Brazil's unexpected flu pattern

An extremely interesting article is slated to appear in the American Journal of Epidemiology later this month. I haven't seen it yet but Nature News carried a short piece about it. It comes from a team of experts in seasonal flu patterns at NIH's Fogarty International Center (FIC).

The notion that flu epidemics start in areas of high population density and spread outwards may not hold true for the tropics, hints a study from Brazil.

In that country, new research reveals, flu starts in the less densely populated north and moves towards cities in the south. The result indicates that climate, rather than population density, plays a bigger part in the spread of the disease in Brazil. And that could have implications for how flu is managed in the tropics.

"This flips our understanding of influenza in tropical regions on its head and will hopefully improve control strategies," says Mark Miller of the Fogarty International Center (FIC) for Advanced Study in the Health Sciences in Bethesda, Maryland, who worked on the study. (Matt Kaplan, Nature)

Miller and his colleague Wladimir Alonso at FIC combed through several decades of Brazilian influenza mortality data and morbidity data from 2000 to 2005, expecting to seasonal to flu start in the large cities and transport hubs, like São Paulo and Rio de Janeiro, and then spread to the smaller cities. But instead it started in rural areas of the north 2 to 3 months before the large cities and went south. The climates in these two regions are quite different, the north jungles rainy in the winter, the south hot. July to August, however, is a frosty winter in the south. If climate is really a factor, and these are far from the only observations that says it does, we don't know how. This is part of the enduring mystery of seasonality in flu.

Brazilian public health planners are interested in these data because it suggests vaccination programs might need to be timed differently in different parts of the country. That's an important practical application. But the data are one more reminder that we don't understand one of the most commonly characteristics of influenza in human populations. It has a seasonal periodicity. There is variation in the amplitude and phase but the general pattern remains.

What causes it?

More like this

Revere, are you participating on the HHS Pandemic Flu Leadership blog? It's foundering, and there hasn't been much in the way of epidemiologic presence there. This would be great info to share. It seems that HHS is floundering a la Katrina, and I can't understand why.

N=1: No, have read a couple, but it didn't seem to be science oriented and wasn't sure there was anything I could add. I'm not sure what the idea was and Greg Dworkin was ably holding down an online community view.

"Didn't seem to be science oriented" is the understatement of the year. Too bad.

By Galen McBride (not verified) on 12 Jun 2007 #permalink

As far as I was able to ascertain, it's just a PR gimmick that the HHS marketing contractor, Ogilvy, is using that is purporting to be soliciting input and carrying on a dialogue with the public. All of the posts authored by HHS officials have been canned talking points, and there has been zero response to commenters.

I think this is going to have a deleterious effect on preparedness planning and development. It just reeks, and I am sorry that GreG Dworkin and the Flu Wiki folks are entangled in it.

HHS Pandemic Flu Leadership (or lack thereof) Blog is very educational,
if you can get members of the public
to go look in their taxpayer-funded sausage factory.

(Public that thought "bird flu had gone away" or that it wasn't even on the govts' radar because there was no threat, or they woud hear about it all the time on the news...
Tell them they're missing all the interesting comments,
and that it would be better not to think they can wait
for any Congressional hearings on the matter.)

By crfullmoon (not verified) on 12 Jun 2007 #permalink

did they observe how it spreads from human to human
and slowly travels southward in a wave ?

or did it survive in the south somehow and then revives
when winter comes

the sequences should show it.

anon: These are mortality data and clinic records from the later period. There are no sequences to look at.

Revere, In the Amazonian countryside here(Bolivia)we have beautiful Vampire Bats(Desmodus Rotundus)They are a virus vector between mammals. They also do large birds and the occasional sleeping drunk. If H5N1 comes here it could have a better chance to evolve. One can see the short flying distance downwind from West Africa and wonder why it has not arrived yet to Brazil.

Lou: Bats have always been one of the question marks in my mind. They are an ideal reservoir, they are mammals and they are everywhere. So far I don't know of any documented evidence of bat infection but I don't think it's been studied much (or at all).

just try to infect some bats with flu in a lab.
Why hasn't it been done ? What's the problem ?

22 H3N2-HA-sequences from 2004,Brazil at genbank.
We just need the exact locations and the time of sampling.
And coordinate it with the epidemiological data.
And some more segments were useful, I assume this could be done,
the samples are still there in some refrigerator ?!

(A/Brazil/1759/2004(H3N2))
(A/RioGdoSul/212/04(H3N2))
(A/RioGdoSul/211/04(H3N2))
(A/Parana/306/04(H3N2))
(A/Parana/308/04(H3N2))
(A/RioGdeSul/213/04(H3N2))
(A/RiodeJaneiro/26/04(H3N2))
(A/RioGdoSul/411/04(H3N2))
(A/Parana/291/04(H3N2))
(A/Parana/313/04(H3N2))
(A/RioGdoSul/417/04(H3N2))
(A/StaCatarina/380/04(H3N2))
(A/RioGdoSul/406/04(H3N2))
(A/RiodeJaneiro/17/04(H3N2))
(A/StaCatarina/379/04(H3N2))
(A/Parana/312/04(H3N2))
(A/RioGdoSul/214/04(H3N2))
(A/Minas Gerais/156/04(H3N2))
(A/Minas Gerais/160/04(H3N2))
(A/Minas Gerais/154/04(H3N2))
(A/Parana/298/04 (H3N2))
(A/Minas Gerais/163/04(H3N2))

Genomic evaluation of hemagglutinin sequences from recent Influenza
virus samples, characterized during 2003-2004 epidemics in South
and South East Brazil
JOURNAL Unpublished

Motta,F.C., Rosado,A.S. and Siqueira,M.M.
TITLE Comparison between denaturing gradient gel electrophoresis and
phylogenetic analysis for characterization of A/H3N2 influenza
samples detected during the 1999-2004 epidemics in Brazil
JOURNAL J. Virol. Methods 135 (1), 76-82 (2006)

http://www.ncbi.nlm.nih.gov/sites/entrez?cmd=Retrieve&db=pubmed&dopt=Ab…