摘要
目的:分析不同年龄段住院患者的共病模式及核心疾病,并探讨共病对住院费用的影响。方法:收集浙江省宁波市医疗中心李惠利医院2018年1月1日至2023年6月30日18岁及以上的成人住院患者电子病历数据,针对53种特定疾病,分析不同年龄组住院患者共病现状。采用关联规则挖掘共病模式,利用复杂网络分析方法探索共病的核心疾病,再进一步采用广义估计方程分析不同共病模式对患者住院费用的影响。结果:359402例成年住院患者的共病患病率为38.51%,其中男性和老年群体的共病患病率相对较高,分别为43.60%和58.29%。关联规则分析挖掘出15种常见的共病模式,并提示在不同年龄组有差异。全住院患者最常见的共病模式为“糖尿病→高血压”(支持度=7.04%,置信度=62.17%,提升度=2.17),青壮年组最常见的共病模式为“血脂异常→慢性肝病”(支持度=1.19%,置信度=53.17%,提升度=6.04);中年组为“糖尿病→高血压”(支持度=4.84%,置信度=50.28%,提升度=2.15);老年组为“冠心病、糖尿病→高血压”(支持度=2.38%,置信度=77.43%,提升度=1.63)。复杂网络分析显示,不同年龄组的共病核心疾病不同。青壮年组共病核心疾病为慢性肝病(度数中心性=50,介数中心性=0.055,接近中心性=0.963),中年组为高血压、慢性肝病、糖尿病(度数中心性=52,介数中心性=0.022,接近中心性=1.000),老年组为高血压、糖尿病、恶性肿瘤、慢性肝病、甲状腺疾病、贫血及心律失常(度数中心性=52,介数中心性=0.009,接近中心性=1.000)。广义估计方程分析结果显示,绝大部分共病模式对住院费用增加的影响显著,且不同共病模式住院费用增加的风险不一致。其中,“心力衰竭→高血压”、“脑卒中→高血压”、“恶性肿瘤、糖尿病→高血压”、“脑卒中、糖尿病→高血压”、“糖尿病、心力衰竭→高血压”等共病模式患者的住院费用为未患目标疾病患者住院费用两倍以上。结论:不同年龄住院患者的共病模式及核心疾病存在差异,不同共病模式对费用增加的影响不同,对不同年龄段、不同共病模式的患者应采取个性化管理和治疗。
Objective:To analyze the multimorbidity patterns and core diseases among hospitalized patients in different age groups and to explore the impacts of multimorbidity patterns on hospitalization costs.Methods:Electronic medical records of adult inpatients(aged≥18 years)from Ningbo Medical Center Lihuili Hospital between January 1,2018,and June 30,2023 were collected.The multimorbidity status involving 53 specific diseases was analyzed across different age groups.Association rule mining was used to identify common multimorbidity patterns.Complex network analysis was used to identify core diseases within the multimorbidity networks.Generalized estimating equations(GEE)were used to analyze the impact of different multimorbidity patterns on hospitalization costs.Results:The prevalence of multimorbidity among the 359402 adult inpatients was 38.51%,with higher rates observed in males(43.60%)and elderly patients(58.29%).Association rule mining identified 15 common multimorbidity patterns,which exhibited differences across age groups.The most prevalent multimorbidity pattern overall was“diabetes→hypertension”(support=7.04%,confidence=62.17%,lift=2.17).In the young adult group,the most prevalent pattern was“dyslipidemia→chronic liver disease”(support=1.19%,confidence=53.17%,lift=6.04).In the middleaged group,it was“diabetes→hypertension”(support=4.84%,confidence=50.28%,lift=2.15).In the elderly group,it was“coronary heart disease,diabetes→hypertension”(support=2.38%,confidence=77.43%,lift=1.63).Complex network analysis revealed that the core diseases within multimorbidity networks differed across age groups.The core disease identified in the young adult group was chronic liver disease(degree centrality=50,betweenness centrality=0.055,closeness centrality=0.963).Core diseases in the middle-aged group included hypertension,chronic liver disease,and diabetes(all with degree centrality=52,betweenness centrality=0.022,closeness centrality=1.000).Core diseases in the elderly group comprised hypertension,diabetes,malignant tumors,chronic liver disease,thyroid disease,anemia,and arrhythmia(all with degree centrality=52,betweenness centrality=0.009,closeness centrality=1.000).Generalized estimating equations analysis indicated that,most multimorbidity patterns were significantly associated with increased hospitalization costs.However,the magnitude of cost increase varied across different multimorbidity patterns.Specifically,hospitalization costs for patients with patterns such as“heart failure→hypertension”,“stroke→hypertension”,“malignant tumor,diabetes→hypertension”,“stroke,diabetes→hypertension”,and“diabetes,heart failure→hypertension”were more than double those of patients without any target diseases.Conclusions:Multimorbidity patterns and core diseases among hospitalized patients differ significantly across age groups,and different patterns exert varying impacts on hospitalization costs.These findings underscore the necessity for age-stratified and multimorbidity pattern specific management strategies.
作者
李涛
徐小林
程杨杨
林开
LI Tao;XU Xiaolin;CHENG Yangyang;LIN Kai(School of Public Health,Zhejiang University,Hangzhou 310058,China;Preventive Healthcare Department,Ningbo Medical Center Lihuili Hospital,Ningbo 315000,Zhejiang Province,China;Geriatric Medicine Department,Ningbo Medical Center Lihuili Hospital,Ningbo 315000,Zhejiang Province,China)
出处
《浙江大学学报(医学版)》
北大核心
2025年第4期423-433,共11页
Journal of Zhejiang University(Medical Sciences)
基金
国家自然科学基金(72474197)。
关键词
共病
住院费用
住院患者
关联规则
复杂网络分析
广义估计方程
单中心研究
Multimorbidity
Hospitalization costs
Inpatients
Association rules
Complex network analysis
Generalized estimating equations
Single-center study
