TRIR vs LTIFR vs DART vs SIF Rate: Which Metric Fits Board Decisions?

TRIR, LTIFR, DART, and SIF rate are not four names for the same job. A board uses one metric to read the rearview mirror, another to compare sites, a third to see disability impact, and a fourth to govern fatal risk. When leaders blur those jobs, the dashboard looks simpler and the decisions get worse.

Across 25+ years of executive EHS work and more than 250 cultural transformation projects supported by Andreza Araujo, the pattern has been consistent. The room gets calmer when each metric has one owner and one purpose, and the room gets noisier when a single number is asked to prove that the whole system is healthy. As Andreza Araujo writes in Safety Culture: From Theory to Practice, repeated decisions reveal culture faster than slogans do. That is why the right question here is not which metric is best in general, but which metric should own which decision.

This comparison is for board members, plant managers, EHS leaders, and HR partners who need a metric stack that survives production pressure. If you want the broader language behind metric design first, read Safety Indicators Explained: 4 Types Leaders Should Not Mix and Safety Metric Dictionary before you redraw the review pack.

Key Takeaways

• TRIR is useful for recordkeeping and external trend context, but it should not be the main proof that fatal risk is controlled.
• LTIFR can help compare sites and periods, although it can hide how local classification rules shape the result.
• DART adds a disability and restricted-work lens, which makes it useful for operations and HR, but it still arrives after the event.
• SIF rate is the strongest of the four for board-level fatal-risk governance when the definition is strict and paired with control verification.
• The best dashboard does not pick one winner. It assigns each metric to the decision it can actually support.

Why These Four Metrics Are Not Interchangeable

TRIR, LTIFR, and DART are outcome metrics. They summarize harm that already happened, which makes them useful for reporting and trend review, but not enough for deciding whether the control stack is still holding. SIF rate is different because it is supposed to ask a sharper question about serious exposure, yet it only works when the company defines serious potential clearly and keeps the metric tied to field evidence.

That distinction matters because OSHA recordkeeping gives TRIR its administrative shape, while company practice usually defines LTIFR and SIF rate. In other words, the board should not assume that a single rate means the same thing across regions, business units, or contractors. The article Metric Ownership: 5 Traps That Turn Safety Dashboards Into Theater makes the same point from a governance angle, and it matters here because a metric without ownership becomes decoration.

The serious-risk context also matters. BLS fatal occupational injury reporting exists because the board needs a way to think about what can kill people, not only about what lands in a logbook. James Reason's latent-failure logic points in the same direction. The bad outcome is visible first, but the decision that allowed it often happened much earlier.

Evaluation Criteria

Before comparing the metrics one by one, use six criteria. The board should ask what each metric owns, how close it sits to fatal risk, how easy it is to game, how comparable it is across sites, how quickly it changes action, and how well frontline leaders can explain it without translation.

These criteria are practical, not mathematical. They help a board decide where each measure belongs in the governance stack, which is more useful than debating whether one number is universally good or bad.

TRIR

TRIR, the Total Recordable Incident Rate, is the metric most leaders know first because it fits the language of recordkeeping and external benchmarking. In U.S. practice it is usually normalized to 200,000 work hours, which makes it easy to compare periods and sites as long as the organization keeps the case definition stable. That stability is the catch, because the metric becomes less trustworthy when case classification drifts or when managers start treating the number as a performance target instead of a reporting tool.

TRIR is good for seeing whether the organization is reporting incidents consistently. It is weak for seeing whether the most serious controls are still healthy. A site can lower TRIR while still carrying poor isolation discipline, weak permit quality, or serious contractor exposure. The article SIF Rate vs TRIR vs Precursors: Which Metric Fits explains that gap from a more direct fatal-risk angle, and the gap matters because low recordables can hide a high-energy system.

Use TRIR at board level as context, not as the lead voice. If the board gives TRIR the final word, the room starts optimizing for case count instead of exposure control. That is the wrong incentive, and it is one reason TRIR belongs beside control verification, precursor review, and field checks rather than above them.

LTIFR

LTIFR, the Lost Time Injury Frequency Rate, answers a different question. It asks how often injuries are turning into lost time, and that can be useful when leaders want to compare sites, regions, or periods with a measure that reflects disruption to work. HR and operations often like LTIFR for that reason, because the metric catches cases that have business impact even when the recordable volume is small.

The weakness is that LTIFR depends on local rules about lost time, restricted work, return to work, contractor hours, and case classification. If those rules change, the metric can change without the risk changing at the same pace. That does not make LTIFR useless. It makes it sensitive to definitions, which is exactly why it should not be the only number that drives the board conversation.

Use LTIFR when you need comparability across a group and when you can hold the classification rule steady. It works well as a management trend metric, especially when the board wants to know whether a site is causing more work disruption than the rest of the portfolio. It is less useful when the real question is whether a fatal-risk barrier is still intact.

