In this episode, we’ll be discussing the paper entitled, “Managing risk in hazardous conditions: improvisation is not enough” by Rene Amalberti and Charles Vincent (2019), published in BMJ Quality & Safety. Though the paper is focused on the healthcare industry, we can extrapolate the findings to safety in other industries. We’ll discuss the need for different or modified levels of acceptable safety measures in “degraded operations” when 100% adherence to safety rules is simply not possible.
The paper’s abstract reads:
Healthcare systems are under stress as never before. An aging population, increasing complexity and comorbidities, continual innovation, the ambition to allow unfettered access to care, and the demands on professionals contrast sharply with the limited capacity of healthcare systems and the realities of financial austerity. This tension inevitably brings new and potentially serious hazards for patients and means that the overall quality of care frequently falls short of the standard expected by both patients and professionals. The early ambition of achieving consistently safe and high-quality care for all has not been realised and patients continue to be placed at risk. In this paper, we ask what strategies we might adopt to protect patients when healthcare systems and organisations are under stress and simply cannot provide the standard of care they aspire to.
Discussion Points:
Quotes:
“I think it’s a good reflection for professionals and organistions to say, “Oh, okay - what if the current state of stress is the ‘new normal’ or what if things become more stressed? Is what we’re doing now the right thing to be doing?” - David
“There is also the moral injury when people who are in a ‘caring’ profession and they can’t provide the standard of care that they believe to be right standard.” - Drew
“None of these authors share how often these improvised solutions have been successful or unsuccessful, and these short-term fixes often impede the development of longer-term solutions.” - David
“We tend to set safety up almost as a standard of perfection that we don’t expect people to achieve all the time, but we expect those deviations to be rare and correctable.” - Drew
Resources:
The Safety of Work on LinkedIn
David: You're listening to The Safety of Work Podcast episode 102. Today we're asking the question, what is the right strategy when we can't manage safety as well as we'd like to? Let's get started.
Hey, everybody. My name is David Provan. 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.
Drew, this question, what's the right strategy when we can't manage safety as well as we'd like to, surely, that must be being asked in lots of organizations. Should we dive straight in? Would you like to introduce the paper we're discussing, and then we can get straight into it?
Drew: Sure. I think the paper does a pretty good job of introducing the topic itself, so let's just introduce the paper itself. The authors this time are Rene Amalberti and Charles Vincent. Regular listeners should be familiar with the name Amalberti. We talked about him in episode 85, where we discussed his famous paper, The paradoxes of almost totally safe transportation systems.
Charles Vincent, I haven't read much of his work, but he is a big figure in clinical and patient safety. He's currently a professor of psychology at Oxford. But for many years, he was director of an NIHR Research Center for Patient Safety and Clinical Improvement at Imperial University in London. So two giants of authors of this paper.
The title is Managing risk in hazardous conditions: improvisation is not enough. It's published in 2019—it's going to be interesting once we get further into this paper—in the British Medical Journal quality and safety, so a very reputable journal. The paper, though, I should note is labeled as a viewpoint paper. This isn't representing original research or even literature review.
The paper is basically an essay presenting opinions or talking about controversial issues. As best I can tell, though, they still actually peer review the paper. Anytime that it's (for example) referencing other research, you should consider that to be a reasonably high quality.
David: Thanks, Drew. I guess that's why I spent a little bit of time on the authors, because a viewpoint paper is a nice way to describe it. There are some interesting questions here that we'll ask because we'll talk about the intersection of Safety I and Safety II. And we'll talk about the intersection of (I guess) absolute safety and best possible safety at a point in time. It's a real tension that is described in this paper that is real for many organizations.
I guess lots of industries and systems are becoming ever more complex and under increased pressure. I guess this paper sets out that I think the opening sentence is something like, healthcare systems are at a stress level like never before. This paper was submitted in February 2019. I wish these authors knew what was to come in 12 months in terms of the healthcare system and obviously Covid. Any thoughts about a system that is always more stressed than it ever has been before until the next stress that comes along?
