The third of the recently diagnosed H5N1 cases in Egypt has now died, bringing that country’s total to 18 cases with 10 deaths, the largest outside asia, southeast asia or Indonesia. The case count for 2006 now shows more cases (114) and more deaths (79) than any previous year. And the virus was more deadly, at last measured by a case fatality ratio (deaths divided by total confirmed cases). Indeed the number of deaths in 2006 exceeded the total of deaths for 2003, 2004 and 2005 combined (79 versus 78), although the number of cases exceeded last year’s by only 18%, compared to an 88% increase in deaths over 2005.
The difference is the frightful case fatality ratio registered by Indonesia (82%) compared with the other eight countries with confirmed cases (combined CFR of 58%) and the fact that Indonesia had almost half the cases (55 of 114). Thus the apparent increase in case fatality is being driven entirely by the Indonesian experience. Whether this represents a difference in the virus (a difference in virulence in the Indonesian virus) or a difference in case ascertainment (which affects the denominator, the confirmed cases) or a difference in medical management (which affects the numerator, the proportion of cases that die) or all three combined in various ways, we don’t know.
As in previous years, the second half of the calendar year has seen far fewer cases than the first half. This is the usual pattern. From July, there have been 21 human cases in five countries; from January to June there were 93, an apparent steep drop-off. In 2005 the first 6 months toll of 59 was also followed by a drop-off to 38 cases and in 2004 the numbers were 32 and 14 (data from here with additions of the three Egyptian cases in December). Thus the first half of 2006 was worse than the first half of 2005, but the autumn of 2005 was worse than this year. The general pattern is clear, however. Human cases increase markedly in the first half of the year, and the increase starts now, in January, which has been the peak month in each of the past three years (23, 20 and 25 cases, respectively). We don’t know if this January will be a repeat, but we should be prepared and expecting it.
Any epidemiologist will tell you that trying to read the tea leaves with this much data is fruitless, although it doesn’t prevent just about everyone else from trying. There are more cases, true, but there is also a lot more virus around. H5N1 has spread to poultry and bird populations in a wide geographic area. We don’t know whether it has changed in lethality or not. The same can (and should) be said of transmissibility.
But we are in a bird flu earthquake zone. People in California who know they are living over a fault zone, we don’t know and can’t predict whether The Big One will happen today or tomorrow or in ten years, or conceivably never. In California they still build their buildings and bridges to withstand earthquakes that have happened in the distant past and may or may not happen again and who knows when. They don’t write news stories about “fear of earthquakes overblown” or “what ever happened to The Big One?” for good reason. The conditions for an earthquake haven’t gone away in California and people still feel tremors regularly.
Just consider the bird flu case counts and poultry outbreaks another form of tremor.
And rebuild your public health and social service infrastructures “to code.”