And yes, the decision to yank J&J’s vaccine was the trigger.

Daily vaccinations were rapidly growing for every demographic under 50, and stable for 50-64, until the FDA’s “pause” (as though they could unclick that button and undo the damage).

They abruptly u-turn:

When you account for the fact so many older people have already been vaccinated, you can even see the impact in the 65+ groups.

The share of unvaccinated getting vaxd each day fell for *every adult age group*.

Another way to look at it is what daily vaccination rates were for each age group at any given point in the vaccination campaign.

Before the J&J pause, younger groups were *not* much less enthusiastic than older people to get vaccinated. Now, the trends are not good.

Is this ALL the FDA’s fault? There’s no way to prove the counterfactual.

But the different public reactions to how the UK and Europe handled the reports of AZ blood clots suggests that suspensions are in fact more harmful to public trust than warnings.
https://yougov.co.uk/topics/international/articles-reports/2021/03/22/europeans-now-see-astrazeneca-vaccine-unsafe-follo

Either way, the vaccination program was mortally wounded on April 13. If we don’t manage to revive it, we could miss the herd immunity threshold.

Existing trends have us ending up at 64% of the adult population vaccinated. Hopefully, with natural immunity, that will be enough.

But while Israel has gotten herd immunity with only 60% vaccinated, the adult vaccination rate is 85% — which goes up to 94%, with natural immunity from recovered cases (who are yet to be vaccinated).

If we go on as we’re going, we’ve got no shot at that. https://twitter.com/FT__Dan/status/1387160243633758208?s=20

In case you’re wondering: no, growth rates were not generally slowing for people under 50 before the pause. Modest declines in growth for 40s and 50-64, but 18-39 was still accelerating.

I should add:

Of course vaccinations were going to peak sometime, young people were going to have lower uptake, and not all the decline will be explained by disruption of J&J or by hesitancy. But it’d be very weird if this was just a coincidence.

This suggests that a huge chunk of the lost first doses (and missing growth) was J&J supply shock directly. This may be less hesitancy per se than J&J going to different people.

(It doesn’t show first doses for Moderna/Pfizer, so it’s hard to see the impact on them, up or down.)

Using CDC data on 1st doses and OWID for doses by manufacturer, I *think* you can subtract the two and get a rough idea of what's going on with 1st doses of mRNA.

It looks like they had plateaued before 4/13, but did also fall off. The bump before the decline was all J&J.

These data may rely on different reporting systems, because J&J tends to peak on different days than overall first doses.

(Weirdly, OWID has J&J peaking on weekends, while CDC has doses tanking then.)

@youyanggu’s projections suggest that we were heading for a peak in overall new vaccinations soon anyway.

But the pause may have moved up the peak and made the decline sharper than predicted, due to supply disruption and any knock-on effects. https://covid19-projections.com/path-to-herd-immunity/#comparison-projected-vs-actual

Here’s the comparison through 4/24 (last day with complete data).

We were already undershooting the model a bit, prior to J&J ramp up. Decline since the pause is a bit sooner/sharper than predicted. *If* gap is due to J&J, we’re missing 4-500k/day or 4.6m doses thru 4/24.

This *could* be explained by absence of J&J shots, which had been 400k/day when it was paused.

But real world is messy, models aren’t perfect, and the decline may have happened anyway.