Influenza predictions

Following yesterday’s [rather idle] post, Alessandra Iacobucci pointed out the sites of Research on Complex Systems at Northwestern and of GLEaM at Indiana University that propose projections on the epidemic. I have not had time so far to check for details on those projections (my talk for this morning session is not yet completed!).

I would have liked to see those maps in terms of chances of catching the flu rather than sheer numbers as they represent population sizes as much as the prevalence of the flu. The rudimentary division of the number of predicted cases by the population size would be a first step.

4 Responses to “Influenza predictions”

  1. Alessandra Says:

    I’m sorry, the parentheses stuck at the end, breaking both links. Here they are

    The nyt link might require a free subscription.

  2. Alessandrs Says:

    GLEaM and RoCS numbers are indeed old and the updates incredibly stops at the beginning of June. Maybe the explanation of this neglect is that their projections for the end of May were too optimistic (or “a bit off”, However the projection models seem interesting to a laywoman like me, and that is why I pointed them out to Christian. The GLEaM team uses long range airline fluxes data, while the RoCS team uses the results of a previous investigation on geographic circulation of dollar bills, whimsically called “Where’s George?”. They failed, but maybe the reason is bad calibration, or the bad quality of the initial data, since they started the simulations shortly after the first alert at the end of April. There is now another group in Toronto working on H1N1 spreading prediction based on air traffic patterns (, apparently with more success, but I could not find their estimations. Anyway, I was not able to came upon no other “sketches” of mathematical models on the H1N1 subject so far.

    The CDC estimation that a million US citizens have already had H1N1 made me panic a lot! I read the whole transcription of the source interview ( I found it unclear in some points. As far as I understood, there is “modeling in progress”, but I couldn’t find any further details. Is the estimation based on phone surveys and symptomatic? If so, it is not clear how they consider the possibility of simply having a cold, or other flu-like viruses.

    I think we have to be informed, alerted, aware, prepared, etc., but it is obvious that no one can measure the full extent of the pandemic within a reasonable error, nor the probability of all the possible mutations and coupling with other viruses.
    That induces enough anxiety already. Freaking out people with apocalyptic declarations, such as that of Britain’s health minister Andy Burnham, who said yesterday that his country could have 100 000 new cases of swine flu every day by August (!!!), is just criminal.

    Now I’ll go panicking in my corner…

  3. Those numbers are old. It’s spreading rapidly within many populations now – here’s a news report with a CDC source estimating that maybe a million Americans have already had it:

    Reported and confirmed cases are orders of magnitude below this because most people aren’t being tested. Other reports I’ve seen suggest it is more contagious than the normal flu but the symptoms are less severe.

    And by the way, thanks for videoing the meeting the other day. It would have worked really well if the sound quality could be improved – perhaps a decent external microphone would do it. I think this should become standard at meetings and seminars to include a broad an audience as possible.

    • Thanks for the update.

      About the videoing of the meeting, I wholeheartedly agree many things could be improved, from the sound to the focus, to having an external camera operated by someone so that the speaker is filmed as well. It was impossible to do by myself, but I will try harder the next time! As you say, this should be part of all conferences. Or may they should even be all conferences, in order to cut on traveling…

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: