A Data Set Study Of Health Information Across Whole Life Covering All Groups

Health informatization constructions at home or abroad started with the hospital information system development, and then advanced to electronic health records building. The ultimate goal is to achieve the integration and sharing of regional health information resource. The data from health information system is huge, complicated and overlapping in content. Lack of unified data collection standard and exchange format makes data sharing among different systems, departments or regions very difficult. And the duplication of data collection also results in a waste of resource. Only determine the data set of health information and standardize the data exchange format, the accuracy of information collection, collation and management can be assured, and then the information sharing can be achieved. All of these will contribute to improve the quality of medical services, adapt to health reform, finally to meet people's growing demand for health.The purpose of this study is to analyse the major health problems and health services across whole life from diffident angles, and to discuss what data should be collected and the extent of importance, necessity and availability of these data. It will provide scientific supports for establishing health information data across whole life covering all groups and help to strengthen information sharing and increase the utilization rate of the health data.Based on the collection and analysis of a vast amount of policy documents, operation form and investigating the running health information system, literature review and brainstorming were used to build the preliminary data set and determine the tentative expression rules of recording items. Delphi study was used to ensure the integrity, scientificity and authority of the content of data set. The final content of the data set and expression rules of recording items were demonstrated by the experts from related areas.1. Data set schemaFirst, multidimensional time series model was constructed, including X, Y, Z axis. X axis represented ten stages of human lifetime. They were fetal gestation, neonatal period, infancy, early childhood, preschool, school age, adolescence, adolescence, middle age and old age. Y axis represented the 11 modules, including the residents' health records, women's health, child health, immunization programs, basic medical care, rehabilitation therapy, management of chronic non-communicable diseases, management of infectious diseases, health education, family planning and registration of vital events. Z axis represented the items corresponding to the special life stage and life events.Secondly, the tree structure of data set was constructed based on the logic components construction of open electronic health records theory. TAccording to certain attributes of information, the data set was divided into multiple levels, and took on a tree structure, which showed progressive and specific step by step. Each recording item was in the bottom of the tree structure.Finally, according to the major health problems, health events and health services of the ten life stages, put different information into the appropriate level, so as to complete the function structure of the data set.2. Determine the content of the data setThe effect of Delphi study was evaluated to show that the result was credible. Because of the similarity of consultation method and its evaluation methods, the counseling effect of one and two categories were showed only. The authority coefficient were (0.83±0.06) and (0.83±0.05) in round 1 and round 2. There was no significant difference (t=0.767, P=0.456). Compared with the first round, the coefficient of variation decreased(P<0.001), and cooperation index increased (P<0.005) in the second round. All of these showed that Delphi study was credible in determining the health information date set. The final determined data set including 9 sub data sets, and contained 81 first level items,534 second level items,3848 third level items,1480 forth level items,553 fifth level items,61 sixth level items and 24 seventh level items. According to the scores gaved by experts, the items of the data set were arranged in a matrix, the more score the item got, the position of the item was closer to the top left corner. And the less it got, the position was closer to the lower right corner.3. Item description rules and rangeEach item of the data set was defined and standardized based on the national standard, the professional standard made by the Ministry of Health or brainstorming. For example, gender description quoted person's sex code of GB/T 2261.1-2003. Medical insurance type quoted the professional standard made by the Ministry of Health.lt should be marked with health insurance category code CV0209.01 of public data element standards for health record. Item description rules and range based on brainstorming should be described as name, definition and range. For example, vaccination type of children wasdefined as the type of the vaccination children received, range code is 1-Basic vaccination,2-repeated vaccination,3-strengthened vaccination,4-Emergent vaccination.The health information data set built based on extensive literature and current policies review, brainstorming and Delphi study is the most complete and systemic up to the present day. It is the basis of developing the health information system across whole life covering all groups. And it will lay a solid foundation for the further development of health informatization.