What Factors Lead to Mortality Rate in the U.S. in 2021 During the Pandemic?
Contents:
Disclaimer
I am not a professional statistician and I did this report on my free time. Some of the math may be a little "fuzzy", but I am confident the gist of the report is fairly sound and has merit.
Introduction
This is an addendum to last year's report:
What Factors Lead to Mortality Rate in the U.S. in 2020 During the Pandemic?.
Some people have referred to the pandemic as a Pandemic of the Unvaccinated. With the high percentage of unvaccinated people receiving treatment at many hospitals, it certainly does appear that the vaccines have been helpful in reducing the number of deaths. However, a somewhat more in-depth look at the data suggests that health insurance and income may have been even more significant factors in 2021.
Multivariable Analysis
As in last year's report, a multivariable linear regression is used to look for correlations between many variables and mortality rate. This time around, two variables are dropped (Mask Mandate Days and Lockdown Days) and one new variable is added (% Vaccinated). The analysis yields the formula:
Y = 4097 - 59.73 * X1 + 11.42 * X2 + 3.110 * X3 - 13.86 * X4 + 44.99 * X5 + 2.155 * X6 + 1.212 * X7 - 24.99 * X8 - 38.22 * X9 + 67.34 * X10 - 19.46 * X11
The dependent variable Y is excess death rate (the number of deaths per 1,000,000 people). The 11 independent variables are listed in Table 1:
Table 1. Independent Variables Sorted by Decreasing Impact on Excess Death Rate
* - Correlation with excess death rate is not statistically significant.
! - The metric of population density used here is the average number of people per mile, assuming all people within a county are evenly distributed.
# - For this metric, black people are considered 0% white and Latinxs and Native Americans are considered 50% white.
Pandemic of the Unvaccinated? Or Pandemic of the Uninsured, Poor and Black & Brown People?
Surprisingly, this analysis suggests that in 2021 the uninsured, poor, and black/brown populations may have been at greater risk of death than the unvaccinated! In the 2020 analysis, % Uninsured (X10) and Median Income (X8) were not a factor. % White People (X4) is the only significant factor that remains the same. And while % Vaccinated (X11) is correlated with a lower mortality rate, the correlation is not statistically significant. However, it should be noted that if we retain all 11 factors, even those that are not statistically significant, the maximum potential gain from vaccination is greater than that from insurance or income. To illustrate, using the regression equation (above Table 1), if 100% of the population were vaccinated, it would have saved about 250,000 or half of the approximately 500,000 excess deaths in the United States in 2021 (as of November 8th). By contrast, if 100% of the population were insured, it would have saved about 200,000 people--less than 250,000, but still a lot. This brings to mind some questions. Why isn't the correlation with vaccination stronger? And why did insurance and wealth become big factors?
I dare say that the correlation with vaccinations would be significant if better data were used in the analysis. However, one thing that presumably lessens the apparent impact of the vaccines is that once people were vaccinated, they probably took more risks and exposed themselves and others to the virus more. Such a relationship is suggested in the analysis by the fact that % Biden Voters (X7) goes from being very beneficial in 2020 to not helpful at all in 2021. A separate regression of % Biden Voters (X7) and % Vaccinated (X11) shows a strong correlation--Biden voters apparently were much more likely to get vaccinated. So if they then increased their exposure to other people, you'd expect the benefit of being a Biden voter to go down.
As for the increased mortality risk in 2021 associated with lack of insurance and lower income, two things come to mind. First, effective treatments, such as monoclonal antibodies, have become available, but these treatments often come with a cost, more so to the uninsured. Second, pandemic unemployment benefits have ended in some states, forcing some people back to work in dangerous environments--crowded, indoors. And presumably this applies more often to poor people. It's a tale of two pandemics, if you will. Initially the risks were more equal and we were in it together more. And now capitalism kicks in more.
The third statistically significant variable is color. As in 2020, black and brown people are apparently at greater risk. Some of this presumably is biology (e.g. lower levels of Vitamin D) but some of it may be due to discrimination (e.g. less access to healthcare, beyond what is already accounted for with other variables, such as insurance, income, and population density.)
The final enormous surprise: in 2021 it actually decreases your risk to be obese! It seems like there has to be some mistake here, after everything we've heard about the risks of being obese, but maybe there's some logic to this. In the 2020 analysis, obesity was correlated to excess death, but it was also correlated to being a Trump voter. And in the analysis, being a Trump voter came out as being the significant risk factor, not obesity. Presumably, Trump voters were more likely to downplay the virus and were more likely to put themselves at risk. But in the 2021 analysis, being a Trump voter no longer appears to be a significant risk factor. However, in addition to being correlated to Trump voters, it turns out obesity is also correlated to having a low income and being black or brown. And the correlation to each of these three variables: politics, income and color, is statistically significant. And in the main analysis, income and color were the significant risk factors, not obesity. So it appears that other factors, such as politics, income, and color, have acted as confounding variables during the pandemic to exaggerate the perceived risk of being obese. The 2021 analysis even suggests that being obese is advantageous. Although this seems like a stretch, since Joe Rogan and others have been preaching the danger of obesity, perhaps enough obese people have taken greater precautions that it's noticeably lowered the risk to this population.
Conclusions
The results from the 2021 analysis are quite a bit different from those of 2020. They suggest a tale of two pandemics. In 2020, the virus surprised the country, putting people at greater risk who chose not to stay apart or who lived in close proximity. In 2021, the virus is no longer novel. There are vaccines and treatments. But instead of being merely a pandemic of the unvaccinated, it is considerably more complicated. I am confident the vaccines saved many lives. However, being uninsured and having a low income are risk factors, presumably because these populations have less access to treatment (and/or are less likely to seek it out).
Using a not-exactly-sound statistical extrapolation, it is estimated that 200,000 lives would have been saved if everyone were insured. This happens to be the same number that Jimmy Dore reported (
2020 analysis), but since he was using suspect logic and only data from 2020, I'll not apologize but instead chalk it up to lucky coincidence. And as for Joe Rogan and his doom and gloom concerning obesity, I'm doubling down on my assertion that the risk of obesity during the pandemic has been greatly exaggerated. ;-)
The one consistency between years has been that black and brown people appear to be at greater risk.
Go back to top for the
Disclaimer.