Understanding the Effect of Categorization of a Continuous Predictor with Application to Neuro Oncology

Neuro-Oncology Practice ◽

10.1093/nop/npab049 ◽

2021 ◽

Author(s):

Ruchi Gupta ◽

Courtney N Day ◽

W Oliver Tobin ◽

Cynthia S Crowson

Keyword(s):

Prognostic Factor ◽

Use Cases ◽

Continuous Variables ◽

Continuous Version ◽

Loss Of Information ◽

Race Ethnicity ◽

Continuous Predictors ◽

Continuous Predictor

Abstract Many Neuro-Oncology studies commonly assess the association between a prognostic factor (predictor) and disease or outcome, such as the association between age and glioma. Predictors can be continuous (e.g., age) or categorical (e.g., race/ethnicity). Effects of categorical predictors are frequently easier to visualize and interpret than effects of continuous variables. This makes it an attractive, and seemingly justifiable, option to subdivide the continuous predictors into categories (e.g., age< 50 years vs. age ≥50 years). However, this approach results in loss of information (and power) compared to the continuous version. This review outlines the use cases for continuous and categorized predictors and provides tips and pitfalls for interpretation of these approaches.

Download Full-text

Preoperative peripheral blood lymphocyte/monocyte ratio and survival in patients with resected stage IV melanoma.

Journal of Clinical Oncology ◽

10.1200/jco.2012.30.15_suppl.8600 ◽

2012 ◽

Vol 30 (15_suppl) ◽

pp. 8600-8600

Author(s):

Nicole M. Rochet ◽

Luis F. Porrata ◽

Lisa A. Kottschade ◽

Travis Edward Grotz ◽

Svetomir Markovic

Keyword(s):

Prognostic Factor ◽

Proportional Hazards ◽

Complete Blood Count ◽

Stage Iv ◽

Independent Prognostic Factor ◽

Monocyte Count ◽

Continuous Variables ◽

Melanoma Patients ◽

Chi Square Analysis ◽

Stage Iv Melanoma

8600 Background: The prognosis of stage IV melanoma patients remains poor. Published results have suggested that components of the complete blood count have significant prognostic value in several malignancies. Among the most studied were the absolute lymphocyte count (ALC), and absolute monocyte count (AMC) on clinical outcomes of patients with lymphoid malignancies. Thus, we sought to investigate if the pre-operative ALC (ALC-PO), AMC (AMC-PO) and ALC/AMC ratio (ALC/AMC-PO) affects the risk of disease recurrence and survival after complete surgical resection of metastatic melanoma. Methods: We studied 66 stage IV, resected melanoma patients followed at Mayo Clinic from 2000 to 2010. Log rank chi-square analysis was used to determine the best cut-off values for each pre-operative variable, while proportional hazards models were used to compared survival. Results: The median follow-up of the cohort was 24 months (range: 2.3 – 117 months). ALC-PO, AMC-PO and ALC/AMC-PO, as continuous variables, were all identified as prognostic factors for both relapse-free survival (RFS) and overall survival (OS). The best cut-off values for ALC-PO, AMC-PO and ALC/AMC-PO were 1.9; 0.62; and 2.05, respectively. Using Kaplan-Meier analysis, patients with an ALC-PO ≥ 1.9 x 109/L experienced superior OS and RFS compared with ALC-PO < 1.9 x 109/L patients [median OS of 58 months vs. 34 months, p < 0.04; median RFS of 14 months vs. 5 months, p < 0.009]. Conversely, a low AMC-PO (<0.62 x 109/L) was associated with better OS and RFS compared with higher AMC-PO (≥ 0.62 x 109/L): [median OS of 47 months vs. 14 months, p < 0.007; median RFS of 9 months vs. 5 months, p < 0.02]. When the ALC-PO and AMC-PO were combined as an ALC/AMC ratio, the group with an ALC/AMC-PO ≥ 2.05 experienced a superior OS and RFS compared to patients with ALC/AMC-PO < 2.05: [median OS of 49 months vs. 12 months, p < 0.0001; median RFS of 10 months vs. 4 months, p < 0.0001]. Multivariate analysis showed ALC/AMC-PO to be an independent prognostic factor for RFS and OS (HR = 0.32, p < 0.003; HR = 0.23, p < 0.002). Conclusions: Our study showed, that ALC/AMC-PO ratio is an independent prognostic factor for RFS and OS in patients undergoing resection of metastatic (stage IV) melanoma.

