Leading vs lagging indicators in construction safety

Lagging indicators: what already happened

These are outcomes. They are important because they show real harm, but they arrive after the damage and a small company may go years without a serious event, which tells you little about whether it is safe.

• Recordable injuries and illnesses
• Days away, restricted or transferred cases
• Fatalities
• Lost workdays and workers' compensation costs
• Incident rates built from the counts above

Leading indicators: what is being done

These are inputs and activities, and they can be improved before anyone is hurt.

• Inspections completed, and the share of findings closed on time
• Near misses reported and investigated
• Job hazard analyses completed before high-risk tasks
• Training sessions delivered and attendance
• Safety observations and the share that were safe behavior
• Time to fix a reported hazard

A strong leading indicator is one the team can control and that has a believable link to preventing harm. Counting how many toolbox talks happened is easy. Measuring whether hazards reported at the talk were fixed is better.

The rate math, with invented numbers

Recordkeeping rates scale the case count to the hours worked, so a large company and a small one can be compared. The standard base is 200,000 hours, which is 100 full-time workers for a year. The formula is the number of cases times 200,000, divided by total hours worked.

Say a contractor logs 4 recordable cases over 400,000 hours worked.

• 4 × 200,000 = 800,000
• 800,000 ÷ 400,000 = 2.0

That is a total recordable incident rate of 2.0. The DART rate uses the same arithmetic with only cases involving days away, restricted work or transfer. Your own recordability decisions follow 29 CFR Part 1904, covered in the recordkeeping guide. Check your hand work with the incident rate calculator, and see the worked formulas in the CHST formulas guide.

Why one number is never enough

A low injury rate can come from good practice or from luck, or from workers not reporting. A rate built on few hours swings widely with a single case. That is why leading indicators matter: they show whether the system is working before the next incident answers the question the hard way.

Be careful with indicators that can be gamed. If supervisors are scored on a low injury count, some will under-report. Pairing a lagging number with leading ones, and rewarding reporting of near misses, reduces that risk.

How the exam uses this

Expect questions like these:

• Classify it. A list of metrics, and you pick the leading one. Inspections completed, yes. Recordable cases, no.
• Pick the weakness. A company points to a low injury rate as proof of safety. The answer notes that lagging data alone is slow and can hide risk.
• Do the math. A case count and hours, and you compute a rate. Keep the 200,000 base in mind and watch for the unit trap: rates use hours, not workers.
• Choose a better metric. The strongest answer measures an activity linked to prevention, not a number that punishes reporting.

Pick your three

If you are building a program, start small. Choose one lagging indicator, such as your recordable rate, and two leading ones that tie to your top hazards, such as the percent of planned inspections completed and the average days to close a finding. Review them monthly with the crew leads, not only with managers.

Where to practice

Work the rate problems until they take under a minute, then use the CHST study plan to place Domain 3 in your calendar. Take the free 10-question check to see how this domain scores for you, with every answer explained and no signup.

FAQ

Is a near miss leading or lagging?

Near-miss reporting is usually treated as leading, because it gives information before anyone is harmed.

Why does the rate use 200,000 hours?

It stands for 100 full-time workers for a year, which lets you compare employers of different sizes.

Can a leading indicator be misleading?

Yes. Activity counts such as meetings held can look good while hazards stay unfixed, so pair them with measures of what was corrected.