In today’s episode, David and Drew explore the concept of behavioral substitution as a strategy for decluttering safety practices, examining a 2023 paper from the JBI Evidence Implementation journal titled "The Big Six Key Principles for Effective Use of Behaviour Substitution." The discussion centers on how healthcare's approach to de-implementing low-value practices can inform safety professionals' efforts to replace ineffective safety measures with more valuable alternatives.
You’ll hear six key principles for effective behavioral substitution, drawing parallels between healthcare and safety contexts. They discuss how these principles can guide both the removal of ineffective practices and the implementation of new ones, emphasizing the importance of considering practical needs, existing skills, and organizational resources when making such changes. The episode provides valuable insights for safety professionals looking to improve their organization's safety practices through evidence-based substitution strategies.
Discussion Points:
Quotes:
"You can't swap out something that people believe works for something that they don't believe works." - Drew Rae
"A lot of the safety, if not all the safety work we do in organisations is about anxiety reduction, not necessarily about improving safety.” - David Provan
"Rather than thinking about decluttering as just what we can reduce or take away, it may be more useful to think about it as a process of gradually swapping out each thing that's not working well." - Drew Rae
"If you can't explain the substitute behavior with the same ease which you can explain the behavior that you want to be implemented, then people have to work a bit harder and they might go. Why are we making this all so complex?" - David Provan
“That's the point they're making here, is like maybe the patient doesn't need care, but that doesn't mean that we shouldn't acknowledge their need for care and their need to be taken seriously.” - Drew Rae
Resources:
The Safety of Work on LinkedIn
David: You are listening to the Safety of Work podcast, episode 129. Today we’re asking the question, how can we use swapping as a strategy for decluttering? Let’s get started.
David: Hey everybody. My name’s David Provan and I’m here with Drew Rae, and we’re from the Safety Science Innovation Lab at Griffith University in Australia. Welcome to the Safety of Work podcast. In each episode, we ask an important question in relation to safety of work or the work of safety, and we examine the evidence surrounding it.
Today’s paper, and today’s topic was something that sparked an interest with you recently and you sent it through. Do you want to just talk a little bit about the background of this idea as to where we get swapping as a strategy for decluttering from?
Drew: I think you pointed out in the episode notes that [Jean-Christophe Lacoste] recently had a thread up on LinkedIn that had a lot of discussion about behavior-based safety, and the way we tend to either support it or dismiss it in the safety wars.
One of my beefs has always been that both the people who dismiss it and the people who support it are often talking about very outdated ideas about what behavioral safety is. We’re talking about 70 years of psychology research and going back to the first 10 of those years isn’t fair for anyone.
But then I realized that I was sitting in a workshop this week, and I encountered a whole subfield of behavioral science that I was not aware of. This is the type of thing that I think we do a lot in safety is we don’t realize that other people have the same problems and may be talking about addressing those problems.
You and I have written a lot about decluttering. There’s apparently another name for decluttering, which is de-implementation, which sits within the field of implementation science, which draws on modern research in behavioral science.
It’s not an exact parallel because when we’re talking about de-implementation, particularly in healthcare which is what we’re going to do, it’s really a wide spectrum of what we are de-implementing.
At one end is the traditional stuff that we talk about in decluttering, so management systems, processes, and safety practices that never had any evidence behind them. They were just introduced and have become bureaucracy, and there’s no evidence that they add to the safety of work.
At the other end are very specific clinical practices. They might have originally been evidence-based or originally been best practice, but they’ve gradually become outdated, and now clinicians are doing things that the current best evidence says that they shouldn’t be doing.
Or sometimes it’s just like creep, like people gradually start ordering more and more tests even though there’s no evidence that ordering those tests actually help the patients. All of this falls under the body of implementation science and de-implementation.
I thought that at the very least, there were some interesting parallels and probably actually even a lot of overlaps between what we think about as decluttering. I thought we all had something to learn from reading and thinking about this paper.
David: I really, really enjoyed reading this paper. I think two weeks ago when we talked about supervision in social work and now we’re talking about de-implementation healthcare clinical practice, and maybe it might be a bit of a trend this year if we can keep up our consistent cadence like we’ve done at the start of the year, that we might even help our listeners with more research and evidence from adjacent domains to Safety Science.
