Table of Contents
Simulating health and social care delivery: Can the tension
between individual patient path ways and strategic planning
beaddressed?
Steffen Bayer, James Barlow, Richard Curry
Imovation Studies Centre, The Business School, Imperial College London
South Kensington Campus, London, SW7 2AZ, United Kingdom.
Telephone: +44 (0)20 7594 5985
s.bayer@imperial.ac.uk, j.badow@impaiial.ac.uk, richard.cunry@imperial.ac.uk
In this presentation we discuss the use of system dynamics to understand the delivery of
health and social care. We report on the development of a simulation model to address
the issues of elderly careina local health care systemin the South East of England. We
explore the appropriate level of aggregation for a model to address the policy
challenges faced by local care organisations. While we recognise the value of a highly
aggregated systems dynamics approach and the danger of inappropriate
disaggregation, we argue in the presentation that features of the care system (such as
the recurrence of acimissions to the hospital) make it desirable to include some aspects
of patient history in a flow model of patient path ways. We examine the feasibility of a
more disaggregated approach and the influence of this choice for the understanding of
the system and the development of beneficial policy options.
As Westem socielies age and the cost of providing health and social care rises
innovative approaches to the delivery of savices ae becoming increasingly important
Such new approaches include an increased emphasis on rehabilitation, the delivery of
care in new settings (such as in intermediate care facilities and the cient/patients own
home) and the support of care delivery by technical means (e.g. tdecare systems).
Health and social care delivery is a complex system whose devdopment is strongly
influenced by the actions of different actors such as health and social care provides,
funders, as well as wineable individuals and their families. Actions in one part of the
care system often have unexpected and unwanted consequences elsewhere. Costs and
benefits of any change are uneverly distributed across the system Systemic thinking is
therefore required to plan and implerent changes effectivdy and successfully. In the
UK, the govemment has recognised this, at least in principle and promoted a “whole
system approach” in health and social care planning.
The System Dynamics approach is wel suited for the study of policy issues in the
ywovision of health and social care for the eddelly. The published work in the ara
generally derives major insights from the analysis of the system strucume on a rdalivdy
aggregated level. Wolstenholme (1993 and 1999), Linard (1996) and Walker and
Haslett (1999) are some examples of past work in this area Clearly, this analysis on a
more aggregae levd fits wel into the spirit of the system dynamics approach, and
exploits its strength.
In addition to the tensions between the gods, polides and actions of diffeent
organizations, there are also tensions within organizations A mejor charectedstic of
Management in health and sodal cae aryganizaiions is the tension between decision
meking on diffeent levds. These ae the levd of the individia patients (with ther
particular history) and the strategic level within an orgenization In addition to being a
“prindpal - agent’ problem this is dso an expression of the diffaent type of
infonmetion available on a strategic leva and on the levd of an individual patient
The aggregate, systemic outcome arises as a consequence of the care pathways of
individual patients, which in themselves are not rdevant for policy leva and strategic
decision-making. The system dynamicist can (and should) ignoe the lower led
derents and rdationships which do not influence the policy recommendetion -
unnecessary details are noise which obscure the understanding of the system. However,
inasmuch as the collective effects of a more delailed disaggregation influences systemic
outcomes and policy recommendations, they cannot safely be ignored.
The individual history of somebody in a care situation matters in influencing the further
path of care delivery. Many problems of the health care system (such as repeated
hospital stays of inadequately supported at tisk individiels in the commmity) or
expected effects of policies (such as an emphasis of rehabilitation and prevention
cannot be adequately studied in an ahistoric “ance through perspective” where each
episode is new and changes to flow rates as patients cycle to through the same cae
setting camot be capmed. However induding the history of individual patients is
catainly neither desirable if our aim is to foster a policy rdevant, system led
understanding of care delivery, nor feasinle within a well-constructed system dynamics
model.
This presentation explores to what extent it is possible to capume the policy relevant
effects of the care paths and history by segmenting patient groups in care settings (eg.
the hospital) and including different flow rates betwen and acooriing to these
segments. This atfempt to better understand aggregate effects has to be distinguished
from segmentation to inboduce atdiitiona detail. Diffeent options for such a
segmentation (eg. according to the mimber of previous hospital visits or to the home
care package received before the hospital stay) have consequences in tems of data
requirements and the structure and complexity of the system
In the presentation this discussion is applied to a model of health and social care
delivery which wes devdoped for a location in the South East of England in order to
understand the impact of healthcare imovations such as telecae on the local care
economy. The model allows assessing the impact of choices regarding the capacity (and
type) of available cae savices and regarding refenal policies on the diffaent
components on the care system
We enphaesize that an appropriate balance between aggregation and segmentation can
help to devedop valuable insights into key strategic and policy issues while minimising
the danger of obscuring the model structure and introducing unnecessary complexity. In
patticula; this approach allows the design of policies which address the important issue
of the recurrent admission of patients into the hospital. We argue in the presentation
that this attempt can to some extent the tension between individual patient path ways
and strategic planning can support the implementation of savice imovetions which
have systemic benefits for the health and social system.
References
Linard, K. T. 1996. System Dynamics Modelling and Aged Care, Modelling the Future
- Techniques & Directions, Australian Institute of Health & Welfare Workshop.
Richardson, G P, Anderson, D F and Wu, Y J. 2002. Misattribution in Welfare
dynamics: The Puzzling Dynamics of Recidivism. Proceedings of the XX Intemational
Conference of the System Dynamics Society, Palermo.
Walker, B. C. 2000. The dynamics of local rules in hospital admission parocesses.
Moresh University.
Walker, B. C. and Haslett, T. 1999. System Dynamics and Action Research in Aged
Care, Proceedings of the XVII Intemational Conference of the System Dynamics
Society, Wellington.
Wolstenholme, E.F. 1993. A case study in community care using systems thinking.
Joumal of the Operational Research Society 44: 925-934.
Wolstenholme, E.F. 1999. A patient flow perspective of U.K. Health Services:
Exploring the case for new "intermediate care" initiatives, System Dynamics Review
15: 253-271.
1 This important problem in care delivery for the dderly is similar to problems in many other policy
areas, Such as the problems of recidivism in wdlfare reform. (Richardson etal., 2002)
Back to the Top