A dynamic approach to FMEA-RPN risk classification: comparing with the classical FMEA method

Area of Science:

• Healthcare Risk Management
• Ergonomics
• Clinical Informatics

Background:

• Traditional Failure Mode and Effects Analysis (FMEA) using fixed thresholds inadequately addresses dynamic risks in high-acuity settings like Neonatal Intensive Care Units (NICUs).
• Existing risk assessment methods lack consideration for contextual and system dynamics crucial for patient safety in specialized care environments.

Purpose of the Study:

• To introduce a novel dynamic classification model for prioritizing risks in high-risk healthcare environments.
• To enhance risk assessment by integrating contribution percentages and percentile ranks within a decision matrix for context-sensitive prioritization.

Main Methods:

• Identified 21 ergonomic failure modes in a NICU setting through observations, interviews, and literature review.
• Calculated Risk Priority Numbers (RPNs) using classical FMEA and a proposed dynamic model incorporating a decision matrix.
• Assessed model robustness via sensitivity analysis and evaluated agreement with practitioner classifications using weighted measures.

Main Results:

• The dynamic model reclassified previously medium-risk factors (e.g., long shift hours, alarm sounds) as very high risk.
• Sensitivity analysis confirmed model robustness to changes in Occurrence, Severity, and Detection ratings.
• Achieved a high agreement rate of 93.7% between the model's classifications and nurse assessments.

Conclusions:

• The dynamic framework offers context-sensitive risk prioritization by considering failure mode contribution and RPN distribution.
• This approach provides a transparent and reproducible method for risk-focused improvement planning in critical care settings.
• The model demonstrates superior risk classification compared to classical FMEA for NICU ergonomic failures.