=Paper=
{{Paper
|id=None
|storemode=property
|title=Resilience as individual adaptation: preliminary analysis of a hospital dispensary
|pdfUrl=https://ceur-ws.org/Vol-696/paper7.pdf
|volume=Vol-696
}}
==Resilience as individual adaptation: preliminary analysis of a hospital dispensary==
https://ceur-ws.org/Vol-696/paper7.pdf
CEUR Proceedings 4th Workshop HCP Human Centered Processes, February 10-11, 2011
Resilience as Individual Adaptation:
Preliminary Analysis of a Hospital Dispensary
Mark-Alexander Sujan (m-a.sujan@warwick.ac.uk)
Warwick Medical School
University of Warwick, Coventry CV4 7AL, UK
Simone Pozzi (simone.pozzi@dblue.it), Carlo Valbonesi (carlo.valbonesi@dblue.it)
Deep Blue Human Factors Research & Consulting
Rome, Italy
Catherine Ingram (Catherine.Ingram@hhtr.nhs.uk)
Hereford Hospitals NHS Trust
Hereford, UK
Abstract Compared to more traditional approaches to safety,
resilience engineering stresses the need to analyse in an
This study aimed to identify and to describe common integrated way what makes a system work as well as
forms of resilient behavior in a hospital dispensary. 16 what causes it to fail. Success and failure stem from the
narratives submitted by pharmacy staff were analyzed same processes, or system characteristics, hence they
qualitatively. Common forms of resilient behavior
identified include: personal negotiation, creating shared
should be understood as generated by the same system
awareness, thinking ahead, seeking help, prioritizing properties.
activities. Most of these forms of behavior rely on personal Hollnagel particularly criticises all those approaches
initiative and experience and there is little organizational that try to curb or constrain human variability as a source
awareness and support. of disturbances. According to Hollnagel, people and
organisations always need to adjust flexibly to the
Introduction operating conditions, in order to make optimal use of
Hollnagel (Hollnagel, 2006a) argues that “to understand finite resources and time. Human variability is the core
how failure sometimes happens, one must first driver of such a flexible adjustment, so it should not be
understand how success is obtained”. Such a statement eliminated, but rather seen as extremely useful. It may
well represents the line of thought known as Resilience also engender failure, but most of the times it ensures that
Engineering, advocated by Hollnagel himself, Nancy the system adjusts successfully to internal or external
Levenson, David Woods and Sidney Dekker among disturbances and keeps operating at a satisfying level of
others (Hollnagel, Woods, & Leveson, 2006). These performance.
authors maintain that safety is better managed by also In healthcare, where Reason’s (Reason, 1997) model of
focusing on what the system does well, rather than organizational accidents has been highly influential in
simply concentrating on the failures. Even though shaping many patient safety initiatives, it is rather
resilience has been defined in many different ways, uncommon to look at how people successfully cope with
depending on the main emphasis the authors want to disturbances and disruptions. A literature search in
convey, a commonly agreed definition may read as Quality & Safety in Health Care, one of the leading
follows: “the intrinsic ability of an organisation (system) journals dealing with patient safety, produced few hits on
to maintain or regain a dynamically stable state, which the topic of resilience. There is, therefore, a need for
allows it to continue operations after a major mishap empirical studies that explore and describe resilience in
and/or in the presence of a continuous stress” (Hollnagel, healthcare settings.
