Thanks Peter, it's been an interesting session.

Thanks for all the great questions and I look forward to more conversations in the future!

For those of you interested in understanding more about our experience, this is a chapter we wrote about it (a later version was published in "Delivering Resilient Healthcare" by Hollnagel et al)

In some ways, disasters represent a fundamental challenge to our understanding of risk. We have focussed on managing small risks but taken for granted the bigger risks that can arise. This has been seen in the oil industry, military history etc and also in healthcare in past. I think we fell into the trap of thinking our modern society was immune to these risks but it has been clear we are not.

The next stage will be one of how we frame these disasters? What lessons will we learn? How much will we question our previous assumptions? Will we choose blame or learning? Will we value adaptation or see these events as aberrant?

Safety-II is sometimes portrayed as only "learning from what goes well". This really sells it short as it is a way of understanding that safety is something we have to create, it isn't the natural state to be defended.

You definitely see different things from different models. We initially saw that both were required but now we probably have a Safety-II worldview that informs all the tools we use and the way we think about issues. We still have checklists, procedures, investigations but they are focussed in a different way. It has changed everything but it is definitely not a battle between Safety-I and Safety-II.

Followup: I absolutely agree that it is not right to load all the adaptive capacity onto frontline workers. We need to see resilience as a system property, not just an issue of indiviuals and teams.

This is a really key point. You need to have adaptive capacity and "room for manoeuvre" before it is needed. This is a central tension - what is waste and what is "slack"?

I share your concerns that the lessons from the largest, most open demonstration of resilient performance will go unseen and we will once again go back to the industrial, efficiency focussed models of the past. It is up to all of us to ensure that we stop that happening.

We need to be "prepared and prepared to be unprepared" and that only happens with adaptive capacity.

We were lucky enough to have support from our local manager for simulation as part of "business as usual" rather than education. Seeing the team's reflecting on work as a way to improve their performance and care. One day in ICU is ~$8000 so our programme was relatively cheap, mainly just the time together. Once we started to see changes in the unit (over time) it became a given to continue.

The next issue is how do you roll out a way of looking at the world, understanding it will look different in different settings? This is the net challenge that I am partnering with HQSC to work on.

Hi Peter!
Definitely a soft launch! We changed the questions we asked and the way we approached change. It is more of a change in the way you see the world. By doing a "here is the latest and greatest thing" type launch we risk trying the same old "upgrade the components" approach, waiting for the next new thing.

This is a central issue in "engineering a culture" vs "culture is what emerges" as you know

Within our unit, we have made significant changes. This reflects a change in our underlying safety model. We are definitely learning much more about the way our system normally works, how the risks are dynamic and how the behaviours of our staff usually make sense.

However, we all exist in social structures that reinforce a certain way of looking at the world. The Health & Disability Commission, for example, is grounded in proportioning responsibility regarding individual practitioners and institutions. This creates the context for the way in which incidents are approached and reinforces the old linear accident models. We are working to change this wider context as I know Worksafe are.

Sounds perfect! We'll just sort out this pandemic thing and then time to move on. HQSC a great partner in this too.

Safety-II doesn't mean no rules, it means ensuring that the rules are followable. So, we are trying to dampen down unwanted variability but not suppress the ability to navigate the complexity of work.

This means having a deep understanding of the "work as done", by taking the time to talk to people about real work. Curiosity is a great way to build trust.

Absolutely!
Two issues: we have changed our focus from blame to learning "why did that seem the right thing to do at the time?" You get a much better understanding of the realities of daily work.

In terms of adaptive capacity then yes, this was a key to the way we coped with Whakaari/White Island and in our preparations for COVID. Both required a massive reorientation of the service in different ways but built on the previous work.

This is the future for all of us. The need to adapt may not be as ever-present as in ICU but technological and societal changes mean we all exist in a dynamic world. We need to move from just "doing things right" to "doing the right things", ensuring we are constantly updating our understanding of wider risks.

Context is everything!

Hi Marion!
I can't really speak for the MOH but I think that in healthcare, in general, we have tended to treat these as separate issues. There is a lot of "wellbeing" work going on but I am concerned that the connection between system performance and human wellbeing is not seen.

Working in an environment that supports real work and you feel valued is both a win for wellbeing and performance. Hoping for change by bringing in HF to healthcare

Love this question!
This is a really fascinating field and related to issues of power and psychological safety.

Healthcare has traditionally approached this through two ways: empowerment and encouraging people to "speak up".

However, empowerment may inadvertently send the message "I have the power and I choose to give it to you (until I decide not to)". What we need to be thinking about as leaders is how to create the space and expectation for contribution. This means realising we are situationally dependent on those we lead. I recommend Ed Schein's "Humble Inquiry" as well as anything by Amy Edmondson.

The second issue of "speaking up" is related. The message here is that staff should be responsible for overcoming the hierarchical barriers. In our team simulations, we found that when people know what was going on, knew what their role was and there was an expectation of active followership then the issue of "speaking up" disappeared.

We had been frustrated by a lack of progress with current models of safety and quality improvement. The mismatch between "Tayloristic" approaches and the realities of ICU was pretty clear; ask any intensive care person a question and the answer is always "it depends". It depends on the context and path to get there, there is never one right answer in all situations.

As part of understanding this, I read Hollnagel's "Resilient Healthcare" book and could immediately see the implications for much of what we were doing. We started to look at some of our issues differently and change how we approached them. It took 2-3 years to get the major changes in the ICU as this is a way of seeing the world, not a toolkit to be implemented.

I am now part of the Resilient Healthcare Network, an international collaboration of practitioners, researchers and academics interested in this area.

Thanks, Sheri. Great question!

Very often in healthcare, as I'm sure it is in other industries, there is a big focus on compliance. "Fill in the falls or pressure area assessment, fill in the early warning scores a daily care plan" etc. We risk sending a message that good work is just complying with rules. Ironically when something goes wrong, we say staff should have used their "critical reasoning" to see what was happening. We encourage passivity and then are surprised when people don't change when the unexpected happens.

We have spent several years building our teams to be focussed on adaptability: developing a shared understanding of risks, understanding what the expected course looks like so we can recognise when "not expected"; building psychological safety, active followership, routine debriefs focussed on 1% improvement etc. It has been several years of preparation work to build the capacity for resilient performance. Luckily we get to practice this every day!

This was initially rolled out via our simulation programme but a key aspect is that it is reinforced as "usual" by senior staff modelling curiosity about the realities of work for the team.

I've attached the framework we developed as part of this work and you can see the focus on adaptability to change as a core element.

Good morning all! It's great to get to "talk" to you all in these interesting times. Peter asked me to reflect on why we started exploring Safety-II approaches in the Middlemore Critical Care.

Healthcare faces some pretty big challenges: the increasing complexity of the patients' conditions and technology; increasing demands for greater and more timely care that meets the needs of individuals and their families; limited resources including global competition for skilled staff; staff wellbeing and high rates of burnout (especially in high acuity areas like ICU); and finally the limited ability of current approaches have had in improving these issues.

We often treat these as separate issues when they may really be different aspects of the same problem: complexity.

Intensive care is an extreme example of these issues: we look after the sickest patients from our communities and around the country; our world is dynamic (think White Island/Whakaari), ambiguous (often having to act before we are certain), and complex (with many interactions between the disease, patient, technology and healthcare system). Models based on compliance or "one size fit all" are fundamentally mismatched to the nature of this world.

The central issue is therefore about "how people cope with complexity" and how can we support them to do so. This viewpoint, consistent with Safety-II, allows us to align patient-experience, quality and safety, staff wellbeing and overall system performance in a way that is cohesive and brings the different areas together. By designing for "as much as possible to go right" we can not only remove some of the classic tensions (safety vs production) but we can also engage with our staff with the key insights they bring about ways to improve both their work and the care we provide.

It has changed almost every aspect of how we work (training, incident review, intervention design etc) and transformed the unit to one focussed on adaptability in these uncertain times.