"An abundance of caution" or a dearth of courage?
California trots out the pandemic's worst cliche to justify a dreadful decision
Note: California reversed this decision on January 21 after a 4 day pause.
The pandemic has produced many maddening moments. The latest came Sunday when California halted administration of more than 330,000 Moderna vaccines across nearly 300 sites following six non-lethal allergic reactions at a single site. The instructions sent to providers explained that the pause represented “an extreme abundance of caution” before helpfully adding “There are not immediate replacement doses during the pause.”
The decision to halt distribution of that batch is indefensible. My back-of-the-envelope math estimates that it will cost ~10 lives per day as long as it lasts. If the vaccine is ultimately destroyed or expires, it will cost hundreds of lives — more on that later.
It’s also not news that the vaccine can cause allergic reactions in patients. If there’s real evidence that this batch of vaccines is more dangerous to patients, public health officials haven’t shared it with us. So why throw the brakes on administering this batch?
The giveaway is the now-ubiquitous phrase “an abundance of caution,” which serves to preclude meaningful cost-benefit analysis. The speaker simultaneously admits overreaction and signals that discussing tradeoffs would be distasteful. That’s fine when, say, explaining why you arrived at the airport two hours before your flight. But when public officials invoke the phrase, it feels like an abdication of responsibility.
Admittedly, the pandemic hasn’t always made it easy to run a cost-benefit analysis. There’s always uncertainty about the impact of any given policy. The more fundamental challenge is the incommensurable nature of the outcomes involved. How do you weigh a dozen young adults engaging in suicidal ideation against a 77-year-old woman dying of Covid?
But the bureaucratic fondness for extremely abundant caution seems to be less about uncertainty and more about ass-covering. Nobody will be fired for the extra deaths that will occur as a result of this pause — for they’ll be the sort of diffuse, statistical deaths that are hard to link to this decision. For the officials involved, the safer path is to pause distribution — no matter how small the actual risk from this batch of vaccines.
To be clear, I’m not exaggerating when I say that this could lead to hundreds of deaths. Let’s lay out some facts and assumptions so we can run the numbers.
California has averaged more than 500 Covid deaths per day this January. In a state with 40 million residents, that represents 1 death per 80,000 people per day. Note that the trend is upwards, not downwards, but we’ll assume it remains constant for the next month.
Californians over 65 are 15.7% of the population but 74.8% of deaths. If that ratio has held in January, then the daily death rate in that group is about 1 out of every 16,500.
The vaccines would all be given to people in the over-65 bucket.
California’s vaccination sites have far more capacity than doses, so “pausing” distribution of this batch simply means that fewer doses will be given out until this pause is lifted.
When the pause is lifted, the doses will be administered on the same schedule they originally would have been administered on but shifted by the length of the pause.
With two doses per person, the 330,000 doses amount to 165,000 fully vaccinated people.
After being fully vaccinated, nobody dies of Covid. (This may be slightly too optimistic, but note that the widely quoted 95% efficacy refers to cases prevented, not deaths.)
With these assumptions, we see that each day of delay in distributing these vaccines leads to 165,000 * 1/16,500 = 10 excess deaths per day. The pause has already lasted 3 days, so that’s 30 excess deaths so far (although those deaths will occur in the future.) If the pause lasts a month, we would see 300 excess deaths. If it lasts longer (or more to the point if the vaccine doses are ultimately destroyed) the impact will be even greater than 300 deaths but modeling beyond the immediate future requires more assumptions.
Of course, these figures probably underestimate the impact since they only focus on deaths prevented within the 165,000 direct recipients of the vaccine. But vaccination indirectly benefits those that come into contact with the vaccinated; that’s the entire point of herd immunity. It seems plausible that the true impact is hundreds of deaths higher than my estimate above.
Maybe I’m wrong. Perhaps California can shelve a third of a million doses without slowing distribution this week (although that would be a damning indictment of everything else about their vaccine program.) Or perhaps a half dozen non-lethal reactions at a single site really do signal a risk that outweighs hundreds of deaths. If so, the California Department of Public Health should find a modicum of courage and make those arguments. For now, though, they’re content to hide behind “an extreme abundance of caution.”