立場新聞 Stand News

Wuhan pneumonia in context – why we need not overreact

2020/2/4 — 11:13

編按:作者認為,今次武漢肺炎疫情引起全球恐慌,從坊間的一些反應來看,似乎是有點過了頭。從數字上看,武漢肺炎的人口死亡率 (population death rate) 與流感相若,甚至較低,而一般來說,隨著醫療界對疫症的了解加深,死亡率有望進一步降低。同時,正因為全民都嚴陣以待,將大大有助控制疫情擴散。不過,他指出今次疫情於年初爆發,距離天氣回暖尚有一段日子,故疫症延展的時期可能比沙士長。作者認為,在全民防疫的心態下,相信未來數星期,感染數字有望回落。

《立場》編輯補充,目前仍需進一步研究,包括對相關新型冠狀病毒的源頭及擴散的追蹤,才能對武漢肺炎疫情有較全面的了解。

The following is an updated version of the author’s original article on his blog, linkedin, and fb pages:

There seems a universal (well at least in HK) panic about the Wuhan pneumonia outbreak, with at least 95% of the people we see in the streets wearing masks – a spectacle – and many organisations completely shutting down (including the government) for fear of contagion…

廣告

This is despite to date only 15 people in HK have been confirmed cases. The status at the time of writing looks like this:

廣告

If you look at the numbers, however, the situation is perhaps much less frightening than the media (and officials) have whipped up:

1. Death rate similar/weaker than normal flu? As table below shows, the death rate of diagnosed cases in the epicentre, Wuhan, may be high at 5.2%, but by the time you look at the province at large, it drops to 3.1%, and overall China-wide, the ratio becomes a low 2.1%. HK’s death rate was a lucky big Zero:

Table 1: WARS statistics

Table 1: WARS statistics

Compared to other headline flu pandemics reported (see FR1 below), where death rate range around 0.03-0.08% of population, the current rate of death, even in Wuhan, is a tiny fraction (see right column in Table 1, at 0.0024%) – ie there needs a 100x increase in deaths from current Hubei levels (0.0007%) to match past pandemics alone – ie. 16k–49k in casualties in total. Luckily, the Hubei death toll is a low 414 as of today;

2. The death rate have been dropping usually the early deaths tend to be more heavily reported/collected as a statistic compared to the actual infected population, while the catching up by the latter leads to a more normalised, actual rate in time (red line below). Public awareness may also put a stop to accelerated growth in death rates, which the blue line below could well be suggesting, falling from 3% to early 2% in the past few days. A consistent departure from the red trajectory will mean we can largely return to normal activities:

 

Chart 1: death rate tends towards the disease’s true virility in time

Chart 1: death rate tends towards the disease’s true virility in time

3. Vigilance is ultra-high this time compared to SARS (when prevention only scaled up after several weeks of casualties), people are armed to the teeth within days, thanks to the magnifying effect of social media and international measures (which are unprecedented – chartered flights to retrieve nationals from China, for example). As a result, the spreading of the disease will likely be much less rapid. Here is a map showing which country to be in with  any epidemic outbreak. Not surprisingly, OECD countries come top:

Figure 2: which countries are best prepared for an outbreak?

Figure 2: which countries are best prepared for an outbreak?

4. Possible longer drag than SARS One feature that is not talked about is the timing of the current outbreak – instead of a March incident like in SARS, we are in January, that means the period of time before it becomes too warm for virus to spread diseases will be longer than during the SARS episode, which ended in June 2003 as temperature rose.
This time round, the outbreak will potentially be lengthened by two months. Hopefully the growth in case numbers (red line in Chart 2 below) will see a tailing off much faster this time compared to SARS (blue line) on increased vigilance.

Chart 2: Case growth rate may decay like the SARS outbreak

Chart 2: Case growth rate may decay like the SARS outbreak

5. Virility tracking will give the answer the viciousness of the current outbreak has yet to be proven, but if the trends discussed thus far hold, the situation stands good chance stabilising at current levels.

In the next few weeks we should have our answer, but if anyone even remembers the not too distant 2009 avian flu outbreak, which is so far 1000x higher than WARS in death rates (red dotted line, Chart 3), people seem much more ready, if not more than necessarily prepared this time round.

Chart 3: log chart comparing current death rates to past pandemics

Chart 3: log chart comparing current death rates to past pandemics

Reference – further reading 1 (FR1)

Flu causes more death than you may think!

Every flu season, a lot of people die from this disease, mainly from complications such as pneumonia

According to CDC, 140,000–810,000 people are hospitalized each year since 2010 in the United States for the flu illness and its complications and between 12,000 and 61,000 people die each year from the flu since 2010. The number of flu deaths every year varies. That is 1.5%-25% of hospitalised flu victims die each year!

According to the World Health Organization (WHO), about 3 to 5 million cases of severe flu illness and about 250 000 to 500 000 flu season deaths worldwide occur due to the influenza virus. That is anywhere between 5% to 17% of deaths.

Really big pandemics in context

The world has seen five pandemics during the last century, which took a large number of lives. Here are the figures of deaths that occurred in the United States and Worldwide during those years.

  1. 1889 Russian Flu Pandemic – about 1 million flu deaths (0.07% global population)
  2. “Spanish flu” A of 1918-19 caused the highest number of influenza-related deaths: approximately 500,000 deaths occurred in the U.S. and 20 million worldwide. That figure is more than the total number of deaths caused by the World War one — 16 million. As a matter of fact, during that year, the flu had killed more people than any other illness in recorded history. (1.1% global population)
  3. “Asian flu” A of 1957-58 caused 70,000 deaths in the United States and about one million to two million deaths worldwide (0.07% global population)
  4. “Hong-Kong flu” A of 1968-69 resulted in 34,000 deaths in the United States and an estimated one million to three million people died worldwide. (0.08% global population)
  5. 2009 H1N1 Flu Pandemic – about 18,300 deaths in the United States and up to 203,000 deaths worldwide (0.003% global population)

So amongst the most feared pandemics in history, only the Spanish flu was of truly frightful scale when 1.1% of population perished.

https://healthvigil.com/flu-season-deaths-us-worlswide/

Reference – further reading 2 (FR2)

Trends in Recorded Influenza Mortality: United States, 1900–2004, by Peter Doshi

In results section author concluded: “An overall and substantial decline in influenza-classed mortality was observed during the 20th century, from an average seasonal rate of 10.2 deaths per 100 000 population in the 1940s to 0.56 per 100 000 by the 1990s. The 1918–1919 pandemic stands out as an exceptional outlier.”

發表意見