Evaluation On Quality Of Care In Tertiary Cancer Hospitals In China Based On Risk Adjustment

Background:With process of industrialization,urbanization,aging and change of residents’ lifestyle,cancer burden in China is gradually increasing,and domestic demand for cancer care is increasingly prominent.While there is a gap on five-year survival rate of cancer patients between China and developed countries,and among different regions of China.As an indispensable part of medical system,tertiary cancer hospitals have their responsibilities on quality improvement of cancer care in China.Quality management is the core task of hospital management,and quality evaluation is the key step to promote hospital’s quality improvement.High-quality quality evaluation studies for cancer hospitals in China are lacked.Previous studies on quality evaluation for cancer hospitals in China did not adapt frontier evaluation theories and methods,did not use risk adjustment technology,and lacked large-scale empirical verification.Therefore,it is of theoretical and practical value to develop a set of quality evaluation tool,evaluation methods and risk adjustment strategies for tertiary cancer hospitals in China,and to carry out the nationwide empirical investigation.Objectives:Firstly,to develop a set of quality evaluation tool for tertiary cancer hospitals in China.Secondly,to construct and validate risk adjustment models of quality indicators,as well as explore their impact on hospital performance.Thirdly,to adopt the appropriate method to comprehensively evaluate the quality of care in tertiary cancer hospitals in China,and put forward empirical strategies for quality improvement.Methods:(1)The systemic literature reviews related with basic concepts,frameworks,methods and practices of quality evaluation for cancer hospitals were conducted globally,so as to create our quality evaluation framework and clarify collected quality indicators.The modified Delphi method was used to develop a set of indicators for tertiary cancer hospitals in China,and indicator weight was identified using the analytic hierarchical process method and percentage weight method.Finally,related questionnaires were developed for data collection in subsequent empirical practice.(2)Based on the fifth round of "External Evaluation of the National Healthcare Improvement Initiative" project,32 tertiary cancer hospitals in China and 5174 cancer inpatients in surveyed hospitals were enrolled.Multilevel logistic regression was used to construct risk adjustment model for patient satisfaction,health improvement and discharge education indicator.Risk adjustment factors were explored using social-demographic information(gender,age,education level,family income,etc.)and disease-related factors(cancer type,cancer stage,self-reported health status,length of hospitalization,etc.).Besides,hospital-level standardized rate was calculated and the impact of risk adjustment on hospital performance was explored.(3)The comprehensive score method,traditional TOPSIS method and SPA-TOPSIS method were used to evaluate structure quality,process quality,outcome quality and overall quality among 32 surveyed cancer hospitals.Besides,evaluation results based on these three methods were compared using correlation coefficients and scatter plots.Results:(1)Based on Donabedian’s "structure-process-outcome" dimensions and IOM’s "effective,efficient,safe,equitable,accessible and patient-centered" dimensions,a set of quality indicators for tertiary cancer hospitals in China was developed including 3 first-level indicators,15 second-level indicators and 47 third-level indicators.The weights of first-level indicators for structural quality,process quality and outcome quality were 0.1634.0.2970 and 0.5395,respectively.Indicators with top three largest weight among second-level indicators were patient safety(0.2106),treatment effect(0.1634)and diagnosis and treatment standard(0.0918).Indicators with top three largest weight among third-level indicators were five-year survival rate(0.0782),health improvement(0.0744)and patient satisfaction(0.0582).Hospital questionnaire and patient questionnaire were developed to collect indicator data of surveyed cancer hospitals.(2)Risk adjustment models were constructed for patient satisfaction,health improvement,and patient discharge education.The area under receiver operating characteristic curve for all models was above 0.7,and their calibration graphs performed well.The statistically significant factors in risk adjustment models included gender,age,cancer type,cancer stage,self-reported health status,length of hospitalization and actual respondent.The ranking of the hospitals’ performance was altered after risk adjustment,which indicated risk adjustment was of great significance for quality evaluation among tertiary cancer hospitals in China.(3)The structural quality of tertiary cancer hospitals in the eastern region of China is better,the process quality and outcome quality of those in the central region are relatively good,while tertiary cancer hospitals in the western region remain substantial improvement room in structure,process and outcome quality.The three ranking lists of tertiary cancer hospitals using three different comprehensive evaluation methods were relatively consistent(correlation coefficient above 0.9),indicating the robustness of hospital ranking.Pairwise comparison of three ranking lists showed that hospital rank based on SPATOPSIS method was the basically same as that from comprehensive score method,while hospital rank based on traditional TOPSIS method showed substantial difference,which is probably due to the phenomenon in the traditional TOPSIS method that "one hospital was closer to the positive ideal point and meanwhile was closer to the negative ideal point".Compared with the traditional TOPSIS method,the SPA-TOPSIS method may be a better option in this evaluation practice.Conclusion:The development of scientific evaluation tool,the use of risk adjustment method,and the selection of appropriate comprehensive evaluation method are of great value for quality evaluation in tertiary cancer hospitals in China.The evaluation tool,risk adjustment models and comprehensive evaluation methods constructed in this study can provide an important reference for the scientific measurement of quality of care in cancer hospitals in China.There are substantial differences in quality of care among tertiary cancer hospitals in China.A keen attention should be paid by policy makers and hospital administrators and discrepancy of hospital performance remains to be narrowed.