目的 分析四种综合评价方法在医院医疗工作质量评价中的应用价值.方法 2012年1月至2016年1月4年间收集我院包括急诊、门诊、出院病例数在内的共计12项工作内容作为指标进行详细统计,分别使用主成分分析法、逼近理想点法、密切值法和秩和比法对我院医疗工作的质量进行综合评估,同时对4种评价结果进行排序.结果 经综合评价发现,我院连续4年的医疗工作质量逐年提升,研究中采用主成分分析法、逼近理想点法、密切值法评价结果的排序完全一致,显示2015年1月至2016年1月间我院医疗工作质量最高,而2012-2013年间我院医疗工作质量最差;秩和比法在结果排序上存在些许差异,总体来看依然是不断提升的状态.等级一致性检验结果表明4种评价方法具有一致性(χ2=2.649,P>0.05).结论 主成分分析法、逼近理想点法、密切值法和秩和比法具有一致性,均可用于医院医疗工作质量综合评价,在实际应用过程中,可以根据数据类型、目的差异性进行合理的选择.%Objective To analyze the application value of four kinds of comprehensive evaluation methods in the quality evaluation of hospital medical work. Methods From January 2012 to January 2016, a total of 12 work items including emergency department, outpatient department and discharged cases were collected as indicators for detailed statistics. The methods of principal component analysis, technique for order preference by similarity to ideal solu-tion (TOPSIS), rank sum ratio (RSR) and close-value method were used to evaluate the quality of medical work in our hospital, and the results were sorted at the same time. Results The comprehensive evaluation showed that the quality of our hospital's medical work had been increasing year by year for four years, with the same sorting results of principal component analysis, technique for order preference by similarity to ideal solution (TOPSIS), close-val-ue method. Research showed that the quality of medical work was the highest during January 2015 to January 2016, and the quality of medical work was the worst during 2012 to 2013. There was some difference for RSR in the sort-ing results, but still showed a rising trend. The consistency test results showed that 4 evaluation methods were con-sistent (χ2=2.649, P>0.05). Conclusion The methods of principal component analysis, technique for order preference by similarity to ideal solution (TOPSIS), rank sum ratio (RSR) and close-value method were consistent, which can be used for the comprehensive evaluation of hospital medical quality. In the actual application process, the appropriate meth-od can be reasonably selected according to data type and objective differences.
展开▼