Understanding and Evaluating the Implementation of Integrated Care: A ‘Three Pipe’ Problem

Abstract

In their 2004 systematic review on the diffusion of innovations in service organizations, Greenhalgh et al concluded that there was a lack of any robust understanding in how complex health service innovations can be implemented and sustained (or not) across contexts and settings [1]. An underlying implication from this work was the need for more 'realistic evaluation' methodologies to help unpick how outcomes may result from the intricate interplay between multi-component interventions in different contexts and settings [2]. One such advance has been the recent development of the COMIC Model for the comprehensive evaluation of integrated care interventions [3]. Derived from work undertaken in the recently concluded EU-funded Project INTEGRATE [4], the COMIC Model (Context, Outcomes and Mechanisms of Integrated Care interventions) uti-lised the realistic synthesis approach to study the interplay between contexts, mechanisms and outcomes across selected case examples of integrated care, including for diabetes in Dutch Care Groups. The authors were able to demonstrate how such an approach brought insights into understanding how, when and why integrated care interventions influenced outcomes in these specific cases. As a conceptual tool, realistic synthesis provides a useful template to provide an in-depth narrative description of the various factors that may influence outcomes and, potentially, to then take the lessons from one evaluation and test them across a range of different contexts. Indeed, other work within Project INTEGRATE formulated a bench-marking tool by creating a set of generic factors influencing the implementation of integrated care that appears to have face validity across condition-specific groups (e.g. diabetes , COPD, geriatric conditions and mental health) and across different contexts and settings of deployment [4]. In this edition of IJIC, a special collection of perspective papers on the building blocks of integrated care has shed further light on some of the more critical components [5-10]. What these amply demonstrate is that the successful implementation of integrated care requires an effective composition of a complex set of interventions at the micro-, meso-and macro-levels. Moreover, effective implementation is as much relational as it is technical. In other words, the influence of pre-existing values, cultures, politics and relationships (both personal and organisational)