Download Full-text

Using Lifespan Developmental Theory and Methods as a Viable Alternative to the Study of Generational Differences at Work

Industrial and Organizational Psychology ◽

10.1017/iop.2015.47 ◽

2015 ◽

Vol 8 (3) ◽

pp. 342-346 ◽

Cited By ~ 12

Author(s):

Hannes Zacher

Keyword(s):

Generation X ◽

Generational Differences ◽

Statistical Power ◽

Empirical Studies ◽

Developmental Theory ◽

Viable Alternative ◽

Continuous Variables ◽

Work Context ◽

Loss Of Information ◽

Comprehensive Theory

I agree with Costanza and Finkelstein (2015) that it is futile to further invest in the study of generational differences in the work context due to a lack of appropriate theory and methods. The key problem with the generations concept is that splitting continuous variables such as age or time into a few discrete units involves arbitrary cutoffs and atheoretical groupings of individuals (e.g., stating that all people born between the early 1960s and early 1980s belong to Generation X). As noted by methodologists, this procedure leads to a loss of information about individuals and reduced statistical power (MacCallum, Zhang, Preacher, & Rucker, 2002). Due to these conceptual and methodological limitations, I regard it as very difficult if not impossible to develop a “comprehensive theory of generations” (Costanza & Finkelstein, p. 20) and to rigorously examine generational differences at work in empirical studies.

Download Full-text

Continuous Predictors: Why Splitting Continuous Variables Into Categories Is Undesirable

Best Practices in Logistic Regression ◽

10.4135/9781483399041.n5 ◽

2017 ◽

pp. 131-169

Keyword(s):

Continuous Variables ◽

Continuous Predictors

Download Full-text

Do not group quantitative variables

Epidemiology with R ◽

10.1093/oso/9780198841326.003.0010 ◽

2020 ◽

pp. 219-222

Author(s):

Bendix Carstensen

Keyword(s):

Utility Function ◽

The Other ◽

Optimal Decision ◽

Continuous Variables ◽

Decision Point ◽

Independent Variables ◽

Optimal Decisions ◽

The Relationship ◽

Continuous Predictor ◽

Predicted Risk

This chapter explores the problems caused by categorizing quantitative variables (here termed continuous variables). Optimum decisions are made by applying a utility function to a predicted value. At the decision point, one can solve for the personalized cutpoint for predicted risk that optimizes the decision. Dichotomization on independent variables is completely at odds with making optimal decisions. To make an optimal decision, the cutpoint for a predictor would necessarily be a function of the continuous values of all the other predictors. Moreover, categorization assumes that the relationship between the predictor and the response is flat within intervals; this assumption is far less reasonable than a linearity assumption in most cases. Categorization of continuous variables using percentiles is particularly hazardous. To make a continuous predictor be more accurately modelled when categorization is used, multiple intervals are required.

Download Full-text

Abstract 1286: Macrophages in the breast tumor microenvironment as a prognostic factor among race/ethnicity

10.1158/1538-7445.am2015-1286 ◽

2015 ◽

Author(s):

Ana M. Santander ◽

Tulay Koru-Sengul ◽

Feng Miao ◽

Merce Jorda ◽

Stephan Gluck ◽

...

Keyword(s):

Prognostic Factor ◽

Tumor Microenvironment ◽

Breast Tumor ◽

Race Ethnicity

Download Full-text

Relationship between Pretreatment Serum Albumin Levels with the Risk of Malignant Pleural Mesothelioma

International Journal of Statistics in Medical Research ◽

10.6000/1929-6029.2020.09.08 ◽

2021 ◽

Vol 9 ◽

pp. 69-82

Author(s):

Sabyasachi Mukherjee ◽

Keyword(s):

