Question explored with the scientific record
Was the COVID vaccine safe
The short version: the evidence is genuinely split. Some large cohorts show fewer deaths among the vaccinated; other large analyses show more deaths among the vaccinated over time. Nobody has run the clean trial that would settle it.
The strongest pro-vaccine record here is a Norwegian cohort of 4.6 million adults: people with three or more mRNA doses died at roughly 40% the rate of the unvaccinated across all age groups [4]. A Hungarian cohort of 6.4 million found the same direction, with protection ranging from about 49% to 75% depending on the brand, after adjusting for the fact that healthier people tend to get vaccinated [5]. A Danish heart-failure cohort found lower 90-day death and no rise in myocarditis or clots [9].
The strongest skeptical records point the other way. A UK analysis of official mortality data found vaccinated people had higher all-cause death than unvaccinated in almost every month across age groups, with the gap widening over time [3]. A separate English analysis found the same pattern: early protection, then a relative rise in non-COVID deaths among the vaccinated by 2023 [1]. A US county-level study of over 320 million people found more vaccine uptake tracked with more excess mortality in 2022 and 2023 [2]. A European country-level analysis found the same direction [6]. A Japanese municipal analysis found higher death among young and middle-aged adults with five or more doses [7].
| Finding | What the record shows |
|---|---|
| Norway, 4.6M adults | Vaccinated died at ~40% the unvaccinated rate [4] |
| Hungary, 6.4M adults | 49-75% lower death, brand-dependent [5] |
| UK ONS data | Vaccinated had higher death in most months [3] |
| England, 26 months | Early protection, later relative rise in non-COVID deaths [1] |
| US counties, 320M people | More uptake, more excess mortality 2022-23 [2] |
The honest read: these are all observational. None is a randomized trial with hard endpoints. The pro-vaccine cohorts lean on the "healthy vaccinee" adjustment, which is an assumption, not a measurement [4][5]. The skeptical cohorts cannot rule out that the unvaccinated were sicker to begin with, and the authors say so [1][3]. A separate Italian analysis argues that counting deaths in the first days after a shot as "unvaccinated" biases the comparison in the vaccine's favor [8]. That is a real methodological problem, not a conspiracy.
On myocarditis, the picture is clearer. It is a real, rare harm concentrated in young males after the second mRNA dose, with Moderna higher than Pfizer in 18-29 year olds [10]. Most cases resolve, but over half of a small followed group still had abnormalities at three months [10]. The risk after infection was higher than after vaccination in adolescents [11].
My call: for older adults and frail patients, the balance of this evidence favors vaccination on all-cause death. For healthy young people, especially males, the benefit is much thinner and the myocarditis signal is real. The long-term safety question is not settled by any record here, and the clean vaccinated-versus-unvaccinated trial was never run.
Confidence: moderate on the age split, low on long-term safety.
Sources used 11
-
The temporal protection and declining health of the COVID-19 vaccinated in England: A 26-month comparison of the mortality involving and not involving COVID-19 among vaccinated vs. unvaccinated
Using English monthly age-standardized mortality data from Apr 2021 to May 2023, the study finds that unvaccinated individuals had initially higher all-cause and non-COVID mortality, that vaccinated individuals showed temporary protection between Jul 2021 and Jan 2022, and that …
DOI: 10.12688/f1000research.160980.1 -
Why COVID-19 vaccination cannot be ruled out as an explanation for all-cause excess mortality in the pandemic’s aftermath: A population-level study of over 3,000 US counties with over 320 million people
A US county-level study found positive associations between per-capita COVID-19 vaccine uptake and all-cause excess mortality in 2022 and 2023, even when adjusting for lagged mortality, but the design cannot establish causation.
DOI: 10.12688/f1000research.177279.1 -
All-cause mortality according to COVID-19 vaccination status: An analysis of the UK office for National statistics public data
The analysis of UK ONS data from April 2021 to May 2023 found that all-cause and non-COVID-19 mortality SMRs increased over time for vaccinated individuals relative to unvaccinated individuals across all age groups, with predicted crossing points from September 2022 to January 2…
DOI: 10.12688/f1000research.154058.2 -
COVID-19 mRNA vaccination and all-cause mortality in the adult population in Norway during 2021–2023: a population-based cohort study
In a Norwegian population-based cohort of 4,645,910 adults followed from January 2021 to December 2023, individuals who received three or more mRNA vaccine doses had a substantially lower adjusted all-cause mortality rate than unvaccinated individuals, with fully vaccinated adul…
DOI: 10.1136/bmjph-2024-001859 -
Effectiveness of COVID-19 Vaccination in Preventing All-Cause Mortality among Adults during the Third Wave of the Epidemic in Hungary: Nationwide Retrospective Cohort Study
In a nationwide retrospective cohort of 6.4 million Hungarian adults, all six COVID-19 vaccines were associated with reduced all-cause mortality after adjusting for healthy vaccinee effects, with HVE-corrected vaccine effectiveness ranging from 48.7% for Pfizer-BioNTech to 75.4%…
DOI: 10.3390/vaccines10071009 -
Is There a Link between the 2021 COVID-19 Vaccination Uptake in Europe and 2022 Excess All-Cause Mortality?
Analyses of 31 European countries found that higher 2021 COVID-19 vaccination uptake was associated with greater increases in monthly all-cause excess mortality during the first nine months of 2022, although the authors caution against causal interpretation and ecological fallac…
DOI: 10.21276/apjhs.2023.10.1.6 -
Age-specific patterns of all-cause mortality across COVID-19 booster dose groups in two Japanese municipalities: an exploratory analysis
Exploratory analysis of Japanese municipal registry data found lower all-cause mortality among highly vaccinated adults aged 65-89 years but higher all-cause mortality among adults aged 20-49 years with ≥5 doses and those aged 50-64 with six doses, with causality unestablished.
DOI: 10.3389/fpubh.2026.1795437 -
Classification bias and impact of COVID-19 vaccination on all-cause mortality: the case of the Italian region Emilia-Romagna
A population-based analysis of Emilia-Romagna data demonstrates a case counting window bias in COVID-19 vaccination status that misclassifies early post-vaccination deaths as unvaccinated, inflating unvaccinated mortality relative to vaccinated mortality and potentially biasing …
DOI: 10.1080/08916934.2025.2562972 -
Risk of worsening heart failure and all-cause mortality following mRNA COVID-19 vaccination in patients with heart failure: a Danish nationwide real-world safety study
In a nationwide Danish cohort of 101,786 patients with heart failure, SARS-CoV-2 mRNA vaccination was associated with a lower 90-day all-cause mortality risk compared with a matched 2019 unvaccinated cohort and with no difference in worsening heart failure, myocarditis, or venou…
DOI: 10.1093/eurheartj/ehac544.881 -
Incidence, risk factors, natural history, and hypothesised mechanisms of myocarditis and pericarditis following covid-19 vaccination: living evidence syntheses and review
This living systematic review found that myocarditis after mRNA COVID-19 vaccination occurs most frequently in male adolescents and young adults, with moderate-certainty evidence that Moderna is associated with higher incidence than Pfizer in 18-29 year olds and low-certainty ev…
DOI: 10.1136/bmj-2021-069445 -
Myocarditis in Adolescents (12-17 years) Associated with the Pfizer-BioNTech (BNT162b2) Vaccine: A Systematic Review and Meta-analysis
In a meta-analysis of four observational studies, BNT162b2-associated myocarditis in adolescents was rare, with male predominance and higher incidence after the second dose.
DOI: 10.47672/ajhmn.1387