Drew: David, we're not a political podcast, but this is raising questions of which was worth Covid or 10 years of Tory government. I think that's a genuine question you asked, and that's one of the points they put forward in this paper. We can't just say things are bad now, we've got to wait till we get better, because things often never do get better.
Were often under stress for years or careers. It's not like we're just in a temporary state of, oh, we just have to be stressed at the moment. We'll wake up, we've got some more time, and then we'll worry about safety.
David: I fell into that personal trap of going, oh, I'll just be busy for another couple of months and then not be busy anymore. I think it's a good reflection for professionals and organizations to go, oh, okay, what if the current state of stress is the new normal or what if things become more stressed, is what we're doing now, the right thing to be doing, and that's the center of this paper.
It is healthcare-centric. Obviously, this paper is in the BMJ Quality & Safety journal. We'll try and extrapolate out to other industries as well. But specifically in relation to health care, the authors say, there are a lot of new and serious hazards for patients, and there are a lot going on in hospitals. Again, this was before Covid, in terms of patient volumes, the complexity of health conditions, the reduced or shortfalls in staffing.
It means that the overall quality of care falls short of the standard expected by both patients and professionals. Given the situation in the healthcare sector, the standard of care that patients expect and the standard of care that professionals expect to provide is not able to be achieved.
Drew: I think this is a question that comes up for a lot of people. Actually, David, it was the same day that you had sent me this paper to get ready for the episode. I got an email from a healthcare safety professional asking almost that exact same question. What do you do as an organization when you've got this standard of care that you know you aspire to achieve?
Right now, the job involves shortcuts, workarounds, and all the things that your safety management system says are bad for safety. You know you just can't snap your fingers and go back to safe. You can't say, well, okay, if we can't keep safe, well, then we've got to go out of business, because you're a hospital. Your whole business is saving lives that aren't going to be saved if you're not there providing the care, even if you're not providing the care as well as you want to.
David: It's this idea that this is the aspiration of care in the hospital setting. If we take it outside of that setting, it's like this aspiration that companies have of zero injuries or zero harm. We know we're not there yet.
Our organization is under a lot of pressure. We've got to keep producing. What's the best thing to do now? Is it to continue to try to manage as if we're at this aspirational state, or is there a different approach that could be more effectively taken in that moment?
Drew: I think it goes beyond the zero harm argument. You could be experiencing a low rate of injuries and still be telling yourself, okay, yes, we've got some injuries occurring. It's not perfect, but we're very satisfied with what we're doing to prevent injuries. We're very satisfied that our systems are working well. We're following the systems.
This paper asks the question, what happens when you're not even at that state, when you can't say we had the injuries despite doing everything we could? You know you're not doing everything you can, you just don't have enough time, don't have enough resources to do even the things that you know are the next things you need to do in order to improve.
David: Good clarification, Drew. This paper goes on to talk about what they described the section as the evolution of poor performance. It's introduced that teams and organizations, most of our listeners, I guess, would understand this, that teams and organizations constantly have to adapt in times of increased demand, increased pressures. These adaptations are usually improvised. They can vary widely depending on the team and who's in charge.
In a healthcare setting, what happens when we're above the capacity of the beds? What happens when we don't have enough people on shift? In these situations, local teams need to adapt to try to figure out how to keep the system doing what it needs to deliver. These improvisations can vary. Different teams will deal with these pressures in different ways.
I guess the paper says that in the short-term, teams figure out how to adapt and cope with these problems for a few hours, a few shifts, a few weeks, and with the expectation that things will become easier. However, if the pressures continue, then the paper talks about the deterioration in working conditions and deviation from practices. What we might talk about in safety science is this drift into failure could start to become a sustained adaptation through formal or informal working practices.
Drew: I think sometimes we forget when we hear terms like normalization of deviance from Diane Vaughan. Normalization of deviance drifts into failure. They sound like people just being slack. I think we sometimes forget that the drift isn't just accidental, I just don't care one day, and then I just start constantly working badly.
