Document 2313
 DOCN  M94A2313
 TI    Prediction on HIV and AIDS incidence in Japan.
 DT    9412
 AU    Kakehashi M; Anderson RM; Department of Public Health, Hiroshima
       University School of; Medicine, Japan.
 SO    Int Conf AIDS. 1994 Aug 7-12;10(1):332 (abstract no. PC0261). Unique
       Identifier : AIDSLINE ICA10/94370262
 AB    OBJECTIVE: To predict the numbers of AIDS cases and HIV infected persons
       in Japan. METHODS: Back calculation for short term prediction and
       mathematical modeling for long term prediction were performed. Annual
       reported numbers up to 1992 in Japan formed the basis for prediction.
       The data were pooled into homosexual and non homosexual (heterosexual,
       here after) groups. In the back calculation we assumed parametric form
       of HIV incidence trend. Confidence intervals were calculated by
       bootstrap method. The mathematical model covers heterosexual
       transmission involving pair formation and prostitutes. We carried out
       computer simulation setting the parameters in plausible ranges. RESULTS:
       1) The result of back calculation for heterosexual group is illustrated
       in Fig. 1. The exponential trend seemed fit better to the observed HIV
       incidence than the quadratic exponential one does. Neither trend fitted
       in homosexual group. 2) According to the computer simulation, the steady
       state will be attained after many decades. Exponential growth in earlier
       stage seemed justified except for very early years. DISCUSSION AND
       CONCLUSIONS: The number of heterosexual AIDS cases was predicted to
       increase while the trend in homosexual cases was more difficult to
       ascertain. Accumulation of observed data and adjustment for reporting
       delay will increase precision of the prediction. TABULAR DATA, SEE
       ABSTRACT VOLUME.
 DE    Acquired Immunodeficiency Syndrome/*EPIDEMIOLOGY/TRANSMISSION  Adult
       Computer Simulation  Confidence Intervals  Female
       Homosexuality/STATISTICS & NUMER DATA  Human  HIV
       Infections/*EPIDEMIOLOGY/TRANSMISSION  HIV Seroprevalence/*TRENDS
       Japan/EPIDEMIOLOGY  Male  Models, Statistical  Sex Behavior  MEETING
       ABSTRACT

       SOURCE: National Library of Medicine.  NOTICE: This material may be
       protected by Copyright Law (Title 17, U.S.Code).