New for August 16th
I will try to keep new information and updates in the same post for a while until I have a chance to do a comprehensive re-overview of everything.
The 16 August update from WHO indicates a large uptick in the daily number of cases. Over the two days of the most recent reporting period an average of 76 cases per day have been identified as confirmed, probable, or suspect, with a total of 76 deaths over that period of time.
Good news is that the situation in Nigeria doesn't seem to be developing. There were no new cases over the this reporting period, and one death. The last new cases in Nigeria were reported on August 6th for the period between August 2 and august 4. So over ten days without a new case in Nigeria probably means that the "outbreak" is burned out.
Meanwhile, there is another suspected/possible case in another country. This has happened a number of times, where a suspected Ebola case is identified. Hong Kong, the Philippines, the US, etc. have had these, and of all of the cases none have been shown to be Ebola except possibly one, and that
is in Saudi Arabia did not work out either, the Saudi case was not Ebola. The new possible case is in South Africa.
The updated graph showing the increase in daily cases is inserted below as before.
And now, a personal note to Laurie Garrett. Laurie wrote this post, and I wanted to comment on it but the commenting system there did not work for me. (Perhaps one has to be subscriber.)
I love you work. It was your book, based on your Thesis at the Kennedy School, that got me interested in tropical diseases. Well, that and at the same time going to the tropics, running a makeshift health clinic there, and getting some of the diseases. I often point people to your earlier writing on influenza to find out about the true pre-Wakefield anti-vax movement, to see how the US handling of Swine Flu made it very difficult if not impossible to have a sensible national vaccine program that was not byzantium (which is what we have now)
But I think your article on not being scared enough about Ebola has some problems. I agree that this outbreak has not been taken seriously. I nave noted in my own writing that WHO and CDC, even, are coddling the public about some of this. I also noted, which I don't think you did, that Ebola "in Africa" is Ebola in America already. One of my neighbors died of Ebola, and one of his relatives in Liberia did as well, and some of my other neighbors lost relatives, I've heard. This is because Liberians and other Africans live in communities with one foot in Africa and one foot here. Those who died of Ebola did so in West Africa, but they are still neighbors who live here part time or African-based relatives of neighbors who live here full or part time.
So yes, for many reasons, be concerned.
Here's where I don't agree with you.
First, while the cures and vaccines are truly not deployed as you point out, you are more negative than necessary. In fact you are hyperskeptical. A common phrase in hyperskepticism is "there is not a shred of evidence of..." Well, there is not a shred of evidence that my four year old is upstairs eating a peanut butter and jelly sandwich instead of the nice dinner I made my family, but that does not mean it is anywhere near impossible. In fact, he's probably eating a peanut butter and jelly sandwich, though I can produce not a shred of evidence from here in my basement that he is. There is in fact good scientific reasons to think that the cures that have been brought to the brink of testing are likely to work, and Ebola is not like Malaria (not even close) or even viral cousins such as Influenza when it comes to vaccine prospects. The prospects are good, if only someone would work on them.
Next problem: No, Saudi has not Ebola. No, there have not been a number of actual, non-panic-based cases of Ebola outside of the affected area other than my neighbor and those he infected, in Nigeria.
Next:No, Nigeria does not actually seem to be having an outbreak. No new cases in ten days is good news. It may be over there.
We'll see about South Africa.
But yes, I do agree with you on two other points. First, all of the health care molecules have moved to one corner of the proverbial room suffocating other health care efforts in the affected countries. This is a big deal. Second, yes, it really is possible despite major media and major organizations insisting it is highly unlikely for this outbreak to seed an outbreak pretty much anywhere in the world. Not that likely. But I won't say that there is not a shred of evidence that it could happen (citing that every single case outside the zone for which confirmation was completed has not worked out). I'll just say that we have no freakin' clue how likely it is, but it is not zero and the consequences would be dire.
