Influenza-Like-Illness (ILI)

The Distribute project began in 2006 as a distributed, syndromic surveillance demonstration project that networked state and local health departments to share aggregate emergency department-based influenza-like illness (ILI) syndrome data. Preliminary work found that local systems often applied syndrome definitions specific to their regions; these definitions were sometimes trusted and understood better than standardized ones because they allowed for regional variations in idiom and coding and were tailored by departments for their own surveillance needs. Originally, sites were asked to send whatever syndrome definition they had found most useful for monitoring ILI. Places using multiple definitions were asked to send their broader, higher count syndrome. In 2008, sites were asked to send both a broad syndrome, and a narrow syndrome specific to ILI.

Objective

To describe the initial phase of the ISDS Distribute project ILI syndrome standardization pilot.

Syndromic surveillance systems significantly enhance the ability of Public Health Units to identify, quantify, and respond to disease outbreaks. Existing systems provide excellent classification, identification, and alerting functions, but are limited in the range of statistical and mapping analyses that can be done. Currently available commercial off-the-shelf (COTS) statistical and GIS packages provide a much broader range of analytical and visualization tools, as well as the capacity for automation through user-friendly scripting languages. This study retrospectively evaluates the use of these packages for surveillance using syndromic data collected in Ottawa during the 2009 pH1NI outbreak.

Objective

The objective of this study was to create and evaluate a system that uses customized scripts developed for COTS statistical and GIS software to (1) analyze syndromic data and produce regular reports to public health epidemiologists, containing the information they would need to detect and manage an ILI outbreak, and (2) facilitate the generation more detailed analyses relevant to specific situations using these data.

http://Google.org developed a regression model that used the volume of influenza-related search queries best correlated with the proportion of outpatient visits related to influenza-like illness (ILI) model to estimate the level of ILI activity. For calibration, the model used ILINet data from October 2003 to 2009, which report weekly ILI activity as the percentage of patient visits to health care providers for ILI from the total number patient visits for the week. Estimates of ILI in 121 cities were added in January 2010.

Objective

This paper compares estimates of ILI activity with estimates from the Centers for Disease Control’s ILINet from October 2008 through March 2010.

The New York City (NYC) Department of Health and Mental Hygiene monitors visits daily from 49 of 54 NYC emergency departments (EDs), capturing 95% of all ED visits. ED visits for influenza-like illness (ILI) have reflected influenza activity in NYC, better than the more broadly defined fever/flu and respiratory syndromes, but the correlation with H1N1 is unknown. 

Laboratory-confirmed influenza and respiratory syncytial virus (RSV) were made reportable in NYC in February 2008. DOHMH receives electronic reports of positive tests. 

As part of 2009–10 influenza surveillance, five hospitals were selected for ‘sentinel’ surveillance of hospitalized influenza cases, to test all patients with a respiratory condition for influenza. Sentinel hospitals ensured that patient medical record numbers were in the daily ED syndromic file and in the electronic laboratory reports.

Objective

To determine the correlation of the ILI syndrome with laboratory-confirmed H1N1 and RSV during the October 2009 to March 2010 H1N1 season in NYC.

During the spring of 2009, a public health emergency was declared in response to the emergence of the 2009 Influenza A (H1N1) virus. Owing to the response, timely data were needed to improve situational awareness and to inform public health officials. Traditional influenza surveillance is time-consuming and resource intensive, and electronic data sources are often more timely and resource saving. Collaboration began between the Centers for Disease Control and Prevention (CDC), the International Society for Disease Surveillance, and the Public Health Informatics Institute to expand syndromic Emergency Department (ED) surveillance through the Distribute project.

Distribute collects aggregate, daily or weekly reports of influenza-like illness (ILI) and total patient visits to EDs from participating health jurisdictions, stratified by age group and other variables. Additional variables included the three digit zip code of the patient’s residence as well as the disposition and temperature, however not all jurisdictions collect these variables. Distribute data are typically extracted from ED-based electronic health data systems. The ILI definition is determined by the participating jurisdiction that can be a city, county, or state. At the time of analysis, the network consisted of 33 jurisdictions.

Because ILI data reported to Distribute had not been systematically compared with data reported through other surveillance systems, CDC planned an evaluation of the Distribute data, which included a comparison to the Influenza-like Illness Network (ILINet). 

ILINet is a collaborative effort between the CDC, local and state health departments and primary health care providers. The network currently consists of approximately 3000 healthcare providers in all 50 states, Chicago, the District of Columbia, New York City, and the US Virgin Islands. Enrolled providers send CDC weekly reports via internet or fax that consist of the total number of patients seen for any reason and the number of those patients with ILI by age group. ILI is defined as fever (temperature of X1001F (37.8 1C)) and a cough and/or sore throat in the absence of a known cause other than influenza.

Objective

To compare ILI data reported to the Distribute surveillance project to data from an existing influenza surveillance system, the US Outpatient ILINet.

Current methods for influenza surveillance include laboratory confirmed case reporting, sentinel physician reporting of Influenza-Like-Illness (ILI) and chief-complaint monitoring from emergency departments (EDs).

The current methods for monitoring influenza have drawbacks. Testing for the presence of the influenza virus is costly and delayed. Specific, sentinel physician reporting is subject to incomplete, delayed reporting. Chief complaint (CC) based surveillance is limited in that a patient’s chief complaint will not contain all signs and symptoms of a patient.

A possible solution to the cost, delays, incompleteness and low specificity (for CC) in current methods of influenza surveillance is automated surveillance of ILI using clinician-provided free-text ED reports.

Objective

This paper describes an automated ILI reporting system based on natural language processing of transcribed ED notes and its impact on public health practice at the Allegheny County Health Department.

The Syndromic Surveillance Program (SSP) of the Acute Disease Epidemiology Section of the Georgia Division of Public Health, provides electronic influenza- like- illness (ILI) data to the Center for Disease Control and Prevention’s Influenza-like Illness Surveillance Network Program that characterizes the burden of influenza in states on a weekly basis.

ILI is defined as a fever of 1001, plus a cough or sore throat. This definition is used to classify ILI by the SSP, as well as in diagnosis at the pediatric hospital system. During the 2009 H1N1 pandemic, the SSP was provided a daily data transfer to the Center for Disease Control and Prevention to heighten situational awareness of the burden of ILI in Georgia. Throughout the peak of the pandemic, data from the pediatric hospital system identified when the percentage of daily visits for ILI had substantively increased. The data includes patient chief complaint (CC) data from emergency department visits for two facilities at Facilities A and B. The data received by SSP does not include diagnosis data.

Patient emergency department discharge data (DD) for ‘FLU’ was provided to SSP retrospectively to compare with the CC data routinely collected and analyzed. The data was derived from the pediatric health system’s month end, internal, syndromic surveillance report based upon emergency department visits, and including physician’s diagnosis at the time of patient’s discharge. The case definition of ‘FLU’ from the pediatric health system facilities is acute onset of fever, with cough and/or sore throat in the absence of a known cause other than influenza.

Objective

The objective of this study is to describe the difference between patient CC, ILI data provided daily to the Georgia SSP during the 2009 H1N1 pandemic, and patient DD subsequently provided for comparison with the SSP from its participating pediatric hospital system, and its two affiliated emergency rooms.