Gm. Raab et al., BAYESIAN FORECASTING OF THE HUMAN-IMMUNODEFICIENCY-VIRUS EPIDEMIC IN SCOTLAND, Journal of the Royal Statistical Society. Series A. Statistics in society, 157, 1994, pp. 17-30
The most commonly used method for acquired immune deficiency syndrome
(AIDS) forecasting (backprojection) assumes that the incubation distri
bution of the time from infection with human immunodeficiency virus (H
IV) to AIDS is known, and makes no explicit use of information on the
number of individuals who have been infected with HIV. Now that we are
further into the epidemic our predictions are more sensitive to assum
ptions about the incubation distribution which may now be influenced b
y pre-AIDS treatment. At the same time we have gained more knowledge a
bout the HIV infection curve in Scotland as a result of retrospective
testing of stored samples. In addition to AIDS case reports data are c
ollected in Scotland on CD4 counts for HIV positive patients under med
ical care, which give incidence data on CD200 cases (two consecutive C
D4 counts below 200 or an AIDS diagnosis). We have developed Bayesian
methodology to make use of our partial knowledge of both the incubatio
n distribution and the infection curve to make short-term forecasts of
AIDS and CD200 cases. The range of forecasts from this approach inclu
des the uncertainty in our specification of both the incubation distri
bution and the infection curve.