|Grant Number:||5R01CA040644-19 Interpret this number|
|Primary Investigator:||Breslow, Norman|
|Organization:||University Of Washington|
|Project Title:||Statistical Methods in Cancer Epidemiology|
DESCRIPTION (Applicant's abstract): Epidemiology plays a major role in the identification of carcinogenic agents and in the quantification of dose time response relationships upon which regulation and preventive strategies are based. Epidemiology as a science depends critically upon statistics. The goal of this project is the development of more efficient statistical designs and methods of analysis for both analytic and descriptive studies. There are two major areas of emphasis. First, many studies involve the estimation of a large number of related quantities, such as multiple disease rates in small areas for identification of "hot spots" and construction of disease incidence maps. Evaluation of community based intervention programs and meta-analyses of data from multiple studies likewise require consideration of "random effects" to represent unexplained heterogeneity at the level of the community or the study. A major goal is the development, evaluation and implementation of hierarchical statistical models that allow for the efficient estimation of both random and fixed effects in such settings. Second, two-phase case- control studies, exposure stratified case-cohort studies and other complex stratified designs are of great value in limiting the collection of costly data to those subjects who are most informative regarding disease/risk factor associations. An important example is the validation substudy conducted to alleviate the effects of measurement error. Here "gold standard" measurements are made for a small number of subjects in a random subsample. More efficient methods of study design will be developed based on stratification of the validation sample using measurements available for all subjects. More efficient methods of statistical analysis will be developed using the tools of modem semiparametric inference. Such designs and efficient new analysis methods can dramatically reduce study costs while yielding estimates that are almost as good as when "gold standard" measurements are made for everyone. The methods used to achieve these goals include mathematical and statistical analysis, computer simulation and application to important datasets collected by cancer epidemiologists and other public health scientists.
Are statistical contributions to medicine undervalued?
Authors: Breslow NE
Source: Biometrics, 2003 Mar;59(1), p. 1-8.
Saddlepoint approximations for small sample logistic regression problems.
Authors: Platt RW
Source: Stat Med, 2000 Feb 15;19(3), p. 323-34.
Generalized linear mixed models for meta-analysis.
Authors: Platt RW, Leroux BG, Breslow N
Source: Stat Med, 1999 Mar 30;18(6), p. 643-54.
Approximate hierarchical modelling of discrete data in epidemiology.
Authors: Breslow N, Leroux B, Platt R
Source: Stat Methods Med Res, 1998 Mar;7(1), p. 49-62.
Weighted likelihood, pseudo-likelihood and maximum likelihood methods for logistic regression analysis of two-stage data.
Authors: Breslow NE, Holubkov R
Source: Stat Med, 1997 Jan 15-Feb 15;16(1-3), p. 103-16.
Power of affected sibling method tests for linkage.
Authors: Tierney C, McKnight B
Source: Hum Hered, 1993 Sep-Oct;43(5), p. 276-87.
A conditional analysis for two-treatment multiple-period crossover designs with binomial or poisson outcomes and subjects who drop out.
Authors: McKnight B, Van den Eeden SK
Source: Stat Med, 1993 May 15;12(9), p. 825-34.
Descriptive epidemiology and survival analysis of nasopharyngeal carcinoma in the United States.
Authors: Burt RD, Vaughan TL, McKnight B
Source: Int J Cancer, 1992 Oct 21;52(4), p. 549-56.
Testing the null hypothesis in small area analysis.
Authors: Cain KC, Diehr P
Source: Health Serv Res, 1992 Aug;27(3), p. 267-94.
Approaches to the analysis of case-control studies of the efficacy of screening for cancer.
Authors: Weiss NS, McKnight B, Stevens NG
Source: Am J Epidemiol, 1992 Apr 1;135(7), p. 817-23.
Analysing the relationship between change in a risk factor and risk of disease.
Authors: Cain KC, Kronmal RA, Kosinski AS
Source: Stat Med, 1992 Apr;11(6), p. 783-97.
Linkage analysis of malignant melanoma with the chromosome 1 markers D1S47 and PND.
Authors: Blossey H, Guo SW, McKnight B, Tierney C, Thompson E, Wijsman E
Source: Cytogenet Cell Genet, 1992;59(2-3), p. 182-4.
Power and detectable risk of seven tests for standardized mortality ratios.
Authors: Samuels SJ, Beaumont JJ, Breslow NE
Source: Am J Epidemiol, 1991 Jun 1;133(11), p. 1191-7.
Further studies in the variability of pock counts.
Authors: Breslow N
Source: Stat Med, 1990 Jun;9(6), p. 615-26.
Statistical issues in the analysis of data from occupational cohort studies.
Authors: Breslow N
Source: Recent Results Cancer Res, 1990;120, p. 78-93.
Efficiency gains from the addition of controls to matched sets in cohort studies.
Authors: Raboud J, Breslow NE
Source: Stat Med, 1989 Aug;8(8), p. 977-85.
Logistic regression analysis and efficient design for two-stage studies.
Authors: Cain KC, Breslow NE
Source: Am J Epidemiol, 1988 Dec;128(6), p. 1198-206.
Logistic regression for stratified case-control studies.
Authors: Breslow NE, Zhao LP
Source: Biometrics, 1988 Sep;44(3), p. 891-9.
Nonparametric estimation of relative mortality functions.
Authors: Breslow N, Langholz B
Source: J Chronic Dis, 1987;40 Suppl 2, p. 89S-99S.
Methods of estimation in log odds ratio regression models.
Authors: Breslow NE, Cologne J
Source: Biometrics, 1986 Dec;42(4), p. 949-54.
A general estimator for the variance of the Mantel-Haenszel odds ratio.
Authors: Robins J, Greenland S, Breslow NE
Source: Am J Epidemiol, 1986 Nov;124(5), p. 719-23.
Estimators of the Mantel-Haenszel variance consistent in both sparse data and large-strata limiting models.
Authors: Robins J, Breslow N, Greenland S
Source: Biometrics, 1986 Jun;42(2), p. 311-23.
Short-term assays to predict carcinogenicity. Statistical analysis of data from in-vitro assays of mutagenesis.
Authors: Breslow N, Kaldor J
Source: IARC Sci Publ, 1986;null(83), p. 457-81.