Blog

Workplace Violence Doesn’t Only Happen in Hospitals

Small storefront business in daylight, one employee alone with a customer at the door

Key Takeaways

Hospitals dominate the workplace violence numbers because hospitals are where somebody counts. Incidents at a twenty-person business enter no dataset at all.
Exposure follows the situation a worker is standing in. Industry codes report none of it.
A quiet year measures your reporting path, and in a small business that path runs through you.

Every story you’ve read about workplace violence was set in a hospital, and the reason has less to do with where the danger lives than with where somebody bothered to count it. The federal survey behind the headline numbers sorts injuries by industry. Healthcare fills the top of that table every year, and an owner reading it files the whole subject under another industry’s problem.

That filing decision is the most expensive one on this page. It’s also why workplace violence prevention gets treated as a hospital purchase rather than a small business decision.

We’ve been putting safety systems into places where people work alone with the public for close to a decade, and the pattern holds. The businesses with the least visibility into their own exposure are the small ones, and the reason is structural rather than careless. You can’t see what nobody reports, and nobody reports into a dataset that was never built to hold them.

Why does workplace violence always sound like a hospital problem?

Because healthcare is where the counting happens, and the counting there is thorough. Across 2021 and 2022, healthcare and social assistance accounted for 72.8% of every workplace violence case recorded in private industry, according to the Bureau of Labor Statistics. That share is real, and it’s also why every article on the subject opens in an emergency department.

Look at what has to exist before an incident becomes a number. A hospital has an incident reporting system, a risk manager whose job is to run it, an accreditation body that asks what the reports say, and a legal department that wants the record to exist. An assault on a nurse passes through all four before it reaches a federal survey. Run the same incident through a fourteen-person dental practice and a patient’s brother gets aggressive at the front desk, the receptionist is shaken, the day continues, and nothing gets written down anywhere.

So the concentration in the data measures two things at once: where violence genuinely runs highest, and where the machinery for recording it exists. Healthcare scores high on both counts. Small business scores high on neither, then reads its own absence from the table as safety.

What actually puts a worker at risk, if not the industry?

The situation they’re standing in. Alone with a customer. A back office after closing with the day’s money in it. One person handling a stranger who has decided they’ve been wronged and has nobody else to say it to. Those conditions turn up in a clinic, a hotel, a repair shop and a real estate office, and not one of them appears in an industry code.

That last part is worth sitting with, because it explains why you can’t find the number you’re looking for. The federal survey records what happened and the industry of the employer it happened to. It has no field for whether the worker was alone, whether a door was closed, or what time the shift ended. The condition that moves the risk was never collected. The honest reading is that nobody has ever counted this the way it would have to be counted.

The rate outside healthcare runs lower and it stays well above zero. Private industry as a whole recorded 2.9 violence-related injury cases per 10,000 full-time workers in the Bureau of Labor Statistics count, against 14.2 in healthcare and social assistance. Healthcare carries the higher rate by a wide margin, and every point of that 2.9 is built from the same handful of situations, spread across every sector the survey covers. Our position on this is settled: the shape of the danger is a closed door with one person on each side of it, and risk concentrates wherever a person’s standing to object is weakest. The person who can least afford to push back is usually the one standing in the situation. Pushing back is how you get moved off the schedule. A safety arrangement that only works where other people can see you has missed the place where harm happens.

We build for exactly that situation: one person, a closed door, nobody watching. If you want to know what covering it looks like in a building your size, ask us.

Contact Us

Why is “nothing has happened here” the weakest evidence you have?

Because you’d be the last person to hear about it. The absence of reports measures your reporting path, and in a business your size that path runs through you. Three things have to happen before an owner learns that something occurred:

The staff member names it. They have to file what happened as an incident rather than a rough shift with a difficult customer. Most people take the second reading, because the second reading lets them go back to work.
They decide it’s worth telling. They weigh whether raising it changes anything, and whether it marks them as the person who makes a fuss.
They tell you specifically. There’s no HR function to route it through in a small business. Reporting means walking up to the person who writes the schedule and signs the paycheck.

Underreporting is a rational response to a process that rarely protects the person who uses it. People stay quiet in unsafe conditions because they need the job, the rent is due, and the arithmetic of speaking up rarely favors them. Any read on your own exposure that skips past that is starting in the wrong place.

