The Impact of Healthy and Sustainable Food Choices Interventions on Wasted Food: Nutritional and Environmental Aspects of Plate Waste Produced in Italian Hospital Canteens

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Although I acknowledge the merits of the authors’ research efforts, I believe that this topic requires a deeper analytical approach, particularly regarding modeling and data analysis, to meet the standards expected for publication in a high-impact journal.

Title
The title is not sufficiently clear regarding the main focus of the investigation.

Abstract
The writing needs improvement, as the ideas appear fragmented and the study design and main contributions are not clearly presented.

Keywords
The selected keywords do not adequately reflect the core themes of the manuscript or facilitate efficient indexing and retrieval.

Materials and Methods
Several calculations are not clearly described. The authors mention estimated values for energy, carbon footprint, and water footprint; however, the equations, assumptions, and methodological considerations used for these estimations are not included in the manuscript. Furthermore, it is unclear why these variables were selected if their correlations or relationships with the other study variables were not evaluated.

The terms “nutritional composition” and “environmental indicators” also require clarification. The manuscript should explain the criteria and rationale used for selecting these indicators.

The statistical analyses are mainly descriptive, and the study lacks multivariate analysis. I would recommend incorporating supervised learning or other advanced analytical approaches to strengthen the interpretation and predictive value of the results.

Conclusion
The first paragraph reads more as a summary of the study procedures than as a conclusion. This section should be rewritten to emphasize the main findings, implications, and contributions of the research.

Author Response

Responses to Reviewer #1.

Although I acknowledge the merits of the authors’ research efforts, I believe that this topic requires a deeper analytical approach, particularly regarding modeling and data analysis, to meet the standards expected for publication in a high-impact journal.

Response: We thank the reviewer for the feedback. To address the reviewer's comments, we performed additional analyses to further investigate factors associated with plate waste. The results are reported below, addressing each specific comment, and in the reviewed manuscript.

Title
The title is not sufficiently clear regarding the main focus of the investigation.

Response: We changed the title as follows: “The Impact of Healthy and Sustainable Food Choices Interventions on Wasted Food: Nutritional and Environmental aspects of Plate Waste Produced in Italian Hospital Canteens”

Abstract
The writing needs improvement, as the ideas appear fragmented and the study design and main contributions are not clearly presented.

Response: The abstract has been revised to incorporate the reviewer’s suggestions and to improve readability.

Keywords
The selected keywords do not adequately reflect the core themes of the manuscript or facilitate efficient indexing and retrieval.

Response: We thank the reviewer for the feedback. We have reviewed the keywords and proposed the following: plate waste; food waste; service food waste; workplace canteen; hospital cafeteria; sustainability; food choices

Materials and Methods
Several calculations are not clearly described. The authors mention estimated values for energy, carbon footprint, and water footprint; however, the equations, assumptions, and methodological considerations used for these estimations are not included in the manuscript.

Response: We thank the reviewer for the feedback. We added a more exhaustive explanation and 2 methodological citations in the manuscript. We did not include equations to calculate energy, carbon and water footprint in the text, because we used data from pre-existing databases, the Italian Food Composition Database for Epidemiological Studies in Italy (Gnagnarella et al., 2022; www.bda.ieo.it) for energy and nutrient data, and the Su-Eatable life database by Petersson et al (2021) for carbon and water footprint data. Estimations were made based on databases and using the EuroFIR standard methodology for calculating energy and nutrient composition of complex food recipes (reference added in the text). To clarify the process, the recipes and portion sizes for each dish offered during the observation period were requested from the canteen staff. Each ingredient of each recipe was matched with the foods of the Italian Food Composition Database for Epidemiological Studies (BDA) version 2022 and with those of the SU-EATABLE LIFE dataset to obtain the energy and nutrient profile as well as the environmental indicators (i.e., carbon footprint – CF; water footprint – WF) for each standard portion of the final served dish. This process was performed in accordance to the standard recipe approach (Vàsquez Caicedo et al., 2008). Then we calculated energy, nutrients and environmental indicators of the wasted part of each tray based on the amount of waste in grams for each recipe in each tray (with a proportion from the standard portion size in grams).

Furthermore, it is unclear why these variables were selected if their correlations or relationships with the other study variables were not evaluated.

