COVID19: Is A Second Wave Inevitable?
by Raywat Deonandan, PhD
Epidemiologist & Associate Professor
University of Ottawa
(I add my credentials to these COVID-19 blog posts in case they get shared. I want readers to know that my opinion is supposedly an educated and informed one)
Aside from questions about safe school openings or the likelihood of a vaccine coming to market soon, the most common question I get these days has to do with the fabled “second wave.” In fact, I’ll be on a UK chat show in a few hours talking about this very thing. So I felt that now was a good time to write down my thoughts on the matter, lest my mind go completely blank as soon as that camera goes on.
As with all of these COVID posts, if you identify an error, please let me know in the comments below. I especially welcome scientists with relevant expertise to educate me if I’ve wandered too far off the compound.
What Are Epidemic Waves?
There really isn’t a formal definition. You know it when you see it. Incident cases increase over time, peak, then decrease over time. And after a lull of indeterminate time, the incidence rate rises again, peaks again, then declines once more. I just described two waves, and I used the incident cases to define the curve. But I just as easily could have used deaths caused by the disease, or even the proportion of all deaths in the population that were caused by that disease.
But there isn’t a universal and perfect metric for determining when the first wave is over and the second has begun. Nature doesn’t give us perfectly smooth curves, but rather jagged spikes and a whole lot of outliers. So, frankly, you don’t typically know when the first wave is over until it’s either well over, or until the next wave is clearly underway.
It ain’t very scientific. It’s just a convenient way to talk about general trends. And some infectious diseases researchers I know outright refuse to talk about waves. Dr Eric Toner at Johns Hopkins was quoted as saying, “When you’re underwater, it’s hard to tell how many waves are passing over your head.”
Do Pandemics Always Come in Waves?
Short answer: not necessarily, but often. Like, really very often.
Historically, many of the most famous respiratory disease pandemics did indeed come in waves. We only really started talking about the wave nature of these diseases after the Spanish Flu of 1918. That monster came in three waves: Spring of 1918, Fall of 1918, and the Winter of 1919. Canadian Epidemiologist (and a man I admire greatly for his scientific and moral leadership during this crisis) Dr David Fisman tweeted this chart of epidemic curves for four influenza pandemics, clearly showing that they presented in waves of mortality:
Even the MERS virus of 2012, which came and went fairly quickly, showed something resembling a couple of waves:
I’ve heard the first SARS virus of 2002-2003 described as a wave-less pandemic, probably because it’s more symptomatic nature (compared to COVID) allowed for good early identification and contact tracing, which in turn allowed public health workers to quash it before it got good traction in the human population.
But, as you can see below, its epidemic curve definitely shows some wave-like behaviour, no?
So, based on the behaviour of past viruses with similar characteristics, it seems likely that the COVID virus will also present to us in waves.
Why Do They Come In Waves?
There are two possible answers to this question: (i) there is something intrinsic to the virus that prefers and eschews certain seasons to others; and/or (ii) there is nothing special about the virus, and this is all about human behaviour.
My area of expertise is not in infectious diseases, so I can’t say this definitively. But when it comes to influenza and coronaviruses, it seems that it’s pretty much always the second case: that human behaviour drives the wave form. These viruses see cases typically crest in the cold winter months and recede as the warm weather returns.
I liken the role of behaviour to the predator-prey model in ecology, which is described by the Lotka–Volterra equations, which of course give wave-like functions:
The predator-prey model works like this. Imagine a closed ecosystem has a population of rabbits who feed on the grass, and a population of foxes who feed on the rabbits. As the rabbit population increases, the fox population happily eats them, and also increases. But then there are too many foxes, and they eat so many rabbits that the rabbit population declines. But then there are too few rabbits, and the foxes start to die off. But now that there are fewer foxes, the rabbit population can grow again…. until there are so many foxes that they once again kill off most of the rabbits.
And so on, and so on. If you were to plot the population sizes of rabbits and foxes over time, you’d get a Lotka–Volterra function, which looks suspiciously wave-like:
What Does This Have to Do With COVID?
