Covid, Contagion, and Looking Forward

I have Covid. This is the first time I’ve tested positive, though I’m pretty sure I had it in February/March 2020 after traveling in Europe.

Covid is becoming a universal experience, but it’s a cruddy disease, and the idea that we’re stuck with it—and its repeated bouts of damage, since immunity doesn’t last—feels dismal. 

It didn’t have to be this way. When Covid first surfaced, its R0 number—the number of susceptible people who will catch it from a single infected person, on average—was about 2.5. Although that’s much higher than the flu, which has an R0 of about 1.28¹, a disease with an R0 of 2.5 can be stopped. But the world’s response didn’t stop Covid, and within a year or so, the virus had begun its evolution into one of the world’s fastest-spreading diseases.

These numbers are kind of fuzzy, right?

Yes, rankings get complicated. And the data on Covid’s increasing contagiousness is depressing and often confusing. But looking away from depressing and confusing data does not make the underlying data better (though it may make you feel better in the short term!). So, let’s dive in to the fuzzy world of R0s and R-effs (we’ll get to what that is).

Measles is widely acknowledged as the world’s most contagious disease, based on its R0 number of 12 to 18. (Yep, there’s lots of uncertainty in the real world.) But measles is one-and-done: get it, recover from it, and you’re immune for life.

Covid doesn’t work like measles: we can catch it repeatedly as immunity wanes over months to years. But Covid is highly contagious. Is it measles contagious, on an R0 scale? Probably not—but R0 is mostly irrelevant at this point, since most of us have some Covid immunity, so we’re not 100% susceptible. What’s more relevant is a constantly-changing measure called the effective reproduction rate (r-eff). That’s the number of people who will catch a disease from a single infected person, given the levels of immunity in the community at the time. And immunity levels are always changing.

Covid just seems to keep going, though, despite all this imprecise immunity math.

Yes, what’s clear is the virus’ trajectory over time: its ability to spread keeps increasing. Original omicron (BA.1) was much more contagious than Delta, and the BA.5 strain was more contagious than original Omicron, and XBB1.5 was more contagious than that, and so on. But there are nuances; it’s not simple addition.

The R0 and r-eff don’t tell the whole story, either. There’s another component of spread: the incubation period. Measles has a 10- to 12-day incubation period, but Covid’s incubation period now averages 2-4 days (though it can be longer). I’ll quote Dr. Roby Bhattacharyya in an El Pais article from January 2022:

To demonstrate omicron’s infectious power, Bhattacharyya has done a back-of-the-envelope calculation to imagine what a race between omicron and measles, another of the world’s most-contagious viruses, would look like. One person with measles infects 15 others on average if none are vaccinated, compared to the six people infected by omicron. But the key lies in the so-called “generation time”: i.e. the number of days that elapse between when the first person is infectious and when those they infect also become infectious. With measles, that takes about 12 days. In the case of omicron, this only takes four to five days. “One case of measles would cause 15 cases within 12 days. One case of omicron would give rise to another six at four days, 36 cases at eight days and 216 after 12 days,” explained Bhattacharyya.²

Bhattacharyya used an R-number of 6 for original omicron (BA.1). The current EG.5 and XBB strains are more contagious than that original strain was. And even though population immunity has increased since then, Covid immunity isn’t an on-or-off light switch, but a spectrum of resistance. As another nuance, Covid is weirdly asymmetric: some people infect almost no one, or only household contacts, while a few people are super-shedders, according to this 2023 Nature study (you can read a longer preview here).

My head hurts thinking about this. But more importantly, what can we do?

No one knows when Covid will stop becoming more contagious, so all we can do in the meantime is further reduce our susceptibility. At a macro level, that can take the form of therapeutics, policy, or funding and investment.

On the therapeutics front, in addition to the tools we already have, one interesting development in China and India is the rollout of nasal vaccines. The idea is that by delivering the vaccine to the nasal mucosa—where Covid enters the body—immune cells would develop close to the scene of the crime (the infection location). It’s like standing up a neighborhood fire department rather than waiting for firefighters in the next county (antibodies and T-cells at distant body sites) to realize there’s a problem and get moving. For a deeply scientific explanation of this, Nature Medicine has a great article about mucosal immunity and mucosa-associated lymphoid tissue (MALT).

Medicines available to fight Covid have also improved vastly, and as the virus evolves, so will the medicines.

Okay, let’s talk policy.

On the policy front, one (somehow) controversial approach to reduce the threat from new diseases would be to ban gain-of-function research. To be clear, we don’t know how Covid arose, there is widespread disagreement, and I believe we will never know. But the pandemic made it crystal clear that novel viruses are disruptive, damaging, and unpredictable. This is true in every sense: physical, mental, emotional, societal, economic, political. Novel pathogens are not worth taking any chances with. None. And gain-of-function research takes those chances on the regular, whether or not an incident ever results.

The premise of gain-of-function research is, as I understand it, that by creating new variants of pathogens, we learn how they work so we could make therapeutics faster if a pathogen naturally evolved in a way previously explored in the lab. I think this doesn’t hold water. We are bringing risky pathogens into existence in order to stop other hypothetical risky pathogens? Taking on actual risk to prevent hypothetical risk? This list of biosecurity incidents alone argues this is a bad idea. As always, incentives play a role.

AI adds another wrinkle to the mix: it could make gain-of-function research much easier, but on the other hand, it will also make it much easier to identify and develop new therapeutics. My view is that we should use AI in a narrow, well-controlled, and heavily tested way to make new therapeutics, not new pathogens. Nothing seems hard about this conclusion. Banning gain-of-function research probably reduces a lot of risk right now, even more than in the past.

You mentioned therapeutics, policy, and funding and investment. How about that last one?

Sure. Instead of funding gain-of-function research, how about funding public health, which fell flat on its face during the pandemic because of years of underinvestment in the lead-up to Covid’s emergence. Shoring up public health might mean: 

• Strengthening information sharing among local, state, and federal public health departments.

• Cutting red tape for testing, reporting, and treatment. 

• Establishing rapid-response channels to take anomalous reports from doctors and act fast to investigate.

• Funding a broad effort to proactively build trust within communities. As the state of healthcare for many people worsens, local public health departments could offer free services, maybe even resurrect the idea of doctor visits for the ill or housebound like in ages past. But they’d need money to do this. 

We are not doomed to a world of failing health infrastructure and disintegrating trust. The pandemic made things worse, but it also showed us some ways to make things better: first, we should do everything we can to reduce the risk of new diseases emerging. (We are already not doing this with bird flu, but we can and should do better.) And if a new disease does emerge, nipping it in the bud with strong, high-trust, and community-cooperative public health is by far the best strategy, before it can get its evolutionary flywheel spinning. Building up that trusted, accessible local infrastructure, neighbor to neighbor, doctor to patient, should be a high priority.

1. Biggerstaff, Matthew, Simon Cauchemez, Carrie Reed, Manoj Gambhir, and Lyn Finelli. “Estimates of the reproduction number for seasonal, pandemic, and zoonotic influenza: a systematic review of the literature.” BMC Infectious Diseases, September 4, 2014. https://bmcinfectdis.biomedcentral.com/articles/10.1186/1471-2334-14-480

2. Ansede, Manual. “Omicron: The fastest-spreading virus in history.” El Pais, January 3, 2022. https://english.elpais.com/usa/2022-01-03/omicron-the-fastest-spreading-virus-in-history.html