Unusual Correlations: Populations That Are at a Higher Risk for COVID-19
Despite the pandemic officially coming to an end last year, the threat of COVID-19 is still looming large. We tend to forget that several populations are at higher risk for developing severe and possibly fatal responses to the virus. According to Mayo Clinic, the populations that are at the highest risk of serious COVID-19 symptoms are people aged 65 and older and people with underlying health conditions such as asthma, heart disease, and diabetes.
Using the American Community Survey database from 2021, I have identified three unusual correlations for groups that may be at a higher risk and have yet to be noticed. Rather than assessing risk based on health conditions, these variables consider social, political, and economic conditions. This database contains various demographic and COVID information for every county in the United States. I used a statistic called “COVID Deaths per Capita” to assess COVID deaths. This basically takes the population of each county and divides it by the number of COVID-related deaths for the year. Because the values were so small, I multiplied each ratio by the average county population of the United States so my visualizations could be interpreted more easily.
The first unusual correlation I identified was a higher COVID death per capita in single females. The correlation analysis I ran came back with a coefficient of 0.30 indicating that there is a moderate relationship between the two (see Figure 1). In other words, counties with a higher prevalence of single females, have more COVID deaths. On average, single females make up 26% of county populations in the United States. This minority has an even stronger correlation with unemployment and family poverty with a coefficient of 0.59 and 0.46, respectively. As a result, single females are more likely to face economic difficulties, which could prevent them from affording health care and seeking medical attention.
Next, we have broadband access and COVID deaths. This correlation test indicated that there was a moderate correlation of 0.476 between broadband access and deaths per capita (see Figure 2). In the context of this database, broadband access is measured as the percentage of households in a county with a broadband internet connection. Especially during the early stages of the pandemic, accessing information about COVID-19 was crucial. Internet access allowed individuals to receive up-to-date information on where the virus was spreading, practices to stay safe from the virus, and treatments for the latest strands. In the later stages of the pandemic, people needed to locate and book COVID booster shots, which could only be done using broadband internet. Populations without broadband access could face difficulties in obtaining this important information, which puts them at a greater risk for COVID.
Lastly, I was very interested in exploring a strange correlation between COVID-19 deaths per capita and Donald Trump voters. I used data from both the 2016 and 2020 elections where I found moderate correlations between people who voted for Trump in either of the elections and higher COVID death rates (see Figure 3). I also found that COVID-19 deaths in Trump voters have been trending upwards since 2016. The population that voted for Trump back in 2016 showed a correlation of 0.316 to COVID-19 deaths, then jumped to 0.349 in 2020. Studies have shown that political affiliation can have an impact on adherence to public health guidelines. One study showed that participants who approved of President Trump’s leadership regarding COVID-19 were less likely to engage in personal protective behavior such as mask-wearing, social distancing, and hand washing. Not engaging in safety measures for the virus could put Trump supporters at a higher risk of getting exposed to COVID-19. As the threat of COVID-19 persists, it is important to understand who may be at risk, regardless of how “unusual” these correlations may appear.
