应用数据挖掘精确计算重症医学科平均住院日初探

Data mining in accurate calculation of the average length of stay in intensive care units

目的精确计算客观合理的重症医学科平均住院日。方法应用数据挖掘语言提取患者住院周期全信息,以精确计算平均住院日,应用配对t检验分析传统算法得出的平均住院日与本研究算法得出的平均住科日之间的差异。结果2017年重症医学科的平均住科日为6.66天,比平均住院日的12.03天短5.37天,平均住院日与平均住科日之间的差异有统计学意义(t=-13.614,P<0.05)。结论平均住科日能客观合理反映科室真实的工作效率,尤其适用于患者转入转出频繁的重症医学科。

Objective To accurately calculate average length of stay in intensive care unit (ICU) discipline in an objective and reasonable way.Methods Application of data mining language was used to extract all inpatient data during their hospital stay for accurate calculation of the average length of stay. Paired test analysis was used to calculate the difference between traditional average length of stay and the new one.Results The average length of ICU stay in 2017 was 6.66 days,5.37 days shorter than 12.03 days of the average length of stay,with the difference between the average length of stay and that in the ICUs being significant (t =-13.614,P <0.05).Conclusions The average length of ICU stay can objectively reflect the actual working efficiency of the ICUs,especially fitting intensive care units of frequent patient admissions and discharges.