Safety Performance Metrics and Rates

Why Safety Performance Metrics and Rates Matter

Safety professionals are constantly asked a deceptively simple question: Is our safety program working? Metrics and rates are how you answer it with evidence rather than opinion. They allow you to compare one facility to another, one year to the last, and your organization to its industry peers. They drive insurance costs through the Experience Modification Rate (EMR), influence whether contractors are prequalified for bids, satisfy OSHA recordkeeping requirements under 29 CFR 1904, and give management the data needed to allocate resources where risk is highest.

For the BCSP Certified Safety Professional (CSP) exam, this topic sits squarely within the Risk Management and Program Management domains. You can expect both calculation questions (compute a TRIR, DART rate, or severity rate from raw data) and conceptual questions (distinguish leading from lagging indicators, interpret an EMR, identify the limitations of incidence rates). Because these questions are formula-driven and highly predictable, they represent some of the most reliable points you can earn on exam day if you prepare properly.

What Safety Performance Metrics Are

A safety performance metric is any measurable quantity used to evaluate the effectiveness of a safety and health management system. Metrics are generally sorted into two families:

1. Lagging indicators measure outcomes that have already occurred. They tell you what went wrong after the fact. Examples include injury and illness incidence rates, DART rates, severity rates, fatality rates, workers' compensation costs, and the EMR. Lagging indicators are easy to count, widely benchmarked, and required for regulatory reporting, but they are reactive and can be misleading in small populations or over short time periods.

2. Leading indicators measure activities and conditions that predict or prevent future outcomes. Examples include percentage of scheduled inspections completed, near-miss reporting rates, training completion rates, time to close corrective actions, number of safety observations or behavior-based safety contacts, job hazard analyses completed versus planned, management safety walk-throughs, and safety perception or climate survey scores. Leading indicators are proactive, but they must be validated to ensure they actually correlate with reduced injuries rather than simply measuring activity for its own sake.

A mature program uses a balanced scorecard of both types. ANSI/ASSP Z10 and ISO 45001 both emphasize monitoring, measurement, analysis, and performance evaluation using a mix of indicators tied to the organization's objectives.

The Core OSHA Incidence Rate Formula

The foundation of nearly every rate question is the OSHA incidence rate formula:

Incidence Rate = (Number of Cases × 200,000) ÷ Total Hours Worked by All Employees

The constant 200,000 represents the hours worked by 100 full-time employees in one year: 100 workers × 40 hours per week × 50 weeks (allowing two weeks of vacation). Using this base normalizes the result so that any rate can be read as cases per 100 full-time workers per year, which allows apples-to-apples comparison between a 50-person shop and a 5,000-person plant.

Important rules for the hours denominator:
• Count only hours actually worked. Exclude vacation, sick leave, holidays, and other paid non-working time.
• Include hours for all employees on the payroll, including temporary workers you supervise day to day.
• For salaried employees whose hours are not tracked, estimate using a reasonable assumption (commonly 2,000 hours per year per full-time employee).

The Major Lagging Rates You Must Know

Total Recordable Incident Rate (TRIR), also called Total Case Incidence Rate (TCIR)
TRIR = (Total OSHA recordable cases × 200,000) ÷ Hours worked
This counts every case recorded on the OSHA 300 Log: fatalities, days-away cases, restricted or transferred cases, and other recordable cases (medical treatment beyond first aid, loss of consciousness, physician-diagnosed significant injury or illness, and special cases such as needlesticks, medical removal, standard threshold shifts in hearing, and work-related tuberculosis).

DART Rate (Days Away, Restricted, or Transferred)
DART = (Number of cases involving days away from work, restricted duty, or job transfer × 200,000) ÷ Hours worked
Note that DART counts cases, not days. A case with 40 days away counts as one case. DART replaced the older Lost Workday Case Incidence Rate (LWDII) terminology.

Days Away From Work Case Rate (DAFWII)
DAFWII = (Cases with at least one day away from work × 200,000) ÷ Hours worked
This is a subset of DART; it excludes cases involving only restriction or transfer.

Severity Rate (Lost Workday Rate)
Severity Rate = (Total number of days away, restricted, or transferred × 200,000) ÷ Hours worked
This measures the seriousness of incidents, not just their frequency. Under OSHA recordkeeping rules, days are counted as calendar days (including weekends and holidays), the day of injury is not counted, and the count for any single case is capped at 180 days for days away and another 180 days for restricted or transferred days.

