<p>Sure, don't hoard toilet paper – but if policymakers fear fear itself, they'll downplay real dangers to avoid "mass panic". Fear's not the problem, it's how we <em>channel</em> our fear. Fear gives us energy to deal with dangers now, and prepare for dangers later.</p>
<p>Honestly, we (Marcel, epidemiologist + Nicky, art/code) are worried. We bet you are, too! That's why we've channelled our fear into making these <strong>playable simulations</strong>, so that <em>you</em> can channel your fear into understanding:</p>
<p>This guide (published May 1st, 2020. click this footnote!→<supid="fnref1"><ahref="#fn1"rel="footnote">1</a></sup>) is meant to give you hope <em>and</em> fear. To beat COVID-19 <strong>in a way that also protects our mental & financial health</strong>, we need optimism to create plans, and pessimism to create backup plans. As Gladys Bronwyn Stern once said, <em>“The optimist invents the airplane and the pessimist the parachute.”</em></p>
<p>So, let's make a very, <em>very</em> simple "epidemic flight simulator"! In this simulation, <iconi></icon> Infectious people can turn <icons></icon> Susceptible people into more <iconi></icon> Infectious people:</p>
<p>It's estimated that, <em>at the start</em> of a COVID-19 outbreak, the virus jumps from an <iconi></icon> to an <icons></icon> every 4 days, <em>on average</em>.<supid="fnref2"><ahref="#fn2"rel="footnote">2</a></sup> (remember, there's a lot of variation)</p>
<p>If we simulate "double every 4 days"<em>and nothing else</em>, on a population starting with just 0.001% <spanclass="nowrap"><iconi></icon>,</span> what happens? </p>
<p><strong>Click "Start" to play the simulation! You can re-play it later with different settings:</strong> (technical caveats: <supid="fnref3"><ahref="#fn3"rel="footnote">3</a></sup>)</p>
<p>This is the <strong>exponential growth curve.</strong> Starts small, then explodes. "Oh it's just a flu" to "Oh right, flus don't create <em>mass graves in rich cities</em>". </p>
<p>But, this simulation is wrong. Exponential growth, thankfully, can't go on forever. One thing that stops a virus from spreading is if others <em>already</em> have the virus:</p>
<p>The more <spanclass="nowrap"><iconi></icon>s</span> there are, the faster <spanclass="nowrap"><icons></icon>s</span> become <spanclass="nowrap"><iconi></icon>s,</span><strong>but the fewer <spanclass="nowrap"><icons></icon>s</span> there are, the <em>slower</em><spanclass="nowrap"><icons></icon>s</span> become <spanclass="nowrap"><iconi></icon>s.</span></strong></p>
<p>But, this simulation is <em>still</em> wrong. We're missing the fact that <iconi></icon> Infectious people eventually stop being infectious, either by 1) recovering, 2) "recovering" with lung damage, or 3) dying.</p>
<p>For simplicity's sake, let's pretend that all <iconi></icon> Infectious people become <iconr></icon> Recovered. (Just remember that in reality, some are dead.) <spanclass="nowrap"><iconr></icon>s</span> can't be infected again, and let's pretend –<em>for now!</em>– that they stay immune for life.</p>
<p>With COVID-19, it's estimated you're <iconi></icon> Infectious for 10 days, <em>on average</em>.<supid="fnref4"><ahref="#fn4"rel="footnote">4</a></sup> That means some folks will recover before 10 days, some after. <strong>Here's what that looks like, with a simulation <em>starting</em> with 100% <spanclass="nowrap"><iconi></icon>:</span></strong></p>
<p>And <em>that's</em> where that famous curve comes from! It's not a bell curve, it's not even a "log-normal" curve. It has no name. But you've seen it a zillion times, and beseeched to flatten.</p>
<p><strong>NOTE: The simulations that inform policy are way, <em>way</em> more sophisticated than this!</strong> But the SIR Model can still explain the same general findings, even if missing the nuances.</p>
<p>Actually, let's add one more nuance: before an <icons></icon> becomes an <spanclass="nowrap"><iconi></icon>,</span> they first become <icone></icon> Exposed. This is when they have the virus but can't pass it on yet – infect<em>ed</em> but not yet infect<em>ious</em>.</p>
<p>(This variant is called the <strong>SEIR Model</strong><supid="fnref6"><ahref="#fn6"rel="footnote">6</a></sup>, where the "E" stands for <icone></icon>"Exposed". Note this <em>isn't</em> the everyday meaning of "exposed", when you may or may not have the virus. In this technical definition, "Exposed" means you definitely have it. Science terminology is bad.)</p>
<p>For COVID-19, it's estimated that you're <icone></icon> infected-but-not-yet-infectious for 3 days, <em>on average</em>.<supid="fnref7"><ahref="#fn7"rel="footnote">7</a></sup> What happens if we add that to the simulation?</p>
