基于关联规则的住院患者跌倒伤害程度特征分析  

Analysis of the characteristics of the degree of fall injuries among inpatients based on association rules

作　　者：刘媛 蔡燕 王聪[2] 蒋艳[2] Liu Yuan;Cai Yan;Wang Cong;Jiang Yan(Geriatric Medicine Center,West China Hospital of Sichuan University/West China School of Nursing,Sichuan University,Chengdu 610041,China;Evidence-Based Nursing Research Office,West China Hospital of Sichuan University,Chengdu 610041,China)

机构地区：[1]四川大学华西医院老年医学中心/四川大学华西护理学院,成都610041 [2]四川大学华西医院循证护理研究室,成都610041

出　　处：《中华现代护理杂志》2025年第11期1442-1448,共7页Chinese Journal of Modern Nursing

摘　　要：目的挖掘住院患者跌倒伤害数据中的关联规则,分析不同跌倒伤害程度的患者特征。方法本研究为临床数据挖掘研究。采用便利抽样法,选取2019年6月—2023年5月四川大学华西医院发生跌倒伤害的228例住院患者为研究对象。收集患者一般资料、疾病资料、跌倒事件特征资料,以伤害程度为后项,运用Apriori算法深入挖掘数据,并结合Fisher精确检验对关联规则进行有效性验证。邀请5名副高级及以上职称的护理专家对挖掘的规则进行专题讨论,筛选出具有临床价值的关联规则。结果228例患者中,女112例、男116例,年龄≥60岁147例。根据伤害程度,将患者分为轻度伤害(n=91)、中度伤害(n=78)、重度伤害(n=59);受伤部位最多见于头面颈部(59.65%,136/228);根据设定的Apriori算法参数,生成680条以伤害程度为后项的强关联规则。去除冗余规则后,保留331条规则,保留规则均通过Fisher精确检验(P<0.05)。最终通过专题讨论,筛选出12条具有临床价值的强关联规则。跌倒重度伤害的特征包括年龄>75岁、四肢或躯干受伤、贫血、合并≥3种慢性病、部分自理、多药共用且使用特殊药物、后夜跌倒等。中度伤害的特征包括头面颈部受伤、男性、年龄<60岁、有头晕史、部分自理、后夜跌倒等。轻度伤害的特征包括年龄≤75岁、合并1~2种慢性病、家属陪护、多药共用且使用特殊药物、低白蛋白、如厕行为等。结论住院患者的一般情况、疾病特点以及跌倒特征等都会对跌倒伤害程度造成影响,不同风险因素交互下的跌倒伤害风险存在差异。Objective To mine the association rules in the data of fall injuries among inpatients and analyze the characteristics of patients with different degrees of fall injuries.Methods This study was a clinical data mining research.Using the convenience sampling method,228 inpatients who suffered fall injuries in West China Hospital of Sichuan University from June 2019 to May 2023 were selected as the research subjects.General information and disease information of the patients,and characteristics of fall events were collected.Taking the degree of injury as the consequent,the Apriori algorithm was used to deeply mine the data,and Fisher's exact test was combined to verify the validity of the association rules.Five nursing experts with the professional title of associate senior or above were invited to conduct a special discussion on the mined rules,and the association rules with clinical value were screened out.Results Among the 228 patients,there were 112 females and 116 males,and 147 patients were aged≥60 years.According to the degree of injury,the patients were divided into mild injury(n=91),moderate injury(n=78),and severe injury(n=59).The most common injured part was the head,face,and neck(59.65%,136/228).According to the set parameters of the Apriori algorithm,680 strong association rules with the degree of injury as the consequent were generated.After removing redundant rules,331 rules were retained,and all the retained rules passed Fisher's exact test(P<0.05).Finally,through a special discussion,12 strong association rules with clinical value were screened out.The characteristics of severe fall injuries included being older than 75 years old,injuries to the limbs or trunk,anemia,having three or more chronic diseases,partial self-care ability,using multiple drugs and special drugs,and falling during the late night.The characteristics of moderate injuries included injuries to the head,face,and neck,being male,being younger than 60 years old,having a history of dizziness,partial self-care ability,and falling duri

关 键 词：意外跌倒 跌倒伤害 伤害程度 住院患者 关联规则 

分 类 号：R473[医药卫生—护理学] R473.5[医药卫生—临床医学]