The World Health Organization approach: projections of non-paediatric HIV infection and AIDS in pattern II areas.
A model based on HIV prevalence rates that can be easily used to provide shortterm projections of nonpediatric AIDS cases in a specific population is proposed. The drawbacks are the limited shortterm scope of projecting infections and cases, the lack of epidemiological features, and the inability to predict consequences of behavioral and social changes in the population. Data requirements are HIV serological survey data and current information on the epidemiology and natural history of HIV infection. The approach uses back decomposition of HIV prevalence at a given point into incident infections. Incubation and survival parameters provided by the Population Division of the Department of International Economic and Social Affairs of the UN Secretariat are used to estimate AIDS cases. 4 epidemiological assumptions are made: 1) that HIV infection began to spread at the end of the 1970s and there were few HIV positive cases in 1980, 2) that the rate of increase was the greatest in the early and mid-1980s, 3) that the cumulative prevalence cure of HIV infection is more likely to be logistic in shape, and 4) that the role of competing causes of adult mortality is not included in the model. For shortterm projections of <3 years, the following assumptions are made: 1) transfer rate from HIV infection to clinical AIDS is 22% within 5 years, 50% within 10 years, and 77.5% within 15 years; 2) the progression function from HIV infection to AIDS is linear, beginning at 5.5% from the 2nd year of infection and a median at 10 years; 3) there are no significant changes in urban/rural and age distribution of HIV infection; and 4) AIDS cases survive a median of 1 year from diagnosis with a limit of 5 years' survival. Estimates were derived using combined 1985 data from Kenya, Malawi, Uganda, and Tanzania. Projections are based on the year 1989. Hypothetical population totaled 16.1 million, with 42% of the urban population <15 years and 50% of the rural population <15 years. Prevalence rates and hypothetical population are given by 5-year age groups and urban/rural status. Cumulative mortality rates are 50% after 1 year, 80% after 2, 89% after 3, 92% after 4, and 100% after 5 years. The results were that 12% of the urban population and <1% of the rural population were HIV infected. 547,000 were HIV positive in 1989 of which 82% were >5 years.