Paul Romer (@paulmromer): Two NYT pieces assess the future course of pandemic

One is incoherent, preachy, incomplete, and reassuring without justification.

The other is coherent, clear, and suggests concern.

Guess which is by med/health professionals?

Sam Lord / Samuel J. Lord (@samjlord): @paulmromer Besides the headline, I didn’t think the piece was “reassuring.” They seem to both give similar core arguments: our current vaccination approach is insufficient.

Paul Romer (@paulmromer): @samjlord See this sentence:

"More important, we may not need to achieve that goal [herd immunity] in order to escape the pandemic."

Sam Lord / Samuel J. Lord (@samjlord): @paulmromer I suppose it depends on what “escape” means here. I read it to mean “go back to life as normal without the large number of deaths we experienced before the vaccine.”

Paul Romer (@paulmromer): @samjlord I do not understand.

If we return to life as we lived it in 2019, by definition being below the herd immunity threshold means that we will face exponential growth in the number of cases, hospitalizations, and deaths.

Sam Lord / Samuel J. Lord (@samjlord): @paulmromer 60% of the US adult population and 85% of the elderly—those most vulnerable to covid—have had at least 1 dose already, so subsequent outbreaks will likely not kill as many people. Future outbreaks may be no worse than flu. Which is why we need to vaccinate the world.

Paul Romer (@paulmromer): @samjlord If the average age of the people who get infected is lower, the infection fatality rate will be lower.

But with exponential growth in cases, you are still going to see exponential growth in deaths.

With > 150 million susceptibles, many will die. Hospitals will be overwhelmed.

Sam Lord / Samuel J. Lord (@samjlord): @paulmromer Good points. I see now how the piece has flaws.

Still, I think a reach herd immunity by vaccinating every single last reluctant American is impossible and counterproductive, because humans in other countries are begging for the vaccine.

@zeynep's piece is excellent.

Paul Romer (@paulmromer): @samjlord @zeynep There are 45 million US citizens over age 65. If we treat 1 vax as full immunity, there are 0.15 * 45 => 7 million susceptible.

If we use the age specific IFR from Levin et al, => 400K deaths if all of them get infected.

And IFR for 40-65 yr olds is not trivial. But you see.

Paul Romer (@paulmromer): @samjlord @zeynep Levin et al paper:

Assessing the Age Specificity of Infection Fatality Rates for COVID-19: Systematic Review, Meta-Analysis, and Public Policy Implications, by ...

Paul Romer (@paulmromer): @samjlord @zeynep by Andrew T. Levin, William P. Hanage, Nana Owusu-Boaitey, Kensington B. Cochran, Seamus P. Walsh, and Gideon Meyerowitz-Katz, October 2020.

Bill Hanage (@BillHanage): @paulmromer @samjlord @zeynep As one of the authors - what are you picking as the age there for IFR? It increases so rapidly that you can’t treat >65 as a single cohort, and the numbers and vaccination rates in each age group end up mattering

Paul Romer (@paulmromer): @BillHanage @samjlord @zeynep Yes.

I used your formula (capped at the value for age = 85) and combined it with data on the population by age that I estimated by using a cubic spline that replicated the published numbers by age for 5 year groups.

Bill Hanage (@BillHanage): @paulmromer @samjlord @zeynep Sorry if I missed this, but did you include varying vaccination rates by age?

Paul Romer (@paulmromer): @BillHanage @samjlord @zeynep I relied on this from @EricTopol for the estimate of 85% over age 65 w at least one dose.

I don't have more detailed data by age, but I doubt that % with two doses exceeds 85% at older ages. So these two sources of error probably offset each other.

Buzz Hollander MD (@buzzhollandermd): @paulmromer @BillHanage @samjlord @zeynep @EricTopol I honor your cubic splines - but I'll observe that infection-acquired immunity may be high among vaccine-eschewing seniors; & that herd immunity dynamics would benefit that cohort at some point well before 100% infection. But agree... this ain't over! May is just a good month.

Paul Romer (@paulmromer): @buzzhollandermd @BillHanage @samjlord @zeynep @EricTopol Sure.

There are lots of ways to refine a back-of-the-envelope calculation.

The CDC estimates for the rate of infection during 2020 among people age > 65 is 17%. So taking account of this, you might get the number of potential deaths among ages > 65 down from 400k to say 300K.

Paul Romer (@paulmromer): @buzzhollandermd @BillHanage @samjlord @zeynep @EricTopol A back-of-the-envelope calculation tells you the order of magnitude. In this case, we are talking about hundreds of thousands of potential deaths.

At this scale, I don't understand what it means to say that without reaching herd immunity we can "escape the pandemic."

Buzz Hollander MD (@buzzhollandermd): @paulmromer @BillHanage @samjlord @zeynep @EricTopol My envelope might be less pessimistic than yours; but I grant there’s also the <65s to worry about (20% of deaths); and I do think the vulnerable population you describe is not going to fare as well w/o the favorable May/June seasonality which also hid some policy warts in 2020.

Paul Romer (@paulmromer): @buzzhollandermd @BillHanage @samjlord @zeynep @EricTopol My point is not that we should be more or less pessimistic. It is that I have no idea what the authors meant when they wrote:

"More important, we may not need to achieve that goal [herd immunity] in order to escape the pandemic."