ObjectiveTo explore the selection problem of independent variables and stepwise regression method for multiple logistic regression analysis. MethodsAccording to the data of the case-control investigation for coronary heart disease, age (X1), hypertension history (X2), hypertension family history (X3), smoking (X4), hyperlipidemia history (X5), animal fat intake (X6), weight index (X7), type A personality (X8), and coronary heart disease (CHD, Y) were analyzed by SPSS 18.0 software. The multiple logistic regression analysis was done and the differences of risk factors were compared among 6 kinds stepwise regression variable selection method. ResultsThe univariate analysis showed that no difference was found between CHD group and non-CHD group in age distribution (P=0.116). But the multivariate logistic regression analysis showed that, comparing to population over 65 years old, age was a protective factor on the low age groups (OR< 45=0.100, 0.000 to 0.484, P=0.020; OR45-54=0.051, 0.003 to 0.975, P=0.048). If the age was defined as categorical variable, the risk factors for coronary heart disease were animal fat intake (X6), type A personality (X8), hypertension history (X5) and age (X1), respectively (P < 0.05). If the age was defined as a continuous variable, the effect of age (X1) was not statistically significant (P=0.053). The common risk factors were intake of animal fat (X6) and type a personality (X8) by six kinds method of stepwise variable selection. In addition, the risk factor also included hyperlipidemia history (X5) (forward-condition, forward-LR, forward-wald), hypertension family history (X3), age (X1) (backward-condition, backward-LR) and hypertension history (X2) (backward-wald). ConclusionStepwise regression method should be used to analyze all the variables, including no statistically significant independent variables in univariate analysis. If the categorical variable is regarded as continuous variables, some information may be lost, and even the risk factors may be missed. When the risk factors are not the same by several stepwise regression variable selection method, it should be combined with clinical and epidemiological significance, as well as biological mechanisms and other professional knowledge.
Objective To analyze the substitution mechanism of surrogate endpoints for traditional Chinese medicine (TCM) clinical efficacy evaluation of chronic heart failure (CHF). Methods To obtain data from the occurrence of surrogate endpoints and cardiogenic death of patients with CHF in 7 hospitals. The causal relationship between surrogate endpoints and cardiogenic mortality was inferred by the Bayesian network model, and the interaction among surrogate endpoints was analyzed by non-conditional logistic regression model. Results A total of 2 961 patients with CHF were included. The results of Bayesian network causal inference showed that cardiogenic mortality had a causal relationship with the surrogate endpoints including NYHA classification (P=0.46), amino-terminal pro-B-type natriuretic peptide (NT-proBNP) (P=0.24), left ventricular ejaculation fraction (LVEF) (P=0.19), and hemoglobin (HB) (P=0.11); non-conditional logistic regression analysis showed that NYHA classification had interaction with NT-proBNP, LVEF, and HB prior to and after adjusting confounders. Conclusions The substitution capability of surrogate endpoints for TCM clinical efficacy evaluation of CHF for cardiogenic mortality are NYHA classification, NT-proBNP, LVEF, and HB in turn, and there is a multiplicative interaction between the main surrogate endpoint NYHA classification and the secondary surrogate endpoints including NT-proBNP, LVEF, and HB, suggesting that when the two surrogate endpoints with interaction exist at the same time, it can enhance the substitution capability of surrogate endpoints for cardiogenic mortality.