<p>Not much changes! How long you stay <icone></icon> Exposed changes the ratio of <spanclass="nowrap"><icone></icon>-to-<iconi></icon>,</span> and <em>when</em> current cases peak... but the <em>height</em> of that peak, and total cases in the end, stays the same.</p>
<p>Short for "Reproduction number". It's the <em>average</em> number of people an <iconi></icon> infects <em>before</em> they recover (or die).</p>
<p><strong>R<sub>0</sub></strong> (pronounced R-nought) is what R is <em>at the start of an outbreak, before immunity or interventions</em>. R<sub>0</sub> more closely reflects the power of the virus itself, but it still changes from place to place. For example, R<sub>0</sub> is higher in dense cities than sparse rural areas.</p>
<p>The R<sub>0</sub> for "the" seasonal flu is around 1.28<supid="fnref8"><ahref="#fn8"rel="footnote">8</a></sup>. This means, at the <em>start</em> of a flu outbreak, each <iconi></icon> infects 1.28 others <em>on average.</em> (If it sounds weird that this isn't a whole number, remember that the "average" mom has 2.4 children. This doesn't mean there's half-children running about.)</p>
<p>The R<sub>0</sub> for COVID-19 is estimated to be around 2.2,<supid="fnref9"><ahref="#fn9"rel="footnote">9</a></sup> though one <em>not-yet-finalized</em> study estimates it was 5.7(!) in Wuhan.<supid="fnref10"><ahref="#fn10"rel="footnote">10</a></sup></p>
<p>In our simulations –<em>at the start & on average</em>– an <iconi></icon> infects someone every 4 days, over 10 days. "4 days" goes into "10 days" two-and-a-half times. This means –<em>at the start & on average</em>– each <iconi></icon> infects 2.5 others. Therefore, R<sub>0</sub> = 2.5. (caveats:<supid="fnref11"><ahref="#fn11"rel="footnote">11</a></sup>)</p>
<p>But remember, the fewer <spanclass="nowrap"><icons></icon>s</span> there are, the <em>slower</em><spanclass="nowrap"><icons></icon>s</span> become <spanclass="nowrap"><iconi>s.</icon></span> The <em>current</em> reproduction number (R) depends not just on the <em>basic</em> reproduction number (R<sub>0</sub>), but <em>also</em> on how many people are no longer <icons></icon> Susceptible. (For example, by recovering & getting natural immunity.)</p>
<p>When enough people have immunity, R < 1, and the virus is contained! This is called <strong>herd immunity</strong>. For flus, herd immunity is achieved <em>with a vaccine</em>. Trying to achieve "natural herd immunity" by letting folks get infected is a <em>terrible</em> idea. (But not for the reason you may think! We'll explain later.)</p>
<p><strong>NOTE: Total cases <em>does not stop</em> at herd immunity, but overshoots it!</strong> And it crosses the threshold <em>exactly</em> when current cases peak. (This happens no matter how you change the settings – try it for yourself!)</p>
<p>This is because when there are more <spanclass="nowrap">non-<icons></icon>s</span> than the herd immunity threshold, you get R < 1. And when R < 1, new cases stop growing: a peak.</p>
<p><strong>If there's only one lesson you take away from this guide, here it is</strong>– it's an extremely complex diagram so please take time to fully absorb it:</p>
<p>It's a paradox. COVID-19 is extremely contagious, yet to contain it, we "only" need to stop more than 60% of infections. 60%?! If that was a school grade, that's a D-. But if R<sub>0</sub> = 2.5, cutting that by 61% gives us R = 0.975, which is R < 1, virus is contained! (exact formula:<supid="fnref12"><ahref="#fn12"rel="footnote">12</a></sup>)</p>
<p>(If you think R<sub>0</sub> or the other numbers in our simulations are too low/high, that's good you're challenging our assumptions! There'll be a "Sandbox Mode" at the end of this guide, where you can plug in your <em>own</em> numbers, and simulate what happens.)</p>
<p><em>Every</em> COVID-19 intervention you've heard of – handwashing, social/physical distancing, lockdowns, self-isolation, contact tracing & quarantining, face masks, even "herd immunity"– they're <em>all</em> doing the same thing:</p>
<p>So now, let's use our "epidemic flight simulator" to figure this out: How can we get R < 1 in a way <strong>that also protects our mental health <em>and</em> financial health?</strong></p>
<p>Around 1 in 20 people infected with COVID-19 need to go to an ICU (Intensive Care Unit).<supid="fnref13"><ahref="#fn13"rel="footnote">13</a></sup> In a rich country like the USA, there's 1 ICU bed per 3400 people.<supid="fnref14"><ahref="#fn14"rel="footnote">14</a></sup> Therefore, the USA can handle 20 out of 3400 people being <em>simultaneously</em> infected – or, 0.6% of the population.</p>
<p>That's what <ahref="http://www.imperial.ac.uk/mrc-global-infectious-disease-analysis/covid-19/report-9-impact-of-npis-on-covid-19/">the March 16 Imperial College report</a> found: do nothing, and we run out of ICUs, with more than 80% of the population getting infected.
