A Systems Thinking Approach To Investigating Delayed Discharges In The UK
Sangeeta Sardiwal
London South Bank University
Abstract
Outlined is the systems based process used to investigate delayed discharging, which combines
parts of the Soft Systems Methodology (SSM) with Systems Dynamics (SD).At the early stages
use of hard system approaches were inappropriate to investigate this type of complex problem.
SSM was used to consider the soft problems present, expressing the problem situation as a rich
picture. This informed construction of a SD model to investigate policy interventions for
reducing delayed discharges. This process had practical implications for structuring qualitative
data to inform the SD methodology.
Introduction
The aim of this paper has been to outline the systems based process used to investigate delayed
discharging in the NHS in the UK.
Delayed discharging happens when patients that are medically fit to return home become stuck
within hospital through no fault of their own. There are numerous causes of delayed discharge,
such as the lack of social care provision and insufficient numbers of social workers to carry out
patient assessments. This research has focused on delayed discharging of elderly patients, within
hospital concerning the NHS and Social Services. Delayed discharges are multi-faceted
occurring at different stages of the patient process.
The NHS and social care system has large amounts of dynamic complexity, involving large
amounts of information, stakeholders and technology. This arises because the system changes
over time and is governed by feedback, dominant structures. The system is adaptive and
characterised by tradeoffs. This was seen when social services used an informal policy of
prioritisation to cope with large patient backlogs (Sardiwal, 2009). Patients waiting for social
service provision to be set up downstream along the patient process were prioritised over patients
that needed initial social service assessments.
Given the dynamic complexity that exists there has been a need to use systems thinking. Systems
thinking focuses on the whole system and the relationship between its parts, rather than on the
individual parts. Having a holistic view of the problem situation has been extremely important
given the increased conflict, pressure from different areas and policy resistance from
stakeholders. It has been important to consider the perspectives of different stakeholders and not
to make any assumptions about what things are like in practice. MacDonough (2010) argued that
social worker practitioners felt that their roles were fulfilling despite the pressures such as targets
on managers, which can be a different view portrayed by the public.
Soft System Dynamics Methodology
The systems based process used to investigate delayed discharge has been highlighted in figure
1. The methodological process has combined stages 1, 2, 5 and 7 of the Soft Systems
Methodology (Checkland, 2000) with System Dynamics (Forrester, 1968). The process
incorporated three main aspects. These were firstly to find out about the problem situation in the
real world (stages 1 and 2), secondly to use systems thinking to devise a suitable intervention
(stages 3 to 5) and thirdly to take action (stage 6 to 7). A key focus of the process was on
gaining a practical insight into the issue of delayed discharges.
7. Policy changes to improve the
problem situation (Checkland’s stage 7
1. Mess problem situation of delayed ‘Action to improve the problem
discharge (Checkland’s stage 1 situation)
situation considered problematic)
2. Problem situation expressed in 6. Comparison of two models-
WHAT IS -REAL rich picture (Checkland's stage 2 wn be Ma hee oe be
y a heckland’s stage
WORLD # problem situation expressed Comparison of model and real
world)
3. Causal loop diagrams/ annotated
system dynamics models (added to
Checkland's methdology)
WHAT MIGHT BE-
SYSTEMS THINKING}
(LOGICAL
ANALYSIS)
4. System dynamics model of the
fleet of seperate information systems
(added to Checkland's methodology) 5, System dynamics model of what might
be, modeling the effects of ature integrated
information systems (added to Checkland’s
methodology)
Figure 1: Soft System Dynamics Methodology
The soft systems methodology and system dynamics methodologies have been previously
integrated in different ways. Rodriquez-Ulloa and Paucar-Caceres (2004) presents a ten stage
approach that incorporates all the stages of the soft systems methodology. For example the use of
stage 3 the problem orientated root definitions and stage 6 determining the cultural and desirable
change needed.
The main reason that the soft systems methodology was combined with system dynamics was to
help understand the problem situation that has lead to the reference mode of increased delayed
discharges among elderly patients. This would aid in the formulation of a systemic structure, the
annotated system dynamics models to start the system dynamics process. The process of
combining system dynamics with the soft systems methodology was devised to reflect this, and
only the essential stages were incorporated.
