Incorporating Social Media Monitoring Software into Health Department Foodborne Illness Surveillance and Response

 描述(由申请者提供)：南内华达州卫生区(SNHD)有机会扩展最近与罗切斯特大学的团队成员及其在谷歌[X]的分包商一起进行的一项试点研究，并回答信息学与社交媒体监控软件(SOMS)的合作，名为Nememsis。软件Nememsis在用户从食品店发帖后，通过跟踪用户长达两周的时间来识别暗示食源性疾病的推文。然后，Nememsis软件根据生病推文的数量和严重程度为食品机构分配分数。在试点项目中，研究人员正在观察，与盲目分配的对照组相比，得分高的餐厅的健康检查结果是否也往往更差。在这项拟议的研究中，SNHD计划通过将现有的食源性疾病数据(包括食源性疾病投诉和客户卫生投诉)纳入软件并允许其影响分数来扩展该软件。然后，根据严重程度，评分将被用作确定何时何地需要监管反应的工具，如例行检查或流行病学调查。通过使用这项新技术，SNHD可能能够更快地识别未报告的食源性疾病病例，并迅速调查病例和潜在的暴发，从而减少疾病在社区中的传播。此外，如果将Nememsis软件与其他食源性疾病预报器统一起来的干预措施能够更快地识别食源性疾病暴发，则需要在实地调查食源性疾病问题上花费的宝贵时间将减少。在拟议的研究研究中，SNHD将通过监测实验室确认的沙门氏菌病和O157：H7大肠杆菌的比率、食源性疾病报告的数量和顾客对餐厅卫生的投诉数量，来评估干预措施是否对社区食源性疾病产生影响。其次，SNHD将评估干预措施是否会影响调查经核实的食源性疾病投诉和经核实的客户卫生投诉所花费的时间。这些数据将在研究的整个过程中收集，然后与基线数据进行比较，并分析差异。

 DESCRIPTION (provided by applicant): The Southern Nevada Health District (SNHD) has the opportunity to expand upon a recent pilot study conducted in conjunction with team members from University of Rochester and their subcontractors at Google[X] and Answer Informatics working with a social media monitoring software (SoMMS) entitled nEmesis. The software nEmesis identifies tweets suggestive of foodborne illness by following a user for up to two weeks after the user posts from a food establishment. The nEmesis software then assigns a score to the food establishment based on the quantity and severity of sick tweets. In the pilot program, researchers are observing whether restaurants with high scores also tended to have worse health inspection outcomes as compared to blind assigned controls. In the proposed research, SNHD plans to expand the software by incorporating existing foodborne illness data, including foodborne illness complaints and customer's sanitation complaints, into the software and allowing it to influence the score. The score will then be used as a tool to determine when and where a regulatory response is needed such as a routine inspection, or an epidemiological investigation, depending on the severity. By using this novel technology, SNHD will potentially be able to identify cases of unreported foodborne illness faster and investigate cases and potential outbreaks quickly resulting in the reduced spread of disease in the community. Furthermore, if the intervention of unifying the nEmesis software with other foodborne illness predictors is able to identify foodborne illness outbreaks more quickly, less valuable time will need to be spent in the field investigating foodborne illness concerns. In the proposed research study, SNHD will evaluate whether the intervention has an effect on foodborne illness in the community by monitoring rates of lab‐confirmed salmonellosis and E. coli O157:H7, quantity of foodborne illness reports, and quantity of customer complaints on restaurant sanitation. Secondly, SNHD will evaluate whether the intervention has an effect on time spent investigating verified foodborne illness complaints and verified customer sanitation complaints. This data will be collected over the life of the research and then compared to baseline data and analyzed for differences.