With a pandemic like COVID, the prey (the rabbits) are…. us! And the predator (the foxes) are the virus! But the axes are different. Instead of measuring the population size of rabbits, we’re measuring the extent of human social interaction. And instead of measuring the fox population, we’re plotting the number of people infected by the virus.
As people socialize more, the virus finds more victims to infect, and the epidemic curve trends upward. In response, people start socially distancing and wearing masks and staying at home more. So the virus finds fewer victims, and its incidence rate drops. First wave over.
But then silly humans look outside their doors and see no foxes about, and assume it’s safe to go outside. They start socializing again. And guess what happens? The virus finds purchase once more, and the incidence rate climbs again…. until people remember that this is a serious thing, and adopt public health measures again, causing this second wave to retreat.
Does That Mean The Weather Has No Effect?
The weather definitely has an effect. The higher humidity of summer probably means that exhaled respiratory droplets fall to the ground faster, thus lessening slightly the need for physical distancing. When scaled up to the population level, this might result in diminished overall transmission.
Summer months are also when people are outdoors more. And as we all know now, when you’re outside you are (a) less likely to be in large groups, (b) more likely to far away from other people, and (c) have the benefit of very good ventilation so that both droplets and aerosol particles are less likely to land on you.
There are reasons that the common cold is more …common (heh heh)… in the Fall in Winter. And I just gave those reasons.
What Determines Whether There Will Be A Second Wave of COVID?
The trajectory of this disease will be determined by human response. This disease doesn’t waft into your house through an open window. It is carried in by a person. You get it from people. And people increase or lessen their chances of getting it through their behaviours.
As the Fall months approach, the weather will get cooler, so that people will spend more time indoors where distancing is more difficult and ventilation is poorer.
As business and bars and restaurants open, people will spend more time indoors and among a multitude of different contacts.
Schools are opening. As I noted in an earlier post, the risk to individuals in schools is low. But when scaled up the population level, schools are a threat to pandemic control. Schools are a pandemic accelerant and have a good chance of providing the spark for new outbreaks.
As the months tread on and COVID fatigue increasingly becomes a real and common thing, many people are forgetting their public health measures. The ridiculous anti-mask and anti-distancing movements are accelerants, as well, that provide fodder for the virus-fox who just cant resist that sweet sweet non-distancing rabbit flesh.
And, of course, there is the very important fact that the vast majority of people remain susceptible. In Canada, probably only 5% or so of people have been exposed to this virus. That means that 95% of us have no immunity.
Tell Me More About Susceptibility!
One of the most common ways of modelling infectious diseases is by using something called the SIR model. (S=susceptible, I=infected, R=removed). I’ve talked a little bit about the SIR model in an earlier post.
But the essence is that, in this idealized model, every population begins with everyone being susceptible to the disease. As the infection is introduced, some people get it and either recover or die. Either way, they are removed from the pool and are no longer susceptible. Over time, as the number of susceptible people diminishes, the infection has nowhere to grow, so dies off. The curves look like this:
If you look at the red curve, that’s essentially what a first wave looks like. It peaks and declines because the number of susceptible people diminishes.
In Canada, we still have 95% of the population being susceptible. So clearly our first wave was not driven down by reducing that number. It was driven down by public health interventions that mimicked depressed susceptibility, i.e. many people were rendered “uninfectable” through mask wearing and social distancing.
So if those public health investments were to wane, the true susceptibility of our population would once again assert itself, the predator virus would once more find purchase in the form of ripe victims, and that red curve would once more climb toward a new peak.
So Is This Inevitable?
I think that given the nature of school openings, the cold weather, the fact that businesses must remain open, and the fact of the existence of idiots protesting the need for public health measures, it is highly likely we will see the start of a new wave of COVID this coming Fall.
What is not inevitable is that the size of that wave has to be large enough to disrupt our lives. By this I mean that there is still a strong role to play in mask-wearing, general hygiene, and distancing. If enough people continue to do these things, we stand a very good chance of keeping that second peak to a very low height indeed.
So stay vigilant, friends. How this disease unfolds is 100% dependent on our actions. This is a good thing. It means our fate is in our own hands. We can choose to get through this with minimal pain. Let’s make good choices.