Severity Rate: How to Calculate It in 30 Minutes

Severity rate is the safety metric that shows how heavy injury and illness consequences are by relating lost or restricted days to hours worked. It does not replace TRIR, DART, or SIF review, because it answers a different question: not how many cases happened, but how much work capacity was damaged after those cases.

2.5 million nonfatal workplace injuries and illnesses were reported by U.S. private industry employers in 2024, according to the BLS annual release. This guide shows an EHS manager how to calculate severity rate in 30 minutes, then audit whether the number reflects actual risk or only clean recordkeeping.

Why does severity rate matter for EHS managers?

Severity rate matters because frequency alone can hide the operational weight of harm. A site with 2 minor first-aid-like recordables and a site with 2 injuries that remove workers from normal duty for 120 calendar days may look similar in a count-based review, although the second operation carries a very different human, production, and control-failure burden.

As Andreza Araújo argues in Muito Além do Zero, glossed in English as Far Beyond Zero, the absence of accidents or a clean number does not prove capability. The same critique applies to severity rate. A low number can mean real prevention, but it can also mean underreporting, weak return-to-work review, or a denominator inflated by hours that dilute visible pain.

EHS managers should treat severity rate as a trigger for investigation, not as a verdict. Read it beside lagging indicator limits, near-miss quality, SIF precursors, and corrective-action closure, because severe outcomes rarely appear without earlier signals that the dashboard could have captured.

Step 1: Define which days count?

Severity rate starts with a clear rule for counted days, usually days away from work, restricted work days, transferred days, or a locally defined lost-time day set. OSHA's recordkeeping rule under 29 CFR Part 1904 requires many employers with more than 10 employees to maintain OSHA Forms 300, 300A, and 301, and OSHA explains that those forms support recording, reporting, and electronic submission.

The first trap is changing the numerator after leadership sees the number. Across 25+ years leading EHS at multinationals, Andreza Araújo has observed that metrics lose credibility when the organization debates definitions only after a bad month. The definition should be written before the period begins, approved by EHS and operations, and applied even when it makes the graph uncomfortable.

For a 30-minute review, open the injury log and mark 3 columns: days away, restricted days, and transferred days. Decide whether your severity rate will include only days away or a broader lost-capacity total. Keep both views if possible, because days away show absence while restricted days show work redesigned under medical or operational limits.

Step 2: Pull total hours worked?

Total hours worked is the denominator that makes severity rate comparable across teams, departments, and months. The standard OSHA incidence-rate logic uses 200,000 hours as a base for 100 full-time workers, and many organizations adapt the same base for severity calculations so that one month can be compared with another.

The denominator is not neutral. If contractors, overtime, temporary labor, or cross-charged maintenance hours are excluded inconsistently, the rate becomes a political number rather than a safety signal. This is why safety KPI audits for false confidence should test the hour source before interpreting any trend.

Pull hours from payroll or workforce planning, not from a rounded estimate in a slide. Separate employee and contractor hours when the exposure profile differs, then keep a combined view for executive review. If the site cannot produce hours within 30 minutes, the first corrective action is not a new chart. It is fixing the data pipeline.

Step 3: Apply the severity rate formula?

The practical severity rate formula is counted lost-capacity days divided by total hours worked, multiplied by 200,000. If a plant had 84 counted days and 420,000 hours worked, the severity rate would be 40.0 because 84 divided by 420,000, multiplied by 200,000, equals 40.

200,000 hours is a normalization base, not a proof of safety maturity. As Andreza Araújo writes in Diagnóstico de Cultura de Segurança, indicators only become useful when they help leaders understand what the system is producing. A formula can standardize the number, but it cannot explain whether a worker delayed reporting pain for 3 weeks or whether a supervisor redesigned the job too late.

Use a visible calculation sheet with 5 fields: reporting period, counted days, total hours, multiplier, and final rate. Then add a note field for unusual events, such as a single severe case, shutdown hours, or a temporary labor surge. Without the note, the rate may be mathematically correct and operationally misleading.

Step 4: Separate frequency from severity

Frequency and severity answer different management questions. Frequency asks how often injury or illness cases occur, while severity asks how much work capacity is lost after those cases, which means a month with fewer cases can still be worse if one injury produced long absence or extended restriction.

The market often treats lower frequency as proof of progress because it is easier to celebrate. Andreza Araújo's critique of zero-accident targets warns against that shortcut. In Far Beyond Zero, the core warning is that a protected number can become more important than protected people, especially when bonuses, public rankings, or executive praise depend on the graph.

Build a 2-by-2 view. One axis is frequency, such as TRIR or recordable count. The other is severity rate. A low-frequency, high-severity quadrant requires executive attention because the organization may be experiencing fewer events while still failing at high-consequence controls.

