A safety dashboard can tell
A safety dashboard can tell a reassuring story: recordables down, lost-time cases down, rates trending in the right direction. Leaders relax. Teams get recognized for another “injury-free” month.
- By Jesse Jacobs
- Feb 18, 2026
A safety dashboard can tell a reassuring story: recordables down, lost-time cases down, rates trending in the right direction. Leaders relax. Teams get recognized for another "injury-free" month. Yet on the floor, people still talk about close calls, sore backs, small cuts handled quietly, and hazards everyone knows, but nobody wants to put in writing.
If that sounds familiar, you may not have a data problem. You may have a trust problem. Safety metrics don't just measure performance—they shape it. When the metric becomes the goal, people learn how to protect the numbers rather than protect each other. The result is a culture where the scoreboard looks great, but risk remains the same (or worsens) because the system has trained people to keep bad news out of the official record.
Why "good numbers" can mislead
Most organizations rely heavily on lagging indicators such as Total Recordable Incident Rate (TRIR), Days Away, Restricted or Transferred (DART) rate, lost-time injury rates, and severity. They are useful for tracking outcomes and benchmarking. But they share a fundamental weakness: they depend on events being reported and classified correctly.
Research on the U.S. Survey of Occupational Injuries and Illnesses (SOII) has documented concerns about undercounting in employer-reported injury and illness data and has explored why cases may be missed or misclassified. In comparisons between employer logs and other sources, mismatches were often attributed to recordkeeping errors, data transfer issues, and misunderstandings of reporting rules, rather than to the absence of harm [1].
In other words: "good numbers" can be produced by the way a system is designed.
A five-question trust check
Before you redesign anything, run a quick diagnostic. Ask a cross-section of workers (different shifts, crews, and languages) these five questions and compare answers. The goal is not to "score" people—it is to find where fear, friction, or confusion is blocking the flow of information.
- Can you submit a near-miss or hazard report in under two minutes, using the tools you have?
- Do you believe you can report an injury and still be treated fairly—no blame, no hassle, no pay or job impact?
- When someone reports, do they get acknowledged and see what changed?
- What happens at the first point of contact: curiosity and support, or interrogation and disbelief?
- Are rewards, bonuses, or praise tied to "no incidents," or to prevention and learning activities?
If you see big gaps between leadership and frontline answers, start there. A reporting system cannot be "fixed" with new metrics if workers do not trust what happens after they speak up.
Three common ways metrics "lie."
1) Underreporting driven by fear or perceived consequences.
When workers believe reporting will lead to blame, discipline, stigma, or hassle, reporting drops—especially for less visible conditions and near misses. In one workplace case study, fewer than 5% of workers had officially reported a work-related injury or illness in the prior year, while over 85% reported work-related symptoms; 50% had persistent problems; and 30% reported lost time or work restriction. Workers cited reasons such as fear of reprisals and doubts about management's willingness to respond. [2]
2) Misclassification (often from confusion, not malice).
Recordkeeping is complex. "First aid vs. medical treatment," restricted work definitions, and work-relatedness judgments can lead to inconsistent classification. SOII undercount research highlights how incomplete or inaccurate logs and misunderstandings of recordkeeping requirements can contribute to missing cases. [1]
References:
[1] U.S. Bureau of Labor Statistics, Monthly Labor Review (2016). An update on SOII undercount research activities. DOI: 10.21916/mlr.2016.41. https://www.bls.gov/opub/mlr/2016/article/an-update-on-soii-undercount-research-activities.htm
[2] Pransky G., Snyder T., Dembe A., Himmelstein J. (1999). Under-reporting of work-related disorders in the workplace: a case study and review of the literature. Ergonomics. DOI: 10.1080/001401399185874. https://doi.org/10.1080/001401399185874
[3] Lipscomb H.J., Nolan J., Patterson D., et al. (2013). Safety, incentives, and the reporting of work-related injuries among union carpenters. American Journal of Industrial Medicine. DOI: 10.1002/ajim.22128. https://doi.org/10.1002/ajim.22128
[4] OSHA. 29 CFR 1904.35 (Employee involvement; reporting; anti-retaliation). eCFR: https://www.ecfr.gov/current/title-29/subtitle-B/chapter-XVII/part-1904/section-1904.35
[5] OSHA. Recommended Practices for Safety and Health Programs (OSHA 3885). PDF: https://www.osha.gov/sites/default/files/publications/OSHA3885.pdf
[6] Lestari E.D., Kurniawidjaja L.M. (2020). Effectiveness of Web-Based Near Miss Reporting Program on Preventing Occupational Accident. Proceedings of the 5th Universitas Ahmad Dahlan Public Health Conference (UPHEC 2019), Advances in Health Sciences Research, vol. 24, Atlantis Press, pp. 178–183. DOI: 10.2991/ahsr.k.200311.035. https://doi.org/10.2991/ahsr.k.200311.035
[7] OSHA. Using Leading Indicators to Improve Safety and Health Outcomes. PDF: https://www.osha.gov/sites/default/files/publications/OSHA_Leading_Indicators.pdf