Application of the Soft System Dynamics Methodology
At the early stages of research the use of hard system approaches were inappropriate to
investigate this type of complex problem. Delayed discharges had not been well-defined. It was
necessary to use Soft Systems Methodology to consider the soft problems present in the problem
situation (stage 1) and to express the problem situation in the form of a rich picture (stage 2).
The Soft Systems Methodology helped to devise a mental construct for perceiving the problem
situation, capturing the richness of information. The rich picture (see figure 2) was based on the
qualitative data collected from interviews with stakeholders and reflected issues, structures and
processes in the NHS and Social Services system. It was drawn to capture the problem situation
holistically and systemically.
Human activity systems have been captured, such as the social workers that are waiting to go to
the ward to carry out the assessments. They are important as human activity systems can create
purposeful action that can elevate the problem situation.
The rich picture served several purposes. The views of different stakeholders were incorporated
into the rich picture, illustrating divergent views about the problem situation of delayed
discharging. This has been important as often the mental models of stakeholders have been
limited to small areas of the problem situation. It helped to make decisions such as what levels of
aggregation should be used and what states should be included.
The rich picture concerned capturing soft variables such as the socio-technical aspects that had
the characteristics of being objectionable, changeable and loosely measurable. These were
variables that could not be seen in the real world as they had no physical substance. Not all the
variables were quantifiable and as a result could not be transferred into the annotated stock and
flow diagram. For example it was highlighted that staff did not always understand the
importance of having an integrated information system on service provision
There was a need to move from the real world to the virtual systems thinking world. The rich
pictures expressed informed the constructing of the causal loop diagrams in stage 3 (see figure
3). This was moving from the impossibility non-deterministic complexity to carry out
experiments to stimulating and experimenting with a computer model. In stage 4 annotated
system dynamics models informed the construction of a formal system dynamics model which
was used to replicate the reference mode behaviour of delayed discharges.
Continuing in the systems thinking world in stage 5 the system dynamics model provided a
virtual environment to test the effect of policy interventions for reducing delayed discharges, in a
rigorous way without experimenting with the real system. The proposition that integrated
information systems (what might be) could make a difference in reducing delayed discharges
was tested.
wag loony spo
roe \
€ Namal
SN
sonwehs 5
5 id
of hier
Patent
Paes
penee
OEE fm war "
OO Peper / )
ts
\
i carro heap
al
fosr HOSsertéL
we
ero
(3 Lats heres
s ©
rere Pty § e8)
Ee
Yo | ao pomese »
K A atemment
pi ae aD)
as © ward to}
Steam. ereutee —
Figure 2: Rich Picture Of The Health And Social Care System
“bes
cae
a
P ae
ag
ee Brite
is lear: nee
penn 2
le
spa notifications
f. Rorgecson: i
Eee
. fT) ican @ |
ia fea &
- [Peete pYouw a ® ered ce eee
és ee, er pOw | a, a = oe mee Le
ps eng [NHS assessment] sechon 2 section 2 start 8S section 5 f Lpromsea | eave provinion
eee cnn TE mss, Ta cad rote “~ nection A iceriiedeit
‘nonficaton for secton 2 ERY, sage
@ |e /
not refered
cca ov sermoes
Figure 3: Annotated Stock And Flow Diagram Illustrating The Balancing Loops Present
There was a need to compare the conceptual system dynamics model with the real world and
incorporate stage 5 of the Soft Systems Methodology. This has helped to reflect on the systems
thinking process.
Education and debate has taken place with the case study at stage 7 of the Soft System Dynamics
Methodology.
Discussion
The soft system dynamics methodology reduces the limitations of each separate methodology,
soft systems methodology and system dynamics as they are combined together.
Using the soft system dynamics methodology helped to create a separation between the real
world and the systems world. This has not always been clear with the system dynamics
approach, as it has been outlined within the soft systems methodology.
The interviews were extremely important in carrying out the process of inquiry to help provide
qualitative data about the problem situation that could be structured. For example in determining
the boundaries of the problem situation within the health and social care sectors. The interviews
helped to discover the causes of the problems rather than the symptoms of the problems. This has
been important as often what occurs in practice is very different from what managers think are
happening. The use of rich pictures helped to capture the information relationships that existed.
The soft system dynamics methodology provided a flexible approach to structuring data.