Response: The primary aim of this study was to quantify and characterize from the nutritional and environmental perspective food waste in hospital canteens. Therefore, energy, nutrients, and environmental indicators estimations were needed to evaluate how much and what is wasted not only in grams or in terms of typology of foods but also in terms of energy, nutrients, GHG emissions and water. We believe that knowing the composition of wasted food can inform better waste preventing interventions and/or waste management policies.

The terms “nutritional composition” and “environmental indicators” also require clarification. The manuscript should explain the criteria and rationale used for selecting these indicators.

Response: We thank the reviewer for the suggestion and we added more information about this in the manuscript. The reported nutritional composition of PW included all the macronutrients: proteins, lipids, carbohydrates, sugars, fibre, and fatty acids, because they provide an overall evaluation of the nutritional quality of the meal as well as the discarded part of it and are commonly used in nutritional epidemiology to characterize dietary patterns. Environmental indicators were limited n this analysis to carbon footprint (CF) and water footprint (WF), as they are among the most widely used indicators for assessing the environmental sustainability of food systems and were the only indicators available for a consistent number of food ingredients and collected in a comprehensive and harmonized dataset (SU-EATABLE life), therefore resulting in accurate estimations with limited missing data. These indicators allow estimation of the additional environmental burden associated with a meal (that already has an impact on the environment) if part of it is wasted.

The statistical analyses are mainly descriptive, and the study lacks multivariate analysis. I would recommend incorporating supervised learning or other advanced analytical approaches to strengthen the interpretation and predictive value of the results.

Response: We thank the reviewer for this valuable suggestion. To strengthen the analytical framework beyond the descriptive analyses presented in the original version, we performed additional multivariable regression analyses. Specifically, we fitted a multivariable logistic regression model to identify independent predictors of plate waste occurrence (yes/no), including meal energy content, canteen site, timepoint (T0-T1), sex, age, work schedule and tray type (beef, meat without beef, fish, vegetarian, vegan) as covariates. We also conducted a multivariable linear regression analysis among participants with non-zero plate waste to investigate factors associated with the amount of waste generated. These analyses showed that canteen site and meal energy content (scaled per 100 kcal increase) were the main independent determinants of plate waste, whereas the other variables were not significantly associated with the outcome. The results of the multivariable logistic regression model have now been added to the manuscript and are reported in the revised results section. Although we considered more complex supervised learning approaches, the primary objective of this study was explanatory rather than predictive. We are currently applying deep learning in other epidemiological case–control studies; however, we believe that such analyses are beyond the scope of the present study. Therefore, we considered multivariable regression models to be the most appropriate analytical framework, as they allow direct estimation and interpretation of the independent associations between participant and meal characteristics and plate waste outcomes. The predictive performance of the logistic model was also evaluated (AUC = 0.62), showing limited discrimination, suggesting that the available variables explain only part of the variability in plate waste occurrence.

Conclusion
The first paragraph reads more as a summary of the study procedures than as a conclusion. This section should be rewritten to emphasize the main findings, implications, and contributions of the research.

Response: We thank the reviewer for the feedback, we have rewritten the conclusion based on both reviewers’ comments to emphasize the main findings and future implications or our research.

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript addresses a relevant topic and is supported by a solid dataset; however, it currently lacks sufficient analytical depth and conceptual integration. To meet publication standards, revisions are required, particularly regarding causal interpretation, control of confounding factors, and theoretical framing of the findings.

Required Revisions

Introduction

• Clarify explicitly the research gap
• Reformulate objectives to include an analytical/explanatory aim, not only descriptive outcomes.
• Introduce a conceptual framework linking food choice → consumption → waste.

Materials and Methods

• Justify and discuss explicitly the lack of a control group and its implications for causal inference.
• Clarify whether intervention components were identical across canteens; if not, describe differences structurally.
• Address explicitly the issue of menu variability (T0 vs T1) as a confounder.
• Specify how repeated observations per participant were handled (or justify why this is not relevant).
• Provide a clearer description of sampling and participation rate bias.
• Consider adding or justifying the absence of a multivariate statistical approach (e.g., models adjusting for age, sex, canteen, etc.).

Results

• Include at least one adjusted analysis (multivariate or stratified) to identify predictors of plate waste.
• Report effect sizes consistently, not only p-values.
• Improve clarity by reducing redundancy and focusing on key findings.
• Ensure consistency between tables, figures, and text interpretation.

