Menu Close

BJOG Editor’s Choice: Multivariable Regression, Prediction, and Clinicians

Citation

Thorp, John M., Jr. (2015). BJOG Editor's Choice: Multivariable Regression, Prediction, and Clinicians. BJOG: An International Journal of Obstetrics & Gynaecology, 122(3), i.

Abstract

One of my great honours was to mentor a co-founder of SAS Inc., Jane Helwig, as she pursued a second career in obstetrics and gynaecology after great success as a biostatistician. The SAS software opened the gates to let anyone with a computer perform multivariable logistic regression. After a scathing review of a paper we co-wrote from a US competitor journal to BJOG, in which we were ordered by the editor to use the SAS software to regress a data set that in her opinion was not statistically appropriate to use multivariable regression, she confided that one of their great fears in introducing their statistical product was that it would now be possible for clinicians not well grounded in analytics to misuse this powerful tool.

URL

https://doi.org/10.1111/1471-0528.13289

Reference Type

Journal Article

Year Published

2015

Journal Title

BJOG: An International Journal of Obstetrics & Gynaecology

Author(s)

Thorp, John M., Jr.