AIRCHILL Schedule of Assessments & CRF Data Dictionary
Shared clinical-operations module · v0.1
Schedule of Assessments & CRF Data Dictionary
This shared module defines the minimum data structure for AIRCHILL-CA and AIRCHILL-STROKE. Indication-specific fields are flagged below. Final eCRFs should use controlled terminology, audit trails, edit checks, role-based access and validated electronic systems.
Schedule of Assessments
| Assessment | Screen | Randomise / device start | 0–6 h | 24 h | 48–72 h | Discharge | 30 d | 90 d |
|---|---|---|---|---|---|---|---|---|
| Eligibility / consent pathway | ● | verify | ||||||
| Baseline demographics / history / pre-event function | ● | |||||||
| Randomisation / allocation | ● | |||||||
| Device configuration / software / serial | ● | continuous change log | ||||||
| Patient-side gas temperature, humidity, flow, FiO₂, pressures, minute ventilation | baseline | continuous / high-frequency | summary | as applicable | ||||
| Core/systemic temperature | baseline | ● | continuous or protocol frequency | ● | ● | ● | ||
| Haemodynamics / ECG / vasopressors | baseline | ● | continuous / interval | ● | ● | |||
| ABG / gas exchange | baseline where available | prespecified intervals | ● | as indicated | ||||
| AE / SAE / device deficiency | start | continuous | ● | ● | through discharge | review | review | |
| Neurological outcome | pre-event baseline | early | early | ● | optional | primary blinded | ||
| EQ-5D-5L / resource use | ● | ● | ● | |||||
| Stroke imaging core-lab fields | ● | procedural | 24-h imaging | |||||
| Cardiac-arrest CPR/ROSC/WLST fields | ● | ● | ● | ● | ● | ● | ● |
Core CRF modules
- Screening log: screening ID, site, date/time, eligibility, exclusion reason, consent route, randomised Y/N.
- Baseline: age, sex, height/weight, relevant history, medication, pre-event mRS/CPC, frailty where chosen.
- Event timeline: source-synchronised timestamps. CA: emergency call, witnessed status, bystander CPR, first rhythm, ROSC, airway, randomisation, AIRCHILL start. Stroke: last-known-well, door, imaging, anaesthesia, randomisation, device start, puncture, reperfusion.
- Device exposure: serial number, software, circuit, temperature setpoint/actual, humidity, FiO₂, flow, pressures, alarms, interruptions, reason, duration.
- Ventilation: mode, VT, RR, PEEP, plateau/peak pressure, compliance, minute ventilation, SpO₂, PaO₂, PaCO₂.
- Haemodynamics: HR, rhythm, MAP/SBP/DBP, vasopressors, fluids, major interventions.
- Safety: AE/SAE, onset, seriousness, severity, expectedness, causality to device/procedure, action, outcome, reporting dates.
- Device deficiency: failure, malfunction, use error, alarm, corrective action, whether SAE could have occurred.
- Outcome: mortality, cause/date; mRS/CPC; NIHSS for stroke; cognitive/participation measure where selected; blinded assessor details.
- Health economics: EMS resources, ICU/ward days, ventilation days, procedures, RRT, rehab, discharge destination, readmission, long-term care, EQ-5D-5L.
Critical derived variables
- Randomisation-to-device-start.
- Event/ROSC or stroke-onset-to-device-start.
- Time to first measurable thermal effect.
- Cumulative respiratory thermal dose.
- Protocol-adherent exposure duration.
- Any reperfusion delay attributable to study intervention.
- Primary outcome completeness and assessor blinding status.
Data-quality rules
All critical clocks must be synchronised; source hierarchy is predefined; units and ranges are standardised; impossible sequences trigger edit checks; device telemetry is preserved in raw form; post-randomisation edits require audit trail; endpoint adjudication fields are separated from treating-team fields.