Janet Anderson, King’s College London
Discussions regarding the nature of complexity emphasise concepts such as emergence, non-linearity, cross level interactions, and self-organisation as hallmarks of complex systems. Much HF thinking about complexity in systems is based on the analysis of engineered systems in which the design is fully specified and outcome measures are clear. This has led to a focus on the role of automation as a mediator between the operator and the system being controlled, which is seen as adding another layer of complexity. By contrast, in healthcare, clinicians interact directly with patients using technology only for discrete tasks rather than for controlling the system. Furthermore, care is delivered through complex layers of interaction between colleagues, patients, relatives and other stakeholders. In this paper we propose that the existence of multiple conflicting performance measures is a feature of healthcare work which adds great complexity and increases the difficulty of controlling the system to deliver standardised outcomes.
Our discussion is based on extensive ethnographic work in two settings: the Emergency Department (ED) and the Older Person’s Unit (OPU) of a large London teaching hospital. Using a resilience engineering perspective, observations and interviews were conducted based on a theoretical resilience model. Analysis focused on pressures and demands, adaptations and outcomes.
Our data show that there is a multiplicity of factors that determine whether an good outcome has been achieved. Three categories of factors that shape outcomes are;
1.Subjective preferences and individual judgments about priorities for patients and their families/carers. Patient factors such as prognosis, ability to tolerate treatments, and judgements about the quality of life are important, as well as the presence, availability and skills of relatives and carers and support from other healthcare organisations.
2.Imposed targets and standards create organisational imperatives and often focus on efficiency and throughput or the avoidance of specific adverse events, such as never events. This imposes requirements that may compete with other goals of clinical care, creating dilemmas for practitioners who receive little guidance for resolving such conflicts.
3.The context in which care is delivered is crucial. Healthcare organisations are open to disturbances in the environment that create extra demands, especially due to increased patient numbers. Their ability to absorb disturbances and maintain standards of care is key, but is bounded – they cannot stretch indefinitely. Continued functioning under pressure may involve prioritising some goals over others. For example, patient flow may be prioritised over thorough investigation when there is a long queue of patients in the ED.
In this paper we discuss how trade-offs and prioritising judgements are integral to care. We show how targets can be counter-productive by creating hard constraints on how care is delivered, forcing other goals to be traded off. Multiple competing performance measures are a feature of complexity in healthcare and require a nuanced approach to quality improvement rather than the imposition of targets for single indicators that fail to reflect context.