Discussion

• Provide a coherent explanation of paradoxical findings (e.g., decrease in PW where intervention was less effective).
• Move beyond description and develop mechanistic/behavioral interpretations.
• Explicitly discuss the concept of mismatch between food selection and actual consumption.
• Strengthen comparison with existing literature using critical synthesis, not only reporting similarities/differences.
• Clearly separate results interpretation vs speculation.

Limitations

• Expand this section to explicitly include:
• lack of causal design
• menu variability
• intervention heterogeneity
• potential selection bias
• measurement limitations (visual estimation)

Conclusions

• Reformulate to avoid overgeneralization.
• Align conclusions strictly with results.
• Clearly state practical implications (policy / canteen management).
• Highlight the need for future controlled studies.

Figures and Tables

• Improve figure interpretability (reduce complexity, clarify legends).
• Ensure all figures add analytical value, not only descriptive repetition.
• Consider adding a conceptual figure summarizing findings.

Overall

• Reduce descriptive load and increase analytical depth.
• Strengthen internal coherence between objectives, methods, results, and conclusions.
• Ensure that claims are fully supported by the study design.

Author Response

Responses to Reviewer #2.

The manuscript addresses a relevant topic and is supported by a solid dataset; however, it currently lacks sufficient analytical depth and conceptual integration. To meet publication standards, revisions are required, particularly regarding causal interpretation, control of confounding factors, and theoretical framing of the findings.

Response: We thank the reviewer for the valuable feedback. To address the reviewer's comments, we performed additional analyses to further investigate factors associated with plate waste. The results are reported below, addressing each specific comment, and in the reviewed manuscript.

Required Revisions

Introduction:

Clarify explicitly the research gap.

Response: We thank the reviewer for this suggestion. We have revised the Introduction to better highlight the research gap addressed by the present study. The modified paragraph is the following: In the food service context, while several studies have investigated food waste and others have evaluated interventions aimed at promoting healthier and more sustainable food choices, evidence linking these two aspects remains limited. Some authors found no difference in PW after an intervention nudging healthier choices in a worksite canteen [10], while others observed that changing nutritional standards in school catering (e.g., requiring that one of the three components selected is a fruit or vegetable and increasing vegetable portion sizes and variety) also reduced vegetable waste, suggesting that inter-ventions promoting healthy diets may be part of food waste reduction strategies [11].  Therefore, additional evidence assessing both food choices and plate waste is essential to determine whether healthier and more sustainable selections are ultimately consumed rather than discarded, providing a more comprehensive evaluation of the intervention overall success.

 

Reformulate objectives to include an analytical/explanatory aim, not only descriptive outcomes.

Response: We appreciate this comment and the suggestion to incorporate a more advanced analytical approach to strengthen the study. The study objectives and data collection procedures were defined a priori and were primarily designed to describe plate waste and assess changes in food choices and plate waste following the intervention. For this reason, the original analyses focused mainly on descriptive and comparative outcomes. Nevertheless, in response to the reviewer's suggestion, we have strengthened the analytical component of the study by performing additional multivariable regression analyses to identify independent determinants of plate waste occurrence and quantity. These analyses have been incorporated into the revised manuscript and provide a more explanatory interpretation of the findings.

 

Introduce a conceptual framework linking food choice → consumption → waste.

Response: We agree with the reviewer and have expanded the introduction to better emphasize the conceptual relationship between food choice, food consumption, and food waste: “These processes can be viewed as interconnected stages of the same behavioral pathway. In food service settings, individuals select foods according to personal preferences as well as contextual and environmental factors. These choices influence actual consumption patterns and, consequently, the quantity and type of food left uneaten”. Therefore, interventions promoting healthier and more sustainable food choices should be evaluated not only in terms of food selection but also in terms of actual consumption and waste generation. From this perspective, plate waste represents an important indicator of the nutritional, environmental, and practical effectiveness of interventions targeting food choice behaviors.

Materials and Methods

Justify and discuss explicitly the lack of a control group and its implications for causal inference.

Response: We thank the reviewer for this important observation. The study adopted a descriptive design with 2 timepoints, which allows to evaluate changes over time in real-world settings but does not allow causal inference. No control group was included because the intervention was implemented at the organizational level across all participating hospital cafeterias as part of a broader institutional health promotion initiative. Educational material is exposed in the common spaces of the hospital canteen; thus, all users were inevitably exposed to the intervention. We added a specification in the text. Moreover, it was not possible to select a comparable cafeteria without exposure to the intervention, because, each cafeteria is different and, as seen in the latest results, cafeteria site is one of the main determinants of PW. We acknowledge that factors other than the intervention may have contributed to the differences observed between T0 and T1. To address this limitation, we have explicitly acknowledged it in the revised Discussion section and have complemented the descriptive analyses with multivariable regression models to better characterize the determinants of plate waste.

