and the emergency department. Participants (ages 12-18) were active students in 74 traditional schools (i.e. non-online education). Collateral information was gathered from guardians before participants were evaluated. School risk evaluations were performed with each participant, and audio recordings from the evaluations were later transcribed and manually annotated. The BRACHA (School Version) and the School Safety Scale (SSS), both 14-item scales, were used. A template of open-ended questions was also used. This analysis included 103 participants who were recruited from 74 different schools. Of the 103 students evaluated, 55 were found to be moderate to high risk and 48 were found to be low risk based on the paper risk assessments including the BRACHA and SSS. Both the BRACHA and the SSS were highly correlated with risk of violence to others (Pearson correlations>0.82). There were significant differences in BRACHA and SSS total scores between low risk and high risk to others groups (p-values <0.001 under unpaired t-test). In particular, there were significant differences in individual SSS items between the two groups (p-value <0.001). Of these items, Previous Violent Behavior (Pearson Correlation = 0.80), Impulsivity (0.69), School Problems (0.64), and Negative Attitudes (0.61) were positively correlated with risk to others. The novel machine learning algorithm achieved an AUC of 91.02% when using the interview content to predict risk of school violence, and the AUC increased to 91.45% when demographic and socioeconomic data were added. Our study indicates that the BRACHA and SSS are clinically useful for assessing risk for school violence. The machine learning algorithm was highly accurate in assessing school violence risk.