Delaying That Second Dose Is Not Evidence-Based Medicine But It Still Makes Sense


by Raywat Deonandan

Mar 5, 2021

This article was published in The Ottawa Citizen on Mar 5, 2021.


The clinical trial data describing the safety and efficacy of the Pfizer and Moderna COVID vaccines were probably the most scrutinized science papers in history. They showed efficacy scores over 94% when a prime dose was followed by a booster 21 or 28 days later. Yet the National Advisory Committee on Immunization (NACI) now recommends extending that interval to up to 4 months, a substantial deviation from the manufacturers’ directions. This has upset many who feel that this departure is tantamount to experimentation without consent.

But we must remember the goal of our vaccination program. It is not necessarily to give individuals maximum immunity. Rather, it is to get society back to normal. We do so by getting sufficient immunity into the most people as quickly as possible.

The mRNA vaccines are remarkably immunogenic. According to the trial data, a sharp drop in disease incidence was seen in the vaccinated group, starting 10-14 days after the first dose, before receiving the second. At that point, efficacy was observed to be over 92%, which is a staggering number. After the second dose, the efficacy increased only by 2 or 3 percentage points. And real world data from BC showed that a first dose alone prevented 80% of COVID cases among long term care residents.

We therefore know that a single dose is quite capable of offering profound protection. An imprecise but useful way of thinking about the two doses is that the first one creates immunity, while the second dose makes that immunity last longer. So, two doses are still needed for immunity longevity.

However not everyone in the trials received their second dose exactly on schedule. Some delayed up to 42 days but showed no appreciable change in efficacy. This led to the tantalizing possibility that a second dose could be further delayed without compromising overall effectiveness.

NACI’s recommendation is, frankly, without strong evidence. We’ve only been giving these shots for 3 months, so it’s chronologically impossible to know what happens if we delay the booster by 4 months. That does not mean that it won’t work. Rather, it just means we don’t know what the consequences will be. Pursuing such a path is the opposite of adhering to evidence-based medicine, which is the practice of treating patients based solely on what is strongly known.

But public health is not evidence-based medicine. In a public health emergency, non-ideal evidence must sometimes be considered when trying to prevent imminent catastrophe. The evidence supporting mask usage and lockdowns was imperfect, but sufficient to garner broad support, as the need for immediate intervention was dire.

Similarly, we are facing a time-sensitive threat that justifies a less than slavish devotion to the trial findings. We are racing to achieve sufficient population immunity to prevent new hyper-transmissible variants from triggering a 3rd wave of COVID-19. In Ontario, test positivity data are showing a rise amongst those aged 10-29 years that will surely spread to the older demographic, unless we make the elderly immune through vaccination with scarce doses.

We have a small window of opportunity in which to do this. If we succeed, we might avoid the spike in deaths that heretofore has always followed a rise in cases. Maybe we can avoid that 3rd wave altogether.

There is insufficient vaccine supply to do so if second doses are held back in compliance with the original dosing schedule. Delaying the second dose accomplishes one very big thing: it doubles the number of people that we can vaccinate.

The risks are real. Will immunity wane appreciably before a second dose is finally available? Will fading immunity then allow for the evolution of even more dire variants? We don’t know.

So we weigh the magnitude of the imminent threat –new variants and a 3rd wave—against the theoretical threats of waning immunity and newer variants months down the road. By this calculus, the argument for delaying become more palatable.

But the need to enact such delays, and the fear of immunity waning before receiving a booster, can both be assuaged by securing vastly more vaccine doses sooner. Our leaders tinker with dosing schedules out of necessity brought on, not by science or disease, but by slow procurement.


Raywat Deonandan is an Epidemiologist and Associate Professor with the Faculty of Health Sciences at the University of Ottawa