(remember: total cases <em>overshoots</em> herd immunity)</p>
<p>Even if only 0.5% of infected die – a generous assumption when there's no more ICUs – in a large country like the US, with 300 million people, 0.5% of 80% of 300 million = still 1.2 million dead... <em>IF we did nothing.</em></p>
<p>(Lots of news & social media reported "80% will be infected"<em>without</em>"IF WE DO NOTHING". Fear was channelled into clicks, not understanding. <em>Sigh.</em>)</p>
<p>The "Flatten The Curve" plan was touted by every public health organization, while the United Kingdom's original "herd immunity" plan was universally booed. They were <em>the same plan.</em> The UK just communicated theirs poorly.<supid="fnref15"><ahref="#fn15"rel="footnote">15</a></sup></p>
<p>Increased handwashing cuts flus & colds in high-income countries by ~25%<supid="fnref16"><ahref="#fn16"rel="footnote">16</a></sup>, while the city-wide lockdown in London cut close contacts by ~70%<supid="fnref17"><ahref="#fn17"rel="footnote">17</a></sup>. So, let's assume handwashing can reduce R by <em>up to</em> 25%, and distancing can reduce R by <em>up to</em> 70%:</p>
<p><strong>Play with this calculator to see how % of <spanclass="nowrap">non-<icons></icon>,</span> handwashing, and distancing reduce R:</strong> (this calculator visualizes their <em>relative</em> effects, which is why increasing one <em>looks</em> like it decreases the effect of the others.<supid="fnref18"><ahref="#fn18"rel="footnote">18</a></sup>)</p>
<p>Now, let's simulate what happens to a COVID-19 epidemic if, starting March 2020, we had increased handwashing but only <em>mild</em> physical distancing – so that R is lower, but still above 1:</p>
<li><p>This <em>reduces</em> total cases! <strong>Even if you don't get R < 1, reducing R still saves lives, by reducing the 'overshoot' above herd immunity.</strong> Lots of folks think "Flatten The Curve" spreads out cases without reducing the total. This is impossible in <em>any</em> Epidemiology 101 model. But because the news reported "80%+ will be infected" as inevitable, folks thought total cases will be the same no matter what. <em>Sigh.</em></p></li>
<li><p>Due to the extra interventions, current cases peak <em>before</em> herd immunity is reached. In fact, in this simulation, total cases only overshoots <em>a tiny bit</em> above herd immunity – the UK's plan! At that point, R < 1, you can let go of all other interventions, and COVID-19 stays contained! Well, except for one problem...</p></li>
<p>That was the other finding of the March 16 Imperial College report, which convinced the UK to abandon its original plan. Any attempt at <strong>mitigation</strong> (reduce R, but R > 1) will fail. The only way out is <strong>suppression</strong> (reduce R so that R < 1).</p>
<p>Let's see what happens if we <em>crush</em> the curve with a 5-month lockdown, reduce <iconi></icon> to nearly nothing, then finally –<em>finally</em>– return to normal life:</p>
<p>This is the "second wave" everyone's talking about. As soon as we remove the lockdown, we get R > 1 again. So, a single leftover <iconi></icon> (or imported <spanclass="nowrap"><iconi></icon>)</span> can cause a spike in cases that's almost as bad as if we'd done Scenario 0: Absolutely Nothing.</p>
<p>This solution was first suggested by the March 16 Imperial College report, and later again by a Harvard paper.<supid="fnref19"><ahref="#fn19"rel="footnote">19</a></sup></p>
<p><strong>Here's a simulation:</strong> (After playing the "recorded scenario", you can try simulating your <em>own</em> lockdown schedule, by changing the sliders <em>while</em> the simulation is running! Remember you can pause & continue the sim, and change the simulation speed)</p>
<p>This <em>would</em> keep cases below ICU capacity! And it's <em>much</em> better than an 18-month lockdown until a vaccine is available. We just need to... shut down for a few months, open up for a few months, and repeat until a vaccine is available. (And if there's no vaccine, repeat until herd immunity is reached... in 2022.)</p>
<p>Look, it's nice to draw a line saying "ICU capacity", but there's lots of important things we <em>can't</em> simulate here. Like:</p>
<p><strong>Mental Health:</strong> Loneliness is one of the biggest risk factors for depression, anxiety, and suicide. And it's as associated with an early death as smoking 15 cigarettes a day.<supid="fnref20"><ahref="#fn20"rel="footnote">20</a></sup></p>
<p><strong>Financial Health:</strong>"What about the economy" sounds like you care more about dollars than lives, but "the economy" isn't just stocks: it's people's ability to provide food & shelter for their loved ones, to invest in their kids' futures, and enjoy arts, foods, videogames – the stuff that makes life worth living. And besides, poverty <em>itself</em> has horrible impacts on mental and physical health.</p>
<p><em>"Sure, we *could've* done what Taiwan & South Korea did at the start, but it's too late now. We missed the start."</em></p>
<p>To understand how Taiwan & South Korea contained COVID-19, we need to understand the exact timeline of a typical COVID-19 infection<supid="fnref21"><ahref="#fn21"rel="footnote">21</a></sup>:</p>
<p>This is called <strong>contact tracing</strong>. It's an old idea, was used at an unprecedented scale to contain Ebola<supid="fnref23"><ahref="#fn23"rel="footnote">23</a></sup>, and now it's core part of how Taiwan & South Korea are containing COVID-19!</p>