Objective To construct and compare logistic regression and decision tree models for predicting systemic inflammatory response syndrome (SIRS) in patients with type B aortic dissection (TBAD) after interventional surgery. Methods A retrospective analysis was conducted on clinical data of TBAD patients at Peking University Shenzhen Hospital from 2020 to 2024. The patients were divided into a SIRS group and a non SIRS group based on whether SIRS occurred within 24 hours after surgery. Multivariate logistic regression was used to analyze the influencing factors of SIRS occurrence in TBAD intervention patients, and a decision tree model was constructed using SPSS Modeler to compare the predictive performance of the two models. Results A total of 742 patients with TBAD were included, including 579 males and 163 females, aged between 27 and 97 (58.85±10.79) years. Within 24 hours after intervention, a total of 506 patients developed SIRS, with an incidence rate of 68.19%. Logistic regression analysis showed that the extensive involvement of the dissection, the surgical time≥ 2 hours, PET coated stents implanted, serum creatinine, white blood cell count, C-reactive protein, monocyte count (MONO), neutrophil count levels elevated, estimated glomerular filtration rate and decreased albumin levels were independent risk factors for SIRS (P<0.05). The decision tree model selected a total of 10 explanatory variables and 6 layers with 37 nodes, among which MONO was the most important predictor. The area under the decision tree model curve was 0.829 [95% CI (0.800, 0.856)], which was better than the logistic regression model's 0.690 [95% CI (0.655, 0.723)], and the difference was statistically significant (P<0.001). Conclusion The incidence of SIRS after TBAD intervention is high, and the decision tree model has better predictive performance than logistic regression. It can identify high-risk patients with higher accuracy and provide a practical tool for early clinical intervention.
ObjectiveTo investigate the correlative factors for the efficacy of surgical treatment for single segment degenerative lumbar spinal disorders. MethodsFrom October 2008 to November 2010, a prospective non-randomized controlled study was carried out on 179 patients who were diagnosed to have L4-5 degenerative lumbar spinal disorders and underwent surgical treatment. Ninety-seven patients were included in our study, including 64 males and 33 females, aged between 21 and 86 years old, averaging 49.0. The follow up lasted for an average of 18.9 (12-27) months. The correlative factors including age, sex, body mass index, preoperative psychological state and degree of low back pain, surgical methods, combination with adjacent segment degeneration and recurrence state were analyzed. Single and multiple-factor Logistic regression analysis was used to determine the relationship between independent factors and surgical results of lumbar degenerative disease. ResultsAt the last follow-up, Japanese Orthopaedic Association scores were improved to 22.40±3.18 with an improving rate of (68.5±15.7)% compared with the preoperative condition (7.61±3.09), and the difference was significant (t=-33.031, P<0.001). Univariate analysis showed that all factors were variables associated with the surgical results excluding sex and age (P<0.05). Multiple-factor logistic regression analysis showed that the preoperative psychological state, combination with adjacent segment degeneration and surgical methods had important impact on the surgical results (P<0.05). ConclusionSurgical treatment of lumbar degenerative disease is effective. The preoperative psychological state, combination with adjacent segment degeneration and surgical methods are important factors associated with the surgical results.
ObjectiveTo establish logistic regression analysis model to evaluate the diagnostic efficacy of breast imaging report and data system (BI-RADS) ultrasound signs in forecasting malignant risk of breast lesions. MethodUltrasound graphic materials of 1 660 breast lesions diagnosed during January to September 2011 were retrospectively studied and standardized by BI-RADS. Pathology results were regarded as gold standard reference. Ultrasound signs with significant efficacy after single-factor logistic regression were evaluated in multi-factor logistic regression model to predict the malignant risk of breast lesions. ResultsEighteen ultrasound signs of breast lesions on BI-RADS were included in the final regression model. Among them, Cooper ligaments stretch, echogenic halo, skin thickening, axillary lymph node abnormalities, structural distortions and speculation had high OR values of 30 or more and had higher specificity than 90%. The diagnosis values of regressions model were high, with a sensitivity of 84.5%, specificity 95.5% and accuracy 91.4%. The area under ROC curve was 0.964 and prediction accuracy was 91.0%. ConclusionsThe logistic regression model based on BI-RADS ultrasound signs of breast lesions has high diagnostic values in detecting breast cancer.
