Comparative efficacy of a third dose of mRNA-based COVID-19 vaccine in US veterans

Target test specifications

We designed an observational analysis to emulate a hypothetical pragmatic trial of the risk of a third dose of BNT162b2 compared to mRNA-1273 and the outcome of COVID-19 in the VA healthcare system.Supplementary table 3 summarizes the main protocol components.

Eligibility criteria included veteran status. At high risk for his COVID-19 severe between October 20 and November 18, 2021 65 years of age or older or ages 18-64 (with at least one of his coexisting conditions listed in the table) based on the existence of 1)⁸ or 18 years of age or older between November 19, 2021 and February 8, 2022 (based on national guidelines for 3rd dose rollout)^9,Tenreceived a second primary series of mRNA vaccines at least 6 months earlier (based on the same guidelines); known residence outside a long-term care facility; known smoking status and known body mass index in the previous year. Individuals had to have been in the VA healthcare system during the previous year (defined as receiving care at a station eligible to administer the vaccine under study and having at least one primary care visit). ) but for the past 3 days (may indicate onset of symptomatic disease).

The intervention of interest was a third dose of BNT162b2 or mRNA-1273 vaccine. To ensure a balance of key characteristics across groups, eligible veterans in the eligible trials were stratified defined by calendar day of their third dose (5-d bin), calendar month of their second mRNA vaccine dose Within, you will be randomly assigned to one of these two vaccines. Age (5-year bins), gender (male, female), race (white, black, other, unknown), urbanity of residence (urban, non-urban), geographic location (Veterans Integrated Services Network coded as 19 categories of ) and the number of SARS-CoV-2 tests performed in the last 12 months (0, 1, ≥2).

Five outcomes of interest were documentation of SARS-CoV-2 infection, documentation of symptomatic COVID-19 and COVID-19-related hospitalizations, ICU admissions and death. For each eligible individual, follow-up will begin on the day they received their third dose of vaccine (baseline), subject outcome, death, 112 days (16 weeks) after baseline, or the end of the study The period ended on February 15, 2022, whichever comes first.

Our targeted trial will assess the comparative efficacy of a third dose of vaccine during a period spanning the period when SARS-CoV-2 delta and omicron variants were endemic. The Delta subspecies has dropped to his 26% share of the subspecies in circulation in the US as of December 25, 2021. This is because it was quickly superseded by the Omicron variant, which rose to 100% share on February 12, 2022.¹¹To assess comparative efficacy during the period in which the Omicron variant predominated, the recruitment period was 1 January to 1 March 2022, and the SARS-period in which the only results of interest were documented was CoV-2 infection because it was too short to get enough of the rarer and more serious consequences.

Target trial emulation

Both target trials were emulated using the VA Healthcare database³Immunizations were identified using records from the Corporate Data Warehouse and the VA COVID-19 Shared Data Resource. SARS-CoV-2 infection identified using VA COVID-19 National Surveillance Tool^3,12Integrate clinical laboratory data with natural language processing of clinical notes to capture documented diagnoses inside and outside the VA healthcare system. Symptomatic COVID-19 was defined as one or more of the following symptoms documented within 4 days of SARS-CoV-2 infection: fever, chills, cough, shortness of breath or difficulty breathing, sore throat, Loss of taste or smell, headache, muscle/myalgia, diarrhea, vomiting. COVID-19 hospitalizations were defined as hospitalizations within 21 days of documented SARS-CoV-2 infection, COVID-19 ICU hospitalizations were defined as ICU hospitalizations during COVID-19 hospitalization, and COVID-19 deaths were , defined as death within 21 days. 30 days after documented SARS-CoV-2 infection.Supplementary table Four We provide detailed information on all study variables and their confirmation.

Target trial stratification by matching people who received a third dose of BNT162b2 and mRNA-1273 based on calendar day of third dose, calendar month of second dose, age, gender, race We attempted to mimic different randomizations, urbanity of residence, geographic location, and number of SARS-CoV-2 tests performed in the previous 12 months.Uses the same matching algorithm as described in previous studies³.

To investigate possible remaining confounding (for example, due to underlying health conditions or medical-seeking behavior), we included controls for two negative outcomes.⁷First, we assessed the risk of symptomatic COVID-19 in the first 7 days after the third vaccination. Then, during follow-up, he assessed the risk of death from causes other than COVID-19.

statistical analysis

Covariate balance after matching was assessed by plotting mean differences between vaccinated group variable values ​​(standardized by continuous variables), and differences of 0.1 or less were considered acceptable.¹³.

Cumulative incidence (risk) curves for vaccine groups were estimated using the Kaplan-Meier estimator¹⁴We then calculated the 16-week risks for each outcome and compared them between vaccine groups via differences and ratios. We performed subgroup analyzes by time elapsed since completion of the COVID-19 vaccine primary series (6–7, 8, or >9 months) and vaccine type. Primary series (BNT162b2 or mRNA-1273). (1) previously documented SARS-CoV-2 infection, (2) did not receive a homologous third dose compared to the COVID-19 vaccine primary series, (3) eligible to receive a third dose A sensitivity analysis was performed to exclude individuals with Since the dose of mRNA-1273 varies with the booster dose (50 μg), boosters were administered based on the procedural codes available in the “inpatient” and “outpatient” domains to compare vaccine recipients with known doses. Dose could not be specified (50 μg) vs 3rd primary dose (100 μg).

Percentile-based 95% confidence intervals for all estimates were calculated using a nonparametric bootstrapping procedure (including both matching and subsequent analysis) with 500 replicates.

Analyzes were performed using R software version 3.6.0 (R Foundation for Statistical Computing) and SAS Enterprise Guide version 8.2 (SAS Institute).

Report overview

For more information on the study design, please visit Overview of the Nature Portfolio Report Link to this article.