ANSI Z16.1 Rates (Historical, but Still Tested)
The older ANSI Z16.1 standard used a base of 1,000,000 hours instead of 200,000.
Frequency Rate = (Number of disabling injuries × 1,000,000) ÷ Hours worked
Severity Rate = (Total days charged × 1,000,000) ÷ Hours worked
Z16.1 also assigned scheduled day charges for permanent injuries (for example, 6,000 days charged for a fatality or permanent total disability). The standard has been withdrawn, but exam questions sometimes specify the 1,000,000 base, so always read which constant the question wants.

Average Days Charged per Case (Severity-Frequency Index)
Average Days per Case = Severity Rate ÷ Frequency Rate, or simply Total days lost ÷ Number of lost-time cases. This tells you how severe the typical case is.

Fatality Rate
BLS reports fatal occupational injuries per 100,000 full-time equivalent workers, calculated as (Fatalities × 200,000,000) ÷ Hours worked. Be careful: the base is different from the incidence rate base.

Experience Modification Rate (EMR)
The EMR is an insurance metric, not an OSHA metric. It compares an employer's actual workers' compensation losses to the expected losses for employers of similar size in the same classification.
EMR = Actual losses ÷ Expected losses (with actuarial weighting and ballast applied)
An EMR of 1.0 is average. Below 1.0 means better than expected (lower premiums); above 1.0 means worse than expected (higher premiums). The EMR typically uses three years of loss data, excluding the most recent policy year, so it lags current performance. It also weighs frequency more heavily than severity through the split-point mechanism (primary versus excess losses), which is why many small claims can hurt an EMR more than one large claim. Contractors are often required to have an EMR below 1.0 to bid on work.

Benchmarking Against Industry Data

The Bureau of Labor Statistics (BLS) publishes incidence rates annually by NAICS industry code and establishment size through the Survey of Occupational Injuries and Illnesses (SOII). Comparing your TRIR and DART to the BLS average for your NAICS code is the standard way to determine whether performance is above or below industry norms. OSHA also uses these rates to target inspections through programs such as Site-Specific Targeting, and establishments with high DART rates relative to their industry are more likely to be selected.

Limitations and Statistical Considerations

The CSP exam expects you to understand that rates can mislead:
• Small denominators are volatile. A 20-person company with 40,000 hours and one recordable has a TRIR of 5.0. One more case doubles it to 10.0. Small changes in case counts create wild swings that are statistical noise, not real trends.
• Injuries are rare events that follow roughly a Poisson distribution. Judging a trend requires enough data (hours and cases) to be statistically meaningful. Control charts (u-charts or c-charts) can help distinguish special-cause variation from random variation.
• Lagging rates can be manipulated through underreporting, aggressive return-to-work that converts days-away cases to restricted cases, or incentive programs that discourage reporting. OSHA considers rate-based incentive programs that deter reporting to be potentially retaliatory under 1904.35.
• A low TRIR does not mean low risk of a catastrophe. Process safety incidents and fatalities are often decoupled from personal injury rates. This is why serious injury and fatality (SIF) and potential-SIF metrics have become important, along with process safety metrics such as API RP 754 Tier 1 through Tier 4 indicators.
• Rates normalize by hours, not by exposure. Two departments with equal hours may have vastly different hazard exposures.

Worked Calculation Examples

Example 1 - TRIR: A plant worked 1,250,000 hours and recorded 15 recordable cases. TRIR = (15 × 200,000) ÷ 1,250,000 = 3,000,000 ÷ 1,250,000 = 2.4.

Example 2 - DART: Of those 15 cases, 4 involved days away and 2 involved restricted duty only. DART = (6 × 200,000) ÷ 1,250,000 = 1,200,000 ÷ 1,250,000 = 0.96.

Example 3 - Severity Rate: The 6 DART cases produced a total of 95 days away and 40 restricted days (135 total days). Severity Rate = (135 × 200,000) ÷ 1,250,000 = 27,000,000 ÷ 1,250,000 = 21.6 days per 100 full-time workers.

Example 4 - Reverse calculation: A company has a TRIR of 3.0 and 600,000 hours. How many recordables occurred? Rearranging: Cases = (TRIR × Hours) ÷ 200,000 = (3.0 × 600,000) ÷ 200,000 = 9 cases.