<p>(It also lets us use our limited tests more efficiently, to find pre-symptomatic <iconi></icon>s without needing to test almost everyone.)</p><p>(It also lets us use our limited tests more efficiently, to find pre-symptomatic <spanclass="nowrap"><iconi></icon>s</span> without needing to test almost everyone.)</p>
<p>Traditionally, contacts are found with in-person interviews, but those <em>alone</em> are too slow for COVID-19's ~48 hour window. That's why contact tracers need help, and be supported by –<em>NOT</em> replaced by – contact tracing apps.</p>
<p>(This idea didn't come from "techies": using an app to fight COVID-19 was first proposed by <ahref="https://science.sciencemag.org/content/early/2020/04/09/science.abb6936">a team of Oxford epidemiologists</a>.)</p>
<p>Wait, apps that trace who you've been in contact with?... Does that mean giving up privacy, giving in to Big Brother?</p>
<p>Heck no! <strong><ahref="https://github.com/DP-3T/documents#decentralized-privacy-preserving-proximity-tracing">DP-3T</a></strong>, a team of epidemiologists & cryptographers (including one of us, Marcel Salathé) is <em>already</em> making a contact tracing app – with code available to the public – that reveals <strong>no info about your identity, location, who your contacts are, or even <em>how many contacts</em> you've had.</strong></p>
<p>Along with similar teams like TCN Protocol<supid="fnref24"><ahref="#fn24"rel="footnote">24</a></sup> and MIT PACT<supid="fnref25"><ahref="#fn25"rel="footnote">25</a></sup>, they've inspired Apple & Google to bake privacy-first contact tracing directly into Android/iOS.<supid="fnref26"><ahref="#fn26"rel="footnote">26</a></sup> (Don't trust Google/Apple? Good! The beauty of this system is it doesn't <em>need</em> trust!) Soon, your local public health agency may ask you to download an app. If it's privacy-first with publicly-available code, please do!</p>
<p>But what about folks without smartphones? Or infections through doorknobs? Or "true" asymptomatic cases? Contact tracing apps can't catch all transmissions... <em>and that's okay!</em> We don't need to catch <em>all</em> transmissions, just 60%+ to get R < 1.</p>
<p>(Rant about the confusion about pre-symptomatic vs "true" asymptomatic. "True" asymptomatics are rare:<supid="fnref27"><ahref="#fn27"rel="footnote">27</a></sup>)</p>
<p>Isolating <em>symptomatic</em> cases would reduce R by up to 40%, and quarantining their <em>pre/a-symptomatic</em> contacts would reduce R by up to 50%<supid="fnref28"><ahref="#fn28"rel="footnote">28</a></sup>:</p>
<p>Thus, even without 100% contact quarantining, we can get R < 1 <em>without a lockdown!</em> Much better for our mental & financial health. (As for the cost to folks who have to self-isolate/quarantine, <em>governments should support them</em>– pay for the tests, job protection, subsidized paid leave, etc. Still way cheaper than intermittent lockdown.)</p>
<p>We then keep R < 1 until we have a vaccine, which turns susceptible <spanclass="nowrap"><icons></icon>s</span> into immune <spanclass="nowrap"><iconr></icon>s.</span> Herd immunity, the <em>right</em> way:</p>
<p>(Note: this calculator pretends the vaccines are 100% effective. Just remember that in reality, you'd have to compensate by vaccinating <em>more</em> than "herd immunity", to <em>actually</em> get herd immunity)</p>
<p>But what if things <em>still</em> go wrong? Things have gone horribly wrong already. That's fear, and that's good! Fear gives us energy to create <em>backup plans</em>.</p>
<p>What if R<sub>0</sub> is way higher than we thought, and the above interventions, even with mild distancing, <em>still</em> aren't enough to get R < 1?</p>
<p>Remember, even if we can't get R < 1, reducing R still reduces the "overshoot" in total cases, thus saving lives. But still, R < 1 is the ideal, so here's a few other ways to reduce R:</p>
<p>You're right. Masks don't stop you from getting sick<supid="fnref29"><ahref="#fn29"rel="footnote">29</a></sup>... they stop you from getting <em>others</em> sick.</p>
<p>To put a number on it: surgical masks <em>on the sick person</em> reduce cold & flu viruses in aerosols by 70%.<supid="fnref30"><ahref="#fn30"rel="footnote">30</a></sup> Reducing transmissions by 70% would be as large an impact as a lockdown!</p>
<p>However, we don't know for sure the impact of masks on COVID-19 <em>specifically</em>. In science, one should only publish a finding if you're 95% sure of it. (...should.<supid="fnref31"><ahref="#fn31"rel="footnote">31</a></sup>) Masks, as of May 1st 2020, are less than "95% sure".</p>
<p>However, pandemics are like poker. <strong>Make bets only when you're 95% sure, and you'll lose everything at stake.</strong> As a recent article on masks in the British Medical Journal notes,<supid="fnref32"><ahref="#fn32"rel="footnote">32</a></sup> we <em>have</em> to make cost/benefit analyses under uncertainty. Like so:</p>
<p>Cost: If homemade cloth masks (which are ~2/3 as effective as surgical masks<supid="fnref33"><ahref="#fn33"rel="footnote">33</a></sup>), super cheap. If surgical masks, more expensive but still pretty cheap.</p>
<p>Benefit: Even if it's a 50–50 chance of surgical masks reducing transmission by 0% or 70%, the average "expected value" is still 35%, same as a half-lockdown! So let's guess-timate that surgical masks reduce R by up to 35%, discounted for our uncertainty. (Again, you can challenge our assumptions by turning the sliders up/down)</p>
<p>Masks <em>alone</em> won't get R < 1. But if handwashing &"Test, Trace, Isolate" only gets us to R = 1.10, having just 1/3 of people wear masks would tip that over to R < 1, virus contained!</p>