Under pressure people work around, under pressure people adapt, and then that pressure stays, so the change conditions stay. Eventually, those changing conditions just become the way we do work, which is not the way we originally wanted to do work, but it's just the way things are done now.
There’s some research that even says that even if the pressures don't continue, once you've pushed into that new pattern, that new pattern is just the normal. You've lost the previous high standard forever until someone actually finds some way to pull you back. You don't just bounce back once the pressure is gone.
David: I think that's a great point about not just bouncing back. The paper cites a review of 58 studies from eight countries that found that in the healthcare sector, workarounds were common in all settings studied. Essentially, 58 studies, 8 countries, workarounds are present in all settings. That's a fairly absolute conclusion from that research or literature review.
I guess the conclusion was while these workarounds may aid short-term productivity, like getting over the short-term pressures and tension, the studies found that they do actually pose a variety of threat to the patient. This idea that teams will adapt when they need to isn't without risk and unintended consequence, which is why we talk sometimes about this need for guided adaptability, not just adapt, because we need to adapt in a way that maintain safety. It appears that that's not actually the way that these adaptations occur.
Drew: They talk about threats to patients here, but I think it's also threats to staff members in a couple of ways. This isn't an original thought of mine. I apologize if the person who gave it to me is our listener, because I've just forgotten who I heard this from. They were talking about how the way we get resilience as an organization in times of stress is we rely on individuals to be resilient.
We're getting that extra adaptive capacity that the organization needs by burning out the people. What's the easiest way for an organization to get extra capacity? It's not to hire an extra person, it's for the people who are there to work harder and longer. What's the way to get flexibility, it's for people to do things outside their job description, to help each other out. But when they do that, for more than just a short period of time, it just burns everyone out. That's a direct physical harm to people to have that.
But there's also the moral injury when people are in a caring profession, and they can't provide the standard of care that they know is or believe to be the right standard. These people suffer because they can see the adverse consequences to the patients, but they just don't have the capacity to give those patients a higher standard, more attention.
David: I wouldn't mind just reading a paragraph out of this paper because I think it gives a sense of this idea that maybe things aren't actually going to get better in many organizations, and it frames the rest of the paper. This is a quote and a sighted quote.
"In the last round, which was 2014 to 2018, of mandatory French hospitals certification, reviewers found one or more areas of substandard care in over 60% of 2218 French hospitals. Poorly performing hospitals are given 3–12 months to resolve these problems.
However, in practice, for a variety of reasons, more than 10% of all French hospitals were unable to return to an acceptable standard within a year. In France and in other countries, plans are put in place to deal with poor performance, but sustained improvement may take years to achieve. Services, therefore, continue to run in an unsafe mode with local adaptations and features, but seldom with any planned attempt to manage the ongoing risk."
Drew, my conclusion from this is that at any point in time—I assume that French healthcare will be no different to other healthcare and many other industries—that even when we find problems in two-thirds of all of our activities over a 12 month period, there's still a significant portion of things that we've been unable to fix, unable to resolve, even to an acceptable standard.
Drew: It's one thing when an audit tells you that that sign needs to be positioned three inches higher. Someone can come along, move the sign, hammer it back in, and you fix the problem. An audit tells you you have a deficient management system. You don't fix a management system just by rewriting a document. You consistently improve all of your safety investigations just by writing a new procedure for investigations.
The organization's going to be operating below the standard you want. It takes a long time to fix some of these things. That assumes that you've got the spare capacity and the resources to put into improvement rather than just using up all the capacity just to run things as they're being run now.
David: In this challenge, there's a lot of effort that goes into improving safety activities, and that can draw on the same resources that go into delivering the primary functions of the organization and imposing another goal conflict or trade-off in those operational teams, which is if I've already got this adaptation, this frontline work around, then they're already working beyond capacity. Where are they going to find the resources to address some of these more systemic issues? I think that goes to the heart of the paper as well. Are you ready to go to the next section we go on?