There’s a second reason your own read comes back clean, and it sits underneath the reporting problem. The situations that carry the exposure are normal operating conditions. Somebody opens up alone because that’s the shift. The office manager stays late to close the books because that’s when the books get closed. None of it gets raised with you, because raising it would mean questioning how the business runs, and everybody involved knows the business runs that way for reasons.

Here’s the honest limit of what anyone knows: nobody has measured how much goes unreported outside healthcare. Every underreporting study solid enough to cite was run in hospitals, and hospital figures don’t transfer to a five-person office. That gap is worth noticing on its own. The workplaces least able to see the problem are the ones nobody has studied.

What does one incident actually cost a business your size?

More than the injury, over a longer stretch, through three separate doors. They arrive months apart on different pieces of paper, which is why most owners never connect them into one number.

Start with the kind of injury violence produces. Among California workers’ compensation claims filed for PTSD with no physical injury alongside it, between 2009 and 2018, violence accounted for 48.9%. Nearly half of an entire category of claim traces back to something somebody did to a worker, and it’s the category an owner is least likely to watch happening.

Those claims also run long. The median was 132.5 days away from work, against 91 days for the same diagnosis from any other cause. Four and a half months, in an operation where one absence means the owner covers the shifts personally, pays the overtime, or does both.

Where it lands	What it looks like
The claim	An average of $47,316 across all industries and all causes
The time away	A median of 132.5 days for violence-related PTSD claims
The premium	Frequency counts for more than severity in your experience modification
The team	The people who watched how you handled it, which no rating plan measures

That $47,316 average comes from National Safety Council figures covering every industry and every cause, and it’s the number your insurance carrier is working with while you’re still thinking of the event as an upsetting afternoon.

The third row is the one owners see coming least often. The experience modification that sets your workers’ compensation rate cares more about how often you have claims than about how bad they are, and NCCI sets the dollar threshold that decides it, which since late 2023 has varied state by state rather than running to one national number. So one moderate claim can move your rate about as much as a severe one, and it stays in your rated experience for years after the person involved has recovered or moved on. The incident closes. The pricing of it doesn’t.

Why does workplace violence prevention keep waiting for somebody else to decide?

Because nothing forces it. No inspector arrives, no renewal form asks the question, and the subject sits below payroll and the broken freezer until either a mandate or an incident moves it up the list. Both of those are worse moments to decide in than a quiet Tuesday.

Mandates are coming, in more states and covering more kinds of workplace than they used to, and knowing what applies to you is worth an afternoon. What a mandate does to the decision is the part worth understanding now. When a purchase is driven by a requirement rather than a problem, price becomes almost the whole decision. The question turns into what’s the cheapest thing that satisfies the rule, and the cheapest compliant thing gets chosen against the rule rather than against your building, your hours and your people.

Safety and dignity get handed out like a privilege in this economy, and small businesses are where that shows up most plainly. There’s no committee, no program and no budget line standing between what the owner decides and what the staff actually get. That’s the whole case for deciding on your own terms while the decision is still yours.

What should you be able to answer before you spend anything?

Two questions. They’re worth putting to anyone selling you something, and they’re worth answering about whatever you already have in the building.

If someone presses for help, does help know which room to go to? An alert saying something is wrong somewhere on the premises sends people looking. An alert that names the room sends them arriving.
Does it hold up when conditions are bad? Power out, network down, phone in a drawer on silent. Those are the exact conditions that show up alongside the worst moments, and anything quietly depending on all three has a gap in it.

If noise is genuinely all you need, a cheap alarm off a shelf is a fine answer, and we’ll say so plainly. It’s faster and it costs less. Those two questions are what separate that from something you’re relying on: whether help reaches the right room, and whether it works every single time.

Ask them of any vendor. Then ask them of your current arrangement, which is probably a buddy system and a cell phone, and see how it answers.

Hospitals really do carry the highest rate of workplace violence in the country, and they’re also the only workplaces that count carefully enough to prove it. Your business sits in the part of the table nobody measured, which is a different thing from sitting somewhere safe.

So the question worth carrying out of this is smaller and harder than whether you work in a high-risk industry. Which of your people, this week, will be alone with somebody they can’t predict? You already know the answer. You knew it before you opened this page.

What changes now is that a quiet year stops working as proof, and those situations become the thing workplace violence prevention is actually about. No one should be afraid while trying to earn a wage. In a business your size, nobody but you decides whether they are.

WORKPLACE VIOLENCE PREVENTION