基于贝叶斯估计的诊断试验ROC曲线回归模型

目的 阐明基于贝叶斯估计的ROC曲线回归模型.方法 通过实例对比分析,介绍WinBUGS软件ROC曲线回归模型参数估计与应用.结果 基于贝叶斯估计的ROC曲线回归模型不仅可考虑(平衡)协变量对诊断试验结果准确性评价的影响,而且可计算不同协变量取值条件下的ROC曲线下面积;不同先验分布的选取在一定范围内模型参数估计结果较稳定,可作为临床诊断试验结果分析的依据.结论 基于贝叶斯估计的ROC曲线回归模型,可有效地解决受协变量影响的临床诊断试验准确度评价问题.     

决策树模型在手足口病合并脑膜脑炎重症化危险因素中的应用

目的 探讨C5.0模型结合Logistic回归分析（以下简称C5.0组合模型）在手足口病（hand,foot and mouth disease,HFMD）合并脑膜脑炎重症化影响因素分析中的应用价值。方法 整群抽取2015年4月~2017年7月河南郑州某医院收治的324例HFMD患儿,应用SPSS 21.0进行Logistic回归分析,应用SPSS Modeler 18.0建立C5.0模型,利用单因素Logistic回归分析结果建立C5.0组合模型,比较三种方法的差异。结果 重症化危险因素结果分别为:C5.0组合模型显示心率≥ 140次/min、血糖升高、手足抖动与烦躁不安是其危险因素;C5.0模型显示心率≥ 140次/min、血糖升高、意识障碍、手足抖动和呕吐是其危险因素;Logistic回归模型显示心率≥ 140次/min、手足抖动、血糖升高、呕吐、烦躁不安、中性细胞比率升高是其危险因素。三种模型的灵敏度、特异度、约登指数与受试者工作特征曲线（receiver operating characteristic,ROC）曲线下面积分别如下:C5.0组合模型依次为95.7%,94.2%,0.90,0.946;C5.0模型依次为80.7%,88.3%,0.69,0.845;Logistic回归模型依次为98.0%,70.0%,0.68,0.840。结论 C5.0模型结合单因素Logistic回归在预测HFMD合并脑膜脑炎重症化危险因素方面价值优于C5.0模型和Logistic回归模型。     

Logistic回归模型拟合SARS发病及流行特征

目的利用Logistic回归模型预测SARS发病及流行趋势,研究SARS发病与流行特征及相应预防措施对疾病流行的影响,为预防与控制SARS提供科学依据.方法使用Logistic回归模型拟合广东省、香港特别行政区、北京市及山西省累计发病数据,并绘制出SARS流行曲线.结果对上述地区发病人数进行拟合,各地曲线拟合的决定系数R2均大于0.99,预测值与实际值符合,无显著性差异.从SARS流行曲线推算出各地发病高峰及时间分布,预防措施对疾病流行曲线有明显影响.结论 Logistic回归模型对SARS流行的拟合效果好,可用于理论上探讨预防措施对SARS发病及流行特征的研究.     

Logistic回归模型拟合SABS发病及流行特征

目的 利用比Logistic回归模型预测SARS发病及流行趋势，研究SARS发病与流行特征及相应预防措施对疾病流行的影响，为预防与控制SARS提供科学依据。方法 使用Logistic回归模型拟合广东省、香港特别行政区、北京市及山西省累计发病数据，并绘制出SARS流行曲线。结果 对上述地区发病人数进行拟合，各地曲线拟合的决定系数R2均大于0.99，预测值与实际值符合，无显著性差异。从SARS流行曲线推算出各地发病高峰及时间分布，预防措施对疾病流行曲线有明显影响。结论 Logistic回归模型对SARS流行的拟合效果好，可用于理论上探讨预防措施对SARS发病及流行特征的研究。     

logistic回归模型在ROC分析中的应用

目的探讨logistic回归模型在有协变量或多指标联合诊断试验ROC分析中的应用。方法根据疾病状态建立logistic回归模型,通过形成的预测概率或联合预测因子为分析指标,并结合非参数模型和双正态模型建立ROC曲线。结果通过实例阐述了整个分析过程,并说明了该试剂盒的有效性,同时利用两种模型得到了一致的结果。结论ROC分析中结合logistic回归模型简单有效,尤其适用于有协变量或多指标联合诊断试验的分析评价。     

