Influenza

Missed opportunities for influenza vaccination in office-based settings occur when patients (who are inclined to accept influenza vaccination if a provider recommends it) remain unvaccinated after a fall/winter healthcare visit. Healthcare providers can be very influential in encouraging patients to obtain influenza vaccination, but little is known in real-time during annual campaigns of how many and what type of providers are actually giving vaccinations in office settings. Many factors affect the ultimate population coverage including taking advantage of opportunities to vaccinate during medical visits. This suggests that provider vaccination behavior, if leveraged, could result in higher rates of influenza vaccine coverage. “Big” healthcare data in the form of high volume streams of electronic healthcare reimbursement claims (eHRCs) can potentially be used to track influenza vaccine administration practices in office-based settings in near real-time, thus empowering public health officials to provide this feedback to practitioners and potentially modify behaviors.

Objective

This paper describes the results of formative research to develop a new metric for public health officials to use in near-real-time tracking of the weekly participation of office-based providers in community influenza vaccination campaigns.

ILINet is used nationwide by sentinel healthcare providers for reporting weekly outpatient visit numbers for influenza-like illness to CDC. The Florida Department of Health receives urgent care center (UCC) data through ESSENCE from participating facilities. Seminole County is unique in that its four sentinel providers located in separate UCCs report into both systems, and all their discharge diagnoses are available through ESSENCE. However, the reported number of patients being discharged from those providers with diagnoses of influenza is not equivalent to the number of cases reported into ILINet. Data from the two systems were therefore compared both among and between the individual sentinel providers in order to determine the extent of the variation over four influenza seasons.

Objective

To compare influenza-like illness (ILI) data reported to the Centers for Disease Control and Prevention (CDC) U.S. Outpatient Influenza-like Illness Surveillance Network (ILINet) with discharge diagnosis data for influenza from the same reporting source obtained through the Electronic Surveillance System for the Early Notification of Community-based Epidemics (ESSENCE) in Seminole County, Florida.

Champaign and Urbana, Illinois are considered twin cities that share the University of Illinois. Due to different geographic recruitment procedures, Champaign and Urbana public elementary schools offer a particularly novel opportunity to examine the H1N1 outbreak among students. Urbana schools recruit from specific geographic areas (neighborhoods) designated by the school district whereas Champaign schools are non-selective in their composition where students residing in Champaign can attend any school within the city.

Objective

The goal of this project is to examine the patterns of school absenteeism during the H1N1 pandemic of 2009 comparing two contiguous school districts with very different enrollment policies.

Several countries prospectively monitor influenza-attributable mortality using a variation of the Serfling seasonal time series model that uses sinusoidal terms for seasonality. Typically, a seasonal model from previous years is used to forecast current expected mortality. Using laboratory surveillance time series data in the model may enhance interpretation of the surveillance information.

Objective

To demonstrate use of routine laboratory-confirmed influenza surveillance data to forecast predicted influenza-attributable deaths during the current influenza season. We also assessed whether including information on influenza type produced better surveillance forecasts.

Influenza is a contagious disease that causes epidemics in many parts of the world. The World Health Organization estimates that influenza causes three to five million severe illnesses each year and 250,000-500,000 deaths. Predicting and characterizing outbreaks of influenza is an important public health problem and significant progress has been made in predicting single outbreaks. However, multiple temporally overlapping outbreaks are also common. These may be caused by different subtypes or outbreaks in multiple demographic groups. We describe our Multiple Outbreak Detection System (MODS) and its performance on two actual outbreaks. This work extends previous work by our group by using model-averaging and a new method to estimate non-influenza influenza-like illness (NI-ILI). We also apply MODS to a real dataset with a double outbreak.

Near real-time emergency department chief complaint data is accessed through Florida’s syndromic surveillance system: Electronic Surveillance System for the Early Notification of Communitybased Epidemics-Florida (ESSENCE-FL). The Florida Department of Health relies heavily upon these data for timely surveillance of influenza and influenza-like illness (ILI). Hospital discharge data available from the Florida Agency for Health Care Administration (AHCA) captures information about influenza-associated ED visits and is considered complete. The delay in receiving the data (up to a year) hinders timely evidence-based decision making during the influenza season. Previous analyses (comparing the complete AHCA hospital discharge data to the ESSENCE-FL ILI syndrome and Influenza sub-syndrome) have shown ESSENCE-FL is a timely, effective tool to monitor influenza activity in the state and that the Influenza sub-syndrome most closely approximates influenza season activity in Florida. Adults > 65, pregnant women and children < 5 are at increased risk for morbidity and mortality from influenza infection. This investigation aims to determine if syndromic surveillance can be used to characterize in near real-time influenza infection in adults > 65, pregnant women, and children < 5 by comparing ED visits for influenza and ILI in ESSENCE-FL to historical AHCA records of people who incurred ED charges at a Florida hospital with diagnosed influenza.

Objective

To determine if emergency department (ED) based syndromic surveillance can be utilized to characterize in near real-time influenza infection in three high-risk populations: a) adults > 65, b) pregnant women, and c) children < 5.

Influenza-like illness (ILI) remains a significant public health burden to both the general public and the U.S. Department of Defense. Military personnel are especially susceptible to disease outbreaks owing to the often-crowded living quarters, substantial geographic movement, and physical stress placed upon them. Currently, the military employs syndromic surveillance on electronic reporting of clinical diagnoses. While faster than traditional, biologically-focused monitoring techniques, the military surveillance system proved inadequate at detecting outbreaks quickly enough in a recent study conducted by the CDC. Recently, research has included novel data sources, like social media, to conduct disease detection in real-time and capture communities not traditionally accounted for in current surveillance systems. Data-mining techniques are used to identify influenza-related social media posts and train a model against validated medical data. By integrating social media data and a medical dataset of all ILI-related laboratory specimens and doctor visits for the entire military cohort, a more comprehensive model than presently exists for disease identification and transmission will be possible.

Objective

To integrate existing influenza surveillance data sources and social media data into an accurate and timely outbreak detection model embedded into dashboard biosuveillance analytics for the Department of Defense.

 Numerous methods using social media for syndromic surveillance and disease tracking have been developed. Many websites use Twitter and other social media to track specific diseases or syndromes.1 Many are intended for public use and the extent of use by public health agencies is limited.2 Our work builds on 4 years of experience by our multi-disciplinary team3 with a focus on local surveillance of influenza. 4,5

Objective

Create a flexible user-friendly geo-based social media analytic tool for local public health professionals. With the goal of increasing situational awareness, system has capability to process, sort and display tweets with text terms of potential public health interest. We continue to refine the Social Media and Research Testbed (SMART) via feedback from surveillance professionals.