Quality improvement interpreted as a complex adaptive system : implications and opportunities : a thesis submitted in partial fulfilment of the requirements for the degree of Doctor of Philosophy at Massey University, Palmerston North, New Zealand

Abstract

The effectiveness of quality improvement (QI) methods in healthcare has been challenged, especially under circumstances of high complexity. This thesis examines the implications for quality improvement if complex socio-technical systems such as healthcare are interpreted as complex adaptive systems (CAS). The research followed a mixed-method design. Informed by the complex systems and quality management literature, a conceptual model for quality improvement within CAS was developed — the complex quality improvement network (CQIN). An agent-based simulation model was then used to establish the plausibility and face validity of the model constructs and their interaction. Thematic analysis and crisp-set qualitative comparative analysis (QCA) were then used to examine the evidence for CQIN constructs within published quality improvement case studies. One applied case study was also conducted for deeper insight into the practical difficulties of interpreting a real-world quality improvement project as a CAS. Finally, the findings of the simulation modelling and the secondary data analysis were integrated into a Bayesian network model. Empirical evidence, in the form of consistency across cases and coverage within cases, was found for eleven of the twelve CQIN constructs. Multiple sets of sufficient conditions for reported improvement success were identified across cases. These sets were minimised to four strategies for successful quality improvement; i) strengthening agent network communication paths; ii) building shared understanding of problem and context amongst networked agents; iii) increasing problem-solving effectiveness; and iv) improved system signal integration. If the evolutionary foundations for CAS are in some way inhibited, the likelihood of quality improvement success is reduced. Healthcare quality improvement can be plausibly simulated using fundamental CAS principles. The first contribution to quality improvement discourse is the CQIN model, a CAS model of change applied specifically to quality improvement. A second contribution of this research is a complex quality improvement risk assessment model, the CQIN Bayesian Network. Practitioners can use this model to examine and test identified CAS-informed improvement strategies. The individual CQIN constructs make a third contribution by providing new categories of causal factors for the comparison of disparate quality improvement case studies.