Using Risk-Based Safety Approaches to Reduce Serious Injuries & Fatalities: An Interview with Pam Walaski

Risk-Based Safety for Fatality Prevention Image

Risk management can be a powerful tool in occupational safety and health. And when applied strategically, it can help safety managers avoid serious injuries and most importantly fatalities at work.

That’s good to know, because as our risk management/safety management expert Pam Walaski explains in the interview below, the rates of occupational fatalities haven’t been going down as of late and we can do better.

In the interview below, Pam (who’s also done interviews with us on Risk Management and Safety Management Systems as well as Five Steps to Implementing Risk-Based Safety Approaches) explains the current and sad reality of occupational fatalities, explains why risk management is a tool we can use to combat these fatalities, and shows us how to get started.

As always, thanks to Pam.

Also, know we’ve included a free starter’s guide to using risk-based occupational safety management at the very bottom of this article.

You can watch/listen to the discussion in the video below or click MORE to read a transcript (the guide’s down there, too).

Introducing Pam

Convergence Training:

Hi everybody and welcome. This is Jeff Dalto of Convergence Training, back with another webcast in our webcast series, and we have a return visitor.

This is Pam Walaski. Pam is the Senior Program Director for STC, which stands for Specialty Technical Consultants. You might remember Pam from in the past, we’ve talked to her and she gave a webinar introducing using systems thinking, safety management systems, and risk-based approaches to safety. And then we follow that up with five easy steps to implementing risk-based safety programs. And today we’re going to talk about using risk-based approaches to reduce serious injuries and fatalities.

You might hear a couple acronyms for serious injuries and fatalities, you might hear SIF or FSI.

And so with that, hello, Pam and welcome. How you doing today Pam?

Stats about Fatalities at Work (There Are Too Many)

Pam Walaski:

Good Jeff, how are you? It’s good to be back. As always. I enjoy talking to you.

Convergence Training:

Yeah, me too. Thanks very much. And we continue to learn a lot about from you about using risk for occupational safety and health.

And maybe you could you start by giving us some of the trends about fatalities in workplace?

Pam Walaski:

Yeah, the the easiest way to describe the trend is that it’s been pretty well stuck. You know, I think people describe it as flat-lined.

I think you haven’t seen much movement in fatalities in 15 years. As you begin to look at the BLS census for fatal occupational injuries in 2017, it was up above 5,000. That’s a second time since 2008. It’s usually around 4,700 to 5,000. That’s generally where it is. Some people would say you have to kind of mediate that number with the fact that we have more people working and with more people working there are more hours, and with more hours, there’s more chance for fatalities.

But I think what you have to also keep in mind is that the fatality rate isn’t moving either. And so we’ve been somewhere between three and four fatalities per 100 workers, way back since 2007. Last year, it was 3.5. So we’re not seeing any movement, it really is pretty well stuck. We may go up a little bit one year and down a little bit the next year. But, when you think about that, I always say you have to be very careful because that is a number, but that is also somebody’s mom and somebody’s dad, and somebody’s brother, somebody who was loved and cared about by hundreds of people who worked with lots of people. And the impact of their loss is astronomical. So you really can’t get into sort of “it’s five less than last year, so that’s good.” As long as we’re not seeing major shifts in the number of fatalities every year, then we’re not making progress. And that’s really to me the simplest way to answer that question.

Why Fatalities at Work Aren’t Decreasing

Convergence Training:

Yeah, good point. I think we all can get numbed by data, so we agree with you about that.

Any reason why we’re stuck?

Pam Walaski:

In my observations, I think there’s a couple of reasons.

I think we continue to look at workplace safety from a compliance perspective. So in other words, we find a hazard in the workplace, we look up what our regulatory agencies say we should do to fix it, and we fix it. And then we walk away and we find a hazard somewhere else in our workplace, and we look up what the regulatory agency has to say about it, we fix it. And a couple months later, we come back and wouldn’t you know there’s the same hazard as the original one a couple months ago is back again. So I think you find this sort of circular process that never really makes any headway. Safety by hazard, if you will. So safety by managing risk, which we’re going to talk a lot about today, is really the way to go. It’s the way that we are able to identify those hazards, quantify the risk, and then bring it down to an acceptable level.

Side note: See our article on compliance & risk-based approaches to safety management.

