AI PILOT MEASUREMENT WORKSHEET — MD Transform — v1.0, October 18, 2026
Protocol: https://mdtransform.com/evidence.html#protocol
Planning template. No outcomes are implied by blank fields.

WORKFLOW AND OWNERSHIP
Practice / anonymous study ID:
Workflow owner and clinical escalation owner:
Tool / version / configuration:
Eligible unit of work and completed outcome:
Inclusion / exclusion criteria:
Data permissions and approved collection location:

DESIGN (complete before starting)
Baseline dates:                       Pilot dates:
Baseline sampling method:            Pilot sampling method:
Target sample and rationale (not a statistical power claim):
Staffing, role, volume and case-mix variables to record:
Primary measure and prespecified success threshold:
Quality measure / material error definition / stop criteria:
Audit sampling method and reviewer:
Fallback / escalation process:

ONE ROW PER ELIGIBLE ATTEMPT (use anonymous task IDs)
Task ID | period | role | preparation min | handling min | review min | correction min | exception min | elapsed min | completed correctly? | error category | excluded? reason
Record missing observations explicitly. Do not enter patient identifiers.

REPORT
Eligible baseline n:                  Eligible pilot n:
Audited baseline n:                   Audited pilot n:
Excluded / missing n and reasons:
Mean active effort per task:          Median / range:
Correct completions / attempts:
Material corrections / audited outputs:
One-time costs:                       Recurring costs and period:
Concurrent changes and limitations:
Decision / owner / next review date:
Capacity actually redeployed (if observed):