Data Analytics

In today’s fast paced world, information is available (and expected) instantaneously. Social media has only fueled this expectation as it has permeated all aspects of our lives. More and more of the population is turning to social media outlets to share their thoughts and update their status, especially during disasters. With all these conversations occurring, it is only reasonable to assume that health status is part of the information being shared. In fact, studies by Johns Hopkins University and Harvard University have shown that social media reporting can serve as an early indicator and warning of emerging health issues within a community. Whether people are talking about being sick themselves or fear of illness in the community, there is a wealth of knowledge to be gained by tapping into this information. Yet gaining insight and understanding from social media data can be problematic. The unstructured nature of the data, the presence of social media “spam”, and the frequency of reposting information makes social media a noisy data source. Being able to harness this data would provide the opportunity to use social media as an effective situational awareness and early warning tool for biosurveillance missions. But how do you accomplish this? There are tens of millions of conversations happening on social media every day that would need to be sifted through to get to the health related topics. No public health entity has the time or staffing for that endeavor.

Objective

The goal of the Now Trending website is to provide a web based tool that pulls out relevant Twitter conversations concerning illness and disasters and provides meaningful analytics on how those conversations are trending. The website gives the user the ability to view trends overall and for specific geographic areas.

The NNDSS is a nationwide collaboration that enables all levels of public health (local, state, territorial, federal and international) to monitor, control, and prevent the occurrence and spread of state-reportable and nationally notifiable diseases and conditions. The NNDSS data are a critical source of information for monitoring disease trends, effectiveness of prevention and control programs, and policy development. To provide timely NNDSS data, state and territorial health departments voluntarily report notifiable disease incidence data to CDC when they become aware of these cases and as per recommended national notification timeframes. These provisional data are published each week in Morbidity and Mortality Weekly Report (MMWR). Great strides have been made exploring and exploiting new and different sources of disease surveillance data and developing robust statistical methods for analyzing the collected data. However, there have been fewer efforts in the area of online dissemination of surveillance data. Appropriate dissemination of surveillance data is important to maximize the utility of collected data.

Objective

The purpose of this study was to identify ideas for an enhanced dissemination of the US National Notifiable Diseases Surveillance System (NNDSS) provisional data.

The Child Health Epidemiology Reference Group (CHERG) has predicted around 43 million pneumonia cases in India. It is recognized that for huge nation like India, which accounts for 23% of global pneumonia burden, the national estimates may hide regional disparities. In this context, we have generated Indian state specific burden of severe pneumonia, pneumococcal pneumonia and pneumonia deaths through use of mathematical model.

Objective

This presentation highlights the use of mathematical model to estimate burden of disease in absence surveillance data. We estimated the burden of severe pneumonia, pneumococcal pneumonia and pneumonia deaths in Indian states using a mathematical model through application of vaccine probe methodology and attributable fraction.

Absenteeism has been considered as a potential indicator for the early detection of infectious disease outbreaks in population, especially in primary schools. However, in practice this data are often characterized by an excess of zeros and spatial heterogeneity. In a project on integrated syndromic surveillance system (ISSC) in rural China, Random effect zero-inflated Poisson (RE-ZIP) model was applied to simultaneously quantify the spatial heterogeneity for “occurrence” and “intensity” on school absenteeism data.

Objective

To describe and explore the spatial heterogeneity via Random effects zero-inflated Poisson model (RE-ZIP) for absenteeism surveillance in primary school for early detection of infectious disease outbreak in rural China.

Early detection of a disease outbreak using pre-diagnostic textual data is available in biosurveillance systems with the integration of data such as chief complaints. Social media has been identified as an additional pre-diagnostic data source of interest. Textual data analysis in public health is usually based on a keyword search and often involves a complex Boolean combination of terms that produce results with many false alarms. Epidemiologists may wish to query the data differently based on the event of interest, yet the process is laborious to weed out uninteresting content. Specialized detectors that decide on the topical relevance of keyword search usually require developers to adapt methods to new uses, which is a time- and cost-prohibitive activity. Users need the ability to rapidly build text content detectors on their own.

Objective

To demonstrate a framework for user-customizable text processing that can improve the efficiency and effectiveness of mining text for biosurveillance, with initial application to Twitter.

LHDs are operating in a changing data environment. As household telephone use declines, national surveys are not sampling large enough populations to report representative local health statistics. As a result, reliable indicators from surveys such as the Behavioral Risk Factors Surveillance Survey (BRFSS) are becoming scarce. Soon, these indicators may not be sufficient for county assessments. NC DETECT primarily uses data from emergency departments, the Carolinas Poison Center, and the Pre-hospital Medical Information System (PreMIS) to identify outbreaks and facilitate emergency response. However, while built to aggregate “real-time” data, NC DETECT also provides a source for rich, long-term indicators. The challenge for LHDs is that they may not have the knowledge, training, or technical assistance needed to fully utilize NC DETECT services. This project capitalizes on available human, organizational, and technical resources to increase LHD situational awareness and to demonstrate the usefulness of both “real-time” surveillance data as aggregate indicators of county health, and of low-cost prototyping using Excel’s more advanced Business Intelligence (BI) features.

Objective

This project aims to fill a growing county-level health data gap through the development of a low-cost, Excel-based surveillance tool. This prototype utilizes emergency department data (ED) collected by NC DETECT, a state-wide syndromic surveillance system, in order to visualize, monitor, and compare annual local health indicators for use in local decision making. In this way, the project aims to increase noncommunicable disease surveillance capacity and improve situational awareness within North Carolina local health departments (LHDs).

Standardized electronic pre-diagnostic information is routinely collected in Alberta, Canada. ARTSSN is an automated real-time surveillance data repository able to rapidly refresh data that include school absenteeism information, calls about health concerns from Health Link Alberta; a provincial telephone service for health advice and information, and emergency department visits categorized by standardized chief complaint. Until recently, real-time ARTSSN data for public health surveillance and decision making has been underutilized.

Objective

We developed early warning algorithms using data from ARTSSN and used them to detect signatures of potential pandemics and provide regular weekly forecasts on influenza trends in Alberta during 2012-2014.

The National Science and Technology Council, within the Executive Office of the President, established the Pandemic Prediction and Forecasting Science and Technology (PPFST) Working Group in 2013. The PPFST Working Group supports the US Predict the Next Pandemic Initiative, and serves as a forum to accelerate the development of federal infectious disease outbreak prediction and forecasting capabilities. Priorities include identification, evaluation, and integration of disparate biosurveillance and other data streams for prediction/forecasting; characterization of the decision context for US Government use of prediction/forecasting models; and development of a common US Government vision for federal prediction/forecasting capabilities. The Working Group comprises 18 federal departments and agencies, as well as the National Security Council, Office of Science and Technology Policy (OSTP), and Office of Management and Budget. OSTP, the Centers for Disease Control and Prevention, and the Department of Defense chair the Working Group.

Objective

To accelerate the development of US federal infectious disease outbreak prediction (i.e., identification of future time and place of a disease event) and forecasting (i.e., disease spread) capabilities.

Dengue is endemic in Singapore, with epidemics of increasing magnitude occurring on a six-year cycle in 1986/7, 1992, 1998, 2004/5, 2007 and 2013. The incidence per 100,000 population ranged from 87.2 to 105.6 in 2009-20121 , and surged to 410.6 in 2013. The mean weekly number of dengue cases over a five-year period provides an indication of the baseline level. We illustrate an adjustment that has been made to the computation of the baseline level due to increased testing for dengue in 2013.

Objective

To make adjustment of historical trends to accurately reflect the baseline level of dengue cases in Singapore, in view of increased testing for dengue in 2013.