ObGyn Intelligence
ObGyn Intelligence ← All Tools — tools.obmd.com
Adverse Outcome Investigation Checklist

ObGyn Adverse Outcome Investigation Checklist

NTSB-methodology case review framework — Amos Grünebaum, MD · ObGyn Intelligence · obmd.com

1
Duty
Doctor-patient relationship established
2
Breach
Deviation from standard of care
3
Causation
Breach directly caused the damage
4
Damage
Measurable adverse outcome resulted
NTSB principle applied to medicine: When a plane crashes, investigators do not ask whose fault it was. They ask what happened at every step, what decisions were made and why, and what system change prevents the next crash. The goal is not blame. The goal is never again.
0
Checked
0
N/A
0
Remaining
0%
Complete
🗂️
Case identification
Complete before beginning the review — anonymized throughout
▼
🤝
Phase 0 — Duty: was this your patient?
Element 1 — the threshold question. Without duty, nothing else applies.
0/0▼
Establishing the relationship
A formal doctor-patient relationship confirmed between the named clinician and this patient critical
Duty arises when a clinician agrees to diagnose or treat. A single clinical encounter suffices.
Chart confirms admission, registration, or a clinical encounter with notes or orders under the named clinician
Clinician's role documented precisely: primary attending, covering physician, consultant, or triage-only contact
Edge cases — where duty is contested
Curbside consultation — was specific advice given informally without chart documentation? Duty may still attach depending on specificity and whether it was acted upon
Hallway consults are a significant source of contested duty in ObGyn.
Covering physician — scope and documentation of coverage reviewed
Triage-and-discharge — whether duty was established during a triage-only encounter has been assessed
Telemedicine or telephone advice — whether remote guidance without in-person examination created duty of care has been evaluated
Duty determination documented in writing: confirmed / contested / absent critical
🚨
Phase 1 — The event: what happened
Build the factual sequence. No interpretation yet.
0/0▼
Event reconstruction
Written chronological timeline constructed from the medical record — not from memory or verbal accounts critical
Source each entry to a specific chart note. Minute-by-minute for acute events; hour-by-hour for longer trajectories.
Timeline includes: who was present, what was observed, what was ordered, what was administered, and when
The point at which the outcome became irreversible identified as precisely as possible — the "last safe moment" critical
NTSB equivalent: the moment a crash became unrecoverable. This anchors the entire causation analysis.
Time from recognition of deterioration to first clinical intervention calculated and documented
EFM strips reviewed in full if applicable — not summarized, reviewed in their entirety critical
Vital signs trend reviewed for the 6-hour window preceding the adverse event
Lab values and their timing relative to clinical decisions reviewed
Harm assessment (Element 4: Damage)
Nature and extent of harm formally documented — maternal, neonatal, or both critical
Short-term vs. long-term harm distinguished (transient vs. permanent injury)
Confirmed that harm was an adverse outcome — not a known complication appropriately disclosed and consented critical
A foreseeable complication occurring without error is not necessarily a breach.
📋
Phase 2 — Clinical record: the black box
The chart is the flight data recorder. What does it actually say?
0/0▼
Documentation completeness
Complete medical record secured and confirmed unaltered critical
Nursing notes, physician notes, and consultant notes all reviewed — not just the attending's notes
All verbal orders documented and countersigned per protocol
Informed consent documentation present and complete for all procedures performed
Documentation gaps identified and their timing noted — undocumented periods are as important as documented ones critical
Medication administration record reviewed — drug, dose, route, timing, and who administered
Anesthesia record reviewed if applicable
Operative report reviewed if applicable — EBL, unexpected findings, surgical decisions
Prenatal record
Full prenatal record reviewed — were risk factors identified antenatally?
Growth scan results reviewed — was FGR, macrosomia, or oligohydramnios identified or missed?
GBS status, blood type, and key antenatal results documented and acted upon
⚖️
Phase 3 — Standard of care: was good medicine followed?
Element 2 (Breach) — compare each decision against the accepted standard
0/0▼
Defining the standard
Relevant ACOG and SMFM guidelines identified for each decision point in question critical
Applicable standard of care stated in writing — not assumed. "What should a reasonably competent ObGyn have done?"
Where the standard is unsettled, this acknowledged explicitly — not resolved by assumption
Clinical decision review
Each major clinical decision reviewed: Was there a decision? Was it documented? Timely? Consistent with the standard? critical
Triage and assessment at admission reviewed — was risk stratification performed and appropriate?
EFM interpretation reviewed against ACOG/NICHD criteria — Category I, II, or III correctly applied?
Decision-to-incision time for cesarean reviewed against the 30-minute benchmark for emergency CS
Oxytocin use reviewed — dosing, escalation, response to category II/III tracing, documentation of rationale
Shoulder dystocia management reviewed if applicable — HELPERR steps documented?
PPH management reviewed if applicable — quantitative blood loss measured, hemorrhage protocol activated at the right threshold?
