Content area
Full Text
European Journal of Epidemiology (2007) 22:717 Springer 2007 DOI 10.1007/s10654-006-9078-x
METHODS
Patterns of bias due to dierential misclassication by casecontrol status in a casecontrol study
Po-Huang Chyou
Biostatistics, Biomedical Informatics Research Center, Marsheld Clinic Research Foundation, 1000 North Oak Avenue, Marsheld, WI, 54449, USA
Accepted in revised form 3 November 2006
Abstract. Background: Casecontrol study is still one of the most commonly used study designs in epidemiological research. Misclassication of case control status remains a signicant issue because it will bias the results of a casecontrol study. There exist two types of misclassication, dierential versus nondierential. It is commonly accepted that nondierential misclassication will bias the results of the study towards the null hypothesis. Conversely, no reports have assessed the impact and direction of dierential misclassication on odds ratio (OR) estimate. The goal of the present study is to demonstrate by statistical derivation that patterns exist on the bias induced by dierential misclassication. Methods: Based on a 2 2 case control study design, we derive the odds ratio without misclassication, and those with misclassication according to: (1) controls are misclassied as cases by exposure status; (2) cases are misclassied as controls by exposure status; and (3) both controls and cases are misclassied by exposure
status simultaneously. Furthermore, mathematical derivations are shown for each of the ratios of the two odds ratios with and without misclassication.These methods are carried out by simulation analyses. Results: Simulation analyses show that quite a number of biased odds ratios tend to move away from the null hypothesis and result in approaching zero or innity with increasing proportion of misclassication among cases, controls, or both. These patterns are associated with the exposure status and the values of unbiased odds ratio (<1, 1, or >1).Conclusions: Our ndings suggest that, unlike nondierential misclassication, dierential misclassication of casecontrol status in a casecontrol study may not weaken the exposureoutcome association towarding the null hypothesis. Care needs to be taken for interpreting the results of a casecontrol study when there exists dierential misclassication bias, a practical issue in epidemiological research.
Key words: Casecontrol study, Misclassication bias, Odds ratio
Introduction
Casecontrol study is still one of the most commonly used study designs in epidemiological research [16]. Misclassication of casecontrol status remains a signicant issue because it will...