hull losses, 2023
annually in US hospitals
reduction with WHO checklist
The Model
How Aviation Does It — and Why It Works
Modern commercial aviation does not operate on memory. It does not operate on experience alone. It operates on phase-specific, challenge-response, digitally verified checklists integrated directly into cockpit avionics — updated in real time, and impossible to skip without a system flag.
Every flight divides into discrete phases: preflight, engine start, taxi, takeoff, climb, cruise, descent, approach, landing, shutdown. Each phase has its own checklist. Each item is challenged verbally by one crew member and confirmed by the other. If the gear is not down, the aircraft tells you. If the flaps are not set, the aircraft tells you.
Glass cockpit systems like Garmin G1000 and Gulfstream avionics display checklists that respond to sensor data — some items auto-complete when the system detects the correct configuration. Electronic Flight Bags replaced paper charts and cards across major carriers through the 1990s and 2000s. The result: zero major commercial airline fatalities in the United States in 2023.
This did not happen because pilots got smarter. It happened because the system got better.
“Aviation does not trust the expert. Aviation trusts the system. The expert uses the system.”
Atul Gawande — The Checklist Manifesto, 2009The Comparison
Aviation vs. Medicine: A Checklist Reckoning
| Domain | Aviation | Medicine |
|---|---|---|
| Format | Digital, integrated, phase-specific | Paper, laminated cards, printed PDFs |
| Verification | Sensor-linked; auto-flags missed items | Manual, memory-dependent, no enforcement |
| Challenge-Response | Mandatory two-person verbal confirmation | Inconsistent; rarely structured |
| Updates | Real-time via avionics software | Months to years behind the evidence |
| Emergency Protocols | Memory items + always-current printed backup | Varies by institution; often outdated |
| Compliance Monitoring | Flight data recorder captures all checklist states | Virtually none at point of care |
| Outcome | Near-zero preventable deaths | 98,000+ preventable deaths/year (US) |
The difference is not competence. Surgeons, obstetricians, and anesthesiologists are among the most rigorously trained professionals in existence. The difference is system design. Aviation accepted this reality decades ago. Medicine has not.
The Evidence
We Already Know Checklists Save Lives. The Problem Is Implementation.
Haynes et al. (2009) demonstrated a 36% reduction in major complications and a 47% reduction in deaths across eight hospitals on four continents following introduction of the WHO checklist. The checklist had 19 items. It fit on one page. It took under two minutes to complete.
Haynes AB, et al. A Surgical Safety Checklist to Reduce Morbidity and Mortality in a Global Population. N Engl J Med. 2009;360(5):491–499.
The California Maternal Quality Care Collaborative implementation of structured hemorrhage checklists and bundles reduced maternal mortality from obstetric hemorrhage by 55% over five years in participating hospitals. The intervention was a checklist. The barrier was institutional culture, not evidence.
Main EK, et al. Reduction of severe maternal morbidity from hemorrhage using a state perinatal quality collaborative. Am J Obstet Gynecol. 2017;216(3):298.e1–298.e11.
Observational data consistently shows paper checklist completion rates well below 50% in real-world settings — even in hospitals that report “full implementation.” Items are skipped. Cards become outdated. No one flags the gap. Digital tools with mandatory completion fields and audit trails change clinical behavior in ways paper never can.
Russ SJ, et al. A qualitative evaluation of staff attitudes towards the WHO surgical safety checklist. J Eval Clin Pract. 2015;21(2):237–47.
Phase-Specific Safety
Aviation’s Most Important Lesson for Obstetrics
Aviation does not use a single checklist. It uses a different checklist for each phase of flight. A preflight checklist and a landing checklist are not interchangeable — and a triage checklist and a postpartum hemorrhage checklist should not be either. In obstetrics, the phases are clear.
Phase 1
Admission & Triage
Risk stratification, vital baselines, EFM interpretation, IV access, blood type, hemorrhage risk score.
Phase 2
Labor
Oxytocin protocol compliance, fetal monitoring triggers, epidural safety, progress assessment thresholds.
Phase 3
Delivery
Operative setup, shoulder dystocia response, neonatal team notification, cord blood gases.
Phase 4
Postpartum
Hemorrhage quantification, uterotonic escalation, VTE prophylaxis, hypertension surveillance.
Paper cannot adapt to phase in real time. A digital system presents the right checklist at the right moment — triggered by clinical events, elapsed time, or provider input. That is not a vision of some future technology. It is available now.
The Argument Against Paper Is Not Theoretical
A paper checklist does not know what time it is. It does not know whether the item was actually completed or merely initialed. It cannot alert a second provider when a critical threshold is crossed. It cannot be updated when guidelines change. It cannot be audited after an adverse event with any reliability.
A paper checklist is better than no checklist. But calling it a safety system in 2025 is the same as calling a sextant a navigation system. Both work. One works far better. The question is why we are still choosing the sextant.
“The answer is not incompetence. It is inertia. And inertia is a choice.”
The LiveEvidence Approach
What a Digital Clinical Decision System Actually Looks Like
LiveEvidence tools are not static PDFs. They are interactive, evidence-grounded, phase-specific clinical decision support systems built directly from peer-reviewed data, with every number traceable to its source publication. This is the aviation model applied to medicine: the right information, at the right phase, verified against current evidence, accessible at the point of care.
Phase-Triggered
The right decision framework for the current clinical moment, not a generic reference.
Evidence First
Every tool opens with its source publications in Vancouver format. The evidence is never hidden.
Absolute Risk
Every risk estimate in absolute terms alongside relative. No misleading statistics.
No Invented Data
If evidence does not exist for a specific scenario, the tool says so explicitly.
Always Current
Digital tools can be updated when evidence changes. A laminated card cannot.
Dual Audience
Clinician-depth and patient-language modes. One interface, two conversations.
“Evidence that sits unread helps nobody. A checklist that cannot be verified helped nobody. The system aviation built works. It is the system medicine needs to build.”
Amos Grünebaum, MD — Professor of Obstetrics & Gynecology | LiveEvidence.com