What changed
KAISEN tests subgroup measurement, diagnosis, mitigation and drift monitoring across synthetic clinical tasks. Threshold adjustment consistently reduced disparity, while calibration varied, proxy misspecification silently broke causal diagnosis and drift thresholds failed to transfer between cohorts.
Why it matters
A fairness pipeline can produce reassuring averages while its diagnosis, mitigation variance or monitoring assumptions fail in the setting where decisions are made.
What to do
Report intervention variance, minimum detectable effects and deliberate failure tests. Retune drift thresholds on each cohort instead of copying one successful audit unchanged.
Caveat
Every result is synthetic and establishes no clinical validity. Use the released methodology to test audit components, not to select real medical thresholds.