Drew: Yeah, let's go on to the next section. Our readers may remember from the Amalberti episode, immediately followed by the Rasmussen episode, that Amalberti is a bit of a fan of Rasmussen, and pretty much can't explain anything in safety without drawing upon Rasmussen's work.
That's exactly what he does here. He leaps into this with a brief reminder of what we talked about in episode 86, Rasmussen's paper on risk management in a dynamic society.
In particular, there's a bit he represents. There's no perfect frozen state of safety. Safety is all about anticipation and adjustment to manage risks. What he's really talking about is this dynamic process. But if you just do that in the short-term, that's not going to fix things in the longer term.
Dynamic adjustment might explain how people handle crises, but how do they handle long periods? That's where that title, innovation is not enough, he doesn't believe that this short-term adaptability of people can solve the problem of systemic under-resourcing or systemically falling below standards. He tries to then set out a more practical framework for how you operate in these conditions. David, do you want to take us through the different strategies?
David: I think I might just share a little bit more about this idea of improvisation is not enough, because this paper actually talks specifically about Safety II. Amalberti, I think that they go on and say that this is a dynamic problem. Safety II suggests that in a high-risk industry. We need the ingenuity and adaptability of people to basically fix the problems in the system. They probably shot over the bow at maybe some of these other authors and say—
Drew: I can see the quote you're thinking about here, David.
David: It says, “We may have many elegant descriptions of the resourcefulness, the ingenuity of human beings in coping with hazard and coping with crisis.” However, while we have these descriptions of success, none of these authors share how often these improvised solutions have been successful versus unsuccessful.
These short-term fixes are adaptive. They impede the development of longer-term solutions. They can add risk to the system by creating new processes that are not known to managers and they're not known to other members of the team, they can add additional steps into a process, and in the longer-term, like you said earlier, normalize these deviations from standard practice.
I guess, most importantly to these authors, the existing literature on Safety II offers little guidance as to how we might best prepare and support people and organizations to manage these expected pressures and crises. It feels like Amalberti is coming around a little bit in this literature. Maybe Nancy Levington came into this literature when she talked about Safety III.
Drew: It's almost like he's saying, stop telling me about resilience. Stop telling me about adaptive capacity. Tell me how we actually do that stuff.
David: Also, we'll talk about these principles soon. Tell me how we manage the things that we know are there, not necessarily start to prepare for these imaginary things that might happen. If we've got all these pressures, tensions, goal conflicts, and trade-offs in our system, tell us how we're going to manage those specifically, not necessarily how we might manage some potential future thing. We know it's always going to be both, but I think this paper is trying to refocus where the organization's effort and attention is directed.
Drew, the authors outlined four principles. Maybe I'll mention each of these briefly and then just ask you for a comment on each if you're happy with that.
Drew: Okay, sounds good.
David: The first principle they talk about would be, we must give up hope of waiting for things to return to normal so we can continue to innovate and improve our system. However, we should face the fact that unsafe practices, risks, pressures, tension, and dangerous conditions are always going to be present in our operations. There's no real point in hoping for things to end up in this nice, perfect, stable, normal, operational space.
Drew: I don't know that I've got a comment to make on that one, David. I think that's almost like a framing device.
David: Right, we framed it like that. The second principle they say is we accept that we can never eliminate all risks and hazards. There's nothing wrong with eliminating risks, where this is feasible and impossible. We need to balance these preventative actions with a wider portfolio and safety strategy that is explicitly aimed at managing dynamic threats and pressures.
Drew: I like this idea that we tend to set safety up almost as a standard of perfection. I guess we don't expect people to achieve it all the time, but we expect those deviations to be rare and correctable. I think what Amalberti's hinting at here is that if we set that standard too high, then failure to meet it is routine, and that's a problem. Because once people know that they're below the standard, they don't know what is genuinely acceptable.