One of the things I think since we’ve written a lot about decluttering and that paper was a 2018 paper, at least I felt that when we talked about decluttering safety—that term is very widely known in the safety world—I felt like people were going to go off and use it as an opportunity to make a lot of change in their organization.
But for me, the decluttering in industry around safety is so that the pace has been quite glacial. I think what was missing there is different strategies for doing these types of things. Not just taking something away, but what we’ll talk about today, which is in the title about swapping and what that can look like.
I’m going to enjoy this. I think just maybe to help with a clear example, and you gave a couple of examples there, but this idea of de-implementation which is one of the examples in the paper, is like for example if a patient presents with these symptoms in these circumstances, then the treatment strategy should be X. Like prescribing a certain medication or ordering a certain test.
And like you said, as the evidence moves on and now our understanding of treatment for certain presenting patients moves on, we actually want to get very experienced clinicians to change the way that they think through and problem-solve in relation to a particular patient. Like don’t use this type of medication or don’t order these tests in these types of circumstances because it’s either unnecessary, or at worst, we’ve actually learned that it can harm the patient.
These are changes that you actually really want to make. But one of the underlying challenges with all of us as humans is that when we’ve got a lot of experience with behaving in a certain way in certain situations, that can be very difficult to change.
I’m not one of these people who says behavior doesn’t have a role to play in safety, because ultimately what we want to achieve in safety is people in our organizations to behave in certain ways in certain situations. I think there’s a lot of science that can help us here.
Drew: I thought a good parallel that we could use here is the doctor ordering an unnecessary test for a patient, feels very like management wanting to investigate every minor incident that happens. It’s a very natural response, but the evidence might say that’s not actually the best thing to do for the patient.
How do we then change that reaction when you’ve got the patient demanding, hey doc, I want to do something, or you’ve got the manager saying, but we’ve got to do something because someone got hurt? Often just saying no, we are not going to do anything is not the right answer or a helpful answer. We need to think about what would actually be a good answer to give instead.
David: Exactly. We’re starting to allude to this idea of swapping and what we’ll talk more about in I think as substitution. Would you like to introduce the paper?
Drew: The title is called The Big Six: Key Principles for Effective Use of Behavior Substitution and Interventions to De-Implement Low-Value Care.
David: You probably don’t mind that as a title. I think they starting with the big six probably, as someone who loves people who invest time, effort, and thought into the title of a paper, just says what it does, but it’s pretty catchy at the start.
Drew: Numbered lists are a nice way to structure a paper. I don’t think you need it in the title, but it’s very clear, like key principles. The paper says in the title exactly what the paper does. I love the clarity.
The authors, Andrea Patey, Jeremy Grimshaw, and Jill Francis are all at healthcare research institutes primarily in Canada. I think one of them might be in Australia.
The lead author, particularly Dr. Patey, is an implementation scientist specializing in de-implementation. I think she’s the perfect person to write this paper because her whole career has been studying how to roll out new healthcare practices and how to roll back unwanted healthcare practices.
One of the questions she asks, which I think is a really challenging question, is is there actually a difference in between the strategies we need to roll out something new and the strategies we need to roll back something old? I think that is still a little bit of an open question. Is the psychology just the exact same or do people have different psychology and therefore need different behavioral incentives when you appear to be going backwards rather than forwards?
JBI Evidence Implementation published in 2023, so very recent. Most of what’s published in this journal is very empirical papers talking about the science of implementation. This is more of a summary or commentary piece, just because we thought that would be more accessible for our audience rather than picking one of the specific evidence-based papers that sits behind this work.
David: I think that idea of de-implementation being the same or different is an interesting question because in Edgar Schein’s humble inquiry work, he talks about one of the most difficult challenges with learning is the unlearning that needs to happen alongside the learning.
I think that’s a really interesting open question about how we think through that as a change exercise, whether we’re putting something in or taking something away.
Drew: Should we get to the basic point of the paper?
David: Yeah. Do you want to start with the research aim or what their paper’s set out to do? And then we can talk about the details.
Drew: Okay. This is a particular strategy for de-implementation, and the strategy is called substitution. In other words, if you’ve got a practice that you think is ineffective and possibly even harmful, then one thing that people try to do is rather than just saying stop it, they say do this instead. It’s a very specific strategy that is used.