比例优势模型实现ROC分析的方法及其应用前景分析

目的 探讨比例优势模型在ROC分析中的应用前景。方法 比较比例优势模型与双正态模型等经典方法所计算的ROC曲线下面积及其标准误；采用灵敏度残差平方和与决定系数两个指标评价参数模型的拟合优度。结果 在一般情况下，由比例优势模型所得到的Roe曲线指标结果与经典方法很接近；对于有序分类资料和连续型资料，该模型的拟合效果均较好；但由于该模型获得的Roe曲线形状单一，有些情况下该模型的拟合不理想。结论 与经典的方法相比，比例优势模型有其自身的特点，实际应用时应慎重做出选择。     

数据存在共线性时采用主成分回归分析与投影寻踪回归分析的效果比较

目的比较主成分回归分析与投影寻踪回归分析在数据存在共线性时效果之差别。方法利用实际数据从拟合效果和预测效果两方面评价两种建模方法的优劣。结果主成分回归模型的决定系数为0.8172,相对误差绝对值的平均值为6.42%,预测误差的均方为0.61;投影寻踪回归分析各模型的决定系数为0.8851~0.9944,相对误差绝对值的平均值为1.11%~4.81%,预测误差的均方为0.03~0.38。结论本实例数据(存在一定共线性)分析结果表明,投影寻踪回归分析的拟合效果与预测效果均优于主成分回归分析。     

胰十二指肠切除术后腹腔感染风险预测模型的构建及验证

目的 构建一种新型的胰十二指肠切除术(PD)后腹腔感染(IAI)风险预测模型,为临床诊断、治疗提供理论依据。方法 回顾性分析2016年6月-2022年6月在天津市第三中心医院行PD术的243例患者的临床相关资料,利用LASSO回归分析筛选PD术后IAI的危险因素,Logistic回归分析并建立PD术后IAI风险预测模型,C-index、受试者工作特征(ROC)曲线、校准图和决策曲线分析法(DCA)对其进行评估。结果 243例行PD术患者,96例发生了术后IAI,感染率为39.5%; LASSO回归分析筛选出肿瘤大小、胰瘘、胆瘘、腹腔出血和血肌酐(Scr)水平是PD术后发生IAI的危险因素(P<0.05);多因素Logistic回归分析显示,胰瘘、胆瘘、腹腔出血和Cr水平是PD术后IAI的独立危险因素(P<0.05);该模型C-index为0.779(95%CI:0.738～0.821),ROC曲线下面积(AUC)为0.778,显示出良好的区分度,校准曲线显示该模型的预测风险概率与实际发生概率基本吻合,DCA曲线显示该模型应用价值高。结论 PD术后发生IAI的风险较高,LAS...     

应用logistic回归和ROC曲线评价甲状腺球蛋白及其抗体对甲状腺癌和良性甲状腺结节的鉴别诊断

目的应用logistic回归和ROC曲线探讨甲状腺球蛋白(TG)及甲状腺球蛋白抗体(ATG)对甲状腺癌和良性甲状腺结节鉴别诊断的价值。方法以病理学检查结果作为诊断金标准,采用电化学发光法测定147例甲状腺癌患者、220例良性甲状腺结节患者及150例健康者的血清TG和ATG,通过ROC曲线和logistic回归评价TG和ATG单项及两项联合检测结果。结果通过ROC曲线评价TG和ATG,其曲线下面积(AUC)分别为0.704和0.710。TG和ATG两项指标联合应用的诊断效率最好,其值为0.780。结论通过测定血清TG和ATG,并构建logistic回归模型和应用ROC曲线分析,能方便、有效地评价TG和ATG联合检测结果在甲状腺癌和良性甲状腺结节的鉴别诊断价值,有助于术前区分结节的性质,从而指导临床采取合理有效的治疗手段。     

