【和訳】Many Hammers, Many Dances: A Key Driver of Diversity in COVID-19 Policies

スマートニュース メディア研究所より発表された新型コロナウイルスに関する論考『Many Hammers, Many Dances: A Key Driver of Diversity in COVID-19 Policies』を和訳いたしました(翻訳責任:江本伸悟)。著者の了承を頂いた上で、ここに紹介させていただきます。DeepL翻訳を下訳に用いています。明らかな誤訳がありましたらお知らせいただけますと幸いです。

Many Hammers, Many Dances: A Key Driver of Diversity in COVID-19 Policies

Jun 19 · 29 min read
Ken Suzuki, Rich Jaroslovsky, Yuhei Nishioka, Chao Song, Tonghwi Soh
  • Why are there so many different policies and strategies toward COVID-19? Is there a single “right” policy to combat the disease?
  • COVID-19に向けて、さまざまな方針や戦略があるのはなぜか?この病気と闘うための唯一の「正しい」政策はあるのだろうか?
  • Our analysis identified four types of exit strategies being used by nations in dealing with COVID-19.
  • 私たちの分析では、COVID-19に対処するために各国が使用している4つのタイプの出口戦略を特定した。
  • We found that a key factor accounting for the diversity of COVID-19 policies is the infection fatality rate (IFR), the proportion of deaths among all infected individuals. The IFR of COVID-19 is about 1%, much lower than the reported case fatality rate (CFR), which is about 5.5%.
  •  COVID-19の方針の多様性を説明する重要な要因が、感染致死率(IFR:全感染者中に対する死亡者の割合)であることが明らかになった。COVID-19のIFRは約1%であり、症例致死率(CFR:感染の報告されている人数に対する死亡者の割合)約5.5%よりもはるかに下回っている。(訳注:CFRは通常、単に致死率あるいは致命率と訳されることが多いが、ここでは症例致死率と訳す)

The world has now been living for months with the COVID-19 pandemic. While the number of new cases continues to rise in some nations — Brazil, India and Mexico, for example — a growing number of countries have already passed their first peak of the outbreak and are in a phase of containment. In the terminology of “The Hammer and the Dance,” the construction popularized by Tomas Pueyo in a viral Medium post[34] about the pandemic, these countries have moved beyond the Hammer phase — suppressing the disease’s rapid spread through lockdowns and other high-impact social restrictions. Yet a close look finds that not all of these countries have entered the Dance, the extended period of living with and adjusting to the disease’s constant though diminished impact.

Pueyo, T. (2020, March 20). Coronavirus: The Hammer and the Dance. Medium, Retrieved from https://medium.com/@tomaspueyo/coronavirus-the-hammer-and-the-dance-be9337092b56

Indeed, it is increasingly apparent that there are many different hammers and many different dances, depending largely on policy choices made by individual governments. Italy and Spain, for example, have seen prolonged(長引かせる) Hammer phases as the number of new infections has fallen only slowly. On the other hand, countries such as China and South Korea have kept the number of new cases down to near zero and are now well into the Dance. Then there’s Sweden, which eschewed(避ける) the Hammer of lockdowns and has tried to move directly to the Dance, pinning its hopes(ピン留め; 期待を寄せる) on a more rapid achievement of herd immunity while avoiding the massive economic damage inflicted(押し付ける) by Hammer policies.
Who is right, and who is wrong? Why have, say, East Asian countries fared(やっていく) better than the United States, Brazil, Sweden? Why do we see such a variety of policies around the world? This paper hypothesizes some explanations. It suggests that the infection fatality rate (IFR) of COVID-19 is much lower than the reported case fatality rate (CFR) based on testing. This is, obviously, good news. But it also means countries have been more likely to adopt policies consistent with their own historical and cultural experiences that may or may not be the most efficient or effective ways to combat the disease.
Four Types of Exit Strategies After Outbreak
In examining the policy responses of countries that have succeeded in suppressing the outbreak, we have come to understand four different types of exit strategies. These four categories are:
  • The Hammer and the Dance (Zero)
  • The Hammer and the Dance (Glider)
  • The Hammer and the Hammer
  • Herd Immunity
  •  ハンマー&ダンス(ゼロ)
  •  ハンマー&ダンス(グライダー)
  •  ハンマー&ハンマー
  • 集団免疫

So, to again pose our earlier question, who is right and who is wrong? As it turns out, the world may provide multiple answers. Let’s take a closer look at the characteristics of each type.
Type1: Hammer and Dance (Zero)(タイプ1 ハンマー&ダンス(ゼロ))
Although they differ in their methods, countries and regions such as China, South Korea, Taiwan, Japan, Australia, New Zealand and Norway have all succeeded in controlling the number of new cases and their Rt (Effective Reproduction Number) — the average number of secondary cases per infectious case in a population, which shows the rate at which the pandemic is spreading — have now reached a level signifying no significant increase in the number of infections. Since the number of new infections during the Dance period is near zero, we named this scenario “The Hammer and the Dance (Zero)”.