<p>Okay, this isn't an "intervention" we can control, but it will help! Some news outlets report that summer won't do anything to COVID-19. They're half right: summer won't get R < 1, but it <em>will</em> reduce R.</p>
<p>For COVID-19, every extra 1° Celsius (2.2° Fahrenheit) makes R drop by 1.2%.<supid="fnref35"><ahref="#fn35"rel="footnote">35</a></sup> The summer-winter difference in New York City is 15°C (60°F), so summer will make R drop by 18%.</p>
<p>Summer alone won't make R < 1, but if we have limited resources, we can scale back some interventions in the summer – so we can scale them <em>higher</em> in the winter.</p>
<p>But we wouldn't have to be 2-months-closed / 1-month-open over & over! Because R is reduced, we'd only need one or two more "circuit breaker" lockdowns before a vaccine is available. (Singapore had to do this recently, "despite" having controlled COVID-19 for 4 months. That's not failure: this <em>is</em> what success takes.)</p>
<p><strong>Even under a pessimistic scenario, it <em>is</em> possible to beat COVID-19, while protecting our mental and financial health.</strong> Use the lockdown as a "reset button", keep R < 1 with case isolation + privacy-protecting contract tracing + at <em>least</em> cloth masks for all... and life can get back to a normal-ish!</p>
<p>Sure, you may have dried-out hands. But you'll get to invite a date out to a comics bookstore! You'll get to go out with friends to watch the latest Hollywood cash-grab. You'll get to people-watch at a library, taking joy in people going about the simple business of <em>being alive.</em></p>
<p>So now, let's plan for some <em>worse</em> worst-case scenarios. Water landing, get your life jacket, and please follow the lights to the emergency exits:</p>
<li>COVID-19 is most closely related to SARS, which gave its survivors 2 years of immunity.<supid="fnref36"><ahref="#fn36"rel="footnote">36</a></sup></li>
<li>The coronaviruses that cause "the" common cold give you 8 months of immunity.<supid="fnref37"><ahref="#fn37"rel="footnote">37</a></sup></li>
<li>There's reports of folks recovering from COVID-19, then testing positive again, but it's unclear if these are false positives.<supid="fnref38"><ahref="#fn38"rel="footnote">38</a></sup></li>
<li>One <em>not-yet-peer-reviewed</em> study on monkeys showed immunity to the COVID-19 coronavirus for at least 28 days.<supid="fnref39"><ahref="#fn39"rel="footnote">39</a></sup></li>
<strong>Here's a simulation starting with 100% <spanclass="nowrap"><iconr></icon></strong>,</span> exponentially decaying into susceptible, no-immunity <spanclass="nowrap"><icons></icon>s</span> after 1 year, on <em>average</em>, with variation:</p>
<p>In previous simulations, we only had <em>one</em> ICU-overwhelming spike. Now, we have several, <em>and</em><iconi></icon> cases come to a rest <em>permanently at</em> ICU capacity. (Which, remember, we <em>tripled</em> for these simulations)</p>
<p>Counterintuitively, summer makes the spikes worse <em>and</em> regular! This is because summer reduces new <spanclass="nowrap"><iconi></icon>s,</span> but that in turn reduces new immune <spanclass="nowrap"><iconr></icon>s.</span> Which means immunity plummets in the summer, <em>creating</em> large regular spikes in the winter.</p>
<p><strong>(After playing the recording, try simulating your own vaccination campaigns! Remember you can pause/continue the sim at any time)</strong></p>
<p><strong>To be clear: this is unlikely.</strong> Most epidemiologists expect a vaccine in 1 to 2 years. Sure, there's never been a vaccine for any of the other coronaviruses before, but that's because SARS was eradicated quickly, and "the" common cold wasn't worth the investment. </p>
<p>Still, infectious disease researchers have expressed worries: What if we can't make enough?<supid="fnref40"><ahref="#fn40"rel="footnote">40</a></sup> What if we rush it, and it's not safe?<supid="fnref41"><ahref="#fn41"rel="footnote">41</a></sup></p>
<p>1) Do intermittent or loose R < 1 interventions, to reach "natural herd immunity". (Warning: this will result in many deaths & damaged lungs. <em>And</em> won't work if immunity doesn't last.)</p>
<p>2) Do the R < 1 interventions forever. Contact tracing & wearing masks just becomes a new norm in the post-COVID-19 world, like how STI tests & wearing condoms became a new norm in the post-HIV world.</p>
<p>3) Do the R < 1 interventions until we develop treatments that make COVID-19 way, way less likely to need critical care. (Which we should be doing <em>anyway!</em>) Reducing ICU use by 10x is the same as increasing our ICU capacity by 10x:</p>
<p><strong>Here's a simulation of <em>no</em> lasting immunity, <em>no</em> vaccine, and not even any interventions – just slowly increasing capacity to survive the long-term spikes:</strong></p>
<p>Maybe you'd like to challenge our assumptions, and try different R<sub>0</sub>'s or numbers. Or try simulating your <em>own</em> combination of intervention plans!</p>
<p><strong>Here's an (optional) Sandbox Mode, with <em>everything</em> available. (scroll to see all controls) Simulate & play around to your heart's content:</strong></p>
<p>This basic "epidemic flight simulator" has taught us so much. It's let us answer questions about the past few months, next few months, and next few years.</p>
<p>Plane's sunk. We've scrambled onto the life rafts. It's time to find dry land.<supid="fnref42"><ahref="#fn42"rel="footnote">42</a></sup></p>