It’s not actually empirically supported as a broad strategy. There might be particular examples of where it’s worked or where it hasn’t worked, but it’s an area where we need more evidence that it’s actually a good strategy. But it is absolutely a common and very intuitive strategy to try rather than just saying don’t do it. You don’t want the toddler putting something in their mouth. Rather than just saying stop, you put something else in their hand that’s maybe safer to chew on.
David: That’s a great example parents could relate to, and I think the paper made the point quite early that, like you said, there hasn’t been researched that thoroughly. However, the research suggests that just telling people not to do something and leaving it open to then decide how they’re going to deal with that or what they might do instead might not be as effective as giving someone a clear thing to do in that situation instead.
Drew: Just before we go too further, I just realized I fell into the classic trap of behavioral science, which is I just completely infantilized the people who are doing these low-value practices, and using little kids is such an easy analogy to reach for. But yeah, this isn’t that. That might communicate the basic idea because you humans are human even when they’re little kids.
But these are adults who have good reasons for doing what they’re doing. These are medical professionals who’ve been doing things that have been effective for them in the past. And to have a patient in front of them who has needs, just not doing something is not an acceptable answer often in that situation.
What they’re going to do instead, if you tell them not to do one thing, maybe they’ll ignore you because they don’t have an alternative. Maybe they’ll just pick up something else to do instead, which has an even lower evidence base or has more potential for harm. We are actually leaving the situation uncontrolled if we just try to stop.
David: Maybe I could talk through an example, an extension of how I interpreted part of the writing as a way of helping make it clear, and maybe test even with yourself my understanding of this.
For example, a patient turns up to see a general practitioner and presents with certain types of symptoms, which maybe in the past is a very obvious candidate for a course of antibiotics. As we learn over time more about antibiotics resistance and we really want to minimize the amount that we just give people antibiotics, we may not want our general practitioners to use medications like that unless it was absolutely necessary.
However, in the paper telling a general practitioner not to give someone antibiotics without what to do instead doesn’t leave them able to necessarily know, well I’ve got a person who thinks they’re sick and they’ve come to me for some treatment, so they’re looking for me to do something, so what am I going to do?
This idea of substitution is like, well, okay, prepare a pamphlet of material, talk the person through their symptoms, hand them some information, tell them what they need to look out and monitor for, and tell them in what circumstances they might need to come back for further care. It still validates their attendance at the clinic, and it gives the general practitioner a strategy for how to de-implement this immediate prescription of antibiotic medication.
In the absence of that substitution or what to do in that setting, it can leave the general practitioners really vulnerable to, well I need to offer some kind of treatment, so what do I do?
Drew: That’s a great example, and I’m sure listeners can draw parallels between that when we are trying to take away ineffective safety practices. I don’t know how much pushback I’ve had with Take 5s, that people want to know, okay, so if we’re not going to do it, what do we do instead? And I just try saying over and over, well it doesn’t work, so just stop it. And that doesn’t work as an answer.
The same thing with TRIFR, your injury frequency rates. They provide no information, so why don’t we just get rid of them because senior management is demanding a number. You can’t take away people’s number without putting something in its place.
David: And I think in the context of this paper, it’s like someone going to a doctor thinking that they’re sick and being told, go home, you’re not sick. It’s a similar situation and this paper talks about it quite well.
This paper you mentioned it’s an essay. It doesn’t actually do its own original research. Do we want to talk a little bit more about some of the background or we want to just dive into what this paper does actually do?
Drew: The main thing I’d say is that part of the motivation for this paper is that the strategy of substitution has not been well-tested. If you are going to test something, you have to create the conditions under which it might succeed. There’s no point in testing something badly and then saying, oh, it didn’t work when you didn’t try hard.
What they’re doing is they’re laying out, if it works, these are the things that we would need to do in order to make it work so that other people can try it as a strategy, making sure that they’re following principles that would give it the best chance to succeed. Then they go through each of these things, and under each one of them is linked to the underlying psychology or behavioral science principles that indicate why we would need to do this.
David: That’s a great point because the risk there, in the absence of this proposed way of conducting this research, is people start substituting certain behaviors with all other types of behaviors. Then maybe the well-researched intervention concludes that the practice doesn’t work.
