Modeling African AIDS

Tom Keske tkeske at mediaone.net
Wed May 17 20:26:14 EST 2000


> infection both to women and to men. Adding to the infectionrate is the
lack
> of will to use condoms by men in the presented study. The presented study
> was only ablut to start but might give some explanations to why the
> computerprogram gives such big difference in ability to calculate the
> American and the African development of spread of AIDS.
> I believe using a computer software to calculate a sexually transmitted
> disease is very difficult then the infection rate seems to be affected by
> sexual practices, which are very difficult to control for since people do
> not easily talk about them, and the variation might be very high.
> Health care would also be expected to interfere with the spread of a
> diseasese. Was that taken into account when designing the software?


The software doesn't have "built-in" assumptions about practices
or infectivity rates, or whether infectivities are fixed or time-varying-
it can handle anything.  You establish the assumptions that you want
it use by way of an interactive dialog.   If you want to try different
data and assumptions, you just plug it in.    The program itself
doesn't change.   I took great care to make sure that the raw
data, and not the software, would be the limiting factor.

I'm making some generous assumptions to err of the side of
"worst case"- no condoms, at all.  Total promiscuity.  Rates of
sexual contact that are as high as for high-risk gay men in
San Francisco in the pre-AIDS late 1970s.

Still, the rates of infection come out to be this low.  It took
on the order of 450 years to get even  9000 of the
100,000 infected.

If African strains are more infectious to heterosexuals, that
would certainly be important to know, and would certainly
have impact on America- they would show up here,
almost inevitably, and portend a much worse future
problem.

I believe that a prominent researcher (I'll have to dig
through my files) has pooh-poohed the reports of African
HIV strains being more dangerous to hetersexuals.  There
are few statement ever made in relation to AIDS
epidemiology that have never been contradicted,
it would seem.

Tom Keske







More information about the Microbio mailing list