When One Site Learns About a Hazard, the Company Has Recognized It: The General Duty Clause Across Locations

A regional operator has a near-miss at one of its warehouses. A pedestrian walks into a forklift lane that has no physical separation, the driver brakes in time, and a supervisor writes the event up. The site fixes it with bollards and a repainted walkway. The report is filed in the site's folder and, if the company's systems are typical, goes no further. Eighteen months later, a worker at a different building, run by a different manager under the same company name, is struck in a nearly identical layout. When the investigation begins, someone asks a question that has nothing to do with the second site: what did the company already know?

That question sits at the heart of the General Duty Clause, section 5(a)(1) of the Occupational Safety and Health Act, 29 U.S.C. 654(a)(1). For an organization with one location, the clause is a background rule. For an organization with many locations, it is a rule about how information moves, and most portfolios are not set up to answer for it.

What the clause requires, and when it applies

Section 5(a)(1) requires each employer to furnish employment and a place of employment free from recognized hazards that are causing or are likely to cause death or serious physical harm to employees. OSHA turns to it where no specific standard addresses the hazard. Where a particular standard in Part 1910 or Part 1926 covers the condition, enforcement ordinarily proceeds under that standard, so the clause is the fallback for hazards that the specific rules do not reach.

In OSHA's Field Operations Manual, a citation under the clause generally rests on four elements. The employer failed to keep the workplace free of a hazard to which its employees were exposed. The hazard was recognized. The hazard was causing or likely to cause death or serious physical harm. And there was a feasible and useful way to correct it. Each of these can be argued, and each is a place where the facts about a company's own history matter.

The element that matters for this discussion is recognition. OSHA can show that a hazard is recognized in several ways, including through industry knowledge and through general understanding, but the one that carries the most weight in a contested case is often the employer's own: what the company's records, programs, communications and decisions show it already understood. Internal incident reports, near-miss entries, insurer loss-control findings, audit results, worker complaints and prior corrective actions can all serve as evidence. The employer does not need to have written the word "hazard" for the record to be read that way.

An employer is a company, not a location

The multi-site consequence follows from a plain observation. The statute speaks to the employer. For a company with forty locations, the employer is the company, even if each location runs its own morning meeting and keeps its own binder. Whether an adjudicator would attribute one site's knowledge to the entire enterprise depends on the facts and the forum, and that is a question for counsel. What operators cannot assume is that the boundary around a site's folder is a boundary around what the company is treated as knowing.

That is uncomfortable, because it runs against how large organizations naturally operate. Sites keep their own records. Local managers hear about problems first and fix them within their own budgets. Regional leaders learn about some of it and corporate learns about less. In many portfolios, the people with authority to change a standard practice never see the incident that would justify it. The clause turns that gap from an internal inefficiency into an enforcement exposure: the company knew, in the sense that its own records say so, and the question becomes what it did about it.

There is a practical point on the other side of the same coin. The fourth element, feasibility, can draw on what the employer itself has already done. If one location fixed a condition with a particular guard, layout or procedure, and another location with the same exposure did not, the first fix is evidence that a feasible method existed. An enforcement officer or a plaintiff does not need an outside expert to show what could have been done; the company's own best site has already shown it.

Where this shows up in real portfolios

The hazards most often litigated under the clause tend to be the ones that have no dedicated standard or only a partial one. Heat is a current example, and we have written about how it plays out across state lines in the state patchwork for heat illness. Ergonomic exposures are another, discussed in our multi-site ergonomics post. Workplace violence, combustible dust before a standard-specific approach was settled, and various machine and material handling configurations have all been pursued through section 5(a)(1) at different times.

For a multi-site company, the exposure is not the existence of the hazard. Every operation has some. It is the pattern of recorded knowledge without a recorded response. A near-miss that was fixed at one site and never communicated to the others is a typical example. A trend visible in insurer loss runs that no one in operations reviewed is another. An internal audit finding that was marked closed at the site level because the local fix was done, while the identical condition at sibling sites went unexamined, is a third.

It also connects to how OSHA classifies violations. A condition found again after it has been cited elsewhere in the company can be treated as a repeat violation rather than a first-time one, a subject we cover in the same gap at two sites. The General Duty Clause and the repeat classification feed on the same underlying weakness: the company's information about its own sites is not connected.

Operators in states that run their own OSHA-approved plans should also expect those states to have their own authority and approach. Our post on state plans for multi-state footprints covers the structure; the point here is that a general duty obligation exists in some form across jurisdictions, and the details of how it is enforced differ.

What a defensible response looks like

The goal is not to stop recording incidents. Records are a good thing, and a company that stops writing things down to avoid creating knowledge has traded a manageable risk for an unmanageable one. The goal is to make sure that what is recorded at one location gets read somewhere with the authority to act on it.

A workable approach has four parts. First, a common intake: incidents, near-misses, audit findings, worker concerns and insurer recommendations should land in a structure that can be sorted by hazard type across all sites, not only by location. Second, a regular review, at a fixed frequency, by someone with the authority to change a company-wide practice, that looks for the same hazard appearing in more than one place. Third, a recorded decision. When a hazard is identified as recurring, the file should show what was decided, who decided it, and the date, whether the decision was to adopt a company-wide control, to adopt a control at certain sites with a stated reason for the difference, or to accept a risk with the rationale written down. Fourth, a follow-through record showing that the control was actually implemented and checked at each affected site, not only announced.

None of this requires a safety department of any particular size. In many large operators this work sits with operations, HR, or a plant manager who also runs a P&L, and the structure above works for them as well, because it asks for a routine and a record, not a title. What it does require is that someone is accountable for looking across sites at all. Without that person, the company is relying on each site to escalate what it learned, and sites escalate unevenly.

The uncomfortable arithmetic of scale

A company with a handful of sites can run this by phone calls and a shared spreadsheet. A company with dozens cannot, and a company that has grown by acquisition inherits sites whose records follow another firm's conventions, a problem discussed in our piece on the acquired site. The more locations, the more incidents, and the more chances that the same hazard is documented in two places by two different people who never compare notes. The same scale that makes the clause harder to manage makes its consequences larger, because an unremedied pattern is easier to describe as a company-wide failure than an isolated one.

This post is general information about how the clause works, not legal advice, and following it does not guarantee compliance with federal or state requirements. Whether and how a particular company's internal records would be treated in an enforcement or litigation setting is a question for qualified counsel. If you are accountable for safety across many locations and want to see incident, audit and corrective-action information across every site in one place, request a demo or talk to sales to see how WorkSafely Enterprise approaches it.