It may not be that the practice substitution doesn’t work. It just might be that the selected substituted behavior doesn’t work for a range of reasons. They really want to make sure that we’re testing the substitution process by having a way of deciding how we should decide what the substituted behavior should be.
Drew: These are six principles for choosing a substitute behavior to hand in place of trying to declutter a particular practice.
David: So, are we ready to talk about these?
Drew: Okay, but I think we’re going to hate the first one. Principle number one, evidence and rationale. Identify a substitute behavior that has a clinical rationale or strong evidence base for its use. In other words, substitution is not going to work in safety.
David: Well, is that because we feel that there’s not enough evidence base for a lot of the practices that we do, let alone the practices that we might want to substitute in?
Drew: Yes. Good luck finding a substitute that has a strong evidence base.
David: I think you could maybe expand this particular principle, to be a really strong argument for why this substituted behavior is a good thing to do. This is really about the strength of belief. We talked about that in the decluttering thing, how easy is it for people to believe that this is a good thing to do in this situation?
Drew: That’s entirely fair, and that is the underlying change management processes that they’re pointing to. They’re talking about beliefs about consequences, memory, attention, to decision processes. You can’t swap out something that people believe works for something that they don’t believe works.
In healthcare, the way to make people believe that something works is to have an evidence base. Maybe in safety we can get away with just having a strong theory or a strong rationale for why we think the new thing will work better.
David: It looks like in the notes you’ve got some safety examples for each of these, so I don’t want to steal your thunder around that. How might we think about an example in safety?
Drew: Let’s say that the thing that we want to get rid of is investigating every minor incident and having a mandatory rule. Minor incident happens, you’ve got to investigate. We say instead we are going to do learning teams. That is only going to work if people have a genuine reason to believe that learning teams are going to work better than the incident investigations. You can’t just say this is the latest thing, this is a good thing. People have actually got to have good reason to believe that it’s an appropriate substitution.
David: So the first principle around evidence and rationale is to make sure that the thing that we are substituting for the particular behavior or practice that we want to de-implement has a really strong case for why it is a better thing to do.
Drew: And that reinforces the whole idea we want anyway, that we should be doing things that have a strong evidence base. it doesn’t just help for this change. It helps for all of our future changes. We get people into this cycle of talking about evidence, thinking about evidence, thinking about rationale, and that’s our reasons for doing things rather than shiny new objects.
David: That’s our first principle. Maybe just a reminder for listeners about that open question about are these principles that you would use to select something to substitute for something else, potentially the same principles that you might use to decide to do something completely new as well? A strong argument for doing something and that feels like one that it’s a good one to think about in maybe both of those contexts, but let’s see how it holds up for the remaining five.
Drew: Good question. The second one is about the objective. This is really interesting because they actually talk about two objectives. They say identify a substitute behavior that serves the clinical objective and the practical objective.
In the case of a patient, your clinical objective might be making the patient better, but your practical objective might be making sure the patient knows that they’ve been taken seriously. And you got to do both of those things.
It may be that clinically the best thing to do for the patient is tell them to go away, but it’s actually important that they feel validated that they turned up to the doctor. You don’t want to discourage them from doing that or make them immediately just go off to some quack because you weren’t taking them seriously.
David: In safety, I think it was Erik Hollnagel, maybe a decade or so now, who said that organizations need to both be safe and feel safe. His comment was about the latter getting in the way, the latter taking priority over the former. A lot of the things we do in safety is about feeling safe rather than being safe.
I know you’ve had a lot of attention in the last year about your idea that a lot of the safety, if not all the safety work we do in organizations, is about anxiety reduction not necessarily about improving safety. I love this paradox here if you like, or this duality here that we need to make sure that what we are doing has a safety of work–type of risk reduction, safety improvement benefit, but it also makes the people feel how we might want them to feel as well. That power was really quite striking to me.
Drew: This is where I think I’m quite at odds with people like Hollnagel. Hollnagel talks about feeling safe rather than being safe. He’s dismissing the feeling safe. He’s saying we should focus on the being safe.
My point is, maybe it is entirely valid that the stuff that we are doing is to make people feel safe. That might be a legitimate need. Or at the very least it’s a need that we have to address if we want to reform safety. We can’t ignore that need. That’s the point they’re making here.
