Higher Ultra-Processed Food (UPF) Intake Is Associated with Lower Food Literacy in Greek Adults with Overweight or Obesity: Results from a Cross-Sectional Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I would like to commend the author for this well-structured and clearly articulated manuscript entitled “Higher Ultra-Processed Food (UPF) Intake Is Associated with Lower Food Literacy in Mediterranean Adults with Overweight or Obesity.” The topic is timely and highly relevant, addressing an important public health issue within the Mediterranean context. The study provides valuable insights into the association between food literacy and dietary habits. However, several aspects require further clarification and improvement before the manuscript can be considered suitable for publication.

 

Title

Add "cross-sectional study" and specify "Greek" population.

 

Abstract

Report absolute median differences 

 

Introduction

Add  the secondary hypothesis posits BMI status as an effect modifier

 

Methods

Add  CONSORT-style flow diagram.

You mention cutting 49 people for "implausibly low" intake, what was the exact cutoff? 

Given that the 26-item food literacy instrument was originally developed and validated in a Korean population, the authors must provide a detailed rationale justifying its applicability to Greek adults.

A single 24-hour recall ranks individuals adequately but underestimates usual UPF intake due to high intra-individual variability (CV 25-40%). In Limitations, authors should acknowledge this regression dilution bias and note the absence of weekday/weekend distribution data, standardized composite dish deconvolution protocols.

Tables effectively utilize medians/IQR for non-normal data. Table 3 requires inclusion of household size (referenced in Methods). Tables 5-6 exhibit inconsistent presentation (medians mixed with means±SD); standardization to medians/IQR throughout is preferred, with sample sizes explicitly stated per cell. Absolute score differences with 95% confidence intervals (e.g., 14-point total literacy decrement in overweight/obesity HUPFI group) provide clinical context absent from p-values alone.

 

Discussion

I recommend that the authors discuss if low literacy causes UPF eating OR if UPF eating worsens literacy skills.

I recommend that the authors explicitly include and justify the following points in the study’s limitations: the potential bias associated with single-day recall; the limited representativeness of a predominantly young and highly educated sample compared to the general Greek population; the likelihood of underreported self-reported BMI; and the absence of blood tests or other objective measurements.

Author Response

I would like to commend the author for this well-structured and clearly articulated manuscript entitled “Higher Ultra-Processed Food (UPF) Intake Is Associated with Lower Food Literacy in Mediterranean Adults with Overweight or Obesity.” The topic is timely and highly relevant, addressing an important public health issue within the Mediterranean context. The study provides valuable insights into the association between food literacy and dietary habits. However, several aspects require further clarification and improvement before the manuscript can be considered suitable for publication.

 

We thank the Reviewer for the positive and constructive comments on our manuscript and for the time devoted to its evaluation. We have carefully addressed all comments and revised the manuscript accordingly, as detailed below.

 

Comment 1. Title: Add "cross-sectional study" and specify "Greek" population.

Reply 1. We thank the Reviewer for this input! Relevant changes have been made to the title. Please see the Title of the paper.

 

Comment 2.  Abstract: Report absolute median differences 

Reply 2. We thank the Reviewer for this helpful suggestion. We therefore focused on presenting the most essential results, including the absolute between-group differences and the subgroup analysis among participants with overweight/obesity. Due to the strict 200-word limit imposed on the Abstract, it was not possible to include further detailed information.

 

Comment 3. Introduction: Add the secondary hypothesis posits BMI status as an effect modifier

Reply 3. We thank the Reviewer for this valuable contribution! The secondary hypothesis has been added. Please see lines 87-88.

 

Comment 4. Methods: Add CONSORT-style flow diagram.

Reply 4. Thank you for this helpful suggestion! We have added a CONSORT-style flow diagram (Figure 1) detailing recruitment, exclusions, and analytic sample. Please see Figure 1.

 

Comment 5. You mention cutting 49 people for "implausibly low" intake, what was the exact cutoff? 
Reply 5. We thank the Reviewer for this comment. We assessed plausibility using the reported energy intake to basal metabolic rate ratio (EI:BMR). Participants reporting energy intake ≤ 85% of their estimated BMR were excluded as having implausibly low energy intake, consistent with the widely used EI:BMR plausibility approach and the Goldberg cut-off framework for identifying dietary misreporting and diet reports of poor validity in nutritional epidemiology. We have revised the results section of the manuscript to improve clarity. Please see lines 198 - 199.

