Conceptualising contexts, mechanisms, and outcomes: Dilemmas and the road ahead in Health Services Research

Background Realist methods, including theory building, review, synthesis and evaluation are gaining popularity in health services research (HSR) because of their apparent applicability to the complexity of care and focus on... [ view full abstract ]

Background
Realist methods, including theory building, review, synthesis and evaluation are gaining popularity in health services research (HSR) because of their apparent applicability to the complexity of care and focus on causal mechanisms. Interventions can be evaluated in terms of how participants ‘reason or behave in response to new resources’. Understanding how outcomes are generated by mechanisms working differentially in different contexts is at the core of realist thinking, but how contexts and mechanisms are conceptualised remains under-explored. As realist methods are increasingly used in multidisciplinary and cross-sectoral projects, it is a priority to better conceptualise these core aspects.

Aim
To illustrate and explore the dilemmas of categorising and conceptualising contexts, mechanisms and outcomes when building theory using a realist approach in real world health and social care settings.
Round table method
• (10min) Introductions
• (20min) Present our propositions (see below) using examples from analyses carried out in two ‘realist’ projects carried out in the south west of UK:
• A mixed methods ethnographic and ‘value stream mapping’ study of decisions about admission made in emergency settings
• Development and evaluation of an intervention for prison leavers with mental health problems
• (10min) Panel members discuss these propositions in turn.
• (50min) Further discussion with panel members and audience regarding these propositions to build on adapt and expose uncertainties and further theoretical work needed to develop the propositions. All participants will be encouraged to illustrate their points, wherever possible, with examples drawn from realist research.

Propositions
1. A generic framework is useful for considering different typical scenarios in health service research settings:
a. Such a model would encompass organisational structures, processes and cultures as well as practitioners’ interactions with patients/clients/individuals
b. Other key interactions include practitioner-practitioner interactions and individual-other practitioner ) interactions.
c. Practitioners bring resource/s to an individual which might change their thinking, skills or behaviours
d. The way the system supports practitioners is critical. The resource could be the manager/supervisor (providing guidance), a piece of technology (prompting or assisting data flow), or a space (for example in emergency rooms) which allows an individual practitioner to work in a different way.
e. Together these form an interlocking logic model of how organisations along with individual practitioners can generate better outcomes for patients/clients within systems.

2. It is important to conceptualise the internal logic of health service research context-mechanism-outcome chains.
a. Our findings build on the concept of intervention ‘dose’ (Rycroft-Malone et al) in that outcomes may be graded; they may be either changes in patterns of thinking and knowledge, or actual behaviours.
b. Outcomes are likely to be the result of multiple mechanisms interacting or at adding cumulatively.
c. Mechanisms can have effects on multiple different outcomes
Given the lack of context-independent magic bullets it becomes even more important to understand when mechanisms can be triggered within certain contexts to produce small changes in outcomes. It is valuable, given these complexities, to build up an interlocking theory of potential causal links, with contingencies.