Prognostic Factor ◽

Serum Albumin ◽

Malignant Pleural Mesothelioma ◽

Asbestos Exposure ◽

Disease Diagnosis ◽

Parametric Estimation ◽

Pleural Mesothelioma ◽

Significant Prognostic Factor ◽

Continuous Variables ◽

Pretreatment Serum

Background: Malignant Pleural Mesothelioma (MPM) is a very rare and aggressive form of cancer. Recently it was found that pretreatment Serum Albumin (SA), the main circulating protein in blood is a significant prognostic factor for MPM patients. The objective of this present article is to show the relationship between pretreatment Serum albumin (SA) levels with the risk of MPM. Methods: Generalized additive model (GAM), an advanced regression analysis method has been introduced here to find this mathematical relationship between the response variable (SA) and the cofactors. Results: The main determinates of SA are identified - asbestos exposure, hemoglobin, disease diagnosis status (patients having MPM) are the factors having significant association with SA, whereas duration of asbestos exposure, duration of disease symptoms, total protein (TP), Pleural lactic dehydrogenise (PLD), pleural protein (PP), pleural glucose (PG) and C-reactive protein (CRP) are the significant continuous variables for SA. The non-parametric estimation part of this model shows Lactate dehydrogenase (LDH) and Glucose level are the significant smoothing terms. Additionally it is also found that, second and third order interactions between cofactors are highly significant for SA. Conclusions: The findings of this present work can conclude that - serum albumin may play the role of a very significant prognostic factor for MPM disease and it has been established here through mathematical modeling. Few of the findings are already been exist in MPM research literature whereas some of the findings are completely new in the literature.

Download Full-text

Epstein-Barr virus and its prognostic value in a cohort of Peruvian women with cervical cancer

10.1101/2020.08.04.20167841 ◽

2020 ◽

Author(s):

Denisse Castro ◽

Juana Vera ◽

Percy Soto-Becerra ◽

Marco López-Ilasaca ◽

Alejandro Yabar ◽

...

Keyword(s):

Cervical Cancer ◽

Prognostic Factor ◽

Epstein Barr Virus ◽

Independent Prognostic Factor ◽

Prognostic Models ◽

Continuous Variables ◽

Internal Validation ◽

Prognostic Effect ◽

Barr Virus ◽

Epstein Barr

Aim: We aim to evaluate the prognostic effect of Epstein-Barr virus (EBV) infection on overall survival (OS) in Peruvian women with cervical cancer. Methods: We conducted a retrospective cohort study. Polymerase chain reaction technique was used in paraffin-embedded tumor tissue for the detection of EBNA-1 and LMP-1. We used a multiple Cox proportional-hazard regression to estimate adjusted hazard ratios (aHR) for death and 95% confidence intervals (95% CI). In order to model continuous variables without categorization, we used a multivariable fractional polynomial approach. We performed a stability analysis using bootstrapping for internal validation. Results: A total of 99 patients with cervical cancer were included. The prevalence of EBV in cervical cancer specimens was 22.2% (n=22). The 1-year and 5-year OS rates were 81.8% (95% CI 58.5-92.8) and 45% (95% CI 23.9-64.1) in the EBV-positive group compared to 78.8% (95% CI 67.7-86.4) and 37.8% (95% CI 25.7-49.8) in the EBV-negative group, respectively. In the multivariate analysis, positive EBV status was an independent prognostic factor for improved OS (aHR: 0.32; 95% CI 0.16 to 0.67; p=0.002) compared to negative EBV status. Conclusions: EBV status is an independent prognostic factor for OS in cervical cancer. Evaluation of EBV status could be used as a clinical prognostic biomarker and to improve currently available prognostic models such as the FIGO system. Future prospective studies will be needed to validate these theories.

Download Full-text

Diminishing Barriers to the Praxis Examination in Communication Sciences and Disorders

Perspectives on Global Issues in Communication Sciences and Related Disorders ◽

10.1044/gics1.2.64 ◽

2011 ◽

Vol 1 (2) ◽

pp. 64-71

Author(s):

Luis F. Riquelme

Keyword(s):

Geographic Location ◽

The United States ◽

Success For All ◽

Speech Language Pathology ◽

Barriers To Success ◽

Call To Action ◽

Language Pathology ◽

National Certification ◽

Race Ethnicity ◽

Passing Score

Abstract Passing the Praxis Examination in speech-language pathology or audiology can be a difficult task. A passing score is the entry to a list of requirements for national certification (CCC-SLP, CCC-A) and for state licensure in the United States. This article will provide current information on the examination and address barriers to success that have been identified over the years. A call to action may serve to refocus efforts on improving access to success for all test-takers regardless of race/ethnicity, ability, or geographic location.

Download Full-text