Just for yonks, I grabbed the WHO confirmed Avian Flu cases list and did a little plot of cumulative cases vs time…


The blue curve is cumulative cases; the red curve is cumulative deaths.

The date is the approximate date of WHO report, with 28 Jan 2004 as the zero date (11 cases, 8 deaths). The dates are approximate because I just counted months between the slightly uneven spaced report dates and assigned 30 days per month. Since the data is coarse grained this should not matter over this length time span.

If you fit it with an exponential, the doubling time is just about one year.

If I were feeling energetic I’d fiddle with a proper log-lin fit and see if it is trending to saturation (logistic growth) or faster-than-exponential.

But, to be honest, the data are too scrappy and selection biased (early cases probably overlooked and fraction of deaths overreported) for that to be a useful exercise, yet.
Jan 2007 will be interesting.


  1. #1 kanagawa
    July 17, 2006

    Uhm… The last “data point” you plotted is the Total column. I.e., the sum for all previous data points. You need to exclude that one. Had you done that then you would’ve noticed that, according to this data, actual confirmed cases decreased in 2006 over 2005 numbers.

  2. #2 Steinn Sigurdsson
    July 17, 2006

    yes, this is a cumulative case plot…
    and we’re only half-way through 2006…

New comments have been temporarily disabled. Please check back soon.