By Jon Zibell, VP of Global Alliances & Marketing, SciSure
Earlier in my career, I was doing consulting work inside research labs. Before I looked at a single report or requested data, I asked the people in the lab a simple question: do you feel safe here?
The answers can be startling. Some people say, “Of course, we’ve got a great team. They invest in us and care about our safety.” Others pause and say, “Most of the time.”
I bring this up because the most common way laboratories measure safety can also be the weakest. Zero incidents this quarter. Zero recordables this year. It is an easy number to report and an easy one to celebrate. But an incident log can mean your controls are working – or it can mean it wasn’t reported. From the outside, those two labs look the same.
In a survey of about 2,400 scientists commissioned by UCLA’s Center for Laboratory Safety, 86% said they believed their labs were safe places to work. Just under half had been injured at some point. More than a quarter of junior researchers said they’d had an injury they never reported to a supervisor.
So, if “zero incidents” is not the measure, what is? In my experience, below are a few components of a good lab safety program; one that’s likely to be clear on whether it’s working:
- Hazard-based training tailored to the actual work,
- Automated reporting that gives real-time and actionable insights,
- An EHS team that researchers see as partners.
All three can be measured, but most labs are not measuring them yet.
Get your denominators in order
Years ago, I sat in a presentation from an organization that had recently lost a researcher in a fatal lab incident. The slides that followed were all about progress. “We used to deliver 100 training courses per year. Now we deliver 250.” More flame-resistant lab coats. More PPE (personal protective equipment) on the shelves.
I sat there the whole time with one question in my head: What is the denominator? If you went from 100 training courses to 250, is that good? Should the number be 500? 2,000? Are we celebrating progress, or a slightly smaller shortfall?
Afterwards, I caught the speaker in the hallway. I told him I liked the progress and asked what the targets were. How much training should have been delivered? How many people should have been wearing flame-resistant coats? He looked at me and said, “I don’t think we know that. We’re just driving the number up the best we can.”
That conversation has stayed with me ever since. You cannot work toward a number if you do not know what it should be.
“We delivered more training” is one of the things I hear often from EHS teams, and it’s a slippery metric. A count tells you about effort and improvement. It says nothing about coverage. The useful version is a ratio: training completed against training required, people protected against people exposed.
And the denominator moves: a new chemical arrives, a protocol changes, new equipment goes in, or a process scientist starts splitting time between the development lab and pilot plant. Each of those changes what “fully trained” means. If you set your targets once a year, you’re measuring against a goal that no longer exists.
Training built around the lab, rather than the job title
Good safety training is targeted and tailored at every level. Someone in a physical plant needs different training from someone in an office, even if they both sit through the same IT security course. But it gets really complicated inside the lab.
Most lab training is assigned through HR filters and a standard Learning Management System (LMS). Your job title says you work in a lab, so you get the standard lab training list. Anything specific to your lab gets sorted out locally by someone in your department, if it gets sorted out at all.
Your job title doesn’t know what you’re exposed to. To train someone properly, you need to know:
- Which hazards are in the spaces they work in, from chemicals and biological agents to the equipment itself
- What the people around them are working with, since exposure doesn’t stop at your own bench
- Every space they move between, because someone who works in three labs needs to be trained for all three
- What PPE they need, based on the highest level of risk they’ll face
Training has three jobs here: prevent the incident, prepare people to respond if it happens anyway, and make sure they have the protection to limit the damage if all else fails.
The response part is the one that slips. Safety Partners, an EHS consultancy that publishes an annual collection of real laboratory incidents, described a case in its ninth edition where a colleague at a nanotechnology core facility lost consciousness and the people around them didn’t follow the medical emergency procedures they’d been trained in. In their words, employers in that edition learned that “just doing an initial training is inadequate.”
So, the measure I’d watch is coverage: how many people are trained for the hazards they’re exposed to today, and how quickly that gap closes when something in the lab changes. The course count won’t tell you either.
When reporting is a chore, risk sits in someone’s inbox
The second place safety programs break down is reporting. Incidents, near misses, inspection findings: all of it depends on someone taking the time to log it and someone else taking the time to act on it. Every extra step in between is another reason it does not happen.
Inspections are a good example. When we start working with a new team, we ask them to walk us through everything from the moment they decide to inspect a lab to the moment the last finding is closed. It’s rarely simple. An email goes out to schedule it. The inspector takes notes in an app on a phone or tablet. Those notes become a PDF, and the PDF goes back out by email. Then the reminders start. “Can you close out these findings?” “Any update on what we found last month?”
That back and forth frustrates everyone. Lab Operations wants the labs running smoothly. Scientists want to get back to their work. EHS wants the findings resolved, compliance achieved, and risk mitigated. Meanwhile, the findings sit there for days, weeks, and sometimes months.
I think of it as a lit wick. The moment a hazard is identified, the wick is burning, and every day it stays open, you’re carrying that risk. The goal is to put it out fast, document that you did, and move on with a clear compliance trail of resolved findings.
That gives you a simple measure. Start a stopwatch when an inspection is scheduled and stop it when every finding is verified as resolved and the report is closed out. If you can’t say with confidence how long that takes, end to end, something in the process isn’t working. A finding that’s closed on paper but never checked hasn’t really been closed.
Incident reporting follows the same logic. The tenth edition of Safety Partners’ incident collection asks investigators to “look beyond what may seem like an immediate and obvious cause, such as not following a procedure.” Was there time pressure? Was the training adequate? Those questions only get asked if the report exists.