DART

DART, the Days Away, Restricted, or Transferred rate, is useful because it captures the burden of cases that affect work capacity and job design. For plant leaders, that matters. A DART shift can tell you that the work system is creating cases that are serious enough to remove people from normal duty or move them into restricted tasks, which means the operation is carrying more than a simple bruise count.

At the same time, DART is still a lagging measure. It reacts after the event, and it can still be distorted by local medical practice, case classification, and return-to-work decisions. A site can look calmer on DART while the serious exposure remains untouched. The article Leading Indicators: 6 Gaps That Make Safety Activity Look Preventive is useful here because it shows why activity alone does not prove control.

Use DART when the board wants to understand workload impact, disability burden, and operational disruption. Do not use it as the main fatal-risk compass. It can tell you that harm is affecting people and schedules, but it cannot tell you whether the system is still one step away from a much worse event.

SIF Rate

SIF rate, when it is defined well, is the metric closest to board-level fatal-risk governance among the four. It asks which serious injury or fatality exposures are present, whether the organization has credible precursors, and whether the control stack is being verified where the energy is highest. That is why many leaders use it together with critical control checks and serious near-miss review rather than as a standalone injury statistic.

This metric has one important condition. It must be defined tightly enough that the board can tell the difference between a real serious-risk signal and a prettier version of TRIR. If the company labels every inconvenience as SIF potential, the metric loses force. If the company defines it too narrowly, the board misses the exposures that matter most. The article Critical Control Verification: 7 Audit Traps That Leave SIF Risk Untouched is the right companion because SIF rate only means something when the controls are verified in the field.

Patrick Hudson's maturity thinking and James Reason's latent-failure logic both point to the same practical conclusion. If the organization wants to know whether fatal risk is moving in the right direction, it must look earlier than the injury count and deeper than the monthly summary. SIF rate belongs in that earlier and deeper conversation. It is the strongest of the four for board governance, provided that it is paired with precursor indicators and control assurance.

Decision Matrix

The table below uses a simple 1 to 5 scale. Five means the metric is strong for that criterion. The scores are practical, not universal, because local definitions and governance rules still matter.

Read the matrix as a governance map. TRIR speaks well to outside audiences, LTIFR helps compare operational burden, DART shows impact on work capacity, and SIF rate owns the serious-risk conversation. A board that wants one number to tell the whole story is asking for a simplification that the work cannot support.

Recommendation Per Context

For the board, make SIF rate the lead metric and keep TRIR as context. That order protects the board from mistaking recordkeeping cleanliness for serious-risk control. The board should still see LTIFR and DART, but those numbers should answer disruption questions, not decide whether fatal-risk barriers are healthy.

For the plant manager, use LTIFR and DART to see whether the operation is producing work disruption, then pair those numbers with SIF rate and critical-control checks. That mix helps the manager move resources before a small case pattern turns into a larger control failure. If the manager needs a stronger weekly routine, How to Run a Leading-Indicator Quality Audit in 30 Days gives a useful next step.

For the EHS leader, keep the metric stack explicit. Say which number is for reporting, which is for comparison, which is for disability impact, and which is for fatal-risk governance. That one sentence removes a lot of confusion and makes it much harder for a dashboard to drift into theater.

For HR, DART matters because it shows how cases affect work design, accommodation, and return to duty. HR should not try to own TRIR or SIF rate alone, because those measures need operational evidence. HR helps most when it gives the board a clear view of restricted work, transfers, and the human cost of repeat cases.

FAQ

Is SIF rate better than TRIR?

Not in every situation. SIF rate is better for fatal-risk governance, while TRIR is better for recordkeeping context and external trend review. The right answer is usually both, with clear jobs for each one.

Should the board stop reviewing TRIR?

No. The board should stop letting TRIR act like the final safety score. It still has value as a reporting and benchmarking metric, but it should sit beside stronger fatal-risk evidence.

Why does LTIFR still matter?

LTIFR still matters because it shows how often injuries are causing lost time, which helps leaders understand operational disruption. The metric becomes weak when people change case definitions or use it as a substitute for serious-risk review.

Can DART be gamed?

Yes. DART can move if local classification, return-to-work practice, or restricted-work rules change. That is why the number should be checked against field evidence and not read as a stand-alone truth.

What should sit beside SIF rate?

SIF rate should sit beside precursor indicators, critical-control verification, serious near-miss review, and field evidence. That combination shows whether the risk path is changing before the injury count does.

If your current dashboard still mixes reporting, comparison, and control assurance into one block, rebuild it in that order instead. Andreza Araujo's books, advisory work, and safety culture framework are built for that kind of cleanup, and the store at loja.andrezaaraujo.com is the best place to start.