In a recent article published in BMC Public Health, researchers analyzed the association between time-varying rates of vaccination against coronavirus disease 2019 (COVID-19) and the risk of hospitalization for cases of COVID-19 (CHR), a proxy for the burden of disease at the individual level and the burden of disease on health care systems of population level in different waves of variant severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in the United States (US).
survey: Association between vaccination rates and COVID-19 health outcomes in the United States: a population-level statistical analysis. Image Credit: Nhemz/Shutterstock.com
Background
There were 1.1 million deaths in the US due to COVID-19 by March 1, 2023. Vaccines against COVID-19 have most effectively limited the disease and its impact, including the socioeconomic burden on individuals and the nation’s health care system.
However, studies evaluating the effectiveness of vaccines against COVID-19 have relied on individual-level data, confounded by unquantified factors and inconsistent quality.
Thus, for the US, high-resolution data were not available at the population level, reflecting the real-world relative associations between available COVID-19 vaccines and CHR against COVID-19 over time.
About the research
The present study used generalized additive models (GAMs) to examine the relationship between rates of vaccination against COVID-19 and CHR in 48 US states between April 19, 2021, and March 1, 2022.
The study model captured nonlinear dynamics, accounting for dynamic (time-varying) and static (time-constant) factors that potentially contribute to COVID-19 CHR and disease transmission.
The former were natural immunity derived from previous severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, government policies, activity-related levels of population engagement, and local health infrastructure.
In contrast, the latter includes each country’s Social Vulnerability Index (SVI), race/ethnicity, comorbidities, and health care costs considered significant in previous studies.
In addition, the model accounted for COVID-19 incidence rates (CIR) as a separate outcome variable. The study framework used a variable transformation method to capture spatial variation in relative associations.
The study period spanned the pre-Delta, Delta, and Omicron waves of COVID-19, and they evaluated all of them independently in this study.
Results
The study results point to several key findings about the effects of the COVID-19 vaccine at the population level in the US
GAMs using the relative CHR of COVID-19 (RCHR) as the outcome variable showed variance explained values between 46.8% and 72.3% for variant waves.
Furthermore, the correlation between observed and predicted RCHRs shows strong positive correlations ranging from 0.67-0.81.
Vaccination at the population level was significantly associated with reduced CHR of COVID-19.
Interestingly, past SARS-CoV-2 infections (one to four months) showed strong negative associations with RCHR in different waves; however, this effect remains variable and inconsistent at both the individual and population levels.
Activity-related levels of population engagement (eg, gym attendance), government policies, and local health infrastructure added to the explanatory power of the study model, favoring the importance of reporting them on population-level outcomes of COVID-19 vaccines.
However, their associations were inconsistent over time and across variants. For example, the relationship between CHR of COVID-19 and hospital visits went from negative to positive between the pre-Delta to Delta and Omicron waves.
Additionally, the observed correlation between relative weekly testing frequency and RCHR was negative and decreased from the pre-Delta wave to the Omicron wave.
Additionally, US states with higher SVI consistently show higher RCHR, and Medicaid spending per person shows a consistent negative relationship with RCHR.
GAMs using RCIR as the outcome variable showed lower performance, suggesting a more dynamic relationship in terms of COVID-19 transmission, especially during the Omicron wave.
The variance explained for the Omicron-Booster-RCIR model was 17%, suggesting that booster vaccination provided additional protection against severe COVID-19 during Omicron waves. However, their effect on Omicron infection itself was limited.
Conclusions
The study provides robust evidence for the efficacy of the COVID-19 vaccines against CHR of COVID-19 in different waves of variants in the United States.
Despite the emergence of new variants, vaccines remain effective and have remarkably mitigated the adverse effects of COVID-19 and the socioeconomic burden on healthcare systems. These data can help inform future public health policies in the US
Future studies should identify other factors that may capture the dynamics of COVID-19 transmission during the Omicron period.
Additionally, studies should explore the complex and evolving nature of COVID-19 transmission.
