Shots Across The Bow

So, is it TERRORISM! or what?

Maybe you should reconsider having a blog category called "science".

Regarding your assertion that the flu season hasn't yet peaked, that conclusion is not supported by the second chart.

Since the points only show level and not rate, it's impossible to tell whether the last point is still on the increasing side of the curve or whether the peak occured between the last two points and the last point is merely higher up on the decreasing side.

We won't know until next month's data comes out.

Stormy, technically, you are correct. It is possible, though highly unlikely, that the peak occurred midweek, and we are on the downside. However, this is not a random phenomena, where we can't make predictions based on previous observations. Based on those observations, namely the shape of curves from previous years, and the relatively short (weekly) sampling period, it is possible that week 51 was the peak, but it is most likely that the peak is still to come.

On the other hand, going from a mathematical point of view, for our purposes we don't have to calculate the exact derivative to determine the rate of increase in infections; identifying the points of inflection and estimating from there works just fine. On that basis, it is safe to say that the rate of increase has not peaked yet.

Saying we won't know until the next data set is released is true, just like saying that we won't know if the sun will come up tomorrow until dawn is also true.

But I'd put money on either of them.

My reason for thinking so is that the in the two other years shown don't slow down significantly before hitting the peak.

The other thing I note is there appears to be an inverse relationship between severity and duration of the flu outbreak; the worse it is, the sooner it ends.

The inverse relationship is directly related to the limited size of the at-risk population I referred to in the post. The harder the bug hits, the faster it races through the pool of vulnerable people, which means it tends to die out quicker, unless it's a strain that affects a wider range of people. Even though the Type A strain is stronger than most, it is still relatively common, so most folks will have some immunity to it.

I see your point about the sharp peaks of the previous years, but I believe the moderate drop is due to the relative weakness of the available vaccine leading to a less dramatic drop in the rate of increase.

But, as you say, this is speculation, and we won't know who is correct until we get a couple more data points.

Take care.

Gee Rich, that was swell! I like how you completely avoided the thrust of my comment, that whatever the flu season is doing this year it completely blows SARS out of the water. Specifically, flu is on track to kill its usual 36,000 Americans this year while the SARS epidemic over which you promoted hysteria managed to kill fewer than 800 people WORLDWIDE.

In any case, once again you ended up shooting off lots of rounds right into the water. As an article from WebMD posted today as I write this (Dec. 31)shows (http://my.webmd.com/content/article/79/96056.htm?z=1728_00000_1000_nb_05), only now MIGHT we reached an "epidemic threshold." Which is to say that I was completely right in my column of Dec. 25th in which I stated, "Flu may become epidemic, but it's not yet."

but wait, there's more! According to the San-Diego Union-Tribune ("http://www.signonsandiego.com/news/metro/20031231-9999_6m31flu.html), again of December 31, "This year, the flu has hit 46 states, prompting CDC to say it is nearing epidemic levels." Sorry Rich, but "nearing" is not "at." So even the CDC is wavering on whether it's epidemic, but Rich Haley, in all his epidemiological wisdom, is a better spokesman for the CDC than the CDC.

Finally, despite my own assertion about your shells missing, one claim of yours did blow me out of the water in the sense of flabbergasting me. You claim that flu having started earlier contributes to your position. What hooey! Flu doesn't begin on a given day like Christmas. It's always earlier or later than average. That has nothing to do with epidemic levels.

Keep growing that beard, Rich, and crawl back into that spider hole!

Michael Fumento

From the CDC data linked in my post comes the following statement:

"During week 51, 7.8% of all deaths reported by the vital statistics offices of 122 U.S. cities were due to pneumonia and influenza. This percentage is at the epidemic threshold of 7.8% for week 51."

Rather than go to secondary sources, since we all know how often reporters get things wrong, I went to the primary source, the CDC, to get my data, and the primary source says plainly that the flu is at an epidemic level.

And the contention that timing has no impact on what levels would be considered epidemic simply ignores reality. The mortality chart produced by the CDC quite clearly shows that the level of infection defined as epidemic varies over time, based on deviation from a seasonal baseline compiled from historical data.