Comment 6. Given that the 26-item food literacy instrument was originally developed and validated in a Korean population, the authors must provide a detailed rationale justifying its applicability to Greek adults.

Reply 6. We thank the Reviewer for this useful comment! As there is no food literacy questionnaire developed or validated in the Greek population, we have provided a detailed rationale justifying its applicability in our population. Please see lines 129 - 142.

 

Comment 7. A single 24-hour recall ranks individuals adequately but underestimates usual UPF intake due to high intra-individual variability (CV 25-40%). In Limitations, authors should acknowledge this regression dilution bias and note the absence of weekday/weekend distribution data, standardized composite dish deconvolution protocols.

Reply 7. We thank the Reviewer for this valuable comment. We have revised the Limitations section accordingly. Please see lines 338 - 354.

 

Comment 8. Tables effectively utilize medians/IQR for non-normal data.

Comment 8.1. Table 3 requires inclusion of household size (referenced in Methods).

Reply 8.1. We thank the Reviewer for this valuable contribution! Table 3 is updated with the inclusion of household size.
Comment 8.2. Tables 5-6 exhibit inconsistent presentation (medians mixed with means±SD); standardization to medians/IQR throughout is preferred, with sample sizes explicitly stated per cell. Absolute score differences with 95% confidence intervals (e.g., 14-point total literacy decrement in overweight/obesity HUPFI group) provide clinical context absent from p-values alone.

Reply 8.2.  We thank the Reviewer for this constructive comment. Following your suggestion, we have revised the manuscript accordingly and implemented the requested modifications in Tables 3, 4, 5, and 6.

 

Comment 9. Discussion: I recommend that the authors discuss if low literacy causes UPF eating OR if UPF eating worsens literacy skills.

Reply 9. We thank the Reviewer for this valuable comment! We have added a paragraph discussing the bidirectional relationship between food literacy and UPF consumption and acknowledged the limitations of the cross-sectional design regarding causal inference. Please see lines 258 – 270.

 

Comment 10: I recommend that the authors explicitly include and justify the following points in the study’s limitations: the potential bias associated with single-day recall; the limited representativeness of a predominantly young and highly educated sample compared to the general Greek population; the likelihood of underreported self-reported BMI; and the absence of blood tests or other objective measurements.

Reply 10. We thank the Reviewer for this comment. We have included these points in our limitations and use them, also, to form our thoughts on future research (stated in the conclusion section). Please see lines 338 - 376.

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for submitting this timely and relevant manuscript investigating the association between ultra-processed food (UPF) intake and food literacy among Mediterranean adults. Below are several critical points that may help strengthen the methodological clarity and overall impact of the article:

• Recruitment via snowball sampling and social media may lead to selection bias (e.g., younger, more health - conscious participants). Please clearly acknowledge this limitation and its implications for generalizability.
• It is unclear whether a power calculation was conducted to determine the appropriate sample size. This is particularly important given subgroup comparisons (e.g., LUPFI vs HUPFI). Please clarify.
• A single 24-hour recall may not adequately capture usual dietary intake. There should be at least 3 days including 1 weekend. Please discuss this limitation and consider recommending repeated recalls or FFQs in future research.
• Height and weight were self-reported, which may introduce inaccuracies in BMI classification. This should be explicitly stated as a limitation.
• Dividing participants based on the sample-specific median limits comparability with other studies. Consider referencing established UPF intake cutoffs (e.g., <30%, >50%).
• Please include internal consistency data (e.g., Cronbach’s alpha) for the Greek version of the food literacy questionnaire used in this study.
• Potential confounding variables (e.g., physical activity, income, urban/rural setting) were not accounted for.
• Expand the discussion to include possible psychological or contextual mechanisms linking low food literacy and high UPF intake, such as stress, cooking skills, or time scarcity.

 

 

Author Response

Thank you for submitting this timely and relevant manuscript investigating the association between ultra-processed food (UPF) intake and food literacy among Mediterranean adults. Below are several critical points that may help strengthen the methodological clarity and overall impact of the article:

 

We sincerely thank the reviewer for the positive and thoughtful evaluation of our manuscript and for highlighting its timeliness and relevance. We greatly appreciate the constructive comments and suggestions provided, which have helped us improve the methodological clarity and overall quality of the manuscript. Below, we address each point raised in detail and describe the corresponding revisions made to the text.

 

Comment 1. Recruitment via snowball sampling and social media may lead to selection bias (e.g., younger, more health - conscious participants). Please clearly acknowledge this limitation and its implications for generalizability.

