Evaluating the Association of COVID-19 Vaccination and New-Onset Diabetes as an Outcome Público
Lee, Sein (Spring 2024)
Abstract
While the general consensus is that the COVID-19 vaccine is critical for protecting individuals from severe outcomes and is safe, there have been mixed reports of adverse diabetes outcomes and increased risk of new-onset diabetes following vaccination. There are methodological challenges when investigating such an association, such as confounding, surveillance bias, and appropriately addressing COVID-19 vaccination as a time-varying exposure. This study investigates associations between new-onset diabetes and COVID-19 vaccination as a time-varying exposure among adults, and whether this association varies by pre-vaccination demographics and underlying diabetes risk.
This study was based on electronic medical records of adults enrolled in Kaiser Permanente. The outcome of interest is new-onset diabetes, and the exposure is COVID-19 vaccination. Sociodemographic and clinical covariates of interest include sex, age, race, smoking, exercise, BMI, COVID-19 infection, blood pressure, and Social Vulnerability Index (Center for Disease Control and Prevention 2020). Descriptive and exploratory analyses were performed using R and SAS. An unadjusted Cox time-variant proportional hazards model was created with just the exposure variable, and adjusted for all covariates of interest.
Out of 728,120 study participants, 505,763 were vaccinated at some time during the study period of March 2020 and December 2022 and 222,407 were not. 19,477 patients were diagnosed with new-onset diabetes during the study period. The unadjusted Cox time-variant model yielded a hazard ratio of 2.457 (95% CI: 2.322, 2.600) for vaccination. However, in the final Cox time-variant model adjusting for potential confounders, the hazard ratio was 1.144 (95% CI: 0.984, 1.330). Covariates associated with increased risk of new-onset diabetes were COVID-19 infection, BMI, obesity, blood pressure, smoking status, number of pre-pandemic ambulatory visits, age, and SVI percentile.
The findings suggest no association between COVID-19 vaccination and new-onset diabetes during our study period. The results emphasize the importance of control for confounding factors in the study of COVID-19 exposures and diabetes outcomes, especially during the early availability of the vaccine when access was prioritized to those at high risk for severe consequences of COVID-19 infection. There is also a need for careful analysis and clear communication of results regarding the safety of COVID-19 vaccination with respect to risk of new-onset diabetes for public health officials and policymakers.
Table of Contents
Introduction..........................1
Methods................................4
Data..................................4
Study Measures..................4
Analysis.............................5
Results...................................6
Discussion..............................9
Appendices.............................11
References..............................21
About this Master's Thesis
| School | |
|---|---|
| Department | |
| Subfield / Discipline | |
| Degree | |
| Submission | |
| Language |
|
| Research Field | |
| Palabra Clave | |
| Committee Chair / Thesis Advisor | |
| Committee Members |
Primary PDF
| Thumbnail | Title | Date Uploaded | Actions |
|---|---|---|---|
|
|
Evaluating the Association of COVID-19 Vaccination and New-Onset Diabetes as an Outcome () | 2024-04-08 11:35:58 -0400 |
|
Supplemental Files
| Thumbnail | Title | Date Uploaded | Actions |
|---|