Premium
Regression Analysis of Tumour Prevalence Data
Author(s) -
Dinse Gregg E.,
Lagakos S. W.
Publication year - 1983
Publication title -
journal of the royal statistical society: series c (applied statistics)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.205
H-Index - 72
eISSN - 1467-9876
pISSN - 0035-9254
DOI - 10.2307/2347946
Subject(s) - statistics , regression analysis , regression , geography , mathematics
SUMMARY This paper proposes a logistic regression model for comparing treatment groups with respect to tumour prevalence. The prevalence test commonly used to compare treatments in animal tumorigenicity experiments (Hoel and Walburg, 1972; Peto et al. , 1980) is essentially equivalent to a likelihood score test derived under a logistic model that expresses tumour prevalence as a function of time and treatment. The more general regression context suggests an alternative to the convention of grouping observations into arbitrarily chosen intervals. The model also incorporates covariates, provides a framework for estimating the strength of a dose‐response relationship and for testing a central assumption underlying the usual prevalence test, and is computationally simple to analyse.