Respiratory Syncytial Virus (RSV)
National Syndromic Surveillance Program
The National Syndromic Surveillance Program (NSSP) consolidates data from many sources and on a wide range of subjects to provide timely, complete, and accurate information in partnership with federal agencies, local and state health departments, and academic and private sector partners. Among these data sources, emergency department (ED) visit data resulting from influenza, respiratory syncytial virus, and COVID-19 respiratory infections are available on data.gov. 83% of United States ED departments provide data to NSSP within 24 hours, and percentage occurrences at the state and county level are released weekly 1,2.
Program Landing Page:
data.gov API Source:
Data Dictionary
| Signal | Description |
|---|---|
geography
|
State and county-level information on participating healthcare facilities within a given Health Service Area (HSA). The representation of sub-state regions aligns with the National Center for Health Statistics’ definition of an HSA, where regions align with patterns of care seeking in hospital systems. State boundaries correspond to the standard census boundaries applicable for that decennial period. Here, geolocation is coded in Federal Information Processing System (FIPS) 3. Find the relationship between FIPS codes and HSA county codes from the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program Health Service Areas (HSA) webpage: NCI Modified Table 4,5. |
time
|
Qualifying emergency department visits related to a specific pathogen observed in a given Morbidity and Mortality Weekly Report (MMWR) week and reported by participating sites. New data releases from source: once per week on Fridays Earliest date available: 2022-10-01 |
percent_visits_rsv
|
Percent of emergency department visits coded as being caused by RSV identified using the following DD and clinical terms. Calculated as qualifying rsv diagnoses out of all emergency department visits for the applicable group stratification. ICD-10-CM: B97.4 Respiratory syncytial virus (RSV) as the cause of diseases classified elsewhere (B97.4), Respiratory syncytial virus pneumonia (J12.1), Acute bronchitis due to respiratory syncytial virus (J20.5), and Acute bronchiolitis due to respiratory syncytial virus (J21.0). ICD-9-CM: Acute bronchiolitis due to respiratory syncytial virus (466.11), Pneumonia due to respiratory syncytial virus (480.1), and Respiratory syncytial virus (079.6). SNOMED CT: Respiratory syncytial virus infection (55735004), Healthcare associated respiratory syncytial virus disease (408684006), Pneumonia caused by respiratory syncytial virus (195881003), Bronchopneumonia caused by respiratory syncytial virus (10625551000119103), Respiratory syncytial virus bronchiolitis (57089007), Acute bronchiolitis caused by respiratory syncytial virus (195739001), Respiratory syncytial virus bronchitis (79479005), Acute respiratory syncytial virus bronchitis (195727009), Respiratory syncytial virus laryngotracheobronchitis (72204002), and Respiratory syncytial virus pharyngitis (31309002). |
percent_visits_flu
|
Percent of emergency department visits coded as being caused by influenza identified using the following DD and clinical terms. Calculated as qualifying influenza diagnoses out of all emergency department visits for the applicable group stratification. ICD-10-CM: Influenza due to certain identified influenza viruses (J09), Influenza due to other identified influenza viruses (J10), and Influenza due to unidentified influenza virus (J11). ICD-9-CM: Influenza with pneumonia (487.0), Influenza with other respiratory manifestations (487.1), Influenza with other manifestations (487.8), Influenza due to identified avian influenza virus with pneumonia (488.01), Influenza due to identified avian influenza virus with other manifestations (488.09), Influenza due to identified 2009 H1N1 influenza virus with pneumonia (488.11), Influenza due to identified 2009 H1N1 influenza virus with other manifestations (488.19), Influenza due to identified novel influenza A virus with pneumonia (488.81), Influenza due to identified novel influenza A virus with other manifestations (488.89). SNOMED CT: Influenza caused by influenza A virus subtype H1N1 (442696006), Influenza caused by Influenza A virus (442438000), Influenza (6142004), and Pneumonia and influenza (195878008). |
percent_visits_covid
|
Percent of emergency department visits coded as being caused by COVID-19 identified using the following DD and clinical terms. Calculated as qualifying COVID-19 diagnoses out of all emergency department visits for the applicable group stratification. ICD-10-CM: Pneumonia due to coronavirus disease 2019 (J12.82) and COVID-19 (U07.1). SNOMED CT: Disease caused by severe acute respiratory syndrome coronavirus 2 (840539006), Suspected disease caused by severe acute respiratory coronavirus 2 (840544004), and Severe acute respiratory syndrome coronavirus 2 (840533007). |
Coverage Map
Emergency departments located in HSA counties across the entire United States, including the District of Columbia and Guam, can submit their data to NSSP, as well as to their local and state health departments. The map below illustrates the representation of this data 6.