Reply 1. We thank the Reviewer for this comment. The selection bias as well as the possible limitations for generalizability, are clearly stated as a consequence of our snowball sampling method in our limitations section. Please see lines 358 – 361 and 377 - 380.

 

Comment 2. It is unclear whether a power calculation was conducted to determine the appropriate sample size. This is particularly important given subgroup comparisons (e.g., LUPFI vs HUPFI). Please clarify.

Reply 2. We thank the Reviewer for this important point. Given that the primary between-group comparisons were conducted using a nonparametric Mann-Whitney (Wilcoxon rank-sum) test due to non-normality, we have now added a sensitivity power analysis in G*Power (v3.1) using the Wilcoxon-Mann-Whitney test (two groups), two-tailed, with α=0.05 and power=0.80.

Based on the achieved subgroup sizes, the minimum detectable effects (Cohen’s d) with 80% power were:

• Overall comparison (LUPFI n=150 vs HUPFI n=151): d≥0.35
• Normal-weight subgroup (LUPFI n=102 vs HUPFI n=88): d≥0.44
• Overweight/obesity subgroup (LUPFI n=48 vs HUPFI n=63): d≥0.58

We have added these details to the Methods section. Please see lines 181 – 189.

 

Comment 3. A single 24-hour recall may not adequately capture usual dietary intake. There should be at least 3 days including 1 weekend. Please discuss this limitation and consider recommending repeated recalls or FFQs in future research.

Reply 3. We thank the Reviewer for this valuable comment, with which we agree and thus, we have now addressed this limitation in the manuscript. Please see lines 338 – 354.

 

Comment 4. Height and weight were self-reported, which may introduce inaccuracies in BMI classification. This should be explicitly stated as a limitation.

Reply 4. We thank the Reviewer for this important comment. We agree that the use of self-reported height and weight may introduce measurement error and potential misclassification of BMI. Thus, we have added a relevant statement in the Limitations section. Please see lines 369 - 371.

 

Comment 5. Dividing participants based on the sample-specific median limits comparability with other studies. Consider referencing established UPF intake cutoffs (e.g., <30%, >50%).

Reply 5. We thank the Reviewer for this valuable comment! This point is addressed in lines 383 - 387.

 

Comment 6. Please include internal consistency data (e.g., Cronbach’s alpha) for the Greek version of the food literacy questionnaire used in this study.

Reply 6. We thank the Reviewer for this useful comment! We have now added internal consistency estimates for the Greek version of the Food Literacy questionnaire in the current sample. Please see lines 135 – 142.

 

Comment 7. Potential confounding variables (e.g., physical activity, income, urban/rural setting) were not accounted for.

Comment 7. We thank the Reviewer for this valuable comment! Unfortunately, we have not collected this data. We have added and discussed this limitation to the relevant section. Please see lines 361 – 367.

 

Comment 8. Expand the discussion to include possible psychological or contextual mechanisms linking low food literacy and high UPF intake, such as stress, cooking skills, or time scarcity.

Reply 8. We thank the Reviewer for this comment. We have expanded the Discussion to address potential psychological and contextual mechanisms linking food literacy and UPF intake, including stress, time scarcity, and cooking self-efficacy, supported by relevant literature. Please see lines 258 -270.

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors, I appreciate the opportunity to review your interesting and important work. I would like to offer some recommendations that I believe could improve your research:
- What did the 24-hour follow-up consist of? Didn't you consider that participants might not remember exactly everything they consumed during the reference 24 hours? I consider this a significant limitation.

- How were the online interviews conducted? Were they conducted via teleconference?

- What were the inclusion criteria? This wasn't clear in the text. Since recruitment was done through social media, how did you ensure that participants were truthful?

- Self-reported weight and height can lead to significant errors in the results. Did you offer any alternatives for those who didn't know their weight and/or height?

- What tool was used to assess food literacy? It's important to describe this because it's key to this research.

- What do you mean when you refer to household measures to estimate the amount of food and beverages consumed? The results and discussion are adequate.
Thank you.

Author Response

Dear authors, I appreciate the opportunity to review your interesting and important work. I would like to offer some recommendations that I believe could improve your research:

 

We thank the Reviewer for offering recommendations that will improve our work. We have addressed all comments and the relevant answers are provided below.