 

Clarify whether intervention components were identical across canteens; if not, describe differences structurally.

Response: The intervention components were not identical across the participating canteens. The specific characteristics of the intervention implemented in each canteen are described in detail in Section 2.1 Food Choice Intervention. These differences reflect the organizational and operational characteristics of each setting and were taken into account in the analyses by considering cafeteria site as a study variable and by describing all PW variables separately by canteen.

 

Address explicitly the issue of menu variability (T0 vs T1) as a confounder.

Response: We thank the reviewer for highlighting this point. Menu variability between T0 and T1 may represent a potential source of confounding and has now been explicitly addressed in both the Materials and Methods and the Limitations sections. For canteen 1 and 2, the food service provider remained the same at both timepoints. Therefore, menus were developed according to the same nutritional and organizational guidelines, although some variation in specific recipes and dishes may have occurred from one year to the next. In C3, which is managed internally and does not follow predefined cyclical menus, greater variability in the dishes offered may have occurred. We acknowledge that differences in menu composition between T0 and T1 could have influenced both food choices and plate waste independently of the intervention.

 

Specify how repeated observations per participant were handled (or justify why this is not relevant). Response: Since participants could visit the cafeteria multiple times during the observation week, repeated observations for the same individual in a different day (therefore with a different meal) were included in the main analysis. The primary objective of the study was to characterize the overall food waste generated in the cafeteria rather than to estimate individual behaviors; therefore, the trays were initially analyzed as independent observations without considering the users whom they belong. However, in order to reduce the potential influence of repeated measurements by regular users, additional analyses stratified by sex and age were conducted, in that case considering only the first recorded visit for each participant. This was reported in section 2.4 Statistical Analysis.

 

Provide a clearer description of sampling and participation rate bias.

Response: We provided a clearer description in the limitation section. Since participation in the study was not mandatory, it is plausible that people who were more interested in the topic or had more time during lunch break, were more likely to participate. However, response rate was almost 50% of users in all canteens.

 

Consider adding or justifying the absence of a multivariate statistical approach (e.g., models adjusting for age, sex, canteen, etc.).

Response: We thank the reviewer for this valuable suggestion. We performed additional analyses to further investigate factors associated with plate waste and added in the text results from a multivariate logistic regression model.

First, we fitted a multivariable logistic regression model using the occurrence of plate waste (yes/no) as the dependent variable and including meal energy content, cafeteria site, sex, age, work schedule, tray type (beef, meat without beef, fish, vegetarian, vegan), and timepoint (T0, T1) as covariates. The analysis showed that higher meal energy content (scaled per 100 kcal increase) was significantly associated with a greater likelihood of plate waste (p < 0.001). Moreover, compared with Cafeteria 1, participants eating in Cafeteria 2 (OR = 0.46; 95% CI: 0.36–0.57) and Cafeteria 3 (OR = 0.53; 95% CI: 0.41–0.68) had a significantly lower probability of generating plate waste. No significant associations were observed for sex, age, work schedule, tray type, or timepoint.

Second, among participants who generated plate waste, we performed a multivariable linear regression analysis using the amount of waste (g) as the outcome. Greater meal energy content was associated with higher amounts of waste, whereas Cafeterias 2 and 3 were associated with significantly lower waste quantities compared with Cafeteria 1. Male sex was also associated with lower waste amounts. Then, we evaluated model assumptions and performed a sensitivity analysis using a log-transformed waste variable because of the marked right-skewness of the waste distribution. Although the transformation improved residual distribution, some deviations from normality persisted. Therefore, results from the linear model should be interpreted with caution, and we decided to report in the manuscript only data from the multivariate logistic regression.

Results

Include at least one adjusted analysis (multivariate or stratified) to identify predictors of plate waste.

• Report effect sizes consistently, not only p-values.
• Improve clarity by reducing redundancy and focusing on key findings.
• Ensure consistency between tables, figures, and text interpretation.

