WHO's 123 New Cases/Day = 100% EVD Spread Per Month

[franzpick]

WHO's new cases at 3685 in today's 9-4 report for 8-31, up from 3069 cases on 8-26 in their last report 8-28, are up 616 in 5 days, or 123 cases per day. Were that daily increase to continue for 30 days, we will see 3690 more cases / 3685, or a 100% case increase in the next month.

Taking instead the average daily case increase over WHO's last TWO reports covering 8-21 to 8-31 gives 3685 - 2615 = 1070 new cases, divided by 11 days, gives an average increase of 97 cases/day, X 30 days = 2910 more cases in 30 days, adding up to 6595 cases, or a 79% increase in 1 month.

The same math for the 20 day increase from the 1975 cases WHO reported on 8-11, to 8-31's 3685 cases still averages a high 86 cases/day, giving a 70% increase to 6265 cases by Oct.1. Clearly, the average daily new case averages being reported are going up, not down, and even if the daily increases were to drop 50% to say, 50 new cases per day, the monthly totals would continue to expand at 1500 plus/month, or 40% every 30 days:

http://www.who.int/csr/don/archive/disease/ebola/en/

WHO's previous regular August reports every 2, 2 and 3 days, Wed-Fri-Mon gave some immediate indication of trend changes. Lacking those reports, and rather than waiting 4-5 plus days for WHO's next report, and not having a previous report to compare it to, any recent 5 day trend changes can nevertheless be approximated by calculating the daily average new case increases by referring to the DAILY reports from the national ministries of health of Liberia, S. Leone and Guinea, all of whom issue DAILY reports:

https://www.internationalsos.com/ebola/index.cfm?content_id=420&language_id=ENG

Unfortunately these MOH reports are poorly organized and presented, but when and if I can discern the recent 5 day new case daily averages from the reports for the 3 countries I will post daily an approximation of current changes in the rate of the EVD spread.

 

[JimO]

Re: WHO's 123 New Cases/Day = 100% EVD Spread Per Month

I have read that ebola Zaire has an R0 of around 1.8 and that the timeframe cases have doubled during periods when it has been spreading during this outbreak (exclusive of a lull during July) has been approximately 24 days. Without effective controls being put into place, this would result in close to 4,000 deaths by the end of September and as many as 40,000 deaths by the end of the year. Beyond that, if the outbreak is not brought under control, the numbers become unimagineable. There are approximately 16 to 17 million people in Guinea, Liberia, and Sierra Leone. With an R0 of 1.8 in an unprotected population and lack of controls, up to 75% of the population could ultimately become infected. With a CFR close to 70% of resolved cases, the long-term effects of failure to curtail this outbreak could rival some of worst disease epidemics in modern history. Also, these numbers only apply to the three countries. Any spread to other countries, particularly Nigeria, would increase these numbers.

 

[franzpick]

Re: WHO's 123 New Cases/Day = 100% EVD Spread Per Month

"... the pathogen must effectively transmit within a group (R0>1), and then the pathogen must persist within a group long enough to allow for movement among the groups. Therefore, the factors affecting disease persistence--such as infectious period, group size and recruitment of new susceptibles--are as important as the local transmission rates in predicting the spread of pathogens across a metapopulation."

http://www.ncbi.nlm.nih.gov/pubmed/17251146?dopt=Abstract

Thanks Jim, sure looks grim (and over my head).

I'm hoping that data capture is somehow overstating the new case numbers, which would subsequently taper down, and also that containment is improving faster than the villagers, as well as executives and MDs, can flee, and faster than families can continue to do unreported and dangerous burials.

 

[Vibrant62]

Re: WHO's 123 New Cases/Day = 100% EVD Spread Per Month

Franzpick - thanks for doing the maths.

It is also worth bearing in mind that these numbers are a gross underestimate, as there are many additional cases outside treatment centers as there is simply no room for more patients, and many individuals are being turned away.

