HIV and ID Observations

Pufferfish, from United States Exploring Expedition (1838-1842)

The future ain’t what it used to be, said one very wise man.

He might have also said, It’s difficult to make predictions, especially about the future, but alas we’ll have to credit that profundity to someone else.

Still, both these statements embody the insurmountable difficulty of making accurate predictions — a problem starkly evident during pandemic times. How many times have we watched people give conflicting views of when, or how, this thing is going to play out, even in the short term? 

Two recent examples come to mind. Dr. Mike Osterholm, Director of the Center for Infectious Disease Research and Policy at the University of Minnesota, repeatedly warns of an additional surge this spring as more contagious variants take hold. This headline calls him “Dr Doom.”

On the flip side, Dr. Marty Makary from Johns Hopkins wrote in mid-February that we’ll likely be mostly done with COVID-19 in April — which, if I’m doing the math, starts 11 days from now. They can’t both be right.

Our ability to make accurate predictions is highly flawed, dependent on innumerable forces we can only begin to understand. 

We base these predictions on our background, our education, our knowledge base, plus an unconscious force that sends them in various directions — optimistic or pessimistic, confident or timid, contrary or mainstream.

The bold ones get the most attention, especially if backed by impressive credentials. If Larry Summers says we’re heading into ruinous economic territory with the stimulus package, who am I to question him? Or Janet Yellen, who predicts the exact opposite?

Enter Zeynep Tufekci — sociologist, computer programmer, and Associate Professor at University of North Carolina. You might expect an epidemiologist, or infectious diseases specialist, or virologist to have the best record in laying out the most likely way forward as COVID-19 continues its march around the globe, now 15 months in. But again and again I have found hers to be among the most logical voices, mostly in pieces published in The Atlantic and The New York Times.

And importantly, I’m far from the only one to hold this view.

Is it her diverse educational and vocational background? The fact that she’s a true “citizen of the world,” having lived in multiple places? That she works really, really hard to get things right? That’s she’s also wicked smart, to coin the Bostonian phrase to describe the smartest person in the room?

Probably all of the above. 

Zeynep kindly joined me recently on this Open Forum Infectious Diseases podcast to discuss how she ended up in her interesting current position, and her approach to COVID-19 — how we missed the mark for well over a month on the seriousness of the problem, our missteps on masks, the continued penchant for “beach scolding,” how we undersell the vaccines, and the general timidity of the biomedical community in questioning authority.

And yes, she finishes by speculating how this might end. 

Highly recommended.

Transcript here. Also available on Spotify, Apple Podcasts, etc.

Feb 23, 2020.

On March 7, 2020, right before CROI here in Boston, a bunch of us ID types planned to get together for a pre-conference dinner. A mixture of Bostonians and out-of-towners who hadn’t seen each other for a while. A chance to catch up before our busiest (and most important) scientific meeting.

What happened?

One person landed in Boston, and promptly got the next flight back to Los Angeles.

Another canceled his flight and never left home. A shame, too, since he was bringing his new partner (whom I still haven’t met). I hear she’s really fun and interesting.

One had to stay home due to a no-travel ruling issued for all his faculty. (Those edicts had not yet come from Harvard, but would a week later.)

A fourth had arrived early for the conference, but was so busy setting up the now “virtual” CROI that dinner for her was out of the question.

And of course, attendees from Europe (especially) already had pulled out of the conference en masse.

You get the idea.

Our numbers for dinner were substantially down, but we still assembled with (mostly) locals. Indoors. Windows closed. It’s winter here in early March. Plenty of hand-washing and, befitting the fomite-centric perspective at the time about SARS-CoV-2 transmission, extra use of an alcohol-based hand sanitizer — something I’d never purchased before.

That evening was filled with lots of discussion about the Biogen conference. About China, Italy, Spain, and Iran, and what was undoubtedly coming our way soon. About our frustration with the lack of COVID-19 testing — we ID doctors spent all of February complaining about that, with no answers coming from anywhere. Just painful silence.

And, yes, lots of anxiety.

Now, as CROI starts up again — virtually, of course — I think back to that night. A dinner indoors with friends and colleagues, possible a year ago today but unimaginable since, could now again happen in our (fully vaccinated) future.

Not saying I know when it will happen — just that it could. Hope.

Wow, it’s been a memorable year.

We don’t know. That part is easy.

Also easy is that case numbers really are falling — it’s not just reduced testing — and it’s happening pretty much everywhere.

