Regression for skewed biomarker outcomes subject to pooling.

摘要: Epidemiological studies involving biomarkers are often hindered by prohibitively expensive laboratory tests. Strategically pooling specimens prior to performing these lab assays has been shown effectively reduce cost with minimal information loss in a logistic regression setting. When the goal is perform continuous biomarker as outcome, analysis of pooled may not be straightforward, particularly if outcome right-skewed. In such cases, we demonstrate that slight modification standard multiple linear model for poolwise data can provide valid and precise coefficient estimates when pools formed combining biospecimens from subjects identical covariate values. x-homogeneous cannot formed, propose Monte Carlo expectation maximization (MCEM) algorithm compute maximum likelihood (MLEs). Simulation analytical methods essentially unbiased parameters well their errors appropriate assumptions met. Furthermore, show how one utilize fully observed inform strategy, yielding high level statistical efficiency at fraction total cost.