Blower S M, Hartel D, Dowlatabadi H, Anderson R M, May R M
Department of Pure & Applied Biology, Imperial College of Science & Technology, London, U.K.
Philos Trans R Soc Lond B Biol Sci. 1991 Feb 28;331(1260):171-87. doi: 10.1098/rstb.1991.0006.
A data-based mathematical model was formulated to assess the epidemiological consequences of heterosexual, intravenous drug use (IVDU) and perinatal transmission in New York City (NYC). The model was analysed to clarify the relationship between heterosexual and IVDU transmission and to provide qualitative and quantitative insights into the HIV epidemic in NYC. The results demonstrated the significance of the dynamic interaction of heterosexual and IVDU transmission. Scenario analysis of the model was used to suggest a new explanation for the stabilization of the seroprevalence level that has been observed in the NYC IVDU community; the proposed explanation does not rely upon any IVDU or sexual behavioural changes. Gender-specific risks of heterosexual transmission in IVDUs were also explored by scenario analysis. The results showed that the effect of the heterosexual transmission risk factor on increasing the risk of HIV infection depends upon the level of IVDU. The model was used to predict future numbers of adult and pediatric AIDS cases; a sensitivity analysis of the model showed that the confidence intervals on these prediction estimates were extremely wide. This prediction variability was due to the uncertainty in estimating the values of the models' thirty variables (twenty biological-behavioural transmission parameters and the initial sizes of ten subgroups). However, the sensitivity analysis revealed that only a few key variables were significant in contributing to the AIDS case prediction variability; partial rank correlation coefficients were calculated and used to identify and to rank the importance of these key variables. The results suggest that long-term precise estimates of the future number of AIDS cases will only be possible once the values of these key variables have been evaluated accurately.
建立了一个基于数据的数学模型，以评估纽约市异性传播、静脉注射吸毒（IVDU）和围产期传播的流行病学后果。对该模型进行了分析，以阐明异性传播和IVDU传播之间的关系，并对纽约市的艾滋病毒流行情况提供定性和定量的见解。结果证明了异性传播和IVDU传播动态相互作用的重要性。对该模型进行情景分析，以对纽约市IVDU社区中观察到的血清阳性率水平稳定现象提出一种新的解释；所提出的解释不依赖于任何IVDU或性行为的改变。还通过情景分析探讨了IVDU者中异性传播的性别特异性风险。结果表明，异性传播风险因素对增加艾滋病毒感染风险的影响取决于IVDU的水平。该模型用于预测未来成人和儿童艾滋病病例数；对该模型的敏感性分析表明，这些预测估计值的置信区间非常宽。这种预测的可变性是由于估计模型的30个变量（20个生物行为传播参数和10个亚组的初始规模）的值存在不确定性。然而，敏感性分析表明，只有少数关键变量对艾滋病病例预测的可变性有显著贡献；计算了偏秩相关系数并用于识别和排列这些关键变量的重要性。结果表明，只有在准确评估这些关键变量的值之后，才有可能对未来艾滋病病例数进行长期精确估计。
