论文部分内容阅读
目的:通过处理全国410家样本医院的全身抗感染药季度采购数据,阐述与简单线性回归预测法相比,引入成分数据分析法预测药品费用结构变化的优势。方法:处理2006年1季度至2009年4季度全国410家样本医院的全身抗感染药季度采购数据,采用成分数据分析法预测2010年1季度至2012年4季度用药费用结构;比较该方法与简单线性回归预测法的预测值与实际值间的相异度。结果:成分数据分析法的预测值与实际值间的欧几里德距离与Aitchison距离在2010年1季度至2011年2季度均小于简单线性回归预测法,但在2012年1季度至2012年4季度均大于简单线性回归预测法。结论:与简单线性回归预测法相比,成分数据分析法应用于短期预测医院用药费用结构时可表现出更高的准确性。
OBJECTIVE: To address the advantage of using composition data analysis to predict the change of drug cost structure compared with the simple linear regression prediction method by processing the quarterly procurement data of systemic anti-infectives in 410 sample hospitals across the country. Methods: The quarterly procurement data of systemic anti-infectives from 410 hospitals in the first quarter of 2006 to the fourth quarter of 2009 were processed and the cost structure of medication was predicted from the first quarter of 2010 to the fourth quarter of 2012 using the component data analysis method. Compared with the simple The Degree of Disparity between Predicted and Actual Value of Linear Regression Prediction. Results: The Euclidean distance and Aitchison distance between the predicted value and the actual value of the method of component data analysis are less than the simple linear regression prediction method from the first quarter of 2010 to the second quarter of 2011, but from the first quarter of 2012 to the fourth quarter of 2012 Quarter is greater than the simple linear regression prediction method. CONCLUSIONS: Compared with the simple linear regression prediction method, the method of compositional data analysis can be used to predict the cost structure of hospital medication in the short term, which shows higher accuracy.