湖南省手足口病普通病例EV71阳性率与重症率的曲线回归分析

目的 探索湖南省2011年1月~2015年12月手足口病普通病例EV71阳性率与重症率之间的关联。方法 应用SPSS 18.0对2011-2014年的肠道病毒71型（enterovirus,EV71）阳性率与重症率建立直线回归模型和曲线回归模型,并用2015年的数据验证模型,应用组内相关系数（intraclass correlation co-efficient,ICC）评价观测值与预测值的一致性。结果 湖南省手足口病普通病例EV71阳性率和重症率的关联,用三次曲线回归模型的拟合优度最高（校正的R2=0.687）,二次曲线回归模型次之（校正的R2=0.594）,直线回归模型最差（校正的R2=0.420）。三次曲线回归模型对重症率的预测值与观测值之间的两因素混合效应模型单个测量绝对一致ICC值为0.497。结论 利用三次曲线回归模型,用普通病例EV71阳性率预测重症率所得预测值与实际值的一致性一般。     

Evaluation of Diagnosis-Based Risk Adjustment Among Specific Subgroups: Can Existing Adjusters be Improved by Simple Modifications?

Managed care organization use risk adjustment systems to allocate resources and evaluate provider performance. Managers of healthcare organizations need statistical methods to determine whether a particular risk adjustment system can be applied successfully to their organization's unique population and setting, and, if not, whether simple modifications can be effective. We demonstrate methods that can be used to evaluate risk adjustment systems in populations and in subgroups within those populations. We evaluate the use of two diagnosis-based risk adjustment systems, Adjusted Clinical Groups (ACGs) and Diagnostic Cost Groups (DCGs), to explain healthcare utilization within three subgroups of veterans who used Department of Veteran Affairs (VA) healthcare services: homeless individuals, individuals with post-traumatic stress disorder (PTSD), and individuals with spinal cord disorders (SCD). ACG and DCG models are modified to better predict mean level of use for each subgroup and explain the variation in use within the group by adding indicators for each of the three conditions. Predictive ratios (PRs) and R-squares are presented within each of the subgroups for base and revised models. Both models performed well for PTSD (PRs = 0.90 and 0.95, DCG and ACG, respectively), while the DCG model fit better for SCD (PRs = 0.93 and 0.72, respectively); both models underpredicted substantially among the homeless (PRs 0.67). Adding indicators for each subgroup forces perfect prediction of mean use within subgroups and substantially improved discrimination within groups. Overall R-squares moderately improved when indicators were added.     

Predicting EuroQol (EQ-5D) scores from the patient-reported outcomes measurement information system (PROMIS) global items and domain item banks in a United States sample

Background  Preference-based health index scores provide a single summary score assessing overall health-related quality of life and are useful as an outcome measure in clinical studies, for estimating quality-adjusted life years for economic evaluations, and for monitoring the health of populations. We predicted EuroQoL (EQ-5D) index scores from patient-reported outcomes measurement information system (PROMIS) global items and domain item banks. Methods  This was a secondary analysis of health outcome data collected in an internet survey as part of the PROMIS Wave 1 field testing. For this study, we included the 10 global items and the physical function, fatigue, pain impact, anxiety, and depression item banks. Linear regression analyses were used to predict EQ-5D index scores based on the global items and selected domain banks. Results  The regression models using eight of the PROMIS global items (quality of life, physical activities, mental health, emotional problems, social activities, pain, and fatigue and either general health or physical health items) explained 65% of the variance in the EQ-5D. When the PROMIS domain scores were included in a regression model, 57% of the variance was explained in EQ-5D scores. Comparisons of predicted to actual EQ-5D scores by age and gender groups showed that they were similar. Conclusions  EQ-5D preference scores can be predicted accurately from either the PROMIS global items or selected domain banks. Application of the derived regression model allows the estimation of health preference scores from the PROMIS health measures for use in economic evaluations.     

