By Ciara O’Sullivan | October 19, 2021
The COVID-19 Health Inequities in Cities dashboard was created to compare and track inequities related to the COVID-19 pandemic across a multitude of dimensions, including across subgroups of individuals within cities, across neighborhoods within cities, and across cities. The newest addition to the dashboard focuses on describing and comparing inequities by race/ethnicity in COVID-19 outcomes over the course of the pandemic.
City specific trends of racial disparities over time
Within the “City Report: Inequities: Outcomes” tab, users can select a city of interest and examine trends in COVID-19 outcomes over time for different race/ethnicity sub-populations. Users can select whether to visualize case incidence, mortality, or hospitalization as their outcome of interest and filter by crude or age-adjusted rates. The trends are presented by quarters of the year, each being an aggregate of 3 months: Quarter 1 (Q1) is January-March, Q2 is April-June, Q3 is July-September, and Q4 is October-December.
Inequities by Race/Ethnicity over time
This chart compares various COVID-19 outcomes by race/ethnic groups and yearly quarters. Hover over the lines to see an explanation.
Figure 1. COVID-19 incidence over time by race/ethnicity in San Francisco, CA
Figure 1 shows the change in age-adjusted cases per 100k from Q1 2020 to Q3 2021 in San Francisco stratified by race/ethnicity. Here we can see that that Hispanic individuals in San Francisco saw the highest incidence rates of COVID-19 among the five race/ethnicity groups from the early pandemic until Q2 2021 when incidence rates among non-Hispanic(NH) Black individuals became the highest.
Hovering over each data point in the figure will expand a detail box that gives the exact case/100K population value, the rate ratio as compared to NH Whites and an explanation of the disparities between race/ethnicity groups. For example, in San Francisco during Q4 2020, Hispanic individuals experienced 6 times the incidence rate of COVID-19 cases compared to NH White individuals.
Please note that, for all these figures, we only include data points if they represent 20 or more deaths. This “suppression” of data is done to avoid very noisy values that can make these comparisons challenging. While this is a commonly used approach, it obscures data for smaller population subgroups (for example NH-Native American individuals). A more complete picture of disparities in this pandemic requires more complex analyses using, for example, vital registration data (Shiels, 2021).
Trends of racial disparities over time across cities
Another new feature within the “Compare Across Cities: Inequities: Outcomes: Individual Level” tab allows the user to view trends in COVID-19 outcomes by race/ethnicity for multiple cities simultaneously. Users can compare crude or age-adjusted rates over time for one of three outcomes described above. Figure 2 shows age-adjusted hospitalization rates over time among different race/ethnicity groups for 15 cities.
Inequities by Race/Ethnicity over time Across Cities
This chart compares various COVID-19 outcomes by race/ethnic groups and yearly quarters, using the CDC COVID-19 Case Surveillance Restricted file. Hover over the lines to see an explanation.
Figure 2. COVID-19 Hospitalization rates by race/ethnicity over time in various US cities
Using Figure 2, users can efficiently compare how different stages of the pandemic impacted various race/ethnicity groups across different cities. For example, during Q4 2020, Los Angeles saw much higher hospitalization rates among Hispanic and NH Black individuals compared to other racial/ethnic groups whereas Phoenix saw the greatest hospitalization rates among NH Native American individuals during the same period. Again, users can hover over any individual data point to obtain details on the exact rate, rate ratios compared to NH Whites, and explanations of the differences.
Tracking race and ethnicity data quality over time across cities
The final update, also in the “Compare Across Cities: Inequities: Outcomes: Individual Level” tab, includes a visualization displaying the percentage of data missing race/ethnicity information from each city and the change over time. Figure 3 shows how four highlighted cities (Minneapolis, Portland, San Francisco, and Washington) have varying percentages of missing race/ethnicity information within the incidence data reported to the CDC over time. While only cities that met certain thresholds for the level of cumulative missingness within the data are included in the analysis, we can see in Figure 3 that the percent of cases missing race/ethnicity data ranges from 3% in Washington Q2 2020, all the way to 41% in Portland Q3 2021. Interestingly, for many cities, the percentage of missingness in race/ethnicity data trends upward over time. A decrease in the data quality due to missingness makes it more difficult to fully understand and address health inequities.
Race/Ethnicity Data Quality in COVID-19 Outcomes by City Over Time
This panel displays how many of the reported COVID-19 outcomes in a given city is missing Race/Ethnicity information over time, using the CDC COVID-19 Case Surveillance Restricted file. You can hover the highlighted trend lines for details.
Figure 3. Percent of reported cases missing race/ethnicity information over time in select US cities
The emergent nature of the COVID-19 pandemic has brought constant changes and additional challenges; there have been several waves of infections, novel variants, changing policies, and innovative preventative measures, like vaccines, introduced which all change the landscape of this pandemic. This dashboard is an important tool to help evaluate how inequities related to race/ethnicity may have changed over time in the context of all these developments. Understanding a fuller picture of the disproportionate impacts of this pandemic on different populations is vital to ensure that health equity is at the forefront of the response to the next health crisis.
References
Shiels, Meredith S., et al. “Racial and ethnic disparities in excess deaths during the COVID-19 pandemic, March to December 2020.” Annals of internal medicine (2021).