2006b, p. 16). Cook et al. (Cook, Render, & Woods, 2000) introduce
Resilience engineering thus emphasises the ability of a the useful notion of gaps or discontinuities in care. Due
system to maintain control even when faced with (major) to the structural characteristics of healthcare and the
disruptive events. It also specifies how such an ability intrinsic complexity, a major activity of healthcare
should be able to cope with both internal and external workers is to cope with the resulting gaps and
events, namely with its internal variability (e.g. technical discontinuities in care. In other words, normal successful
failures, human action, etc.) and stress engendered by everyday performance is not the result of prudent system
external variability (e.g. weather conditions, problems of and safety barrier design only, but rather of the technical
nearby systems, etc.). Some authors (Woods, 2006) also work of people within the system who anticipate, detect
highlight how a resilient system should to able to adapt and bridge the various gaps they encounter. It is
not only to known disturbances, but also to problems important to bear in mind the emphasis here on everyday
outside of the “design envelope”, that is to problems that performance, rather than on failure. These activities
were not anticipated by system designers and happen in a intended to deal with complexity and bridging the gaps
totally unexpected manner (or timing). are so tightly interwoven with other technical work that
48
CEUR Proceedings 4th Workshop HCP Human Centered Processes, February 10-11, 2011
often they cannot be distinguished from it (Cook et al., Analysis
2000). The submitted narratives were analyzed qualitatively
In a previous paper (Pasquini, Pozzi, Save, & Sujan, using the Nvivo software package. The software tool
2010), we elaborated a model where risk factors were facilitates deep analysis of non-numerical or unstructured
pushing the system out of control, counter-acted by data, such as narratives and interviews. It supports a
resilient behaviours (or other resilient features). The range of qualitative research methods including grounded
model was based on the authors’ experience of how a real theory, the approach taken in this project. The
safety critical system works and achieves its functions, preliminary analysis and coding were done
but lacked a detailed description of the resilient collaboratively by a domain expert (CI) and a safety
characteristics. expert (MAS). The codes were generated from the data.
Emerging themes were discussed in a review meeting
with human factors experts (SP, CV).
The last part of the analysis was conducted separately
by two analysts (MAS, CV) on the basis of an emerging
research hypothesis. The two analysts later compared the
results of their work with a third HF expert (SP), (i) to
agree on the coding of specific events and (ii) to establish
a shared set of codes (which will be used to continue the
analysis in following research studies).
The purpose of the above process was to prevent
idiosyncratic interpretations by a single analyst by
Figure 1. The variability of a system needs to be involving three partially separate strands of work. To
maintained under control, by counterbalancing counterbalance potential divergence between the three
disturbance factors (upward arrows) with resilient analysts, the initial research hypothesis served as a guide
behaviours (downward arrows). to orient the separate strands in a common direction. A
third requirement was not to spoil the richness of field
In this paper we present preliminary results of a data by imposing an overly strict a priori interpretation –
qualitative analysis of resilient forms of behavior in a for instance, ideas deriving from previous researches or
hospital dispensary. The next section describes the from the theoretical framework.
setting, as well as the data collection and analysis The theoretical resilience engineering framework was
methods used. Then the results of the analysis are brought into play only as a common theoretical
understanding of the data, but did not orient the
presented with empirical examples. The concluding
identification of the codes themselves. As stated above,
section outlines further work. the codes were generated by the data, with an empirical
bottom-up approach. A detailed comparison of the results
Methods with “standard” resilience engineering frameworks will
be performed in follow-up studies.
Setting
The hospital is a main provider of acute services for the Results
West of England and parts of Wales and has a capacity of The preliminary analysis identified three main themes,
259 inpatient beds. The pharmacy department employs under which most of the codes could be clustered:
50 staff, the majority of which work in the dispensary on disturbances, feelings of frustration, coping strategies.
a rotational basis, and there are 8 staff who are based These are explained below.
permanently in the dispensary.
Disturbances
Data collection Situations that were described as disturbed or challenging
Data was collected as part of the Health Foundation Safer in the narratives were coded as disturbance. This node
Clinical Systems Program (SCS). The program was arose after merging initially used nodes such as
commissioned in 2008 and involves 4 NHS organizations concurrent activities, excessive demand and absences
with the aim of developing systems approaches to into a single higher-level node. A disturbance can range
delivering more reliable and safer care. The data used in from mild or frequently recurring disturbances to a crisis
this paper was collected by inviting staff in the situation. Disturbances can be caused by internal
dispensary in one hospital to submit narratives about (absence due to annual leave, people engaged in multiple
something that caused them hassle during the previous activities) or external (absence due to sickness, external
week. Staff were encouraged to use their own language demand) factors. The immediate consequence of
and style in order to promote the idea of the narratives
disturbances is a rise in demand and queues building up.
being a kind of reflective “safety diary”. No further
Disturbances require adaptation and coping strategies.
