Predictive accuracy of novel risk factors and markers: A simulation study of the sensitivity of different performance measures for the Cox proportional hazards regression model

Predicting outcomes that occur over time is important in clinical, population health, and health services research. We compared changes in different measures of performance when a novel risk factor or marker was added to an existing Cox proportional hazards regression model. We performed Monte Carlo simulations for common measures of performance: concordance indices ( c, including various extensions to survival outcomes), Royston’s D index, R2-type measures, and Chambless’ adaptation of the integrated discrimination improvement to survival outcomes. We found that the increase in performance due to the inclusion of a risk factor tended to decrease as the performance of the reference model increased. Moreover, the increase in performance increased as the hazard ratio or the prevalence of a binary risk factor increased. Finally, for the concordance indices and R2-type measures, the absolute increase in predictive accuracy due to the inclusion of a risk factor was greater when the observed event rate was higher (low censoring). Amongst the different concordance indices, Chambless and Diao’s c-statistic exhibited the greatest increase in predictive accuracy when a novel risk factor was added to an existing model. Amongst the different R2-type measures, O’Quigley et al.’s modification of Nagelkerke’s R2 index and Kent and O’Quigley’s [Formula: see text] displayed the greatest sensitivity to the addition of a novel risk factor or marker. These methods were then applied to a cohort of 8635 patients hospitalized with heart failure to examine the added benefit of a point-based scoring system for predicting mortality after initial adjustment with patient age alone.

Download Full-text

Testing for Covariate Effect in the Cox Proportional Hazards Regression Model

Communication in Statistics- Theory and Methods ◽

10.1080/03610920802536958 ◽

2009 ◽

Vol 38 (14) ◽

pp. 2333-2347 ◽

Cited By ~ 6

Author(s):

Karthik Devarajan ◽

Nader Ebrahimi

Keyword(s):

Regression Model ◽

Proportional Hazards ◽

Cox Proportional Hazards ◽

Covariate Effect ◽

Cox Proportional Hazards Regression ◽

Proportional Hazards Regression

Download Full-text

MODELLING THE DETERMINANTS OF UNDER-FIVE CHILD MORTALITY RATES USING COX PROPORTIONAL HAZARDS REGRESSION MODEL

FUDMA Journal of Sciences ◽

10.33003/fjs-2020-0404-496 ◽

2021 ◽

Vol 4 (4) ◽

pp. 401-408

Author(s):

M. C. Musa ◽

O. E. Asiribo ◽

H. G. Dikko ◽

M. Usman ◽

S. S. Sani

Keyword(s):

Regression Model ◽

Child Mortality ◽

Proportional Hazards ◽

Mortality Rates ◽

Cox Proportional Hazards ◽

Cox Proportional Hazards Model ◽

Predictor Variables ◽

Under Five ◽

Cox Proportional Hazards Regression ◽

Proportional Hazards Regression

An under-five childhood mortality rates in Nigeria is still high, despite efforts of government at all levels to combat the menace. This study examined some factors that significantly affect under-five child mortality. A sample of mothers with children under the age of five from Nigeria Demographic and Health Survey data (NDHS, 2013 & 2018) was used to assess the effect of some selected predictor variables (or covariates) on childhood survival. Cox proportional hazards model is essentially a regression model popularly used for investigating the association between the survival time and one or more predictor variables. The results from final fitted Cox proportional hazards regression model that the covariates, contraceptive used by the mother, state of residence, birth weight of child and type of toilet facility used by the h-ousehold were found to be significantly associated with under-five survival in the North Central Region of Nigeria. All the calculations are performed using the R software for statistical analysis.

Download Full-text

Prognostic Significance of Serum Free Amino Acids in Head and Neck Cancers

Cells ◽

10.3390/cells8050428 ◽

2019 ◽

Vol 8 (5) ◽

pp. 428 ◽

Cited By ~ 3

Author(s):

Vit Vsiansky ◽

Marketa Svobodova ◽

Jaromir Gumulec ◽

Natalia Cernei ◽

Dagmar Sterbova ◽

...

Keyword(s):

Amino Acids ◽

Amino Acid ◽

Regression Model ◽

Head And Neck ◽

Cell Lines ◽

Proportional Hazards ◽

Cox Proportional Hazards ◽

Transformed Cells ◽

Cox Proportional Hazards Regression ◽

Proportional Hazards Regression

Despite distinctive advances in the field of head and neck squamous cell cancer (HNSCC) biomarker discovery, the spectrum of clinically useful prognostic serum biomarkers is limited. As metabolic activities in highly proliferative transformed cells are fundamentally different from those in non-transformed cells, specific shifts in concentration of different metabolites may serve as diagnostic or prognostic markers. Blood amino acids have been identified as promising biomarkers in different cancers before, but little is known about this field in HNSCC. Blood amino acid profiles of 140 HNSCC patients were examined using high-performance liquid chromatography. Cox proportional hazards regression model was used to assess the prognostic value of amino acid concentrations in serum. Colony forming assay was used to identify the effect of amino acids that were significant in Cox proportional hazards regression models on colony forming ability of FaDu and Detroit 562 cell lines. In the multivariable Cox regression model for overall survival (OS), palliative treatment was associated with an unfavourable prognosis while high serum levels of methionine have had a positive prognostic impact. In the relapse-free survival (RFS) multivariable model, methionine was similarly identified as a positive prognostic factor, along with tumor localization in the oropharynx. Oral cavity localization and primary radio(chemo)therapy treatment strategy have been linked to poorer RFS. 1mM serine was shown to support the forming of colonies in both tested HNSCC cell lines. Effect of methionine was exactly the opposite.

Download Full-text