So, I don't want to tone-troll your article. You went for breathless, and you got to breathless, and that's probably a good choice, you have the credibility to pull it off and people need to hear much of what you said. But no, Ebola is not leaking out of the zone now, and yes, there is better hope I think for the existing (as in on the table, not deployed or even tested) cures and vaccines (and by the way, the "ethical considerations" are a red herring, that is true for all drug development, but seems only mentioned frequently with respect to Ebola).
End of rant. Again, love your work.
End of August 16th update
Yesterday I made the optimistic statement that the number of new cases a day may be leveling off, as for two reporting periods in a row, representing five days, the new cases were about half of the previous reporting period, normalized to a per-day estimate.
Today's report from WHO covers two days and indicates 128 more cases, so the number of new cases per day for the latest known period is actually higher than at any previous time during this outbreak. Pursuant to this I've replaced the pertinent graphic below. I was optimistic, but I also provided caveats. The caveats won.
Is the current Ebola Outbreak subsiding?
At some point, the Ebola Outbreak in West Africa has to slow down and stop. The disease is too hot to not burn itself out, and it has no human reservoir. Ebola accidentally broke into the human population earlier this year or late last year, probably once (see below), and despite the regular increase in daily reported cases over the last several weeks, the disease must at some point begin to level off.
The latest two updates from WHO indicate that the Ebola outbreak may be leveling off now, tough it is too early to be certain. The following graph shows the approximate number of new cases reported per day by WHO. This is calculated by taking the number of new cases in a report and dividing by the number of days covered by that report. A given estimate of daily new cases may be quite off for a number of reasons. First, even if there is a long term upward or downward trend, there is likely to be a lot of randomness in the data. Second, this is the number of cases reported in that time period, not the number of cases that manifest. It is likely that some cases manifest during the reporting period are not recorded yet, and cases manifest for the prior reporting period are included in the current reporting period. Over several reporting periods this would, obviously, even out, but a given number of days in a reporting period may be off by a day or so. So, these caveats mean that we should be very cautious in interpreting this graph.
Note that what appears to have been a fairly steady increase in number of cases, with about the same number of ups as downs but with the ups adding to a higher sum, since late June, has been followed by two reporting periods with decreases in numbers of new cases. Note also, however, that in late May the number of new cases per day went up fairly quickly then dropped again before a new steady rise occurred. If we use a moving average of 3 data points, which would combine sets of 2-4 days each to obtain something close to a 10 day effective moving average, the upward trend is more evident than any recent downward trend:
The next two WHO reports may clarify this trend.
Mortality Rate Is Decreasing
The mortality rate for this outbreak continues to decrease slightly, which is probably a result of increase effectiveness of the response to the outbreak, despite all the news stories about how things seem out of control.
The current mortality rate is dropping below 55% given confirmed, probable, and suspected cases and deaths. But the rate varies across different categories. The outbreak-long rate for all cases and all deaths is currently 55%, and looking only at confirmed cases and deaths, it is 56%. The mortality rate for all previous African Ebola outbreaks, taking total reported cases and total reported deaths, is 66%.
This is the largest outbreak ever, and then some
Currently there are over 1,800 confirmed, probable, or suspected cases reported in the West African outbreak, and 1176 confirmed cases. Using just the confirmed cases, to be conservative, the present outbreak is 277% larger than the next largest outbreak, which was in 2000 in the Gulu, Masindi, and Mbarara districts of Uganda, with 425 cases. The total number of confirmed cases for the present outbreak represents about 49% of all of the prior African Ebola outbreaks combined.
Patient Zero Identified?
Patient Zero, who we assume is the person to whom the disease jumped from its usual animal reservoir, directly or indirectly, may have been a toddler in Guinea. The two year old child died in December 2013, which is quite a bit before this outbreak came on everyone's radar screen, and after which it was fairly low level for a while.
