Choosing a Covid Vaccine

So many COVID-19 vaccines, so much confusion. 

First off, let’s all acknowledge that you, individually, are unlikely to be able to choose which vaccine you can have. Most places, at any given timepoint, will only have access to one.

That being said, there is a constant stream of news about the various vaccines that is not being contextualized for people. So this post is the help you understand what’s not being said about how we should choose vaccines. 

Vaccines Are a Public Sector Purchase

What makes vaccines different from other massive purchases governments make? Literally nothing. Vaccines are just a medicine. And medicines are just products. When you want to think about any product purchased for a lot of people - school uniforms, software for employees, bike/scooter share services - you consider the same three things

1. All in cost - purchase prices + maintenance + insurance + other

2. Efficacy - success at doing the thing you want

3. Feasibility - ability to play with existing & future environment

A manual bike share might make the most sense in a temperate flat city, but electric bikes might be needed in a hilly city with the ability to pick up bikes, and e-scooters might be better in a hilly city with more money but that has less bike parking. No one solution is best for everyone, especially because each city values things differently. 

All in Cost

There’s a sticker price for each vaccine (BioNTech = $19.50, Moderna = $25-$37) but that number doesn’t take into account a lot of other cost. How much does it cost to ship the vials from the manufacturing location? How much are special freezers to hold the vials once they arrive? And what is the labor & administrative cost to administer 2 separate doses a month apart?

At $10 per dose, the J&J single-shot vaccine is not half of the cost of BioNTech, it’s potentially a tenth of the cost; or less. And there is a finite amount of money governments can spend. Opting for a vaccine that is twice as expensive could mean that only half as many people get it. Or, perhaps more likely, that governments will have to forgo infrastructure, education, or other critical expenses in the near future. No one is talking about this very real tradeoff enough. 

Efficacy

I specifically defined efficacy as “success at doing the thing you want” because every product has multiple features on which it can be evaluated. Back to our rideshare analogy, “what you want” could be “complete shopping trips” or “help people commute to work” or “work for a long time without needing repair” or something else. Different solutions will be the best depending on which goal you focus on. 

So what are we trying to achieve with vaccines?

1. Get people back to normal lives as fast as possible?

2. Prevent infections from the current or future variants?

3. Allow infection but limit severe illness? 

4. Limit hospitalizations?

5. Limit deaths?

It matters! And we should be explicit about the goal before governments start shopping. 

Here’s what we know as of publication date

BioNTech, Moderna, AstraZeneca, J&J Janssen, Sputnik V data

Since March 2020, our ostensible goal has been to “flatten the curve” and “not overwhelm hospitals”. Plus, arguably most people are scared of hospitalization and death, not illness per se. However, many governments, especially in light of the growing number of coronavirus variants, are now focused on preventing any kind of infection. 

If preventing  any and all infection is your goal, BioNTech is probably the way to go. (Although Feasibility, discussed below, should be a part of that discussion) If you are trying to keep people alive, it perhaps doesn't matter which vaccine you choose.

Lastly, it’s important to consider baseline risk. In someone young and healthy with a baseline risk of 5%, reducing that risk by 75% is the same as a 95% reduction in someone whose baseline risk was 20%. Both end up at 1% risk. 

Feasibility

Feasibility is a partial result of cost, and a determiner of efficacy, but it’s so important, I want to call it out separately. If a product is cheaper to buy, more cash is freed up to use to solve problems when they occur. In that way, lower cost can increase feasibility. On the other hand, you could have the perfect solution (high efficacy) but if you can’t get it to the people who need it, (feasibility) your efficacy in practice is zero. 

Imagine the perfect dress...that’s in someone else’s closet. Or a breakthrough drug... that we can’t actually make. Feasibility is everything. 

While the mRNA vaccines are awesome in their protection, I’m seriously concerned that as we move farther from dense urban settings to rural populations, we will not be able to store the BioNTech vials or get people to drive hours for a second dose. Plus, 2 doses one month apart is actually like conducting 600 million vaccinations instead of 300 million. That is not a trivial logistical difference. If people only get one dose because they can’t drive to a center twice, they are actually worse off than getting the J&J vaccine which is intended at 1 dose. 

Practical challenges to implementation matter. A lot. And it’s usually where everything goes wrong. People plan for the best case scenario, and actually get the normal case scenario. 

Mix & Match

I guess you can tell that I’m quite bullish on J&J’s vaccine, whereas the news articles have focused on the 66% infection prevention. And no, it’s not Jersey pride. It’s possibly because I estimate my personal risk from the coronavirus as quite low, and I’m terrified of needles. I don’t want two shots when I can get one. 

But beyond that, for most governments, the optimal outcomes are achieved by limiting hospitalizations & deaths (both of which J&J achieve) and paying way less per vaccine. This frees up cash to do other things, like educate children, which we have woefully failed to do for the last year. 

Even in “rich” countries, it probably makes sense to use the mRNA vaccines only on the high risk population, essential workers, or perhaps when people live less than an hour from a distribution site. After that, I think we should all move to J&J. It seems counterintuitive, but makes more cents.