Maybe the patient doesn’t need care, but that doesn’t mean that we shouldn’t acknowledge their need for care and their need to be taken seriously. We’re not going to give them the good care, or at least they’re not going to accept the good care if they feel that we’ve dismissed or haven’t taken seriously their anxiety.
David: I think this is the point about Take 5s, or recordable injury rates, or investigating minor incidents, or choose your aspect of safety practice that you think doesn’t solve the first which is the actual safety outcome, is very much playing a role in the second, which is people feeling like they’re taking safety seriously and doing what they need to do.
I like that principle, just those multiple objectives and satisfying those when selecting a substitute behavior, which is why we probably see, like you mentioned earlier, a lot of the whole just stop doing it has not really been a widely adopted de-implementation approach.
Drew: And I confess there that this is something that I’ve had to learn because I have tried the ‘just stop doing it there’s no evidence for it’ strategy. It’s from trying that and just seeing how little effect it has that’s got me to realize that it is important to take these other purposes into account rather than dismissing them.
Your example of that might be to stop and think like, okay, maybe investigating every minor incident doesn’t create learning and doesn’t create practical improvement, but is it meeting some values-based need that the organization has to signal that every time someone gets hurt, that’s not acceptable to us? We care about that. We are not dismissing the injury. We’re not dismissing the danger. If we’re going to substitute it, we need to substitute it with something that also carries that same signaling effect.
David: And I think I also saw some success with an organization in that exact example where it was very difficult to satisfy the second objective, which is we do not investigate these things by saying, yes we do. There are three questions on an incident report form, which is what we call an initial incident investigation or a basic incident or give-it-a-name that satisfies the second one. But it means that people can still get that need satisfied.
Yes, we do still report everything and we do ask if there’s an immediate issue that needs to get resolved. It gets done all on the same day by the same person and the same supervisor. People can still go, yes, we do still report. We still see if there’s something that needs to be done. But when we say investigation, we don’t mobilize everything for everything. I think the nuance around de-implementation and decluttering is what we are learning over the last five or six years.
Drew: I think this feeds directly into the next principle, is that we need to identify a substitute behavior that’s easily explainable to patients. For patients here, I would just directly substitute management and the workforce. We have to be able to clearly and simply say, what is this new thing that we are doing and why are we doing it instead?
David: To carry on the example of helping general practitioners be able to communicate with patients, that this is what we now know about antibiotics and antibiotic resistance, and this is why we are doing the things the way we’re doing, here’s some information, here’s what to look out for, and here’s when to come back. Something that can be actually quite easily explained.
I think it’s a great parallel with managers and workers when we want to de-implement or substitute safety practice in the organization, how easy is it to understand that new practice.
Drew: We are going to take away TRIFR, and in return we’re going to give you a two hour–long lecture on statistics, variability, sum distributions, and confidence intervals.
David: Or we’re going to take…
Drew: It’s not easy to explain sometimes, and creating those simple explanations is hard.
David: And here’s an index with 10 or 15 different variables combined to be 97.5 on a scale of 1–364. It’s got to be something that I think we’re going to talk about a few things in the next remaining principles that are things that should be no more difficult. I think if you can’t explain the substitute behavior with the same ease, which you can explain the behavior that you want de-implemented, then people have to work a bit harder and they might go, why are we making this all so complex? Why can’t you just tell me how many people we hurt? And if it’s less this month we must be doing better. You need to be able to explain your substitute with as much ease.
Drew: The next one, identify a substitute behavior that is no more time-consuming than the undesired behavior. In other words, are you expecting someone to actually start neglecting other duties in order to swap the behaviors over?
David: If we think if we go straight to the safety domain now, we know for Rasmussen’s dynamic risk work, we know that investing in safety in the short-term borrows from available capacity to apply to core activities. We know that that creates an instant trade-off or goal conflict with achieving other things as well as the safety work, safety practices.
I really like this as a principle, which is if something used to take 10 minutes and you now want someone to do something that takes 15 minutes, then that’s going to be a very difficult thing to do.
Drew: I think this one probably applies even more to that idea of new things in safety rather than decluttering. Because I don’t think anyone tries to declutter by swapping in something that is more onerous. But we do often try to reform safety, and particularly in the new view and Safety Differently, we try to introduce new alternate activities that are actually higher workload.
