Archive

Tag Archives: 2014 ebola epidemic

Hi all,

Today, 14 October 2014, we read of a report by WHO which states that by 1 December we could see as many as 10,000 deaths from Ebola per week. Oddly, the mainstream press finds this newsworthy, despite the fact that this very same mathematical statement was made by WHO and CDC weeks ago when the same numbers were provided in a different form. We are told that as of today, there are 4500 dead and 9000 infected. Going off of the average reported error factor of about 2, this means that there are more likely 9000 deaths and 18,000 infected as of now. There is a disturbing problem going on here which begs comment in the harshest terms. Magical thinking is undermining the ability not just of the public, but of the authorities, to comprehend this disaster on account of three salient features now evidenced incontrovertibly by the statements and actions such as the one just noted:

  1. Mathematical illiteracy
  2. Inability to assess the assumptions of the math
  3. Lack of realism; or inability to reduce the general to the specific

What about log base 2 of 18000 can some people not understand? And where people do get the significance of exponential growth, what about the core assumptions is not clear? The only assumptions behind this math are pretty dry:

  1. Nature will not do something incredibly improbable in the next 12 months to favorably alter the math.
  2. There is no infrastructure on Earth that can locate, extract and “isolate” 70% of a population of 36,000 in sub-Saharan Africa scattered in three different jurisdictions over uncounted thousands of square miles of African city and jungle. (I’ll explain that number shortly). And certainly 4000 Marines can’t do it.

WHO and CDC are counting on the magical thinking of isolating 70% of the infected population, based on the idea that the reproduction rate of the virus could be rendered linear under those conditions. But implicit in this thinking is the requirement that all those persons be identified and extracted from the general population, and then isolated not by merely housing and feeding them, but by bio-containment isolation. And if we have 18,000 infected now, give or take a few thousand if it pleases you – it doesn’t much matter – then no infrastructure having a viable mathematical impact on this situation exists now and will not likely exist within the doubling time of the viral spread. Therefore, let’s not be silly and let’s assume 2*18,000=36,000. It certainly isn’t going to happen with 2000 beds provided by the U.S. military.

I’ve been trying to explain this now for over two months (in various forms) and am convinced no one comprehends a thing I’m saying.

Let me be clear, the assumptions given above are highly likely to hold and the logarithmic relation given will dominate the outcome. What part of that sentence is confusing? Do I really need to work this logarithm to demonstrate what highly likely means? Try this:

log2(7*109) – log2(36000) = about 17.5 months

I’m assuming that the doubling is every 4 weeks, which is generous, and I’m assuming the world’s population is at least 7 billion. Yes, the doubling rate will increase as the total number of infected increases, but I’ll ignore that for the moment.

We have about 18 months before there won’t be anyone left to discuss this because the mortality rate WHO just released is another failure to reduce the general to the specific; namely, that when people start dying in large numbers the death rate will approach 90% or greater consequent to the sheer anarchy and chaos that will result from the viral mortality of 70%, not to mention the lack of food for anyone remaining.

Let me be clear one more time:

This reduces the problem to two options. We can do one or both. We can develop a real bang up vaccine really, really fast and/or we can isolate populations with force. Of course, the West can relax since their superior health care systems might prevent the assumed “seeding” numbers in the thousands, like we have now in West Africa. In that case, the West can expect to be alone when the dust settles … and to be a little hungry. Hope they’re happy with that. But no, there is no need to panic in the West because clusters will indeed likely be snuffed out. But I hope we can see why …

That doesn’t much *&^% matter.

And speaking of the West, CDC’s recommendations to hospitals are a circus of failures to reduce the general to the specific; a process otherwise known as deduction. “Meticulous” guidelines cannot be followed stochastically in a general hospital environment when those hospitals are using BSL-2. Someone in USG with a brain needs to implement BSL-4 in regional hospitals … right now.

I think nature is about to “inform” us as to just how dumb we really are and future observers will quite likely regard many actions already taken or omitted as criminal negligence of the highest order.

Over the past couple of years I’ve been trying to get the message out that religion and political ideology are vehicles of misplaced emotion that undermine IQ and are squeezing humanity into destruction in a death spiral of ignorance and superstition. I’m afraid my message won’t be heard until billions die.

– kk

Advertisements

The following is a thought experiment and it should be taken seriously, but not as an inevitability either. Nature can upset these rates of viral diffusion in any number of ways and there are too many variables to know if this estimate is valid.

If the number of infected persons should exceed 100,000, then it is unlikely that any international effort will have sufficient resources to make any mathematically non-negligible difference in the rate of diffusion of the virus. At that time, the only means of stopping the diffusion of the virus will by external quarantine.

Will we see that number? If we assume that the total number of infected is three times the estimated amount, as some close to the ground have stated, and the reported figure is 5347 as of 18 September, then the likely more accurate count is 3*5347=16041 persons as of 18 September. Therefore, 100,000 infected persons is notionally reached on the x’th doubling:

105 = 16041*2x

Solving for x and multiplying by the observed doubling time we get:

[log2 (105) – log2 (16041)] * 3 = 7.92 weeks

Defeating this trend with 3000 military personnel and several tons of equipment is improbable in that time. In fact, making any appreciable difference in this rate is improbable when we consider the fact that it will take at least 30 days for this effort to fully stand up. If we cannot defeat this function in this time, then the point in time that the probability of containing the virus is maximum is when that point is reached; that is, in 7.92 weeks. USG should be prepared for that time, which is about 18 November.

