Introduction
The following document contains details related to the data collection, cleaning, and analysis of metrics displayed on the Cook County Department of Public Health (CCDPH) influenza surveillance dashboard. While the primary focus of this dashboard is influenza, metrics are also included for COVID-19 and Respiratory Syncytial Virus (RSV) to create a more comprehensive, overall picture of the respiratory viral season.
For data grouped by week, CCDPH uses MMWR weeks(1), with the week ending date displayed. MMWR weeks are used by CDC to report the incidence of national notifiable diseases in the Morbidity and Mortality Weekly Report (MMWR). MMWR weeks run from Sunday through Saturday. When historical data is displayed, data points are aligned by week number, i.e., Week 40 for any selected past season will be displayed at the tick mark for the week ending date corresponding to Week 40 of the current season. Most years have 52 MMWR weeks; however, some have 53 MMWR weeks (e.g. 2014, 2020). Graphing these years in comparison to 52-week years can be challenging. Because the last week of the calendar year is epidemiologically important for flu transmission due to holiday gatherings and school closures, we graph these weeks together. When seasons beginning with a 53-week year are selected, Week 53 data appears at the Week 52 mark. All other data points prior to Week 53 are moved forward one week (i.e., Week 52 becomes Week 51, etc.).
CCDPH’s jurisdiction covers Cook County, except areas that have their own state-certified local health department. These places include the cities of Chicago, Evanston, Oak Park, Skokie, and Stickney township. Whenever possible, we display metrics for CCDPH jurisdiction; however, when this level of geographic granularity is not possible, metrics may be presented at the county level or include other Cook County suburbs listed above.
How to Get Data
If you would like to work with the data displayed in this dashboard, you can submit an External Data Request using this form.
Emergency Department Visits and Hospital Admissions
CCDPH monitors two metrics for hospital visits associated with respiratory illness: the percent of all emergency visits for a selected respiratory illness and the percent of all hospital admissions for a selected respiratory illness. Data for both metrics comes from the National Syndromic Surveillance Program(NSSP). All acute care hospitals in Illinois are required to send a limited set of patient data for emergency department (ED) encounters and inpatient admissions to NSSP(2). Metrics include encounters for residents of suburban Cook County zip codes, regardless of which Illinois hospital the patient visited. To identify encounters associated with respiratory illnesses, we use CDC-designed queries that flag visits based on discharge diagnosis codes. In addition to discharge diagnosis codes specific to flu, COVID-19, and RSV, we also monitor trends for broad respiratory illness to more accurately capture respiratory illness lacking a specific diagnosis, related respiratory pathogens, and respiratory pathogen sequelae, such as pneumonia. Codes corresponding to a given respiratory illness query are included below.
For emergency department encounters, CCDPH monitors the percent of all ED visits with a selected respiratory diagnosis. Visits are grouped by encounter date and are counted regardless of whether the person was admitted or discharged. For hospital admissions, only individuals admitted to the hospital are included (these patients may or may not have been emergency room patients first). Using the percent of visits associated with the diagnosis, as opposed to counts, limits the impact of artificial increases or decreases in total visits due to data quality changes.
Test Positivity and Influenza Subtyping
Laboratory test data for percent positivity calculations comes from a sentinel network of laboratories who voluntarily submit de-identified, aggregated respiratory testing data for the purpose of public health surveillance. Antigen tests and nucleic acid amplification tests (NAATs) are included in percent positivity calculations; serology (antibody) tests are excluded. Percent positivity is calculated by dividing the number of positive tests over the number of tests performed. Each sentinel laboratory may submit data for one or more respiratory pathogens. Sentinel laboratory data is obtained through the following sources:
- National Syndromic Surveillance Program: Includes testing data from LabCorp and Quest for residents of suburban Cook County zip codes. Data is available from 2022 and includes influenza, RSV, and COVID-19.
- The National Respiratory and Enteric Virus Surveillance System: Includes testing data from five Chicago-based hospital and public health laboratories. Data is available from 2016 and includes influenza, RSV, and COVID-19.
