Answers are grounded in this paper and cite their sources.
Personal injury rates such as TRIR and LTIR cannot warn an organisation that a major accident is building. They count frequent, low-consequence harm to individuals, while major accidents come from degraded barriers. Longford (1998), Texas City (2005) and Deepwater Horizon (2010) all occurred at sites with good personal safety records. This practitioner working paper proposes four leading indicators of barrier health that any process or power plant can report monthly from records it already keeps: (1) safety-critical safeguards bypassed or inhibited, reported by age; (2) overdue inspection, proof-test and maintenance work on safety-critical elements; (3) temporary management-of-change authorisations past their expiry date; and (4) the verified effectiveness rate of incident actions, reported alongside closure rate. For each indicator it gives a definition, calculation, data source and suggested trigger for management action, and it sets out ownership, controls against gaming, and the fit with ISO 45001 performance evaluation. The paper completes a series with "The Change Nobody Risk-Assesses" (MOC recognition failures) and "Closed Is Not Effective" (verification of incident recommendations).
No takes yet. Share an insight, caveat, or question.
Adnan Zafar (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: