Analyzing human factors affecting severe maternal morbidity (SMM) using fuzzy Bayesian network (FBN)

Area of Science:

• Obstetrics and Gynecology
• Medical Quality Improvement
• Human Factors Engineering

Background:

• Severe maternal morbidity (SMM) is a critical indicator of obstetric care quality.
• Human error is frequently associated with SMM events.
• Understanding human factors is crucial for improving maternal outcomes.

Purpose of the Study:

• To analyze human factors contributing to SMM.
• To apply fault tree analysis (FTA) and fuzzy Bayesian networks (FBN) for SMM cause analysis.
• To identify specific human errors impacting obstetric care quality.

Main Methods:

• Utilized morbidity data and expert interviews.
• Employed fault tree analysis (FTA) to structure causal relationships.
• Applied fuzzy Bayesian networks (FBN) with and without common cause failures (CCFs) to estimate error probabilities.
• Incorporated L-NOR gate for simplifying conditional probability tables (CPTs).

Main Results:

• Key contributors to SMM included delayed resuscitation, inadequate hemorrhage management, and poor team coordination.
• These factors demonstrated the highest influence on SMM occurrence.
• Final SMM probabilities were estimated as 0.0196 (FFT), 0.0193 (FBN without CCFs), and 0.0167 (FBN with CCFs).

Conclusions:

• The integrated FTA and FBN approach, especially with the L-NOR gate, offers a more accurate modeling of complex cause-and-effect relationships in SMM.
• Strategies to reduce SMM include enhancing team coordination, improving hemorrhage management, and enforcing standard protocols.
• Findings support evidence-based policy-making to elevate obstetric care standards.