Realist programme theory development in practice: the example of the IMPACT realist review on antimicrobial prescribing for doctors-in-training

Published papers reporting on realist reviews often neglect to explain the process of programme theory development in enough detail or only present post-hoc rationalised accounts. Without adequate attention to the process of... [ view full abstract ]

Published papers reporting on realist reviews often neglect to explain the process of programme theory development in enough detail or only present post-hoc rationalised accounts. Without adequate attention to the process of developing programme theory, there is diminished scope for methodological development and capacity building. Explicitly setting out the process of programme theory development, including the involvement of stakeholders, allows for increased methodological rigour and transparency.

This presentation will discuss programme theory development as part of the IMPACT review of interventions to improve antimicrobial prescribing for doctors-in-training. In this review, we departed from the most common approach, which focuses on identifying the programme theory of interventions themselves, to concentrate instead on the programme theory of the overall process under study – that is antimicrobial prescribing. To do this, we looked at how doctors-in-training are involved in antimicrobial prescribing, i.e. how the process of antimicrobial prescribing for doctors-in-training works more generally in different healthcare settings, rather than looking at how it was assumed interventions designed to change the anti-microbial behaviour of doctors-in-training would work.

Our approach to iterative programme theory development allowed us to incorporate a wider breadth of literature towards understanding how mechanisms triggered in particular contexts produce certain outcomes. Step 1: An initial programme theory of the process of antimicrobial prescribing was developed drawing on the experiences of the project team and informal searching of the literature. Step 2: Using the programme theory as a preliminary frame for the review, the literature emerging from the main search of electronic databases was screened for relevance and coded to understand how and why different facilitators and barriers of antimicrobial prescribing influenced the practices of doctors-in-training. Step 3: Explanations were built and refined iteratively, through multiple rounds of familiarisation with the literature and formalised NVivo coding. Step 4: Through this process of theory building we identified hierarchical relationships as core aspects in antimicrobial prescribing practices for doctors-in-training. Step 5: Using this increased understanding, we carried out an additional literature search specifically on hierarchies to allow for an in-depth explanatory focus.

Drawing on medical hierarchies as a lens for configuring explanations of antimicrobial prescribing provided a focus for weaving together different forms of evidence from different study types. Through iterative rounds of refining, refuting and corroborating aspects of the programme theory, the additional literature provided an explanatory backbone to the configuration of contexts, mechanisms and outcomes.

Apart from analysing and configuring the literature, programme theory was also influenced by the stakeholder group recruited for the IMPACT review. Consisting of patients, health professionals and policy-makers, the role of the group was significant in confirming and refining aspects of the programme theory. Instead of a uniform approach to stakeholder involvement in realist reviews, however, we would advocate for an explicit rationale on what different members of a stakeholder group can contribute at different stages of the review process.

This comprehensive account of programme theory development from the IMPACT review could contribute to relevant methodological debates on what constitutes good practice on this issue.