Response: We thank the reviewer for these valuable suggestions. We performed a multivariable logistic regression analysis and reported the results in this section, as described in the above point of this review report (Materials and methods). We also reported ORs and AUC. Then we improved the result section reducing redundancies (directly referring to data in tables) and focusing on key data, and checked for consistency between tables, figures, and text interpretation

Discussion

Provide a coherent explanation of paradoxical findings (e.g., decrease in PW where intervention was less effective).

Response: We thank the reviewer for this insightful comment. We agree that the apparent reduction in plate waste in the canteen where the intervention appeared less effective in modifying food choices deserves further studies. One possible explanation is that users who changed their food choices in response to the intervention may have selected foods perceived as healthier or more sustainable, such as vegetables, without these foods necessarily matching their personal preferences. Consequently, part of the selected food may have been wasted. Conversely, users who maintained their usual food choices may have selected foods more closely aligned with their preferences and therefore consumed a larger proportion of the meal. This interpretation highlights the importance of considering not only the nutritional and environmental quality of food choices, but also the palatability and acceptability of the foods offered. Promoting healthier and more sustainable choices may not automatically translate into lower plate waste if the selected foods are not fully consumed. However, because no information on user satisfaction, food preferences, or motivations for plate waste was collected (as reported in the limitations), this explanation remains speculative. We have clarified this point in the Discussion section. Another point already described in this section that may partially explain this paradox is that in C2 where the intervention was less effective, meals chosen were less energetic, and this is related to the final amount of PW.

 

Move beyond description and develop mechanistic/behavioral interpretations.

Response: We appreciate this suggestion. We have expanded the Discussion section with results from the regression model to provide a broader interpretation of the findings, particularly regarding the relationship between food choice and plate waste. Nevertheless, the information collected in the present study does not allow a detailed investigation of the psychological and behavioral mechanisms underlying users' choices and consumption patterns. A more comprehensive understanding of these mechanisms would require the assessment of factors such as values, attitudes, motivations, food preferences, and contextual influences. Future research could benefit from the adoption of behavioral frameworks, such as the Value–Attitude–Behaviour (VAB) hierarchy or similar models, combined with qualitative and quantitative approaches aimed at investigating the determinants of food choices, food consumption, and plate waste in institutional food service settings.

 

Explicitly discuss the concept of mismatch between food selection and actual consumption.

Response: We agree with the reviewer and have strengthened the manuscript by explicitly discussing the potential mismatch between food selection and actual consumption. We introduced a conceptual framework linking food choice, food consumption, and plate waste in the Introduction and further developed this concept in the Discussion. The results highlight that improvements in food choices do not necessarily correspond to improvements in actual food consumption. In 2 canteens, the intervention appeared to encourage healthier food selections, particularly vegetables, but this was not consistently accompanied by reductions in plate waste. These findings suggest that food choice and food consumption should be evaluated jointly when assessing the effectiveness of interventions aimed at promoting healthier and more sustainable eating behaviors.

 

Strengthen comparison with existing literature using critical synthesis, not only reporting similarities/differences.

Response: We thank the reviewer for this suggestion, however, direct comparisons between studies remain challenging because workplace and institutional food service systems differ substantially in terms of service model, menu planning, portioning practices, pricing systems, and cultural context. Nevertheless, our findings appear more consistent with studies conducted in European institutional food service settings than with those conducted in other geographical contexts. Furthermore, available evidence suggests that service organization may substantially influence plate waste generation, with fixed-portion systems generally showing different waste patterns compared with buffet-style services or pay-by-weight models. We have reported in the text this: “Despite the extreme heterogeneity in adult food service establishments regarding cultural context, service model, menu planning, portioning practices, pricing systems, which makes comparison challenging, …..”

 

Clearly separate results interpretation vs speculation.

Response: We thank the reviewer for this observation. We have revised the Discussion to more clearly distinguish interpretations directly supported by the study findings from hypotheses and speculative explanations. Whenever behavioral or organizational mechanisms are proposed to explain the observed results, these are now explicitly presented as possible interpretations rather than conclusions supported by direct evidence.

Limitations

• Expand this section to explicitly include:
• lack of causal design
• menu variability
• intervention heterogeneity
• potential selection bias
• measurement limitations (visual estimation)

Response: We thank the reviewer for this suggestion. The Limitations section has been expanded to explicitly state the lack of a causal study design, menu variability between T0 and T1, heterogeneity of the interventions across cafeterias, and visual estimation. The selection bias was already stated, but we expanded the explanation.

