A job hazard analysis (JHA) breaks one task into its steps and lists the hazards and controls for each step. The foreman or competent person usually writes it with the crew at the work area before work starts. Once the crew signs it, the JHA is also a record with a date, a project, a task, the crew’s names and a risk score for every step. Count those records across your jobs each week and you have a safety measure you can read before anyone is hurt.
Leading and lagging indicators
OSHA’s guide to leading indicators defines lagging indicators as counts of events that already happened, such as the number or rate of injuries, illnesses and fatalities. Leading indicators measure the events that lead up to an incident, so a contractor can act on a failing control or an open hazard before it turns into one.
The guide lists suggested indicators by program area. For worker participation, one of them is the number of workers involved in developing task-specific job hazard analyses. For hazard identification, it suggests the percentage of daily, weekly or monthly inspections completed, and the percentage of hazards abated in the same day, week or month they were found. OSHA also asks that each indicator be specific, measurable, accountable, reasonable and timely, and that it be stated as a number, rate or percentage you can follow over time.
The lagging numbers for construction
The Bureau of Labor Statistics (BLS) counted 1,034 fatal occupational injuries in private construction in 2024, out of 5,070 for all US workers. The 2024 counts are final. Falls, slips and trips caused 389 of the construction deaths, more than any other event. Specialty trade contractors accounted for 606 deaths. Roofing contractors accounted for 120, and 96 of those were falls.
| Event or exposure | Private construction | Specialty trade contractors | Roofing contractors |
|---|---|---|---|
| Falls, slips, trips | 389 | 253 | 96 |
| Transportation incidents | 244 | 134 | 5 |
| Exposure to harmful substances, environments | 187 | 106 | 13 |
| Contact incidents | 161 | 82 | 2 |
| Violent acts | 46 | 27 | |
| Explosions and fires | 5 | ||
| All events | 1,034 | 606 | 120 |
How a JHA is scored
Free Safety Report, the free JHA site that ResponseVault runs, scores each job step as likelihood times consequence, each on a scale of 1 to 5. A likelihood of 1 means nobody has seen it happen on comparable work, and 5 means it will probably happen today without controls. A consequence of 1 is an injury that needs no first aid, and 5 is a fatality or permanent disability. The score, from 1 to 25, falls into five bands:
- 1 to 2, Negligible. Proceed with normal supervision.
- 3 to 4, Low. The crew reviews the controls on the JHA.
- 5 to 9, Moderate. Proceed only with the listed controls in place, confirmed by the supervisor.
- 10 to 16, High. Do not start until added controls bring the score down, with supervisor sign-off.
- 20 to 25, Extreme. Stop and redesign the task.
Each step is scored twice, first as if no controls were in place and then with the controls the crew will use. The second score is the residual risk, and it decides whether work starts. Free Safety Report says these bands are its own convention and that many companies use different ones, so use your company’s matrix if it has one.
The site’s worked example is a crew setting beam cages and tying a slab mat on an elevated deck. Four of the six steps start in the High band. With controls such as perimeter guardrail, opening covers, tag lines and impalement caps, none of the six is High.
| Job step | Before controls | Band before | After controls | Band after |
|---|---|---|---|---|
| Fly beam cages to the deck | 15 | High | 5 | Moderate |
| Set cages in beam forms at the slab edge | 15 | High | 5 | Moderate |
| Carry and spread slab bars | 12 | High | 6 | Moderate |
| Tie mat intersections | 8 | Moderate | 4 | Low |
| Work near column dowels | 10 | High | 3 | Low |
| Walk on the finished mat | 8 | Moderate | 4 | Low |
Five measures from JHA records
Each measure below needs only the JHA records and the daily log for the same days.
- JHA coverage. JHAs submitted divided by crew tasks started, with the count of crew tasks taken from the daily log. A task that starts without a JHA lowers the rate.
- Worker participation. Crew signatures on JHAs divided by the workers the daily log puts on those crews. It turns OSHA’s suggested count of workers involved in JHAs into a rate.
- Steps still High after controls. The count of job steps on signed JHAs whose residual score is 10 or more. Under Free Safety Report’s bands this count should be zero, because a High step does not start.
- High steps before controls, by task type. OSHA’s JHA booklet, Publication 3071, gives priority to jobs with the highest injury rates and to jobs that could cause severe injuries even with no accident history. This share shows which of your tasks those are.
- Revisions after work started. OSHA 3071 recommends reviewing a JHA after any injury or illness on that job and training every affected employee when you revise it. Free Safety Report has the crew initial each change, so count revisions that carry no new initials.
A worked example
The figures below are an example with round numbers. A general contractor with self-performing concrete and framing crews tracks JHA coverage and participation across three jobs for six weeks. The superintendents count crew tasks and crew members from the daily logs, and the safety manager counts JHAs and signatures.
| Week | Crew tasks | JHAs | Coverage | Crew members | Signatures | Participation | Steps still High |
|---|---|---|---|---|---|---|---|
| 1 | 40 | 28 | 70.0% | 180 | 117 | 65.0% | 2 |
| 2 | 42 | 31 | 73.8% | 190 | 133 | 70.0% | 1 |
| 3 | 38 | 30 | 78.9% | 170 | 126 | 74.1% | 1 |
| 4 | 45 | 38 | 84.4% | 200 | 160 | 80.0% | 0 |
| 5 | 44 | 39 | 88.6% | 195 | 166 | 85.1% | 0 |
| 6 | 40 | 37 | 92.5% | 180 | 162 | 90.0% | 0 |
| Six weeks | 249 | 203 | 81.5% | 1,115 | 864 | 77.5% | 4 |
Over the six weeks the crews started 249 tasks and submitted 203 JHAs, for coverage of 81.5%. Participation was 77.5%, with 864 signatures from 1,115 crew members on those tasks. Both rates rose every week. In the first three weeks, four signed JHAs carried a step that was still High after controls, which under the matrix means work on that step should not have started. The safety manager can review each of those four JHAs with the foreman who signed it.
With JHA responses in one CSV and crew tasks from the daily log in a second sheet, coverage in Excel is a COUNTIFS on each sheet by project and week, with one result divided by the other. A coverage rate above 100% means crews filed JHAs for tasks the daily log missed.
Where the records live
A paper JHA gets counted only when someone types it in. The Free Safety Report builder needs no account, prints a PDF for the crew to sign, and saves the JHA in the browser of the device that made it. The JHA leaves that device only if someone emails the PDF or copies a share link. That works for one crew’s JHA on one day. A weekly count across jobs needs every submission in one place.
ResponseVault, the sponsor of this post, sells software for that step. It turns a paper form into a digital form that crews fill out on a phone or tablet. These claims come from ResponseVault:
- Dropdowns pull projects and vendors from Procore, and subcontractors can submit a form from a link without an account.
- Each submission can update Procore and email a PDF.
- All responses for a form sit in one grid that exports to CSV.
- Webhooks fire when a response is created, updated or deleted, and a REST API returns responses with filters for dates.
- A project database combines form submissions with data from Procore and other systems for reporting in Excel or Power BI.
ResponseVault says the average time to deployment is four weeks, from an onboarding questionnaire to the form in use on site. The app is listed on the Procore App Marketplace as Job Site Forms by ResponseVault. The listing shows no price and directs buyers to contact ResponseVault.
How to start on your jobs
Pick one task type that scores High before controls, such as deck edge work or roofing, on one active job. For two weeks, have the foreman write a JHA before each crew task, and have the superintendent count crew tasks and crew members from the daily log. At the end of each week, compute coverage and participation, and list every signed JHA with a step still scoring 10 or more after controls. Report those figures next to your recordable injury rate at the monthly safety meeting.