Limitations
The following summarizes specific data collection, processing, and reporting characteristics that might impact the interpretation of results. These reflect source limitations provided by the Companion Guide: NSSP ED Data on Respiratory Illness and internal PopHIVE data processing methods 6. Relevant PopHIVE data processing scripts are directly linked to as needed.
To comply with data use and sharing agreements, county-level data are suppressed from New Hampshire, South Dakota, and Washington.
If the proportion of records with a discharge diagnosis is low or the counts of observed events are sufficiently low, reports are suppressed to ensure data privacy. This suppression can impact the resolution at both the county and state levels.
Data are reported based on the electronic diagnostic codes (listed above) generated during an emergency department visit, rather than directly from the viral detection assay. Testing practices, and code assignment standards, are also expected to differ for the viruses, hospital systems, and for specific age groups.
Data are updated by NSSP as qualifying discharge diagnoses (DD) codes are added and changed.
During data processing, where available, county-level values are shown; otherwise, they are replaced with state-level values (
ingest.R).
Epic Cosmos
Cosmos is a dataset created in collaboration with a community of Epic health systems representing more than 300 million patient records from over 1633 hospitals and 37,900 clinics from all 50 states, D.C., Canada, Lebanon, and Saudi Arabia. Summary statistics were obtained using the SlicerDicer tool in Epic Cosmos.
Program Landing Page:
Epic Session ID:
Epic Session ID:
Epic Session ID:
2726769 for “Number of ED Encounters by State of Residence (U.S.) and Age at Time of Visit Range”
2726803 for “RSV ED visits State Week Age”
2747766 for “Positive RSV Tests and RSV tests and Number of Encounters by State of Residence (U.S.) and Age in Years Range”
ED visits: Percent of ED visits recorded as being due to RSV, obtained from facilities using the Epic electronic health records platform and participating in Epic Cosmos. Epic Cosmos captures varying amounts of the population in different regions, so the data might be a more representative measure of viral activity in the community in some areas than others.
Where there are fewer than 10 cases per time period, the data are suppressed. These values are filled in with a value halfway between 0 and the minimum reported value on the time series. These values are noted with suppressed_flag=1
RESP-NET
The CDC’s Respiratory Virus Hospitalization Surveillance Network (RESP-NET) monitors laboratory-confirmed hospitalizations associated with influenza, COVID-19, and respiratory syncytial virus (RSV) among children and adults. Pathogen-specific installments of the surveillance program (COVID-NET, RSV-NET, and FluSurv-NET) operate distinctly, yet together they contribute to RESP-NET’s surveillance and data collection efforts.
Data are collected from hospitals in select counties and county equivalents across up to 14 states, representing more than 30 million people, approximately 10% of the U.S. population 7. RESP-NET defines pathogen-specific surveillance areas around participating hospitals, allowing rates to be accurately reported. Each program collects data on age, sex, and race and ethnicity demographics. Officials strive to capture all cases within the catchment area to ensure a representative understanding of who is impacted by the disease burden 7.
Program Landing Page:
data.gov API Source:
Data Dictionary
| Signal | Description |
|---|---|
rate_covid
|
A qualifying event is defined as an individual within the catchment area who tests positive for SARS-CoV-2 through a laboratory test ordered by a healthcare professional within 14 days prior to, or during, hospitalization 8. Hospitalization rates are defined as the number of residents within a defined area who test positive, divided by the total population of that area, and are reported per 100,000 persons. The type of respiratory illness was extrapolated from the Surveillance Network = COVID-NET column and reshaped into a standalone column using values reported in the Weekly Rate column.
|
rate_rsv
|
A qualifying event is defined as an individual within the catchment area who tests positive for respiratory syncytial virus (RSV) through a laboratory test ordered by a healthcare professional within 14 days prior to, or during, hospitalization 9. Hospitalization rates are defined as the number of residents within a defined area who test positive, divided by the total population of that area, and are reported per 100,000 persons. The type of respiratory illness was extrapolated from the Surveillance Network = RSV-NET column and reshaped into a standalone column using values reported in the Weekly Rate column.
|
rate_flu
|
A qualifying event is defined as an individual within the catchment area who tests positive for influenza through a laboratory test ordered by a healthcare professional within 14 days prior to, or during, hospitalization. A positive result can be obtained from any of the following laboratory assays: viral culture, direct or indirect fluorescent antibody staining, rapid antigen test, or molecular assay 10. Hospitalization rates are defined as the number of residents within a defined area who test positive, divided by the total population of that area, and are reported per 100,000 persons. The type of respiratory illness was extrapolated from the Surveillance Network = FluSurv-NET column and reshaped into a standalone column using values reported in the Weekly Rate column.
