Sustainability and nutritional composition of food choices in hospital canteens: a pre-post intervention study

This pre-post intervention study of three Italian hospital canteens found that while educational and environmental measures led to partial improvements in food choices, the meals remained nutritionally imbalanced and environmentally unsustainable, suggesting that modifying the actual food supply is necessary for more significant impact.

The Gist

Imagine a hospital not just as a place where people go to get better, but also as a giant cafeteria where hundreds of doctors, nurses, and staff grab their lunch every day. The researchers in this study wanted to see if they could gently nudge these workers toward eating meals that are better for their bodies and better for the planet.

Think of the hospital cafeteria as a garden. The researchers wanted to see if they could change the way the garden was tended to make the "fruit" (the food choices) healthier and more sustainable.

Here is the story of what they did and what happened, broken down simply:

The Mission: A Gentle Nudge

The team knew that telling people "eat less meat and more veggies" often doesn't work. So, instead of shouting instructions, they tried a mix of signs and structural changes.

They set up three different "gardens" (Canteens 1, 2, and 3) and tried different tricks in each:

• The Signs (Education): They hung up colorful posters showing the "Perfect Plate" (like a map for a healthy meal) and used QR codes for more info. They also put up "Social Norm" signs that said things like, "Hey, 80% of people here chose a salad today!" to make people feel like they were part of a healthy crowd.
• The Layout (Environment):
  • In one canteen, they moved the salad and fruit to the front of the line. Instead of grabbing a heavy steak first, you had to walk past the veggies first.
  • In another, they turned the salad and fruit into a self-service buffet, letting people grab as much as they wanted (within reason).
  • In the third, they just stuck to the posters because the kitchen was too small to change the layout.

The Results: A Mixed Bag

After a few months, they took photos of everyone's trays to see what actually happened. It was like checking the harvest after a season of gardening.

1. The Good News (The "Green" Wins):

• Less Beef: In the big canteens, people started picking beef less often. Since beef is like a "carbon monster" (it creates a lot of pollution), eating less of it lowered the environmental footprint of the meals.
• More Veggies: In the canteen where they put the salad at the front of the line and made it a buffet, people actually grabbed more salad. It's like when you put a bowl of candy at eye level; people grab it. When you put veggies at eye level, they grab those too.
• Better Fiber: Because people ate more veggies, their fiber intake went up in two of the canteens.

2. The Not-So-Good News (The "Sticky" Problems):

• Fat is Still High: Even with the posters and layout changes, the meals were still very greasy. It's like trying to make a healthy smoothie but accidentally adding a whole stick of butter. The fat content remained stubbornly high.
• Carbs are Low: The meals were missing out on healthy carbohydrates (like whole grains).
• The "Reverse Line" Didn't Work: In the canteen where they forced people to pick veggies before the main meat dish, it didn't really work. People still grabbed the meat. It seems that just changing the order of the line isn't enough if people are hungry, in a rush, or just really craving a burger.
• Fruit is Tricky: People didn't eat much more fruit. The researchers think this is because fruit is seen as a "dessert" or a snack, not a main part of the meal. Also, peeling an apple takes effort, while grabbing a pre-cut salad is easy.

The Big Picture: Why It Matters

The study shows that you can't just put up a sign and expect magic.

• Context is King: What worked in the big cafeteria (self-service salad) didn't work the same way in the small one.
• The Menu Matters Most: The researchers realized that even with great posters, if the kitchen doesn't offer tasty, healthy options, people won't eat them. It's like trying to teach someone to swim in a pool that has no water.
• Beef is the Villain: The biggest factor in the environmental impact was simply whether the meal contained beef. Removing beef from the menu is the single most effective way to lower the "carbon footprint" of a lunch.

The Takeaway

The hospital canteens are like a training ground for healthy habits. The study proved that while gentle nudges (signs and layout changes) can help a little bit—especially with getting people to eat more salad—they aren't a magic wand.

To truly fix the problem, the hospital needs to do two things at once:

1. Keep the nudges: Keep the signs and the smart layouts.
2. Fix the Menu: Actually cook and serve more delicious, plant-based meals that people want to eat.

If they do both, they might just turn the hospital cafeteria into a true model for a healthy, green future.

Technical Summary

1. Problem Statement

Hospital canteens represent a critical setting for public health intervention, serving both patients and a large workforce. However, existing literature indicates that food choices in these settings are often nutritionally suboptimal (high in lipids, low in carbohydrates and fiber) and associated with a high environmental impact (high carbon and water footprints). While political action and educational initiatives are necessary to shift dietary habits toward the "Planetary Health Diet," deeply ingrained eating habits are difficult to change. There is a need to evaluate the efficacy of combined educational (information dissemination) and environmental (nudging, structural changes) interventions in real-world hospital settings to determine if they can successfully guide users toward healthier and more sustainable food choices.

2. Methodology

Study Design:
A cross-sectional pre-post intervention study conducted in three hospital canteens (C1, C2, C3) in northeastern Italy.

• Timepoints: Two index weeks in September 2022 (T0, pre-intervention) and September 2023 (T1, post-intervention).
• Sample: 2,851 lunch trays analyzed (1,227 at T0; 1,624 at T1).
• Participants: Hospital employees. Data collection involved photographing trays, linking them to unique codes, and collecting socio-demographic data via questionnaires.

