I feel compelled to respond to @MattYglesias post "Give the vaccine to the elderly" in which he misinterprets ACIP objectives, pushes for simplicity over saving lives, and draws conclusions that aren't consistent with the evidence. Let me explain 1/13 https://www.slowboring.com/p/vaccinate-elderly

Yglesias conflates the 2 ACIP prioritization goals: (1) Prevention of morbidity/mortality & (2) Preservation of societal functioning. I admit having 2 goals is confusing. For now let's focus on Goal 1, leaving Goal 2 and considerations of equity aside. 3/13

The thrust of his argument for prioritizing based on age alone (after LTCF) is based on 2 principles: simplicity and preserving societal trust. He argues ACIP's "complexity itself is a problem for both for implementation and ethics" & allows too much gaming of the system. 4/13

We have wide agreement that age is a good metric for prioritization because it strongly influences COVID19 risk of hospitalization and death. Someone over 65 is 2.6 times more likely to be hospitalized with COVID19 than someone age 18-44. 5/13

But hospitalization and death are also strongly correlated with several underlying medical conditions. *Within* each age-sex-race category people with one underlying medical condition are 2.5 times as likely to be hospitalized than someone with no conditions. 6/13

Failing to *also* prioritize based on underlying medical conditions would lead to greater hospitalizations and more death. It would be in direct opposition to our stated goal of the prevention of morbidity and mortality. 7/13

The exact argument Yglesias makes for prioritizing the elderly - that they're the most vulnerable - also applies to prioritizing people with underlying medical conditions. Let's consider whether the complexity of prioritization is counterproductive. 8/13

We would need to vaccinate 2.5 condition-free people to prevent the same number of hospitalizations as vaccinating one person with an underlying medical condition (NNV). Can we more than double the pace of vaccinations simply by expanding eligibility? Almost certainly not. 9/13

We would also need implausibly high rates of line-skipping to make prioritization counterproductive and justify abandoning it. And in the coming weeks we will surely find better ways to adhere to the prioritization without slowing down the process. 10/13

@MattYglesias argues that using more than just age to prioritize vaccination is counterproductive because it introduces too much complexity and leaves too much room for gaming the system. But that view simply isn't supported by the evidence. 11/13

However, I love his simple principle of "The _____ are most vulnerable. Let's just take of them first." And it lines up well with ACIP's prioritization for Goal 1. Maybe his views and ACIP's prioritization do not diverge so much after all. 12/13

The solution is to build a vaccination program that delivers doses to high priority groups quickly & efficiently without wasting doses. It will give us the biggest reductions in morbidity & mortality. And it will help speed the process for the rest of us when the time comes.13/13