摘要
为建立农村贫困地区门诊医疗服务利用影响因素的原因结构模型,对各因素进行定性、定量,对全国8省10个国家级贫困县31个乡93个行政村10 790人 1996年和1997年的医疗服务利用进行了调查。以年千人就诊次数作为反映医疗服务利用的结果指标。针对医疗服务利用的影响因素之间关系复杂,对可能的13个影响因素进行了多重共线性诊断。利用岭回归分析层层挑选建模变量,应用通径图分析各影响因素的作用。通径分析表明:年千人患病次数、65岁及以上人口百分比、15岁及以上文盲百分比、职业4个因素对年千人就诊次数有直接作用;性别比例和人均收入2个因素有间接作用。这些影响因素的总作用大小分别为:0.89、0.33、0.32、-0.10、-O.29和0.08。本文运用通径分析结合岭回归分析,对关系复杂的自变量与应变量之间的解释与实际较为符合。
To construct the causal structure model of the influential factors of outpatient medical service utilization (MSU) in poor rural areas of China,an investigation on the health condition and household MSU was carried out in 10 poor counties in China. 10 790 people in 93 villages in 31 towns were visited from March to May, 1997. The annual number of outpatient visits per 1 000 people was used as the result indicator, as well as the dependent variable,of the MSU.To deal with the complication among the factors related to MSU,the multicollearity of the 13 could- be influential factors of the utilization was diagnosed. For there are shortcomings in the design of path diagram, this article adopted ridge regression method before designing the diagram based on specialty knowledge. The four factors - annual prevalence, population percent of the aged 65 and above, population percent of the illiterate aged 15 and above, occupational structure, are proved to be directive influential factors by path analysis and their total influence were 0.89,0.33,0.32. - 0.10 respectively. Sex ratio and per capita income are demonstrated to be indirect factors and their total influence are - 0.29 and 0.08. Path analysis combined with ridge regression to explain the complicated relation between independent and dependent variables was used successfully in this paper.
出处
《中国初级卫生保健》
2000年第6期7-11,共5页
Chinese Primary Health Care
基金
联合国儿童基金会和世界银行资助
关键词
影响因素
医疗服务利用
农村
贫困地区
中国
path analysis, influential factors, medical service utilization(MSU), poor rural areas, ridge regression
