Correction the Bias of Odds Ratio resulting from the misclassification of exposures in the study of environmental risk factors of lung cancer using Bayesian methods

Inability to measure exact exposure in epidemiological studies is a common in many epidemiological studies, especially when information on exposure is obtained retrospectively, as in any case–control design. Depending on the extent of misclassification, results may be affected. Existing methods for solving this problem require a lot of time and money and for and for some of exposures it is not practical. Recently, new methods have been proposed in 1:1 matched case–control study that have been solved these problems to some extent. This paper extend existing Bayesian method to adjust for misclassification in matched case–control Studies with 1:2 matching. The standard Dirichlet prior distribution for a multinomial model is extended to allow separation of prior assertions about the exposure–disease association from assertions about other parameters. Information that exist in literature about association between exposure and disease used as prior information about OR. Correction of misclassification was investigated using Sensitivity analysis. The results of naïve Bayesian model were similar to the classic model. The second Bayesian model that use prior information about the OR, heavily affected by this information. The third model provides maximum bias adjustment for heavy metals, tobacco and opium. This model showed that heavy metal is not an important risk factor although raw model (logistic regression Classic) detected this exposure as an influencing factor on the incidence of lung cancer. Sensitivity analysis showed that third model is robust regarding to different levels of Sensitivity and Specificity. Result of this study showed that although in most of exposures the results of the second and third model were similar but the proposed model able to somewhat correct the misclassification.