GLADIOLI study

Group Model Building to Assess Dietary Health Inequalities in English Local Authorities (GLADIOLI) study

Background and study aims

People living in more deprived areas of the UK are more likely to have poorer diets and higher rates of diet-related illnesses such as type 2 diabetes. Around 15% of UK households have recently experienced food insecurity. Health inequalities linked to diet are influenced by food costs, availability of healthy options, and wider food system factors.

While national government policies (such as taxes or subsidies) could help, change at this level can be slow and there is limited evidence that current approaches are reducing inequalities. Local authorities (LAs) in England have more flexibility to act locally and work with community organisations to develop practical solutions.

This study will test whether a structured workshop approach called Group Model Building (GMB) can help local authorities better understand dietary health inequalities (DHI) and develop stronger, more effective local policies to reduce them.

The study will:

Test whether the intervention increases the number and strength of local policies aimed at reducing dietary health inequalities.
Explore how local stakeholders understand dietary health inequalities and how their understanding changes over time.
Calculate the costs of delivering the intervention.
The study will use a randomly allocated controlled (comparison) trial (RCT) design, meaning some local authorities will receive the intervention and others will act as a comparison group.

Who can participate?

One Key Informant (someone with responsibility for food or dietary inequalities) each from participating English local authorities (LAs), local stakeholders in each intervention LA, including local decision-makers from relevant sectors and representatives from community food organisations.

What does the study involve?

Recruitment will begin with those in the most deprived 40% of areas.

Local authorities will be randomly allocated to one of two groups:

Intervention group

One or more online preparatory meetings with the research team.
Access to a website summarising local data on dietary health inequalities.
A one-day, in-person facilitated workshop using Group Model Building methods. During the workshop, participants will map out local factors influencing dietary inequalities and identify possible solutions.
Around three short follow-up online meetings over three months to refine and prioritise solutions.

Control group

Continue with usual practice during the study period.
Take part in data collection only.
Receive intervention materials after the study ends.
Local authorities in the control group will receive the workshop materials at the end of the study.

Data collection

All local authorities will take part in an interview at the start of the study and again 12 months later to document policies aimed at reducing dietary health inequalities.
Workshop participants (intervention group only) will complete a short questionnaire before and after the workshop about their understanding of dietary inequalities, commitment to action, confidence, and collaboration.
In a smaller number of local authorities, researchers will also carry out a qualitative process evaluation. This may include:
- Observing workshops or relevant meetings.
- Conducting interviews shortly after and again 6–12 months later

Participation is voluntary and informed consent will be obtained.

What are the possible benefits of and risks of participating?

Possible benefits:

Improved shared understanding of dietary health inequalities.
Stronger collaboration between local partners.
Development or strengthening of local policies.
Access to structured tools and local data.
Contribution to national evidence about what works to reduce health inequalities.

Possible risks:
Risks are considered low.

The main burden is time commitment for meetings, workshops and interviews.

Discussions may involve complex or sensitive topics, but researchers will facilitate respectfully. Participants may withdraw at any time.