As for SARS and promoting hysteria, the reader is invited to take a look through my archives and try to find any such statements. Just go to the search window and type in SARS. You'll find several pieces on SARS, including one where I said this:
"My problem with Fumento is he was arguing that the reaction to the SARS outbreak was driven by greed, not any genuine concern, and that SARS was a minor problem."

And later this:
"I'm guessing overall mortality will wind up in the 8-9% range, which is significant for modern times, but as long as SARS remains relatively non contagious, will keep it from becoming the next plague."

Yep, that's promoting hysteria all right.

And on a personal note, I'm not allowed to grow a beard. It would interfere with the seal on my respirator.

Regarding SARS, you're right I did say SARS was a minor problem. Fewer than 800 deaths worldwide compared to an average of 36,000 flu deaths annnually in the U.S. alone is a minor problem. End of story. And while I didn't predict it, the dearth of SARS cases so far this season does add to my case.

Regarding flu being epidemic, below is from today's CNN.com. You'll note that even as it says the outbreak is on the decline it also says, "it's not certain if this flu season will be more severe than past outbreaks." In other words, there's STILL no evidence it ever actually reached epidemic threshold. Yes, I know the CDC has gone back and forth on this, but just as they exaggerated AIDS and SARS they also felt the obligation to exaggerate flu for the same reason -- they felt it necessary to bring more attention to the problem. The CDC is a political animal and has been since it made the leap with AIDS. You just fell for it because, as with SARS, you're operating way outside your league. Again, leave the epidemiology to the experts. And cozy up that spider hole.

CDC: Flu outbreak on decline

(CNN) --Influenza activity appears to be on the downturn in the United States, even as federal officials report more than 90 children have died of the flu this season.

The Centers for Disease Control and Prevention said that 38 states still had widespread activty, down from 42 reported previously.

So far, 93 children under the age of 18 have died of the flu this season, with about 60 percent of them under 5 years old. Whether this number is comparable to past outbreaks is hard to tell, the CDC said. The agency normally doesn't track children's flu deaths so no firm figures for the past exist.

The CDC said that further studies on the impact flu has on children were in development, as well as ways to determine risk factors for severe illness and hospitalization.

This year's outbreak began earlier than usual and has spread more widely than usual, but it's not certain if this flu season will be more severe than past outbreaks, or if it just peaked early. On average, flu kills about 36,000 people in the United States each year.

Before the new CDC numbers were released, Dr. Anthony Fauci -- director of the National Institute of Allergy and Infectious Diseases -- said that the flu season's severity and length is never predictable, but there are some signs to look for.

"When you see states that have widespread activity, when they start to turn the corner, then you have a pretty good indication that we're going in the right direction," he said on CNN's "American Morning."

The especially virulent strain of the virus that is predominant this season was not included in this year's flu vaccine. Fauci said efforts were being made to remedy that and to develop new ways of making the vaccine.

"That's the thing that from the research standpoint we are working on very diligently right now," Fauci explained. "Of developing new types of approaches, of being able to quicken and give greater efficiency to the process of being able to identify the strain and getting it into the vaccine in time."



Find this article at:
http://www.cnn.com/2004/HEALTH/01/08/sprj.flu03.flu/index.html

With a significant drop for week 1 of 2004, it looks even more likely that the peak was between those last two data points.

Do I get a cookie? ;>

Sorry stormy, but no cookie. Check out this post for the story straight from the CDC.

As you can see from those charts, week 51 showed the first decrease, where you speculated that the peak may have been between weeks 50-51. In week 52 the flu rate soared, so the peak could not have been between weeks 50 and 51. Since I wrote that post this afternoon, they've updated the link to show week 53, which shows the sharp decrease Fumento is bragging about.

What's cool is that this year's curve looks to be showing the same thing you noticed earlier, that intensity and duration of an outbreak are indirectly proportional.
Take care

No cookie? Curses, foiled again!