Amy Ising, Lucas M Neuroth, Lubna S Hossain, Julie M Kafka, Anna E Waller
{"title":"Improving the Performance of a Syndromic Surveillance Definition for Sexual Violence.","authors":"Amy Ising, Lucas M Neuroth, Lubna S Hossain, Julie M Kafka, Anna E Waller","doi":"10.1097/PHH.0000000000002217","DOIUrl":null,"url":null,"abstract":"<p><strong>Context: </strong>Timely and representative data for sexual violence surveillance in the United States are lacking.</p><p><strong>Objective: </strong>To address this gap, we aimed to improve the identification of sexual violence in syndromic surveillance emergency department (ED) visit data through evaluation of the latest Centers for Disease Control and Prevention (CDC) sexual violence surveillance definition (CDC V3) using North Carolina (NC) syndromic surveillance data and development of an enhanced definition (V4).</p><p><strong>Design: </strong>We created a gold standard dataset of visits representing incident sexual violence, history of sexual violence, or no sexual violence through manual review, developed an enhanced V4 definition based on this review, and then compared the performance of the CDC V3 and V4 definitions against the gold standard.</p><p><strong>Setting: </strong>North Carolina.</p><p><strong>Participants: </strong>Emergency department visits in NC, 1/1/2016-6/3/2024.</p><p><strong>Main outcome measures: </strong>We compared the precision measures (Sensitivity, Specificity, Positive Predictive Value, Negative Predictive Value, and F1) of the CDC V3 and new V4 definitions for sexual violence. We applied both definitions to NC ED visit data for 2019-2023 to compare demographics and crude rates.</p><p><strong>Results: </strong>The CDC V3 sexual violence definition proved highly specific. The V4 definition identified a 6% increase in incident sexual violence ED visits compared to the CDC V3 definition, with additional chief complaint search terms and the selective incorporation of triage notes. Trends in patient sex, age group, race, and ethnicity were similar across both definitions. The F1 score showed improvement for the V4 definition compared to the CDC V3 definition, suggesting that it better optimizes the need to comprehensively identify ED visits for sexual violence while minimizing false positives.</p><p><strong>Conclusions: </strong>Syndromic surveillance ED visit data provide timely, population-based data on this complex topic. Building upon the CDC V3 definition and incorporating triage notes where available allowed us to develop a more sensitive and accurate surveillance definition for incident sexual violence.</p>","PeriodicalId":47855,"journal":{"name":"Journal of Public Health Management and Practice","volume":" ","pages":""},"PeriodicalIF":1.9000,"publicationDate":"2025-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Public Health Management and Practice","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1097/PHH.0000000000002217","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH","Score":null,"Total":0}
引用次数: 0
Abstract
Context: Timely and representative data for sexual violence surveillance in the United States are lacking.
Objective: To address this gap, we aimed to improve the identification of sexual violence in syndromic surveillance emergency department (ED) visit data through evaluation of the latest Centers for Disease Control and Prevention (CDC) sexual violence surveillance definition (CDC V3) using North Carolina (NC) syndromic surveillance data and development of an enhanced definition (V4).
Design: We created a gold standard dataset of visits representing incident sexual violence, history of sexual violence, or no sexual violence through manual review, developed an enhanced V4 definition based on this review, and then compared the performance of the CDC V3 and V4 definitions against the gold standard.
Setting: North Carolina.
Participants: Emergency department visits in NC, 1/1/2016-6/3/2024.
Main outcome measures: We compared the precision measures (Sensitivity, Specificity, Positive Predictive Value, Negative Predictive Value, and F1) of the CDC V3 and new V4 definitions for sexual violence. We applied both definitions to NC ED visit data for 2019-2023 to compare demographics and crude rates.
Results: The CDC V3 sexual violence definition proved highly specific. The V4 definition identified a 6% increase in incident sexual violence ED visits compared to the CDC V3 definition, with additional chief complaint search terms and the selective incorporation of triage notes. Trends in patient sex, age group, race, and ethnicity were similar across both definitions. The F1 score showed improvement for the V4 definition compared to the CDC V3 definition, suggesting that it better optimizes the need to comprehensively identify ED visits for sexual violence while minimizing false positives.
Conclusions: Syndromic surveillance ED visit data provide timely, population-based data on this complex topic. Building upon the CDC V3 definition and incorporating triage notes where available allowed us to develop a more sensitive and accurate surveillance definition for incident sexual violence.
期刊介绍:
Journal of Public Health Management and Practice publishes articles which focus on evidence based public health practice and research. The journal is a bi-monthly peer-reviewed publication guided by a multidisciplinary editorial board of administrators, practitioners and scientists. Journal of Public Health Management and Practice publishes in a wide range of population health topics including research to practice; emergency preparedness; bioterrorism; infectious disease surveillance; environmental health; community health assessment, chronic disease prevention and health promotion, and academic-practice linkages.