That’s why I treat slow, clumsy reporting as a safety risk in its own right. Think back to those junior researchers with injuries they never reported. Some of that is culture, which I’ll come to next. But a lot of it, I’d bet, is effort. If logging a near miss means finding the right form, filling it in, and emailing it to someone who may not reply in a timely manner, plenty of people will decide it isn’t worth it.
Is your EHS team the police, or a partner?
One of the first things I ask when I visit a customer, prospect or partner is whether their EHS team sees itself as a service provider and a collaborator, or whether it operates more like the inspector. A lot of teams tell me they’d love to be more of a partner, but they’re so busy chasing issues that they end up policing.
Say inspection data shows that someone keeps getting flagged for not wearing a lab coat. The policing response is a note: you’re in the red, do better. The collaborative response is a question: “I noticed this came up again. Why aren’t you wearing the proper lab coat?”
The answer is often something you’d never see in the data. “I didn’t know I needed a specific one for this work.” Or, “We don’t have any in this lab.” Every one of those is fixable. None of them get fixed by a reprimand.
That’s the kind of conversation labs want with health and safety, and it’s the kind that builds a strong culture and trust. Use the data to start a discussion. If it only ever produces a list of violations, people learn to keep their heads down, and what you get is check-the-box compliance. You need to be compliant. But compliance on its own won’t build a safety culture that drives down risk.
In fact, I’d go even further to say that building that culture is the responsibility of EHS and compliance teams.
Safety as an accelerator
Most of the organizations we work with exist to improve people’s lives. EHS has a real job to do, keeping labs safe, compliant and properly trained, with regulatory reports that are accurate and up to date. But the wider organization’s job is to get good, reproducible science to patients, quickly.
That’s why I believe health and safety, done well, speeds research up. A policing environment is a friction point. A good EHS team looks for places where science could move faster or more safely, and steps in with a fix that supports the researchers on both counts.
The best research organizations I’ve seen work the same way, with EHS, lab operations and research teams pulling toward the same finish line. That has consequences for measurement. Team metrics matter, because they show each group is doing their job. But they should add up to the organization’s business goals. If EHS is measured only on safety compliance scores and research only on output, you’ve built two teams that pull against each other.
What to measure beyond zero incidents
Incident counts still belong on your dashboard. They just shouldn’t be alone there. Here’s what I’d add:
- How people feel.
Ask researchers directly: do you feel safe here? Are you trained for the hazards around you? Do you have the PPE you need? Do you believe EHS takes your safety seriously? Are there any safety requirements that slow down science? Where are the friction points? Then ask why. Numbers can be bent to tell almost any story. People standing in the lab usually give you a straight answer – because they care. - Training coverage.
Completed against required, for the hazards each person faces right now. Track how long people have been overdue, as well as how many are delinquent. - Time to close.
From inspection scheduled to every finding resolved and verified. - Repeat findings.
The issues that keep coming up across a lab, a site or the whole organization: lab coats not worn, the wrong type of coat, biohazard bins overfilled. Watch whether each one is trending up or down and run an educational campaign on the highest risk findings. - Near-miss reports.
In my experience, more near-miss reports usually means people trust the system enough to use it. Read what they say, too. - Equipment readiness.
Is it calibrated, and is everyone using it trained on proper and safe use?
In a spreadsheet, spotting a pattern in repeat findings means stopping to analyze months of data. On a live dashboard that updates every time a finding is logged, you see the arrow pointing up or down straight away.
That’s what lets you walk into a lab and say, “This has happened a few times now. Is the biohazard bin in the wrong place?” You’re asking about the root cause instead of handing out a citation. And that’s the point where an EHS team stops reacting to problems and starts getting ahead of them. This is the foundation of building trust with your scientists.
Where to start
Pick one lab this week. Ask the people in it whether they feel safe and ask why. Then go back to your own numbers and check each one for a denominator. If you report training delivered, find out how many were required. If you report inspections completed, find out how long the findings stayed open.
Those two exercises will tell you more about your program than a year of zero-incident reports.
In the end, it should be about keeping our scientists safe and productive, so they can improve the lives of patients who desperately need them. EHS teams play a major role in this. Embrace that role and help drive the science forward. The patients are counting on you.
About the author: Jon Zibell
Jon Zibell is Vice President of Global Alliances & Marketing at SciSure, where he leads strategic partnerships with organizations like The Engine (MIT), My Green Lab, and Safety Partners to help life science and research institutions modernize lab operations and compliance. He writes about the operational, safety, and technology challenges facing modern scientific organizations. Jon holds a B.S. in Marketing & Corporate Communications from Bentley University.
References
-
Safety Partners (2025). Incidents, Accidents, and Near Misses in Laboratory Research: Highlighting the Critical Role of Emergency Response Training. Available at: https://safetypartnersinc.com/incidents-accidents-and-near-misses-in-laboratory-research-highlighting-the-critical-role-of-emergency-response-training/
-
Van Noorden, R. (2013). Safety survey reveals lab risks. Available at: https://www.scientificamerican.com/article/safety-survey-reveals-lab-risks/
-
Safety Partners (2025). Incidents, Accidents, and Near Misses in Laboratory Research. Available at: https://safetypartnersinc.com/wp-content/uploads/2025/01/IANM_Publication_Vol_9_DIGITAL.pdf
-
Van Noorden, R. (2013). Safety survey reveals lab risks. Available at: http://www.nature.com/news/safety-survey-reveals-lab-risks-1.12121
