To estimate all-cause excess mortality as the dependent variable, e.g., for 2022, I first carried out the following calculations: (total deaths in 2022/population in 2022)/((total deaths in 2018+2019)/(total population in 2018+2019)). Next, I multiplied the expression by 100.

If a county reported 1-9 deaths, the CDC database coded them as missing. For consistency, I also coded 0 reported deaths as missing for a very small number of counties. For missing data in either 2018 or 2019, I coded them as missing for both years.

The independent variable – counties’ COVID-19 vaccination doses per capita – was modeled in the lagged year. I.e., for the 2022 analysis, I included vaccine data at the end of 2021, and for the 2023 analysis, at the end of 2022.

Vaccine data were included from counties reporting positive values for Completeness_pct ( US Centers for Disease Control, 2025b) (see below for the inclusion of this concept as a control). To model a proxy for doses per capita, I first summarized the number of doses administered in each county for Administered_Dose1_Recip, Series_Complete_Yes, Booster_Doses, Second_Booster_50Plus, and Bivalent_Booster_5Plus. Next, I divided the number by the population sizes in 2021 and 2022, respectively, and multiplied the result by 100. One county reporting more than 500 doses administered per 100 at the end of 2022 was omitted from the 2023 analysis.

For the 2022 analyses, I included lagged dependent variables for 2021 and 2020 as controls, and for the 2023 analyses, I included lagged dependent variables for 2020, 2021, and 2022 as controls. The approach accounts “for historical factors that cause current differences in the dependent variable that are difficult to account for in other ways” ( Wooldridge, 2006, p. 315). Moreover, including “additional lags yields more accurate parameter estimates” ( Wilkins, 2018, p. 393).

The Completement_pct concept “[r]epresents the proportion of people with a completed primary series whose Federal Information Processing Standards (FIPS) code is reported and matches a valid county FIPS code in the jurisdiction” ( US Centers for Disease Control, 2025b). I include it as a control, since a relatively low value may be a proxy for underreporting of vaccine uptake.

In the following regression analyses, the Completement_pct at the end of 2021 ranges from 59.8 to 99.1, with a mean of 95.0 and a median of 97.5. At the end of 2022, the Completement_pct ranges from 73.4 to 98.9, with a mean of 95.8 and a median of 97.5.

Model 1 shows that a one-unit increase in per-capita vaccination uptake at the end of 2021 was significantly associated with a.042 (95% CI: .030–.055) increase in 2022 all-cause excess mortality. The variance inflation factor (VIF) of 1.43 indicates no multicollinearity concerning vaccine uptake.

Below the bold line, Model 1 reports that the weighted average – i.e., the “overall” – per-capita vaccine uptake at the end of 2021 was 144.3 doses per 100. Using Stata’s margins post-estimation command ( Williams, 2012) with that number, on the Model 1 vaccine estimate, returned a value of 113.8 (95% CI: 113.4–114.1). I.e., when including control variables, the 2022 all-cause excess mortality was 13.8%, based on the county-population-weighted average of vaccine uptake at the end of 2021. Assuming zero vaccine uptake using the margins post-estimation command with that number, on Model 1 estimates, returned a value of 107.7 (95% CI: 106.0–109.4). I.e., the all-cause excess mortality was 7.7% when assuming zero vaccine uptake at the end of 2021.

The findings imply that the weighted average of vaccine uptake was associated with 5.67% (95% CI: 3.89–7.46) higher mortality than assuming zero vaccine uptake. As the US in 2022 had 3,279,857 deaths ( US Centers for Disease Control, 2024a), this implies that the weighted average vaccine uptake was associated with 176,031 (95% CI: 123,590–228,473) additional deaths compared to zero vaccine uptake. (The CIs in this paragraph were achieved by using Stata’s nlcom algebra function on the margins post-estimations.)

Model 2 shows that a one-unit increase in per-capita vaccination uptake at the end of 2022 was significantly associated with a.030 (95% CI: .024–.036) increase in 2023 all-cause excess mortality. The VIF of 1.35 indicates no multicollinearity concerning vaccine uptake.

Below the bold line, Model 2 reports that the weighted average – i.e., the “overall” – per-capita vaccine uptake at the end of 2022 was 194.3 doses per 100. Using Stata’s margins post-estimation command ( Williams, 2012) with that number, on the Model 2 vaccine estimate, returned a value of 106.9 (95% CI: 106.6–107.1). I.e., when including control variables, the 2023 all-cause excess mortality was 6.9%, based on the weighted average vaccine uptake at the end of 2022. Assuming zero vaccine uptake using the margins post-estimation command with that number, on Model 2 estimates, returned a value of 101.1 (95% CI: 100.0–102.2). I.e., the all-cause excess mortality was 1.1% when assuming zero vaccine uptake at the end of 2022.

The findings imply that the weighted average vaccine uptake was associated with 5.73% (95% CI: 4.53–6.93) higher mortality than assuming zero vaccine uptake. As the US in 2023 had 3,090,964 deaths ( US Centers for Disease Control, 2024b), this implies that the weighted average vaccine uptake was associated with 167,566 (95% CI: 134,417–200,715) additional deaths compared to zero vaccine uptake.