Did covid vaccine studies use an active placebo (with side effects)?

Question

I know that the studies done by Pfizer, et al, used standard methods with a placebo control group. If you read about blinded experiments here, it mentions:

A common cause for unblinding is the presence of side effects (or effects) in the treatment group. In pharmacological trials, premature unblinding can be reduced with the use of an active placebo, which conceals treatment allocation by ensuring the presence of side effects in both groups.

As we know, the covid vaccines have some side effects. Anecdotally, I know many people who reported feeling a bit sick for a day after getting one. So my question is, did this "unblind" the participants in the covid vaccine trials? Or did they use a placebo that caused some level of irritation, in order to better hide this?

Answer 1

Of the three vaccine products that have been approved or had emergency use authorization in the United States, all used saline placebo in the clinical trial.

From the Pfizer Polack et al paper:

With the use of an interactive Web-based system, participants in the trial were randomly assigned in a 1:1 ratio to receive 30 μg of BNT162b2 (0.3 ml volume per dose) or saline placebo.

From the Moderna Baden et al paper:

Injections were given 28 days apart, in the same arm, in a volume of 0.5 ml containing 100 μg of mRNA-1273 or saline placebo.

From the Johnson & Johnson Sadoff et al paper:

Participants were randomly assigned in a 1:1 ratio, with the use of randomly permuted blocks, to receive either Ad26.COV2.S or saline placebo.

Thus, we should expect that a mild salt water solution would not elicit much of an immune response. This is a potential weakness of the trial design. Nonetheless, the results were sufficient to result in full approval of the Pfizer product.

Answer 2

The AstraZeneca COVID-19 vaccine approved by EU and UK regulatory authorities used a meningitis vaccine as the control in phase 2/3 trials.

From the methods section of their paper:

Participants were randomly assigned to receive either the ChAdOx1 nCoV-19 vaccine or the quadrivalent MenACWY protein-polysaccharide conjugate vaccine. MenACWY was used as a comparator vaccine rather than a saline placebo to maintain masking of participants who had local or systemic reactions.

Confidence intervals for side effects overlapped between the treatment and control groups (see the appendix in the above linked paper), though the study wasn't powered to detect moderate differences. It seems these investigators may have met their goal of avoiding unmasking due to side effects.

This doesn't guarantee that unmasking didn't occur for some other reason. I'd note that no masking assessment was reported, though it rarely is. It is possible to assess masking by, e.g., asking participants whether they think they received the treatment or control. Unmasking can occur for a variety of reasons, side effects are just one of them. For example, trials of interventions that are a bit more complex than a pill (e.g., injection, IV) often don't blind staff who administer the intervention and control, but keep the participants, investigators, and the people who collect the data blind. This comes with a risk of unmasking, but is balanced against the challenge (and likely failure) of making the intervention and treatment look exactly alike, push exactly alike, etc.