H1N1 Death totals

The CSV data from the rhizalabs site does seem to show more than 150 deaths worldwide from H1N1. I'm not sure why the number is higher than the totals quoted elsewhere, but the CSV data does point to an excellent source of summary data: the Mexican government's Health - Mexico website. (I'm not sure about the proper site name. The URL is http://portal.salud.gob.mx/.)

In my previous post I wondered about the distribution of ages in fatal cases. The PDF reports from the Mexican health site present exactly that information (albeit for Mexico only), very clearly. See for example page 4 of http://portal.salud.gob.mx/descargas/pdf/influenza/situacion_actual_epidemia_190509.pdf

U.S. Says Older People Appear Safer From New Flu Strain - NYTimes.com

U.S. Says Older People Appear Safer From New Flu Strain - NYTimes.com:

"...federal health officials said on Wednesday that people born before 1957 appear to have some immunity to the swine flu virus now circulating.

Tests on blood serum from older people showed that they had antibodies that attacked the new virus, Dr. Daniel Jernigan, chief flu epidemiologist at the Centers for Disease Control and Prevention, said in a telephone news conference."

It would be interesting to see the distribution of ages in all confirmed incidents. Who aggregates that data?


Self-Serving Status Update

I've continued working on the Mesa H1N1 interactive timeline site. It now includes a graph of cumulative incidents (confirmed, suspected, fatal) over time. The number of new incidents worldwide has grown enough that I'm going to need to refactor the code, to better limit (through aggregation) the number of incident markers which must be placed on the map.

I'm still having trouble with a discrepancy in the time-series graph. The development server yields 86 deaths worldwide as of 20 May 2009, whereas the production server shows 154 deaths. Both are supposed to be using the same code, and the same database. I'm reloading the production database now.

It appears that http://flutracker.rhizalabs.com is still the best source of incident data, although there are a few discrepancies in its records. If anyone knows of an authoritative source of detailed incident data, please let me know via the comments.

More about masks and flu transmission

I really need to learn to read...

How can masks be effective in preventing spread of viral infections, when viral particles are so small and when experiments with guinea pigs show that coughing and sneezing aren't required for transmission?

From the PNAS journal article:

"Our results indicate that droplet transmission of influenza virus occurs between guinea pigs..."

Granted, I haven't dug into the article for information on the droplets' sizes, but this suggests all kinds of variables to examine: mask porosity, how vapor/droplets condense onto masks, etc.

More anecdotal evidence on the effectiveness of masks in preventing secondary pneumonia, at least, comes from "The Great Influenza". From page 211:

"Capps did write the JAMA article. He reported finding the masks [used by patients with respiratory disease at Camp Grant near Rockford, IL] so successful that after less than three weeks of experimenting he had abandoned testing and simply started using them 'as a routine measure'."

They didn't always work. Page 215, describing the September 1918 wave of infection:

"Despite all precautions, despite wearing surgical masks and gowns... 43 percent of the staff... required hospitalization. Ten nurses at this single hospital died."

Technology Review: Swine Flu Genome Hints at Milder Virus

People exposed to the 1957 flu pandemic may have some immunity to the current H1N1.

Technology Review: Blogs: TR Editors' blog: Swine Flu Genome Hints at Milder Virus:

"Ralph Tripp, an influenza expert at the University of Georgia, said that his early analysis of the virus' protein-making instructions suggested that people exposed to the 1957 flu pandemic--which killed up to 2 million people worldwide--may have some immunity to the new strain. That could explain why older people have been spared in Mexico, where the swine flu has been most deadly."

Secrets of the Dead: Killer Flu suggested that the distribution of deaths by age in the 1918 pandemic hinted at a prior outbreak, about fifty years earlier.

Swine Flu Genes From Pigs Only, Not Humans or Birds | Wired Science

Swine Flu Genes From Pigs Only, Not Humans or Birds | Wired Science:

"‘The new neuraminidase gene that came in from Eurasian swine is one we’ve never before seen circulating in humans,’ said Rambaut. ‘That’s one of the reasons it’s spreading rapidly. Very few people will have any immunity to this particular combination, which is what gives the concern that this will be a pandemic rather than just a normal seasonal flu outbreak. It remains to be seen how much and to what extent there is existing immunity.’"

Update 2009/04/30: The following article is more recent than the one above, but is it based on the "earlier reports" mentioned in the Wired article?

