What distinguishes an edit check from medical review?

Why

An edit check may flag a potassium result outside an expected numeric range or an adverse event with an end date before its start. It can detect inconsistent units or incompatible values efficiently and repeatedly, but it cannot determine whether a severe abnormal value is clinically real. Medical review considers the patient history, adverse events, medications, protocol and source records.

For example, an outlier may be a critical safety signal rather than an error to suppress. Define which rules are automatic, which findings need a site query and which require a clinician or safety team. Audit the logic, thresholds, data versions and clinical decisions.

The distinction matters because systematically forcing plausible rare findings into normal ranges introduces bias into safety and efficacy results.