Two-Phase Sampling Designs Extended to Ordinal Outcomes Enable Cost-Efficient Biomarker Studies
Two-phase sampling designs, which reduce costs by restricting expensive measurements (e.g., biomarkers, genomic data) to an informative subset of participants, have been extended for the first time to studies with ordinal outcomes. The authors propose three outcome-informed phase 2 selection strategies and pair them with multiple analysis methods — including ascertainment-corrected MLE, multiple imputation, and sieve MLE — to ensure valid, efficient inference. Simulation studies confirm substantial efficiency gains over simple random sampling, and the framework is validated on a real sepsis clinical trial examining interleukin-6 against a four-level clinical status outcome.