Clinical prediction models to predict the risk of multiple binary outcomes: a comparison of approaches

Clinical prediction models (CPMs) can predict clinically relevant outcomes or events. Typically, prognostic CPMs are derived to predict the risk of a single future outcome. However, there are many medical applications where two or more outcomes are of interest, meaning this should be more widely reflected in CPMs so they can accurately estimate the joint risk of multiple outcomes simultaneously. A potentially naïve approach to multi‐outcome risk prediction is to derive a CPM for each outcome separately, then multiply the predicted risks. This approach is only valid if the outcomes are conditionally independent given the covariates, and it fails to exploit the potential relationships between the outcomes. This paper outlines several approaches that could be used to develop CPMs for multiple binary outcomes. We consider four methods, ranging in complexity and conditional independence assumptions: namely, probabilistic classifier chain, multinomial logistic regression, multivariate logistic regression, and a Bayesian probit model. These are compared with methods that rely on conditional independence: separate univariate CPMs and stacked regression. Employing a simulation study and real‐world example, we illustrate that CPMs for joint risk prediction of multiple outcomes should only be derived using methods that model the residual correlation between outcomes. In such a situation, our results suggest that probabilistic classification chains, multinomial logistic regression or the Bayesian probit model are all appropriate choices. We call into question the development of CPMs for each outcome in isolation when multiple correlated or structurally related outcomes are of interest and recommend more multivariate approaches to risk prediction.     

冠状动脉搭桥术后医院感染风险预测模型构建

目的构建冠状动脉搭桥术(CABG)后医院感染的风险预测模型。方法选择2017年6月-2020年6月在郑州市第七人民医院接受CABG治疗的冠心病患者121例,根据术后住院期间是否发生医院感染,分为感染组34例和未感染组87例。采集感染患者临床标本进行病原菌分离和鉴定,收集患者年龄、有无糖尿病、血清白蛋白(ALB)、体外循环时间、术后引流量、引流管留置时间、术后气管插管时间和住院时间等。采用Logistic回归和卡方自动交互检测(CHAID)模型分析CABG术后医院感染的危险因素,受试者工作特征(ROC)曲线检测模型的预测效能。结果 121例患者CABG术后有34例患者发生医院感染,感染率为28.10%;术后医院感染患者共分离病原菌29株,主要为肺炎克雷伯菌(27.59%)、大肠埃希菌(20.69%)、铜绿假单胞菌(17.24%);Logistic回归分析显示,年龄>60岁、ALB<30 g/L、体外循环时间>120 min、引流管留置时间>7 d、术后气管插管时间>24 h、住院时间>30 d均为CABG术后医院感染的危险因素;CHAID模型分析显示,引流管留置时间、体外循环时间、年龄及术后气管插管时间均为CABG术后医院感染的危险因素,模型预测的准确性为72.70%(P<0.05);ROC分析显示,Logistic回归模型预测医院感染的AUC为0.808,显著高于CHAID模型预测的0.640(P<0.05)。结论 Logistic回归模型可以有效预测CABG术后医院感染的发生,CHAID模型可以显示各变量的相互关系,可与Logistic回归模型互补应用于临床风险因素分析。     

大五人格因素与企业职工工作倦怠的关系

目的:探讨大五人格因素与企业职工工作倦怠的关系。方法:采用工作倦怠问卷(MBI-GS)与大五人格问卷调查了506名某企业职工。结果:分层回归分析的结果显示,在控制了人口统计学变量之后,大五人格因素所解释的工作倦怠的方差变异量,在情绪衰竭、玩世不恭和职业效能三个维度分别为28%、24%和17%。具体而言,宜人性是情绪衰竭、玩世不恭和职业效能的有效预测因子;责任感是玩世不恭和职业效能的有效预测因子;情绪稳定性是情绪衰竭和玩世不恭的有效预测因子;而外向性和开放性两个维度对工作倦怠的预测作用均未达到显著水平。结论:大五人格因素能够有效预测企业职工的工作倦怠。     

心理社会因素在糖尿病发生过程中的作用及机理研究 Ⅳ.Ⅱ型糖尿病发生的危险因素分析

目的：分析影响人类Ⅱ型糖尿病发生的危险因素，比较心理社会因素及生物因素的影响程度。方法：对１３１名糖耐量低减和９１名糖耐量正常者平均追踪１８个月，对影响血糖变化和发生糖尿病的人群的社会心理因素及某些生理因素进行Ｌｏｇｉｓｔｉｃ回归分析和相关分析。结果：在对正常转糖耐量低减的回归分析中，年龄和体重指数的影响最大；但在正常和糖耐量低减转糖尿病的两个回归分析中，生活事件强度的贡献量均排在第１位。结论：心理社会因素在一定的生理条件下对糖尿病的发生起作用，其环节可能在糖耐量低减转糖尿病过程。     

Effect of Age and Sex on Stroke Mortality of Young and Middle-aged Adults in China, 2002–2019, and Predictions to 2030

Objective This study aimed to examine the trends in stroke mortality among young and middle-aged adults in China.Methods Data were obtained from the China national vital registration system. Significant changes in mortality were assessed by Joinpoint regression. Age-period-cohort analysis was used to explain the reasons for the changes. Future mortality and counts were predicted by the Bayesian age-period-cohort model.Results Between 2002 and 2019, a total of 6,253,951 stroke mortality in young ...     

