CDC’s Wastewater Surveillance Data Methodology

How CDC Ensures High-Quality Wastewater Data

CDC receives data from about 1,500 wastewater surveillance sites across the United States each week. CDC's experts work hard to ensure these data are of high quality, comparable across sites, and understandable.

• Protecting data quality
  • Before publishing updated wastewater data, CDC completes data quality checks and reviews the data to ensure there aren’t any issues.
• Ensuring data comparability
  • CDC has developed the wastewater viral activity level (WVAL) metric to allow data to be compared between wastewater surveillance sites and combined into state/territorial, regional, and national levels.
  • The WVAL also standardizes data to see trends over time.
• Increasing understanding of data
  • CDC works with state, tribal, local, and territorial health departments to ensure that metrics and visualizations are easy to understand and use for public health and individual decision making.
• Continuous improvement
  • Wastewater surveillance is an evolving science, and CDC may update methods and visualizations to improve how wastewater data are shared.

Understanding the Wastewater Viral Activity Level (WVAL)

What is the WVAL?

The WVAL compares current levels of COVID-19, flu, or RSV to low viral levels at that location over the last 24 months.

Why is the WVAL important?

For You: If you see increased wastewater viral activity levels, it might indicate that there is a higher risk of infection. You can use these data to make informed decisions on how to protect yourself and others from respiratory viruses.

For Public Health: Public health officials use data from wastewater (sewage), along with clinical data, to understand virus levels in a community.

The WVAL helps public health officials:

• See the bigger picture across a state, region, or the United States
• Compare data across cities, states, and time
• Spot unusual increases in COVID-19, flu, or RSV

What data should I look at?

• Current levels: This shows the current status of the virus in your community.
• Trends: Is the virus increasing, decreasing, or staying the same?
• Other data: Look at other data, such as clinical data, to see if they’re showing similar trends.
  • Example resource: Respiratory Illnesses Data Channel

How is the WVAL calculated?

What’s Needed to Begin

• For each site, we need 8 weeks of data using the same lab method.
  • Why? A minimum amount of data are needed to help avoid extreme fluctuations that may occur with less data.

Step 1: Group Data Together

• The amount of virus present in a sample grouped by:
  • Which virus is being measured
  • Which site the sample came from
  • Which lab and laboratory method are used
• Why? This helps ensure that we’re comparing similar data.

Step 2: Data Validation

• To ensure data are ready for comparison, we:
  • Log transform the data
    • Why? This helps us to see overall trends more clearly by reducing the influence of extreme values.
  • Remove unusually high or low values (outliers)
    • If future data continue the high or low trend, the data are no longer removed.
    • Why? Unusually high or low values can skew the data, leading to inaccurate interpretation.

Step 3: Determine Low Levels for Each Site

• When looking at data from the past 2 years, we select the value that is higher than 10% of values to be used as a baseline for comparison.
  • Why? The baseline shows what is considered a typical low level of the virus for a site. The WVAL shows how far away current levels are from low.
• These values are recalculated on April 1 and October 1 for COVID-19, and on August 1 for flu and RSV each year. All historical WVAL values are then updated with the new baseline.
  • Why? Recalculating the baselines when the virus levels are typically low helps ensure that current levels continue to be compared to accurate low levels.

Step 4: Compare Current Levels to Baselines

• To calculate the WVAL, we compare the current amount of virus to the baseline level, then divide that by the standard deviation.
• The larger the WVAL value is, the further away from low levels it is.
• Why? This provides a standard number of how many “steps” (standard deviations) current levels are from low levels, allowing us to compare levels between sites.

Step 5: Identify Average and Median Values

• For each week, average the values.
• The middle (median) value is used to represent each state, region, and the United States.
• Why? This helps to show weekly and big-picture trends.

Step 6: Categorize Values

• Each virus has its own thresholds because each virus behaves differently.
• The WVAL values are grouped into one of five categories (see below for thresholds): Very low, low, moderate, high, or very high.
• Why? These categories help provide a quick understanding of the levels.

WVAL Aggregation

• State/territory aggregation: The median of site-level WVALs
• Regional aggregation: The median of state/territory-level WVALs
• National aggregation: The median of regional-level WVALs
• States and territories may have a higher density of sampling sites in certain geographic areas, so the median wastewater viral activity level may not represent the wastewater viral activity level for every community in the state or territory.

WVAL Inclusion Criteria

COVID-19

New wastewater sampling sites and sites with a substantial change in laboratory methods are included in national, regional, state, or territorial median values once there are at least 8 weeks of samples reported for that location.

Influenza A and RSV

New wastewater sampling sites and sites with a substantial change in laboratory methods are included in national and state or territorial median values beginning on August 1 of each year once there are at least 8 weeks of samples reported for that pathogen. Sites must begin sampling by October 1 each year for their wastewater viral activity level data to be displayed on the website. If sites begin sampling after October 1, their wastewater viral activity level data will not be displayed until August 1 of the following year.