Chin J, Sato P A, Mann J M
Global Programme on AIDS, World Health Organization, Geneva, Switzerland.
Bull World Health Organ. 1990;68(1):1-11.
After the recognition of AIDS (acquired immunodeficiency syndrome) in the early 1980s, uncertainty about the present and future dimensions of HIV (human immunodeficiency virus) infection led to the development of many models to estimate current and future numbers of HIV infections and AIDS cases. The Global Programme on AIDS (GPA) of the World Health Organization (WHO) has developed an AIDS projection model which relies on available HIV seroprevalence data and on the annual rate of progression from HIV infection to AIDS for use in areas where reporting of AIDS cases is incomplete, and where scant data are available to quantify biological and human behavioural variables. Virtually all models, including the WHO model, have projected large increases in the number of AIDS cases by the early 1990s. Such short-term projections are considered relatively reliable since most of the new AIDS cases will develop in persons already infected with HIV. Longer-term prediction (10 years or longer) is less reliable because HIV prevalence and future trends are determined by many variables, most of which are still not well understood. WHO has now applied the Delphi method to project HIV prevalence from the year 1988 to mid-2000. This method attempts to improve the quality of the judgements and estimates for relatively uncertain issues by the systematic use of knowledgeable "experts". The mean value of the Delphi projections for HIV prevalence in the year 2000 is between 3 and 4 times the 1988 base estimate of 5.1 million; these projections have been used to obtain annual estimates of adult AIDS cases up to the year 2000. Coordinated HIV/AIDS prevention and control programmes are considered by the Delphi participants to be potentially capable of preventing almost half of the new HIV infections that would otherwise occur between 1988 and the year 2000. However, more than half of the approximately 5 million AIDS cases which are projected for the next decade will occur despite the most rigorous and effective HIV/AIDS prevention efforts since these AIDS cases will develop in persons whose HIV infection was acquired prior to 1989. The Delphi projections of HIV infection and AIDS cases derived from the WHO projection model need to be periodically reviewed and modified as additional data become available. These projections should be viewed as the first of many attempts to develop estimates for planning strategies to combat the HIV/AIDS pandemic in the 1990s.
20世纪80年代初艾滋病（获得性免疫缺陷综合征）被确认后，由于对人类免疫缺陷病毒（HIV）感染的现状和未来规模存在不确定性，促使人们开发了许多模型来估计当前和未来的HIV感染人数及艾滋病病例数。世界卫生组织（WHO）的全球艾滋病规划署（GPA）开发了一种艾滋病预测模型，该模型依赖于现有的HIV血清流行率数据以及从HIV感染发展到艾滋病的年进展率，用于艾滋病病例报告不完整且缺乏数据来量化生物学和人类行为变量的地区。几乎所有模型，包括WHO模型，都预测到20世纪90年代初艾滋病病例数将大幅增加。由于大多数新的艾滋病病例将在已感染HIV的人群中出现，所以此类短期预测被认为相对可靠。长期预测（10年或更长时间）则不太可靠，因为HIV流行率和未来趋势由许多变量决定，其中大多数仍未被充分了解。WHO现已应用德尔菲法预测从1988年到2000年年中HIV的流行率。该方法试图通过系统地利用知识渊博的“专家”来提高对相对不确定问题的判断和估计质量。2000年HIV流行率的德尔菲预测平均值是1988年510万的基础估计值的3至4倍；这些预测已被用于获取到2000年成人艾滋病病例的年度估计数。德尔菲法参与者认为，协调一致的HIV/艾滋病预防和控制计划有可能预防1988年至2000年期间原本会发生的近一半新的HIV感染。然而，尽管采取了最严格和有效的HIV/艾滋病预防措施，但预计在未来十年中仍将出现约500万艾滋病病例中的一半以上，因为这些艾滋病病例将在1989年之前感染HIV的人群中出现。随着更多数据的获取，需要定期审查和修改源自WHO预测模型的HIV感染和艾滋病病例的德尔菲预测。这些预测应被视为在制定20世纪90年代抗击HIV/艾滋病大流行的规划策略估计方面的众多尝试中的首次尝试。
