If industry safety numbers are murky, how can construction

After months analyzing OSHA’s safety data, Construction Dive’s findings are discouraging: The death rate for U.S. construction workers has flatlined for at least a decade. Injury rates from the four most common hazards also haven’t declined. And OSHA’s program for serial offenders hasn’t snagged all of the worst contractors in commercial or residential construction.

Moreover, the quality and value of OSHA data for initial violations are widely criticized by the industry, citing lag times and the frequency that citations and fines are challenged, and later reduced or erased. OSHA could not provide final figures for the total amount of citations that were resolved or paid.

The rate of fatal injuries in construction is virtually unchanged

Rate of fatal work injuries per 10,000 full-time workers

“It’s no secret that when OSHA shows up it’s too late to be safe,” said Ron Taylor, a lawyer at Venable in Baltimore who has represented Whiting-Turner for over 30 years.

Turner, Whiting-Turner were inspected the most by OSHA

Number of inspections by OSHA, per firm and subsidiaries

Taylor has worked for years to reduce Whiting-Turner’s violations. The Baltimore, Maryland-based contractor had 80 initial violations between 2012 and 2021 and fines of $186,119. Taylor said he trimmed those numbers to 50 violations and the company paid just $39,705.

“All litigation has the opportunity, after initiated, to be resolved … the vast majority settle or resolve before hearing,” he said.

In response, OSHA said its “goal is to get employers to fix the hazards. Sometimes, we lower the penalty to get protections earlier … so focusing on the fines is not always the best measure. When setting a penalty, OSHA begins with the maximum penalty, then adjusts based on a variety of factors.”  

OSHA said it “continues to look at ways to improve its data analysis and collection to better protect workers and ensure equity in enforcement, including making sure sufficient data and information are available for use in targeting enforcement and compliance assistance.”

Nevertheless, a major challenge to improving safety is that there is no universally accepted, publicly available metric for construction that measures how safe a company or jobsite is, which hobbles regulators trying to craft more targeted standards and programs. Bad data also make it harder for workers and project owners to sidestep contractors that run risky jobsites.

The Fatal Four still account for same portion of worker deaths

Deaths on construction jobsites, 2011-2020

A look at the numbers

Construction Dive studied the country’s 30 largest general contractors by revenue including subsidiaries but excluding joint ventures. Addresses were cross-checked for accuracy. Many companies are privately held and not required to disclose such details, but almost all provided them. The results were shared with OSHA as well as with the 10 contractors with the most initial violations.

Measuring safety on sites is challenging. Construction Dive analyzed both total violations and violations per inspection over the past decade, but focused on violations per inspection in order to more fairly compare companies that had been inspected by OSHA at different rates.

Construction Dive also found issues with accuracy, consistency and upkeep in OSHA’s database. Violations per inspection was deemed the more reliable metric based on the time lag between when a violation was contested and resolved versus when it’s recorded in a database.

Every contractor interviewed claimed OSHA’s initial violations were an unreliable safety benchmark because final violations and fines were much lower. 

Moreover, initial violations do “not reflect the safety culture and climate of a project; the volume or type of work being performed; and do not reflect the ultimate resolution of citations,” said Cindy DePrater, New York City-based Turner Construction’s chief environmental health and safety officer. As America’s largest contractor, Turner had the most OSHA inspections during the decade analyzed by Construction Dive, but one of the lowest number of citations per inspection.

The number of inspections for the country’s largest contractors had no bearing on violations, demonstrating that having a high number of inspections does not necessarily correlate with having higher violation rates. Only one in three inspections ended with a citation, indicating the largest players know the rules and most have solid safety programs. 

To keep workers safe, contractors rely on a mix of some type of incident tracking, preventative practices, OSHA guidelines, employee training and technology-based tools. Providence, Rhode Island-based Gilbane, for instance, sees OSHA as an important partner in its safety efforts, but also has plans that exceed OSHA regulations in many areas, according to spokesperson Lynn Rasic.

“Safety is Gilbane’s highest priority on every project … investing in consistent and quality training for both labor and project management staff contributes to stronger safety and quality management across the construction industry,” she said.

Most companies didn’t receive violations with every inspection

Initial violations per inspection from 2012 to 2021 for the country’s top 30 contractors by revenue

The company with the most initial violations per inspection was Los Angeles-based Tutor Perini Corp. with 1.5 violations per inspection. (Read an article exploring the contractor’s safety record in a later installment.)

Tutor Perini also maintained initial violations were not an accurate safety metric, but did not provide its final numbers nor comment on its violations per inspection. 

Tutor Perini declined to be interviewed but defended its track record in a statement: “As one of the largest civil works contractors in the United States, we regularly compare our safety performance against industry averages, and our overall safety record is strong.”

Flawed metrics

The holy grail of safety measurements, experts agreed, is workers’ compensation insurance rates, which increase with injuries and hit a firm’s bottom line. The base experience modification rate is 1 — lower is good, higher bad. But these rates are not publicly available and laws vary by state, making comparisons difficult. 

In the 1990s, OSHA attempted to use workers’ compensation numbers with its Maine-200 programs, but abandoned them because it couldn’t get the data it needed, according to Whiting-Turner’s lawyer Taylor.

OSHA did not comment on why the program was disbanded, but agreed that worker’s compensation data is valuable. One major obstacle, however, is mustering the political will to align state laws that vary widely.