Comparison of Cox proportional hazards regression and generalized Cox regression models applied in dementia risk prediction

Abstract Background Cox proportional hazards regression models are used to evaluate associations between exposures of interest and time-to-event outcomes in observational data. When exposures are measured on only a sample of participants, as they are in a case-cohort design, the sampling weights must be incorporated into the regression model to obtain unbiased estimating equations. Methods Robust Cox methods have been developed to better estimate associations when there are influential outliers in the exposure of interest, but these robust methods do not incorporate sampling weights. In this paper, we extend these robust methods, which already incorporate influence weights, so that they also accommodate sampling weights. Results Simulations illustrate that in the presence of influential outliers, the association estimate from the weighted robust method is closer to the true value than the estimate from traditional weighted Cox regression. As expected, in the absence of outliers, the use of robust methods yields a small loss of efficiency. Using data from a case-cohort study that is nested within the Multi-Ethnic Study of Atherosclerosis (MESA) longitudinal cohort study, we illustrate differences between traditional and robust weighted Cox association estimates for the relationships between immune cell traits and risk of stroke. Conclusions Robust weighted Cox regression methods are a new tool to analyze time-to-event data with sampling, e.g. case-cohort data, when exposures of interest contain outliers.

Download Full-text

Gene selection in Cox regression model based on a new adaptive penalized method

International Journal of Advanced Statistics and Probability ◽

10.14419/ijasp.v8i1.30566 ◽

2020 ◽

Vol 8 (1) ◽

pp. 16

Author(s):

Oday Isam Alskal ◽

Zakariya Yahya Algamal

Keyword(s):

Regression Model ◽

Survival Data ◽

Proportional Hazards ◽

Gene Selection ◽

Cox Regression ◽

Area Under The Curve ◽

Cox Proportional Hazards ◽

Cox Proportional Hazards Regression ◽

Proportional Hazards Regression ◽

Selection Operator

The common issues of high dimensional gene expression data for survival analysis are that many of genes may not be relevant to their diseases. Gene selection has been proved to be an effective way to improve the result of many methods. The Cox proportional hazards regression model is the most popular model in regression analysis for censored survival data. In this paper, an adaptive penalized Cox proportional hazards regression model is proposed, with the aim of identification relevant genes and provides high classification accuracy, by combining the Cox proportional hazards regression model with the weighted least absolute shrinkage and selection operator (LASSO) method. Experimental results show that the proposed method significantly outperforms two competitor methods in terms of the area under the curve and the number of the selected genes.

Download Full-text

Prognostic impact of a lymphocyte activation-associated gene signature in GBM based on transcriptome analysis

PeerJ ◽

10.7717/peerj.12070 ◽

2021 ◽

Vol 9 ◽

pp. e12070

Author(s):

Yujia Lan ◽

Erjie Zhao ◽

Xinxin Zhang ◽

Xiaojing Zhu ◽

Linyun Wan ◽

...

Keyword(s):

Regression Models ◽

Proportional Hazards ◽

Lymphocyte Activation ◽

Risk Groups ◽

Gene Signature ◽

Cox Proportional Hazards ◽

Low Risk ◽

Training Dataset ◽

Cox Proportional Hazards Regression ◽

Proportional Hazards Regression

Background Glioblastoma multiforme (GBM) is a highly, malignant tumor of the primary central nervous system. Patients diagnosed with this type of tumor have a poor prognosis. Lymphocyte activation plays important roles in the development of cancers and its therapeutic treatments. Objective We sought to identify an efficient lymphocyte activation-associated gene signature that could predict the progression and prognosis of GBM. Methods We used univariate Cox proportional hazards regression and stepwise regression algorithm to develop a lymphocyte activation-associated gene signature in the training dataset (TCGA, n = 525). Then, the signature was validated in two datasets, including GSE16011 (n = 150) and GSE13041 (n = 191) using the Kaplan Meier method. Univariate and multivariate Cox proportional hazards regression models were used to adjust for clinicopathological factors. Results We identified a lymphocyte activation-associated gene signature (TCF3, IGFBP2, TYRO3 and NOD2) in the training dataset and classified the patients into high-risk and low-risk groups with significant differences in overall survival (median survival 15.33 months vs 12.57 months, HR = 1.55, 95% CI [1.28–1.87], log-rank test P < 0.001). This signature showed similar prognostic values in the other two datasets. Further, univariate and multivariate Cox proportional hazards regression models analysis indicated that the signature was an independent prognostic factor for GBM patients. Moreover, we determined that there were differences in lymphocyte activity between the high- and low-risk groups of GBM patients among all datasets. Furthermore, the lymphocyte activation-associated gene signature could significantly predict the survival of patients with certain features, including IDH-wildtype patients and patients undergoing radiotherapy. In addition, the signature may also improve the prognostic power of age. Conclusions In summary, our results suggested that the lymphocyte activation-associated gene signature is a promising factor for the survival of patients, which is helpful for the prognosis of GBM patients.