Correlates of health-related quality of life in upper aerodigestive tract cancer patients

Reflecting a limited understanding of the definition and determinants of health-related quality of life (HRQoL), the majority of research in this field has concentrated upon the effect of disease- and treatment-related variables. That work specifically investigating HRQoL among upper aerodigestive tract (UADT) cancer patients is no exception to this observation. Treating subject-related and non-subject-related variables separately, the aim of this study was to investigate predictors of global HRQoL rating in a sample of UADT cancer patients, concentrating upon the relative importance of sociodemographic and clinical variables. A cross-sectional study design was used with a sample of 188 UADT cancer patients. Global HRQoL was assessed using the EORTC QLQ-C30 instrument, global domain (global QoL). Other study variables were collected by subject interview and chart review. Two multivariate regression models were independently developed, containing, respectively, subject-rated and non-sunbject-rated variables. In the model containing subject-rated predictors of global QoL, emotional, breathing, physical, financial, pain and appetite problems were significant predictors (F = 14.6, p < 0.0001 and r2 = 0.54). Among non-subject-rated sociodemographic and clinical variables tested, unemployment, older age, female gender, being dentate and a more advanced disease stage predicted worse global QoL rating, while oral as opposed to pharyngeal or laryngeal cancer predicted a better global QoL rating (F = 5.1, p < 0.0001 and r2 = 0.21). In the latter model, a greater proportion of the variance was explained by sociodemographic variables than by clinical variables.     

Multicollinearity in prognostic factor analyses using the EORTC QLQ-C30: identification and impact on model selection

Clinical and quality of life (QL) variables from an EORTC clinical trial of first line chemotherapy in advanced breast cancer were used in a prognostic factor analysis of survival and response to chemotherapy. For response, different final multivariate models were obtained from forward and backward selection methods, suggesting a disconcerting instability. Quality of life was measured using the EORTC QLQ-C30 questionnaire completed by patients. Subscales on the questionnaire are known to be highly correlated, and therefore it was hypothesized that multicollinearity contributed to model instability. A correlation matrix indicated that global QL was highly correlated with 7 out of 11 variables. In a first attempt to explore multicollinearity, we used global QL as dependent variable in a regression model with other QL subscales as predictors. Afterwards, standard diagnostic tests for multicollinearity were performed. An exploratory principal components analysis and factor analysis of the QL subscales identified at most three important components and indicated that inclusion of global QL made minimal difference to the loadings on each component, suggesting that it is redundant in the model. In a second approach, we advocate a bootstrap technique to assess the stability of the models. Based on these analyses and since global QL exacerbates problems of multicollinearity, we therefore recommend that global QL be excluded from prognostic factor analyses using the QLQ-C30. The prognostic factor analysis was rerun without global QL in the model, and selected the same significant prognostic factors as before.     

Regression analysis based on pairwise ordering of patients' clinical histories

When a medical treatment influences a variety of outcomes, describing the global effect of treatment can be difficult. Traditional approaches specify how treatment affects each separate outcome. This can be done with separate models for each outcome, or by using a combined multivariate model. Describing the overall effect of a treatment thus requires combining these separate effects in some fashion and can be difficult to explain. In this paper, I specify a regression model for use with multiple outcomes where the outcome histories for each pair of patients are ranked. Pairs of patients with different lengths of follow-up are evaluated solely over the common follow-up interval. The logit of the probability that the outcome for patient i is better than that of patient j is assumed to depend on a linear function of the difference of the covariate vectors (for example, treatment indicators) for persons i and j. Thus covariates directly affect the entire clinical history, rather than directly affecting specific outcomes that comprise the history. The idea is that ranking outcomes is more relevant and interpretable than statistically combining separate effects. An estimating equations approach for estimation is described and an example of a clinical trial involving patients with heart failure is provided.     

Management of multidimensional clinical scores: which solutions for items aggregation?