That's always the problem with very ambitious goals. Sure, you can say that they're aspirational, but then it doesn't tell you what the bottom line is. If you're not actually expected to meet the rule, what rules are you expected to follow? I think we're just setting people up for failure when we create these rules that people can't follow. It's not fair to force someone into a trade off, where they've got to decide either follow the rule or get the job done, and we expect them to do both.
Amalberti is going to touch on this as we go. I think it is reasonable for the people who do set the rules to actually make those trade-offs. Rather than writing aspirational rules, they should actually decide, okay, what genuinely is acceptable and achievable given the resources that we have?
David: That's a great point. The third principle following that, that they've talked about is that our principal focus should be on expected problems and hazards, which I alluded to earlier. Even though the existing literature on adaptation focuses on the management of surprises and unexpected problems, there are a lot of things that we know about.
In this hospital setting, the pressures on beds, staffing, equipment, sick patients, these things aren't unexpected. We know that they will happen. They're entirely familiar. These situations are quite different from sudden, unexpected, and unusual crises, in that they're the focus of much of the literature. If there are things that we know could challenge our operational systems, resources, and performance, then we should invest in the capabilities and strategies to deal with those.
Drew: I don't know about you, but I was reminded of Griffith's rules for when you can ask for an extension on an assignment. If you're a part time student, you're not allowed to just say, I was busy at work, because we know that people who have jobs are going to be busy at work. You might not know exactly when you're going to be busy or how you're going to be busy, but you do know that if you're trying to mix work and study, there are times when you're going to be busy.
Amalberti is saying that we've got these, you could call them like predictable surprises or certain uncertainties. Things that we don't know exactly when it's going to happen, but we do know that we're regularly going to face these types of things, so we can plan ahead for them. And we can have ideas in advance about what we're going to do when we're in these things. They shouldn't be unexpected crises.
The other thing he's like aiming out there is this is really quite different from this abstract resilience, where you just have random capacity for unknown things. It's also different from very explicit hazard management that we do, where we've got very precise known events that we put controls around. This is anticipation of things that are reasonably expected to happen. Because they've happened before, they'll happen again. We're not trying to prevent them, we're not trying to control them, we're trying to be ready to deal with them.
David: It's like the known unknowns. They're the things that we understand but we're not necessarily aware of when they will be presented. But having strategies is a sensible thing to do.
Finally, the fourth principle, we must acknowledge from the start that the management of risk when an entire unit or an organization is stressed, requires engagement and action at all managerial levels. Negotiating new priorities and strategies in a strict organization requires coordinated action between executives, middle management, and frontline staff.
It's this idea that we can't just let these pressures hit the frontline and expect the frontline to adapt, absorb, and manage these pressures. These pressures and tensions in the organization need to be understood, managed, and responded to at all levels of the business.
I think this is particularly true in healthcare, where maybe this paper was highlighting that a lot of these pressures that are placed in (say) a hospital health care setting, are expected to be dealt with by the staff on the ward as opposed to by the organization.
Drew: Thanks for that explanation, David, because I was actually struggling to interpret this point. It sounded to me a little bit like a motherhood statement. Management of risk needs to involve people at all levels. We need to communicate well, we need comprehensive training.
If the alternative to that is just trust the frontline to adapt, then I think this is actually very clear. It particularly then applies to these goal conflicts. Goal conflicts can't be something that you're just continuing to delegate until they hit someone who can no longer delegate it.
Sometimes people higher in the organization need to make the decision about the trade-off and need to say, right, okay, this is what we're going to sacrifice. This is what we're going to prioritize, and then other people can just follow that instruction instead of being left to reconcile impossible demands.
David: This paper then goes on from that statement about all levels and those four principles to talk about what training for managing organizational threats look like at these three levels. There were a few examples from other industries that I don't think we'll talk about, because I wasn't overly impressed by some of those examples that were presented.
In terms of this description of what training might look like, the author suggested at the executive level, there should be a focus on how management negotiates between these competing priorities, like particularly safety and other objectives in both the short- and the longer-term.
