基于省级DRG平台的死亡类与非计划重返类指标评价应用研究

Evaluation application of death indicators and unplanned return indicators based on provincial diagnosis-related groups platform

目的探讨死亡类指标与非计划重返类指标在医疗质量评价上的效果与关系。方法收集2017年某省DRG数据平台31家三级公立综合医院共计836976条病案数据,采用多个死亡类指标(低与中低风险组死亡率、高风险组死亡率、粗死亡率、风险调整死亡率)与非计划重返类指标(31天非计划重返入院率、31天非计划重返手术率)进行相关性比较。结果死亡类指标彼此之间具有相关性,而非计划重返入院率与非计划重返手术率不相关(r=0.305)。非计划重返入院率和低与中低风险组死亡率(r=-0.227)、高风险组死亡率(r=-0.098)、实际死亡率(r=-0.130)、风险调整死亡率(r=0.010)均不相关;且非计划重返手术率和低与中低风险组死亡率(r=0.105)、高风险组死亡率(r=0.030)、实际死亡率(r=-0.004)、风险调整死亡率(r=-0.141)均不相关。结论死亡类指标与非计划重返类指标在实际管理技术层面与评价效果上均不尽相同,两者互为有力补充,可构成理想的质量评价指标组合。在今后应用DRG指标进行医院质量评价时,应予以充分重视。

Objective To explore the relationship between death indicators and unplanned return indicators on healthcare quality evaluation.Methods A total of 836976 medical record data were collected from 31 tertiary public general hospitals in a diagnosis-related groups(DRG)data platform in 2017.Multiple death indices(low and low-risk risk group mortality,high-risk group mortality,crude mortality,and risk adjusted mortality)and unplanned return indices(31-day unplanned readmission rate and 31-day unplanned return to surgery rate)were calculated.Pearson′s correlation coefficient was used to examine the relationships among those indices.Results Death indicators were correlated with each other,but the unplanned readmission rate was not correlated with the unplanned reoperation rate(r=0.305).There was no correlation between unplanned re-entry rate and death rate.The correlation coefficients were as follows:unplanned readmission rate versus low and low-risk group mortality(r=-0.227),versus high-risk group mortality(r=-0.098),versus actual mortality(r=-0.130),versus risk adjusted mortality(r=0.010);unplanned reoperation rate versus low and low-risk group mortality(r=0.105),versus high-risk group mortality(r=0.030),versus actual mortality(r=-0.004),versus risk adjusted mortality(r=-0.141).Conclusions The indicators of death and the indicators of unplanned return are not the same in terms of actual management technology and evaluation effect.They are complementary to each other and can form an ideal combination of quality evaluation indicators.