<p>Teams of epidemiologists and policymakers (<ahref="https://www.americanprogress.org/issues/healthcare/news/2020/04/03/482613/national-state-plan-end-coronavirus-crisis/">left</a>, <ahref="https://www.aei.org/research-products/report/national-coronavirus-response-a-road-map-to-reopening/">right</a>, and <ahref="https://ethics.harvard.edu/covid-roadmap">multi-partisan</a>) have come to a consensus on how to beat COVID-19, while protecting our lives <em>and</em> liberties.</p>
<p><strong>For everyone:</strong> Respect the lockdown so we can get out of Phase I asap. Keep washing those hands. Make your own masks. Download a <em>privacy-protecting</em> contact tracing app when those are available next month. Stay healthy, physically & mentally! And write your local policymaker to get off their butt and...</p>
<p><strong>For policymakers:</strong> Make laws to support folks who have to self-isolate/quarantine. Hire more manual contact tracers, <em>supported</em> by privacy-protecting contact tracing apps. Direct more funds into the stuff we should be building, like...</p>
<p>Don't downplay fear to build up hope. Our fear should <em>team up</em> with our hope, like the inventors of airplanes & parachutes. Preparing for horrible futures is how we <em>create</em> a hopeful future.</p>
<p>These footnotes will have sources, links, or bonus commentary. Like this commentary! <ahref="#fnref1"rev="footnote">↩</a></p>
<p><strong>This guide was published on May 1st, 2020.</strong> Many details will become outdated, but we're confident this guide will cover 95% of possible futures, and that Epidemiology 101 will remain forever useful.</p>
<p>“The mean [serial] interval was 3.96 days (95% CI 3.53–4.39 days)”. <ahref="https://wwwnc.cdc.gov/eid/article/26/6/20-0357_article">Du Z, Xu X, Wu Y, Wang L, Cowling BJ, Ancel Meyers L</a> (Disclaimer: Early release articles are not considered as final versions) <ahref="#fnref2"rev="footnote">↩</a></p>
<p><strong>Remember: all these simulations are super simplified, for educational purposes.</strong> <ahref="#fnref3"rev="footnote">↩</a></p>
<p>One simplification: When you tell this simulation "Infect 1 new person every X days", it's actually increasing # of infected by 1/X each day. Same for future settings in these simulations –"Recover every X days" is actually reducing # of infected by 1/X each day.</p>
<p>Those <em>aren't</em> exactly the same, but it's close enough, and for educational purposes it's less opaque than setting the transmission/recovery rates directly.</p>
<p>“The median communicable period [...] was 9.5 days.” <ahref="https://link.springer.com/article/10.1007/s11427-020-1661-4">Hu, Z., Song, C., Xu, C. et al</a> Yes, we know "median" is not the same as "average". For simplified educational purposes, close enough. <ahref="#fnref4"rev="footnote">↩</a></p>
<p>For more technical explanations of the SIR Model, see <ahref="https://www.idmod.org/docs/hiv/model-sir.html#">the Institute for Disease Modeling</a> and <ahref="https://en.wikipedia.org/wiki/Compartmental_models_in_epidemiology#The_SIR_model">Wikipedia</a> <ahref="#fnref5"rev="footnote">↩</a></p>
<p>For more technical explanations of the SEIR Model, see <ahref="https://www.idmod.org/docs/hiv/model-seir.html">the Institute for Disease Modeling</a> and <ahref="https://en.wikipedia.org/wiki/Compartmental_models_in_epidemiology#The_SEIR_model">Wikipedia</a> <ahref="#fnref6"rev="footnote">↩</a></p>
<p>“Assuming an incubation period distribution of mean 5.2 days from a separate study of early COVID-19 cases, we inferred that infectiousness started from 2.3 days (95% CI, 0.8–3.0 days) before symptom onset” (translation: Assuming symptoms start at 5 days, infectiousness starts 2 days before = Infectiousness starts at 3 days) <ahref="https://www.nature.com/articles/s41591-020-0869-5">He, X., Lau, E.H.Y., Wu, P. et al.</a> <ahref="#fnref7"rev="footnote">↩</a></p>
<p>“The median R value for seasonal influenza was 1.28 (IQR: 1.19–1.37)” <ahref="https://bmcinfectdis.biomedcentral.com/articles/10.1186/1471-2334-14-480">Biggerstaff, M., Cauchemez, S., Reed, C. et al.</a> <ahref="#fnref8"rev="footnote">↩</a></p>
<p>“We estimated the basic reproduction number R0 of 2019-nCoV to be around 2.2 (90% high density interval: 1.4–3.8)” <ahref="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7001239/">Riou J, Althaus CL.</a> <ahref="#fnref9"rev="footnote">↩</a></p>
<p>“we calculated a median R0 value of 5.7 (95% CI 3.8–8.9)” <ahref="https://wwwnc.cdc.gov/eid/article/26/7/20-0282_article">Sanche S, Lin YT, Xu C, Romero-Severson E, Hengartner N, Ke R.</a> <ahref="#fnref10"rev="footnote">↩</a></p>
<p>This is pretending that you're equally infectious all throughout your "infectious period". Again, simplifications for educational purposes. <ahref="#fnref11"rev="footnote">↩</a></p>
<p>Remember R = R<sub>0</sub> * the ratio of transmissions still allowed. Remember also that ratio of transmissions allowed = 1 - ratio of transmissions <em>stopped</em>. <ahref="#fnref12"rev="footnote">↩</a></p>
<p>Therefore, to get R < 1, you need to get R<sub>0</sub> * TransmissionsAllowed < 1. </p>
<p><ahref="https://www.statista.com/statistics/1105420/covid-icu-admission-rates-us-by-age-group/">"Percentage of COVID-19 cases in the United States from February 12 to March 16, 2020 that required intensive care unit (ICU) admission, by age group"</a>. Between 4.9% to 11.5% of <em>all</em> COVID-19 cases required ICU. Generously picking the lower range, that's 5% or 1 in 20. Note that this total is specific to the US's age structure, and will be higher in countries with older populations, lower in countries with younger populations. <ahref="#fnref13"rev="footnote">↩</a></p>