The executive cannot and should not simply just say safety is the most important over other domains. They call out that this is a naive approach, which is often made, and that short-term impact on safety margins in response to other pressures can be accepted but only if they're actively managed.
This talks about maybe developing senior executives in an organization to have a far more deeply-understood discussion, decision around short-term trade offs, long-term trade offs, where safety risk is being borrowed for financial performance, how to provide additional resources and controls around those risks, rather than just say, do this and make sure that you always do it safely.
Drew: I just want to repeat that quote because I think it is both heresy and absolutely true. This paper literally says, a short-term impact on safety margins in response to financial pressures can be accepted. Of course it can, of course it is. But if we deny that, then we're never going to actively manage it or communicate it.
That's what the problem is. We implicitly understand that there's going to be an impact on safety, but we don't clearly say to the organization, we are accepting this extra risk, this is how we're going to accept it, and these are the limits of the extra risk that we're going to accept.
David: At a point in my career, I was instructed in response to financial pressures on the organization to reduce safety staff and costs by 60%, and to ensure that there was no increase in safety risk to the organization as to absolute statements.
Drew: That would be great if 60% of your staff were currently having no impact on safety. Not so good if any of them were doing any work at all.
David: That's a good point. All right, enough said.
Middle managers, this is sometimes referred to as the clay layer in organizations (I guess) where the transition happens between the senior and the frontline senior management, the frontline in both directions. I guess this paper says that middle managers need to act as mediators or buffers between the frontline and the executive, exactly as that.
They need to have a good sense of the real conditions on the frontline. It's like a real finger on the pulse of work has done. And they need to have this portfolio of possible interventions that can be deployed at times of high workload or other pressures. If we've got an expected pressure that can arise, we need to have a portfolio of interventions. I guess for them as middle managers, a critical task for them is to be clear about what’s standard to absolute.
In healthcare, they talk about handwashing. No matter how much pressure gets into the system, we're not really going to tell frontline people that it’s no longer acceptable now for them to no longer wash their hands, which can be relaxed. The example I give here is the timing and frequency of observations of vital signs. Maybe I can't get patients to check their vitals every 30 minutes and maybe 45 or 50 minutes is okay when we're still triaging people in A&E.
This is an explicit managed adjustment to the pressures that's like this is infinitely preferable to this general degradation of standards. The thing is if you are going to have a degradation of standards, at least choose the ones that you are okay to degrade and which ones you're not.
Drew: David, you've actually spent more time in railways than I am, so correct me if I'm getting the explanation wrong. This struck me exactly as the way railways handle degraded operations. Ideally, we'd want every time all of our signals aren't working or something like that, we would want to shut down the railway, but we can't do that. Instead, they've got very clear rules and expectations about how you go.
You tell everyone we are now operating in degraded mode, and all the drivers know what that means. That means that you come to a signal, you stop for a certain amount of time, you proceed at a certain speed. If you have visibility, you're allowed to go up to a certain speed. If you don't have visibility, you have to go down to a particular speed.
Even though it's less safe than normal, and everyone knows it's less safe than normal and there's a higher chance of accidents, it's not an open slather. People still have rules and expectations that they're supposed to follow in that degraded mode. They don't just decide for themselves. Things aren't working, I'll do what I can. Everyone's still operating under a set of very clear expectations that can't degrade further.
David: I think there's an acceptance that all operations operate in a degraded state. Nothing's ever as perfect as they were when they were first built or first designed.
I went to aircraft here, because there's an MEL, a minimum equipment list in aircraft. It's always assuming that certain equipment on the aircraft isn't quite right. An instrument will be out or something won't be there, but there's this minimum equipment list. It's going, well, you can do without this particular function, you can do without this particular piece of equipment, but here's the list that you can't do without.
I guess that's what this paper's saying. If your businesses are under pressure, under resource constraint, we know that teams are going to adapt, we know that shortcuts are going to be taken, we know that standards are going to drop. Be honest with yourselves and your organization, particularly middle management in this day, and work with your people to be clear on which standards actually can't lift and which can, either for a short period of time or for a long period of time.
