Study On The Quality Assessment Of Cost Data In Health Economics Analysi

Objectives:Exploring and establishing a set of methods and frameworks suitable for data quality assessment in health economic evaluation for a longer period of time,and assessing the data quality of cost analysis data of project,Evaluation of Immunization Effect of Hepatitis B Vaccine in Beijing and Analysis of Health Economics.This study classified the cost data from the implementation of the hepatitis B vaccine immunization project in Beijing and then collated and analyzed the data according to the method of quality assessment.The purpose of this study is to provide data quality reference for the evaluation of health economics,and lay a foundation for the accuracy and authenticity of the evaluation of health economics.Contents:This study classifies,collates,and evaluates the cost data since the implementation of the hepatitis B project immunization program in Beijing from 1992 to 2013,and evaluates the cost data based on commonly used assessment dimensions and evaluation indicators of data quality.The data included are the data from the city and county level disease prevention and control center for the 22 years from 1992 to 2013,and the cost of hepatitis B vaccination for medical staff in hospitals and community health service centers.1.The main contents of the research and analysis include the following aspects:(1)Labor costs:Including the labor costs of 18 disease prevention and control centers in Beijing and sub-districts,wage number and its work factor of municipal and county-level disease control centers,staff related to hepatitis B immunization work,and workers related to hepatitis B immunization work;Single intramuscular injection and immunization costs of hospital and community health service center.(2)Vaccine cost:mainly refers to the cost of vaccine acquisition(3)Consumables cost:low-cost consumables costs incurred during vaccination(4)Other costs:Including construction area related data,cold storage related data,vehicle use related data,water,electricity,heating costs related data,conference training information publicity fees,etc.The sources of data used for the assessment are mainly the yearbook of health work of the CDC,work reports,other yearbooks,work reports,and data obtained from investigation methods.2.Data evaluation(1)Data dimensions.The dimensions of this study data assessment are accuracy,integrity,relevance,availability,and consistency.(2)Data indicators.This study quantitatively and qualitatively evaluates the dimension of data quality assessment,in which the accuracy and integrity are quantifiable.Accuracy evaluation indicators include logic error rate,proportion of discrete points,and proportion of abnormal values;Integrity assessment indicators include agency non-reported rate,indicator missing rate.Method:This study uses a literature approach,a descriptive study,and a composite index to perform data quality assessments.It uses the literature method to summarize the relevant literatures of data quality assessment.Based on the comparison of various assessment dimensions,the evaluation dimensions and indicators that are used in the assessment are used to assess the data quality.Accuracy,integrity,relevance,availability,and consistency were selected as assessment dimensions.Calculate the evaluation index for accuracy and completeness,calculate the logic error rate,the proportion of discrete points,the proportion of abnormal values,the rate of unreported institutions,the rate of missing items,and the missing rate of each county.The expert consultation method is used to solve data item index interpretation and data filling problems in data collection.Descriptive research method to describe and analyze data,and intuitively display the overall data and calculate the indicators.This study uses timing diagrams and box plots to visualize the data.The timing diagram can express the changing trend of data and intuitively show the turning point in the process of data change.Box plots can highlight discrete points.The difference between the discrete points and the anomalous values can lead to the gap between the data discreteness and the anomaly.Comprehensive index method used to evaluate data quality of district disease and prevention centers.Calculate the index and comprehensive index for each of the four assessment indicators for each district and county,and then comprehensively evaluate and sort the composite index.Result:1.Data accuracy assessment.Logical error rates range from 0 to 11.11%:number of immunization staff and construction area logic errors are 0;Heating costs have the largest logic error,with a logic error rate of 11.11%.The proportion of discrete points varies from 0 to 7.83%:director and deputy director’s total payroll points are 0,the unit employee water bill has the largest discrete point,which is 7.83%.Outliers account for 2.02%to 5.56%:The chief’s gross salary anomaly accounts for the smallest percentage,which is 2.02%,The largest proportion of outliers are wages for immunization workers,unit employees’ water bills,electricity bills per unit area,and heating costs per unit area.2.Data integrity assessment.The number of unreported organizations varies from 1-2.The unreported rate of agencies such as the director’s salary,hepatitis B immunization worker’s salary,hepatitis B vaccine purchase fee,etc.reached 11.11%.Most organizations with other data have a non-reported rate of 0.There are many missing indicators,with values between 0.00%and 11.36%.Among them,the missing rate of the hepatitis B vaccine purchase cost and the low-value consumable product fee was the highest,reaching 11.36%.3.Data relevance assessment.The data obtained from the survey method is in good agreement with the data required for the research analysis and can be directly used for cost analysis.The relevant data obtained by consulting the relevant literature is not very consistent with the data actually needed.4.Data availability assessment.Agency-supplied data and data from the literature are well available.Field survey data availability is worse than agency availability.5.Data consistency assessment.Outcome coefficients in the Outpatient CDC and reporting data have the best consistency.Conclusion:1.Accuracy,integrity,relevance,availability,and consistency of data are important indicators for evaluating data quality.In the assessment of health economics data,it is necessary to conduct an in-depth analysis of each indicator,as well as a comprehensive analysis of the overall situation of the indicator.In this study,a descriptive statistical analysis method and a comprehensive index method were used to conduct a comprehensive analysis of individual indicators and overall indicators.The conclusions of the study have a high degree of credibility.2.High data accuracy.The logic error of most of the data is small,but there are many logical errors in the data related to the cold storage and the heating and water heating fees,which should be taken seriously.From the aspect of the proportion of outliers,the proportion of outliers calculated by the direct method is below 5.60%.3.The data integrity is within the normal range.The missing rate of data indicators in this study is mostly below 10%,which can be replaced when missing data is used to replace.4.High relevance.From the degree of coincidence,the data collected and the actual data required are highly correlated.Except for the poor agreement between the 4.2%data from the literature source and the required data.5.Good availability.The availability of data from the easy to the difficult is the data obtained by the organization,the annual report of the relevant organization,the literature data,and the site survey data.6.The consistency of some of the data is poor.The results of the consistency of the on-site survey data and the reported data are that only the data in the outer suburbs are highly consiste nt,and the consistency of the other two are relatively low.