<p>“Number of ICU beds = 96,596”. From <ahref="https://sccm.org/Blog/March-2020/United-States-Resource-Availability-for-COVID-19">the Society of Critical Care Medicine</a> USA Population was 328,200,000 in 2019. 96,596 out of 328,200,000 = roughly 1 in 3400. <ahref="#fnref14"rev="footnote">↩</a></p>
<p>“He says that the actual goal is the same as that of other countries: flatten the curve by staggering the onset of infections. As a consequence, the nation may achieve herd immunity; it’s a side effect, not an aim. [...] The government’s actual coronavirus action plan, available online, doesn’t mention herd immunity at all.” <ahref="#fnref15"rev="footnote">↩</a></p>
<p>From a <ahref="https://www.theatlantic.com/health/archive/2020/03/coronavirus-pandemic-herd-immunity-uk-boris-johnson/608065/">The Atlantic article by Ed Yong</a></p>
<p>“All eight eligible studies reported that handwashing lowered risks of respiratory infection, with risk reductions ranging from 6% to 44% [pooled value 24% (95% CI 6–40%)].” We rounded up the pooled value to 25% in these simulations for simplicity. <ahref="https://onlinelibrary.wiley.com/doi/full/10.1111/j.1365-3156.2006.01568.x">Rabie, T. and Curtis, V.</a> Note: as this meta-analysis points out, the quality of studies for handwashing (at least in high-income countries) are awful. <ahref="#fnref16"rev="footnote">↩</a></p>
<p>“We found a 73% reduction in the average daily number of contacts observed per participant. This would be sufficient to reduce R0 from a value from 2.6 before the lockdown to 0.62 (0.37 - 0.89) during the lockdown”. We rounded it down to 70% in these simulations for simplicity. <ahref="https://cmmid.github.io/topics/covid19/comix-impact-of-physical-distance-measures-on-transmission-in-the-UK.html">Jarvis and Zandvoort et al</a> <ahref="#fnref17"rev="footnote">↩</a></p>
<p>This distortion would go away if we plotted R on a logarithmic scale... but then we'd have to explain <em>logarithmic scales.</em> <ahref="#fnref18"rev="footnote">↩</a></p>
<p>“Absent other interventions, a key metric for the success of social distancing is whether critical care capacities are exceeded. To avoid this, prolonged or intermittent social distancing may be necessary into 2022.” <ahref="https://science.sciencemag.org/content/early/2020/04/14/science.abb5793">Kissler and Tedijanto et al</a> <ahref="#fnref19"rev="footnote">↩</a></p>
<p>See <ahref="https://journals.sagepub.com/doi/abs/10.1177/1745691614568352">Figure 6 from Holt-Lunstad & Smith 2010</a>. Of course, big disclaimer that they found a <em>correlation</em>. But unless you want to try randomly assigning people to be lonely for life, observational evidence is all you're gonna get. <ahref="#fnref20"rev="footnote">↩</a></p>
<p><strong>3 days on average to infectiousness:</strong> “Assuming an incubation period distribution of mean 5.2 days from a separate study of early COVID-19 cases, we inferred that infectiousness started from 2.3 days (95% CI, 0.8–3.0 days) before symptom onset” (translation: Assuming symptoms start at 5 days, infectiousness starts 2 days before = Infectiousness starts at 3 days) <ahref="https://www.nature.com/articles/s41591-020-0869-5">He, X., Lau, E.H.Y., Wu, P. et al.</a> <ahref="#fnref21"rev="footnote">↩</a></p>
<p><strong>4 days on average to infecting someone else:</strong> “The mean [serial] interval was 3.96 days (95% CI 3.53–4.39 days)” <ahref="https://wwwnc.cdc.gov/eid/article/26/6/20-0357_article">Du Z, Xu X, Wu Y, Wang L, Cowling BJ, Ancel Meyers L</a></p>
<p><strong>5 days on average to feeling symptoms:</strong> “The median incubation period was estimated to be 5.1 days (95% CI, 4.5 to 5.8 days)” <ahref="https://annals.org/AIM/FULLARTICLE/2762808/INCUBATION-PERIOD-CORONAVIRUS-DISEASE-2019-COVID-19-FROM-PUBLICLY-REPORTED">Lauer SA, Grantz KH, Bi Q, et al</a></p>
<p>“We estimated that 44% (95% confidence interval, 25–69%) of secondary cases were infected during the index cases’ presymptomatic stage” <ahref="https://www.nature.com/articles/s41591-020-0869-5">He, X., Lau, E.H.Y., Wu, P. et al</a> <ahref="#fnref22"rev="footnote">↩</a></p>
<p>“Contact tracing was a critical intervention in Liberia and represented one of the largest contact tracing efforts during an epidemic in history.” <ahref="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6152989/">Swanson KC, Altare C, Wesseh CS, et al.</a> <ahref="#fnref23"rev="footnote">↩</a></p>
<p><ahref="https://www.apple.com/ca/newsroom/2020/04/apple-and-google-partner-on-covid-19-contact-tracing-technology/">Apple and Google partner on COVID-19 contact tracing technology </a>. Note they're not making the apps <em>themselves</em>, just creating the systems that will <em>support</em> those apps. <ahref="#fnref26"rev="footnote">↩</a></p>
<p>Lots of news reports – and honestly, many research papers – did not distinguish between "cases who showed no symptoms when we tested them" (pre-symptomatic) and "cases who showed no symptoms <em>ever</em>" (true asymptomatic). The only way you could tell the difference is by following up with cases later. <ahref="#fnref27"rev="footnote">↩</a></p>
<p>Which is what <ahref="https://wwwnc.cdc.gov/eid/article/26/8/20-1274_article">this study</a> did. (Disclaimer: "Early release articles are not considered as final versions.") In a call center in South Korea that had a COVID-19 outbreak, "only 4 (1.9%) remained asymptomatic within 14 days of quarantine, and none of their household contacts acquired secondary infections."</p>
<p>So that means "true asymptomatics" are rare, and catching the disease from a true asymptomatic may be even rarer!</p>
