Heavily funded by taxpayers, how durable are the COVID-19 vaccines? With unprecedented legal mandates in countries like America, investigators from Sweden’s Umeå University sought out to determine the actual durability of leading COVID-19 vaccine products. While initial efficacy appears high–helping to defend people from severe COVID-19–mounting breakthrough infections and waning immunity are known to have triggered ongoing booster programs. How durable are these young products? With limited evidence and a lack of specificity by vaccine product, a trio of researchers from the northern region of Sweden capitalized on Swedish national population data to determine if the vaccine’s effectiveness lasted even six months. Overall, this study determines any benefit beyond six months isn’t certain—and appears to dissipate. And what could be considered a bombshell of a story, is that the Umeå -based authors determine that the Pfizer-BioNTech vaccine known as BNT162b2 or “Comirnaty,” the only vaccine formally approved in the United States, shows notable waning efficacy, particularly among men, the elderly, and people with comorbidities. By month six, the data indicates little to no effectiveness, raising serious concerns about the strategy behind the mass vaccination program now underway.
Swedish law mandates that all healthcare providers in this Nordic nation must report health-related data, including vaccination, to two registries, including the Swedish Vaccination Register as well as the SmiNet register, both managed by the Public Health Agency of Sweden. The law requires 100% coverage of the entire population of approximately 10.2 million.
Designing a retrospective cohort study, the Umeå University study team included Peter Nordstrom, Marcel Ballin, and Anna Nordstrom. Dr. Peter Nordstrom is a geriatric physician at the university’s Department of Community Medicine and Rehabilitation, Unit of Geriatric Medicine. Marcel Ballin completed his doctorate in geriatric medicine in the Department of Public Health and Clinical Medicine while Ann Nordstrom teaches public health at the university. The team secured approval by the Swedish Ethical Review Authority.
Utilizing a retrospective study design, the investigators utilized Swedish nationwide registries including the Swedish Vaccination Register and SmiNet Register (both from the Public Health Agency of Sweden) as well as Statistics Sweden (the national agency for statistics) slicing and dicing the data based on observational study inclusion criteria which included all individuals (N=3,640,421) vaccinated with at least one dose of any COVID-19 vaccine up to May 26, 2021. The following vaccine products are included in the analysis:
Looking at a total of 5,833,003 subjects, the population was selected based on vaccination status and COVID-19 infection up to October 4, 2021.
Vaccinated individuals were randomly sampled from the nation’s total population and matched on birth year, gender, and resident city/town/village. The Umeå University derived the study cohort from this broader population. They matched each inoculated person (equals two doses) 1:1 to one randomly sampled unvaccinated person based on year of birth, gender, as well as a baseline established with the date that the individual received their second COVID-19 vaccine dose in both vaccinated and unvaccinated populations.
Applying other exclusion criteria and repeating the matching procedure five times, the investigators narrowed the cohort to 842,974 matched pairs of vaccinated/unvaccinated subjects.
A second study cohort using less stringent matching criteria was formed for use in a “forthcoming sensitivity analysis.” The study authors matched each vaccinated person to the rest of the cohort based on age including a five-year buffer in age with each pair. Repeating this process ten times to support the matching of unvaccinated persons with several vaccinated persons leading to 1,983,315 per pair (N=3,966,630).
The team employed proportional hazards models with 95% confidence intervals (CI), as well as restricted cubic splines (four knots in default positions) to produce time-to-event for the outcomes (symptomatic infection/severe disease) based on vaccination status (vaccinated/unvaccinated).
Leveraging registry data plus national data from Statistics Sweden, the team executed Cox regression analyses to calculate hazard ratios (HR). They adjusted for the matched samples base on 95% CIs based on “standard errors by the VCE procedure and ROBUST option in Stata.
Results showed that the COVID-19 vaccines used in Sweden wane in effectiveness over time, corresponding with other observational study results. For example, the one approved vaccine in America, Pfizer’s BNT162b2 (“Comirnaty) starts off at 92% effectiveness yet by month number four (4) wanes in effectiveness to 47% (95% CI, 39-55, P<0·001). After month 6 the authors detected no effectiveness (23%; 95% CI, -2-41, P=0·07) and this obviously is a key rationale for booster programs in wealthy nations such as America, the UK, and Israel.
Following other studies, Moderna’s vaccine has shown to be a bit more durable as its mRNA-1273 vaccine effectiveness wanes less than the Pfizer-BioNTech vaccine does. For example, the team reports that the overall effectiveness waned slightly slower at 59% (95% CI, 18-79) by day 181 onwards. The Swedish study results aligned with others that AstraZeneca’s ChAd0x1 nCoV-19 afforded less protection while effectively waned faster with no effectiveness by month four (66%; 95% CI, 41-80). Overall vaccine effectiveness declined with male and the elderly.
What about protection against severe COVID-19? The aggregate of all three vaccines in Sweden vaccine product performance waned from 89% (95% CI, 82-93, P<0·001) at day 15-30 to 42% ((95% CI, -35-75, P=0·21) from day 181 (month 4) and beyond. Again, sensitivity analysis demonstrated even worse performance with males, the elderly, and immunocompromised.
Of course, the observational design includes inherent limitations. While the study team adjusted their approach to factor in cofounders the “possibility of residual and unmeasured confounding remains,” declared the authors. While certain exclusion criteria were applied (previous confirmed infection) probabilities are high that at least some individuals previously infected were included in this large national population analysis. It’s possible that if such individuals were part of the unvaccinated cohort, then natural immunity associated with previous SARS-CoV-2 infection could have weakened vaccine effectiveness estimates. Other limiting elements can be reviewed in the source.
Note, that this study hasn’t been peer-reviewed and should not be cited for medical evidence yet.
While the development of the COVID-19 vaccines marks a notable accomplishment in vaccine development history, and undoubtedly contributes to protection against severe disease and death during the pandemic, actual durability isn’t well studied. While COVID-19 ushered a life sciences boom that can benefit humanity for decades to come, unfolding events necessitate a frank, systematic, and comprehensive investigation into the true benefits of these version 1.0 vaccine products.
Mounting evidence in the form of observational studies from Israel, the U.S., and the U.K. to now a national population study from Sweden suggests significant challenges in actual durability of COVID-19 vaccines.
The product effectiveness wanes considerably within a few months. Vaccine performance depends on the actual underlying product as well as various demographic elements such as age, gender, comorbidities, and the like.
Accumulating data points amount to some challenges in the popular narrative about the overall effectiveness of mass vaccination as the only approach to eradicate SARS-CoV-2, the virus behind COVID-19.
TrialSite suggests an urgent need to better understand COVID-19 vaccine durability and overall effectiveness. With mandates and authoritarian leaning governing urges across multiple nations, an open and honest discussion about vaccine durability must be high on the list of national priorities.