Drew: If you're in an industry that's not so high tech, if something like construction, you could think of this in terms of the more golden rules you have, the less they're really golden rules and more just aspirations. You can have aspirations by all means. You'll have these things we would like to achieve, this is the standard of safety we want, and then also have this lower setting that no matter what, these are the things that we never, ever break even when we don't have enough time and attention to do the other things.
David: It's this idea that if everything's a priority, then nothing's a priority, particularly if you throw all of those things at a resource constrained system. They did go on to talk at the frontline level, particularly frontline leadership. You should be able to be provided with a range of simple compensator strategies, particularly in the short-term.
In healthcare, they use huddles at each work shift, where they can adjust roles and priorities to best adapt to the immediate pressures and concerns. A shift start huddle, which is like, okay, what's going on now? Gee, there are only five of us. Okay, we've got 15 patients. How are we going to actually adjust this? What's going on for this shift?
That's a dynamic work organization that can happen on the frontline at a shift level. People need the skills to be able to collaborate and work that out with their team, and also know what limits of their adaptation to maintain an acceptable standard, even if it's not the ideal standard, let alone the aspirational standard.
Let's move on. I'll let you talk to this one, Drew. The authors, I guess, being good psychologists and safety scientists, put a small section in here on a research and development agenda which I thought you'd quite like. The first part they just talk about doing a whole bunch of descriptive studies.
In any industry, they try to find out what are the common types of pressures? What are the degraded conditions? How are they dealt with by organizations? Can we create a taxonomy of these pressures? A bunch of strategies. I love this agenda of actually describing the problem first.
Drew: Often, when I'm talking about research, I talk about describing just because I love descriptive research. Amalbertie is clearly thinking more in problem-solving mode here, but he's almost like following our manifesto to the letter before we wrote the manifesto.
He's basically saying, describe first. Initially, it's a descriptive exercise with a view to identifying a set of strategies and then testing out those strategies. He's saying, don't try to just solve this problem. Don't come up with solutions. Describe what's going on, describe what people are already doing to adapt, describe what people are trying, come up with some strategies that are a blend of what people are already doing, things that seem like they might work, and test out those strategies.
I love that as a research program that anyone could pick up that approach in any industry (I think) and do a project just with a view to understanding degraded operations in this industry.
In fact, one of my students, Angie Galbraith actually did quite a similar project looking at coal mines in adverse weather. She was looking particularly at the big trucks that go on the circuit, looking at just how adverse weather is one of those things that it's a known unknown or an unknown known. It's something that you know is going to happen, you just don't know when it's going to happen.
How do people adapt? When do they adjust? How do they decide how to stop? Describe that first and then turn that into strategies and guidance for other people.
David: I like both the descriptive as well as the practical nature of what they're suggesting that science can contribute. I guess the paper does after this research agenda, talk about this, and everything that's been spoken about and said, the authors anticipate some resistance to this shift in perspective away from this vision of absolute safety at all times towards this active dynamic management of risk. This question is that the resource should say, can we evolve organizations away from absolute safety as the only acceptable goal?
Drew: Saying we can anticipate some resistance to this shift in perspective is the most polite way of saying, we're going to get hate mail for writing this, aren't we?
David: I think in the context of patient safety and healthcare to talk about knowing and active compromises and trade offs around clinical care, I think they were treading carefully.
Drew: And particularly since they're not just talking about absolute safety in the sense of zero accidents. This isn't the zero harm versus not zero harm debate. They're actually saying we can't make the highest standard of care what we're aiming for. We as an organization cannot promise we will always do our best. We have to say, sometimes we won't. We need some second best options.
We need to actually write into our procedures, write into our management training what second best looks like, because having a best and a second best is far better than having a best and then just below that, a gradual slipping of standards that's always aspirational that's never actually coping with the realities of what the situation's like.