However, working with what data we have still gives an indication of how things are progressing and its not good.

 

[Laidback Al]

Re: WHO's 123 New Cases/Day = 100% EVD Spread Per Month

There is no doubt that the number of reported Ebola cases is continuing to increase as shown in the chart below. However, there are constraints on the growth rate of cases and the number of people that will be infected. There is no reported infectious disease that has an attack rate of 100%.

 

[Sharky]

Re: WHO's 123 New Cases/Day = 100% EVD Spread Per Month

This is a continuation of an existing trend; the doubling time for total deaths has been a little less than 30 days for several months now. As Vibrant62 says, it's extra concerning because we know the numbers are low. Samaritan's Purse estimated at least 25 to 50% low; could be worse.

At some stage, I expect the reporting to get even worse, as it becomes that much more difficult to keep track.

In the long term, communities will develop population immunity, since survivors are immune to re-infection from the same strain. I wonder if any estimates have been made for how long that could take?

 

[franzpick]

Re: WHO's 123 New Cases/Day = 100% EVD Spread Per Month

"In the long term, communities will develop population immunity, since survivors are immune to re-infection from the same strain. I wonder if any estimates have been made for how long that could take?"

Not qualified to respond to that in any manner, so, all seriousness aside, I'll guess that with a 50% CFR it takes until half the population is gone. Doubling their current (1000) fatalities every 30 days would reduce Liberia's 4 Mil population to 2 Mil in just 10 months. YMMV (math).

 

[sharon sanders]

Re: WHO's 123 New Cases/Day = 100% EVD Spread Per Month

at 16:24 on September 05, 2014, EDT.

Ebola cases could grow by thousands per week if current spread continues: WHO

Helen Branswell, The Canadian Press


TORONTO - The World Health Organization is warning that the speed of spread in West Africa's raging Ebola outbreak could escalate in the coming weeks, with thousands of new cases a week by early October.

With the current rate of growth in the hundreds of new cases per week and no easing of the rate of transmission, a jump into the quadruple digits seems to be where this unprecedented outbreak is heading, an epidemiologist working in the office of WHO Director General Margaret Chan said Friday.

more...

http://www.cjbk.com/NationalCP/Article.aspx?id=434307

 

[franzpick]

Re: WHO's 123 New Cases/Day = 100% EVD Spread Per Month

Helen Branswell, The Canadian Press:

"...the speed of spread in West Africa's raging Ebola outbreak could escalate in the coming weeks, with thousands of new cases a week by early October."

WHO's August reports reveal that we are now WITHIN DAYS of a "thousands of new cases a week" rate, and not 3 1/2 weeks away to 'early October': WHO's latest increase of 615 cases was for a 5 day period, 123 cases/day, which equates to a current new case rate already at an 861 per week rate. A 16% rise from the current 123 new case rate, to 143 cases per day, gives us 1000 cases per week.

Daily new case averages today from INTERNATIONALSOS.c, for the last 5 days, for the 3 countries with out-of-control areas, are Liberia 60 cases/day, Sierra Leone 24/day and Guinea 24/day, averaging 108 new cases per day, with many infected being turned away each day from already full treatment-confinement centers, and not included in the daily new case totals.

I will follow daily changes in the new case numbers, and post significant changes in the average 3 country new case rate, and doubling period, which is now at 30 days.

https://www.internationalsos.com/ebola/index.cfm?content_id=397&language_id=ENG

 

[Chris]

Re: WHO's 123 New Cases/Day = 100% EVD Spread Per Month

Ian Mackay has written about daily changes today at his blog. I find his maps and case data to be informative:

http://virologydownunder.blogspot.com.au/2014/09/case-number-changes-between-ebola-virus.html

 

[franzpick]

Re: WHO's 123 New Cases/Day = 100% EVD Spread Per Month

"Unverified sources are reporting that there will be a nationwide (Sierra Leone) "lock down" from 18 to 21 September. Everyone is required to stay in their homes on those days, in an effort to stop further spread of Ebola. Health workers will perform contact tracing and monitoring during that time, and identify those with symptoms of the disease."

