Random Analytics: H7N9 Infographic (to 27 Apr 2013)
by Shane Granger
***** Please note that this infographic of the Avian Influenza A(H7N9) virus was updated with public source information from late 27 Apr to early 28 April 2013 CET/EST *****
Sometime ago I was watching a documentary series on the Bikini atoll nuclear testing which included the 1 March 1954 Castle Bravo hydrogen bomb test. Due to some ‘unknown unknowns’ at the time the test was more powerful than predicted and contaminated huge amounts of the Pacific, including by chance the crew of the Diago Fukuryū Maru (Lucky Dragon 5), a Japanese fishing vessel some 40-miles (64-kilometres) away.
Now the radioactive winds were not supposed to travel in that direction, nor was the explosion predicted to be as large as it was. In some respects, the 23-crew of the Lucky Dragon 5 were just in the wrong place at the wrong time.
Infographic Details
In the past 24-hours there has been confirmation of 2 new cases. No new fatalities have been reported. This brings the total for China to 120-cases including 23-deaths and Taiwan to 1-case without loss of life. (It should be noted that I do not include the asymptomatic Beijing case in my count).
To date 19% of all known cases have been fatal. For context the Case Fatality Rate of SARS was 10.9%.
There were six confirmed discharges in the past 24-hours, all from Zhejiang which takes the confirmed recoveries to 20 (16.8%). Note: All known paediatrics and children have recovered.
The most recent fatality reported was on the 26th April 2013 via Jiangsu Health (in Chinese).
Random Chance
If the confirmed infection area, made up of two municipalities, eight provinces and Taiwan were a country it would span a territory roughly 1.227-million km², have a population of 596.1-million and a nominal Gross Domestic Product of $5.1-trillion (USD, 2011). For context that would make it the 25th largest country for land mass (slightly larger than South Africa), the third most populous country (almost double that of the United States) with the 4th highest GDP (approximately $600Bn less than Japan) and qualify it for membership into the G20.
Yet for the massive amount of territory H7N9 inhabits it has still only made 121 people sick enough to seek treatment and killed 23-persons. Even in Zhejiang, the province with the most confirmed cases you only have a 1:1,200,000 (approximate) chance of catching the disease!
Like the fishermen on the Lucky Dragon 5, currently to get the disease (and then to die of it) you need to extremely unlucky.
One thing about statistics though, they are just that until you become a victim.
Random Analytica 
This is an interesting statistical analysis. My question is this: has your system considered the math regarding the very real fears about what H7N9 is doing while we are unable to detect or control it? Influenza, biological speaking, is very different from many other bugs or natural and man-made disasters. Every time we don’t control it, it has another opportunity to mix in a way that very quickly becomes problematic for us. One in a million chances is very small…until you include math to reflect that this virus has the chance to mutate per every encounter, not every clinically identifiable disease. Every encounter. And if we can’t see it, we can’t interrupt the chain of encounters. It may peter out on its own – there are plenty of examples of times that flu or other bugs have done that. If it doesn’t, we have a problem with 1 in 5 people who get sick dying – the WHO pandemic plan forecasts additional economic and political consequences. I do think it is worth noting how the infamous Gain of Function research was done – passing the virus through ferrets until it became transmissible between the ferrets (the closest analogues for humans). We don’t know exactly what evolutionary pressure the virus is under right now, but we do know that when you pass it through another being, it has another change to make the necessary adjustments, whether or not the being becomes ill. And that is the actual statistical problem. I would be very interested in seeing if you could analyze that problem, even if you had to use information about a different strain of flu to grapple with it.
Mageen,
Many thanks for your kind comments, much appreciated.
Short answer is no, but its a moving feast (and you never know). In the meantime if you check out two blogs you may be able to dig something out. Virology Down Under (Dr. Mackay) and H5N1 (Crawford Kilian) are excellent sources of technical input on H7N9.
Hope that helps.
Shane