Factual · Powerful · Original · Iconoclastic
Initial reports seemed to indicate swine flu was both vastly more contagious and more lethal than seasonal flu. You don't get headlines like: "Bug Outbreak Will Kill Millions Scientist Warns" and "Pandemic Could Kill Up To 120 Million, Warn Experts" for nothing, nor over 300 million references as of May 5 when entering "swine flu" into the Google News search engine. In an earlier blog, I noted that it appears the new flu is far less contagious than the old. But how dangerous is the virus itself?
"What the epidemiologists are seeing now with this particular strain of [swine flu]," Homeland Security Secretary Janet Napolitano said yesterday, "is that the severity of the disease . . . is not stronger than regular seasonal flu."
In fact, swine flu is considerably less severe. Here's how we know.
There are no good data on Mexican cases or deaths and at any rate flu deaths are much higher in underdeveloped countries. It seems only now the media are catching onto the link between poverty and illness.
Our chief concern, naturally, is the U.S, and here we do have good data.
Seasonal flu, according to the Centers for Disease Control and Prevention (CDC), has a death rate between 0.06% and 0.24%. (Others put the figure in the middle at about 0.1%.) The Agency derives its figure from the ratio of estimated annual infections 15 to 60 million to that of the estimated 36,000 deaths.
As this is written, the CDC reports about 400 swine flu cases with no American deaths. (The one death attributed to the U.S. was a Mexican national who sought help here after becoming sick at home.) Therefore, so far we have a denominator but no numerator.
So to provide a numerator, let's assume for sake of argument that one American dies right now. We would therefore have a death rate of 0.25%. That's at the top end of the CDC's seasonal flu range, right? Wrong. Remember, the death rate isn't calculated per case, it's per infection. And we know that most flu infections are too mild for people to seek medical help and hence become part of any database. So how can we figure out how many U.S. swine flu infections there might be? By reference to seasonal flu.
Because seasonal flu is not on the list of nationally notifiable diseases, that must come from information gathered by the World Health Organization and the U.S. National Respiratory and Enteric Virus Surveillance System. Sentinel laboratories across the nation collect suspect blood samples and forward them to these agencies, which forward their findings to the CDC.
This past flu season, detected flu infections were about 40,000. That's our number of confirmed cases. Divide that 40,000 into the CDC estimate of 15 to 60 million U.S. infections and you get a ratio of confirmed cases to unidentified infections of 375 to 1,300 to one.
During a period of heightened alertness you'd expect a lot more people to see a doctor, thus causing this ratio to drop. But even if it were just 100 to one, the U.S. swine flu death rate (again, assuming somebody dies right now) would be 0.0025% - vastly below that of seasonal flu. If ten Americans suddenly dropped dead, it would be 0.025% - still vastly lower than the seasonal flu rate.
And after having 400 cases with no fatalities, it's a calculated bet that ten Americans aren't about to drop dead.