Construction is one of the most hazardous industries, and many of the practices that make a site safer, such as planning the work, clearing constraints, and giving each crew its own space, also make it more productive and reduce rework.
This guide covers pre-task planning, stop-work authority, near-miss reporting, and the difference between leading and lagging indicators, including how the total recordable incident rate is calculated. A worked example shows how to read rising near-miss reports and falling injuries, and why to be careful about it. Requirements vary by jurisdiction; follow those that apply to your site.
Before You Start
Why Safety and Quality Travel Together
Rushed Work Is Both Unsafe and Defective
The pressure that produces shortcuts also produces errors. A site that plans well is usually safer and has fewer defects.
Injuries Are Lagging Indicators
Counting injuries tells you what already went wrong. Leading indicators show whether the safeguards are working now.
Workers See the Hazards First
Crews know where a task is unsafe. A site that welcomes reports learns about hazards before someone is hurt.
Planning Prevents Exposure
Sequencing, zoning, and clearing constraints reduce crowding, awkward workarounds, and improvisation.
Planning Safety Into the Work
| Practice | What it does | Quality link |
|---|---|---|
| Pre-task planning or job hazard analysis (JHA) | Crew and foreman walk through the task, list hazards, and agree controls before starting | Also confirms materials, tools, and sequence |
| Toolbox talks | Short, regular briefings on the day's hazards and lessons | Shares quality lessons too |
| Zoning and takt planning | Gives each trade its own space, avoiding stacking | See Takt Planning |
| Stop-work authority | Anyone can stop unsafe work without penalty | Applies equally to quality problems |
| Near-miss reporting | Records events that could have caused harm | Learning input for root cause analysis |
| Layered observation | Leaders observe work on a schedule | Like a layered process audit |
Leading and Lagging Measures
| Type | Measure | Note |
|---|---|---|
| Lagging | Total recordable incident rate (TRIR) = recordable incidents × 200,000 / hours worked | In the U.S., OSHA recordkeeping (29 CFR 1904) defines recordable cases. 200,000 hours is 100 full-time workers for a year. |
| Lagging | Days away, restricted, or transferred (DART) rate | Same formula with only cases involving days away, restricted work, or transfer. |
| Leading | Percent of tasks with a completed JHA before work | Measures planning discipline. |
| Leading | Near-miss reports per month, and the share investigated | A rise often reflects a healthier reporting culture. |
| Leading | Corrective actions closed on time | Shows whether hazards are actually fixed. |
Worked Example: Six Months on a Site
A project logs 350,000 worked hours over six months with 4 recordable injuries. The figures are illustrative.
- TRIR = 4 × 200,000 / 350,000 = 2.29.
- JHA completion was 92%, meaning 460 of 500 sampled tasks had a JHA completed before work began.
- Near-miss reports rose from 6 in month 1 to 18 in month 6, and the team investigated 15 of the 18 in month 6.
How to read it. Rising near-miss reports are not a sign that the site is getting more dangerous. They usually mean workers trust the process and are speaking up, which lets the team act on hazards before they injure someone. In this case, the team's actions included re-routing a crane path and adding edge protection after reports from two crews.
A caution. The decline in injuries coincides with better reporting, but the data cannot prove one caused the other. Injury counts on one site over six months are small numbers and can swing by chance. Do not congratulate the team on a quiet quarter or reward a low injury count in a way that discourages reporting, since that is a well-known way to hide injuries.
The Hierarchy of Controls on a Construction Site
The hierarchy of controls ranks ways of reducing risk from the least to the most reliable. The best protection removes the hazard; the weakest depends on each worker doing the right thing every time.
Examples. To reduce the risk of falls from an edge: eliminate the exposure by prefabricating at ground level; use guardrails or a safety net as engineering controls; use a harness and lanyard as PPE, with a trained rescue plan. To reduce the exposure to dust from cutting: substitute a method that produces less dust, use water suppression or local exhaust, and then use respiratory protection as the last layer.
