Comparing performance between log-binomial and robust Poisson regression models for estimating risk ratios under model misspecification

Background Log-binomial and robust (modified) Poisson regression models are popular approaches to estimate risk ratios for binary response variables. Previous studies have shown that comparatively they produce similar point estimates and standard errors. However, their performance under model misspe...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:BMC medical research methodology Jg. 18; H. 1; S. 63 - 12
Hauptverfasser: Chen, Wansu, Qian, Lei, Shi, Jiaxiao, Franklin, Meredith
Format: Journal Article
Sprache:Englisch
Veröffentlicht: London BioMed Central 22.06.2018
BioMed Central Ltd
Springer Nature B.V
BMC
Schlagworte:
Asthma
Bias
Data analysis
Economic models
Estimates
Health Sciences
Link function misspecification
Log-binomial regression
Medicine
Medicine & Public Health
Methods
Model misspecification
Parameter estimation
Patients
Ratios
Regression analysis
Research Article
Risk ratio
Robust (modified) Poisson regression
Statistical Theory and Methods
statistics and modelling
Statistics for Life Sciences
Theory of Medicine/Bioethics
Robust (modified) Poisson regression
Risk ratio
Model misspecification
Link function misspecification
Log-binomial regression
ISSN:1471-2288, 1471-2288
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Background Log-binomial and robust (modified) Poisson regression models are popular approaches to estimate risk ratios for binary response variables. Previous studies have shown that comparatively they produce similar point estimates and standard errors. However, their performance under model misspecification is poorly understood. Methods In this simulation study, the statistical performance of the two models was compared when the log link function was misspecified or the response depended on predictors through a non-linear relationship (i.e. truncated response). Results Point estimates from log-binomial models were biased when the link function was misspecified or when the probability distribution of the response variable was truncated at the right tail. The percentage of truncated observations was positively associated with the presence of bias, and the bias was larger if the observations came from a population with a lower response rate given that the other parameters being examined were fixed. In contrast, point estimates from the robust Poisson models were unbiased. Conclusion Under model misspecification, the robust Poisson model was generally preferable because it provided unbiased estimates of risk ratios.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ObjectType-Article-2
ObjectType-Feature-1
content type line 23
ISSN:1471-2288
1471-2288
DOI:10.1186/s12874-018-0519-5