Although the soft systems methodology has helped towards constructing a mental construct,
there has been a lack of dynamic coherence between the behaviour and structure of the problem
situation expressed within this methodology (Lane and Olivia, 1998). There has been a lack of
understanding towards the intuitive behaviour of complex systems incorporated in the soft
systems methodology. This goes further than the soft systems methodology, as an inability of the
human mind to have the capacity to handle complex problems in this way. Given this the system
dynamics methodology is extremely valuable, as system dynamics addresses the issue of how the
behaviour of the system is controlled by its structure.
System dynamics helped to provide a useful insight into the effects of integrated information
systems on delayed discharges. It has been important not to assume that system dynamics is
solely a hard systems approach. Hard system approaches suggests that there is an end goal.
System dynamics methodology is both quantitative and qualitative. What was evident in
applying the soft system dynamics methodology was the importance of qualitative system
dynamics.
The system dynamics methodology was flexible to receiving the input of the rich picture from
the soft systems methodology to inform the annotated system dynamics model. The results of the
system dynamics model showed that integrated information systems had a significant difference
in reducing delayed discharges (Sardiwal, 2009). It was recognised that implementing integrated
information would not be the end goal to eliminating delayed discharges. It has been important to
continue from stage 5 of building the system dynamics model by working in the real world and
applying the soft systems methodology. With the soft systems methodology the goal is not
reached, as the process leads to action in a continuous learning cycle. This has implications for
ensuring the process is not goal orientated.
Contribution to Operational Research
The systems thinking methodology of combining stages of the soft systems methodology with
the system dynamics (seen figure 1) has been documented in previous literature. This paper
extends the literature by highlighting a practical example of this multi-methodological approach.
This methodological approach proved to have some significant impact on the case study
concerned. The rich picture was used as a communication tool with stakeholders, which helped
to create buy in to the system dynamics model. The Social Services and the NHS trusts
concerned with had a greater awareness of what the model was based on. They gained
understanding of what were the most important issues in the situation. Use of the soft system
dynamics methodology helped to open these methods to a wider audience.
A more feasible solution for reducing delayed discharges was devised by considering the
different perspectives of stakeholders within the situation and involving stakeholders in the
model building process.
The multi-methodological process has involved moving between the real world and the systems
thinking world at different stages of the process. Application of this process has shown to have
practical implications for aiding the structuring of the qualitative data and helping to represent
interactions in complex systems in a clear way, which has been used to inform the system
dynamics process. The rich picture helped to focus on key variables, and the feedback loop
structure in place that could be used to inform the annotated system dynamics model.
Through structuring of the qualitative data it can help modellers early on in the research identify
what strategies could be used to improve the performance of the NHS and Social Services
system. This could be particularly helpful to inexperienced modellers.
Outlining this has potential benefits to a wide range of stakeholders. This research framework
can be used to investigate other similar complex problems as it is very accessible. This helps to
improve the communication of these systemic methods to other fields.
Conclusion
The process behind the soft system dynamics methodology has shown to be very useful. The
value of the soft system dynamics methodology has largely been to help explore and understand
the problem situation given the large socio-technical issues present with the problem situation.
The information systems between the NHS and Social Services are very much separate systems.
It is recommended that further research should be carried out to investigate how in practice the
NHS and Social Services can move from having separated to fully integrated information
systems.
References
Checkland, P (2000) Systems thinking, Systems practice. John Wiley & Sons, New York, NY
Forrester, J.W (1968) Industrial Dynamics, Massachusetts Institute of Technology, The M.I.T
Press
Lane, D.C. and Olivia, R. (1998) The greater whole: towards a synthesis of system dynamics and
soft systems methodology. European journal of operational research, 107 (1). pp. 214-235.
MacDonough, J (2010) Can I practice the social work I believe in within the Statutory Sector?
Neo-Liberalism verses Social Justice, 17" February 2010 [Lecture], London, London South
Bank University
Rodriqez-Ulloa, R, Paucar-Caceres, A (2004) Soft System dynamics methodology (SSDM): A
combination of soft systems methodology (SSM) and system dynamics, System Dynamics Society
Conference, 25"-29" July 2004, Keble College, Oxford
Sardiwal, S (2010) Coping Policies within Social Services in the UK, System Dynamics UK
Chapter Conference, 1g” February 2010, London South Bank University
Sardiwal (2009) Evaluating the Effect of Integrated Health and Social Care Information Systems
on Delayed Discharging of Patients, The 27th International System Dynamics Conference,
26th — 30" July, 2009, Albuquerque, New Mexico, USA