Step 5: Compare severity rate with DART and TRIR

Severity rate should be compared with DART and TRIR because each metric shows a different layer of harm. TRIR counts recordable cases, DART narrows the view to days away, restriction, or transfer, and severity rate adds weight by showing how many days the case removed from normal work capacity.

The comparison prevents a common dashboard mistake. A lower TRIR can coexist with a worse severity rate, and a stable DART can hide longer restrictions. That is why a manager reviewing only DART rate pitfalls still needs a severity view to ask whether the same type of case is getting heavier.

Use 3 monthly questions. Did recordable frequency rise or fall? Did DART cases become longer or shorter? Did severity rate concentrate in one department, task, or supervisor group? The answers point the review toward work design rather than toward a generic request for more awareness.

Step 6: Audit the outliers before averaging

Outliers can dominate severity rate because one long-duration case can outweigh many minor cases. A single back injury with 96 restricted days, for example, can create a larger severity signal than 6 short cases combined, even when the monthly incident count looks moderate.

This is where mature leadership resists the temptation to dismiss the case as an exception. During the PepsiCo South America tenure, where the accident ratio fell 50% in 6 months, Andreza Araújo learned that sustainable improvement depends on treating uncomfortable data as a source of learning, not as a public-relations problem.

Review the top 3 cases by counted days before presenting the average. For each, document the task, energy source, body part, supervisor group, medical restriction, control failure, and corrective-action status. If the same job family appears twice, the outlier is probably not an outlier. It is a pattern with a small sample size.

Step 7: Link every severe case to corrective actions

A severity-rate review is incomplete until every high-day case connects to a corrective action with an owner, due date, verification method, and effectiveness check. A worker can return to normal duty after 21 days while the original exposure remains open for another 60 days, which means the metric improved before the risk changed.

In more than 250 cultural-transformation projects supported by Andreza Araújo's team, a recurring weakness is the split between medical closure and operational closure. The clinic can close the file, HR can close the absence code, and operations can still run the same task with the same awkward lift, pinch point, or staffing pressure.

Connect severity cases to corrective-action closure metrics. The dashboard should show whether the exposure was eliminated, engineered down, administratively controlled, or merely accepted until the next worker absorbs the cost.

Step 8: Segment by department, shift, and task

Segmentation shows whether severity is spread across the site or concentrated in a few work systems. A plant-wide severity rate can look acceptable while night-shift maintenance, manual handling in dispatch, or contractor shutdown work carries most of the lost-capacity days.

BLS maintains injury, illness, and fatality data with industry incidence-rate views, which is useful for broad benchmarking. Internal prevention, however, needs a sharper lens than industry averages. The EHS manager should know the 3 tasks producing the heaviest lost-capacity burden, not only whether the site sits above or below a sector rate.

Segment severity rate by department, shift, task, contractor status, tenure band, and injury type. Then compare those cuts with observation quality and near-miss quality. If manual handling produces high severity but low near-miss volume, the issue may be reporting climate as much as biomechanics.

Step 9: Present the rate without creating underreporting

Severity rate should be presented as a learning signal, not as a scorecard that punishes teams for reporting injuries honestly. When leaders shame a department for a high severity month, they may reduce the next month's number by teaching people to hide pain, delay treatment, or avoid formal reporting.

ILO states that safe and healthy working conditions are fundamental to decent work, and the ILO safety and health at work program frames prevention as a system responsibility. That framing matters because severity-rate governance should protect the worker's voice as much as it protects data accuracy.

Open the review with 3 statements. Reporting is protected. The purpose is to understand exposure. Leaders will be evaluated on verified control improvement, not on hiding the red. Those sentences lower fear and make the metric more reliable.

Comparison: Severity rate as arithmetic versus management signal

Severity rate becomes useful when the calculation is treated as the beginning of control review. The table below separates a narrow arithmetic use from the stronger management use that exposes work-design failures.

Every month that severity rate is reviewed only as arithmetic allows the organization to miss the task, shift, or control failure that is already converting injury into lost capacity.

Conclusion: Calculate severity rate, then challenge it

Severity rate is worth calculating because it reveals the weight behind injury and illness cases, but it becomes valuable only when leaders challenge the numerator, denominator, segmentation, and corrective-action trail. A 30-minute calculation can start the review, although the real work begins when the number points to a job that must change.

If your EHS dashboard needs to separate clean recordkeeping from real safety, Andreza Araújo's safety culture diagnostics and ACS Global Ventures consulting can help connect severity rate with field verification, leadership rhythm, and control effectiveness. Start at Andreza Araújo and bring the same discipline to indicators that you expect from high-risk work.