Equivalence of prospective and retrospective models in the Bayesian analysis of case-control studies

The natural likelihood to use for a case-control study is a 'retrospective' likelihood, i.e. a likelihood based on the probability of exposure given disease status. Prentice & Pyke (1979) showed that, when a logistic regression form is assumed for the probability of disease given exposure, the maximum likelihood estimators and asymptotic covariance matrix of the log odds ratios obtained from the retrospective likelihood are the same as those obtained from the 'prospective' likelihood, i.e. that based on probability of disease given exposure. We prove a similar result for the posterior distribution of the log odds ratios in a Bayesian analysis. This means that the Bayesian analysis of case-control studies may be done using a relatively simple model, the logistic regression model, which treats data as though generated prospectively and which does not involve nuisance parameters for the exposure distribution. Copyright Biometrika Trust 2004, Oxford University Press.

[1]  J C Wakefield,et al.  Point-source modelling using matched case-control data. , 2000, Biostatistics.

[2]  Sylvia Richardson,et al.  Stochastic search variable selection , 1995 .

[3]  G Parmigiani,et al.  A Bayesian Hierarchical Approach for Combining Case‐Control and Prospective Studies , 1999, Biometrics.

[4]  Kathryn Roeder,et al.  A Bayesian semiparametric model for case-control studies with errors in variables , 1997 .

[5]  D. Rubin.  The Bayesian Bootstrap , 1981 .

[6]  Paul Gustafson,et al.  A Bayesian approach to case-control studies with errors in covariables. , 2002, Biostatistics.

[7]  M. Pike,et al.  Bias and efficiency in logistic analyses of stratified case-control studies. , 1980, International journal of epidemiology.

[8]  J. Sethuraman,et al.  Convergence of Dirichlet Measures and the Interpretation of Their Parameter. , 1981 .

[9]  Pertti Mutanen,et al.  Exact Bayesian analysis of two proportions , 1987 .

[10]  Sylvia Richardson,et al.  Markov Chain Monte Carlo in Practice , 1997 .

[11]  Guido Consonni,et al.  A Bayesian Method for Combining Results from Several Binomial Experiments , 1995 .

[12]  T. Raghunathan,et al.  Monte Carlo methods for exploring sensitivity to distributional assumptions in a Bayesian analysis of a series of 2 x 2 tables. , 1994, Statistics in medicine.

[13]  K. Roeder,et al.  A Semiparametric Mixture Approach to Case-Control Studies with Errors in Covariables , 1996 .

[14]  A Bayesian partition model for case‐control studies on highly polymorphic candidate genes , 2002, Genetic epidemiology.

[15]  Magnus A. McGee,et al.  Simple Bayesian Analyses for Case‐Control Studies in Cancer Epidemiology , 1993 .

[16]  A. W. van der Vaart,et al.  On Profile Likelihood , 2000 .

[17]  M Zelen,et al.  Case-control studies and Bayesian inference. , 1986, Statistics in medicine.

[18]  R J Marshall,et al.  Bayesian analysis of case-control studies. , 1988, Statistics in medicine.

[19]  S. Baker.  The Multinomial‐Poisson Transformation , 1994 .

[20]  R. Pyke,et al.  Logistic disease incidence models and case-control studies , 1979 .