Abstract
BayesSenMC: an R package for Bayesian Sensitivity Analysis of Misclassification
Highlights
Many epidemiological studies are concerned with assessing the risk of an outcome between exposed and non-exposed subjects
In a case–control study, researchers first identify subjects who have the disease of interest and subjects who do not, and ascertain the exposure status of the subjects in each group
The odds ratio is typically used to assess the association between the exposure and disease in the case–control study; it describes the ratio of the exposure odds in the case group to that in the control group
Summary
Many epidemiological studies are concerned with assessing the risk of an outcome between exposed and non-exposed subjects. Sensitivity analysis can be used to evaluate the effects of uncertainties in measurement on the observed results of the study (Greenland, 1996; Lash and Flink, 2003; Chu et al, 2006), in which the mapping from observed to true measurements may be based on prior information or expert opinion about the accuracy of the measurement When such information or opinion is lacking, researchers may overor under-adjust for misclassification with an inaccurate guess, which may, in turn, produce a poor estimate (Gustafson et al, 2006). This article uses the meta-analysis performed by Carvalho et al (2015) to obtain the prior distributions of the sensitivity and specificity of classifying the exposure status of bipolar disorder. The generalized linear mixed-effects model will be used to synthesize all 55 studies to estimate the overall sensitivity and specificity
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