Download Full-text

Development and validation of a prognostic nomogram for lower-grade glioma based on an autophagy-related lncRNA signature

Archives of Medical Science ◽

10.5114/aoms/145435 ◽

2022 ◽

Author(s):

Jiazhe Lin ◽

Nuan Lin ◽

Wei-jiang Zhao

Keyword(s):

Proportional Hazards ◽

Cox Regression ◽

Cox Proportional Hazards ◽

Differentially Expressed ◽

Survival Prediction ◽

Lower Grade ◽

Lasso Regression ◽

Cox Proportional Hazards Regression ◽

Proportional Hazards Regression ◽

Kaplan Meier

IntroductionGliomas account for 75% of the primary malignant brain tumors. The prognosis and treatment planning vary in lower-grade gliomas (LGG) due to their heterogeneous clinical behaviors. The dysregulation of autophagy-related (ATG) lncRNAs plays a crucial role in LGG. We aimed to develop and validate an ATG lncRNA risk signature, and a survival nomogram with integration of novel prognostic for LGG patients.Material and methodsDifferentially expressed ATG lncRNAs were screened out based on TCGA and GTEx RNA-seq databases. ATG lncRNA prognostic signature was then established by Kaplan–Meier, univariate Cox proportional hazards regression, Least absolute shrinkage and selection operator (LASSO) regression and multivariate Cox proportional hazards regression, with its predictive value validated by time-dependent receiver operating characteristic (ROC) curves. Kaplan–Meier, univariate Cox regression and multivariate Cox proportional hazards regression were used to screen out clinical and molecular variables. A nomogram was developed and internally validated by ROC and calibration plots.ResultsAn ATG lncRNA risk signature was constructed with six differentially expressed lncRNAs (LINC00599, LINC02609, AC021739.2, AL118505.1, AL354892.2, and AL590666.2). Based on the risk signature, a nomogram was developed by addition of the significant prognostic clinical variables (age and grade) and molecular variables (IDH status and MGMT status).ConclusionsWe identified an ATG lncRNA risk signature and develop a nomogram for individualized survival prediction in LGG patients. A user-friendly free online calculator to facilitate the use of this nomogram among clinicians is also provided: https://linstu2009.shinyapps.io/LGGPRODICTORapp/?_ga=2.3154800.1506830296.1588641469-159983587.1588641469.

Download Full-text

Gaining more flexibility in Cox proportional hazards regression models with cubic spline functions

Computer Methods and Programs in Biomedicine ◽

10.1016/s0169-2607(97)00043-6 ◽

1997 ◽

Vol 54 (3) ◽

pp. 201-208 ◽

Cited By ~ 182

Author(s):

Harald Heinzl ◽

Alexandra Kaider

Keyword(s):

Regression Models ◽

Proportional Hazards ◽

Cubic Spline ◽

Cox Proportional Hazards ◽

Spline Functions ◽

Cox Proportional Hazards Regression ◽

Proportional Hazards Regression

Download Full-text

MO509DOSE-RESPONSE RELATIONSHIP BETWEEN ALCOHOL CONSUMPTION AND THE RISK OF DEVELOPING CKD: RETROSPECTIVE COHORT STUDY

Nephrology Dialysis Transplantation ◽

10.1093/ndt/gfab087.0029 ◽

2021 ◽

Vol 36 (Supplement_1) ◽

Author(s):

Yusaku Hashimoto ◽

Takahiro Imaizumi ◽

Akihiro Hori ◽

Sawako Kato ◽

Yoshinari Yasuda ◽

...

Keyword(s):

Alcohol Consumption ◽

Dose Response ◽

Regression Models ◽

Proportional Hazards ◽

Cox Proportional Hazards ◽

Response Relationship ◽

Dose Response Relationship ◽

Cox Proportional Hazards Regression ◽

Proportional Hazards Regression ◽

Incidence Of Ckd

Abstract Background and Aims Drinking habits are one of the most important modifiable lifestyle factors to prevent the development of chronic kidney disease (CKD). Previous studies showed that it was inversely associated with the risk of developing CKD, but the dose-response relationship between alcohol consumption and the development of CKD is still controversial. In the present study, we aimed to examine whether the amount of alcohol consumed at one time is associated with new onset of CKD in general population. Method Study subjects were 11,162 Japanese aged 45 to 74 years, with an estimated glomerular filtration rate ≥60 mL/min/1.73m2, no proteinuria, no past history of cardiovascular disease, COPD or liver disease. The drinking status was obtained by self-administered questionnaires. We categorized the study subjects into four groups based on the amount of alcohol consumption: <20g/time of ethanol equivalent (lowest); 20-40g/time (low intermediate); 40-60g/time (high intermediate); >60g/time (highest). We set non-drinkers as a reference category. The primary outcome was the incidence of CKD, defined as 25% reduction of eGFR and to less than 60 mL/min/1.73 m2 and/or a dipstick urinalysis score of 1+ or greater (equivalent to ≥30 mg/dL) during the follow up period. We employed Cox proportional hazards regression models to examine the dose-response relationship between baseline alcohol consumption and the risk of CKD. Trend tests were performed using Cox proportional hazards regression models that treated alcohol consumption as a continuous linear term. Results Lowest and low intermediate groups were significantly associated with a decreased risk of CKD (hazard ratio [HR] 0.84; 95% confidence interval [CI], and 0.71–0.99; HR 0.79; 95% CI, 0.66–0.96, respectively) compared to non-drinkers. High intermediate group was associated with a decreased risk of CKD (HR 0.92; 95% CI, 0.70–1.21), and highest group was associated with an increased risk of CKD (HR 1.28; 95% CI, 0.84–1.95), but these associations did not reach statistical significance. There was no dose-response relationship between baseline alcohol consumption and risk of CKD (P-trend = 0.30). Conclusion A J-shape association was observed between self-reported alcohol intake and the incidence of CKD. Moderate alcohol consumption at one time may help reduce the risk of CKD.

