Adverse events and medical errors


  • Definitions
    • Adverse Event: Any injury caused by medical management (rather than underlying disease); may or may not be preventable
    • Medical Error: Failure of a planned action to be completed as intended OR use of a wrong plan to achieve an aim
    • Near Miss: Error that could have caused harm but did not reach the patient (intercepted or by chance) c
    • Sentinel Event: Unexpected occurrence involving death or serious physical/psychological injury
      • Most common root causes: Miscommunication, Excessive team hierarchy, Poor interprofessional collaboration in teams c
  • Types of Errors
    • Active Errors: Occur at point of contact between human and system (e.g., wrong medication administered by nurse)
    • Latent Errors: System/organizational failures that create conditions for active errors (e.g., poor labeling, inadequate staffing)
    • Diagnostic Errors: Missed, delayed, or incorrect diagnosis
    • Treatment Errors: Wrong medication, dose, route, or procedure
    • Preventive Errors: Failure to provide prophylaxis or monitoring

Analysis of medical error

  • High-Yield Cognitive Biases:
    • Anchoring: Fixating on initial data/triage diagnosis despite contradictory evidence.
    • Availability heuristic: Overestimating disease likelihood based on recent/vivid cases.
    • Confirmation bias: Seeking only evidence supporting a favored diagnosis while ignoring contradictory data.
    • Premature closure: Terminating diagnostic workup prematurely after accepting initial working diagnosis.
    • Search satisfying: Halting diagnostic search once a single finding is identified (e.g., missing second fracture on X-ray).
    • Framing effect: Decisions biased by how information is presented (e.g., triage labels, transfer notes).
  • Quality Improvement (QI) Tools:
    • Root Cause Analysis (RCA): Retrospective multidisciplinary review for sentinel events; targets systemic/latent flaws, not individual blame.
      • Process: Multidisciplinary team reconstruct event timeline utilize tools (e.g., “5 Whys”, Fishbone/Ishikawa diagram) formulate process-level corrective action plan establish outcome measures.
    • Failure Mode and Effects Analysis (FMEA): Prospective risk assessment performed before introducing a new process/system.
    • Plan-Do-Study-Act (PDSA): Iterative, rapid-cycle testing of small-scale interventions.
  • Error Disclosure & Ethics:
    • First step: Stabilize pt and treat complications immediately (do not delay care for admin/legal consults).
    • Disclosure rules: In-person, transparent communication of facts + known outcomes + corrective plan + explicit apology.
    • Step 2 CK trap: Never blame colleagues/trainees and never conceal errors to avoid litigation.
  • Prevention & Safety Systems:
    • I-PASS handoff: Illness severity, Patient summary, Action list, Situational awareness/contingency, Synthesis/read-back.
    • Universal Protocol / Surgical “Time-Out”: Completed immediately prior to incision; verifies pt identity, procedure, marked surgical site, and prophylactic Abx with the entire OR team.
    • Forcing function: Engineering hard-stop preventing human error (e.g., unique gas line wall connectors, non-Luer enteral tubing).
    • Closed-loop communication: Verbal orders repeated back verbatim and confirmed prior to drug delivery (vital during resuscitations).

Management of medical errors


  • Initial disclosure
    • Acknowledge errors openly
    • Do not hide errors or other relevant factors
    • Express empathy & give apology as appropriate
    • Avoid blaming or denigrating other team members & providers
    • Allow adequate opportunity for patient questions
  • Follow-up steps
    • Remain engaged in the patient’s care
    • Outline steps to prevent future occurrence

Prevention of medical errors

  • Core Principle: System-focused interventions > Person-focused interventions (education/retraining is the weakest strategy due to reliance on human memory).
  • Safety Culture:
    • Flattened hierarchy: Empowers any team member (nurses, trainees) to voice safety concerns or “stop the line.”
    • Non-punitive reporting: Focuses on near-misses and root causes rather than individual blame.
  • Simulation Training: Safe method to rehearse high-risk, low-frequency events (e.g., ACLS, malignant hyperthermia) and improve interprofessional communication. c
  • Forcing Functions (Most Effective):
    • Engineered constraints that physically or digitally block unsafe actions.
    • Examples: Gas-specific keyed nozzles (prevents wrong-gas hookup), EHR chart lockouts for duplicate orders.
    • Complication: Alert fatigue (mitigate by reserving hard stops for severe risks only).
  • Standardization:
    • Checklists to eliminate omissions (e.g., central line insertion bundles to reduce CLABSI).
    • Uniform equipment across a hospital (e.g., identical code carts and defibrillators).
  • Universal Protocol (Prevents Wrong-Site/Wrong-Procedure Surgery):
    • Pre-procedure verification: Confirm Pt identity, procedure, and consent prior to OR entry.
    • Site marking: Marked by operating clinician while Pt is awake.
    • “Time-Out”: Active verbal pause by the entire surgical team in the OR immediately prior to incision.
  • Simplification & Technology:
    • CPOE: Eliminates handwriting errors and missed transcriptions.
    • CDSS: Automated real-time alerts for drug-drug interactions and renal dose adjustments.
  • Usability Testing: Pre-implementation trial of systems/devices to identify design flaws that provoke dangerous workarounds.

Human factors engineering

  • Highest Reliability
    • Forcing functions
      • Description: Hard stops in design or process to eliminate risk for incorrect use.
      • Example: Each anesthesia gas fits only one compatible socket & is not interchangeable.
    • Computerized automation
      • Description: Automated processes to remove human effort & variations that cause error.
      • Example: Automated vital signs monitoring.
    • Environment & physical layout
      • Description: Workspace design to facilitate correct action & minimize error.
      • Example: Look-alike drugs stocked in different locations.
  • High Reliability
    • Standardization & simplification
      • Description: Uniform processes to minimize variation, complexity & learning curve.
      • Example: Every hospital unit follows the same process for heparin administration.
    • Human-machine redundancy
      • Description: Repetitive step to confirm correct action in an error-prone process.
      • Example: Barcode scanning of medications in addition to visual inspection.
  • Medium Reliability
    • Reminders, alerts & double-checks
      • Description: Processes prompting providers to check actions to reduce errors.
      • Examples: Drug-drug interaction alerts; time-out before procedures.

Prevention of communication errors

  • Optimize Information
    • Prioritize acuity: Discuss the highest-acuity (sickest) pts first; triage discussion based on clinical stability. c
    • Targeted depth: Provide extra detail on critical/unstable pts while streamlining stable floor pts.
    • Limit cognitive burden: Avoid information overload; omit irrelevant past medical history or resolved issues to prevent omission of critical data.
  • Optimize Communication
    • Standardized structure: Follow a validated template or checklist (e.g., I-PASS: Illness severity, Pt summary, Action list, Situation awareness, Synthesis).
    • Dual-channel transmission: Combine concurrent oral and written communication (written sign-out paired with verbal run-through).
    • Interactive questioning: Encourage questions and active discussion throughout handoff rather than saving questions for the end.
    • Closed-loop verification: Use selective read-back for high-risk elements (e.g., urgent to-do items, critical pending lab results, medication changes).
  • Optimize Actions
    • Explicit task delegation: Maintain an updated action list (to-do items) with specific timelines and designated responsible team members.
    • Anticipatory guidance: Provide contingency planning using contingency (“if/then”) statements to guide overnight escalation (e.g., “If SBP < 90, give 500 mL bolus; if unresponsive, call ICU fellow”).