Runtime and Error Count Data for Incremental Processing Evaluation in Healthcare Analytics Pipelines

Supporting data for the manuscript "Correctness-preserving incremental processing for stateful SQL transformations: failure modes and mitigation patterns." Contains pipeline execution runtime statistics and field-level error count comparisons across four production healthcare analytics models (Diagnosis, SchedulingAppointment, SchedulingEventAction, TrackingItem) evaluated on Snowflake Enterprise over a 10-day window. Includes full refresh, corrected incremental, and naive incremental strategy runtimes, plus per-column error breakdowns. No patient-level data is included.