I think the other problem that you see a lot is that we tend to elevate the trivial. So we treat all hazards the same, or we try to treat all hazards the same, or we try to attack all hazards at the same time. But we know that there are some hazards that have a much higher risk for a fatality or a serious injury. And so when we just look at it hazard-by-hazard-by-hazard, we’re not really doing a systemic way of reviewing the things that we do. And so we waste a lot of time on hazards that don’t have the highest risk. And then when we’re busy not paying attention, boom, something happens, somebody gets seriously injured or someone dies. And we’ll look at that. And we’ll sit down and we’ll put a lot of time and resources into investigating it. And we’ll find out that we looked at it before and we didn’t think it was that serious of a problem. So we put some low-level controls on there and said that should be good. And we thought that maybe the chance of that occurring was really so slim, the probability was so low, that it wasn’t something that we really had to worry about.

And we find those kinds of fatalities or serious injuries that pop up in organizations that think they have a good system going on. So those are two of the reasons, I think, that are the predominant problems that we see now as to why we continue to have fatalities coming up, we’re just not really looking at those causes of fatalities and really giving them the attention they deserve.

How to Reduce Fatalities at Work: Risk Management

Convergence Training:

Okay, great.

So I guess you’ve already kind of hinted at the answer your question, but in general, what approach do you recommend for addressing fatalities and reducing the number?

Pam Walaski:

Well, we have to approach it from a risk management perspective. A really strong, embedded risk-assessment process within our systems. And that process helps us identify those hazards that pose the greatest risk to an outcome that’s severe, as well as looking at probability. We don’t want to excuse probability. But we really have to focus ourselves on outcome, we’ve got to quantify that risk, and we have to reduce it to an acceptable level.

And so many organizations don’t really have a good handle within their own organization about what an acceptable risk is. And so they put some attention to it, they give it a control or two, and there’s really no way of saying within the organization, “this is acceptable to us, and we can move on.” So we tend to put low-level controls on high outcomes, very serious-outcome incidents. And so those low-level controls are not going to get us where we need to be.

Think about, for example, working at heights, right? If you put somebody on a ladder, there’s an awful lot of ways that they can fall and hurt themselves. They can reach too far off to the side, they cannot climb with three points of contact, they can have the ladder setup wrong, perhaps there’s not somebody holding at the bottom mean, there are just a lot of ways that somebody working on a ladder can fall. They can hit concrete, they can hit something sharp, they can fall from a reasonably low height and still have a serious injury. So we can decide that if we’re going to have people working at heights on ladders, we need to write a procedure. So we write a procedure, we train people on the procedure, and then we say, “Okay, now go do it, and we’ll talk to you later.” And then somebody gets hurt. And we go back and we sort of wonder why that happened, when we had a procedure that was really a very low-level control, training, which we’re not sure even worked, and we didn’t really look at the outcome of that particular task seriously enough.

If the outcome of a task is a fatality, standard operating procedures and training and low-level controls are not going to be an acceptable way to mitigate that risk. So when we have that risk-based perspective, and we have a task that has the potential for a serious outcome, our risk assessment teams are forced by our organizations acceptable risk criteria to really work to get that risk to an acceptable level. And chances are PPE, or operating procedures, or administrative kinds of tasks are not going to get that task down to that acceptable level.

The other problem or the other approach that I think is really important when we think about risk approaches, as they apply to serious fatalities and injuries, is something that’s called a critical-to-safety-control. So if I’m doing a task that has the potential for a fatality or serious injury as an outcome, that’s a reasonable credible consequence, then the controls that I apply to bring up that risk to an acceptable level are critical to the safety of performing the task. And if those controls fail, or they’re not being followed, then the potential for somebody having that outcome goes up significantly. And so we’re going to use those controls, those controls have to be controls that we can verify are actually being implemented as designed. So if we have a standard operating procedure that talks about how to safely work at heights, we have to have some way of knowing that the procedure is being followed. So we have a conformance rate. We have supervisors who are checking things, if we’re doing training for particular tasks, we have to have some way of knowing that the people who went through the training actually absorb the information and are applying what they learned. Not that they sat through the training, signed their name, ate their donut, and then went on to do whatever it was that they wanted to do. We have to make sure that if we’re giving people fall protection systems, to work at heights, that they’re using them properly, they’re inspecting them, they’re tying off properly, all those things are being done.

So those critical to safety controls. Have to be scored, and conformed, and verified. Otherwise, the chances of those controls failing or not being followed are too high. And then we leave ourselves open to that particular kind of outcome. And I think you see that a lot.

Convergence Training:

Okay, good. I like that example, in general, but it’s also great timing, just before OSHA Safety Stand Down, in a week or two, for fall prevention, so thank you for being timely.

