Workplace Cafeteria

High meat consumption in wealthy countries is associated with health and environmental issues. Studies suggest that food selection is often habitual and influenced by environmental cues. Increasing the availability of vegetarian options in cafeterias has shown promise in previous studies, but randomized controlled trials are necessary to estimate its impact across cafeteria settings catering to a more diverse range of customers (i.e. beyond university students).

This study will assess the impact of replacing one of the meat meals on offer with a vegetarian meal on the proportion of vegetarian meal sales. It will also explore other outcomes such as total site revenue and food waste to determine any unexpected positive or negative effects. Additionally, process and economic evaluations will be conducted to assess the feasibility, acceptability, and cost-effectiveness of the intervention.

Does increasing the relative availability of vegetarian (vs. meat/fish) main meals on the lunch menu increase the relative sales of these dishes?

The intervention involves increasing the availability of vegetarian meal options on worksite cafeteria lunch counters by replacing one meat meal option with a vegetarian alternative. This will be implemented using a stepped-wedge design, where different cafeterias will adopt the intervention at various (randomised) time points over six weeks.

The overall goal is to understand both the intended and unintended effects of interventions aimed at reducing meat consumption in cafeteria settings. The intervention may have unintended consequences on food waste, cafeteria revenue, or health indicators (such as calories purchased). We will monitor these outcomes to ensure we know as much as possible about unintended effects.

The intervention could work by making it more likely that a customer’s preferred meal is within the increased set of vegetarian meal options (rather than the decreased set of meat meal options). Additionally, increasing the number of vegetarian meals on offer might incidentally put more cheaper options on the menu (as vegetarian meals tend to be cheaper), and may change perceived social norms of what others in the cafeteria are eating.

Catering company, cafeteria staff, cafeteria customers.

Worksite cafeterias across England

Anyone who eats in participating worksite cafeterias 

Theory of change. workplace cafe

Eligible sites (large enough menu to be able to swap out one meat meal for a vegetarian meal and still offer at least one vegetarian and one meat choice) were asked to participate. 

All outcome data was provided centrally by the catering company.

Environmental and health impacts of the trial will be modelled using recipe data. Recipes (ingredients, amounts, and nutrition info) will be translated into health and environmental effects of active vs control periods in the trial.

Ethical approval for this study was given by Oxford University’s Medical Sciences Interdivisional Research Ethics Committee (IDREC), under reference number R72710/RE006.

We will explore the perspectives of a range of stakeholders and recipients involved in the field trial to understand what worked in the intervention, for whom, in what circumstances and why. Semi-structured interviews with between 20-30 intervention recipients (cafeteria customers) and intervention implementers (cafeteria chefs and managers) will explore experiences and perceived impacts of the intervention, as well as broader perspectives on sustainable diets in the workplace. Wider contextual influences on intervention experience, such as site location, staff demographics and site ethos, will also be investigated. Interviews with at least 3 central staff in the catering company will explore experiences of the intervention, identifying perceived barriers and facilitators to its effectiveness, and explore plans for utilising trial findings in the future.

Academic paper(s), academic conference presentations.

The trial was conducted in the period from the 18^th Sep until 30^th Oct 2023. Data analysis is underway.