Countries that have kept the number of Rt firmly under one and have few new infections will react to new cases by isolating the infected person along with those who had close contact with him or her. Some countries, such as China and South Korea, will depend on monitoring and behavioral tracking to identify those at risk. Others, like Japan, may depend more on voluntary actions (known there as “Jishuku”).
In any event, this means that — in the absence of major new outbreaks — most people will be able to return to their daily lives. A little ingenuity may be needed in daily life, but it’s basically possible to hope for a speedy revival of consumption and a sharp V-shaped recovery in economic activity. In fact, the purchasing managers’ index of China’s manufacturing and services has already shown a V-shaped recovery.


Type2: Hammer and Dance (Glider)(タイプ2 ハンマー&ダンス(グライダー))
Some countries, mainly in Europe, have been reducing the number of new infections but have not yet succeeded in completely suppressing them. We term this type of nation “The Hammer and the Dance (Glider)” because the graph of new infections during the Dance period glides downward, rather than plunging.
Because economic activity cannot be halted indefinitely, these countries need to implement policies to ensure that the number of new infections does not explode as lockdown restrictions are eased(緩める). Because the risk of such an explosion remains as high as it is, a complete return to their previous lives for residents of these nations won’t be possible for the foreseeable(しばらくの間) future. Instead, they face the prospect of a “New Normal” radically different from their pre-pandemic lives.
The result for these nations is the prospect that economic activity will remain at a sluggish(不景気な, のろのろした) level, an L-shape rather than a V-shape, for an extended period of time.
Type3: Hammer and Hammer(タイプ3 ハンマー&ハンマー)
Countries such as the United States and Brazil[7] do not have uniform nationwide policies. Instead, various federal and state authorities have implemented individual policies, resulting in a patchwork approach and infection situations that vary region by region within each nation.The chart below shows the number of new infections and Rt in various U.S. states, with some already showing a glider pattern such as the Type 2 nations above, while other states are still seeing an increase in new infections.

The spread of COVID-19 remained largely unchecked(歯止めが効かない) in 24 U.S. states, according to a study by the Imperial College London published on May 24. Even so, all 50 states have already begun resuming economic activity, as shown in The New York Times below. This could lead to a rise in the reproduction number and the spread of the infection.

The New York Times. (2020). See How All 50 States Are Reopening, Retrieved May 28, 2020, from https://www.nytimes.com/interactive/2020/us/states-reopen-map-coronavirus.html
The combination of patchwork policies and premature reopening raises the prospect of repeated lockdowns whenever there is an explosive increase in the number of new infections. We have named this scenario “The Hammer and the Hammer.”
パッチワーク政策と時期尚早の経済活動再開の組み合わせは、新規感染が爆発的に増加するたびにロックダウンが繰り返される可能性を高める。私たちはこのシナリオを "ハンマー&ハンマー と名付けた。
Under this scenario, economic recovery can be expected, albeit gradually, as economic activity resumes(再開する). We expect this type of economic recovery to be a U-shape, not a V-shape or an L-shape. In other words, different states will recover their economies differently, and the overall national numbers can be expected to show a gradual recovery after a period of stagnation(停滞). Nevertheless, to at least minimize the prospect of explosions of new infections, a “New Normal” behavioral pattern will be required, as with Type 2.
Type4: Herd Immunity(タイプ4 集団免疫)
Finally, there is the Type 4 model, in which the nation aims to acquire herd immunity. Although experts differ on whether herd immunity will be possible in the case of COVID-19, Sweden’s policy is believed to be predicated on the notion that it is.
Herd immunity is a theory that if enough people in a population acquire immunity to a virus through natural infection or vaccination, it will prevent the spread of infection to non-infected people who do not have immunity. The percentage of people who need to be immune to achieve herd immunity depends on the basic reproduction number of the virus. In the case of COVID-19, a rate of 60–70% has been proposed, but some studies, including a study[16] in the United Kingdom on April 27 that focused on individual heterogeneity, suggest that the actual rate might be 20–40%. By way of comparison, Type 1 countries have seen no more than 1% of the population infected. Even in the United States, the most infected country, estimates are that only about 4% of the population is infected.
These graphs show the number of new infections and Rt in Sweden:.

