Comments on A mathematician at risk: Why is h1n1 a big deal? (and why you should get th...

Thanks Jimmy. Yeah, this is pretty much just the p...

2009-11-06T16:08:58.108-08:00

Thanks Jimmy. Yeah, this is pretty much just the preliminary model from back in the day.

I think reducing the infection rate just moves the cdf up and to the left; so pretty much the same effect as immunization though the shape might be a bit different.

I don't really know about which is more effective, but both should certainly increase the herd immunity effects.

2009-11-06T16:07:47.072-08:00

This comment has been removed by the author.

Fantastic job on taking the noro problem and apply...

2009-11-06T15:07:46.564-08:00

Fantastic job on taking the noro problem and applying it to h1n1. I'm truly impressed.

One thought: You assumed that there is a 0.3% chance of infecting any other person. If we further assume that applying a frequent hand-washing regemin throughout the population pushes the chance down to, say 0.2%, what then happens to the curves?

Can then anything be said about hand washing be more or less important than immunization?

~jimmy

Excuse me for being a little defensive but I reall...

2009-10-28T22:07:36.092-07:00

Excuse me for being a little defensive but I really think there are some serious problems with your comment.

I don't think 1 is an unreasonable assumption. And to be fair, I'm not assuming 2; I've constructed a model that supports point 2. A very simple model with acknowledged flaws. But I think it's a pretty shady rhetorical device to call a model result an assumption.

People also used to think that the only reason why some were spared when Noro went through a closed community. Turns out herd immunity was a simpler explanation.

To be frank, I am very skeptical of any theory whose evidence is that there appears to be too much clustering. We see time and time again that people are very bad at recognizing what level of clustering is just an artifact of chance. There are tools to show over clustering and I think it's necessary to go through at least some attempt at rigor when using it as an argument. There's just been too much woo science that has come out of people who think they see patterns in clusters.

The less people who are immune the more likely ...

2009-10-28T21:31:22.067-07:00
The less people who are immune the more likely an outbreak is to happen and the larger that outbreak is expected to be.

There are a couple of assumptions that a lot of people are making:
1. That because a virus is new, fewer people are immune
2. That because people lack immunity, it is more likely to spread through the population.

Although you realize and stated that your epidemiological model doesn't represent reality, many people think that the scenario you presented accurately depicts how influenza actually spreads. In reality, H1N1 doesn't appear to spread this way at all. Some people seem to get it out of the blue, while others get it along with dozens of others in close proximity. This seems to suggest that there is either an existing immunity, or very selective transmission. Either way it's difficult to accurately predict the spread.