David: And even learning teams and the way that some views on learning teams is this is a two-day activity, it needs 8 or 10 people from all across the organization, you’re immediately going to the organization saying, oh, I want to actually do a learning team on this activity. I need 20 working days of effort to run this learning team.
There are definitely some things that we talk about in contemporary safety improvement. Now I for one think that learning teams need to be done that way, but we are asking for a lot of time from the organization sometimes. Someone goes, no, why don’t you just get one person to do an investigation? It’ll take two days. Why do we need all these people to be involved? So I think this timeline is a really important consideration for implementing and substitution.
Drew: Next one, fit with skills. Identify a substitute learning behavior that has a good fit with existing skills. In other words, is the person going to have to learn a new skill set or can they already do the substitute behavior?
This one I think we get wrong in safety. Not just safety. We get this wrong in organizations all the time. We think we’ll make a change, we’ll roll out the change, and as part of rolling out the change, we will just do a quick training course to equip people with the skills they need to do the change with no consideration for how threatened people are when you’ve just given them something that fundamentally changes what they’re good at to something that they’re not good at, and told them this is now your job.
David: An example that jumps out at me is something like field leadership activities. We’ve talked on an episode before about management by walking around and leadership presence. I think that paper was even a healthcare paper.
But we see that trend in business to go from, I don’t want leaders to be out there with their teams doing checklists and compliance activities. I want leaders to be out there doing humble inquiry, deeply engaging, exploring work has done, the gap between work has done to work as imagined, drift and how all this works and all of this.
You’re going, okay, I’ve got a supervisor who used to go out and fill out a form. Now, I want them to go and have a deep 30-minute conversation with only open questions in a psychologically safe environment with their team. I’ll give them a 30-minute session on how to do that, and then wonder when the program doesn’t quite go as hoped. I know I feel about that example, but I’ve seen that a few times.
Drew: The other one I’ve seen directly in safety is people who get into safety because they like well-defined rules, systems, and procedures. They like writing and managing systems, and working with documents. That’s what they’re good at. Then we tell them, no. Toss out all those documents. Toss out all those systems. Go and talk to people.
It’s not just the discomfort with the new. It’s the difference between feeling really competent, feeling really secure, having found a job that matches your values and needs, and then having those challenged and replaced with something that puts you straight into the discomfort zone.
David: And one of the ways the listeners can use this episode in this paper is just to think through where your challenges are going to come from, because you won’t easily always have all six or find options and strategies that easily meet all of these six. But at least I think this can help serve as kind of like a, oh, this is going to take more time. This is going to be difficult. Oh, this has got new skills, this is going to be really difficult. Or I need to find a way to easily explain this.
I think there may be some times where we just can’t have all these criteria met, but at least I think listeners would know where they’re going to find difficulty and where to focus.
Drew: That almost sounds like a good takeaway. Should we run through the last principle and then talk some more about?
David: Yeah, okay. Sorry about that. I got a bit excited. Okay. The sixth principle?
Drew: The sixth one, nice and simple, cost. Identify a substitute behavior that’s no more expensive to perform than the undesired behavior. What is the point of decluttering if what you actually do is fill the gap with something that is either immediately more expensive? Or the other thing they’ve talked about in some of their other work is it basically grows to become more expensive.
The simplest version of it might be cheaper, but once you actually think through what if everyone did that, suddenly everyone doing that grows into really quite an expensive thing for the organization.
David: We talked through those six. There’s a nice table in the paper that I’ll paste in the comments when we publish on LinkedIn. But these authors didn’t just make these six things up. There’s this aspect of research, theoretical domains framework (or TDF for short), that actually synthesizes these six from other theories, in particular, implementation theory. The lead author has a long list of publications through the use of this theoretical domains framework as well.
I just wanted to add that, even though this is more of an essay-based paper, what do we know in the broader literature and how can we think about this in the context of de-implementation and substitution? The authors didn’t just sit down and go, okay, let’s jump in front of the whiteboard.
Drew: For every one of these, there are multiple theoretical- and evidence-based reasons. Why in order to get someone to change their behavior, you need to consider this principle?