Conclusions

• Reformulate to avoid overgeneralization.
• Align conclusions strictly with results.
• Clearly state practical implications (policy / canteen management).
• Highlight the need for future controlled studies.

Response: We thank the reviewer for these suggestions. The Conclusions section has been revised to ensure that all statements are fully supported by the study findings and study design. We reduced potentially overgeneralized statements, focusing on practical implications, and explicitly highlighted the need for future controlled and/or quasi-experimental studies to better assess the causal effects of interventions promoting healthier and more sustainable food choices.

 

Figures and Tables

• Improve figure interpretability (reduce complexity, clarify legends).
• Ensure all figures add analytical value, not only descriptive repetition.
• Consider adding a conceptual figure summarizing findings.

Response: We thank the reviewer for these constructive comments. Figures and tables were revised to improve readability, including clarification of legends. We carefully reviewed all figures to ensure that they add meaningful information. Given the substantial heterogeneity among the participating cafeterias and the context-specific nature of the findings, we preferred not to summarize the results in a single conceptual figure, as this could oversimplify relationships that are likely to be multifactorial and setting-dependent.

Overall

• Reduce descriptive load and increase analytical depth.
• Strengthen internal coherence between objectives, methods, results, and conclusions.
• Ensure that claims are fully supported by the study design

Response: We thank the reviewer for these valuable comments. The manuscript has been extensively revised to improve its analytical depth and overall coherence. In particular, we strengthened the analytical component through the inclusion of multivariable regression analyses aimed at identifying independent determinants of plate waste occurrence and quantity. We also revised the Introduction to better define the research gap and conceptual framework linking food choice, food consumption, and plate waste. The Methods, Results, Discussion, and Conclusions sections were subsequently updated to ensure consistency with the revised analytical approach. Finally, all interpretations and conclusions were carefully reviewed and rephrased where necessary to ensure that they remain fully supported by the study design and do not imply causal relationships.

Reviewer 3 Report

Comments and Suggestions for Authors

I send a review of manuscript (ID: applsci-4316751) of the authors: Federica Fiori , Diana Menis , Elisa Mansutti , Caterina Liudmila Graziani , Peter Cautero , Daniela Zago , Marco Driutti , Lucia Lesa , Enrico Scarpis , Alessandro Conte , Lucrezia Grillone , Francesca Cortelazzo , Angelica Cosolo , Manuela Mauro , Laura Brunelli * , Maria Parpinel „”.

 

I think that the authors have undertaken an interesting and important study, one that is highly topical, particularly as the problem of food waste is becoming increasingly widespread in the context of various forms of mass catering. It is therefore important to encourage initiatives that promote healthy and sustainable dietary choices, particularly given the food shortages in certain regions of the world. I would like to emphasise that I appreciate the authors’ work in undertaking this research topic.

I think that this manuscript is relevant to the publication in Applied Sciences, but the authors should make a minor revision.

 

1. Introduction:

 

Page, lines 69-70 - … food waste still reached about 132 kg per person in 69
one year (data from 2022)…

Page 3, line 98 - The aim of the work: I would ask the authors, despite the previous explanation of the abbreviation ‘PW’, to use either the full name or the abbreviation together with the full name in this section of the paper.

 

2. Materials and Methods

 

Pages 4-5, lines 176-210 (Statistical analysis) - I think that this section should include information on the standard deviation (± SD) and the number of replicates (n=?), and that homogeneous groups (a,b,c…?) should be identified, with a specified LSD (Least Significant Difference). This information should also be included below the relevant tables – e.g.: Table 4.

 

3. Results

 

Page 5, line 211 - In this section, please include the information mentioned above – supplementary details regarding the statistics.

 

 

Tables

Comments as before.

 

References:

All of the literature presented in the table of contents [1-47] was included in the manuscript.

 

                                                    

 

 

 

Comments for author File: Comments.pdf

Author Response

Responses to Reviewer #3.

Introduction:

Page, lines 69-70 - … food waste still reached about 132 kg per person in
one year (data from 2022)…

Response: We modified the sentence as follows, to clarify it: “food waste still reached about 132 kg per person in 2022”

Page 3, line 98 - The aim of the work: I would ask the authors, despite the previous explanation of the abbreviation ‘PW’, to use either the full name or the abbreviation together with the full name in this section of the paper.

Response: We have updated the text based on the reviewer’s suggestion. 

