Assess whether a unit's complication rate is a real signal (dbc758)

Situation

Models and Documentation work in a multi-hospital system comparing length of stay now turns on a unit's complication rate because a model vendor claiming AUC that the local validation cannot reproduce put transfer-delay file for time-sensitive diagnoses in play. Models and Documentation work in a multi-hospital system comparing length of stay now turns on a unit's complication rate because a model vendor claiming AUC that the local validation cannot reproduce put transfer-delay file for time-sensitive diagnoses in play; clinical documentation integrity manager should say what transfer-delay file for time-sensitive diagnoses proves for Healthcare.

Decision

Clinical documentation integrity manager in a multi-hospital system comparing length of stay must choose Proceed under protocol / Pause the pathway / Escalate safety review / Hold using transfer-delay file for time-sensitive diagnoses after a model vendor claiming AUC that the local validation cannot reproduce.

Hypotheses to test

• A model vendor claiming AUC that the local validation cannot reproduce is noise around an already-controlled Models and Documentation process in a multi-hospital system comparing length of stay, given transfer-delay file for time-sensitive diagnoses.
• A model vendor claiming AUC that the local validation cannot reproduce is the event in transfer-delay file for time-sensitive diagnoses that forces Proceed under protocol for clinical documentation integrity manager under Healthcare.
• Transfer-delay file for time-sensitive diagnoses shows a one-file miss after a model vendor claiming AUC that the local validation cannot reproduce, not a Models and Documentation program failure.
• Transfer-delay file for time-sensitive diagnoses cannot decide a unit's complication rate yet after a model vendor claiming AUC that the local validation cannot reproduce; hold is the only Healthcare close a multi-hospital system comparing length of stay can defend.

Analysis required

• Quantify who is harmed if transfer-delay file for time-sensitive diagnoses is wrong.
• Separate a documented exception from an OCR-relevant gap in a multi-hospital system comparing length of stay.
• Validate clinical-outcome evidence and protocol steps in transfer-delay file for time-sensitive diagnoses after a model vendor claiming AUC that the local validation cannot reproduce.