论文部分内容阅读
目的使用Apriori算法查找某院2014年-2015年住院天数超过30天的超长住院患者病案首页各指标中的关联规则,以期能够分析超长住院的内在原因,并且为缩短患者的住院天数提供思路。方法利用R语言的arules包中的生成关联规则的apriori函数,编写程序将数据导入系统,探索选取的2011版病案首页中超长住院日患者的包括性别、年龄、出院诊断、是否手术等20个指标是否存在关联规则,并分析其原因。结果根据编写的程序,共获得329 834条强关联规则,得到如下的规则住院天数为31天~40天且出院诊断首位为Z的患者,其费用一般为50000元以下;呼吸病区的患者住院期间一般需使用抗菌药物;骨科出院患者的年龄多为19岁~30岁;神内病区、呼吸病区、肿瘤病区出院的超长住院患者通常不进行手术治疗;而心外病区、骨科病区、普外病区则通常要进行手术治疗。结论通过关联规则分析,可以找到超长住院的原因,为减少患者的住院天数提供思路。
Objective To use Apriori algorithm to find out the association rules of each index in the long-term inpatient medical record of a hospital from 2014 to 2015 with more than 30 days in 2014, in order to analyze the inherent causes of over-long hospitalization and to provide ideas for shortening the hospitalization days . Methods Using the apriori function of association rules generated in the arules package of R language, the program was compiled to import the data into the system to explore 20 indicators including gender, age, discharge diagnosis and whether or not surgery were performed on the long stay days of the selected 2011 version of the first case Whether there is association rules, and analyze the reasons. Results According to the procedure, a total of 329,834 strong association rules were obtained. The following rules were followed: the days of inpatient stay were 31 days to 40 days and the first diagnosis of patients with Z was outpatient, the cost was generally less than 50,000 yuan; patients in respiratory wards were hospitalized Antibacterial drugs are generally required during the period; orthopedic patients are mostly 19 years old to 30 years old. In-patients with long-term inpatients discharged from in-patients, respiratory patients and cancer patients are usually not treated surgically. Ward, general ward is usually to be operated on. Conclusion Through the analysis of association rules, we can find out the causes of extra-long hospitalization and provide ideas for reducing the number of hospitalized patients.