The Black Death. The 1918 Flu Pandemic. COVID-19.
黑死病。 1918 年流感大流行。 2019 冠状病毒病 (COVID-19)。
We tend to think of these catastrophic, world-changing pandemics as very unlikely events.
人们往往觉得这些改变世界的 灾难性疫情是极小概率事件。
But between 1980 and 2020, at least three diseases emerged that caused global pandemics. COVID-19, yes, but also the 2009 swine flu and HIV/AIDS.
但是 1980 至 2020 年间, 至少出现了三次 造成全球性疫情的疾病。 没错,有 COVID-19, 还有 2009 年猪流感以及 艾滋病毒/艾滋病。
Disease outbreaks are surprisingly common. Over the past four centuries, the longest stretch of time without a documented outbreak that killed at least 10,000 people was just four years.
疾病爆发普遍得令人吃惊。 在过去的四个世纪里, 没有任何致死量 超过一万人的疾病爆发记录 最长时间间隔仅为四年。
As bad as these smaller outbreaks are, they’re far less deadly than a COVID-19-level pandemic. In fact, many people born after the 1918 flu lived their entire lives without experiencing a similar world-changing pandemic. What’s the probability that you do, too?
虽然这些小规模的疾病爆发很糟糕, 但是它们的致死率远低于 COVID-19 级别的疫情。 事实上,很多出生在 1918 年流感大流行以后的人, 一生中都没遇到过类似级别的 足以改变世界程度的的疫情。 那你遇到的概率是多大呢?
There are several ways to answer this question.
可以从好几个角度回答这个问题。
You could look at history. A team of scientists and engineers who took this approach catalogued all documented epidemics and pandemics between 1600 and 1950. They used that data to do two things. First, to graph the likelihood that an outbreak of any size pops up somewhere in the world over a set period of time. And second, to estimate the likelihood that that outbreak would get large enough to kill a certain percentage of the world's population. This graph shows that while huge pandemics are unlikely, they're not that unlikely. The team used these two distributions to estimate that the risk of a COVID-19-level pandemic is about 0.5% per year, and could be as high as 1.4% if new diseases emerge more frequently in the future.
可以回顾历史。 一组采用了该方法的 科学家与工程师 将 1600 年至 1950 年间所有 被记录的流行病与疫情分类统计。 他们用收集的数据做了两件事。 首先,作图呈现不限规模的疾病 在一定时间段内 在世界上某地爆发的可能性。 第二,预测疾病爆发会扩大到 导致一定比例的地球人口 死亡的可能性。 从这幅图可以看出, 虽然大规模疫情是小概率事件, 但是可能性其实也没有那么小。 研究团队用上述方法估算 COVID-19 级别的疫情 发生的风险概率大约为 每年 0.5% , 如果新型疾病在未来 出现得愈加频繁, 那风险概率可高达 1.4% 。
And we’ll come back to those numbers, but first, let’s look at another way to estimate the likelihood of a future pandemic: modeling one from the ground up.
我们之后会再回来讨论这些数字。 现在,我们来看估测 未来疫情可能性的另一种方法: 从底层开始建立模型。
For most pandemics to happen, a pathogen, which is a microbe that can cause disease, has to spill over from its normal host by making contact with and infecting a human. Then, the pathogen has to spread widely, crossing international boundaries and infecting lots of people. Many variables determine whether a given spillover event becomes a pandemic. For example, the type of pathogen, how often humans come into close contact with its animal reservoir, existing immunity, and so on.
对于大部分疫情来说, 发生的必要条件是 病原体,也就是 一种可以导致疾病的微生物, 需要接触人体并且感染人类, 以达到从正常宿主溢出的目的。 而后,病原体需要广泛蔓延, 穿过国界与地域的限制, 感染很多人。 很多变量决定了一个特定的溢出事件 是否最终会蔓延成为疫情。 例如,病原体种类, 人类与动物宿主 近距离接触的频繁程度, 现有的免疫力,等等。
Viruses are prime candidates to cause the next big pandemic. Scientists estimate that there are about 1.7 million as-yet-undiscovered viruses that currently infect mammals and birds, and that roughly 40% of these have the potential to spill over and infect humans.
病毒是导致下一次大疫情的 潜在罪魁祸首。 科学家预计有 1700 万 尚未被发现的病毒 正在感染着哺乳动物和鸟类, 这其中的 40% 有着 病毒溢出并感染人体的潜力。
A team of scientists built a model using this information, as well as data about the global population, air travel networks, how people move around in communities, country preparedness levels, and how people might respond to pandemics. The model generated hundreds of thousands of virtual pandemics. The scientists then used this catalog to estimate that the probability of another COVID-19-level pandemic is 2.5 to 3.3% per year.
