Ro K K
Health Serv Res. 1973 Winter;8(4):298-308.
For purposes of developing a more refined basis for prediction of hospital utilization using readily available demographic variables, data for some 9000 patients admitted to 22 short-term general hospitals in the Pittsburgh area are analyzed to determine the relationship of age, sex, and race to hospital use. Significant differences in length of stay and number of services used are found for various combinations of these variables when a form of multiple regression is used that allows for interaction effects among the variables.
为了利用现有人口统计学变量为医院利用率预测建立更精确的基础，对匹兹堡地区22家短期综合医院收治的约9000名患者的数据进行分析，以确定年龄、性别和种族与医院使用情况之间的关系。当使用一种允许变量间存在交互效应的多元回归形式时，发现这些变量的不同组合在住院时间和所使用服务数量上存在显著差异。
