TL;DR
This retrospective cohort study of over 2.7 million U.S. adults found that Moderna’s bivalent booster (mRNA-1273.222) was significantly more effective than Pfizer-BioNTech’s (BNT162b2) at preventing COVID-19-related hospitalizations and outpatient visits during the Omicron BA.4/BA.5 wave. The relative advantage of the Moderna vaccine was most pronounced in adults aged 65 and older.
What Problem Does This Paper Address?
As the SARS-CoV-2 virus evolved, the original monovalent vaccines saw a decline in effectiveness, particularly against the highly transmissible Omicron variants. In late 2022, "bivalent" boosters—which target both the original strain and the Omicron BA.4/BA.5 subvariants—were deployed to broaden protection.
While both Moderna and Pfizer-BioNTech produced bivalent mRNA vaccines, clinicians and policy-makers lacked direct, head-to-head "real-world" evidence comparing their effectiveness. This paper addresses that gap by evaluating which vaccine provided superior protection against severe (hospitalization) and moderate (outpatient) COVID-19 outcomes in a massive, diverse U.S. population.
Methods
The researchers conducted a retrospective cohort study using a robust Real-World Evidence (RWE) approach. Key methodological features include:
- Data Integration: The study linked primary care Electronic Health Records (EHR) with pharmacy and medical claims data to create a comprehensive view of patient health and vaccination status.
- Propensity Score Weighting: To ensure a fair comparison, the authors used Inverse Probability of Treatment Weighting (IPTW). This adjusted for baseline differences between the two vaccine groups, such as age, sex, race, geography, and pre-existing health conditions.
- Cox Regression Models: These were used to estimate adjusted hazard ratios (HR), which were then converted into Relative Vaccine Effectiveness (rVE).
- Negative Control: To check for "healthy user bias" or residual confounding, the researchers analyzed hospitalization risk within the first 7 days after vaccination (before the vaccine is expected to work). The lack of difference in this window suggested the groups were well-balanced.
Datasets
The study utilized a nationwide U.S. dataset merging two primary sources:
- Veradigm EHR dataset: Includes data from Allscripts and Practice Fusion platforms.
- Komodo dataset: Provides comprehensive pharmacy and medical claims.
- Study Population: A total of 2,748,358 eligible adults (1,049,575 Moderna recipients and 1,698,783 Pfizer recipients) vaccinated between August 2022 and February 2023.
Key Insights and Ideas
- The "Moderna Advantage": Consistent with previous studies of monovalent vaccines, the Moderna bivalent booster showed a statistically significant edge over Pfizer in preventing both hospitalization and outpatient visits.
- Age-Dependent Benefits: The relative effectiveness of Moderna increased with age. This is critical because older adults face the highest risk of severe outcomes and often exhibit "immunosenescence" (a waning immune response to vaccination).
- Biological Plausibility: The authors suggest several reasons for the difference, including Moderna’s higher mRNA dosage (50 mcg vs. 30 mcg) and clinical data showing that Moderna tends to elicit higher antibody titers, more robust T-cell responses, and higher levels of mucosal-protective IgA.
- Durability: The differences between the vaccines were evident within weeks of administration and persisted throughout the six-month follow-up period.
Key Results
The study calculated the Relative Vaccine Effectiveness (rVE), representing the additional protection offered by Moderna compared to Pfizer:
- Overall Hospitalization (Primary Outcome): Moderna was 9.8% (95% CI: 2.6%–16.4%) more effective than Pfizer.
- Overall Outpatient Visits: Moderna was 5.1% (95% CI: 3.2%–6.9%) more effective.
- Adults ≥ 65 Years: The gap widened significantly in seniors, with Moderna showing 13.5% higher effectiveness against hospitalization and 10.7% higher effectiveness against outpatient visits.
- Sensitivity Analysis: Results remained consistent even when the "cohort entry date" was shifted or when "open" claims data were used, reinforcing the reliability of the findings.
Limitations and Future Work
- Observational Nature: Despite rigorous statistical adjustments, unmeasured confounding factors (like individual masking behavior or exposure levels) could still influence results.
- Short Follow-up: The study only tracked outcomes for up to 6 months. Long-term comparative durability remains to be seen.
- Variant Evolution: The study focused on the BA.4/BA.5 period. Effectiveness against newer subvariants (like XBB or JN.1) requires ongoing evaluation.
- Population Bias: The study only included individuals with at least one healthcare contact in the prior year, potentially biasing the sample toward a more "at-risk" or health-seeking population rather than the general healthy public.
- Funding: It is worth noting that the study was supported by Moderna Inc., and several authors are Moderna employees.
Summary updated