I've long maintained that a likely way for Ebola to get into other species is from ground dwelling mammals, such as chimps, gorillas, or forest antelopes and duikers, ingesting or mouthing the discarded wadges of fruit previously handled by Ebola-carrying fruit bats. From such non-human animals the Ebola would then enter human populations from people butchering bush meat. In this case though, I wonder if the toddler may have been a direct recipient, picking up and mouthing fruit-bat spit covered fragments of fruit found on the ground. A parent's worse nightmare, apocalypse style, to be sure.
The Famous Untested Drug
There has been a bit of complaining about my use of the term "drug" or "cure" for ZMapp, a drug that was developed to fight Ebola but not used until just now. Some have said it is not a drug until it is tested and deployed, and until then, it is a possible cure and not a real cure.
This is wrong. A "possible cure" is when you take an existing compound or substance, apply it to a pathogen or an affected animal model, and get a hopeful result. This possible cure can then be further developed to make, most of the time, nothing because these things generally don't work out. Or, to make a cure. Which can then be tested.
In the case of the treatment now given to three patients (two survived one died), the cure was developed for Ebola based on some pretty solid science and prior experience with similar type cures working for similar diseases, successfully, in the past. The cure was not in "hopeful" or "possible" phase, but rather in developed but untested phase. The WHO convened an emergency panel of experts, yesterday, which decided that the cure should be used in the field under certain circumstances. So now there is an untested but developed cure for Ebola being deployed in West Africa. The WHO discussion on this is here.
Regarding the decrease in mortality, I think that this is partly artifact---a math thing. If you look at a graph of infections and deaths, the "cases" are farther along the exponential curve than the "deaths" . . .this widening gap is partly from the lag between when someone is reported as infected and when they are reported as dead. To get a more accurate (but still hugely ballpark) death rate you would compare the # of cases from, lets say, a couple of weeks ago with the current # of deaths. Then to complete the picture you would keep in mind the WHO's comment about the cases and deaths being "VASTLY underestimated". What a choice of adverb . . . V A S T L Y .
One aside, I disagree about Ebola's inevitable "burnout". Despite Ebola's "hotness", at the end, victims are teaming with the virus and it gets very messy, so the family caretakers invariably are exposed/infected. In about 14 days or so those people will need caretakers, and so on. As long as there is an available pool of uninfected individuals, Ebola will continue to burn on until it is put out.
Addendum: I am assuming that the majority of Ebola victims are being cared for at this time by family members. I did not address the effect of Ebola victims being taken to receive professional health care, which could either serve to slow or speed its spread, depending on knowledge of infection control and on available resources.
Regarding the artifact, I discussed that in my earlier post that dealt more with mortality, suggesting the very effect you mention. But that effect should operate over shorter time scales that we see here, and the drop does not go away when the fop data are massaged as you suggest. I think we see both. There is usually an occupies boundary effect at any moment, plus a longer term spdownward shift other trend.
Burning out certainly is hard to do in such a relatively hoping population area.
I'll post today what I considered posting last Thursday (since Friday) --but did not--only to find that reports on Ebola carried by Al Jazeera's English-language service stated exactly what I'd had in mind:
this epidemic is considerably more serious in scope than has been thought by even some in the professional ranks of epidemiology and I think--and thought Thursday--that your postings have been far too complacent about the capacity for containment and too conservative in your assumptions about the potential numbers of currently infected people.
My impression is that there are perhaps thousands who have risked exposure and who remain so far undiscovered across the affected areas. Moreover, when infected people travel in populous urban areas such as Lagos, the disease's opportunities for spread are virtually open-ended.
I don't know what in the world you mean when you wrote that,
"The disease ... has no human reservoir." Nor can I see any good reason to believe that the disease is "too hot to not burn itself out ..."
A temporary drop in detected infections rates--that is, over the space of a mere few days--should be regarded, really, as telling us nothing effectively valid about what the disease may do in the coming weeks. Only a steady and prolonged decline in infection rates is indicative of a turn in the progress of the spread and I've not seen this when the affected areas as a whole are considered.
