目的　探讨住院病人成本的影响因素 ,并对其进行定量测量。方法　按照国际疾病ICD 9分类编码 ,采用病例组合指数法对医院产出组成进行测量 ;采用STAT ,SPSS软件用最小二乘法、岭回归方法分别建立人均住院成本的回归模型并进行比较 ;对人均住院成本影响因素做相关分析 ;采用岭回归方法建立了引入病例组合指数等 6个变量的人均住院成本的多元回归计量模型。结果　由于引入病例组合指数 ,考虑到了医院产出组成对人均住院成本的影响 ,使该成本分析更趋于合理 ;采用岭回归方法建立的回归模型相对于最小二乘法建立的回归模型而言 ,改善了自变量间多元共线性的效应 ,降低了回归系数的标准误 ,回归模型结果更稳定。结论　定量地测量了病例组合指数等因素对人均住院成本的影响 Objective To explore the influencing factors of inpatient costs and to measure them quantitatively. Methods According to the ICD 9 classification code of the international disease, the hospital output composition was measured by the case-combined index method. The STAT and SPSS software were used to establish the regression model of hospitalization cost per capita by least-squares method and ridge regression method, respectively. Cost-related factors were used for correlation analysis; a multivariate regression model for per capita hospitalization cost, including six variables including case combination index, was established using Ridge regression. As a result of the introduction of case-combination index, the effect of hospital output composition on per capita hospitalization costs was taken into account, which made the cost analysis more reasonable; the regression model established by using the ridge regression method was improved compared to the regression model established by the least-squares method. The effect of multicollinearity between independent variables reduces the standard error of the regression coefficients, and the regression model results are more stable. Conclusion Quantitatively measure the impact of factors such as case combination index on hospitalization costs per capita