- Directly submitted sentinel laboratory data: Three suburban Cook County-based laboratories submit aggregated respiratory testing data directly to CCDPH. Data is available for influenza from 2009 and for RSV from 2020.
All sentinel laboratories submitting testing data for influenza include results for both unsubtyped and subtyped flu results. Most seasonal influenza epidemics will include circulation of both Type A influenza viruses and Type B influenza viruses. The majority of seasonal flu illness is caused by Type A viruses, though many flu seasons see increases in Type B closer to spring. For influenza A viruses, there are two main subtypes: H1N1 (pandemic 2009) and H3N2. Most influenza seasons will see one flu A subtype predominate. While both flu A subtypes can cause severe illness, seasons in which H1N1 predominates tend to be more mild than seasons in which H3N2 predominates(3).
Wastewater Surveillance
People infected with respiratory pathogens can shed virus in their stool, with or without symptoms. The virus can then be measured in wastewater, or sewage, as it makes its way to wastewater treatment plants. This method of surveillance is beneficial because it is not dependent on individuals seeking testing, or even knowing they are sick. However, because wastewater is a complex substance, made up of everything from human and animal feces to rain to industrial substances, measuring viral concentrations is a complicated process. Individual results are highly variable, but form a trend over time that can provide information on disease activity, especially when considered with other surveillance metrics(4).
Wastewater data for Cook County is made available through the Illinois Wastewaster Surveillance System(IWSS). Treatment plants who participate in IWSS volunteer to have samples collected from their facilities twice a week and tested to measure viral concentrations for SARS-CoV-2, Influenza A, Influenza B, and RSV. In Cook County, all seven treatment plants operated by the Metropolitan Water Reclamation District of Greater Chicago participate in IWSS. One plant, the Stickney Reclamation Plant, has two sampling sites, at its north and south influents.
To monitor trends at the county level, CCDPH first normalizes viral concentrations by each plants’ average flow rate to get viral copies per liter. We then sum the viral copies per liter to get a county-wide value (displayed as points on the dashboard). Only days in which all eight sites have been sampled are included. We then use a generalized additive model to create a smoothed model of the data (displayed as a line on the dashboard). Prior to any analysis, we remove extreme outliers for raw viral concentrations (values more than 5 times higher than the upper limit of the sites’ interquartile range).
Reported ICU Admissions
All heathcare providers, laboratories, or other testing sites in Illinois are required to report flu-associated intensive care unit (ICU) admissions, per Illinois Control of Communicable Diseases Code(5). Mandated reporters are required to report cases electronically using Illinois’ National Electronic Disease Surveillance System (I-NEDSS). Local health departments then investigate cases, complete missing information, determine case status (“uniform criteria to define a disease for public health surveillance”)(6), and implement disease control measures as needed. In this dashboard, we include all reported cases with a case status of either ‘confirmed’ or ‘probable’. For graphs by week, cases are grouped by the date they were admitted to the hospital.
Mortality Estimates
CCDPH receives weekly de-identified death certificate data for all deaths that occur in Cook County (regardless of decedent residence). Fields include the immediate cause of death, contributing factors, and other significant conditions. Death certificates including the terms in the table below are considered associated with pneumonia, influenza, or COVID-19 (PIC). All search terms include common misspellings. The percent of PIC-associated deaths, out of all deaths reported, is calculated. Deaths attributed to pneumonia are included because pneumonia is common sequela of flu and influenza deaths are often coded as pneumonia deaths(7). Excluding these deaths would significantly under-estimate mortality attributed to flu. However, since the COVID-19 pandemic, many pneumonia deaths are now caused by COVID-19. Therefore, the dashboard also includes a graph showing the counts of deaths certificates with flu terms and COVID-19 terms (Table 2) to help determine the relative contribution of each disease to PIC mortality.
Activity Level
Activity levels for influenza are ‘Low’, ‘Increasing’ (‘Decreasing’ following the season peak), and ‘High’. The activity level is set by CCDPH influenza epidemiologists after careful review of all surveillance metrics.
Recommendations for individuals and healthcare organizations corresponding to each activity level are offered in the Key Points section of the dashboard.