Urban areas and rural. Red states and blue. Places with broad vaccine rollouts and those with hardly any. North and South America, Europe, Africa, and Asia. Even countries with the B.1.1.7 variant.

Look:

Source: 91-divoc.com/pages/covid-visualization/, downloaded 21 Feb 2021.

Let’s round up some theories:

1. Seasonality. An attractive hypothesis — coronavirus infections pre-SARS-CoV-2 definitely show a seasonal pattern.

And various viral diseases go through communities synchronized with the seasons, especially when school starts or the weather gets colder. Any pediatrician will tell you that.

Note that the term “seasonality” has always been a bit misleading — it refers to infections peaking within seasons, not throughout them. Think of influenza, how sometimes we have an early, sometimes a late seasonal peak in the winter.

The problem with this seasonality theory is that the seasons are flipped in the southern hemisphere. And didn’t cases surge over the summer in many southern U.S. states?

2. Herd immunity. Nearly 28 million Americans have had a confirmed COVID-19 diagnosis reported to the CDC. This represents only a fraction of the true cases — especially the mild or asymptomatic ones — and the CDC estimates that only 1 in 4.6 infections are reported. That could bring us up to half the US population with some degree of natural immunity to infection.

Even as of mid-January, the CDC put the actual case numbers at over 80 million, and certainly it’s higher than that now. And note that in some regions, the actual case counts might be even higher — 5 to 20 times higher, according to one recent publication.

3. Behavior. We know much better now how this virus is transmitted. Avoiding crowds and indoor spaces with poor ventilation — and wearing masks — reduce the risk. But has our behavior actually followed suit?

The holidays are behind us. The Super Bowl was lousy. Not many parties for the Australian Open tennis finals. Spring Break hasn’t happened yet.

One compelling hypothesis, related to herd immunity, is that the people least likely to follow infection control advice — or unable to follow it based on work or living situation — already have had COVID-19 and hence are immune.

The others, not yet infected, watched cases surge in December and January and continue to hunker down and stay safe — or again, have the luxury of staying safe. They might be especially vigilant now that a vaccine is in their not-too-distant future — you know, the pot of gold at the end of the rainbow, the light at the end of the tunnel, or the Holy Grail at the end of the Monty Python movie.

(Sorry about that.)

4. Vaccines. The world is vaccinating like crazy. Demand is off the charts. And in most places, we’re targeting the people most likely to have symptomatic or severe disease.

Plus, the data increasingly suggest that the vaccines reduce not just disease, but also the likelihood of transmission — they reduce infections overall (uninfected people can’t transmit), and those with infection have lower viral loads. 

While the vaccine rollout is not yet broad enough to explain the case number drop on its own, it might be contributing. It certainly could be playing a role in Israel.

5. The virus. Maybe the virus is doing us a favor and becoming less virulent over time. Perhaps some of these variants — if not B.1.1.7 — in order to gain the ability to transmit, also cause less severe disease.

Take the virus’s perspective — yes, think like a virus — and how this would be evolutionarily beneficial. More mild cases, more chance to spread its genetic material to other susceptible hosts. That’s all viruses care about, right?

6. It’s a gemish. This brings us to the most likely explanation for the drop in cases, a gemish — Yiddish for a mixture of things. (It’s pronounced “ga-mish”, in case you want to try it out on your own.)

It could be all of the above explanations, in various proportions, and different in various regions — plus things no one has considered.

And the uncertainty about why cases are dropping again hearkens back to this great H.L Mencken quotation, which over time has morphed into this profound statement:

Every complex problem has a solution which is simple, direct, plausible — and wrong.

I stress the importance of being humble about not knowing why the cases are dropping simply because reliance on one of these factors over another could get us into trouble. For example, this week Dr. Marty Makary, writing in the Wall Street Journal, posited that we are already close to herd immunity, making this bold prediction:

There is reason to think the country is racing toward an extremely low level of infection. As more people have been infected, most of whom have mild or no symptoms, there are fewer Americans left to be infected. At the current trajectory, I expect Covid will be mostly gone by April, allowing Americans to resume normal life.

Warning — if anyone tells you with confidence that they know precisely why cases are dropping, and that they have an accurate crystal ball showing that by April we’ll be safely out of this pandemic — please view it with the appropriate scientific skepticism it deserves.

Look, we can hope this optimistic prediction is correct — we all want that. April isn’t far away, we’ll know soon.

But if there’s one thing a pandemic from a new human disease teaches us, it’s that there’s a lot we don’t know.