Intervention Strategy:
The intervention was tailored to the specific logistical and management constraints of each canteen (external vs. internal catering). It comprised three main pillars:

1. Educational Elements: Posters displaying the Mediterranean Diet Pyramid, the Harvard Healthy Eating Plate, and QR codes for information on ultraprocessed foods.
2. Environmental Restructuring (Nudging):
   • C1: Introduction of self-service for fruits, pulses, and salads (allowing portion control).
   • C2: Reversal of the serving flow (users select side dishes/salad before main courses).
3. Descriptive Norm Messages: Posters displaying statistics on peer behavior (e.g., "X% of users chose fresh fruit") to leverage social norms.

Data Analysis:

• Nutritional & Environmental Calculation: Recipes and portion sizes were matched with the Italian Food Composition Database and the SU-EATABLE LIFE dataset to calculate energy, macronutrients, Carbon Footprint (CF), and Water Footprint (WF).
• Visual Estimation: Two researchers independently estimated portion sizes from tray photos to adjust standard recipe data.
• Statistical Tests: Mann-Whitney U test for T0 vs. T1 comparisons within canteens; Kruskal-Wallis and Dunn's tests for inter-canteen comparisons; Chi-square tests for categorical variables (tray types).

3. Key Contributions

• Real-World Application: Unlike controlled trials, this study evaluates interventions in a complex, operational hospital environment with varying management structures (external vs. internal catering).
• Granular Data: The study moves beyond simple sales data by analyzing actual tray contents via photography, allowing for precise estimation of nutritional and environmental impacts per meal.
• Differentiated Intervention Analysis: It highlights how the same intervention strategies yield different results depending on the structural context (e.g., buffet style vs. reversed flow) and the specific food offer available.
• Sustainability Metrics: It integrates Carbon and Water footprints into the evaluation of hospital canteen performance, providing a holistic view of "sustainable nutrition."

4. Key Results

Nutritional Composition:

• Energy Intake: Results were mixed. C1 and C3 saw an increase in median energy (+30 kcal and +135 kcal, respectively), while C2 saw a decrease (-118 kcal).
• Macronutrients: Despite slight improvements, meals remained high in lipids (30–39% of energy) and low in carbohydrates (39–44% of energy) across all canteens.
• Fiber: Fiber intake fell within recommended ranges only in C1 and C3. In C2, fiber decreased significantly.
• Intervention Impact:
  • C1 (Self-service): The shift to self-service for vegetables and pulses correlated with a significant increase in salad/side dish selection (23.3% to 39.9%) and higher fiber intake.
  • C2 (Reversed Flow): The reversal of the serving flow failed to increase vegetable selection or fiber intake; in fact, fiber decreased. This suggests that structural changes alone, without appealing food options or in the absence of other supports, may not be effective.
  • C3 (Education + Norms): In the absence of structural changes, the combination of posters and social norms successfully increased side dish selection (63.7% to 74.5%) and fiber intake.

Sustainability (Environmental Impact):

• Carbon Footprint (CF):
  • C1 & C2: CF decreased significantly (C1: 1338 → 1227 gCO2eq; C2: 1094 → 966 gCO2eq). This was driven by a reduction in beef and cheese consumption.
  • C3: CF increased dramatically (773 → 1126 gCO2eq). This was attributed to the introduction of beef dishes in the menu during T1, which had zero presence in T0.
• Water Footprint (WF): Remained high across all canteens, with no significant reduction in C1 and C3, though C2 saw a decrease.
• Tray Categorization: Significant shifts occurred in tray composition. C1 and C2 saw a reduction in beef trays and an increase in fish/vegetarian options. C3 saw a dramatic shift from fish to beef trays due to menu changes.

Fruit Consumption:

• No significant increase in fruit consumption was observed in any canteen. The authors suggest fruit is perceived as an optional dessert rather than a structural meal component, making it less responsive to general educational nudges compared to vegetables.

5. Significance and Conclusions

• Partial Success: The intervention led to partial improvements. Educational posters and descriptive norms were effective in smaller, cohesive environments (C3) or when combined with self-service structures (C1). However, the "reversed flow" nudge (C2) was ineffective, highlighting that context matters.
• The Role of the Food Offer: The study concludes that educational and nudging interventions are insufficient if the underlying food offer is not aligned with sustainability goals. The drastic increase in CF in C3 was solely due to the introduction of beef, proving that menu composition is the primary driver of environmental impact.
• Future Directions: To achieve significant results, hospitals must implement parallel interventions on the food offer (e.g., increasing plant-based availability, reducing beef) alongside educational strategies.
• Broader Impact: Hospital canteens are powerful platforms for promoting public health. Even modest improvements in employee dietary habits can reduce long-term healthcare costs and foster a culture of health and sustainability within the healthcare sector.

Limitations: The study notes that daily menu variations (especially in the internally managed C3) and the lack of a uniform intervention across all sites make direct statistical comparisons between canteens difficult. Additionally, the analysis was limited to lunch meals and did not account for the rest of the participants' daily diet.