Unfudging nudging: identifying evidence for poorly-specified interventions in public health

Download poster as PDF

Download poster as PDF

Abstract text
Background:Scoping reviews typically aim to describe a broad evidence base, delineate its boundaries and delimit evidence within it, often as a preliminary stage to systematic reviews. If prior attempts to define target interventions or clarify related theory have been limited, refined conceptual understanding and operational definition of interventions become outputs of the scoping process rather than starting points. Eligible studies may pre-date emerging theoretical frameworks and span research disciplines, and lack a common theoretical basis and terminology. Locating and selecting relevant studies therefore becomes technically challenging, requiring development and application of innovative methods.

Objectives:To identify and describe empirical evidence for the impacts of a nebulous set of interventions (‘nudging’ or ‘choice architecture’ interventions) on health behaviour, in concert with iterative development of an operational definition and a conceptual typology of interventions.

Methods:A scoping review of conceptual and empirical evidence for choice architecture interventions. Highly sensitive electronic searches were executed in parallel with snowball searches. Records retrieved by electronic searches were prioritised for manual screening according to likely relevance using text-mining technologies. Multiple iterative cycles of study screening and coding, identification of conceptual boundaries, internal discussion and consultation with external experts were used to refine an operational definition and conceptual typology of target interventions.

Results:Over 800,000 de-duplicated records were retrieved, with over 54,000 abstracts manually screened. Data were extracted from 346 articles included in the final analysis. The definition and typology of choice architecture interventions provided a robust framework to describe and interpret the identified empirical evidence.

Conclusions:The methods applied in this scoping review offer feasible solutions to difficulties that arise when attempting to identify evidence for poorly-specified interventions, whilst presenting some challenges of their own. They provide a useful template for developing similarly detailed maps of evidence for other complex public health interventions.
Authors
Hollands G1, Shemilt I1, Marteau T1, Thomas J2, Ogilvie D1
1 Behaviour and Health Research Unit, Institute of Public Health, University of Cambridge, UK
2 Evidence for Policy and Practice Information and Co-ordinating Centre, Institute of Education, UK
Presenting author and contact person
Presenting author: 
Gareth Hollands
Contact person Affiliation Country
Gareth Hollands (Contact this person) Institute of Public Health, University of Cambridge United Kingdom