guidance or restrictions were provided and the submitted
narratives varied in length from one paragraph to 5 pages. Depending on the success of these coping strategies the
In the first instance, 16 narratives were submitted (by 13 consequences of disturbances may be negligible
out of 34 members of staff that had been approached). (successful adaptation), or cause delays and lead to
frustration. An example is provided below:
49
CEUR Proceedings 4th Workshop HCP Human Centered Processes, February 10-11, 2011
“We were short staffed due to sickness and annual subject clearly anticipates a potential problem for the
leave, the phones never stopped [...] a technician rang system (e.g. workload is going to increase). S/he then
from the ward he wanted a couple of green profiles warns the Clinical Director, in order to ensure that
podding down to [their ward] - a 2 second job, well so I required actions are taken before the workload increase
thought. When I got there the draw was bare, someone actually happens.
had obviously used the last one without photocopying “About 11am [the Senior Dispensary Assistant] came
anymore. So, I had to leave an already short staffed to let me know that the pharmacist had a lot of work in
dispensary to go to the copier to copy some green his tray and the dispensing [Dispensary Assistants] were
profiles which takes a little longer because you have to looking for work to do. I noticed that one of the ward-
swap the paper in the machine etc.” (Example 1: based technicians was dispensing and so asked [the
disturbance) Senior Dispensary Assistant] to see if she could check
In this narrative, the subject points at internal any patient’s own drugs to help the pharmacist.”
disturbances (e.g. annual leaves, a phone call, the draw (Example 4: personal negotiation)
being empty) and to external ones (e.g. sick leaves), In this case, people report to fellow colleagues that the
showing their impact on an already stressed system “we workload is unevenly balanced between two roles (i.e.
were short staffed […] I had to leave an already short the pharmacist and the dispensing Dispensary Assistants),
staffed dispensary”. making sure that a shared awareness exists of the current
system status. As a consequence to that, people flexibly
Feelings of frustration re-adjust their roles, by shifting one technician to a
Feelings of frustration express the personal emotional support role for the pharmacist.
reaction to working situations and fall under the larger Other strategies to deal with disturbances include
category of consequences of disturbances. However, in prioritization of activities and seeking help from others /
order to emphasise the emotive nature of frustration, it offering help. Coping strategies can be proactive (e.g.
was decided to keep it as a separate category. freeing up resources in case they will be needed later) or
Frustration can vary in its intensity ranging from a reactive (prioritizing activities). Prioritization can be
feeling that one isn’t getting anywhere to being very done based on pre-defined work flows or ad-hoc. Seeking
annoyed and upset with oneself or colleagues. help can differ in terms of the type of help that is
Frustration expresses (is caused by) a dissatisfaction with required, e.g. extension of one’s own capabilities, a
the performance or behaviour of others, of the different set of skills.
organisation or with oneself. The consequences of “In the afternoon a nurse came to the hatch and said
frustration are not clear from the narratives, but may have that there was water on the floor in the waiting area, I
a negative impact on the coping strategies outlined was really busy catching up with my databasing and
below, including unwillingness to communicate with filing but as there were quite a number of people waiting
others (ask for help, provide help) and not sharing and in view of patient safety I left my post to go and mop
information. An example from the narratives: it up. It looked like a cup of water from the water
“Phoned [Location A] with a query about a machine had been knocked over.” (Example 5: ad-hoc
prescription, which I had to explain the full story to 3 prioritisation)
nurses, only to find by the third nurse that the patient had “People off work sick and then there were pre-
already gone home and the prescription was no longer arranged meetings to go to. [...] We managed to keep up
required – very frustrating!”. (Example 2: feeling with our work flow of fast track items and out patients
frustrated) (this hadn’t given the clinical check pharmacist a chance
In this example, the individual reports the feeling of to do standard track).” (Example 6: prioritisation based
frustration following an episode of time consuming on pre-defined workflows)
communication with different individuals at another “About 1.20pm ward based technician came to ask if
location that turned out to be needless since the I’d like some help ACTing as there wasn’t much work in
information was no longer required. the dispensing tray - I said yes please!!” (Example 7:
offering help)
Coping strategies The three examples show three different ways of
prioritising work to optimise time and resources. In the
Strategies to cope with disturbances often involve
first case, optimisation takes the form of interrupting the
personal negotiation and sharing of information about the
current activity to carry out a more urgent one (no harm
current situation in order to create a shared awareness:
can arise from temporarily interrupting the database
“Lead technician made me aware that the CT scanner
work, while someone can slip on the water). In the
had been down and there were 37 patients waiting for an
second case, a staff shortage (caused by sick leaves) is
appointment, if the scanner was fixed later today, we may
managed by relying on two pre-defined workflows (fast
see an impact. This would increase the workload on an
track items and standard track items). The third case is a
already busy day. I told lead technician that I’d chase
case of help self-offer, based on the recognition that
this up with [the Clinical Director] to find out if there
workload is currently low in the ward, but it may be
was anything we could do to prepare for this.” (Example
instead high in the dispensary.
