Principal component analysis based on parallel analysis used in the comprehensive comparison of clinical departments
Wang Zhouqiang1, Liu Yanqiu2, Wu Xiaoqin3,Yang Jiannan4, Song Zhijian5
1.Office of the President, Affiliated Hospital of Southwest Medical University, Luzhou 646000, China;
2.School of Humanities and Management, Southwest Medical University;
3.Department of Urological Surgery, Affiliated Hospital of Southwest Medical University;
4.Affiliated Hospital of Chengdu University;
5.Net Management Center, Affiliated Hospital of Southwest Medical University
Abstract：Objective To analyze the quality of medical work in 2016 statistically in a certain hospital, to provide reference to parallel analysis of comprehensive strengths among departments for hospital administrators so as to increasingly promote the management level, service level and efficiency, and to provide data support for future business development of the hospital.Methods Ten representative indicators that are able to represent comprehensive strengths were chosen from 4 categories including opening bed number, the number of discharge, and the average hospitalization days etc. SPSS 19.0 software was used to standardize 10 representative indicators, and then vista software was used for parallel analysis, which determined the number of the retained factors. Then, the principal components analysis was used for the comprehensive evaluation of some departments of the hospital, and 2 principal component factors were extracted from 10 representative indicators to study on the impacts of the indicators on the changes of the comprehensive strengths of these departments.Results The comprehensive strength evaluation system in clinical departments contained two principal component factors which were synthetic factor and social benefit and operation cost factor. According to the comprehensive score of principal component analysis, A1, A3 and A17 ranked the top three clinical departments, with advanced technology and better comprehensive strength; A10, A20 and A21 were the last three, with scale insufficiency, staff shortage and underdeveloped technology.Conclusion There is a gap in the comprehensive strength of some clinical departments in the hospital, which needs to be further improved to enhance resource allocation. Principal component analysis(PCA) is a good method to analyze the disparity among departments and find real reasons. The number of retained factors is analyzed with parallel analysis, which ensures the analysis results scientifically and also is a key step. The parallel analysis method provides a new method for extracting the number of factors.
王舟强,刘琰秋,吴小琴,杨建南,宋志坚. 基于平行分析的主成分分析法在临床科室综合比较中的运用[J]. 中国医院统计, 2019, 26(3): 161-164.
Wang Zhouqiang, Liu Yanqiu, Wu Xiaoqin,Yang Jiannan, Song Zhijian. Principal component analysis based on parallel analysis used in the comprehensive comparison of clinical departments. journal1, 2019, 26(3): 161-164.
［6］FABRIGAR LR,WACCALLUM RC,STRAHAN EJ.Evaluating the Use of Exploratory Factor Analysis in Psycholofical Research［J］.Psychological Methods,1999,4(3):272-299.