When seeking(求めるものを得ようと努力する) to acquire herd immunity through natural infection, certain policies were implemented to prevent a sudden explosion of infection, but essentially daily economic activities have been carried out as normal. Even today, Sweden still has no active lockdown policy in place(実施されて), although public gatherings of more than 50 people and the use of counters in bars are prohibited. We’ll post later on why Sweden could adopt such a policy.
This illustration shows the relationship between economic damage and medical/health damage for each of the four types described above:

There are many different responses to COVID-19, and we’ve seen a number of “life or economy” debates. What the world is looking for now is an answer to the question, “Which is the right response?” Our answer is that this turns out to be a multiple-choice question.
Why Multiple Answers?(なぜ複数の答えがあるのか?)
Why can’t we have a single answer to this most critical of questions? Because of the nature of COVID-19’s fatality rate; and because of the cultural and historical differences among nations.
1.Estimating COVID-19’s Infection Fatality Rate1.COVID-19の感染致死率の推定)
 The Japanese physicist Torahiko Terada once said, “It’s easy to be too scared of something, or not scared enough, but it’s hard to be scared just the right amount.” Amid the daily welter of news and information about the number of infections and deaths, prevention methods and drug development, we have focused on the question of how lethal COVID-19 really is. Our conclusion helps explain the wide variation of national policy responses to the pandemic. Were the death rate much higher, governments would have no option but the most draconian(過酷な) of policies: quarantines, mandatory testing, lockdowns, the cessation of most economic activity. Were the death rate much lower, most countries would simply go about their business with few or no policy concessions(譲歩, 許容)  to the disease. The actual COVID-19 infection mortality rate is high enough to require a policy response, but not so high as to dictate exactly what that response should be.
The fatality rate we see in public reports is usually reported as a simple formula like this (which we call reported case fatality rate, or CFR):
公的な報告書に記載されている致死率は、通常、次のような単純な式で報告されている(私たちはこれを症例致死率 reported case fatality rate、あるいは CRF と呼んでいる ※訳注:通常、単に致死率あるいは致命率と訳される)。
Case fatality rate (CFR) = Number of deaths after confirmed COVID-19 infections / Number of confirmed COVID-19 infections
As of the first part of May, here were the reported case fatality rates for various affected areas:

However, since the number of confirmed COVID-19 infections can vary depending on the number of tests performed, and since the number of tests performed varies from country to country, the reported case fatality rates (CFR) may not reflect the real impact of the disease. Here we choose to use the infection fatality rate (IFR), calculated by the following formula:
Infection fatality rate (IFR) = actual deaths caused by COVID-19 / actual number of people infected with COVID-19
Many existing studies have already discussed and estimated the likely infection fatality rate (IFR) of COVID-19[27][33]. There are several types of methodologies:
  1. Estimate IFR based on a sample from a specific scene/region where tests are conducted thoroughly (e.g. the Diamond Princess), and infer the IFR of a larger population[29][39][44]
  2. Estimate the number of actual cases based on antibody tests results[37][45]
  3. Use statistical modeling to infer the IFR based on existing testing data (from multiple regions or time-series based)[4][38][53]
  1. 試験が徹底して行われている特定のシーン/地域のサンプル(例:ダイヤモンド・プリンセス)からIFRを推定し、より大きな母集団のIFRを推論する[29][39][44]
  2. 抗体検査の結果から実際の感染者数を推定する[37][45]
  3. 統計的モデルを用いて、(複数の地域または時系列から得られた)既存の検査データに基づいて IFR を推論する[4][38][53]
One meta-analysis, examining the conclusions of more than 250 papers, has estimated that the IFR of COVID-19 is around 0.5%-1.5% [27][33].
While it’s difficult to calculate completely accurate figures for either deaths or infections, these studies may bring us closer to the actual infection fatality rate (IFR). In order to make an inter-regional comparison of infection fatality rates (IFRs), we chose to calculate approximate numbers by looking closely at “excess deaths” and “positive antibody test rates,” which is similar to the second methodology.
“Excess deaths” is a measurement, first proposed by the WHO during influenza epidemics, that compares the number of deaths during the outbreak with the average number of deaths during comparable periods without such outbreaks. It’s a way of including people whose deaths might have occurred at home without a confirmed COVID-19 diagnosis[54]. The Financial Times published a much-noted investigative report[14] on excess deaths on April 27, and we also obtained data on excess deaths in each region, using them to estimate the actual number of COVID-19 deaths, as opposed to the lower, “official” figures.
We also collected and analyzed the data for positive antibody test rates in each region. An antibody test shows whether a person has been infected with COVID-19 in the past, and thus can be used to estimate how many people in a region have already been infected with COVID-19 even if they never showed symptoms. (It should be noted that the accuracy of the COVID-19 antibody test may vary slightly depending on the quality of the kit used.)
Since we were able to estimate the denominator of the above formula this way, the COVID-19 infection fatality rates (IFRs) we estimated for each region are as follow: