The Wuhan Coronavirus - Misinformation
The error in the Guardian article is related to a big problem that's been occurring since the beginning of the coronavirus epidemic; the repeated downplaying of the dangers of the disease by our mainstream media. This has been going on even though several scientific papers have been coming out that describe how the virus actually operates. I've already talked about this problem in an earlier blog article but I guess I have to mention it again. In Adam Kucharski's article, he states:
'In disease outbreak analysis, we can measure the transmission of an infection by looking at how many additional cases each infected person creates on average during each of these steps. We call this the “reproduction number”, and for coronavirus, we estimate it’s about 2 for a typical infected case in China.'
Mr Kucharski does not supply any scientific references to back up this claim. He may be focussing the article on misinformation 'epidemics' on popular media but there is much evidence that this value of around 2 simply isn't correct. It's a very surprising thing to do by an 'Associate Professor and Sir Henry Dale Fellow in the Department of Infectious Disease Epidemiology at the London School of Hygiene & Tropical Medicine'. It makes no sense for Mr Kucharski to criticise others in his article for sloppy research and produce a key number out of nowhere. To show how important the reproductive number is for a disease, here is a short video from Scientific American:
In my last blog article, I talked at length about another epidemiologist, Professor Neil Ferguson. He is a senior epidemiologist at Imperial College. A week ago, roughly, he estimated a reproduction ratio of 2.5 to 3 on the 2nd February, mentioned in this blog article. Other groups estimate even higher reproductive numbers. Here is a Hong Kong scientific paper studying the spread of the disease (pdf) and here is their findings paragraph, part of the paper's abstract:
'The early outbreak data largely follows the exponential growth. We estimated that the mean R0 ranges from 3.30 (95%CI: 2.73-3.96) to 5.47 (95%CI: 4.16-7.10) associated with 0-fold to 2-fold increase in the reporting rate. With rising report rate, the mean R0 is likely to be below 5 but above 3.'
This paper backs up another scientific study (pdf), involving the Department of Biology and Emerging Pathogens Institute, University of Florida and the Medical Research Council-University of Glasgow Centre for Virus Research, that estimates the reproductive rate to be between 3.6 and 4. It would seem that Mr Kucharski's number is at the very low end of recent estimates for the Wuhan coronavirus reproductive number.
One way to validate the reproductive number is to examine the coronavirus epidemic's doubling rate. In other words, how long it takes for the number of cases to double. Professor Ferguson has estimated the Wuhan coronavirus doubling rate to be five days. It is longer than my earlier estimate of three days, but that's okay; I'm very happy to bow to his superior knowledge and methods. Five-day-doubling is still very serious; it describes a fast-spreading disease. Such a doubling rate would fit with a reproductive rate of 4, or even 5. Is this the case? Are the Wuhan coronavirus infections doubling in number every five days, as Professor Ferguson states? It is very difficult to use mainland Chinese data to check this theory as it's highly likely that the Chinese numbers have not been accurate since mid-January. This isn't necessarily a criticism of the Chinese government. Once the Wuhan hospitals became overloaded with patients, they wouldn't have been able to test any more people. Instead, we can gather numbers from infections outside China to check the doubling rate.
Five days ago, in this earlier post, I included the John Hopkins live update for that day. Here is today's live update:
As we can see, there are now 33 infected cases in Singapore, which had 18 five days ago. Malaysia had 8, five days ago; it now has 16. These are good matches to Professor Ferguson's estimated doubling rate. It's worth keeping in mind that Professor Ferguson would almost certainly comment, at this point, that the epidemic's doubling rate will vary greatly depending on the health measures in place. Singapore has far more extensive health resources per person than, say, Myanmar, Indonesia, the Philippines, etc. Singapore's doubling rate should therefore be lower than those countries. Paradoxically, Singapore's better medical resources means that it will be more accurately reporting infections and so they would be a higher rate of detection than another country, skewing the data. Such skewing can make more attentive countries look to be having a worse time than less attentive countries.
All in all, Professor Ferguson's estimate of doubling-time does seem to be matching the reliable reports very well. It therefore makes sense to accept his reproduction rate of 2.5 to 3. Alternatively, we could conclude that the reported infections is only a small fraction of the actual infections, due to people incubating the disease without showing symptoms. This would make a reproductive rate of 4 to 5 seem likely, as stated in the earlier scientific reports. As far as we can tell, according to reports of carriers, a single person can pass on the disease to several people while still in the incubation phase, never mind when they show symptoms. What's more, according to a German patient, they can even be infectious after the symptoms have subsided. All this information, taken together, makes a reproductive rate of only 2 extremely unlikely.
From the beginning of this epidemic, I have been shocked by the efforts of the mainstream press to downplay this epidemic. I am guessing they are doing it to avoid appearing alarmist, or possibly to protect the global economy, or both, but that's only a guess. Misinformation is a serious problem in any epidemic; it's such a shame that much of it is coming from the media sources that we're supposed to regard as reliable.