After a quick scan of Laurie Garrett's linked article (thank you, GL, for presenting that!), I'm in fairly complete agreement with her judgment on practically every essential point. We needn't suppose with Ms. Garrett that her remark,
..."my brilliant Council on Foreign Relations colleague John Campbell, former U.S. ambassador to Nigeria, warned that spread of the virus inside Lagos -- which has a population of 22 million -- would instantly transform this situation into a worldwide crisis"....
is precisely and literally true in its forecast. It may indeed not be "instantly" done, but that won't matter if the spread of the virus (with its eventual mutations) isn't effectively contained.
Prox (if I may call you that) yes, and I mainly agree with her as well. But Nigeria simply isn't having an EBOV outbreak, so one could have used any large densely populated city with certain other characteristics connected to the affected area by airplane.
Lagos is theoretically scary, but not empirically scary, but many are empirically scared because there have been some cases there. That is an important difference that I felt needed to be pointed out.
@ 7 : "Prox (if I may call you that) ..."
Certainly. That's standard practice for such a discussion as this.
Citing you: "But Nigeria simply isn’t having an EBOV outbreak, so one could have used any large densely populated city with certain other characteristics connected to the affected area by airplane."
? Then I'm confused about what constitutes an EBOV* outbreak in Nigeria. What about the following statistics and what virus, if not EBOV or "Ebola,"generally, as I mean that term, is being referenced by them?
(source: Center for Disease Control / 2014 Ebola Outbreak in West Africa
Suspected and Confirmed Case Count: 12
Suspected and Confirmed Case Deaths: 4
Laboratory Confirmed Cases: 11
and this, from Wikipedia:
"There are 11 confirmed cases of Ebola in Nigeria as of 14 August 2014. The first one was an imported case of a Liberian-American, Patrick Sawyer, who traveled by air from Liberia and became violently ill upon arriving in the city of Lagos. On 20 July, Sawyer flew into Nigeria via Lomé and Accra from Liberia, and he died five days later in Lagos. In response, the Nigerian government increased surveillance at all entry points to the country; health officials were placed at entry points to conduct tests on people arriving in the country. Initial reports noted that sixty-nine individuals previously in contact with Sawyer (including airport staff, fellow flight passengers and health workers at the hospital where Sawyer was hospitalised) were placed under close surveillance without symptoms. On 4 August, it was confirmed that the doctor who treated the patient tested positive for the virus strain and is being treated. On 6 August, Nigerian authorities confirmed the Ebola death of a nurse who had treated Sawyer." ...
(*) Unless I'm mistaken, EBOV refers to one of the common strains of the deadly Ebola virus but my reference is to any of the five strains and not strictly and only the EBOV strain and this is the sense in which I use the term "Ebola".
I guess that it is this distinction which is all-important for your analysis---as in *Ebola cases in Nigeria, yes, but not the EBOV.* Unless that is the gist of your point, I'm very confused about how we are in a misunderstanding.
"Then I’m confused about what constitutes an EBOV* outbreak in Nigeria. What about the following statistics and what virus, if not EBOV or “Ebola,”generally, as I mean that term, is being referenced by them?"
It isn't clear how many cases make an outbreak, and it is certainly possible to say that they HAD a tiny outbreak (smaller than any previous instance). But, at least one case was not infected in Nigeria, and the other cases are all closely related to that one case, and there are no new cases. So, Nigeria is not having an outbreak. Close call.
The last new case in Nigeria was before August 6th, so over 12 days ago. Maybe there are additional cases waiting to be discovered, but I'm going with no current activity there.
This is just not "science."
From a sample data set, you cannot take the last two numbers and infer anything. It does not have statistical power.
The last two reported numbers are low - it may be subsiding - this is tabloid journalism, not scientific discussion.