|
Season
|
Active surveillance dates within a season begin in October (MMWR Week 40) and end either at the end of April (MMWR Week 17), June (MMWR Week 16), or September (MMWR Week 39) of the following year. In some seasons, programs concluded surveillance in different months 7–9. Refer to page 6 of the embedded PowerBI dashboard provided below for specific details about the date ranges captured in each season by surveillance program. |
MMWR Year
|
The Morbidity and Mortality Weekly Report (MMWR) reflects the epidemiologic year to which the National Notifiable Diseases Surveillance System (NNDSS) disease report is assigned. MMWR weeks range from 1 to 53, although most years only have 52 weeks. The MMWR week starts on Sunday, and MMWR Week 1 is defined as the first week of the year that contains at least four days in the new calendar year 11. |
MMWR Week
|
The Morbidity and Mortality Weekly Report (MMWR) reflects the epidemiologic year to which the National Notifiable Diseases Surveillance System (NNDSS) disease report is assigned. MMWR weeks range from 1 to 53, although most years only have 52 weeks. The MMWR week starts on Sunday, and MMWR Week 1 is defined as the first week of the year that contains at least four days in the new calendar year 11. |
age
|
Stratification includes 0-<1 yr, 0-4 yr, 12-17 yr, 1-4 yr, 18-29 yr, 18-49 yr, 30-39 yr, 40-49 yr, 50-64 yr, 5-11 yr, 5-17 yr, 65-74 yr, 65+ yr, 75-84 yr, 75+ yr, 85+ yr, Adults, and Pediatrics. Overall represents the combined results for all stratifiers. Formatting was modified to display * Years, and Overall entries were renamed Total. Only the age ranges <1 Years, 1-4 Years, 5-17 Years, 18-49 Years, 50-64 Years, and 65+ Years were retained, while all other entries were coded as either Total or other. These values were originally reported under the Age group column.
|
geography
|
Participating network sites consist of acute-care hospitals spanning 185 counties across up to 14 states. This represents more than 30 million people, approximately 10% of the U.S. population. The RESP-NET dashboard, embedded below, under the About Our Data tab, provides further details about geographical representation for all seasons recorded since 2018. These values were originally reported under the Site column and were translated from state names to corresponding Federal Information Processing System (FIPS) codes 3.
|
Sex
|
Stratification includes Male and Female, with Overall representing the combined results. Only the Overall outcome was retained; all other outcomes were filtered out.
|
Race/Ethnicity
|
Stratification includes American Indian/Alaska Native, NH, AI/AN, non-Hispanic, Asian/Pacific Islander, NH, A/PI, non-Hispanic, Black, NH Black, non-Hispanic, Hispanic, White, NH, White, non-Hispanic. Overall represents the combined results for all stratifiers. Only the Overall outcome was retained; all other outcomes were filtered out.
|
Cumulative Rate
|
Cases detected throughout the span of an infectious season, starting with the first week of October (MMWR Week 40), were summed cumulatively until the end of the surveillance period. Rate calculations used the cumulative value for the numerator, while the denominator remains the same as that for non-cumulative rates. Rates are reported per 100,000 persons. |
time
|
The date of hospital admission is recorded to the nearest week and standardized to report the date as the last day of the week, which is Saturday, on which the event occurred 10. These values were originally reported under the Week Ending Date column.
|
Type
|
Differentiates metrics as Age-Adjusted Rate and Unadjusted Rate. Only the Unadjusted Rate metric was retained; all other outcomes were filtered out.
|
Limitations
The following summarizes specific data collection, processing, and reporting characteristics that might impact the interpretation of results. These reflect source limitations provided by the Respiratory Virus Hospitalization Surveillance Network (RESP-NET) and internal PopHIVE data processing methods, unless otherwise noted 7. Relevant PopHIVE data processing scripts are directly linked to as needed.
The participating surveillance sites have experienced changes over time, resulting in the addition or removal of counties represented in new surveillance years. RESP-NET adjusts the total population estimates to account for these variations.
During the 2020-21 season, the case counts from FluSurv-NET were insufficient to report stratified cumulative or weekly rates. Consequently, only the overall rates were reported.
The RSV-NET data was marked as incomplete from May to June 2022.
The COVID-NET program began in March 2020 and has continued to the present day. Both the FluSurv-NET and RSV-NET programs started in October 2018 and also continue to the present day.
Active surveillance dates within a season begin in October (MMWR Week 40) and end either at the end of April (MMWR Week 17), June (MMWR Week 16), or September (MMWR Week 39) of the following year. In some seasons, programs concluded surveillance in different months 7–9. Refer to page 6 of the embedded PowerBI dashboard provided below for specific details about the date ranges captured in each season by surveillance program.
The number of hospitals participating in each respiratory illness surveillance program varies, but all platforms have overlapping surveillance areas. Fourteen states submit data for influenza surveillance, while thirteen states submit data for COVID-19 and RSV surveillance.
Although the areas surrounding the hospitals are well described, these regions are pathogen-specific and may not directly correspond between the three programs represented in RESP-NET.
