The cost of dichotomising continuous variables

BMJPublished 4 May 2006Open access

Douglas G. Altman, Patrick Royston

DOI10.1136/bmj.332.7549.1080

Citations2,573

Abstract

ABSTRACT Myeloid-derived suppressor cells (MDSCs) are present in elevated numbers in TB patients and have been found to be permissive for Mycobacterium tuberculosis (Mtb) proliferation. To determine whether depletion of MDSCs may improve host control of TB, we used a novel diphtheria toxin-based fusion protein known as DABIL-4 that targets and depletes IL-4-receptor positive cells. We show that DABIL-4 depletes both PMN-MDSCs and M-MDSC in the mouse TB model, and that it reduces the lung bacillary burden of Mtb. These results indicate that MDSC-depleting therapies targeting the IL4 receptor are beneficial in TB and offer an avenue towards host-directed TB therapy.

Keywords

Immunology and MicrobiologyBiochemistry, Genetics and Molecular Biology

Psychological MethodsOn the practice of dichotomization of quantitative variables.

3,342 Citations2002Robert C. MacCallum, Shaobo Zhang +2 more

The authors present the case that dichotomization is rarely defensible and often will yield misleading results.

View PDF

Statistics in MedicineDichotomizing continuous predictors in multiple regression: a bad idea

2,228 Citations2005Patrick Royston, Douglas G. Altman +1 more

It is argued that the simplicity achieved is gained at a cost; dichotomization may create rather than avoid problems, notably a considerable loss of power and residual confounding.

Applied Psychological MeasurementThe Cost of Dichotomization

1,492 Citations1983Jacob Cohen

View PDF

Psychological MethodsOn the practice of dichotomization of quantitative variables.

330 Citations2002Robert C. MacCallum, Shaobo Zhang +2 more

Statistics in MedicineInflation of the type I error rate when a continuous confounding variable is categorized in logistic regression analyses

248 Citations2004Peter C. Austin, Lawrence J. Brunner

It is found that the inflation of the type I error rate increases with increasing sample size, as the correlation between the risk factor and the confounding variable increases, and with a decrease in the number of categories into which the confounder is divided.

Methods of Information in MedicineBuilding Multivariable Regression Models with Continuous Covariates in Clinical Epidemiology

195 Citations2005Willi Sauerbrei, Patrick Royston

It is demonstrated that a structured approach based on fractional polynomials can give a broadly satisfactory practical solution to the problem of simultaneously identifying a subset of 'important' predictors and determining the functional relationship for continuous predictors.

Journal of Clinical EpidemiologyProblems in defining cutoff points of continuous prognostic factors: Example of tumor thickness in primary cutaneous melanoma

55 Citations1997Petra Buettner, Claus Garbe +1 more

The results of the present study demonstrate that optimized cutoff points are extremely data dependent and vary notably even if prerequisites are constant, if the classification of a continuous prognostic factor is necessary, it should not be based on the results of one single study, but on consensus discussions including the findings of several investigations.

Obstetrics and GynecologyTreatment of continuous data as categoric variables in

39 Citations1997G DELPRIORE, Peyman Zandieh +1 more

The treatment of continuous data has improved over the time period reviewed, but clinicians should be aware that continuous data may be mischaracterized as categoric variables in some journal articles and editors will consider requesting r values for all continuous data relations.