Replies

Just a mathematical aside:

If certain key variables are known, it is possible to create a mathematical model that will predict how long it will be until a given disease becomes easily transmissible in humans. The model will not tell you exactly WHEN the change will occur, but it will give you a "best guess" of when (mean, median, and mode), a range of times it may occur in instead, and the level of confidence that the event will occur within a given window of time (i.e., the odds that it will happen during that time period).

As always with statistics, the results will depend entirely on the quality of the data.

The key variables that come to mind are:

1. The SIZE of the subject population. In this case, that means how many infected animals/people there are, and how high the viral load is in the infected animals/people.

COMMENT: It would seem that we have a "large" number of infected birds, and possibly other animals. I am not an epidemiologist and do not know what is considered "large" in that field, but the articles I have read seem to support the conclusion that the current infected population is sufficient to be considered "large". Hence, the virus is currently being given a lot of opportunities to make the changes needed to become readily transmissible in humans.

2. The range of possible genetic changes that would result in a readily human transmissible virus.

COMMENT: We cannot know all of the possible changes that could result in this, but we can certainly make a guess based on comparisons with flu strains that are readily transmissible. NOTE: Most statisticians prefer to use conservative assumptions, and would choose to use only those changes that can be clearly identified when building this type of model.

3. The odds that one of the above identifiable changes will occur when a given virus particle reproduces.

COMMENT: This should be predictable based on the location of the needed changes on the DNA/RNA strand, and how likely the strand is to break at one of those particular points.

Item 3 is probably the key variable in this entire analysis. Early in the course of the disease's evolution the number of needed changes is likely to be quite difficult to quantify. However, when the disease gets to a nearly transmissible stage, which is where H5N1 seems to be now, making these assessments should be possible.

4. How likely it is that an infected animal/person who carries a virus particle that has made this transformation will pass it on.

COMMENT: Further transmission is not likely if the animal dies before a large number of mutated virus particles have been made, or if it is not in contact with other potentially infected hosts.

POSSIBLE HUMAN INTERVENTION:

A. The number of infected animals that comprise variable #1 can be reduced by culling the animal populations that are at risk. This is clearly being attempted, with varying levels of determination.

B. If a genetic transformation sufficient to allow easy human to human transmission, the chain of transmission can be still be stopped if it is caught early enough. If ALL humans who show symptoms of H5N1 are isolated, and those who have been exposed are quarantined, then even a successfully mutated virus variant can be stamped out.

SPECULATION: It certainly seems possible that a readily transmissible human form of H5N1 has begun to spread in a remote part of China. The little we know about China's actions indicates they suspect pigs are now serving as a reservoir for the virus, and that some humans have been infected with a readily transmissible form. Moving 50,000 "medical workers" (i.e., troops) into the province is quite consistent with what I would expect a totalitarian nation to do when faced with this type of crisis.

We will never know how hard China was hit by SARS, but they did manage to eventually stop its spread. Let's hope they can do the same with H5N1. However, since H5N1 is so easily spread by migratory birds, it seems unlikely that it can be kept in the box forever.