Methods for assessing familial aggregation were developed for the matched case-control family study design with adjustment for ascertainment. These models utilize the properties of conditional multivariate normal distributions and random effects to accommodate samp ling through probands and matching effects while estimating familial correlations from quantitative methods. A single ascertainment model is assumed for this design. Methods are illustrated from a family study of obstructive sleep apnea based upon a matched case-control design, and by simulation studies. The latter provide evidence of the desirability of including a matching effect, where appropriate, in terms of bias, mean square error and confidence interval coverage probabilities.
Showing the most recent 10 out of 922 publications