BACKGROUND: Clinical evaluation is a multifactorial process producing as many clinical scores as there are clinical dimensions. For a medicoeconomical analysis however, a single global score would be most useful. The aim of this work was to examine methods allowing relative quantification of items on multidimensional clinical scores in order to determine an appropriate solution for weighted items aggregation. METHOD: Several techniques for item weighting, developed as part of a multi-criteria decision-making tool (AMCD Aide Multi-Critère à la Décision), were examined. Two methods were applied to multidimensional assessment of disability in order to estimate the weight that should be assigned to each of its six items before aggregation in a unique score. The AHP method was used with an interview of 20 functional rehabilitation specialists. The Diakoulaki method was based on the observation of thirty patients with chronic back pain and in thirty patients with vascular hemiplegia. RESULTS: The three weighting schemes were very different from the uniform weighting and were not correlated with each other. The items best weighted by the AHP method were "physical independence" and "mobility". For the Diakoulaki method, they were "social integration" and "economic independence" for the chronic back pain patients, and "awareness of the outside world" and "economic independence" for the vascular hemiplegia patients. Application of these three weighting schemes to theoretical patients produced very different assessments of global disability. CONCLUSIONS: Weighting schemes are useful to quantify the relative importance of individual clinical items. The choice of the weighting method is determinant and depends on the objective of the investigator. If the objective is a global score, the first step is to estimate the weight of the individual items, either when applying a validation process for a new multidimensional clinical score, or secondarily to adapt a validated score to a particular protocol.     

疼痛管理护理模式在腰椎间盘突出症患者术后镇痛中的作用

目的：探讨疼痛管理护理模式对腰椎间盘突出症术后疼痛的影响。方法：将笔者所在科收治并需要手术治疗的72例腰椎间盘突出症患者随机分为两组：疼痛管理组（实验组）和传统对照组（对照组）各36例,实验组给予疼痛管理护理模式治疗术后疼痛,对照组按传统护理方法处理;采用口头描述评定法（12级评定法）评估两组患者术后疼痛程度,对照分析两组患者术后切口镇痛效果及腰腿痛缓解程度。结果：实验组患者术后切口疼痛评估得分及腰腿痛评估得分均明显低于对照组,差异具有统计学意义（P〈0．01）。结论：实施疼痛管理模式护理,能显著减轻腰椎间盘突出症患者术后切口疼痛及腰腿痛,获得良好的手术效果,拱,岛患者住院满意度,形成和谐的医患及护患关系,具有极高的临床推广应用价值。     

Economic impact of shifting the locus of care for neuropathic pain from specialists to general practitioners

We developed a decision-analytic model to examine the economic impact of shifting the locus of care for patients with painful neuropathies from specialists to GPs. The impetus for such a shift was assumed to be a formal education program, focusing on the recognition and treatment of neuropathic pain, conducted for GPs. In the model, all patients with neuropathic pain were assumed to initiate care with their GPs and then be referred to specialists and, ultimately, pain clinics as required for adequate pain control. Two alternative scenarios were examined—the “current” arrangement in which most patients were assumed to be referred for treatment by specialists and pain clinics and a “hypothetical” arrangement in which GPs were assumed to play an expanded role in the treatment of neuropathic pain and which, therefore, often precluded the need for referral. The model was populated with clinical, epidemiologic, and economic data from Norway. A total of 34,951 persons in Norway were estimated to seek care for painful neuropathies each year. The formal education program was assumed to cost 1.5 million Kroner (NOK). Shifting the locus of care from specialists to GPs would result in 4,715 additional GP visits, but 12,123 fewer specialist visits and 7,967 fewer visits to pain clinics. This change would result in estimated savings to the Norwegian health-care system in 2004 of 74.1 million NOK (approx. US $11.9 million). A partial shift in the locus of care of painful neuropathies from specialists to GPs may result in substantial cost savings to the Norwegian health-care system.     

老年骨质疏松症疼痛86例中药配合推拿治疗的效果观察

目的：探讨老年性骨质疏松症疼痛患者行中药配合推拿治疗的临床疗效。方法：选取本院2011年7月-2013年7月期间收治的86例老年性骨质疏松症疼痛患者,按照随机数字表法分为对照组和观察组各43例,对照组采用传统中药内服外敷的方法进行治疗,观察组在对照组基础上辅助推拿进行治疗,观察比较两组的临床治疗效果和腰背疼痛改善情况。结果：经过治疗后,观察组疼痛评分明显低于对照组,差异有统计学意义（P〈0.01）,且观察组的总有效率明显高于对照组,差异有统计学意义（P〈0.05）。结论：通过中药配合推拿治疗老年性骨质疏松症疼痛的疗效显著,能有效缓解患者的疼痛状态,值得临床推广应用。