Dual-Label Intervention to Promote Low-Carbon and Healthy Food Choices in a University Cafeteria

Overview

This study aims to evaluate whether a dual-label intervention, combining carbon -coded footprint color labels and calorie information, can promote low-carbon and healthier food choices among university students in a real-world cafeteria setting. A randomized controlled trial will be conducted with approximately 200 participants, who will be assigned to either an intervention group (with labels) or a control group (without labels). The study a consists of 2-week baseline period, a 4-week intervention period, and a 2-week a follow-up period. The primary outcome is the mean carbon footprint level of selected meals, and secondary outcomes include nutritional quality, body composition, and changes in environmental and health awareness.

Study Design and Setting This study is designed as a prospective, stratified randomized controlled trial conducted in a real-world university cafeteria setting. The aim is to evaluate the isolated effect of a dual-label intervention (carbon footprint color-coding combined with calorie information) on food choice behavior without confounding from nutritional labeling. The cafeteria environment will be maintained consistently across both study arms regarding menu offerings, pricing, and food placement. Participant Recruitment and Randomization Approximately 200 university student volunteers will be recruited. Following a 2-week baseline observation period, participants will undergo body composition analysis and will be stratified by gender, Body Mass Index and baseline dietary carbon footprint level. Within each stratum, participants will be randomly assigned in a 1:1 ratio to either the Intervention Group or the Control Group. Participants will be blinded to the specific study hypothesis (carbon reduction) during recruitment and baseline to minimize the Hawthorne effect. Intervention and Menu Profiling Carbon Footprint Labeling: Life-cycle carbon emissions per 100g will be calculated. Dishes are categorized into three color-coded tiers using quintile distribution: Green (Low Carbon), Yellow (Medium Carbon), and Red (High Carbon). Calorie Labeling: Explicit calorie values (kcal per 100g) will be displayed numerically. Nutritional Quality (Hidden Variable): A Nutrient Density Score (NDS) will be calculated for each dish and classified into four grades (A, B, C, D). Crucially, this nutritional grade will NOT be displayed on labels. It serves solely as an analytical variable to assess the natural alignment between low-carbon choices and nutritional adequacy, thereby isolating the effect of carbon/calorie cues. Study Phases The trial spans 8 weeks in total, divided into three distinct phases: Phase 1: Baseline Period (Weeks 1-2): All participants select lunch from their designated windows (Intervention windows vs. Control windows). No labels are displayed during this phase. Participants will capture and upload meal photos daily via a designated system. Baseline environmental and health awareness questionnaires will be administered. Phase 2: Intervention Period (Weeks 3-6): Intervention Group: Food items at the designated windows will display the dual-label tags (Colored Carbon Footprint Icon + Calorie Number). Control Group: No labels will be displayed at their designated windows. Participants continue daily photo uploads of their selected meals. To verify label exposure, the Intervention Group will be prompted to identify the color and calorie content of their selection during the check-in process. Phase 3: Follow-up Period (Weeks 7-8): All labels will be removed from the cafeteria. Participants continue to select meals from their originally assigned windows and upload photos. This phase assesses the sustainability of any behavioral changes induced by the label exposure. Outcome Measures Primary Outcome: Change in the mean carbon footprint level (scored as Green=1, Yellow=2, Red=3) of selected meals from baseline to the intervention period. Secondary Outcomes: Implicit nutritional quality of selected meals. Changes in body composition (Body Fat Percentage, Skeletal Muscle Mass) measured pre- and post-intervention. Changes in self-reported environmental and health awareness scores. Congruence effect: Analysis of behavioral change when carbon and calorie cues are aligned (e.g., Green + Low Calorie) versus conflicting (e.g., Green + High Calorie). Data Collection and Analysis: Meal selection data will be digitized daily through participant-submitted photographs. Carbon and nutritional scores will be calculated based on the backend recipe database. Statistical analysis will utilize linear mixed models to account for repeated measures within participants, comparing the differential change between the Intervention and Control groups over time.