Sexual and Reproductive Health （SRH） Problems among the Unmarried Youths in Sub-urban Shanghai： Needs for RH Education and Services

~~Subiects 6、Method A selj。dmtnl。ie)。ed ano。Iv。on。que。t。onna。,,e。u)。。。l。a.\.ondl。Jed among 7 304 youths aged betw。en j 7 and 24。n t。o to。n。V the sub。。。。ban a),。·l叶Shanghai.Results A total灯 18.4%叶。ubje’t。ha。,。/j。e,。sc。ual入e。po。。。enc。d,4%J/t)。lal。。had eoper。en。ed on。or more Indu。。d a/,o)t。on,a。。d cullZ].3yf V s。。bject。。ho j。。。,。。。。。tal八eI户erlence usedcontra’咛ti。es}-e只。liar八.Mult。var。ateLo厂。st。。e厂厂。。,l。…     

急诊冠状动脉内支架植入治疗急性心肌梗死——围术期死亡和近期预后的研究

为评价急性心肌梗死(AMI)患者急诊经皮冠状动脉腔内成形(PTCA)和冠状动脉内支架植入围术期死亡和近期预后的影响因素,通过对74例行急诊冠状动脉内支架植入的AMI,术后6个月内的临床随访和冠状动脉造影随访,进行多因素相关回归分析。结果表明,多支病变患者的糖尿病、高血压和高血脂的合并比例明显高于单支病变患者;总围术期内死亡率为5.4％,剔除合并心源性休克患者后,死亡率为1.4％;12例75岁以上的患者中,2例死于心源性休克,1例死于心脏破裂,5例心源性休克患者死亡3例;术中非致命性合并症的发生率在单支和多支病变组之间差异无显著性(P>0.05);多元回归分析表明血流动力学状态和患者的年龄是决定AMI患者围术期死亡的独立相关因素;多支病变患者术后6个月内心脏事件发生率明显高于单支病变组;术前和术后6个月内的患者左心室射血分数(LVEF)明显改善;而且单支和多支病变组之间术后6个月时的LVEF之间无显著性差异。     

股骨上端解剖学观测与人工股骨头置换

目的：为选择人工股骨头提供可靠的解剖学数据。方法：对２５０例正常成人股骨上端进行了观测,并对大小转子间距与股骨头最大径作直线相关分析。结果：大小转子间距与股骨头最大径呈正相关（左侧ｒ＝０．４３１,右侧ｒ＝０．４４６,Ｐ〈０．０１）,大小转子间距与头颈长基本等长。结论：用大小转子间距估计股骨头最大径准确、实用、方便。     

慢性砷中毒自觉症状和皮肤改变影响因素的Logistic分析

目的探讨慢性砷中毒患者自觉症状、皮肤色素异常丑过度角化的影响因素。方法选择山西大同地区山阴县黑疙瘩乡大营村进行调查问卷及体检,结果进行单因素及多因素Logistic逐步回归分析。结果单因素分析提示,年龄、迁入时间、饮含砷井水时间、吸烟、饮茶等与自觉症状、色素异常及皮肤角化显著相关,水果是保护因素。多因素Logistic逐步回归分析结果表明,年龄、迁入时间是自觉症状的主要危险因素;年龄是色素异常的主要危险因素,水果摄入是保护因素;年龄、迁入时间、饮含砷井水时间是皮肤过度角化的主要危险因素,水果摄入是保护因素。结论自觉症状及皮肤改变主要与砷的撮入及蓄积量有关。砷中毒病区地理环境背景也是影响皮肤改变的因素之一。水果等抗氧化物质的摄入可能起保护作用。     

Adverse childhood experiences and the development of Down syndrome regression disorder

Down syndrome regression disorder (DSRD) is a clinical symptom cluster of acute or subacute neurocognitive regression in otherwise health persons with Down syndrome. The objective of this study was to evaluate if adverse childhood experiences (ACEs) were more prevalent in children with DSRD than those with DS alone. A survey-based, cohort-based study was performed. Caregivers of individuals with DSRD with onset of symptoms between age 10 and 30 years and DS alone were administered the ACEs questionnaire via an online REDCap survey. A total of 159 responses were collected after excluding incomplete surveys and those not meeting criteria for DSRD. Individuals with DSRD were not more likely to experience ACEs (p = 0.18, 95% confidence interval [CI]: 0.43–1.17). In those with ACEs prior to the onset of symptoms, the median time prior was 7 months (interquartile range: 5–10). Individuals with DSRD were more likely to report three or more ACEs (52, 33%) compared to those with DS alone (39, 22%) (p = 0.02, 95% CI: 1.08–2.87). Exposure to ACEs were not predictive of response to particular therapeutic interventions although those with multiple ACEs 3 months prior to the onset of symptoms was associated with lower response rates to benzodiazepines and immunotherapy (p = 0.02, 95% CI: −3.64–−1.13). This study provides preliminary data that individuals with DSRD experience ACEs at a similar rate to individuals with only DS alone, although three or more ACEs, often preceding the onset of symptoms, was more prevalent in individuals with DSRD.