Rethinking Post-Incident Reviews: Embracing Contributing Factors Over Root Causes

Incidents in complex technological systems are inevitable, and how organizations learn from them is crucial for future resilience. Traditional post-incident reviews often focus on identifying a singular "root cause," which oversimplifies the multifaceted nature of modern systems. Shifting the focus from seeking root causes to analyzing contributing factors aligns with contemporary scholarship in system safety and human factors engineering.

Failures in complex systems are rarely due to a single point of failure. John Allspaw (2021) emphasizes that labeling an element as the "root cause" often serves more as psychological comfort than a technical explanation, providing a false sense of security by oversimplifying intricacies. Similarly, Richard Cook (1998) argues that catastrophic failures require multiple faults that align in unexpected ways, highlighting that safety is an emergent property of systems, not merely the absence of failure.

Nancy Leveson's work on System-Theoretic Accident Model and Processes (STAMP) underscores the importance of viewing accidents as results of systemic interactions rather than individual component failures (Leveson, 2004). James Reason's Swiss Cheese Model illustrates how defenses against failure are layered, and how holes in these layers can align to permit a hazard, reinforcing that accidents are due to multiple, often latent, failures aligning in unforeseen ways (Reason, 2000).

In my previous article, I critiqued the "Five Whys" method for fostering a culture of blame and inadequately addressing systemic issues. In that article, I proposed shifting to the "Five Hows," focusing on how systemic processes contribute to incidents (Henry, 2023). This approach aligns with Sidney Dekker's notion of "drift into failure," where complex systems degrade over time due to normal performance variations, leading to failure without any singular root cause (Dekker, 2011).

Limitations of Root Cause Analysis

Root Cause Analysis (RCA) has been a staple in incident investigation, aiming to identify the fundamental reason for a failure. While effective in simple systems, its application in complex systems is problematic. The root cause approach focuses on linear causality, neglecting the intricate web of interactions characterizing complex systems.

Embracing Contributing Factors

Adopting a contributing factors approach acknowledges the interplay of various elements within a system. This perspective allows organizations to:

• Develop a Holistic Understanding: By considering all factors that contribute to an incident, organizations better comprehend system complexities.
• Promote a Blameless Culture: Focusing on systemic issues rather than individual faults encourages open communication and learning.
• Enhance System Resilience: Identifying and addressing multiple contributing factors improves overall system robustness.

Incorporating the Five Hows

The "Five Hows" methodology shifts the focus from "Why did this happen?" to "How did this happen?" promoting deeper exploration of mechanisms and interactions leading to an incident. For example:

1. How did the system fail?
2. How did existing safeguards fail to prevent the incident?
3. How did organizational processes contribute to the failure?
4. How did team actions or decisions influence the outcome?
5. How can we modify the system to prevent future incidents?

By iteratively asking "How," organizations delve into operational, organizational, and technical layers contributing to failures, fostering continuous improvement.

A Framework for Effective Post-Incident Reviews

To implement this contributory approach, organizations can adopt the following framework:

1. Comprehensive Event Narrative: Document a detailed, blame-free timeline and account of the incident.
2. Identification of Contributing Factors: List all technical, human, environmental, and organizational factors involved.
3. Systemic Analysis: Use methodologies like STAMP or the Swiss Cheese Model to understand factor interactions.
4. Application of the Five Hows: Explore how each contributing factor influenced the incident and address them.
5. Development of Actionable Solutions: Formulate strategies targeting systemic improvements over quick fixes.
6. Feedback and Learning Loop: Establish mechanisms for sharing lessons and integrating them into practices.

Conclusion and Call to Action

Modern technological systems' complexity necessitates reevaluating traditional post-incident review practices. Organizations must move beyond searching for a singular root cause and embrace analyzing contributing factors. Adopting methodologies like the Five Hows and focusing on systemic improvements enhance resilience, foster a culture of continuous learning, and better navigate future challenges.

Organizations should revise their post-incident review processes to integrate the contributing factors approach and the Five Hows methodology. This shift aligns with best practices and promotes a proactive stance on system safety and reliability. Leaders must facilitate training and discussions to embed this mindset within teams, transforming incidents into opportunities for meaningful growth and improvement.

References

• Allspaw, J. (2021). What We Talk About When We Talk About 'Root Cause'. Retrieved from GitHub .
• Cook, R. I. (1998). How Complex Systems Fail. Cognitive Technologies Laboratory, University of Chicago.
• Leveson, N. G. (2004). A New Accident Model for Engineering Safer Systems. Safety Science, 42(4), 237-270.
• Reason, J. (2000). Human Error: Models and Management. BMJ, 320(7237), 768-770.
• Henry, D. (2023). From Why to How: Rethinking Root Cause Analysis. Retrieved from LinkedIn .
• Dekker, S. (2011). Drift into Failure: From Hunting Broken Components to Understanding Complex Systems. Ashgate Publishing.