Plan it in. Many controls, such as prefabrication, crane lifts, scaffolding access, and edge protection, must be designed into the plan and schedule. Identifying them during planning is cheaper than adding them after a near-miss.
Use more than one layer. Combining controls gives redundancy. If a worker removes a guardrail to pass material, a procedure and a spotter may still protect them.
Pre-Task Planning and Learning From Near Misses
Pre-task planning is a short, structured conversation before work begins. It gives the crew a shared picture of the task, the hazards, and the controls. It is most useful when it is specific to the day's work and the conditions.
- Describe the task and the area, the crews nearby, and the equipment in use.
- Identify hazards for each step, including those created by other trades or by changing conditions.
- Agree the controls, using the hierarchy, and confirm that equipment and protection are in place.
- Confirm authority to stop work. Everyone should know that they can, and should, stop if conditions change or something is unclear.
- Sign off and revisit the plan if the work or conditions change.
Learn from near misses. A near miss is an event that could have caused harm but did not. Encourage reporting without blame, review reports quickly, and feed the lessons to crews and into planning. A site where near misses are common and reported is usually safer than one with few reports, because people have not stopped talking about risk.
Measure leading indicators. Count completed inspections, corrective actions closed on time, pre-task plans done, near misses reported, and safety observations. Lagging measures such as injury rates show outcomes but come too late to prevent them.
Pitfalls. Generic, copied pre-task forms; rewarding a low injury count in a way that discourages reporting; pressure from the schedule that overrides stop-work authority; and poor coordination between trades sharing a space. See the Last Planner Guide and the Takt Planning Guide. This guide is educational and is not safety, engineering, or legal advice. Follow your site's safety program and the regulations that apply to you. Figures in the examples are illustrative.
Self-Assessment Questions
- Do crews complete a hazard analysis before starting each task?
- Can anyone stop unsafe work without fear of penalty?
- Do we track leading indicators as well as injury rates?
- Do we investigate near misses and close the actions?
- Do incentives encourage reporting, not hiding, problems?
Common Mistakes
Rewarding a Low Injury Count
Bonuses tied to injury numbers push people to under-report. Reward leading behaviors like reporting and planning.
Paperwork Hazard Analyses
A JHA filled in after the work is finished protects nobody. Do it with the crew, before the task.
Ignoring Near Misses
Each near miss is a free lesson. If reports go nowhere, they stop.
Treating Safety and Quality Separately
The same planning, constraint removal, and root cause discipline improves both.
Construction Safety and Planning: Frequently Asked Questions
How is the total recordable incident rate (TRIR) calculated?
TRIR equals the number of recordable incidents multiplied by 200,000, divided by the total hours worked. The factor of 200,000 represents 100 full-time employees working 40 hours a week for 50 weeks. For example, 4 recordables over 350,000 hours gives a TRIR of about 2.29. Definitions of recordable cases depend on the applicable recordkeeping rules.
What is a leading indicator in construction safety?
A leading indicator measures safety activities and conditions that prevent injuries, such as the share of tasks with a completed pre-task plan, near-miss reports and how many are investigated, and corrective actions closed on time. Unlike injury rates, which look backward, leading indicators show whether the safeguards are working now.
Why might near-miss reports increase when a site gets safer?
Rising near-miss reports often reflect a stronger reporting culture in which workers trust that speaking up leads to action. That gives the team information to fix hazards before injuries occur. Tracking how many reports are investigated and closed shows whether the increase is being turned into improvement.
Sources and Further Reading
- OSHA, Recording and Reporting Occupational Injuries and Illnesses, 29 CFR Part 1904, and the OSHA construction standards, 29 CFR Part 1926.
- Construction Industry Institute, research on safety planning and leading indicators.
- Sidney Dekker, The Field Guide to Understanding 'Human Error'.
- Lean Construction Institute, resources on safety and last planner planning.