Example 5 - Finding hours from headcount: A firm employs 150 full-time workers for a year. Estimated hours = 150 × 2,000 = 300,000 hours. With 3 recordables, TRIR = (3 × 200,000) ÷ 300,000 = 2.0.

Example 6 - ANSI base: If a question asks for a frequency rate using a 1,000,000-hour base and there were 5 lost-time injuries in 800,000 hours: (5 × 1,000,000) ÷ 800,000 = 6.25.

Exam Tips: Answering Questions on Safety Performance Metrics and Rates

Tip 1 - Identify exactly which rate is being asked. TRIR counts all recordables. DART counts only days-away, restricted, or transferred cases. Severity counts days, not cases. Misreading this is the single most common error. Underline the key word in the question stem before you calculate.

Tip 2 - Use 200,000 unless told otherwise. Default to the OSHA base. If the question explicitly mentions ANSI Z16.1, a 1,000,000-hour base, or fatalities per 100,000 workers, switch constants. The answer choices will often include a distractor that is exactly five times too big or too small, which is what you get by using the wrong base.

Tip 3 - Count cases, not injured people or body parts. One employee injured twice in a year is two cases. One incident that injures three workers is three cases. One case involving both days away and restricted duty is still one DART case.

Tip 4 - Scrub the hours. If the question gives total paid hours and separately lists vacation or holiday hours, subtract them. If it gives headcount only, multiply by 2,000 hours per full-time employee (or 40 hours × 50 weeks) unless a different figure is stated.

Tip 5 - Watch the 180-day cap and calendar-day rule when calculating severity. If a case lists 240 days away, OSHA counts only 180. Weekends count. The day of injury does not.

Tip 6 - Sanity-check the magnitude. Typical TRIRs fall between roughly 1 and 6 for most industries. If you compute 0.03 or 300, you almost certainly inverted the fraction or dropped a zero. Work the arithmetic twice.

Tip 7 - Be ready for reverse and comparative problems. You may be given the rate and asked for hours or cases, or asked which of two facilities performed better. Rearrange the formula algebraically and compare normalized rates, never raw counts.

Tip 8 - Know EMR direction and timing. Lower is better, 1.0 is the industry average, and it reflects three prior years excluding the most recent. If a question asks why a company with a great current year still has a high EMR, the answer is the lag built into the formula.

Tip 9 - Classify indicators correctly. If a choice describes something that has already happened (an injury, a claim, a lost day), it is lagging. If it describes an action, condition, or behavior intended to prevent harm (training, inspections, near-miss reports, corrective action closure), it is leading. Near-miss reporting is a classic leading indicator even though a near miss is an event, because its purpose is preventive and a rising near-miss report rate is usually viewed as positive.

Tip 10 - For conceptual questions, think like a statistician. When asked about the weakness of comparing monthly rates at a small site, the answer centers on small-sample variability and the rarity of events. When asked for the best way to evaluate whether a change is real, think control charts, longer time frames, or pooling data.

Tip 11 - Link metrics to management systems. Questions framed around ANSI/ASSP Z10 or ISO 45001 will reward answers that tie metrics to objectives, management review, and continual improvement rather than to rate reduction alone.

Tip 12 - Manage your time. Rate calculations are quick wins. Set up the formula, plug in the numbers, compute, check the magnitude, and move on. Do not let a single calculation consume five minutes.

Quick Reference Summary

• Incidence Rate = (Cases × 200,000) ÷ Hours
• TRIR counts all OSHA recordables
• DART counts cases with days away, restriction, or transfer
• Severity Rate = (Total days × 200,000) ÷ Hours; cap 180 days per case; calendar days
• ANSI Z16.1 uses 1,000,000 hours
• Fatality rate uses 100,000 FTE workers (200,000,000 hours)
• 200,000 = 100 workers × 40 hours × 50 weeks
• EMR: 1.0 average, lower is better, three-year lag
• Lagging = outcomes; Leading = preventive activities and conditions
• Small denominators create volatile rates; use statistics, not single data points

Master these formulas and concepts, practice a dozen varied calculations until the steps are automatic, and you will convert this exam topic into dependable points while also building a skill you will use throughout your safety career.