Rivera Laura A, Li Ye, Savage Rachel D, Crowcroft Natasha S, Bolotin Shelly, Rosella Laura C, Lou Wendy, Hopkins Jessica, Gemmill Ian, Johnson Ian
Public Health Ontario, 480 University Ave, Toronto, M5G1V2, Canada.
Dalla Lana School of Public Health, University of Toronto, 155 College Street, Toronto, M5T 1P8, Canada.
BMC Public Health. 2017 Feb 15;17(1):199. doi: 10.1186/s12889-017-4073-6.
Despite widespread implementation of syndromic surveillance systems within public health agencies, previous studies of the implementation and use of these systems have indicated that the functions and responses taken in response to syndromic surveillance data vary widely according to local context and preferences. The objective of the Syndromic Surveillance Evaluation Study was to develop and implement standardized supports in local public health agencies in Ontario, Canada, and evaluate the ability of these supports to affect actions taken as part of public health communicable disease control programs.
Local public health agencies (LPHA) in Ontario, which used syndromic surveillance based on emergency department visits for respiratory disease, were recruited and randomly allocated to the study intervention or control group. The intervention group health agencies received standardized supports in terms of a standardized aberrant event detection algorithm and a response protocol dictating steps to investigate and assess the public health significance of syndromic surveillance alerts. The control group continued with their pre-existing syndromic surveillance infrastructure and processes. Outcomes were assessed using logbooks, which collected quantitative and qualitative information about alerts received, investigation steps taken, and public health responses. The study was conducted prospectively for 15 months (October 2013 to February 2015).
Fifteen LPHAs participated in the study (n = 9 intervention group, n = 6 control group). A total of 1,969 syndromic surveillance alerts were received by all LPHAs. Variations in the types and amount of responses varied by LPHA, in particularly differences were noted by the size of the health unit. Smaller health units had more challenges to both detect and mount a response to any alerts. LPHAs in the control group were more likely to declare alerts to have public health significance and to initiate any action. Regression models using repeated measures showed an interaction between the year (Year 1 versus Year 2) and the intervention as well as an interaction between year and sustained nature of the alert. Both of these were linked to the control health units reporting more "watchful waiting".
This study raises questions about the effectiveness of using standardized protocols to improve the performance of syndromic surveillance in a decentralized public health system. Despite efforts to create standardized protocols and engage public health agencies in the process, no significant differences in the effective use of syndromic alerts were observed beyond year 1. It also raises questions about the minimum capacity of the agency and minimum population size that are required for an effective response.
尽管公共卫生机构广泛实施了症状监测系统，但此前对这些系统实施和使用情况的研究表明，根据当地情况和偏好，针对症状监测数据所采取的功能和应对措施差异很大。症状监测评估研究的目的是在加拿大安大略省的地方公共卫生机构中开发并实施标准化支持措施，并评估这些支持措施对影响作为公共卫生传染病控制计划一部分所采取行动的能力。
招募了安大略省使用基于急诊科呼吸道疾病就诊情况进行症状监测的地方公共卫生机构（LPHA），并将其随机分配到研究干预组或对照组。干预组的卫生机构在标准化异常事件检测算法和响应协议方面获得了标准化支持，该协议规定了调查和评估症状监测警报的公共卫生意义的步骤。对照组继续使用其原有的症状监测基础设施和流程。使用日志对结果进行评估，日志收集了有关收到的警报、采取的调查步骤和公共卫生应对措施的定量和定性信息。该研究前瞻性地进行了15个月（2013年10月至2015年2月）。
15个地方公共卫生机构参与了该研究（n = 9个干预组，n = 6个对照组）。所有地方公共卫生机构共收到1969条症状监测警报。不同地方公共卫生机构的应对类型和数量存在差异，特别是按卫生单位规模注意到了差异。规模较小的卫生单位在检测和应对任何警报方面面临更多挑战。对照组的地方公共卫生机构更有可能宣布警报具有公共卫生意义并采取任何行动。使用重复测量的回归模型显示年份（第1年与第2年）与干预之间存在交互作用，以及年份与警报的持续性质之间存在交互作用。这两者都与对照卫生单位报告更多的“观察等待”有关。
本研究对使用标准化协议来提高分散式公共卫生系统中症状监测性能的有效性提出了疑问。尽管努力制定标准化协议并让公共卫生机构参与其中，但在第1年之后，未观察到症状监测警报有效使用方面的显著差异。它还对有效应对所需的机构最低能力和最低人口规模提出了疑问。
