Book of the month: Just culture: balancing safety and accountability by Sidney Dekker (2nd edition)

Preamble

Last month, when I was looking for a picture of the book cover for "Just culture" I noticed that there was now a second edition. What to do? Having spent many hours reading and taking notes on the first edition I didn't want to start all over and I didn't have the time to read another book. So I decided to post my  comments on the first edition, buy the second edition and comment on that this month.

What's New?

5 years after the first book, Dekker wrote this second edition, which he claims is more logical and addresses the needs of organisations who continue to struggle in the creation of a Just Culture. He has also added some material on ethics and on caring for the second victim. A number of new narratives have been added which provide a good intro for the ensuing argument. Dekker also makes a good claim for the importance of stories; stories are often told from one point of view and take on a life of their own, e.g. "The doctor as a murderer" (p.119) Whichever story is more believable or acceptable or promoted will prevail.

What's good about this book?

Dekker does concentrate more on the organisational requirements of a Just Culture. Headings such as "Responding to Failure: The Organization"(p.6) and "Ask What is Responsible, Not Who is Responsible"(p.12) explore the problems faced by organisations. He has also written a new section on what a successful reporting system looks like (voluntary, non-punitive, protected) (p.58) The re-organisation makes the book easier to read and the argument more logical. The additions Dekker has made strengthen his case that safety and accountability must be balanced.

Dekker's addition of the second victim material (p.76-78) is timely. He explains how failure can be devastating to the professional involved and that criminalisation is neither just nor logical.

What's bad about this book?

Dekker's arguments around utility and utilitarianism (p.2) are simplistic. In essence, he argues that utilitarianism would support the removal of a supposed unsafe worker as this would benefit the majority and harm only one person. However utilitarianism does not mean short-termism and the removal of the worker would be considered within a longer-term "greater good". Utilitarianism then would probably not support the removal of this worker for a supposed infarction for all the reasons that Dekker discusses in this book.

Final thoughts

This is a better book than the first edition. It is more logically organised and the writing is clearer. The additions are useful and little has been removed (the book is now 13 pages longer). Should you buy the second edition rather than the first? Yes. If you already have the first edition should you buy the second? Probably not, although you might want to loan it from your medical library.

Constructive Alignment in Simulation: Why we don’t simply replicate what happened in the resus room last week (Dr Al May, Educational Co-ordinator)

Introduction

I have met many people in the simulation world.  Some have been just starting off their journey and some have many years of experience under their belts.  Something that often comes as a surprise to many of those people is my idea that there is more to writing a “scenario” than it first seems.  Everyone is obsessed with how to debrief and often the scenario design and running itself is considered easy (or not considered at all!).  I would like to use this month’s blog to try and explain why I think what I think about the importance of “scenario” design and running. 

-Debriefing isn’t the only tricky part

Learners learn

The first thing you may have noticed is that I have used quotation marks.  The reason for this is because what I am going to talk about is applicable to any of the activity part of simulation, which won’t always be an immersive scenario.  Everything comes down to Constructive Alignment (Biggs, J. Tang, C. (2011) Teaching for quality learning at university, Fourth Edition, Open University Press).  The “Constructive” part comes from constructivist theory; learners use the activity they participate in to construct learning as interpreted through their existing schemata.  One of the reasons that simulation can be so efficient, stems from the ability of the simulation faculty to have tight ultimate control over the activity of the learner. 

-Controlling the activity in simulation by specific design and running of a scenario gives the opportunity to (potentially!) control the resultant learning.

Start with the Learning Objectives

The “Alignment” part refers to aligning teaching and assessment to intended learning outcomes.  Applying this to simulation, alignment should apply to the entire simulation based learning event.  This means that the learning objectives are integral to the design, running and debriefing of the simulation.  The debriefing conversation could be likened to formative assessment in terms of observation of a performance gap and facilitated discussion around how to close that gap.  Ideally, this should be a facilitated self-assessment by the learners. A skilled debriefer will be able to direct the developmental conversation towards the intended learning objectives and this is much easier, and occurs more naturally when the entire activity has been designed around learning objectives. 

-Everything needs to be directed to achieving the learning objectives

Finish with the Learning Objectives

At SCSC, we almost universally use Take Home Messages as an assessment of the debriefed simulation we have delivered.  This final step of the learning event should answer the key question all educators must ask themselves “Did my simulated learning event deliver the learning outcomes I intended?”  We simply ask something like “So what have you learned that you are going to implement tomorrow?” at the end of each debriefing.  This is at the heart of constructivist theory; we only know what the learners have learned if we ask them (or test them!).  At this point, if they don’t say the learning objectives you intended back to you (in some form that demonstrates they’ve “got it”) then you haven’t done your job.  Something, somewhere along the line has gone wrong.  Either the scenario wasn’t designed around the objectives, or you didn’t run it tightly enough, or you didn’t manage to facilitate the debriefing conversation to cover the objectives.  If they haven’t got it at the end, its too late, there is no point in telling them what you think they should have learned from your simulation and debriefing.  If you are going to do that, why not just save a lot of time and money and just tell them stuff in a classroom.  This is similar to The Design Focused Evaluation Approach (Smith, C. D. (2008) Design-Focused Evaluation Assessment & Evaluation in Higher Education, 33(6), 631-645), which seeks learner feedback on the efficacy of the alignment between the intended learning outcomes and the teaching and learning activities students engage in during a course of study. 

-Use Take Home Messages to assess whether you have done your job

So how do I do it?

The first step in designing any scenario should be to set learning objectives appropriate to the target learners.  Although scenarios will often eventually resemble an actual real life scenario, it is important to resist the temptation simply to replicate the real event using simulation.  By setting learning objectives first, the simulated scenario can be designed efficiently and effectively without including any of the extraneous content that is not aligned to the objectives.  One way to do this is to design each state of the scenario around one simple learning objective.  The scenario is then run according to a set plan, moving through scenario states that are mapped to learning objectives.  The learner is guided through the scenario in a realistic way with prompts to help them perform.  The prompts are minor physiology or faculty actions aimed at moving the scenario on, in the direction you have pre-planned without breaking the immersive psychological fidelity.  The transition trigger for moving from one state to the next will be something that the learner must do in the scenario, which is often an action that is closely aligned to a learning objective.  If your participants think they are managing a patient with an allergic reaction for the whole scenario, it will be quite difficult and appear strange for you to try to debrief around managing a patient with concealed haemorrhage, which was what the learning was supposed to be about!  Avoid getting in to those situations by actively designing, driving and guiding the scenario along the lines it needs to go to cover the learning objectives. 

-Designing and Driving is an active process

Conclusion

This is why I think what I think about simulation design and running.  What do you think?

(Dr Al May, Educational Co-ordinator (Faculty Development))

SimTechDay 2013: Backseat drivers take centre stage by Scott Rudnicki-Bayne

Simulation Technicians, we take care of the equipment, set-up scenarios, make the “magic” work. We’re confederates and actors. We keep scenarios on track, the guys and gals who ride in the backseat of the simulation education wagon. The inaugural Scottish Clinical Simulation Centre’s (SCSC) SimTechDay on the 27th of November 2013 was our day to take centre stage. Techs and other interested parties from all over Scotland descended on Forth Valley Royal Hospital (FVRH) in Larbert to network, share ideas and thoughts, and hear from our guest presenters.

Guest presenters:

Mannequin maintenance by Ian White (Laerdal) 

• Ian White, a Laerdal Field Service Engineer, spoke about Laerdal’s commitment to improve after-sales and technical support, whilst also providing a few tips on maintaining the appearance of SimMan
• Kevin Stirling, Lecturer in Simulation, University of Dundee and Finance Director, ASPiH, talked about the role of ASPiH and how Scottish SimTech could integrate this meeting into the national networks of technicians and how ASPiH might be able to support this.
• Andrew Middleton, Scotia UK, who presented a piece on “Effective Video and Audio Recording for Simulation and Debrief” as well as showing us some new AV kit.
• Nick Gosling, from St. Georges Advanced Patient Simulator Centre, talked about Mobile Simulation Strategies, which gave us a lot to consider when taking our simulated patients out to “in-situ” or “mobile” scenarios.
• Sarah of Sarah’s Scars, who showed us how to create realistic-looking burns and an open wound with stage make-up, which will undoubtedly help increase the fidelity of our scenarios.

Presenters were available throughout the day to speak with us about any specific issues we had or developments we’d like to see.

What went well?

Don't do this at home

On reflection some aspects worked out really well. The support from presenters, attendees, sponsors and the SCSC team was fantastic. The ability to network and discuss issues with people in similar fields was both rewarding and motivating. Every presentation provided something for everyone: Ian White showed how to use talcum powder to improve the feel of, and minimise adhesive residue remaining on, SimMan's arms. Kevin Stirling initiated debate on a Tech Room at ASPiH.  Andrew Middleton provided individual support on SMOTs. Nick Gosling shared his huge wealth of knowledge and experience in simulation and Sarah made moulage look simple.

What were the challenges?

It's just a flesh wound.

There were also areas to improve upon, both in terms of organising and hosting an event like this. The programme was overpopulated for the time available, which meant some of the presentations were a bit rushed. Next time I’d try to provide time for the presentations rather than squeezing presentations into the time. I’d also consider using more rooms, for 2 reasons. Firstly, due to the overwhelming interest we sadly had to turn some people down who wished to attend. Secondly, the variety of techs (i.e. some university, some healthcare, some new to the role, some involved for years) had widely varying levels of knowledge and experience. So a larger number of attendees might give us the opportunity to have parallel sessions which could be tailored to suit the two distinct groups, where appropriate. I will also work on my skills as chair, e.g. formally introducing presenters and ensuring appreciation of their time and efforts is verbalised more clearly.  

Final thoughts

All in all a great day, focusing on some of the issues in our wide and varied job descriptions. A big thank you to all who attended, presented and helped make this event the success that it was. The feedback from the attendees was excellent and the planning has already started for SimTechDay 2014.

Scott Rudnicki-Bayne (SimTech, Scottish Clinical Simulation Centre, Larbert)

@5imTech1

It's not my fault, it's the drifting, complex system.

One of the explanations provided by people for not embracing a systems-based approach to incident investigation is that it allows the "bad" individual to escape punishment.

In their book "Crisis Management in Acute Care Settings", St Pierre, Hofinger, Buerschaper and Simon state:

"Misapplication of (Reason's) Swiss-cheese model can shift the blame backward from a 'blame-the-person' culture to a 'blame the system' culture. In its extremes, actors at the sharp end may be exculpated from responsibility"

The concern is that some individuals (Robert Wachter labels them the "incompetent, intoxicated or habitually careless clinicians or those unwilling to follow reasonable safety rules and standards") will not be held accountable. These deviants will blame "the system" for not stopping them earlier or not making it clear enough what a violation was or not training them better. Wachter's 2012 paper "Personal accountability in healthcare: searching for the right balance" argues that lines must be drawn to distinguish simple human mistakes from sub-standard performance.

In his book "Managing the risks of organisational accidents" James Reason also calls for the drawing of a line "between acceptable and unacceptable behaviour" and calls for the replacement of a "no-blame culture" with a "just culture".

Drawing the line

A number of people/organisations have proposed "line-drawing" mechanisms:

• David Marx: who says we must differentiate between "human error" (a slip/lapse), "at-risk behaviour" (wilful shortcuts) and "reckless behaviour" (substantial and unjustifiable risk)
• The English National Patient Safety Agency (NPSA): their decision tree asks us to carry out 4 tests (deliberate harm, incapacity, foresight, substitution)
• Leonard and Frankel: they consider 4 questions (impairment, deliberately unsafe, substitution, prior history)
• Reason (see figure)

James Reason's decision tree (although it looks more like a hedge)

Not "Where?" but "Who?"

In his book "Just Culture", Sidney Dekker argues very convincingly that the line is arbitrary and that the definitions are fuzzy. Without being able to travel back in time and then read the mind of the individual who was at the sharp end of the error how can we be 100% sure that an act was intended or not? Instead Dekker argues that the argument should centre around who gets to draw the line(s): peers, regulators or prosecutors.

Abolishing retrospective blame

One idea that may inform the argument is that it may not be sensible, effective or just to attempt to make "someone" accountable in retrospect once an error has been committed and a patient harmed. Therefore if errors are discovered as part of an adverse event analysis then these should not be used to "blame" individuals. However, as a corollary, it may be appropriate to make people accountable for their current and future behaviour. It is therefore just as important to have systems in place to check and enforce correct behaviours as it is to be able to analyse past events.

As an example, consider the surgical pause as described in the WHO safety checklist. It makes little sense to blame a person or a team for not completing the pause correctly in retrospect once an error has been reported. It is much better to seek an active safety culture which would pick up the fact that the pause is not being done correctly before a patient has been harmed. It is this proactive approach to safety which is still missing in many places.

This post concludes with Don Berwick's thoughts:

"Blame and punishment have no productive role in the scientifically proper pursuit of safety."

References:

• Berwick, D. Quoted at: http://www.england.nhs.uk/2013/12/12/never-events-news/
• Decker, S. Just Culture. Farnham, UK: Ashgate Publishing Ltd. 2007.
• Leonard MW, Frankel A. The path to safe and reliable healthcare. Patient Educ Couns 2010;80:288–92.
• NPSA http://www.ahrq.gov/professionals/quality-patient-safety/patient-safety-resources/resources/advances-in-patient-safety/vol4/Meadows.pdf
• Reason, JT. Managing the risks of organisational accidents. Aldershot, UK: Ashgate Publishing Co. 1997.  (Decision tree from: http://www.coloradofirecamp.com/just-culture/definitions-principles.htm)
• Wachter, RM. Personal accountability in healthcare: searching for the right balance. BMJ Qual Saf 2012;0:1–5.
• Wachter RM, Pronovost PJ. Balancing "no blame" with accountability in patient safety. N Engl J Med. 2009;361:1401-1406