In clinical trials one traditionally models the effect of treatment on the mean response. The underlying assumption is that treatment affects the response distribution through a mean location shift on a suitable scale, with other aspects of the distribution (shape/dispersion/variance) remaining the same. This work is motivated by a trial in Parkinson's disease patients in which one of the endpoints is the number of falls during a 10-week period. Inspection of the data reveals that the Poisson-inverse Gaussian (PiG) distribution is appropriate, and that the experimental treatment reduces not only the mean, but also the variability, substantially. The conventional analysis assumes a treatment effect on the mean, either adjusted or unadjusted for covariates, and a constant dispersion parameter. On our data, this analysis yields a non-significant treatment effect. However, if we model a treatment effect on both mean and dispersion parameters, both effects are highly significant. A simulation study shows that if a treatment effect exists on the dispersion and is ignored in the modelling, estimation of the treatment effect on the mean can be severely biased. We show further that if we use an orthogonal parametrization of the PiG distribution, estimates of the mean model are robust to misspecification of the dispersion model. We also discuss inferential aspects that are more difficult than anticipated in this setting. These findings have implications in the planning of statistical analyses for count data in clinical trials.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1002/bimj.201700218 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
On the ability of the LR method to detect bias when there is pedigree misspecification and lack of connectedness.
Genet Sel Evol
November 2024
Facultad de Agronomía, Universidad de Buenos Aires, C1417DSQ, Buenos Aires, Argentina.
Background: Cross-validation techniques in genetic evaluations encounter limitations due to the unobservable nature of breeding values and the challenge of validating estimated breeding values (EBVs) against pre-corrected phenotypes, challenges which the Linear Regression (LR) method addresses as an alternative. Furthermore, beef cattle genetic evaluation programs confront challenges with connectedness among herds and pedigree errors. The objective of this work was to evaluate the LR method's performance under pedigree errors and weak connectedness typical in beef cattle genetic evaluations, through simulation.
View Article and Find Full Text PDFGeneralized Bayes approach to inverse problems with model misspecification.
Inverse Probl
October 2023
Department of Statistical Science, Duke University, Durham, NC, United States of America.
We propose a general framework for obtaining probabilistic solutions to PDE-based inverse problems. Bayesian methods are attractive for uncertainty quantification but assume knowledge of the likelihood model or data generation process. This assumption is difficult to justify in many inverse problems, where the specification of the data generation process is not obvious.
View Article and Find Full Text PDFModel selection and robust inference of mutational signatures using Negative Binomial non-negative matrix factorization.
BMC Bioinformatics
May 2023
Department of Mathematics, Aarhus University, Aarhus, Denmark.
Background: The spectrum of mutations in a collection of cancer genomes can be described by a mixture of a few mutational signatures. The mutational signatures can be found using non-negative matrix factorization (NMF). To extract the mutational signatures we have to assume a distribution for the observed mutational counts and a number of mutational signatures.
View Article and Find Full Text PDFModel misspecification misleads inference of the spatial dynamics of disease outbreaks.
Proc Natl Acad Sci U S A
March 2023
Department of Evolution and Ecology, University of California, Davis, CA 95616.
Article Synopsis
- - Bayesian phylodynamic models have revolutionized epidemiology by helping researchers track how pathogens spread across different geographic areas, though they rely heavily on initial assumptions about model parameters.
- - Default prior assumptions in these models often lead to unrealistic interpretations regarding pathogen dispersal rates, routes, and the origins of outbreaks, which can skew research findings.
- - The authors suggest improving these models by developing more biologically plausible prior assumptions to enhance the accuracy of epidemiological studies and better inform disease surveillance and monitoring strategies.
Blinded sample size re-estimation for comparing over-dispersed count data incorporating follow-up lengths.
Stat Med
December 2022
Department of Biostatistics, Nagoya University Graduate School of Medicine, Nagoya, Japan.
Blinded sample size re-estimation (BSSR) is an adaptive design to prevent the power reduction caused by misspecifications of the nuisance parameters in the sample size calculation of comparative clinical trials. However, conventional BSSR methods used for overdispersed count data may not recover the power as expected under the misspecification of the working variance function introduced by the specified analysis model. In this article, we propose a BSSR method that is robust to the misspecification of the working variance function.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!