Top

Knee Surgery, Sports Traumatology, Arthroscopy

Published in:

03-11-2022 | REVIEW

Multivariable regression: understanding one of medicine’s most fundamental statistical tools

Authors: Nathan H. Varady, Ayoosh Pareek, Christina M. Eckhardt, Riley J. Williams III, Sophia J. Madjarova, Matthieu Ollivier, R. Kyle Martin, Jón Karlsson, Benedict U. Nwachukwu

Published in: Knee Surgery, Sports Traumatology, Arthroscopy | Issue 1/2023

Abstract

Multivariable regression is a fundamental tool that drives observational research in orthopaedic surgery. However, regression analyses are not always implemented correctly. This study presents a basic overview of regression analyses and reviews frequent points of confusion. Topics include linear, logistic, and time-to-event regressions, causal inference, confounders, overfitting, missing data, multicollinearity, interactions, and key differences between multivariable versus multivariate regression. The goal is to provide clarity regarding the use and interpretation of multivariable analyses for those attempting to increase their statistical literacy in orthopaedic research.

Alin A (2010) Multicollinearity. Wiley Interdiscip Rev. Comput Stat 2(3):370–374. https://doi.org/10.1002/WICS.84CrossRef

Austin PC, Steyerberg EW (2017) Events per variable (EPV) and the relative performance of different strategies for estimating the out-of-sample validity of logistic regression models. Stat Methods Med Res 26(2):796. https://doi.org/10.1177/0962280214558972CrossRef

Beks RB, Bhashyam AR, Houwert RM, van der Velde D, van Heijl M, Smeeing DPJ et al (2019) When observational studies are as helpful as randomized trials: Examples from orthopedic trauma. J Trauma Acute Care Surg 87(3):730–732. https://doi.org/10.1097/TA.0000000000002347CrossRef

Heller G (2010) Proportional hazards regression with interval censored data using an inverse probability weight. Lifetime Data Anal 17(3):373–385. https://doi.org/10.1007/S10985-010-9191-8CrossRef

Hidalgo B, Goodman M (2013) Multivariate or multivariable regression? Am J Public Health 103(1):39. https://doi.org/10.2105/AJPH.2012.300897CrossRef

Jameson SS, Dowen D, James P, Serrano-Pedraza I, Reed MR, Deehan D (2012) Complications following anterior cruciate ligament reconstruction in the English NHS. Knee 19(1):14–19CrossRef

Kim SC, Bateman BT (2020) Methodological challenges in conducting large-scale real-world data analyses on opioid use in musculoskeletal disorders. J Bone Joint Surg Am 102(Suppl 1):10–14. https://doi.org/10.2106/JBJS.20.00121CrossRef

Kutner M, Nachtsheim C, Neter J, Li W (2004) Applied Linear Statistical Models. McGraw-Hill, New York

Lederer DJ, Bell SC, Branson RD, Chalmers JD, Marshall R, Maslove DM et al (2019) Control of confounding and reporting of results in causal inference studies. Ann Am Thorac Soc 16(1):22–28CrossRef

10.

Moons KGM, Altman DG, Reitsma JB, Ioannidis JPA, Macaskill P, Steyerberg EW et al (2015) Transparent Reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): explanation and elaboration. Ann Intern Med 162(1):W1–W73. https://doi.org/10.7326/M14-0698CrossRef

11.

Moons KGM, de Groot JAH, Bouwmeester W, Vergouwe Y, Mallett S, Altman DG et al (2014) Critical appraisal and data extraction for systematic reviews of prediction modelling studies: the CHARMS checklist. PLoS Med 11(10):e1001744. https://doi.org/10.1371/JOURNAL.PMED.1001744CrossRef

12.

Pavlou M, Ambler G, Seaman SR, Guttmann O, Elliott P, King M et al (2015) How to develop a more accurate risk prediction model when there are few events. BMJ. https://doi.org/10.1136/BMJ.H3868CrossRef

13.

van Smeden M, de Groot JAH, Moons KGM, Collins GS, Altman DG, Eijkemans MJC et al (2016) No rationale for 1 variable per 10 events criterion for binary logistic regression analysis. BMC Med Res Methodol 16(1):1–12. https://doi.org/10.1186/S12874-016-0267-3/TABLES/4CrossRef

14.

Varady NH, Feroe AG, Fontana MA, Chen AF (2021) Causal Language in Observational Orthopaedic Research. J Bone Joint Surg Am 103(19):e76. https://doi.org/10.2106/JBJS.20.01921CrossRef

15.

Vittinghoff E, McCulloch CE (2007) Relaxing the Rule of Ten Events per Variable in Logistic and Cox Regression. Am J Epidemiol 165(6):710–718. https://doi.org/10.1093/AJE/KWK052CrossRef

Title: Multivariable regression: understanding one of medicine’s most fundamental statistical tools
Authors: Nathan H. Varady
Ayoosh Pareek
Christina M. Eckhardt
Riley J. Williams III
Sophia J. Madjarova
Matthieu Ollivier
R. Kyle Martin
Jón Karlsson
Benedict U. Nwachukwu
Publication date: 03-11-2022
Publisher: Springer Berlin Heidelberg
Published in: Knee Surgery, Sports Traumatology, Arthroscopy / Issue 1/2023
Print ISSN: 0942-2056
Electronic ISSN: 1433-7347
DOI: https://doi.org/10.1007/s00167-022-07215-9

At a glance: The STEP trials

Springer Medicine

Multivariable regression: understanding one of medicine’s most fundamental statistical tools

Abstract

At a glance: The STEP trials

Springer Medicine

Abstract

Please log in to get access to this content

Other articles of this Issue 1/2023

ACL stump and ACL femoral landmarks are equally reliable in ACL reconstruction for assisting ACL femoral tunnel positioning

Medial patellofemoral ligament reconstruction using nonresorbable sutures yields comparable outcomes to reconstruction with a pedicled quadriceps tendon autograft when performed in addition to bony risk factor correction

Hip microinstability: fact or fiction?

Poorer functional Outcomes in Patients with Multi-Ligamentous Knee Injury with Concomitant Patellar Tendon Ruptures at 5 years Follow-Up

Explosive hamstrings strength asymmetry persists despite maximal hamstring strength recovery following anterior cruciate ligament reconstruction using hamstring tendon autografts

Inter and intra-observer agreement of the 3-dimensional CT based anterior inferior iliac spine classification system shows fair-to-moderate agreement among high volume hip surgeons