3: creating shared awareness)
In this case the disturbance is an internal one (e.g. a
technical failure like the CT scanner being down) and the
50
CEUR Proceedings 4th Workshop HCP Human Centered Processes, February 10-11, 2011
Conclusion Aldershot, UK; Burlington, VT: Ashgate
The hospital dispensary is an environment that faces Publishing Limited.
challenges both due to internal as well as external factors. Pasquini, A., Pozzi, S., Save, L., & Sujan, M.-A. (2010).
Internal factors include issues such as staff on annual Requisites for Successful Incident Reporting in
leave, staff being unavailable due their being out on Resilient Organisations. In E. Hollnagel, D.
wards or being engaged in multiple activities. External Woods & J. Wreathall (Eds.), Resilience
factors relate predominantly to the nature of the work Engineering in Practice: A Guidebook.
(incoming prescriptions) that is dependent on work flows Aldershot, UK: Ashgate.
in other parts of the hospital. The narratives describe an Reason, J. (1997). Managing the risks of organizational
environment that is frequently very busy and stretched, accidents. Hampshire, UK: Ashgate Publishing
but that is at the same time able to adjust and to adapt to Limited.
these challenges. Woods, D.D. (2006). Essential Characteristics of
The preliminary qualitative analysis of the narratives Resilience. In E. Hollnagel, D.D. Woods & N.
identified a number of coping strategies that enable the Leveson (Eds.), Resilience engineering :
dispensary to deal with the challenges in a resilient way: concepts and precepts (pp. xii, 397 p.).
personal negotiation, creating a shared awareness, Aldershot, UK; Burlington, VT: Ashgate
prioritization of activities, offering and seeking help from Publishing Limited.
others. As maintained by Hollnagel, human variability is
a key feature to adjust and adapt to current demands, as
there are cases of role swapping, dynamic prioritization
based on local demands and relative urgency, active
monitoring of workload uneven distribution, or
anticipation of likely problematic future demands.
It is interesting to note that only prioritization of
activities is supported at an organizational level through
pre-defined urgent and standard workflows. The vast
majority of resilient forms of behavior exhibited by the
dispensary are the result of personal initiative,
negotiation and experience. No training is provided for
such skills and there are no mechanisms in place to
capture and to document valuable experiences.
The analysis is a first step towards a more
comprehensive, empirically constructed framework of
resilience in healthcare environments. Such a framework
should allow healthcare organizations to identify training
opportunities in non-technical skills as well as to
institutionalize resilience.
Acknowledgments
This work has been supported in part by the Health
Foundation (Registered Charity Number: 286967) as part
of the Safer Clinical Systems program. We would like to
thank all the pharmacy staff who contributed narratives.
References
Cook, R.I., Render, M., & Woods, D.D. (2000). Gaps in
the continuity of care and progress on patient safety.
British Medical Journal(320):791-794.
Hollnagel, E. (2006a). Resilience-The challenge of the
unstable. Aldershot, UK: Ashgate Publishing
Limited.
Hollnagel, E. (2006b). Resilience – the Challenge of the
Unstable. In E. Hollnagel, D.D. Woods & N.
Leveson (Eds.), Resilience engineering :
concepts and precepts (pp. xii, 397 p.).
Aldershot, UK; Burlington, VT: Ashgate
Publishing Limited.
Hollnagel, E., Woods, D.D., & Leveson, N. (2006).
Resilience engineering : concepts and precepts.
51