Pam Walaski:

Yeah, that’s why I chose it.

Side note: Check out our recent interview with Craig Hamelund of Oregon OSHA about the Safety Stand Down and fall prevention/protection.

Overlooked Risks That May Lead to Fatalities

Convergence Training:

(laughs) Well done. Okay, so what are some kind of risks that people might commonly overlook when they’re performing a risk assessment that later might actually lead to a fatality in maybe an unexpected manner.

Pam Walaski:

I think there are two ways I would categorize those types of risk that we generally tend to look at less critically.

One is something that we might consider a one-off; something that is low-probability even though it’s high-severity, but the probability of it happening, we’ve assessed that risk and we decided that it’s very low. The chances of that happening, we sort of look at particular risk, and we decided it’s really not that significant. And so we don’t put as much time and attention into that particular risk. And we allow that outcome to be dismissed, because we think the chances of that happening aren’t that good.

So what happens is that the risk score for that particular task is lower than we would normally see that risk score be because the probability is so low, but the outcome is high. And so with the risk score being low, then what we decide to do to fix that or what’s acceptable within our organization is not as significant as what we might do if that score were higher.

And you know, you call them one-offs, you call them, you know, low-probability/high-severity outcome events. You hear about them all the time, DuPont and BP and some of the other organizations that have suffered those kinds of significant losses, often we find those kinds of problems.

The other thing that I think is problematic is when we use our incident history as a way to determine what the probability is of a particular event happening. So if you have never had a fatality for a particular task or not even a serious injury, then we tend to artificially deflate the risk, which again leads to the same problem. We feel like we can manage that risk with low-level controls, because we don’t think the risk is very high to begin with.

Or maybe we’re conducting a risk assessment on the task for the first time. We’ve never done this task before; we’re adding something new to our production line; we’re working with a new chemical; or whatever. And we don’t have any incident data to look at because we haven’t done this particular task before. And so again, we dismiss the lack of incident data, rather than saying we need to go find incident data somewhere because we don’t have it, or find some other way of increasing the risk score, because of a lack of data.

So we can look outward for data, we can go to our insurance company, we can go to the Bureau of Labor Statistics (BLS), we can do accident facts from the National Safety Council, there are a lot of ways to get that information.

Or we can set up our risk assessment process, so that it automatically eliminates the lack of data as part of our risk score. And so we would remove the frequency of occurrence data from our floor, and we might look just at the frequency of exposure, how often is somebody doing the task, and the outcome. So we put those two together, and we ignore the fact that we don’t have any occurrence data, rather than having a low occurrence data score causing our eventual risk score to be lower.

And so I think those are some of the things that you see often when somebody has a fatality and they sit down and do an investigation. They weren’t prepared for that particular problem because of some of those types of ways of looking at it.

Precursors and Causes of Occupational Fatalities

Convergence Training:

That’s interesting. I never thought about replacing the probability or likelihood of an event occurring with the exposure rate. So that’s a good piece of information, thank you.

My next question is going to have to do with something called a precursor and something called a cause. And so I wondered at first, before I asked you my question, if you can explain the difference between a precursor of fatalities and the cause of a fatality.

Pam Walaski:

So, a pre-cursor is one of those terms that doesn’t really have a formal definition that everybody agrees to, sort of like some of us say SIFs and some of us say FSIs.

But a pre-cursor is fairly commonly described as something that is an indicator of the potential for a high-outcome event, a high probability for a fatality or a serious injury.

And a cause, of course, if a precursor would be sort of a leading indicator, if you would, and a cause would be how somebody actually died or how somebody actually got seriously injured in a particular workplace. So more of a lagging indicator.

Side note: Check out this article on Lagging & Leading EHS Indicators.

And most people can look at their data, their incident data, and can find lots of common causes for the more serious injuries that they’ve had within the organization.

Convergence Training:

Great, thanks for setting that up. And the reason I’m asking you for that is I know in the past, you have given me some lists of common precursors and common causes of occupational fatalities, and I wonder if you could just tell me, tell us where that list came from, and then give us the list if you could?

Pam Walaski:

Yeah, sure. So I think if you look back to the point at which concern about fatalities and serious injuries began to really bubble up in the literature or in conferences that you might have attended, or things that people were saying, some sort of thought leaders, I think you started to see that in the early 2000s, you know, maybe 2002-2003, people started really looking at the data and noticing that things were not trending downward anymore.