But overall though, the authors do caution that even though each one of these principles holds, it isn’t actually fully established that behavior substitution as a whole works. What we need to do is try some good behavioral substitutions, following as many of the principles as possible, see what happens, and then use that to start building our decluttering strategies.
David: Does this mean you’re now on a lookout for a PhD student who wants to come to Griffith and do a behavior-based safety project? And have you told Sid about that?
Drew: Ooh, I love that idea. Yeah. The honest answer is universities are really quite siloed, and one of the areas that Griffith doesn’t cross well is between the area that the lab is in and the psychology department. Probably it would actually be better for someone in the psychology department bringing one of us on as an additional supervisor rather than us trying to supervise a psychologist down our end.
But yes, don’t let that stop anyone who wants to reach out to consider being a PhD student or sponsoring a PhD student, because taking money from someone who would like us to demonstrate that behavioral science works in safety, that would blow Sid’s mind in a way that I think he and I would enjoy.
David: Excellent. Okay, let’s get into the practical takeaways properly then. How about you get us started please?
Drew: Okay, so the overall message of this paper is rather than thinking about decluttering as just like what can we reduce or what we can take away, it may be more useful to think about as a process of gradually swapping out each thing that’s not working well with something else that more effectively fills the same need.
The underlying assumption there is that most things there are probably filling some need, even if it’s not an evidence-based need. Maybe it’s a practical need. We need to meet that need as we do the decluttering.
David: I’d always felt a little bit like that with the decluttering piece. The intent behind the activity always has some legitimacy. Like inductions—okay, bring people up to speed, that’s a good thing to do. Audits—go and check how things are working, good thing to do. Investigation—something’s gone wrong, you didn’t plan for it, find out what happened.
It is hard to argue with the intent behind a lot of these practices. What we’re more concerned about is the effectiveness of the application. I think that idea of swapping something out with something that feels the same need.
So audit’s not working. Yes, but it’s still important for us to know what’s going on. Okay, so how can we do that? So that we still feel like we’ve got confidence that we know what’s going on, but we’re better meeting that objective of actually understanding and improving safety.
Drew: Absolutely. That way we are not challenging people’s values either. There’s no risk that they think we are doing less safety or caring less about safety.
Number two, the principles here fit equally well with implementation and de-implementation. This is the point that you were getting onto earlier, that probably this is a good way of thinking about any change to safety practices.
We might not be able to meet all of them, but getting one of these badly wrong is often the reason why a safety improvement fails. We should use them as our list of challenges we might meet, and think strategically about how we are going to manage these challenges.
David: And I think, comment here that like you said, they come from implementation. We’ve got these six principles. I think the evidence doesn’t suggest that there’s a reason why they wouldn’t work when trying to progressively substitute activity in the organization.
In the absence of having the evidence for what we know works when substituting and what we know doesn’t work, then this is a reasonable place to start experimenting.
Drew: The third thing I’d put as a takeaway is for regulators and researchers. This is yet another reason why we need to keep doing that slow, hard work of creating an evidence base for each safety practice.
It is not enough to have grand theories that tell us how we should reform safety if we can’t provide people with this clear evidence base, that swapping one practice for another is a good idea along with a nice simple explanation for that specific change rather than just a broad reform agenda.
David: And I think that’s a big shout out to this journal. We haven’t had a paper before from JBI Implementation, but when you look at the about page, that journal is just very specific evidence-based healthcare implementation research.
I’d love to see a journal like this pop up in Safety Science. It might be a bit lean for a little while, but this idea of we want to understand specific safety practices in specific domains and specific circumstances where something works and doesn’t work.
This paper is a three- or four-page paper, and I think the idea of really knuckling down, I’d hate to see the… I don’t know. That’d be a broad claim, but yeah. You comment there about more grand theories in safety, more books, more big grand books in safety is not what we probably need.
Drew: So the question we asked this week was how can we use swapping as a strategy for decluttering?
David: Well, I think choosing something to swap in that maybe satisfies these six principles. Then even more broadly, whenever we’re trying to do something new in safety, maybe think about these same six principles as well.
Drew: So that’s it for this week. We hope you found this episode thought-provoking and ultimately useful in shaping the safety of work in your own organization. As always, you can reach us on LinkedIn or send any comments, questions, or ideas for future episodes to feedback@safetyofwork.com.