Optimizing Aggregated N-Of-1 Trial Designs for Predictive Biomarker Validation: Statistical Methods and Theoretical Findings.

As was incorporated mathematically into the statistical modeling, it was expected that those choosing to terminate participation prior to completion of the full trial would more commonly be those for whom response to prazosin was either clearly significant or clearly minimal—while those who elected to continue for the entire trial would more commonly be those for whom it remained unclear to both participant and researchers alike whether the participant had had a significant, specific positive response to treatment or not.

also in 481 other papers