Leading vs. Lagging Safety Indicators: What to Track to Prevent the Next Injury

Most safety programs are measured by a number that can only go up after someone gets hurt. TRIR, DART, and lost-time rates are useful, but they all count injuries that have already happened. Managing safety with them is like driving with only the rearview mirror.

Leading indicators are the windshield. They measure the activities and conditions that come before injuries, so you can act while there's still time to prevent one. Here's how the two types differ, which leading indicators are worth tracking, and how to keep them from turning into busywork.

Lagging indicators: what already happened

Lagging indicators measure outcomes. The common ones are:

• Total Recordable Incident Rate (TRIR)
• DART and days-away rates
• Workers' compensation claims and costs
• Fatalities and serious injuries
• Lost workdays

They're standardized, easy to compare between sites and industries, and they're what customers, insurers, and regulators ask for, so you need them. Two weaknesses limit their usefulness for prevention, though. They react only after harm is done, and they're statistically noisy at small facilities, where one injury can swing a year's rate. A low TRIR can mean a site is safe, or it can mean the site has been lucky.

(If you need a refresher on the math, see how to calculate TRIR and DART.)

Leading indicators: what's about to happen

Leading indicators measure the inputs to safety: hazards being found, risks being controlled, and people taking part. OSHA's guide, Using Leading Indicators to Improve Safety and Health Outcomes, encourages employers to track them alongside injury rates.

Examples worth considering:

• Near-miss and hazard reports submitted. At first, more is usually better. A rising count usually means people trust the system, not that the site got more dangerous.
• Corrective action closure time. How long does it take from a hazard being reported to the fix being verified?
• Overdue corrective actions. Hazards that were found but never fixed are among the clearest warning signs you can measure.
• Safety observations and inspections completed against plan.
• Training completion, especially for high-risk tasks like lockout/tagout, forklift operation, and confined space entry.
• Toolbox talks and safety meetings held, and how many workers attended.
• Repeat findings: the same hazard showing up in the same area or task more than once.

Traps to avoid

Counting activity instead of effect. "We held 52 toolbox talks" sounds good, but 52 talks that nobody listened to prevent nothing. Pair each activity count with a measure of what changed, such as hazards closed or repeat findings reduced.

Punishing the reporting you want. If a spike in near-miss reports draws criticism, reports will dry up, and you lose your best early-warning signal. Treat reporting volume as a sign of engagement, not a problem.

Tracking too much. Five indicators that someone reviews every week beat thirty that nobody reads. Start with near-miss reports, corrective action closure, and repeat findings, then add more as the habit sticks.

Leaving the data in silos. Near-misses sit in one spreadsheet, inspections in another, injuries in a third system. The pattern that predicts your next injury often only becomes visible when you look at all three together.

Connecting leading to lagging

Leading indicators work best when you can see how they connect to outcomes: the dock with the most forklift near-misses, the press line with overdue guarding fixes, the shift with the most first aid cases. Those links are what turn a dashboard into a prevention plan. They're also hard to find by hand once you have more than a few dozen reports.

That's the problem Infranoto is built for. It keeps incidents, near-misses, observations, and corrective actions in one place, and uses AI to find recurring hazards and connect them to the outcomes they predict. If you'd like to see what your leading indicators are already telling you, request a demo.

Getting started this week

You don't need new software to start. Pick one leading indicator and one action:

1. Start counting near-miss reports per month, and make it easy and blame-free to submit one. Our free near-miss reporting toolbox talk is a five-minute way to introduce it to crews.
2. List every open corrective action with an owner and a due date, then review that list weekly.
3. Keep tracking your lagging rates for benchmarking. The free Incident Rate Calculator compares them against your industry.

The takeaway

• Lagging indicators (TRIR, DART, claims) measure harm that has already happened. You need them for benchmarking, but they're too slow to prevent anything.
• Leading indicators (near-misses, corrective action closure, observations, training) measure the conditions that come before injuries.
• Measure effect, not just activity, and never punish the reporting you're trying to encourage.
• The value comes from connecting them. Recurring hazards in your leading data usually point to where your next lagging number will come from.