Prior to the 2020-21 season, population estimates utilized NCHS bridge-race estimates. Starting with the 2020-21 season, unbridged estimates from the 2020-22 Special Tabulation vintage have been used as denominators.
Data reported within a season are considered to be preliminary and are subject to updates, as reports are received from participating sites. This reporting lag is expected to increase around holidays and during periods of increased hospitalization 8,10.
Surveillance relies on clinical testing ordered by healthcare providers and reported rates do not account for any resulting undertesting, differences between facilities, or diagnostic test sensitivity.
Coverage Map
RESP-NET provides summary tables and visualizations showing the participating states and the date ranges for active surveillance for every season, organized by surveillance program. The RESP-NET PowerBI dashboard displaying this information has been embedded below 7.
For more details, navigate to page 6, titled About Our Data.
RSV-NET
The Respiratory Syncytial Virus Hospitalization Surveillance Network (RSV-NET) is a network that conducts active, population-based surveillance for laboratory-confirmed RSV-associated hospitalizations in children younger than 18 years of age and adults. RSV-NET, along with the Coronavirus Disease 2019 (COVID-19) Hospitalization Surveillance Network (COVID-NET) an the Influenza Hospitalization Surveillance network (FluSuv-NET), comprise the Respiratory Virus Hospitalization Surveillance Network (RESP-NET). The RESP-NET platforms have overlapping surveillance areas and use similar methods to collect data. Because the surveillance areas and age groups included in surveillance have changed over time, trends should be interpreted with caution. RSV-NET is CDC’s source for important data on rates of hospitalizations associated with RSV. Hospitalization rates show how many people in the surveillance area are hospitalized with RSV, compared to the total number of people residing in that area. are preliminary and subject to change as more data become available. Data will be updated weekly.
Program Landing Page:
data.gov API Source:
Data Dictionary
| Signal | Description |
|---|---|
variable
|
|
variable
|
Limitations
Many of the same limitations apply here.
Number of laboratory-confirmed cases of the virus per 100,000 people. RESP-NET is an active surveillance system with a well-defined population, so it is possible to obtain estimates of incidence (cases/population). This indicator is only available for certain states. The data comes from certain counties within the state and does not capture the entire population.
The populations for children under 6 months and those aged 6-12 months are each estimated to be half of the population of children aged 0-1 years.
Coverage Map
RSV-NET provides summary visualizations showing the participating states and an estimate of the percentage of the population represented for each season. The RSV-NET PowerBI dashboard displaying this information has been embedded below 9.
For more details, navigate to page 8, titled About Our Data.
National Wastewater Surveillance Program (NWSS)
Testing for viruses in wastewater can provide an efficient way to identify the levels of virus circulation in the community. The CDC’s National Wastewater Surveillance program obtains samples from select communities around the US and presents the data aggregated by state. These data can provide an early indication of viral activity. Limitations include an inability to link detected back to particular groups of individuals and challenges in comparing viral levels across sites that use different methodologies. The sewage is not necessarily representative of the entire population. These data are made available by the CDC.
Program Landing Page:
cdc.gov API:
Limitations
- Wastewater virus activity level (WVAL) value is calculated by CDC as exp(# of standard deviations above baseline). The WVAL therefore represents a measure of relative intensity. The methods for calculating the baseline vary by virus.
Google Trends API
Google Trends data represent scaled Google search volume by week and state or metropolitan area and are obtained from the Google Health Trends API for non-commercial purposes.
This represents the volume of Google searches for ‘RSV’, statistically adjusted to remove searches related to RSV immunizations. We obtained Google Search volume by week and state for ‘RSV’ and for the category ‘Respiratory syncytial virus vaccine’ (11j30ybfx6). We are interested in searches related to clinical activity, not due to searching for information on the vaccine. These two time series show distinct patterns. Searches for RSV vaccines surge in early-Fall. Searches for ‘rsv’ in general have a small hump in early fall that corresponds to the searches for vaccine information, and a second peak later in the fall and into the winter that corresponds to the seasonal epidemic.
To subtract out the signal of vaccination searches from the general RSV signal, we fit a mixed effects model where the outcome was search volume for RSV by state and week, the covariate was search volume for RSV vaccines by week and state in the summer months, and there was a random intercept for state. There was an interaction term for ‘season’, which allowed the effect of search volume to differ in July-October. Using the regression coefficients, we calculate the adjusted search volume as Adjusted volume = RSV_search_volume - season*2.72*vax_search_volume - (1-season)*3.41*search_volume_vax. The latest raw, unadjusted search volumes are available from here.
National Respiratory and Enteric Virus Surveillance System (NREVSS)
Select laboratories around the United States participate in this system and report the number of positive tests for a virus and the number of tests performed. These data are made available by the CDC.