Technology Review: Hunting for Clues in the Swine Flu Genetic Code:

"The new swine flu virus appears to have done more swapping than usual, with genetic segments from four different sources: North American swine influenza viruses, North American avian influenza viruses, one gene segment from a human influenza virus, and two gene segments that are normally found in swine influenza viruses in Asia and in Europe.

"Parts of it are from the original 1918 virus, parts are similar to the current circulating H1N1 strain from seasonal flu," says Sasisekharan..."

So masks *do* work...

An earlier post expressed skepticism about how effectively masks prevent transmission of flu. Now a clinical trial has shown they can prevent transmission of disease from others to the wearer:

Worried about the flu? Get a mask: Scientific American Blog:

"All of the participants were initially healthy but at risk for catching viruses from their children, who had documented cases of respiratory illness. The researchers found that, after a week, the non-mask wearers were four times more likely to catch a variety of viruses, including the common cold and flu, than those who wore them properly (meaning they strapped them on whenever they happened to be in the same room as their sick children). The masks appeared to be equally effective."

If I'm reading correctly, plain old surgical masks were just as effective at preventing transmission as the more expensive P2 masks, which are "specially designed to filter out water droplets containing viruses".

A few comments on pandemic influenza - fluidinfo

A few comments on pandemic influenza:

"The acting-director of the CDC has already said: "There are things that we see that suggest that containment is not very likely." That is a remarkably candid statement. I think it's very clear that the cat is out of the bag."

via Tim O'Reilly

Influenza Recap

Fear, anger and fatalism over swine flu in Mexico:

"But it may be too late to contain the outbreak, given how widespread the known cases are. If the confirmed deaths are the first signs of a pandemic, then cases are probably incubating around the world by now, said Dr. Michael Osterholm, a pandemic flu expert at the University of Minnesota.

No vaccine specifically protects against swine flu, and it is unclear how much protection current human flu vaccines might offer.

In Mexico City's Zona Rosa neighborhood, teenagers with spiky hair and tight jeans laughed at the danger.
'People are giving too much importance to something that isn't that big of a deal,' said Oscar Zarate, 19, shouting over the loud music and the jostling crowd outside a packed night club."

It will be interesting to see how this develops.

About a year and a half ago I was involved in developing a web application for visualizing results of pandemic simulations. It was intended as a disaster-preparedness application for governments and large (esp. multinational) corporations. The effects of all of the mitigations discussed in the article -- travel restriction, vaccinations, antiviral therapies, etc. -- could be compared in the web app.

Mitigation

I'm a little skeptical of the effectiveness of surgical masks, which seem to be a big feature of the initial response. Are masks really effective at keeping small viruses out of suspension in the atmosphere? The CDC says only that:

"Very little is known about the benefits of wearing facemasks and respirators to help control the spread of pandemic flu."

Recent experiments with guinea pigs showed that coughing and sneezing were not necessary for transmission of flu virus. Instead, atmospheric temperature and humidity determined how well the virus could be suspended in air, which is where it ended up as a result of normal exhalations by the test animals.

Those experiments, incidentally, were inspired by a report of events in southwestern New Mexico:

In an article published in 1919 that details the progression of the 1918 influenza epidemic at Camp Cody, New Mexico, the authors describe a parallel outbreak of pneumonial disease among their laboratory guinea pigs.

"The Great Influenza" gave a good account of the mitigations attempted in 1918 -- including use of surgical masks -- and of their effectiveness. The only thing which really worked then was quarantine. For example, Gunnison County Colorado escaped the 1918 flu with only two deaths, thanks to its stringent quarantine efforts. Silverton lost 10% of its population.


Early Detection

One question which our web application failed to answer was: how do you know when an outbreak is underway, i.e. how do you know when to start mitigation?

It's interesting that Google Flu Trends seems not to have picked up on this outbreak. Granted GFT monitors trends in the United States alone; but the CDC is already aware of cases in California and Texas.

Further Reading -- er, Viewing

PBS has produced some good documentaries on the 1918 flu.

"American Experience: Influenza 1918" (website) focuses on what it was like to live through the pandemic.

Although it's somewhat dated (IIRC it precedes successful recreation of the 1918 virus), "Secrets of the Dead: Killer Flu" (website) is a fascinating detective story about efforts to find the origins, and the keys to the effectiveness, of the 1918 flu.