Current Projects

Theorising and measuring intervention agency, and exploring relationships with intervention effectiveness, equity and acceptability
University of Cambridge
PIs: Professor Martin White and Dr Jean Adams
man checking food nutrition label in supermarket

Dietary risks and physical inactivity accounted for 11% of disability adjusted life years lost globally in 2017, and are key drivers of obesity and related conditions. In the UK, obesity cost the NHS £6.1bn in 2014-5, and wider society £27bn. Many markers of diet quality and activity are socio-economically patterned, contributing to socio-economic differences in health outcomes.

Population approaches to prevention, where interventions are delivered to whole population irrespective of baseline risk, have been proposed to be more effective than those targeted at the small proportion of individuals with highest baseline risk. However, population interventions vary in terms of the level of agency required of individuals for them to benefit. Intervention agency refers to the personal resources (e.g. cognitive, financial, material) on which individuals draw to engage successfully with an intervention. For example, educational interventions (such as mass media campaigns delivered by Change4Life) require substantial agency in terms of accessing, understanding and applying the information provided. In contrast, the reformulation resulting from the Soft Drinks Industry Levy requires little agency from individual consumers to benefit.

There is some evidence that population interventions requiring less individual agency are more effective than those requiring more. The personal resources required to exert individual agency may also be socio-economically patterned. This means that population interventions requiring higher individual agency may be differentially effective according to socio-economic position and exacerbate existing inequalities in health. Interventions that require low levels of individual agency may also be less publicly and politically acceptable. However, there have been few systematic syntheses to confirm these findings and no framework exists to describe intervention agency. Successful interventions may also require agency from a variety of different actors (e.g. those planning or delivering interventions, as well as individual recipients). This has been poorly studied.

In this project we will:

  1. Develop theory concerning how agency is exerted and the different actors who may be required to exert agency in order for population interventions to achieve their effects
  2. Design a new tool to categorise population interventions based on the degree of agency required of all relevant actors.
  3. Systematically review studies of the effectiveness, equity and acceptability of population interventions for diet and physical activity; apply our tool; and determine whether effectiveness, equity and acceptability vary according to the degree of agency required of all relevant actors.


Principal Investigators

Professor Martin White
Dr Jean Adams