On the Uses and Abuses of Regression Models: A Call for Reform of Statistical Practice and Teaching
Statistics in Medicine · 2025-06 · 44 (13-14)
DOI: 10.1002/sim.10244 ↗At a glance
Journal article in Statistics in Medicine, published online on 2025-06-24. Written by John B Carlin and 1 co-authors. Institution: Clinical Epidemiology & Biostatistics Unit (CEBU), Murdoch Children's Research Institute, Melbourne, Victoria, Australia. Major MEDLINE topics: Biostatistics, Models, Statistical. Cited 25 times so far (Crossref); 71 references. The full text is freely available (open access, CC BY 4.0). Funded by: Australian National Health and Medical Research Council.
Abstract (original language)
Regression methods dominate the practice of biostatistical analysis, but biostatistical training emphasizes the details of regression models and methods ahead of the purposes for which such modeling might be useful. More broadly, statistics is widely understood to provide a body of techniques for “modeling data,” underpinned by what we describe as the “true model myth”: that the task of the statistician/data analyst is to build a model that closely approximates the true data generating process. By way of our own historical examples and a brief review of mainstream clinical research journals, we describe how this perspective has led to a range of problems in the application of regression methods, including misguided “adjustment” for covariates, misinterpretation of regression coefficients and the widespread fitting of regression models without a clear purpose. We then outline a new approach to the teaching and application of biostatistical methods, which situates them within a framework that first requires clear definition of the substantive research question at hand, within one of three categories: descriptive, predictive, or causal. Within this approach, the development and application of (multivariable) regression models, as well as other advanced biostatistical methods, should proceed differently according to the type of question. Regression methods will no doubt remain central to statistical practice as they provide a powerful tool for representing variation in a response or outcome variable as a function of “input” variables, but their conceptualization and usage should follow from the purpose at hand.
Subjects
MeSH terms (MEDLINE)
Author keywords
Authors and institutions
- Clinical Epidemiology & Biostatistics Unit (CEBU), Murdoch Children's Research Institute, Melbourne, Victoria, Australia.
- CEBU, Department of Paediatrics, The University of Melbourne, Melbourne, Victoria, Australia.
- Centre for Epidemiology & Biostatistics, Melbourne School of Population & Global Health, University of Melbourne, Melbourne, Victoria, Australia.
ORCID profiles: John B Carlin, Margarita Moreno-Betancur
Funding
- Australian National Health and Medical Research Council
Cite
APA
Carlin, J. B., & Moreno-Betancur, M. (2025). On the Uses and Abuses of Regression Models: A Call for Reform of Statistical Practice and Teaching. Statistics in Medicine, 44(13-14). https://doi.org/10.1002/sim.10244
Vancouver
Carlin JB, Moreno-Betancur M. On the Uses and Abuses of Regression Models: A Call for Reform of Statistical Practice and Teaching. Statistics in Medicine. 2025;44(13-14). doi:10.1002/sim.10244
BibTeX
@article{Carlin2025,
author = {Carlin, John B and Moreno-Betancur, Margarita},
title = {On the Uses and Abuses of Regression Models: A Call for Reform of Statistical Practice and Teaching},
journal = {Statistics in Medicine},
year = {2025},
volume = {44},
number = {13-14},
doi = {10.1002/sim.10244}
}
EUVIMED