Big Data

Health care processes consume increasing volumes of digital data. However, creating and leveraging high quality integrated health data is challenging because large-scale health data derives from systems where data is captured from varying workflows, yielding varying data quality, potentially limiting its utility for various uses, including population health. To ensure accurate results, it’s important to assess the data quality for the particular use. Examples of sub-optimal health data quality abound: accuracy varies for medication and diagnostic data in hospital discharge and claims data; electronic laboratory data used to identify notifiable public-health cases shows varying levels of completeness across data sources; data timeliness has been found to vary across different data sources. Given that there is clear increasing focus on large health data sources; there are known data quality issues that hinder the utility of such data; and there is a paucity of medical literature describing approaches for evaluating these issues across integrated health data sources, we hypothesize that novel methods for ongoing monitoring of data quality in rapidly growing large health data sets, including surveillance data, will improve the accuracy and overall utility of these data.

Objective

We describe how entropy, a key information measure, can be used to monitor the characteristics of chief complaints in an operational surveillance system.

GFT is a surveillance tool that gathers data on local internet searches to estimate the emergence of influenza-like illness in a given geographic location in real time.3 Previously, GFT has been proven to strongly correlate with influenza incidence at the national and regional level.2,3 GFT has shown promise as an easily accessed tool to enhance influenza surveillance and forecasting; however, further geographic validation of city-level data is needed. 1,2,6

Objective

To test if Google Flu Trends (GFT) is predictive of the volume of influenza and pneumonia emergency department (ED) visits across multiple United States cities.

 

The National Strategy for Biosurveillance promotes a national effort to improve early detection and enable ongoing situational awareness of all-hazards threats. Implicit in the Strategy’s implementation plan is the need to upgrade capabilities and integrate multiple disparate data sources, including more complete electronic health record (EHR) data into future biosurveillance capabilities. Thus, new biosurveillance applications are clearly needed. Praedico™ is a next generation biosurveillance application that incorporates cloud computing technology, a Big Data platform utilizing MongoDB as a data management system, machine-learning algorithms, geospatial and advanced graphical tools, multiple EHR domains, and customizable social media streaming from public health-related sources, all within a user friendly interface.

Objective

The purpose of our study was to conduct an initial assessment of the biosurveillance capabilities of a new software application called Praedico™ and compare results obtained from previous queries with the Electronic Surveillance System for the Early Notification of Community-Based Epidemics (ESSENCE).

 

A variety of big data analytics, techniques and tools including social media analytics, open source visualizations, statistical anomaly detection, use of Application Programming Interfaces (APIs), and geospatial mapping, are used for infectious disease biosurveillance. Using these methodologies, policy makers and practitioners detect and monitor outbreaks across the world near real time, in multiple languages, 24/7. The non-infectious disease community, namely critical care, injury, and trauma stakeholders, currently lack this level of sophistication. To respond to most MCIs like a terrorist bombing, validated, real-time information is typically available via closed radio channels and limited to a specific set of emergency responders. Health care workers, policy makers, and citizens reach for news, radio, and Internet sources to characterize casualties and hazards, and increasingly social media. During the Boston Marathon bombing, witnesses began posting tweets seconds after the bombing and 15 seconds before CNN reported the incident. Current trauma data sets are unhelpful for real time response, including trauma registries that are used for hospital performance after an incident, and disaster databases consist of secondary reporting used for academic research purposes.

Objective

Discuss how different big-data analytics, techniques, and tools including open source platforms, cloud analytics, social media, crowdsourcing, and geospatial visualization can be used to quickly achieve situational awareness within seconds of a MCI, for use by pre-hospital responders, healthcare workers, and policy makers.

 

This year’s conference theme is “Harnessing Data to Advance Health Equity” – and Washington State researchers and practitioners at the university, state, and local levels are leading the way in especially novel approaches to visualize health inequity and the effective translation of evidence into surveillance practice.

Objective

Washington is leading the way in especially novel approaches. Our goal is to share some of these innovative methods and discuss how these are used in State and Local monitoring of Health

Many methods to detect outbreaks currently exist, although most are ineffective in the face of real data, resulting in high false positivity. More complicated methods have better precision, but can be difficult to interpret and justify. Praedico™ is a next generation biosurveillance application built on top of a Hadoop High Performance Cluster that incorporates multiple syndromic surveillance methods of alerting, and a machine-learning (ML) model using a decision tree classifier  evaluating over 100 different signals simultaneously, within a user friendly interface.

Objective

To compare syndromic surveillance alerting in VA using Praedico™ and ESSENCE.