With the announcement of a third successful COVID-19 vaccine trial from UK researchers at Oxford, candidates for a viable vaccine are multiplying rapidly. Media reports focused on the exceptionally high efficacy rates of the three forerunners have given all of us the good news we crave, tempered only by the reminder that the logistics of production and distribution will delay the effort of getting everyone vaccinated. Our review of the available Pfizer data, however, also raises some questions about the core research—issues which could apply to the trial from Moderna as well.

One factor that threatens to distort the accuracy of any COVID-19 vaccine trial is the uniquely confounding role of human behavior during this pandemic. Clearly most individuals who are willing to receive an investigational vaccine are also often more likely to take measures such as mask-wearing and social distancing. Although this applies to the placebo and vaccine recipients alike, one might reasonably conclude that the majority of study participants would experience less frequent and concentrated viral exposure than the statistically average person. Accordingly one could expect the 95% efficacy rate to shrink once the vaccine is administered to the public where a significant percentage of people are unwilling or unable to use NPI measures but may, nevertheless, eventually acquiesce to taking a vaccine.

Many of the studies into potential COVID-19 vaccines use antibody assays to directly assess the response to vaccination. In contrast, both Pfizer and Moderna decided to test effectiveness indirectly by administering either the vaccine or placebo followed by monitoring for symptoms and testing of symptomatic subjects. The research concluded with approximately 170 subjects in the Pfizer study who tested positive for COVID-19 being separated into a vaccine and a placebo group, yielding a efficacy rate of 95% in the experimental arm of the study.

We now know from extensive investigation into COVID-19 that there are many people who are asymptomatic but contagious. Moreover, vaccines often reduce the severity of the symptoms of disease. If only participants who report clinical manifestations are tested, there is a risk of missing subjects with mild or no symptoms—possibly due to the effects of the vaccine—who are nonetheless infectious. In a national vaccination program, this could lead to the dangerous situation of increases in the number of asymptomatic spreaders.

It is also necessary to recognize that the Pfizer study recruitment applied extensive exclusion criteria. Individuals with hypertension, diabetes, chronic lung disease, asthma, and immune disorders were all excluded, despite the fact that these are exactly the people most susceptible to the most serious complications of COVID-19, not to mention more likely to contract a virus. Those who smoke or vape and obese persons, the latter which represent more than 40% of the American population, were also excluded, as well as healthcare professionals and others whose occupation puts them at a higher risk. Healthcare professionals, subject to special circumstances (heavy exposure mitigated by PPE) not tested by the Pfizer study, are, incidentally, the first demographic slated to receive the vaccine.

Regarding vaccine efficacy among older patients, the Pfizer press releases and website information tout a 94% rate of effectiveness among subjects 65 years and older. As promising as this number may be, it is difficult to ascertain how many subjects weren actually included in this group. Pfizer boasts a whopping 41% international and 45% domestic percentage of elderly participants in their clinical trial, but these sizable figures were adjusted—ostensibly for optics—by including subjects as young as 56 years of age. Ultimately, how many of the study participants actually fell into the standard senior age bracket of 65+ is not immediately clear, calling into question the true statistical significance of the 94% efficacy rate.

Many have voiced apprehension that current vaccine research may be too accelerated to guarantee efficacy and safety. In the case of the Pfizer study, the goal posts for the study end date were moved all the way from the proposed date in June of 2021, mentioned in their governmental application, to last week. Not unlike the NIH study of the drug Remedesivir—recently shown ineffective against COVID-19 in a WHO study—the research parameters were altered to stop the trial when the target number of infected subjects was reached. This breakneck pace of vaccine development, a mad race to take credit for saving the world while reaping staggering financial reward for medicating most of humanity, increases the probability for research errors.

Desperate for some promising weapons to battle the pandemic, we all hope for the best with these novel vaccines. Nevertheless, it is vital that we cultivate realistic expectations to prepare the public for the possibility of disappointment and to ensure the continued use of the proven NPI precautions. Here, that familiar adage, expect the best, but prepare for the worst, may be the best advice.