Wang Mei-Hua, Chen Han-Kun, Hsu Min-Huei, Wang Hui-Chi, Yeh Yu-Ting
Graduate Institute of Biomedical Informatics, College of Medical Science and Technology, Taipei Medical University, Taipei, Taiwan.
Department of General Surgery, Chi-Mei Medical Center, Tainan, Taiwan.
J Med Internet Res. 2018 Aug 8;20(8):e10886. doi: 10.2196/10886.
Outbreaks of several serious infectious diseases have occurred in recent years. In response, to mitigate public health risks, countries worldwide have dedicated efforts to establish an information system for effective disease monitoring, risk assessment, and early warning management for international disease outbreaks. A cloud computing framework can effectively provide the required hardware resources and information access and exchange to conveniently connect information related to infectious diseases and develop a cross-system surveillance and control system for infectious diseases.
The objective of our study was to develop a Hospital Automated Laboratory Reporting (HALR) system based on such a framework and evaluate its effectiveness.
We collected data for 6 months and analyzed the cases reported within this period by the HALR and the Web-based Notifiable Disease Reporting (WebNDR) systems. Furthermore, system evaluation indicators were gathered, including those evaluating sensitivity and specificity.
The HALR system reported 15 pathogens and 5174 cases, and the WebNDR system reported 34 cases. In a comparison of the two systems, sensitivity was 100% and specificity varied according to the reported pathogens. In particular, the specificity for Streptococcus pneumoniae, Mycobacterium tuberculosis complex, and hepatitis C virus were 99.8%, 96.6%, and 97.4%, respectively. However, the specificity for influenza virus and hepatitis B virus were only 79.9% and 47.1%, respectively. After the reported data were integrated with patients' diagnostic results in their electronic medical records (EMRs), the specificity for influenza virus and hepatitis B virus increased to 89.2% and 99.1%, respectively.
The HALR system can provide early reporting of specified pathogens according to test results, allowing for early detection of outbreaks and providing trends in infectious disease data. The results of this study show that the sensitivity and specificity of early disease detection can be increased by integrating the reported data in the HALR system with the cases' clinical information (eg, diagnostic results) in EMRs, thereby enhancing the control and prevention of infectious diseases.
近年来发生了几起严重传染病的爆发。作为应对措施，为降低公共卫生风险，世界各国致力于建立一个信息系统，用于国际疾病爆发的有效疾病监测、风险评估和早期预警管理。云计算框架可以有效地提供所需的硬件资源以及信息访问和交换，以便便捷地连接与传染病相关的信息，并开发一个跨系统的传染病监测和控制系统。
我们研究的目的是基于这样一个框架开发一个医院自动化实验室报告（HALR）系统，并评估其有效性。
我们收集了6个月的数据，并分析了HALR系统和基于网络的法定传染病报告（WebNDR）系统在这期间报告的病例。此外，还收集了系统评估指标，包括评估敏感性和特异性的指标。
HALR系统报告了15种病原体和5174例病例，WebNDR系统报告了34例病例。在对这两个系统的比较中，敏感性为100%，特异性根据报告的病原体而有所不同。特别是，肺炎链球菌、结核分枝杆菌复合群和丙型肝炎病毒的特异性分别为99.8%、96.6%和97.4%。然而，流感病毒和乙型肝炎病毒的特异性分别仅为79.9%和47.1%。在将报告的数据与患者电子病历（EMR）中的诊断结果整合后，流感病毒和乙型肝炎病毒的特异性分别提高到了89.2%和9至9.1%。
HALR系统可以根据检测结果提供特定病原体的早期报告，从而实现疫情的早期发现并提供传染病数据趋势。本研究结果表明，通过将HALR系统中报告的数据与EMR中病例的临床信息（如诊断结果）相结合，可以提高疾病早期检测的敏感性和特异性，从而加强传染病的防控。