Materials and Methods Pages 4-5, lines 176-210 (Statistical analysis) - I think that this section should include information on the standard deviation (± SD) and the number of replicates (n=?), and that homogeneous groups (a,b,c…?) should be identified, with a specified LSD (Least Significant Difference). This information should also be included below the relevant tables – e.g.: Table 4. Results Page 5, line 211 - In this section, please include the information mentioned above – supplementary details regarding the statistics.  Tables Comments as before.

Response: We thank the reviewer for this comment and for the opportunity to clarify the statistical reporting. Standard deviations are reported when appropriate for the overall sample, therefore we added this missing information in the materials and methods. However, for plate waste variables, the distribution was highly right-skewed and characterized by a large proportion of zero values. Therefore, when analyses were restricted to trays with non-zero waste, data are presented as medians and 25th–75th percentiles, which provide a more appropriate summary of non-normally distributed variables. Regarding the request for homogeneous groups and LSD values, we note that these approaches are typically applied in experimental designs involving multiple pairwise comparisons among treatment groups. In the present study participants evaluated at T0 and T1 were not paired and may have represented different individuals. Consequently, the study was not designed to assess treatment effects within the same subjects over time. The Kruskal–Wallis test was used only to explore differences among canteens at each time point and not to estimate the effect of canteen on the intervention outcome. For this reason, canteens were analyzed separately and no post-hoc grouping procedure based on LSD was performed.

To further strengthen the statistical analysis, rather than performing multiple pairwise post-hoc comparisons, we have now included additional multivariable regression models investigating factors associated with both the occurrence and the amount of plate waste. These analyses confirmed that cafeteria site and meal energy content (scaled per 100 kcal increase) were the main independent predictors of plate waste.

 

References: All of the literature presented in the table of contents [1-47] was included in the manuscript.

Response: We thank the reviewer for the check.

Reviewer 4 Report

Comments and Suggestions for Authors

The manuscript has improved significantly following the first review cycle. The authors have successfully addressed the main structural concerns. From my analysis, a final round of minor editing is required before formal acceptance.

ABSTRACT

The text states that Romanian canteens/genotypes had the highest plate waste, but later implies that the canteen with the fewest improvements in food choices (C2) saw a significant decrease. Please ensure that the distinction between the baseline contribution and the post-intervention shift is made explicitly clear to the reader.

INTRODUCTION

Lines 59–61: The phrase "in addition to 13% lost in the early stages of the food supply chain" is structurally repetitive with the following sentence. Consider rewriting it for greater conciseness.

The text mentions: "higher adherence to the Italian dietary guidelines' indicator was associated with better food waste behaviors." Please briefly define what "better behavior" means in this context (e.g., lower waste production or higher awareness) to avoid ambiguity.

RESULTS

Figure 1: The food categories listed in the legend block include a generic item called "Hard grating cheese." Given that it visually represents an insignificant percentage contribution by weight, please indicate in the methodology or in the caption whether this was evaluated as part of the first/second course dressing or as an independent item.

DISCUSSION

The phrase "a speculative interpretation of this results is that..." contains a grammatical number mismatch. Please change it to "these results" or "this result." (Lines 413–415)

The text states: "due to their high environmental load deriving not to high environmental indicators themselves but to their high wasted amount." The preposition "deriving not to" is grammatically incorrect. Suggestion: "...due to an environmental load driving force resulting not from high indicators themselves, but from the large quantities wasted." (Lines 469–471)

The sentence "the intervention appears to have increased vegetable choices... However, we found that while salad waste decreased, side dish waste parallelly increased..." represents an excellent critical point. Please expand slightly on whether this could be attributed to preparation quality, presentation, or over-portioning of cooked side dishes compared to raw salads. (Lines 519–521)

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Although I appreciate the authors' efforts and responses to the reviewers' comments, most of the concerns raised were not sufficiently addressed. Since the initial review, I have noted several methodological weaknesses and concerns regarding the manuscript's fit within the journal's scope. Furthermore, the incorporation of more current and robust data analysis techniques is necessary to enhance the scientific rigor of the study and make it suitable for publication.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The concerns raised during the previous review round have been adequately addressed. The revised manuscript provides greater methodological transparency, includes additional statistical analyses, offers a more balanced interpretation of the findings, and better acknowledges the limitations of the study.

In my opinion, the major scientific issues have been satisfactorily resolved, and the manuscript now provides a meaningful contribution to the literature on food choice interventions and plate waste in institutional food service settings. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf