1.1 Why falls still dominate construction deaths

1 The numbers, and where they come from

Two federal data sets carry this subject. The Census of Fatal Occupational Injuries (CFOI) counts every work-related death and is close to a true census. The Survey of Occupational Injuries and Illnesses (SOII) estimates nonfatal cases with days away from work from employer logs, and since 2021 it publishes in two-year periods. CPWR — the Center for Construction Research and Training — pulls both apart for construction. The figures in this chapter come from its March 2024 bulletin on falls and its June 2026 bulletin on injury trends by subsector.

The trade breakdown for 2022 is blunt. Roofing contractors alone accounted for 100 fatal falls, slips and trips; residential building construction for 63; commercial and institutional building construction for 31; electrical and other wiring installation contractors for 26. Roofs were the primary source in 165 fatal construction injuries in 2022, up 14.6 percent from 144 in 2021. Ladders were the source in 90, down 12.6 percent from 103. Scaffolds and staging were the source in 51, up from 49.

CPWR's June 2026 bulletin adds the current shape of the problem: construction had the highest number of fatal work injuries of any industry in 2024, and falls, slips and trips were the most common event or exposure for both Construction of Buildings and Specialty Trade Contractors that year. The 2024 totals themselves: 1,034 fatal work injuries in private construction, and construction accounted for 48.8 percent of all fatal falls, slips and trips in private industry.

2 What OSHA writes down every year

OSHA publishes its ten most frequently cited standards after each fiscal year. In fiscal year 2025 — October 2024 through September 2025 — five of the ten were height-related, and the pairing at the top and near the bottom of the list is the single most useful fact in this course.

Read ranks 1 and 7 together. The most cited standard in the United States is the requirement to have fall protection. The seventh most cited is the requirement to teach people how to use it. Those two citations usually land on the same jobsite, in the same inspection, because they are the same failure seen from two angles: the harness was in the truck, and nobody had been shown where to tie off.

3 Who dies, and when

Three patterns in the CFOI data change how you should plan a job.

Small employers carry most of the deaths. From 2011 to 2022, 70 percent of fatal fall injuries in construction occurred among workers at establishments with ten or fewer employees. Those are the firms least likely to have a written program, a rescue plan, or a competent person whose whole job is safety — and most likely to be the crew a general contractor brought in for two days of roof work.

Contracted workers are over-represented. Almost two thirds of fatal falls from 2011 to 2022 — 62.7 percent, about 2,300 deaths — occurred among contracted workers, defined as someone employed by one firm but working under the direction of another firm with overall responsibility for the site. Across all fatal construction injuries the figure is 52.0 percent. Falls concentrate at the seam between employers, which is exactly where "I thought they had their own harnesses" lives.

They happen mid-morning, not at the end of a long day. The most fatal falls in 2022 occurred between 10 a.m. and 12:59 p.m., averaging 38.3 for each hour in that window across the year. Fatigue is real, but the mid-morning peak is when the crew is fully deployed at height and improvising around a problem the plan did not anticipate.

4 Planning is the control that fails first

OSHA's national campaign to prevent falls is built on three words: Plan. Provide. Train. The order matters and the first one is the one that gets skipped, because planning happens before anyone is on site and therefore before anyone feels the hazard.

CPWR's Survey on Underlying Causes of Falls from Heights identified insufficient or ineffective planning as the primary cause of falls from heights. It also found something that should end the argument: when employers did no planning at all, the odds of workers using their personal protective equipment were 71 percent lower. A crew does not use gear that has no anchor to clip to and no place to be stored. Planning is not paperwork that follows the work; it is what makes the gear usable.

1. 1
   Plan — decide, before the job starts, how every task at height gets done: the access equipment, the anchor points, the clearance below, the sequence, and the rescue.
2. 2
   Provide — the right equipment for that plan, in the right sizes, inspected, on site the day it is needed. A 6-foot lanyard where the plan needs a self-retracting device is not "provided."
3. 3
   Train — 1926.503, by a competent person, on the systems actually in use, with a written record. Chapter 4.1 covers exactly what that record has to say.

5 What Subpart M is, and what it is not

29 CFR 1926 Subpart M — sections 1926.500 through 1926.503 plus its appendices — is the general fall protection standard for construction. 1926.500(a)(1) sets the scope: it applies to all construction workplaces covered by Part 1926 except as otherwise provided. Three limits are worth knowing from the start.

First, Subpart M does not apply to employees inspecting, investigating or assessing workplace conditions before construction work begins or after it is complete — 1926.500(a)(2)(i). That is a narrow exemption for a walkthrough, not a licence to send an estimator onto an unguarded roof and call it an assessment.

Second, several kinds of construction have their own fall protection rules that displace parts of Subpart M: scaffolds under Subpart L, cranes and derricks under Subpart CC, steel erection under Subpart R, tunneling under Subpart S, electric transmission and distribution under Subpart V, stairways and ladders under Subpart X. 1926.500(a)(2) lists them. Chapter 4.3 goes through the ones you are most likely to meet.

Third, Subpart M is a construction standard. The same worker doing the same task in a manufacturing plant is under 29 CFR 1910 Subpart D and 1910.140, where the trigger height is 4 feet and the equipment rules are written differently. Chapter 1.2 puts them side by side.

6 How this course is laid out

Part 1 answers when you owe fall protection: the trigger heights, the situations 1926.501(b) enumerates, and the holes and falling objects that account for a large share of deaths among people who were not working at an edge at all.

Part 2 answers what counts: the measurable criteria in 1926.502 for guardrails, safety nets, warning lines, controlled access zones, safety monitoring, personal fall arrest, and positioning devices. These are the numbers you can check with a tape measure and a look.

Part 3 is where compliance and survival separate. A system can meet every criterion in 1926.502 and still kill the worker wearing it if the anchorage is wrong, the clearance below is short, or nobody can reach the worker afterwards. Anchorages, clearance and swing fall, inspection, and rescue get a chapter each.

Part 4 covers the program side — training and its written record, the narrow fall protection plan exception in 1926.502(k), the standards outside Subpart M, and how to assemble all of it into something a foreman can run on a Monday morning.

Every requirement in this course is quoted with its paragraph. When a number matters, the paragraph is next to it, so you can open osha.gov and read the sentence yourself. That habit is worth more than anything you will memorize here.