Distributed data processing for public health surveillance

Journal of the American Medical Informatics Association, 2011

Background Providers have been reluctant to disclose patient data for public-health purposes. Even if patient privacy is ensured, the desire to protect provider confidentiality has been an important driver of this reluctance. Methods Six requirements for a surveillance protocol were defined that satisfy the confidentiality needs of providers and ensure utility to public health. The authors developed a secure multi-party computation protocol using the Paillier cryptosystem to allow the disclosure of stratified case counts and denominators to meet these requirements. The authors evaluated the protocol in a simulated environment on its computation performance and ability to detect disease outbreak clusters. Results Theoretical and empirical assessments demonstrate that all requirements are met by the protocol. A system implementing the protocol scales linearly in terms of computation time as the number of providers is increased. The absolute time to perform the computations was 12.5 s for data from 3000 practices. This is acceptable performance, given that the reporting would normally be done at 24 h intervals. The accuracy of detection disease outbreak cluster was unchanged compared with a non-secure distributed surveillance protocol, with an F-score higher than 0.92 for outbreaks involving 500 or more cases. Conclusion The protocol and associated software provide a practical method for providers to disclose patient data for sentinel, syndromic or other indicatorbased surveillance while protecting patient privacy and the identity of individual providers.

Annals of Emergency Medicine, 2002

2005 International Conference on Machine Learning and Cybernetics, 2005

Health systems around the world are moving to integrate health data sources to improve the efficiency of their services through data sharing. However, major technical and legal questions exist concerning data integration, data quality, data security and privacy in health data usage. In this paper, we discuss the Health Data Integration™ (HDI™) solution currently being developed at the e-Health Research Centre. Firstly we describe the architecture of HDI™ and how it performs the tasks of knowledge based medical system. We then consider two of its core capabilities (1) privacy-preserving similarity linkage, and (2) on-line analytical techniques and report generator component.

Journal of Public Health Management and Practice, 2017

Introduction: Data networks, consisting of pooled electronic health data assets from health care providers serving different patient populations, promote data sharing, population and disease monitoring, and methods to assess interventions. Better understanding of data networks, and their capacity to support public health objectives, will help foster partnerships, expand resources, and grow learning health systems. Methods: We conducted semistructured interviews with 16 key informants across the United States, identified as network stakeholders based on their respective experience in advancing health information technology and network functionality. Key informants were asked about their experience with and infrastructure used to develop data networks, including each network's utility to identify and characterize populations, usage, and sustainability. Results: Among 11 identified data networks representing hundreds of thousands of patients, key informants described aggregated health care clinical data contributing to population health measures. Key informant interview responses were thematically grouped to illustrate how networks support public health, including (1) infrastructure and information sharing; (2) population health measures; and (3) network sustainability. Conclusion: Collaboration between clinical data networks and public health entities presents an opportunity to leverage infrastructure investments to support public health. Data networks can provide resources to enhance population health information and infrastructure. KEY WORDS: data network, electronic health record, informatics, meaningful use, population health, public health practice M assive federal investments 1 have dramatically expanded electronic health records (EHR) adoption and implementation by health care providers and hospitals across the

Medical Care, 2010

Background: Comparative effectiveness research, medical product safety evaluation, and quality measurement will require the ability to use electronic health data held by multiple organizations. There is no consensus about whether to create regional or national combined (eg, "all payer") databases for these purposes, or distributed data networks that leave most Protected Health Information and proprietary data in the possession of the original data holders. Objectives: Demonstrate functions of a distributed research network that supports research needs and also address data holders concerns about participation. Key design functions included strong local control of data uses and a centralized web-based querying interface. Research Design: We implemented a pilot distributed research network and evaluated the design considerations, utility for research, and the acceptability to data holders of methods for menu-driven querying. We developed and tested a central, web-based interface with supporting network software. Specific functions assessed include query formation and distribution, query execution and review, and aggregation of results. Results: This pilot successfully evaluated temporal trends in medication use and diagnoses at 5 separate sites, demonstrating some of the possibilities of using a distributed research network. The pilot demonstrated the potential utility of the design, which addressed the major concerns of both users and data holders. No serious obstacles were identified that would prevent development of a fully functional, scalable network. Conclusions: Distributed networks are capable of addressing nearly all anticipated uses of routinely collected electronic healthcare data. Distributed networks would obviate the need for centralized databases, thus avoiding numerous obstacles.