I think if you go somewhere around 2005, maybe through 2007, you began to see organizations gathering large datasets and looking at precursors and looking at causes. Groups like BSI or ORCHSE, they would gather their organization’s representatives together, and they would share their data and they would look at that data and just kind of crunch those numbers. And they began to really pull some commonalities out.

I also heard a lot of this kind of stuff when I attended a fatality prevention forum, which was held here in Pittsburgh, which is where I’m from. It was sponsored by Indiana University of Pennsylvania, their Safety Sciences department, in conjunction with the Alcoa Foundation, and it was the second of two, the first was in 2007, I did not attend that one. But in 2012, they had a lot of presentations, where the focus of “here are the common pressures and here are the common causes that we’re finding,” they began to talk about that. And so you began to see that coming out.

Another source for that kind of information came from a person who I admire greatly. His name is Fred Manuele, he has written a lot of books, a lot of articles, he’s been quite the thought leader in occupational safety and health. He published a book called Advanced Safety Management in 2008. And one of the things he did in that particular book was he convinced some of his clients to give him copies of their incident investigations. And he looked through over 1000 incident investigations and he started tracking the causes of the actual incident as well as the activity that the people were involved in at the time that the incident happened. And his list of activities and causes was, again, very similar to what other organizations were coming up with.

So between all those sources, we’ve sort of come together with a common group or common list of precursors and causes. Precursors are good because it forces an organization to look at high-risk activities. And really, if you’re going to do a risk assessment, and you have a a thousand employees performing any one of 750 different tasks that have each maybe five or six different hazards, you might think to yourself, “How the heck am I going to get started on this process?” To think about precursors, the kinds of activities that are most likely to cause a fatality or a serious injury, allows you to kind of hone in on those particular activities. That would allow you to pay a little bit more attention to them, to kind of put them at the top of your list of risks that you want to assess first.

So, a precursor would be non-routine operations or operations where there are high-energy sources, like electricity, or thermal, or nuclear; they would be situations where you would have contractors on your site; working at heights would be another one. Those kinds of activities tend to put more people at risk for serious injuries and fatalities.

Causes on the other hand, are again very similar, somebody is working at heights, somebody is engaged in trenching there, they’re failing to perform a lockout-tagout properly, and so there’s an energy source present. Those are the kinds of things that we most often see are causes.

In fact, if you’ve been following the news, or just in the past four or five weeks, there was a spate of trenching fatalities. Everywhere you looked at every time you turned around, there was another fatality occurring in trenching collapse. The last one that I read about the other day was a person who was working in a trench that was 25 feet deep. And there was no shoring, no trenching, no benching, no nothing, he was just working in a very deep trench.

So those are the kinds of things that we know have the highest risk for fatalities and serious injuries. And if you look at the BLS data set, when they put out there census of fatal occupational injuries, you’re going to find a lot of those very similar causes.

And just a last word on that the most frustrating thing for me as a safety professional, is that these causes are nothing new. These are not new tasks or activities that we haven’t performed before. These are things we know how to do safely. People are always going to have to work in heights, people are always going to have to work with high-energy sources. They’re always going to have to do work in excavations. And it’s not that we can eliminate those kinds of tasks, but we know that they pose a high risk, and we know how to protect people from them. But because we fail to really assess the risk and address the risk, we allow people to do things with low-level controls and not really pay attention to the risk significantly enough.

More on Precursors of Occupational Fatalities

Convergence Training:

Okay, great. So I know I’ve got a list of precursors and a list of causes. Could you maybe walk us through your list of common precursors, and maybe give us a little more information about how people can put that knowledge to use?

Pam Walaski:

Well, again, the precursors would be things like non-routine operations, working with high-energy sources, having contractors present in your work site, where you’re doing your normal work. Working at heights is a precursor, that particular type of activity. Those are the things that you should be looking out for and when you’re getting down to begin to do your risk assessment process, those are the things you want to focus your time and attention on first.

Again, because those are your highest risk tasks. And those are the things that are going to require more time and attention and a higher level of controls to get them down to a successfully acceptable risk level.

Convergence Training:

There, I just, I just did an interview with a guy from Oregon OSHA talking about the fall protection issue. And to your point about higher level controls, hile you’re right, inevitably we will always have people working at heights, one of the things he was talking about as well, when can you eliminate it? And you know, if you’re going to do work at heights, can you do that work on the ground first, those kind of controls should be better than just slapping some fall protection on somebody.

Side note: Here’s that recent interview with Craig Hamelund of Oregon OSHA talking about using higher-level controls for fall prevention/protection.