<p>From the same Oxford study that first recommended apps to fight COVID-19: <ahref="https://science.sciencemag.org/content/early/2020/04/09/science.abb6936/tab-figures-data">Luca Ferretti & Chris Wymant et al</a> See Figure 2. Assuming R<sub>0</sub> = 2.0, they found that: <ahref="#fnref28"rev="footnote">↩</a></p>
<ul>
<li>Symptomatics contribute R = 0.8 (40%)</li>
<li>Pre-symptomatics contribute R = 0.9 (45%)</li>
<li>Asymptomatics contribute R = 0.1 (5%, though their model has uncertainty and it could be much lower)</li>
<li>Environmental stuff like doorknobs contribute R = 0.2 (10%)</li>
</ul>
<p>And add up the pre- & a-symptomatic contacts (45% + 5%) and you get 50% of R!</p>
<p>“None of these surgical masks exhibited adequate filter performance and facial fit characteristics to be considered respiratory protection devices.” <ahref="https://www.sciencedirect.com/science/article/pii/S0196655307007742">Tara Oberg & Lisa M. Brosseau</a> <ahref="#fnref29"rev="footnote">↩</a></p>
<p>“The overall 3.4 fold reduction [70% reduction] in aerosol copy numbers we observed combined with a nearly complete elimination of large droplet spray demonstrated by Johnson et al. suggests that surgical masks worn by infected persons could have a clinically significant impact on transmission.” <ahref="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3591312/">Milton DK, Fabian MP, Cowling BJ, Grantham ML, McDevitt JJ</a> <ahref="#fnref30"rev="footnote">↩</a></p>
<p>Any actual scientist who read that last sentence is probably laugh-crying right now. See: <ahref="https://en.wikipedia.org/wiki/Data_dredging">p-hacking</a>, <ahref="https://en.wikipedia.org/wiki/Replication_crisis">the replication crisis</a>) <ahref="#fnref31"rev="footnote">↩</a></p>
<p>“It is time to apply the precautionary principle” <ahref="https://www.bmj.com/content/bmj/369/bmj.m1435.full.pdf">Trisha Greenhalgh et al [PDF]</a> <ahref="#fnref32"rev="footnote">↩</a></p>
<p><ahref="https://www.cambridge.org/core/journals/disaster-medicine-and-public-health-preparedness/article/testing-the-efficacy-of-homemade-masks-would-they-protect-in-an-influenza-pandemic/0921A05A69A9419C862FA2F35F819D55">Davies, A., Thompson, K., Giri, K., Kafatos, G., Walker, J., & Bennett, A</a> See Table 1: a 100% cotton T-shirt has around 2/3 the filtration efficiency as a surgical mask, for the two bacterial aerosols they tested. <ahref="#fnref33"rev="footnote">↩</a></p>
<p><strong>"We need to save supplies for hospitals."</strong><em>Absolutely agreed.</em> But that's more of an argument for increasing mask production, not rationing. In the meantime, we can make cloth masks. <ahref="#fnref34"rev="footnote">↩</a></p>
<p><strong>"They're hard to wear correctly."</strong> It's also hard to wash your hands according to the WHO Guidelines – seriously, "Step 3) right palm over left dorsum"?! – but we still recommend handwashing, because imperfect is still better than nothing.</p>
<p><strong>"It'll make people more reckless with handwashing & social distancing."</strong> Sure, and safety belts make people ignore stop signs, and flossing makes people eat rocks. But seriously, we'd argue the opposite: masks are a <em>constant physical reminder</em> to be careful – and in East Asia, masks are also a symbol of solidarity!</p>
<p>“One-degree Celsius increase in temperature [...] lower[s] R by 0.0225” and “The average R-value of these 100 cities is 1.83”. 0.0225 ÷ 1.83 = ~1.2%. <ahref="https://papers.ssrn.com/sol3/Papers.cfm?abstract_id=3551767">Wang, Jingyuan and Tang, Ke and Feng, Kai and Lv, Weifeng</a> <ahref="#fnref35"rev="footnote">↩</a></p>
<p>“SARS-specific antibodies were maintained for an average of 2 years [...] Thus, SARS patients might be susceptible to reinfection ≥3 years after initial exposure.” <ahref="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2851497/">Wu LP, Wang NC, Chang YH, et al.</a>"Sadly" we'll never know how long SARS immunity would have really lasted, since we eradicated it so quickly. <ahref="#fnref36"rev="footnote">↩</a></p>
<p>“We found no significant difference between the probability of testing positive at least once and the probability of a recurrence for the beta-coronaviruses HKU1 and OC43 at 34 weeks after enrollment/first infection.” <ahref="http://www.columbia.edu/%7Ejls106/galanti_shaman_ms_supp.pdf">Marta Galanti & Jeffrey Shaman (PDF)</a> <ahref="#fnref37"rev="footnote">↩</a></p>
<p>“Once a person fights off a virus, viral particles tend to linger for some time. These cannot cause infections, but they can trigger a positive test.” <ahref="https://www.statnews.com/2020/04/20/everything-we-know-about-coronavirus-immunity-and-antibodies-and-plenty-we-still-dont/">from STAT News by Andrew Joseph</a> <ahref="#fnref38"rev="footnote">↩</a></p>
<p>From <ahref="https://www.biorxiv.org/content/10.1101/2020.03.13.990226v1.abstract">Bao et al.</a><em>Disclaimer: This article is a preprint and has not been certified by peer review (yet).</em> Also, to emphasize: they only tested re-infection 28 days later. <ahref="#fnref39"rev="footnote">↩</a></p>
</li>
<liid="fn40">
<p>“If a coronavirus vaccine arrives, can the world make enough?” <ahref="https://www.nature.com/articles/d41586-020-01063-8">by Roxanne Khamsi, on Nature</a> <ahref="#fnref40"rev="footnote">↩</a></p>
</li>
<liid="fn41">
<p>“Don’t rush to deploy COVID-19 vaccines and drugs without sufficient safety guarantees” <ahref="https://www.nature.com/articles/d41586-020-00751-9">by Shibo Jiang, on Nature</a> <ahref="#fnref41"rev="footnote">↩</a></p>
</li>
<liid="fn42">
<p>Dry land metaphor <ahref="https://www.statnews.com/2020/04/01/navigating-covid-19-pandemic/">from Marc Lipsitch & Yonatan Grad, on STAT News</a> <ahref="#fnref42"rev="footnote">↩</a></p>