Several health plans and other organizations are collaborating with the Centers for Disease Control and Prevention to develop a syndromic surveillance system with national coverage that includes more than 20 million people. A principal design feature of this system is reliance on daily reporting of counts of individuals with syndromes of interest in specified geographic regions rather than reporting of individual encounter-level information. On request from public health agencies, health plans and telephone triage services provide additional information regarding individuals who are part of apparent clusters of illness. This reporting framework has several advantages, including less sharing of protected health information, less risk that confidential information will be distributed inappropriately, the prospect of better public acceptance, greater acceptance by health plans, and less effort and cost for both health plans and public health agencies. If successful, this system will allow any organization with appropriate data to contribute vital information to public health syndromic surveillance systems while preserving individuals' privacy to the greatest extent possible.

2012

Health surveillance practices date back to decades ago. Traditionally, such practices to gather health data have been manual; more recently, however, computerized health information systems have been applied to enhance and facilitate health information acquisition for surveillance. The so-called health surveillance systems put in practice the systematic acquisition of health data, which is stored and processed for expert analysis. This chapter makes a survey of health surveillance systems dedicated to syndromic and epidemiological surveillance, identifying the different design and technological strategies adopted in the development of such systems. The aims of such a survey are: (1) to provide practitioners with some information about the collective expertise of health information system architects in the design and implementation of syndromic and epidemiological surveillance systems; and (2) to pave the way for the establishment of software product lines dedicated to such systems.

Information Fusion, 2012

Biosurveillance systems designed and deployed in the United States and abroad to allow public health authorities to monitor the health of their communities have significant design limitations despite their wide usage. One limitation is the lack of algorithmic solutions to combine local data sources for regional situation awareness. The objective of the current study is to develop and demonstrate the value of automated information fusion methods applied to the distributed neighboring public health sites. A prototype system consisting of distributed Bayesian models was designed to enable informed regional and local cognitive decision support response. The Intelligent Decision Support Network (IDSN) is composed of Bayesian Information Fusion Models (BIFMs) that target a particular syndrome or disease type. Using local data from county health departments in Northern Virginia for the time period between August 2005 and May 2007, we estimated the probability of a gastrointestinal (GI) outbreak in two ways: First, based on data from the local hospitals only; and second, based on the relative probability of outbreak by combining local hospital data and probabilities of GI events from the neighboring counties' BIFMs. Preliminary findings showed that the network of distributed models detected events that would be undetected without multi-jurisdictional data.

CCGrid 2003. 3rd IEEE/ACM International Symposium on Cluster Computing and the Grid, 2003. Proceedings., 2003

Medical data represent tremendous amount of data for which automatic analysis is increasingly needed. Grids are very promising to face today challenging health issues such as epidemiological studies through large image data sets. However, the sensitive nature of medical data makes it difficult to widely distribute medical applications over computational grids. In this paper, we review fundamental medical data manipulation requirements and then propose a distributed data management architecture that addresses the medical data security and high performance constraints. A prototype is currently being developed inside our laboratories to demonstrate the architecture capability to face realistic distributed medical data manipulation situations.

2003

Medical data represent tremendous amount of data for which automatic analysis is increasingly needed. Grids are very promising to face today's challenging health issues such as epidemiological studies through large image data sets. However, the sensitive nature of medical data makes it difficult to widely distribute medical applications over computational grids. In this paper, we review fundamental medical data manipulation requirements and we propose a distributed data management architecture that addresses the medical data security and high performance constraints. A prototype is currently being developed inside our laboratories to demonstrate the architecture capability to face realistic distributed medical data manipulation situations.