 

Comment 1. What did the 24-hour follow-up consist of? Didn't you consider that participants might not remember exactly everything they consumed during the reference 24 hours? I consider this a significant limitation.

Reply 1. We thank the Reviewer for this comment. The 24-hour dietary recall method is a widely used dietary assessment tool for dietetics practice. In this method participants are asked to recall all foods and beverages consumed in the previous 24 hours, including portion sizes, preparation methods, and, when possible, brand names. The method is typically interviewer-administered and often uses the multiple-pass approach to improve completeness and reduce recall bias. One of the method’s limitations is that it relies on participants’ memory, which may lead to underreporting or overreporting of foods and portion sizes. However, the standardized procedure followed (multiple pass protocol) ensures the minimization of such biases. To clearly state all the above we have added relevant information to existing text in the methods and limitations section. Please see lines 149 – 151 and 338 – 354.

 

Comment 2. How were the online interviews conducted? Were they conducted via teleconference?

Reply 2. We thank the reviewer for this comment. The online interviews were indeed conducted through tele-conference. This information was added. Please see line 101.

 

Comment 3. What were the inclusion criteria? This wasn't clear in the text. Since recruitment was done through social media, how did you ensure that participants were truthful?

Reply 3. We thank the Reviewer for this comment. We have amended the presentation of our inclusion and exclusion criteria to make them clearer (please see lines 110 - 117). During recruitment, a broad invitation via social media was used to reach a wide audience. The specific eligibility and exclusion criteria were applied and verified through screening questions during the assessment phase. To further address potential misreporting, participants identified as under-reporting energy intake, as well as extreme age and BMI values, were excluded from the analyses to enhance data reliability. These steps were taken to maximize the accuracy of the self-reported data. Apart from that, as all of these are self-reported data, even phase to phase collection would encompass this limitation. We have added this limitation (absence of objective measurements) in our limitation section (Please see lines 374 - 376).

 

Comment 4. Self-reported weight and height can lead to significant errors in the results. Did you offer any alternatives for those who didn't know their weight and/or height?

Reply 4. We thank the Reviewer for this important comment! We have added the relevant concern in the limitations section. Please see lines 369 - 371. In order to ensure a consistent and uniform data collection procedure across all participants, we did not provide alternative methods for those who might not have known their weight or height. Importantly, in our final sample, all participants were able to provide this information, so this issue did not occur in practice.

 

Comment 5. What tool was used to assess food literacy? It's important to describe this because it's key to this research.

Reply 5. We thank the Reviewer for this comment. You can find the description of the tool in the “2. Material and Methods” section, “2.3. Assessment” sub-section, second paragraph, lines 124 - 142.

 

Comment 6. What do you mean when you refer to household measures to estimate the amount of food and beverages consumed?

Reply 6. We thank the Reviewer for this comment. Household measures are for instance cups, used to measure vegetables, rice or spaghetti and tablespoons, used to measure oils. This is a method, among others, commonly applied in dietetic practice to help the public estimate their food and drink portion sizes (Moursi M et al. Nutrients 2025). Relevant examples of household measures were added within text to clarify this. Please see lines148 - 149.

Reference:

• Moursi M et al. Validation of Two Portion Size Estimation Methods for Use with the Global Diet Quality Score App. Nutrients 2025;17(9):1497

 

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for addressing the reviewer's comments. The authors have correctly implemented all the requested changes, with clear improvements in clarity, methodology description, and data presentation. No further revisions are needed in these aspects.

Author Response

We are pleased that the Reviewer is satisfied with the amendments made by our team in response to his/her comments. We also thank the Reviewer for the valuable feedback, which has helped improve our manuscript!

Reviewer 2 Report

Comments and Suggestions for Authors

The Abstract mentions 319 people, while the Results show 317 after exclusion criteria – this should be corrected.

Author Response

Comment 1. The Abstract mentions 319 people, while the Results show 317 after exclusion criteria – this should be corrected.

Reply 1. We thank the Reviewer for highlighting this point! We sincerely apologize for this oversight and have revised the manuscript accordingly. Please see abstract.

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors,
I sincerely appreciate your thorough responses to each of the issues raised, and I can only congratulate you on your interesting work. I encourage you to continue pursuing this important and engaging line of research. My sincere congratulations.

Author Response

We sincerely thank the Reviewer for the generous and encouraging comments! We greatly appreciate the positive evaluation of our work, as well as the constructive feedback provided throughout the review process, which we believe has helped us improve the quality of the manuscript. Thank you very much!