Bayes判别分析在海南省疟区分层中的应用

本文试对与疟疾流行有关的社会经济和地理诸因素进行Bayes逐步判别分析,进而建立高、中、低三类疟区的判别函数式,以探索疟区分层的新方法。
1990年在海南省12个县选择了不同流行程度的自然村55个,以自然村为调查单位,进行社会经济因素调查,以其33个自然村作为建模样本,22个自然村为考核样本。用包括12项社会经济和地理诸因素的统一调查表,逐户进行访问调查。将收集的资料输入IBM/PC-XT微机进行Bayes逐步判别分析。结果选出地形地貌、反映经济水平的劳动力比例、人均收入、居住条件,以及疟防知识和人的特殊行为(上山作业过夜者的比例)等6项对疟区分层有判别意义的变量,并以此6项指标建立了代表高、中、低3个不同流行水平的判别函数式。经建模样本的回代检验,并对照原疟区分层,正确判别率为91.0%。又经考核样本的外部检验,正确判别率为77.3%。表明此种判别分析方法在疟区分层中具有实用价值,在特定条件下有可能代替发病率、原虫率、媒介等指标进行疟区分层。

Application of Bayes Discriminatory Analysis to Malaria Stratification in Hainan Province

The present study was conducted to explore a new method for stratification of malaria endemicity. Several socioeconomic and geographic factors deemed appropriate to be employed in association with malaria epidemic potential were deal with Bayes discriminatory analysis.55 local villages in 12 counties of Hainan province were selected for socioeconomic survey, among which 33 villages were taken as the modelling sample and 22 villages were the non-modelling sample. The questionnaire contained a total of 12 relevant items. After socioeconomic inquiry was performed, the data were analysed through Bayes discriminatory theorem. Six important factors ware selected, i.e. topography (X1), proportion of labour forces (X4), income per person (X5), house structure (X8), knowledge of malaria control (X10), frequency of over-night stay in mountain forest (X12). Three discriminatory function formulas referring respectively to hyper-, meso-, and hypo-endemicity were set up. These 3 discriminatory function formulas were further applied to fit the endemicity of 33 model-sampling villages and 22 non-model-sampling villages. The agreement rate were 91.0% and 77.3% respectively.