Pam Walaski:

Absolutely, absolutely. And we don’t often spend the time to do that. Because we don’t really have a formal risk-based process that forces us to do that. If we have an acceptable-risk criteria, whether it’s a number, or color, or however we do our risk assessment process that says you may do this particular task and until the risk is at this score, you’re going to find that the chances of you being able to successfully get it down to a low enough level with low-level controls are pretty difficult to do. And so it does force your risk assessment teams to higher-level controls.

And if you can’t implement those higher level controls for one reason or another, say, again, let’s go back to the height example. If you can’t bring the work down to the ground level, you’ve looked at that adjust, you can’t do that, now you’re back to understanding and appreciating that whatever controls that you have in place are critical-to-safety controls, and back to the discussion we were having earlier, those controls take on a whole different level of importance and meaning in terms of making sure that they’re being followed and complied with the way that they were set up to be used. Deviation from that control is going to increase the risk that fatality or serious injuries going to occur.

More on Causes of Occupational Fatalities

Convergence Training:

Great. So if that’s a little discussion on precursors, could we do the same thing with causes? Do you have like a little list of common causes that lead to fatalities often, and maybe some tips about how people can put that information to us?

Pam Walaski:

Sure. And it’s very similar. You can look at precursors and you can look at causes and they’re really just about the same kind of thing. So people who fall people, who are electrocuted, people who are involved in excavations and are caught with a collapse, people who are working in tasks where they may not be aware of the energy sources, contractors working in the area, all those kinds of things are the typical causes of more serious injuries and fatalities.

And just like you use those precursors to elevate your risks in a more proactive way, you should never ignore the data that you might have on the causes of your more serious injuries. So if you’re looking at, again, look at your OSHA logs or your workers comp claims for the past five years or even go back further, and take a look at the most serious injuries and begin to think about how did this happen, when was the this particular injury, and begin to tally that information. And chances are, you’re going to find out that it matches up with those particular types of causes. And that when you did the investigation of that incident, if you did it in a fairly comprehensive way, you found that the control failed in some way, shape, or form. The person didn’t wear the proper PPE that they were supposed to, they didn’t tie off. We didn’t use the protective system that we should have by putting somebody when we put somebody in a trench. We didn’t follow a lockout tag out procedure the way it was written, or the procedure was so poorly written that when people followed it, it didn’t work because we hadn’t really taken the time to verify that that procedure was supposed to be working the way it was.

So again, if you don’t have a significant pattern or history of serious injuries and fatalities, causes can be a little bit trickier. But that’s where we go back to the discussion before. If you don’t have that incident history, then that’s where you need to look elsewhere for data. Contact your insurance broker and find out what they can share with you from other clients who do things that are similar to what you do. Look at BLS data, look at National Safety Council data, talk to your trade association, find out what other people who do what you do, what kinds of problems and causes they’ve been having for their injuries.

Because if you’re an organization that’s never had a fatality or hasn’t had one in a really long time, or generally tends to have minor injuries, you have to think about what was the worst-case scenario of those particular incidents. And instead of just looking at what actually happened, you need to look at what could have happened. And so when you assess the risk of those kinds of tasks, you have to look at the worst case scenario. And assume that the worst case scenario is what you want to protect your workers against.

One thing you have to be careful of with worst-case scenarios, though, is you need to make sure that they’re credible worst-case scenarios, right? You can sit down and you can look at any particular task somebody has to do, and you can imagine a scenario where they could get killed during that task. But it’s really credible, right? If you overemphasize the worst-case scenario, then you end up assessing risks at a much higher level, you sort of go the other way, as opposed to under-valuing the risk score, you sort of inflate the risk score artificially. And so again, you’re back to elevating the trivial and wasting a lot of time on reducing the risks of something that may not be that serious. So you have to be careful about that. And when you’re talking about a worst case-scenario, I always recommend that people use the term credible worst-case scenario, and really have

Jeff Dalto, Senior Learning & Performance Improvement Manager
Jeff is a learning designer and performance improvement specialist with more than 20 years in learning and development, 15+ of which have been spent working in manufacturing, industrial, and architecture, engineering & construction training. Jeff has worked side-by-side with more than 50 companies as they implemented online training. Jeff is an advocate for using evidence-based training practices and is currently completing a Masters degree in Organizational Performance and Workplace Learning from Boise State University. He writes the Vector Solutions | Convergence Training blog and invites you to connect with him on LinkedIn.

Contact us for more information