Changing environments to change behaviour: use of the Typology of Interventions in Proximal Physical Micro-Environments tool

Abstract

BackgroundReflecting widespread interest in concepts of nudging and choice architecture, increasing research and policy attention is being applied to altering aspects of the small-scale physical environment, such as portion sizes or product positioning, to change health-related behaviour at population level. However, there is a lack of clarity in characterising these interventions and no reliable framework incorporating standardised definitions, hampering both the synthesis of cumulative evidence about intervention effects and the identification of intervention opportunities. To address this shortcoming, a new tool—Typology of Interventions in Proximal Physical Micro-Environments (TIPPME)—has been developed, here applied to the selection, purchase, and consumption of food, alcohol, and tobacco products. MethodsTIPPME was developed over three phases (identifying need; developing and elaborating; and reliability testing and finalising). A large-scale systematic scoping review was conducted, which generated a provisional intervention typology. This typology was discussed in two expert workshops to identify possible improvements. A series of structured discussions between the research team were conducted, using consensus methods to agree on any proposed changes to the typology, and resulting in a preliminary version of TIPPME. Two intervention coding exercises to assess how reliably it could be used to characterise interventions were completed by researchers or practitioners with backgrounds in public health or behavioural science (n=33) and members of the research team. The first exercise involved coding 40 short intervention descriptions and the second involved coding 24 full-text papers, all taken from the intervention literature identified by the systematic scoping review. FindingsTIPPME comprises a matrix classification structure defining six intervention types (availability, position, functionality, presentation, size, information) and three spatial foci (product, related objects, wider environment). Coding exercises demonstrated strong levels of agreement between participants in applying intervention types (Fleiss' κ=0·76 [exercise 1], 0·80 [exercise 2]). InterpretationTIPPME provides a framework to reliably classify and describe an important class of interventions, and enable more systematic design, reporting, and analysis of interventions to change health-related behaviour. It makes a distinct, novel contribution to collective efforts to build the cumulative evidence-base for effective ways of changing behaviour across populations. FundingUK Department of Health Policy Research Programme (Policy Research Unit in Behaviour and Health [PR-UN-0409-10109]). DO is supported by the Medical Research Council (Unit Programme number MC_UU_12015/6). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the abstract.