Authorization for another 2 new vaccines, a potentially critical international licensing deal, and a new phase 3 trial: This month saw a…
When several European countries paused use of the Oxford/AstraZeneca Covid vaccine in March 2021, the criticism from many commentators was swift, extreme, and, above all, certain. “A magnificent example of European failure.” That was one of the responses I noted in my commentary at The Atlantic, along with the others in my cartoon – stupid! harmful! potentially fatal! By the end of the month, 12 European countries, as well as Indonesia and Malaysia, had temporarily suspended the vaccine. When it became clear how rare the syndrome was, most resumed use of the vaccine.
It happened again in April, with a second of the adenoviral vaccines – the single-shot Johnson and Johnson vax. This time, it was the US and South Africa that paused the rollout of the vaccine for a short time. Again, predictions of what this would mean seemed to me quite dire.
Most of those countries had other vaccines to rely on, so I didn’t think the decisions would be disastrous for them. There are real differences in how risk-averse to vaccine-related events communities are, and I thought drug regulators being less risk-averse than their communities would be more of a longterm threat than being cautious, or even over-cautious. But what actually happened?
Enough time has passed for some study results to roll in. I found 12 relevant studies, one of which tries to work out if there was a difference in vaccine usage as a result. That one’s for the AZ vax. The rest look at vaccine hesitancy and people’s opinions of the vaccines. One of those is an in-depth qualitative study (for the AZ vax). The other 10 are surveys or polls, 5 each for the AZ and J&J vaxes. They’re all in Europe and/or the US, except for the qualitative study, which was done in Australia. Even though the view you get from these studies is quite narrow, in time, geography, and potential impacts, I think there are a few (tentative) answers in there.
First up, what about the main disaster the critics predicted would befall the countries that paused the vaccines, and perhaps others as well? Did the temporary suspensions increase hesitancy for Covid vaccination, regardless of vaccine, and so lead to reduced levels of vaccination? Apparently not much, if at all.
The suspensions had a definite impact on many people’s views of the suspended vaccines, but the surveys don’t show spill-over safety concerns for other vaccines. A pause may have strengthened the hesitancy of some of those who were already hesitant, though. However, we don’t know if that ultimately changed decisions to vaccinate. (The rate of hesitancy in surveys has fluctuated over time – see for example, in the US.)
Did the suspensions slow down Covid vaccination programs? It doesn’t seem to have made much difference in Europe. It seems unlikely to have made a major difference in the US, where vaccination appeared to be slowing down around then anyway, with over half the population vaxed. It would have had more of an impact, though, in South Africa, where there weren’t other vaccines to fall back on. The vaccination program was delayed for 2 weeks. More on that below.
I think it’s hard to draw a lot more from this batch of studies. I don’t think there was enough study of trust in regulators to draw a conclusion. A dozen studies may sound like a lot, but working out what the impact of these vaccine suspensions might have been is extremely complex. There may be more studies than I found, too – my dig was pretty shallow, and I only looked for studies in English. (All 12 are listed below this post.)
The pauses themselves aren’t the only issues affecting use of, or attitude towards, those 2 vaccines. There was the question of their relative effectiveness as well. Both vaccines were less effective than mRNA and protein subunit vaccines being rolled out in some parts of the world, and more effective than several inactivated vaccines. And both AZ and J&J had major supply problems happening when the countries hit the pause button.
The J&J pause because of adverse events was also tangled up with the safety concerns at a Baltimore factory, starting in March. And as I mentioned earlier, it occurred around the same time that the US appeared to be heading to a vaccination plateau, with the most enthusiastic people already vaccinated (well over half of all adults).
How did the single study of vaccine usage try to look if there was a signal in all that noise?
The authors are Vageesh Jain and Paula Lorgelly, from University College London (Jain 2022). They did a difference-in-differences study – approaching the pausing and non-pausing of the AZ vaccine across Europe as a natural experiment, with the non-pausing countries acting as a comparison group. They used data for 19 countries in the European Economic Area that didn’t suspend the vaccine and 9 countries that did.
Their analyses took a large variety of factors into account, including demographic, socioeconomic, and health system factors, as well as rates of Covid infections and deaths, and vaccine supply. However, the data on vaccine supply couldn’t show when expected vaccines hadn’t actually been supplied on time. A few of the non-pausing countries restricted the age range for getting that particular vaccine, but the authors’ analysis suggest that didn’t make a difference. The result? A very small decrease in vaccination in the weeks after the pause (less than 1%), but it wasn’t a statistically significant difference.
I think the question of short-term impact is clearer in South Africa, though I didn’t find a study analyzing the consequences. The lack of access to vaccines there meant the pause in April hit hard. Well over 70 million doses would have been needed to vaccinate all adults in the country. In January 2021, the government had bought 1.5 million doses of AZ, and were in line for 12 million doses from Covax – but only 1.4 million Covax doses had arrived even by the end of June. In mid-February they’d decided not to roll AZ out because of concerns about adequate effectiveness against the circulating variant. So they pivoted to J&J for vaccinating frontline healthcare workers, but paused when the CDC called for a pause in the US. That 2-week pause happened when South Africa’s about-to-be third-largest Covid wave so far was starting.
The major disaster globally was the lack of access to vaccines, not the pauses. Fortunately, the syndrome that triggered them turned out to be very rare, and we learned what to look for, as well as risk-minimization strategies. The pressure of the pauses may have propelled regulators into getting those answers more quickly, and to that information getting very widespread attention. One of the messages from the authors of the qualitative study in this group, Samantha Carlson and colleagues, seems critical to me (Carlson 2021). Faced with concerns about vaccine safety, we need “to communicate in such a way that addresses people’s concerns, and leaves little opportunity for other concerns to sprout and grow.”
Disclosures: My interest in Covid-19 vaccine trials is as a person worried about the virus, as my son is immunocompromised: I have no financial or professional interest in the vaccines. At the time Covid vaccination was available to me, the AZ vaccine was the only option for my age group in Australia (I had 2 doses and later, 2 boosts with a different vaccine). I have worked for an institute of the NIH in the past, but not the one working on vaccines (NIAID). More about me.
List of studies
Jain 2022 (AZ)
Carlson 2022 (AZ)
Surveys and polls:
Agosti 2022 (AZ)
Comerford 2022 (AZ)
Frankovic 2021 (J&J) (YouGov)
Hamel 2021 (J&J) (Kaiser Family Foundation)
Hsieh 2021 (J&J) (Census Bureau’s Household Pulse Survey)
Petersen 2022 (AZ)
Salmon 2022 (J&J)
Smith 2021 (AZ) (YouGov)
Sønderskov 2021 (AZ)