https://www.internationalsos.com/ebola/index.cfm?content_id=397&language_id=ENG

This is an idea whose time has come, and waiting 2 1/2 weeks may be to organize the tracing team and perhaps to acquire food, water and supplies that might need to be provided to the confined residents: keeping everyone 'home' will facilitate a 'census' count of those who are symptomatic, and might also prevent their access to infected meat (and fruit?) as well as H2H contact with the infected, and with EVD victims and those at unauthorized burials:

https://www.internationalsos.com/ebola/index.cfm?content_id=397&language_id=ENG

 

[franzpick]

Re: WHO's 123 New Cases/Day = 100% EVD Spread Per Month

Thanks Chris: Mackay puts much work into his VDU site, but his use of the recently delayed, and now irregularly timed WHO reports leaves him 1) 4-5 plus days behind the current case numbers available daily from the 3 country MOH's, and 2) without a similarly timed previous report to use for comparison and trend analysis.

My intention is give an indication of what the case increases have been in the last 5 days, easily comparable to the previous 5 day period, without reference to days old and incomparable data from the last WHO report - and graph trend analysis not being part of the problem.

 

[Sharky]

Re: WHO's 123 New Cases/Day = 100% EVD Spread Per Month

Not qualified to respond to that in any manner, so, all seriousness aside, I'll guess that with a 50% CFR it takes until half the population is gone. Doubling their current (1000) fatalities every 30 days would reduce Liberia's 4 Mil population to 2 Mil in just 10 months. YMMV (math).

No. "Population immunity" specifically prevents this type of worst-case scenario. A 50% CFR that killed fully half of the population would mean that everyone was infected.

What happens with population immunity is that the rate of infection eventually declines. Survivors are immune, so as their numbers increase, the number of people that a given infected person can pass the disease to decreases.

I don't know what the numbers are for Ebola (hence my earlier post), but for some viruses, a relatively low fraction of immunity in the population can completely prevent outbreaks.

 

[JJackson]

Re: WHO's 123 New Cases/Day = 100% EVD Spread Per Month

Herd immunity is achieved once the reproductive number (Ro) drops below 1. The Ro is the number of new case each case seeds. An Ro of 1 means the the outbreak is maintained but does not increase or decrease, less than one it dies out, more than one it grows the higher the value the faster it grows. Seasonal flu has an Ro of about 1.35 previous Ebola outbreaks have had Ro's about 1.8.

While a disease may have a typical Ro it is not fixed for the disease and varies with the environment. A flu outbreak in a sparsely populated community of goat herders is not going to spread like the same strain if it gets loose in Calcutta railway station.

As Ebola works its way through a closed population it will eventually slow as its finds it more difficult to find new host who have no immunity but in areas with high population density, and a lot of the right types of inter-personal contacts, the Ro will soar and you will need very high levels of natural immunity to bring it back down to below 1. Interventions to drop the rates of potentially infectious interactions (like quarantines and hand washing/gloves) are aimed at manipulating the Ro.
N.B. that the closed community in question is not Liberia but the planetary population. As a forest fire moves to new areas, once it has exhausted the fuel to hand, so will Ebola. To contain a forest fire we can use a fire break or air bomb it with a fire suppressant, in this analogy the fire break would be a totally effective quarantine of 20 million people with no cross board movement and the suppressant would be a vaccine or cure.

The aim of herd immunity by vaccination is to surround an infected individual with so many un-infectable people the disease can never get a foothold. As we have seen this has worked well with the likes of the MMR diseases until a significant minority opted out of the program believing the disease had gone and the vaccine was risky. Their actions have broken the heard immunity and these old scourges are causing faltering outbreaks again. Thankfully most still take the vaccine so the Ro's have not been able to rise to their pre vaccination levels and these diseases can not cause the epidemics they once did.