目的 探讨成都地区高尿酸血症发生的危险因素。 方法 收集2009年10月－2010年4月在四川大学华西医院体检中心进行健康体检的36 639人的临床资料，对资料进行单因素分析和多因素logistic回归分析。 结果 进行健康体检的36 639人，其中男21 175人，女15 464人。高尿酸血症患者5 233例，患病率为14.3%。年龄＞50岁、男性、饮酒、糖尿病、高血压病、甘油三酯增高、低密度脂蛋白增高和血清肌酐水平增高与高尿酸血症的发生有关。Logistic回归分析显示男性（OR=13.300，P=0.000）、饮酒（OR=4.219，P=0.009）、糖尿病（OR=3.609，P=0.024）是发生高尿酸血症独立危险因素。 结论 成都地区高尿酸血症的患病率略高于全国平均水平，临床治疗和护理高尿酸血症的患者时应积极控制与高尿酸血症发生密切相关的危险因素。
【摘要】 目的 调查四川大学医院管理MBA项目的学员满意度，分析其影响因素，寻求相关启示，为现代医院管理者决策提供参考依据。 方法 以2006年-2010年四川大学医院管理MBA项目学员为研究对象，采用自制的调查问卷，对参与培训的336名学员进行统一的问卷调查，对结果采用logistic回归分析。 结果 发放问卷336份，回收有效问卷320份，有效回收率95%。86.2%的学员表示对培训项目的效果满意。课程内容的实践性、培训组织管理模式、是否有丰富的个案分析等因素与学员满意度有关（Plt;0.05）。 结论 要做好医院管理MBA的培训项目，需要关注影响学员满意度的因素，需在课程内容设计、培训模式改进、案例资料库的甄选等方面努力，从而设计更符合现代医院管理需要的MBA培训课程。【Abstract】 Objective To investigate the satisfaction degree of students in the MBA programs of Sichuan university, analyze the influential factors for the satisfaction degree. Methods Self-made questionnaire was applied in the investigation on 336 students who attended the MBA program of Sichuan University from 2006 to 2010. The investigation results were collected and analyzed by using logistic regression analysis. Results A total of 320 (89.65%) valid questionnaires were retrieved among all the 336 questionnaires assigned. In all the students, 86.2% were contented with the current situation of scientific training program. The content of the course practice, management of training mode and whether there was a rich case analysis were influential factors for students′ satisfaction degree (Plt;0.05). Conclusions The influential factors for students’ satisfaction degree should be taken into consideration in order to better carry out the MBA hospital management training program. More emphasis should be paid on the course design, training mode adjustment and selection of cases, in order to cater for modern hospital managers.
Objective To investigate the feasibility of diagnosis of potential chronic obstructive pulmonary disease (COPD) patients who cannot finish the pulmonary function test via biphasic CT scan. Methods Sixty-seven male individuals aged 43 to 74 (57.0±5.9) years were divided into a COPD group (n=26) and a control group (n=41). All individuals underwent biphasic quantitative CT scan for calculating the proportion of emphysema, functional small airway disease, and normal component of the whole lung and each lobe. Results Based on principle component analysis, two principal components “imaging feature function 1 and imaging feature function 2” were calculated and analyzed by logistic regression, which found that imaging feature function 1 was an independent risk factor of COPD (odds ratio=8.749, P<0.001), and imaging features function 1 could be used to assist the diagnosis of COPD (area under receiver operating characteristic curve=0.843, P<0.001). Conclusion Imaging features function 1 is an independent risk factor for COPD and can assist the diagnosis of COPD.
Early detection of vascular function plays an important role in the prevention and treatment of cardiovascular diseases (CVDs). This paper reports the main studies of the effectiveness of fingertip temperature curve in digital thermal monitoring (DTM) for predicting CVDs, as well as the relationship between parameters from DTM and pulse wave velocity (PWV) detection. A total of 112 subjects [age (42.18±12.28) years, 50% male, 37 with known CVDs] underwent DTM and PWV detection. Results showed that most of parameters related to CVDs were from the declining stage of the digital thermal signal. Binary Logistic regression models were built, and the best one was chosen by ten-fold validation to predict CVDs. Consistency was great between the detection result of PWV and that of the Logistic model of DTM parameters. Parameters from DTM also contained information for early detecting of vascular stiffness. This study indicates that the fingertip temperature curve in DTM has a potential application for predication of CVDs, and it would be used to access vascular function in the initial stage of CVDs.