Here’s why.

At this time USG should (and must if it is to succeed as pointed out) shift to a quarantine solution. At this time, the total number of infected is 100,000. While this may be too many to contain the virus within the population, it is small enough to contain it within the borders of the three affected countries, assuming steps are taken to do so. But simply putting troops on the border may not be enough. A no-fly zone will be required and all arteries of passage outside these countries must be involuntarily (but temporarily) evacuated with a radius of not less than 15 miles from the point where the artery intersects the border (the actual distance would vary depending on the size of the passage and the topology). Refugees should be sent to either Liberia, Sierra Leone or Guinea. And authorities will have to “guard” this area and allow no entry. This is the only solution in the absence of a vaccine. USG should begin now to collaborate with all neighboring countries to give U.S. military personnel access to these “choke points”. USN must enforce a no-sail zone off the coasts of Liberia and Sierra Leone on 18 November.

U.S. military forces, if the numbers exceed 3,000, should be issued all chemical, biological and radiological gear (and I do mean ALL three types) to all personnel. Other nations that supply military personnel should do likewise. Prioritization of vaccination, wherever available, should be ranked thusly:

1. Military personnel and health professionals in the hot zone

2. Gibraltar, Sinai and Panama (my concern is that governments will continue to underestimate the scale of human migratory flow, and how this will cause a wave of successive virus infection flow along migration paths, which will converge on these “ambush sites”; in the sense that these sites, if treated wisely, could serve as points to “ambush” the virus).

3. Populations at the periphery of the hot zone

4. The hot zone

5. The global population, en masse.

If the number of infected increases the probability of containment will fall. When the number of infected in the affected countries reaches 1 million, containment will likely fail in any case. 1 million infected will be reached on:

[log2 (106) – log2 (16041)] * 3 about 4.5 months, or January 15.

At that time, should that occur, USG should apply the same tactic at Gibraltar and the Sinai and the no-fly and no-sail zone should be extended to the continent of Africa. If a vaccine is available, a mass and heavy vaccination of the population in the areas of Gibraltar and Sinai should be performed in a buffer area extending about 100 miles on either side of this boundary. The vaccine should be made available to any person seeking it, regardless of origin and indeed, it should be mandatory (if a refugee arrives in a heavily vaccinated area they are less likely to attempt to continue their exodus). If a vaccine is not available in that quantity, the same area should be evacuated. In either case, the area will require U.S. troops to deny passage to all persons. They should also deny entry into the vaccinated zone. Once this is accomplished, and all available assets should be employed to do so, similar checkpoints should be established at successive national borders from the outbreak. Obviously, if sufficient quantities of a vaccine remain, they should be administered at those locations and to the population generally.

USG should begin collaboration with Spain, Morocco, Israel and Egypt now. As a precautionary measure, USG should do the same with Panama and begin plans to set up a similar boundary in Panama. Plans to establish a buffer in Panama from Atlantic to Pacific by evacuation should begin now. If the efforts to contain the outbreak on the continent fail the virus will have run its course by:

[log2 (109) – log2 (16041)] * 3 about 12 months, or about September 15, 2015.

Resulting in the loss of very roughly 500 million human lives.

If containment to the African continent fails, the virus will run its course and preferentially impact countries that are not well developed or which have large, poor populations. In that case, the virus will run its course in roughly:

[log2 (7*109) – log2 (16041)] * 3 about 14 months, or about November 15, 2015.

Resulting in the loss of very roughly 2.5 billion human lives.

If the (by then) pandemic escapes the African continent, China and especially India are at grave risk because of their large, poor populations confined to a single legal jurisdiction. China will likely not allow foreign assistance on a large scale, though India might. The problem that is being overlooked here is that once the virus has a large pool of infected the ability to contain it drops sharply. In reality, I expect the 2.5 billion figure to be higher. We can expect geopolitical destabilization to occur in this case, with war and conflict becoming a salient feature. The worst case scenario, though not a likely one, is for some authorities such as those in China to use a “snake and nape” tactic. I would caution any such nation (think Pakistan and China) that use of nuclear weapons will only exacerbate the problem as the radiological effect will be as bad or worse than the virus itself. The problem here is trying to guess what a developing nation will do when desperate.

There is no way to know how this function will behave in the future and its doubling rate may change, resulting in large differences in these estimates. However, we cannot ignore the fact that the observed doubling rate is the best information we have for projecting the diffusion of the virus. Another reason to be somewhat skeptical of our starting numbers is the fact that we don’t know if the doubling time is an artifact of the reporting fidelity or a true representation of the rate of diffusion of the virus. However, it is probably imprudent to assume the former.

USG and NIH should place GSK on a war footing now, by the manner of imminent domain or national security if necessary. This should be extended to any other private competency as well, if identified. The likelihood of this virus taking a good hold in any industrialized, wealthy nation is very low, but these events could have cataclysmic economic consequences on the entire world nonetheless.

This entire analysis assumes that no mutation rendering transmission airborne occurs. All suggestions provided are made on the premise of minimizing loss of human life.

– kk

%d bloggers like this: