When Things Go Right: Safety II in an Academic Emergency Department

This study explored variability and ED clinician proactive adaptations in order to enhance safety in the face of demands and pressures.

Data was based on a cross-sectional survey.

Background:

·???????? Patient safety, according to one agency, is defined as “the prevention of errors, injury, or other preventable harm and reduction of unnecessary harm”

·???????? They argue that these types of definitions have “generated a “find and fix” approach of retrospective detection and investigation of past events with a focus on reducing the burden of errors”

·???????? In this perspective, “safety” is defined by what it is not and is studied only in its absence – said to be a Safety-I lens [others have also called the absence of safety ‘unsafety’]

·???????? They argue that excessive reliance on this lens practically results in “drawing reductionist linear cause-and-effect relationships between error and outcome, assuming that deviation from standard practice introduces risk”

·???????? Moreover, this lens fails to adequately learn from the “vast majority of cases in which adaptive actions by clinicians actually keep patients safe”

·???????? “By examining only instances of undesired outcomes and noting clinician practice variability, we have historically failed to recognize that the same practice variability is present during daily expected work and also results in exceptional outcomes”

·???????? Therefore, adaptive practice variability is “not only ubiquitous but also necessary to deliver patient care in our complex healthcare system”

·???????? From a Safety-II lens, performance variability isn’t exceptional but is necessary, and “clinical experts making dynamic practice adjustments within a complex system serve as a vital source of flexibility”

·???????? Therefore, most of the time, “desirable outcomes are achieved despite complex systems and nuanced situations”

·???????? Next they talk about graceful extensibility, observing how variability isn’t necessary a source of potential harm in a system but can instead represent worker real-time adjustments to meet nuanced situations

·???????? Also efforts to excessively dampen variability can introduce more barriers to efficient work, hence introducing more workarounds and variability

·???????? It may also “inhibit graceful extensibility, thus negatively impacting most cases where expert adaptations keep patients safe”

Results

Overall, they found:

·???????? “Clinical team members report keeping patients safe through various forms of behavior adaptations, most frequently reporting collaboration beyond usual expected practice”

·???????? Workers bypassed policies when deemed necessary to meet physical and social needs of patients

·???????? Most respondents believed that existing policies matched their day-to-day work pretty well, and they felt supported in adapting to the needs of patients

·???????? “This work stands as a sampling of resilient, expertise-driven actions of frontline staff”

A number of different adaptive behaviours were documented. This included via collaboration with others, time-based interventions, physical harm prevention, patient/visitor interactions and more.

A number of instances of bypassing policies or workflows to keep patients safe were described. These bypasses were coded into four themes:

·???????? Bypassing policies to meet the physical needs of patients

·???????? Bypassing policies to meet the behavioural or social needs of patients

·???????? Communicating outside of normal routes to meet the needs of a situation

·???????? Bypassing usual technology-based workflows

Some of the respondents indicated experiences of negative emotion, like self-doubt, fear of repercussion or anxiety when strictly adhering to policies, bypassing policies or encountering situations where no policy existed.

Yes, some clinicians felt fear or anxiety when strictly adhering to policies. The authors suggest that “This may stem from the inherent sense that adapting to patient needs is an expected part of their daily work”.

Because of the unpredictable and “and therefore unprescribable nature of patient care in the ED”, this setting seems primed for performance variability and graceful extensibility.

While human variability has typically been viewed as a liability to be controlled, this data reported examples where “nuanced performance variability by healthcare team members contributed to patient safety”.

The research provided some examples of how clinicians adapted behaviours proactively to maximise patient safety and mitigate hazards.

Hence:

“we see that clinician adaptations mitigated the translation of hazards into harm, in contrast to the traditional perspective that would view practice adaptations as increasing the risk of harm”.

Interestingly, most respondents in the survey believed that the prescribed policies reasonably matched their day-to-day work, and they felt supported in adapting to the needs of patients and situations.

The authors highlight the “difficulty in measuring everyday work that results in desired outcomes”, since these events tend to be frequent, small-scale and unremarkable; hence often ignored. Therefore, the respondents in the survey may not be readily cognisant to all of their adaptations, and see the policies as reasonably congruent with their day to day performance.

The authors stated how many team members in their own ED environments “recognize core concepts of Safety II and graceful extensibility playing out in their everyday work, even if they do not know the theoretical background”.

Therefore, the key Safety-II underpinnings seem to have considerable support from clinicians, in at least how they practice day to day (rather than the theoretical ideas).

Also, while the authors expected a greater gap between work-as-imagined and work-as-done, clinicians generally observed a good congruence between them. This may also be due to the fact that since clinicians are so good at adaptation, any risks resulting from gaps” is likely mitigated by ED care team involvement in protocol implementation”. Although they say that the gap may have been reported to have been larger if they identified specific policies, rather than generalities.

Finally, they discuss the blending of approaches, and how the Safety-I perspective it also necessary to optimise patient safety. They argue that existing data analysis and learning has been biased around poor outcomes (incidents and near misses). This “creates a false perception of a linear correlation between practice variation and bad outcomes”.

This approach has omitted the collection and review of when clinicians act outside of protocol, and the conditions that drive these adaptations.

Authors: Boettcher, S., Aranda, J., Pavlic, A., Ladell, M., Williams, K. S., Wilbanks, M. D., & Jacobson, N. (2024). When Things Go Right: Safety II in an Academic Emergency Department.?Cureus,?16(9).