Marburg Toll in Angola Rises to 277

Recombinomics Commentary ^ | May 1, 2005 | Henry Niman

[Judith Anne]

Thus, the possibility of a serious undercount in and around Uige remains a very real possibility.

[Judith Anne]

Pinging the list...

[Judith Anne]

I, and others, have wondered about this for weeks.

A serious undercount, multiple breaches in infection protocols at the HOSPITALS, Marburg cases that suddenly turn out to be merely exercises (but the ministry of health was not informed), empty hospitals because the populace thinks that health workers are spreading the disease (and they may be) Marburg deaths that aren't counted because no lab confirmation is made, local healers giving "injections" to Marburg cases, porous borders, "administrative reclassification" of all cases outside of Uige...

NOT looking good.

[Judith Anne]

Oh, and I forgot to mention, the Angolan president, Jose Eduardo dos Santos (did I get the name right?) out of the country on a visit to South America...

[dc-zoo]

Oh my.. the truth finally starts to trickle out.

[Judith Anne]

See also:

Marburg Spread to Democratic Republic of the Congo?

It's also at Recombinomics, with a news header and commentary.

http://www.recombinomics.com/News/04300501/Marburg_DRC2.html

[dc-zoo]

out of the country on a visit to South America...

"How CONVEEENIENT".

[Judith Anne]

That was my opinion.

[Judith Anne]

s/b "That was my opinion, too."

Sorry.

[xVIer]

Serious undercount...hmmm, I have such a bad feeling about it.
Anyone else getting worried? That becomes unknown growth...
mrsx

[xVIer]

...And the Prez taking off to South America ?
That isn't very reassuring if this is getting loose...
maybe I am having a tin-foil episode...join me, won't you?
mrsx

[xVIer]

277 was a possible slip of the health ministers tongue the other day, remember that? Y'all caught it then! Anyway they probably don't have a clue ...just those dying right in front of them and even those numbers are constantly tampered with!
mrs x

[THEUPMAN]

Listen … is that the sound of HORSES HOOVES I hear?

[redgolum]

My boss is from Brazil, and with their lax extradition laws it is a nice place to go.

If Marburg is spreading so fast that he is running away, then this is going to be real nasty.

[Smokin' Joe]

I wonder how much of the national treasury has fled the country as well?

No, this looks like there is a numbers game going on to keep a panic down, as they are not reporting many cases which may or may not be confirmed (ever) because there is no way to run lab tests.

I would bet there are not many other likely suspects for someone dying this way, and few in circulation at present.

[tdewey10]

Things that don't make sense (TM)

(1) Blaming "Traditional Healers" for the spread.

see e.g.,
http://www.unicef.org/infobycountry/angola_26332.html
http://www.sfgate.com/cgi-bin/article.cgi?f=/c/a/2005/04/25/MNGOBCEFEV1.DTL&type=health

The poblem, according to the unicef article there are only around 150 traditional healers. If they were responsible for the spread, you think they'd all be dead by now.

(2) Blaming home "injections."

Same problem as above. If this was a source in the past, I don't see how it could be a continuing source, given that everyone who was injected or did the injections should be dead by now.

The Uigeians seem to be moderately on-the-ball, their concern with hospitals has turned out to be justified.

The recent BBC video showing no-one going near the dying baby suggests that they are no longer doing any kind of traditional mourning (laying hands on the dead, etc.)

And yet, the virus still spreads.

Possibilities:

(A) It's getting spread by mosquitos. We know that it can be spread via eating dead animals, could it have mutated such that it can stay alive in a mosquito host-body.

(B) It's being transmitted via the air.

(C) Your guess.

[Dog Gone]

I think people are reading too much into the foreign trip. Marburg is not running rampant in Luanda. If it were, we would know it because that is where the international diplomatic community is stationed, as well as the oil workers from the US who rotate their offshore job duties.

Yes, the numbers are suspect and almost certainly wrong. But this is freaking Angola where people still believe in witch doctors and reliable communication is rare. They can't even keep the power on to their hospitals.

I think it's more "clueless" than coverup.

[null and void]

A pale horse by the sound of it...

(Not zebras)

[2ndreconmarine]

Thus, the number of cases and deaths have been averaging 2-3 per day

Fascinating article. Tons of good information. I will address my response in two posts.

First, consider the number of new cases and deaths at 2-3 per day. In addition, we know that there are about 20 active cases at any one time. (Again, using Angola Health / WHO statistics). Well, the simple calculation for the lifetime of a patient after symptoms until death is:

tau_d=active / rate = 21 / 3 = 7-10 days

OK, now we have a good number for the lifetime. Using the model, (c.f. Model Derivation) we know that the growth rate is then:

tau_g = tau_d * (1/((I_c/D_c) - 1))

D_c/I_c = .92, ==> I_c/D_c = 1.08 ==>

tau_g = 11.5 * tau_d

Since tau_d ~= 7 days,

tau_g = 80 days.

So, the growth rate is exponential with a time constant of 80 days. This number is perfectly consistent with the growth rate reported. Moreover, he reports:

Thus, the number of cases and deaths have been averaging 2-3 per day

With 250 cases, we have roughly that the daily new cases should be:

250*(exp(1/80) -1) = 3

It all fits very nicely. So, we have a real measure of the growth that is reliable, assuming reported data are correct. This is nice because the data fit the model perfectly. Indeed, a comparison between the 2nd and 3rd growth plots that we showed yesterday (c.f. Marburg Growth Plots) show the same thing. The calculation for D_c/I_c = .92 matches the reported growth in the third plot exactly.

[2ndreconmarine]

The "stabilization" in cases has really been a stabilization in totals, which include discarded reported patients. The discarding had been at about the same rate as newly reported patients.

A very disturbing point, and also the point you and many other posters to these threads have made.

How many real, but unreported Marburg cases are there?

The concern is not the number of cases, per se, but the fact that the growth rate determinations (by calculation or curve fit) would be wrong. What makes this a real threat is the potential for growth, not the numbers today.

If we take his statements as litteral truth, then:
1. There was no slowing of the growth rate, it was all due to "administrative reclassification."
2. The number of new, unreported cases is identical to the number of new, reported cases.

Therefore, the number of new, reported cases in April was:
277 - 120 = 157

If there were an equal number of unreported cases, 157, then the total number of cases would now stand at:

277 + 157 = 434

These numbers are disturbingly close to the growth rate predictions. The growth curve prior to April 10 suggested a growth rate that would have 500 cases by today (c.f. Growth Plots ). Moreover, we discussed a thread a few days ago (c.f. Additional Active cases post, and subsequent), that there were either 89, 174, or 263 additional, unreported cases. (Depending on the meaning of the English). Well

157 ~ 174

So, this report would be consistent too. Therefore, extending the Recombiniometrics analysis, there may the reported 277 cases, or the true number may be as high as ~450 or so.