Download Full-text

Quantitative assessment of erector spinae muscles and prognosis in elderly patients with pneumonia

Scientific Reports ◽

10.1038/s41598-021-83995-3 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Hiroki Yoshikawa ◽

Kosaku Komiya ◽

Takashi Yamamoto ◽

Naoko Fujita ◽

Hiroaki Oka ◽

...

Keyword(s):

Elderly Patients ◽

Hospital Mortality ◽

Regression Models ◽

Proportional Hazards ◽

Community Acquired Pneumonia ◽

Cox Proportional Hazards ◽

Erector Spinae ◽

Cross Sectional ◽

Proportional Hazards Regression ◽

Survivor Group

AbstractErector spinae muscle (ESM) size has been reported as a predictor of prognosis in patients with some respiratory diseases. This study aimed to assess the association of ESM size on all-cause in-hospital mortality among elderly patients with pneumonia. We retrospectively included patients (age: ≥ 65 years) admitted to hospital from January 2015 to December 2017 for community-acquired pneumonia who underwent chest computed tomography (CT) on admission. The cross-sectional area of the ESM (ESMcsa) was measured on a single-slice CT image at the end of the 12th thoracic vertebra and adjusted by body surface area (BSA). Cox proportional hazards regression models were used to assess the influence of ESMcsa/BSA on in-hospital mortality. Among 736 patients who were admitted for pneumonia, 702 patients (95%) underwent chest CT. Of those, 689 patients (98%) for whom height and weight were measured to calculate BSA were included in this study. Patients in the non-survivor group were significantly older, had a greater frequency of respiratory failure, loss of consciousness, lower body mass index, hemoglobin, albumin, and ESMcsa/BSA. Multivariate analysis showed that a lower ESMcsa/BSA independently predicted in-hospital mortality after adjusting for these variables. In elderly patients with pneumonia, quantification of ESMcsa/BSA may be associated with in-hospital mortality.

Download Full-text

Replacement of potatoes with other vegetables and risk of myocardial infarction in the Danish Diet, Cancer and Health cohort

British Journal Of Nutrition ◽

10.1017/s0007114521000349 ◽

2021 ◽

pp. 1-21

Author(s):

Anne Mette L. Würtz ◽

Mette D. Hansen ◽

Anne Tjønneland ◽

Eric B. Rimm ◽

Erik B. Schmidt ◽

...

Keyword(s):

Myocardial Infarction ◽

Similar Pattern ◽

Proportional Hazards ◽

Cox Proportional Hazards ◽

Heterogeneous Group ◽

Dietary Guidelines ◽

Cox Proportional Hazards Regression ◽

Proportional Hazards Regression ◽

Root Vegetables

ABSTRACT Intake of vegetables is recommended for the prevention of myocardial infarction (MI). However, vegetables make up a heterogeneous group, and subgroups of vegetables may be differentially associated with MI. The aim of this study was to examine replacement of potatoes with other vegetables or subgroups of other vegetables and the risk of MI. Substitutions between subgroups of other vegetables and risk of MI were also investigated. We followed 29,142 women and 26,029 men aged 50-64 years in the Danish Diet, Cancer and Health cohort. Diet was assessed at baseline by using a detailed validated FFQ. Hazards ratios (HR) with 95% CI for the incidence of MI were calculated using Cox proportional hazards regression. During 13.6 years of follow-up, 656 female and 1,694 male cases were identified. Among women, the adjusted HR for MI was 1.02 (95% CI: 0.93, 1.13) per 500 g/week replacement of potatoes with other vegetables. For vegetable subgroups, the HR was 0.93 (95% CI: 0.77, 1.13) for replacement of potatoes with fruiting vegetables and 0.91 (95% CI: 0.77, 1.07) for replacement of potatoes with other root vegetables. A higher intake of cabbage replacing other vegetable subgroups was associated with a statistically non-significant higher risk of MI. A similar pattern of associations was found when intake was expressed in kcal/week. Among men, the pattern of associations was overall found to be similar to that for women. This study supports food-based dietary guidelines recommending to consume a variety of vegetables from all subgroups.

Download Full-text