DART Rate: 6 Distortions Leaders Must Fix

OSHA recordkeeping normalizes injury rates around 200,000 work hours, which means a single restricted-duty case can look mathematically equal across very different risk realities. This article shows how senior leaders should read DART rate without letting it hide SIF exposure, underreporting, contractor risk, or weak corrective action.

Why DART rate is not a safety verdict

DART rate measures cases that involve days away, restricted duty, or job transfer, using the OSHA recordkeeping formula of cases multiplied by 200,000 and divided by hours worked. The metric is useful because it creates a common denominator, although that denominator can mislead directors when workforce size, overtime, and case management practices differ sharply between sites.

As Andreza Araujo argues in Safety Culture: From Theory to Practice, a safety indicator only matters when it changes leadership behavior. A plant with a DART rate of 0.6 may still have uncontrolled energy, poor permit discipline, or weak contractor supervision, which means the number is clean while the operation remains exposed.

The board should treat DART as a lagging symptom, not as proof of cultural maturity. The better executive question is not whether the number improved, but whether the organization can explain what changed in critical controls, supervision routines, and serious-injury prevention.

1. DART rate can punish honest reporting

DART rate becomes distorted when supervisors know that one restricted-duty decision will affect the site dashboard for the year. Because the OSHA 300 log feeds the classification, a minor disagreement over restricted work can change the rate even when the underlying hazard profile has not changed.

Across 25+ years leading EHS at multinationals, Andreza Araujo has observed that pressure on lagging indicators often changes reporting behavior before it changes field behavior. This is the same trap addressed by the existing article on leading indicators that TRIR will never show, where clean lagging numbers can coexist with weak precursor signals.

Executives should audit the path from injury notification to classification. If the medical restriction decision is debated mainly because of its dashboard impact, the metric has started to manage the conversation instead of describing the event.

2. Small sites suffer mathematical volatility

DART rate uses the 200,000-hour multiplier, which approximates 100 full-time workers working 40 hours per week for 50 weeks. In a 35-person site, one DART case can produce a rate that looks catastrophic, while a 2,000-person site can absorb several cases without the same visual shock.

This matters in contractor prequalification, because procurement teams often compare rates without normalizing exposure hours, task severity, or project phase. A contractor performing high-energy shutdown work may look worse than a low-risk supplier simply because the rate is more sensitive to one case.

For a fair reading, segment the dashboard by exposure hours and task profile. A board report should show the raw count, worked hours, DART rate, task risk, and SIF potential, since the rate alone cannot tell whether the organization is improving or merely moving through a small-number cycle.

3. DART rate ignores high-potential near misses

DART rate records injury consequences after a case has occurred, which means it can miss high-potential near misses where nobody was hurt. A dropped load, failed isolation, or uncontrolled line of fire may produce a DART rate of zero, although the next repetition could become fatal.

Andreza Araujo's work with more than 250 companies shows why safety dashboards need SIF potential as a separate lens. In the article on risk matrix failures that hide serious risk, the same pattern appears when teams score probability too low because the event has not yet injured anyone.

A senior EHS manager should require every DART review to include high-potential near misses, critical-control failures, and open corrective actions. The indicator becomes useful when it sits beside the events that almost became injuries, because those events often tell the truth earlier.

3 lenses should appear together in the monthly dashboard: DART rate, SIF potential, and critical-control verification. When one of the three is missing, leadership sees only part of the risk system.

4. Restricted-duty management can mask severity

Restricted duty is legitimate when it protects recovery and keeps the worker safely engaged, although it can also become a cosmetic strategy. If a company designs modified work mainly to avoid days away, the DART category remains visible but the real Severity Rate conversation may be softened.

In her Portuguese title Muito Alem do Zero, translated as Far Beyond Zero, Andreza Araujo argues that zero-centered goals can create behaviors that protect the target rather than the person. DART can produce a similar effect when leaders reward low rates without examining the quality of medical decisions, job transfer practices, and return-to-work conversations.

The fix is procedural. Every restricted-duty case should receive a short executive note that explains the medical basis, the work limitation, the task assigned, and the worker's recovery status. That note protects the worker and the integrity of the indicator.

5. Contractor DART can hide host-company risk

Contractor DART rate often reflects the contractor's reporting system, while the host company controls site access, permits, energy isolation, supervision quality, and simultaneous operations. A low contractor DART rate does not prove that the host organization has designed safe work.

During the PepsiCo South America tenure, where the accident ratio fell 50% in six months, it became clear that safety improvement depended on management routines rather than slogan discipline. The same logic applies to contractors, because host-company routines shape whether a subcontractor sees risk early enough to stop the job.

Prequalification should compare the contractor's DART rate with permit quality, supervisor presence, corrective-action closure, and stop-work evidence. If the contractor has a clean rate but poor field controls, the procurement decision is importing hidden risk.

Each month in which contractor DART is reviewed without field-control evidence creates a false sense of governance, while exposure continues in shutdowns, maintenance windows, and non-routine work.

6. DART rate should be paired with speak-up data

DART rate tells leaders what became recordable, while speak-up data shows whether the organization is willing to surface weak signals before harm occurs. A site with low DART and low reporting may be quieter, not safer.

The connection with speak-up metrics leaders should track is direct. When employees do not challenge a shortcut, report a near miss, or question a permit, the lagging indicator stays clean until the system produces a serious event.

The monthly dashboard should include reporting density, time to supervisor response, percentage of reports closed with field verification, and themes repeated across shifts. 4 voice signals can reveal culture before DART moves, especially in plants where fear or fatigue suppresses early warnings.

7. DART reviews must trigger better investigation

DART cases deserve investigation because they reveal injury consequences, but the investigation should not stop at classification, retraining, or worker behavior. James Reason's Swiss Cheese Model helps leaders examine latent failures, active failures, and weakened barriers without reducing the case to the last person involved.

The existing RCA guidance on avoiding the operator error trap is essential here. A DART case that ends with a generic retraining action has consumed leadership attention without repairing the system that produced the exposure.

For every DART case, require one corrective action tied to a physical control, one tied to a management routine, and one tied to verification. This structure prevents the dashboard from becoming a historical archive and turns the rate into a gateway for risk reduction.

DART rate vs executive safety intelligence

DART rate should stay in the dashboard, although it should never sit alone. The executive view needs a layered interpretation that separates compliance classification from operational risk intelligence.

DART can improve for technical reasons or cultural reasons, and leaders need to know the difference. The companion article on underreporting in safety shows how incentive pressure, supervisor reactions, and field evidence can distort clean numbers.

Conclusion

DART rate is valuable when leaders read it as one evidence stream, but it becomes dangerous when it replaces SIF potential, speak-up, contractor control, and investigation quality.

For leaders who need a dashboard that can survive board scrutiny and still make sense on the shop floor, Andreza Araujo and ACS Global Ventures support safety culture diagnostics, executive dashboards, and practical implementation. Safety is about coming home, and a metric is only useful when it helps that happen.