Preeclampsia recognition and management reviewed if applicable — BP thresholds, magnesium, timing of delivery
Escalation reviewed — was attending called when warranted? Subspecialty consulted when indicated? critical
Informed consent reviewed — options presented? Risks of action and inaction disclosed with absolute numbers? critical
Informed consent is a process, not a signature.
Breach determination
Written determination made: deviation from standard occurred / did not occur / requires expert review critical
If deviation identified: error of commission (wrong action taken) vs. error of omission (action not taken) specified
💬
Phase 4 — Communication: the most common root cause
Where clinical care and human interaction intersect — and most lawsuits originate
0/0▼
Clinician-to-patient
Patient received clear explanation of diagnosis and plan — documented in the chart
Alternatives to the proposed treatment were presented — not just the chosen path critical
Risks of both action and inaction communicated with absolute numbers — not just "there are risks involved"
Patient understanding confirmed — was teach-back or any verification of comprehension performed?
Language and health literacy barriers identified and addressed — interpreter used if needed, documented
Clinician-to-clinician
Handoff communication reviewed — clinical picture clearly transmitted at shift change? critical
Nurse-to-physician communication reviewed — concerns escalated? SBAR or equivalent used?
Chain of command — if a concern was dismissed, was there a mechanism to escalate? Was it used?
Consultant communication — were consultants given complete information? Were their recommendations implemented?
After the event — was the family communicated with promptly, honestly, and with compassion?
🏥
Phase 5 — System factors: beyond the individual clinician
The NTSB always asks about the system, not just the pilot
0/0▼
Staffing & resources
Nurse-to-patient ratio at the time documented and compared to safe staffing standards
On-call structure reviewed — was attending present or reachable within the required timeframe?
Equipment availability confirmed — everything needed for the response available and functioning?
Blood bank response time reviewed if applicable
Protocols & training
Applicable institutional protocol reviewed — does one exist? Is it current?
Staff had completed required competencies and simulation training for this emergency
Whether a similar event has occurred before — if yes, were prior recommendations implemented? critical
Recurring events with no institutional change between them are a system failure, not individual error.
Fatigue reviewed — shift length, consecutive hours, overnight call burden of personnel involved
EHR & technology
EHR alert fatigue reviewed — were clinically relevant alerts suppressed or overridden?
Any EHR downtime or malfunction during the relevant period investigated
Medication ordering system reviewed — CPOE errors, weight-based dosing, allergy alerts
🔗
Phase 6 — Causation: did the breach cause the damage?
Element 3 — the most complex analysis. No assumptions. No hindsight bias.
0/0▼
Causation analysis
Direct causal chain traced from identified deviation to the adverse outcome critical
The "but-for" test: but for the deviation, would the harm have occurred? If yes, the deviation is not the legal cause.
Counterfactual analysis performed — what would the expected outcome have been with standard care? critical
Pre-existing conditions and their independent contribution to the outcome assessed separately
Intervening causes considered — events between the alleged breach and the outcome that may independently explain the harm
Hindsight bias explicitly guarded against — decisions evaluated based on what was knowable at the time
Multiple causation considered — more than one contributing factor? Additive or synergistic?
Published evidence supporting the causal link (or its absence) identified
Four-element summary
Element 1 (Duty): doctor-patient relationship confirmed
Element 2 (Breach): determination made regarding deviation from standard of care
Element 3 (Causation): determination made regarding whether breach caused damage critical
Element 4 (Damage): adverse outcome formally characterized and documented
🛡️
Phase 7 — Prevention: never again
The NTSB's final phase — and the only reason to conduct the investigation
0/0▼
Recommendations
At least one specific, actionable recommendation generated for every root cause identified critical
A review that identifies causes but generates no changes is theater. Every root cause demands a specific response.
Recommendations are SMART: Specific, Measurable, Assignable, Realistic, Time-bound
Each recommendation has a named responsible party and a completion deadline
Recommendations address both individual and system levels
Implementation & follow-through
90-day follow-up review scheduled to assess whether recommendations have been implemented
Findings shared with the full L&D team in de-identified format
If a protocol change is recommended, training scheduled before the new protocol goes live
Mechanism exists to track whether this type of adverse event recurs
Clinician support offered — peer support, wellness resources, debrief. Second victims need care too.
Completed review stored in protected QI database — not in the patient chart, not accessible in litigation

Probable cause determination — four-element summary

Let Claude take a look

Claude reads all entered data — checklist status, findings notes, and four-element selections — and generates a structured analysis. Fill in as much as you have before running.

Analyzing...