A Louisiana Summer — COVID-19 and Hurricanes

Hurricane Laura shortly before making landfall in Cameron Parish

In late April, while New York City was locked down due to the COVID-19 pandemic, I was moved by an article in The New York Times about the racial disparities that resulted in divergent health outcomes for COVID patients in New Orleans. Specifically, the article focused on members of the Zulu club, a Black social organization in the city, and the tremendous loss of many of their members from the virus. This same mortality pattern was unfortunately not limited to New Orleans and has been present in most metro areas across the country, including New York City. Yet, despite this tragic consistency, I’ve reread the Times story several times over the summer months as hurricane upon hurricane reliably curled into the Louisiana coastline as though nature was playing a cruel joke.

I wondered how the assembly line of natural disasters had impacted the spread of COVID throughout the state and decided to analyze both COVID and demographic data for any trends. I created a PostgreSQL database using the NYTimes COVID repository as well as demographic data from the Census Bureau and retrieved parish-level data about the number of new cases and deaths from August 27th, when Hurricane Laura made landfall, until October 31st, several weeks after the landfall of Hurricane Delta on October 9th.

Then, I visualized the data using Tableau and looked for patterns, specifically focusing on median income, total population and the percentage of the population that identifies as Black via the Census. The interactive Tableau workbook can be found here if you want to explore the data yourself!

Here’s what I found:

• Primarily, I was surprised by the apparent lack of a direct correlation between the storm paths and new virus deaths. Possible exceptions might include the parishes of Catahoula and Franklin that saw a rise in cases/deaths and were along the track of Hurricane Delta as well as Allen Parish which lies close to the path of Hurricane Laura.
• There doesn’t appear to be a correlation based on total population; however, both storms traveled over less populous parishes rather than the major population centers of New Orleans and Baton Rouge.
• I included median household income in this analysis because I was curious if greater financial means — larger houses or the ability to escape the storm’s path — would lead to less virus transmission in more affluent parishes. It doesn’t appear that income is the primary factor influencing the dual impact of COVID and hurricanes but there are patterns worth noting. Cameron Parish, which saw the landfall of both hurricanes, has the highest median household income of any parish in southwest Louisiana and a noticeably lower COVID death rate than nearby parishes such as Allen. Conversely, less affluent parishes in the northeast affected by Delta appear to have higher COVID death rates than wealthier adjacent parishes.
• Parish demographics offer an interesting perspective that circles back to the NYTimes story that inspired this project. Black Americans have been more vulnerable to the virus across the country due to factors such as disparate healthcare accessibility, population density and proximity to pollution that can cause and exacerbate underlying conditions such as hypertension. The data does not appear to show a direct correlation between parish demographics and COVID deaths although there are important caveats. Parishes with the highest percentage of Black residents are disproportionately located in parts of the state that did not bear the brunt of the storm that occurred near the coastline. Additionally, the storms did not directly impact the major cities and I am unclear how extensively COVID testing is available in rural parishes with large Black populations so the data may not be comprehensive enough to draw concrete conclusions at this time.

This analysis offers a glimpse into the health impacts of dual disasters but there are other impacts and analyses that may shed further light on the full picture. For example, I used new COVID cases and deaths but using new hospitalization data might reveal different and important trends. Similarly, I used hurricane data as a lens through which to view the spread of COVID but the loss of life and property damage caused by the storms certainly deserve their own focus.