A Response to the ASTM SIF Reporting Standard

The ASTM SIF Reporting Standard

The ASTM SIF Reporting standard will:

• Improve the identification of incidents with Serious Injury and Fatality (SIF) potential.
• Standardize reporting beyond what is required by OSHA. The standard provides four levels of classification with clear definitions.
• Consider the potential severity of the outcome, not just the outcome itself.
• Provide opportunities for the development of new and different leading indicators.

The levels below show how SIF actual and SIF potential incidents fit into the level suggested in the standards:

• Level 1 – SIF Actual / Fatality
• Level 2 – SIF Actual / Serious Injury or Illness
• Level 3 – Minor Injury or Near Miss with SIF Potential
• Level 4 – Non‑SIF Potential Incident

The Origins of the SIF Potential Concept

In 2010, Dr. Tom Krause and his team sponsored the first major study on serious injury and fatality prevention. The study was triggered by data showing that recordable injury rates were declining, while fatal injuries remained level or were increasing.

Tom collaborated with EHS professionals from nine global organizations to study the problem. They concluded that the trend was the result of applying an outdated prevention model. This traditional model assumed that reducing incidents at the base of the safety triangle would automatically reduce serious injuries and fatalities. The problem with this approach is that it does not target the precursors associated with SIF potential (SIFp or pSIF) incidents.

Additionally, the control mechanisms and verification processes needed to be more robust. Identifying and controlling exposures with SIF potential requires a critical risk management system embedded in operations, leadership, and governance.

Adoption of the SIF Potential Model

Krause Bell Group is encouraged to see ASTM adopting the SIF Potential model into reporting criteria. Tom Krause and his team have worked with hundreds of organizations to establish effective SIF Prevention programs. One reason these efforts stall or fail is that organizations continue to reward or punish performance based on lagging indicators, such as the OSHA recordable rate.

We do wish the standard had gone further.

Benchmarking and Metrics

The standard suggests that Level 1 and Level 2 incidents can be benchmarked by converting the data to a rate using the following formula: Number of incidents × 1,000,000 / hours worked.

We agree that these metrics are benchmarkable. However, they have limited value. The number of SIF actual events is small. Within an organization, these events may occur months or even years apart, making it difficult to identify trends or develop effective prevention strategies. While Level 1 and Level 2 rates may replace the OSHA rate, they are not particularly helpful from a prevention standpoint.

The real value lies in Level 3 events — non-SIF actual incidents that have SIF potential (SIFp). However, the standard does not associate a metric with this level, nor does it consider it benchmarkable. The standard states that the number or rate of pSIF incidents is not designed to be benchmarkable, as comparison across organizations, industries, and time periods tends to negatively impact reporting when encouraging more reporting is the desired outcome.

We Respectfully Disagree.

We agree that benchmarking SIFp events across companies is problematic, largely because there is no consistent methodology for determining SIF potential. For example, some companies classify a fall from four feet as having SIF potential, while others use six feet. However, if a global organization establishes an internal SIFp classification standard, benchmarking internally becomes both possible and valuable.

We have seen this with our own clients. When a corporate standard is established, and SIFp data is collected consistently across sites, regions, and business units, organizations gain valuable insight into where their greatest SIF exposures exist and can develop actionable mitigation strategies.

We believe that creating a global benchmarking standard on a set of pSIF metrics, including a rate, is possible. We believe that the model presented in the standard, if adopted across organizations, could lead to a benchmarkable set of pSIF metrics.

Establishing Metrics Does Not Reduce Reporting

Establishing a rate does not negatively impact reporting. If that were the case, the same argument would be applied to Level 1 and Level 2 rates. Leaders will debate whether an injury meets a definition: Is the burn 2.85 inches or three inches? Is the laceration 1.9 inches or two inches? We have seen situations where more time is spent debating recordability definitions than identifying effective controls.

The problem is not the metric, but how the metric is used. When lagging metrics are tied to reward systems, leadership discussions often shift toward the rate itself rather than prevention activities. In large organizations with hundreds of leaders, it is difficult to ensure everyone values accurate reporting over protecting the rate.

This is why we often encourage organizations to establish leading indicators focused on mitigation activities, helping balance the focus between reporting and prevention. From our experience, having a set of metrics associated with SIFp, including a rate, is beneficial. It allows organizations to determine whether they are improving or declining in their control of SIF exposures, while also providing insight into reporting behavior.

Section X1: pSIF Criteria

In Section X1, ASTM provides a model for determining SIF potential (pSIF). The authors are particularly pleased to see this model included. The model originated from an early attempt to develop AI to determine pSIF. Our goal was to create a uniform approach that would eliminate the need for every organization to build a decision tree from scratch. The tool also needed to be intuitive and easy to implement.

To develop the model, the authors reviewed and classified tens of thousands of incidents, refined the exposure list multiple times, rewrote and clarified exposure definitions, and engaged safety professionals to align on pSIF scenarios. The version included in the standard is Version 1. The model has been refined and enhanced since its development in 2021.

The core process for determining pSIF remains the same:

1. Classify the exposure associated with the incident
2. Then determine whether pSIF exists

Not the other way around.

This model provides a secondary benefit: all incidents are classified by exposure, not just those with SIF potential. This gives organizations actionable insight into immediate causal factors across all incidents.

SIF Prevention Is More Than Classification

Overall, Krause Bell Group is pleased with the revised reporting standard. Integrating SIF potential into the standard is a crucial step in advancing SIF Prevention and encouraging broader sharing of effective control strategies.

SIF Prevention has three major components:

• Knowledge and Acceptance
• Reactive Response
• Proactive Precursor Identification and Mitigation

A successful SIF Prevention strategy begins with senior leadership understanding the science behind the approach and recognizing the value the system brings. Once leadership accepts the approach, everyone in the organization must understand the concept and their role in it. This standard will help demonstrate to leaders that this approach has merit.

When an incident occurs, the organization must quickly determine the contributing exposure and whether pSIF exists or not. If pSIF equals yes, the incident should receive elevated attention, including a robust root cause analysis (RCA).

Corrective actions from the RCA should include controls at the top of the hierarchy of controls. Investigation results should then be shared with other locations that have similar exposures, along with requirements to implement the relevant controls.

The final component of SIF Prevention involves proactively identifying SIF hazards, verifying that critical controls are deployed, and ensuring controls are used sustainably.

Conclusion: Measurement Drives Attention and Action

For more than two decades, we have seen how classifying events for SIF potential helps organizations better understand the precursors to serious SIF events and develop effective mitigation strategies. Establishing a standard approach will help more organizations align on the identification of SIF potential events and move us closer to our shared goal of eliminating serious injuries and fatalities in the workplace. If you would like an assessment of your SIF Prevention program, or assistance establishing one, please contact us for a consultation.