David: Drew, we might just conclude and go to some practical takeaways if you're okay with that. I guess in the conclusion of this paper, the authors say, most, if not all, industries are facing many more demands and are more complex than they were in this case 10 or 20 years ago. In many countries, there's this quality divide between the expected standards of practice, and what's actually currently being delivered, and potentially will be unable to be bridged in the foreseeable future.
I guess in healthcare, this is still one year before the pandemic. We of course need to continue to innovate. We need to improve the system, but this in itself may not be enough. I guess these authors, their core argument is, we actually need to stock. This is where we're at right now.
What are the pressures and tensions in our business today? What are the things that we expect to continue to be pressures and tensions for quite some time? And what are our deployable strategies and capabilities to deal with those risks and accept that it's not going to be perfect now or potentially into the future, but we're managing those important things as opposed to just letting our frontline teams try and figure it out indefinitely? Anything you want to conclude before we go to takeaway, Drew?
Drew: Let's put up some takeaways.
David: I'll kick us off and then I think I'll throw a couple in, you throw a couple in. For me, I guess it's just a continuation of that conclusion I just shared. We need to manage the situation and risks that we're currently facing in our organization. Focus our effort there, not focus our effort on promoting the ideal rules, the ideal standards for normal operating situations, but actually just understand and support your business to deal with the situation that they're currently in.
Drew: The second one is to recognize that all operations involve constant tensions and trade-offs. We need to train people how to think about that at different levels. Amalbertie spelt out in this paper, this is how you need to treat it as the executive level. This is how you need to think about it. Middle management, this is how you need to consider it at frontline management. Actually, equipping people both with the understanding that trade-offs are real and with some language and some strategies for how you respond when you need to make those trade-offs.
David: That's good. Like you said, understanding. Following from that point there, understand how work is adapting in response to these pressures and challenges. I guess David Wood says that, in particular, safety professionals need to be obsessed with understanding adaptation.
How are those pressures and challenges actually being faced by teams in the organization? What are the short- and the long-term risks and limits of the system as a result? What are those adaptations that you accept for a short period? What are those adaptations you accept for a long period? What are those adaptations that you need to undermine and need to figure out how to, I don't want to use the word correct, but how to adjust again into some acceptable limit?
Drew: Yup. Where possible, we want to codify this and make it explicit in the same way that we do for other safety standards. We can't do this for everything, but for the things that we can predict, the hazards that we can expect, the conditions that we know will inevitably come, we need to define what second best looks like.
When we are running short on time, what can we stop doing? What standards can we allow to slip? And which are the ones that need to be precise, enforced, meet the ideal even under the bad conditions?
David: Drew, I'll let you continue with the research takeaway as well.
Drew: Okay, let's have a takeaway for researchers. We do need research into the strategies that organizations use under these sorts of expected but unpredictable stresses. Great opportunity for anyone who wants to do a project into degraded operations, both what are degraded operations look like in this industry, and then what strategies do people use? What works well? What doesn't work well? What can we put forward tentatively to suggest to improve handling these known unknowns, these certain uncertainties?
David: I like that. I think that would be very, very fun and useful research to go in with this position that systems aren't perfect, risk is not absolutely eliminated, and how it seems to cope with the pressures, intentions that are on them. And can we make their lives a bit better with some well-researched, deployable strategies?
Drew: I was about to say, now is a great time to do that research, but I think we've made the point that anytime is a great time to do that research.
David: I think, yes. The times might be getting better and better into the future as well as things become more and more complex and more and more degraded, which it's a bit of a cynical position, so I apologize for that.
Drew, the question we asked this week was, what's the right strategy when we can't manage safety as well as we'd like to?
Drew: Amalbertie gives us a wrong answer and a few tentative right answers. He says the wrong answer is to just either insist on the high standard and blame the frontline workers when they can maintain it, or to trust too much in just letting people adapt.
He's given us a few suggestions, though, for how we can put in place more planned adaptation based on an understanding of what we are likely to face and how the organization needs to adapt at different levels to manage the trade-off.
David: Great. 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. Send any comments, questions, or ideas for future episodes to feedback@safetyofwork.com.