科学家团队进行建模分析, 输入模型的信息为病毒溢出概率 以及以下数据: 全球人口,航空旅行网络, 人们在社区内的移动情况, 国家防灾备灾水平, 人们针对疫情采取的应对措施。 模型生成了成百上千种模拟疫情。 科学家用这个样本估测 发生另一次 COVID-19 级别 疫情的概率为 每年 2.5% 至 3.3%。
To get a sense of how these risks play out over a lifetime, let’s pick a value roughly in the middle of all these estimates: 2%. Now let’s build what’s called a probability tree diagram to model all possible scenarios. The first branch of the tree represents the first year: there’s a 2% probability of experiencing a COVID-19-level pandemic, which means there’s a 98% probability of not experiencing one. Second branch, same thing, Third branch, same. And so on, 72 more times. There is only one path that results in a fully pandemic-free lifetime: 98%, or 0.98, multiplied by itself 75 times, which comes out to roughly 22%. So the likelihood of living through at least one more COVID 19-level-pandemic in the next 75 years is 100 minus 22%, or 78%.
为了对这些概率在我们的一生中 会产生何种影响有更直观的认识, 我们取一个这些概率 大约的中间值: 2% 。 现在画概率树状图 来模拟所有可能的情形。 第一个分支代表第一年: COVID-19 级别的疫情 有 2% 的可能性发生, 也就是说 有 98% 的可能性不会发生。 第二个分叉,情况相同。 第三个分叉,也是一样的。 如此再继续 72 次。 只有一条路径通向 完全没有疫情的一生: 98%, 或者是 0.98, 乘以它自己 75 次, 结果大约为 22%。 所以在接下来的 75 年,遇到 至少一次 COVID-19 级别疫情的 概率为 100% 减去 22%, 或者是 78% 。
78%!
78% !
If we use the most optimistic yearly estimate— 0.5%— the lifetime probability drops to 31%. If we use the most pessimistic one, it jumps to 92%.
如果我们使用最乐观的 全年估计 —— 0.5% —— 那么一生中的可能性 降到了 31% 。 如果我们使用最悲观的全年估计, 那结果大幅增长至 92% 。
Even 31% is too high to ignore; even if we get lucky, future generations might not. Also, pandemics are usually random, independent events: so even if the yearly probability of a COVID-19-level pandemic is 1%, we could absolutely get another one in ten years.
即使是 31% , 也大到了我们不能忽视的程度。 即使我们够幸运, 后代们未必足够幸运。 况且,疫情通常是 随机且独立的事件: 所以即使 COVID-19 级别疫情的 全年可能性概率为 1% , 我们在未来的十年内 也很有可能会再遇到一次。
The good news is we now have tools that make pandemics less destructive. Scientists estimated that early warning systems, contact tracing, social distancing, and other public health measures saved over a million lives in just the first six months of the COVID-19 pandemic in the US, not to mention the millions of lives saved by vaccines.
好消息是我们现在已经拥有工具 使疫情带来的伤害没那么严重。 科学家们估计 早期预警系统、接触者追踪、 保持社交距离 以及其他公共卫生措施 仅仅在美国 COVID-19 疫情 发生的前六个月, 就拯救了超过一百万人, 更不要说,还有疫苗 拯救了数百万条生命。
One day, another pandemic will sweep the globe. But we can work to make that day less likely to be tomorrow. We can reduce the risk of spillover events, and we can contain spillovers that do happen so they don’t become full-blown pandemics. Imagine how the future might look if we interacted with the animal world more carefully, and if we had well-funded, open-access global disease monitoring programs, AI-powered contact tracing and isolation measures, universal vaccines, next-generation antiviral drugs, and other tech we haven't even thought of.
有朝一日,另一场疫情 会再次席卷全球。 但我们可以努力让那一天不是明天。 我们可以降低病毒溢出事件 发生的风险, 即使发生了病毒溢出, 我们也能有效遏制, 使其不会蔓延成为全面爆发的疫情。 想象一个这样的未来, 人类与动物世界 接触得更加小心谨慎, 我们拥有资金充足的、开放式的 全球疾病监控项目, 人工智能支持的 接触者追踪和隔离措施, 通用的疫苗, 下一代抗病毒药物, 以及其他我们想象不到的科技。
It’s in our power to change these probabilities. So, we have a choice: we could do nothing and hope we get lucky. Or we could take the threat seriously enough that it becomes a self-defeating prophecy.
相信我们有 改变这些可能性的力量。 所以我们要做出选择:什么都不做, 祈祷我们足够幸运。 或者足够重视这些威胁, 让它们都成为不攻自破的预言。
Which future would you rather live in?
你更想活在什么样的未来里?