Research On The Establishment Of The Data Quality Evaluation Index System For Healthcare Big Data

[Purpose] Healthcare big data has become an important strategic resource,which plays an important role in improving medical quality,promoting personal health,optimizing medical management,improving medical payment,and promoting drug development.data quality is the basis for ensuring that healthcare big data plays its role,and evaluating data quality is an important part of ensuring data quality.At present,the research on the quality evaluation of healthcare big data is relatively scarce,and only pays attention to the results of data quality,ignoring the causes of data quality problems.In order to comprehensively evaluate the quality of healthcare big data and contribute to the development of healthcare big data quality,this study draws on relevant research in china and abroad to construct an evaluation index system,in order to comprehensively and systematically evaluate the quality of healthcare big data.[Methods] This research mainly studies the relevant research in china and abroad through literature research to find the research framework and suitable model,to screen relevant indicators,and to lay a theoretical foundation for the whole research.Through the expert consultation method to verify the scientific and rationality of the indicator system,and it is revised and improved based on expert opinions.Then,based on the expert's score on the importance of the indicators,the analytic hierarchy process is used to determine the weight of each indicator.Finally,the data quality of medical institutions was evaluated through questionnaires,and the sample hospitals were ranked by TOSPSIS method.[Results] Based on the SPO model,this study constructs a primary evaluation index system from three dimensions: structure,process and result.The primary evaluation index system includes 3 first-level indicators,14 second-level indicators and 67 third-level indicators.Then,17 experts in the field of health information were selected for expert consultation,and a total of 14 experts responded.Based on expert consultation,the primary evaluation index system was revised,2 indicators were added,3 indicators were adjusted,and 9 indicators were deleted.The final indicator system included 3 first-level indicators,14 second-level indicators and 60 third-level indicators.The index system is weighted by the analytic hierarchy process.The weights of the three first-level indicators are as follow: structure(0.3167),process(0.2333),and result(0.4500).Based on the index system,the questionnaire was designed to investigate the data quality of 42 medical institutions.The data quality and regional differences of each sample were collected.It was found that the current medical institution data quality has imperfect management system,data reliability and poor accuracy and other questions,and based on the survey results,the data quality of 42 sample hospitals was ranked.[Conclusions] This study explores the application of the SPO model to the construction of the indicator system,and the results show that the indicator system can more comprehensively evaluate the quality of healthcare big data.From the three dimensions of structure,process and coucome is more practical than the traditional focus on data quality results,which can help medical institutions improve the data quality.The evaluation results of the sample medical institutions are consistent with other related research results,indicating that the constructed evaluation index system is scientific and reasonable.