Adopting the Mediterranean Diet: Motivational and Socio-Cognitive Processes in Young Adults

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Abstract:

the abstract should contain same numerical results

in the abstract, conclusions should point out the impact of the results

Main text:

Introduction is too long. Please focus on the main points strictly related to the aim and focus of the manuscript.

At the end of the introduction the specific study's aim is missing. please add

Clarify the measurement model and what is “latent” vs “observed”

Participants are “primarily” university students, and some non-students are included, but the manuscript lacks detail (e.g., proportions, SES proxies, living arrangement, region, employment). These variables can strongly influence MD adherence and perceived control in emerging adulthood.

Consider adding age (and possibly gender) as covariates predicting intention/adherence, or justify why they were excluded.

Briefly discuss overlap and why all regulations were retained simultaneously.

Consider a sensitivity analysis (e.g., combining autonomous regulations, or testing alternative SDT structure) or at minimum acknowledge potential multicollinearity effects.

In Table 3, for external regulation → intention via PBC, the indirect effect CI appears to include zero (reported as -0.258 to 0.024), yet “Mediation = Yes”. If the CI truly crosses zero, mediation should not be claimed. Re-check Table 3 values, significance flags, and narrative interpretation in Results/Discussion accordingly.

Table 2: the correlation between intention and adherence is shown as “.30” without significance marker while others have * / **; please verify formatting and p-values.

 

Author Response

Abstract:

the abstract should contain same numerical results

in the abstract, conclusions should point out the impact of the results

Response: We appreciate the reviewer’s comment. We have revised the abstract to report the main numerical results (standardized coefficients and explained variance) and to better highlight the impact of the findings for healthcare practice and intervention design. These revisions have been implemented in the Abstract section (page 1, lines 24-36).

 

Main text:

Introduction is too long. Please focus on the main points strictly related to the aim and focus of the manuscript.

Response: We appreciate the reviewer’s comment regarding the length and focus of the Introduction. In response, we have revised this section to improve clarity and alignment with the specific aim of the study, rather than simply reducing its length. Specifically, the revised version now articulates the study aim more explicitly and reorganizes the theoretical background to more clearly emphasize the constructs and frameworks directly relevant to the research questions. These revisions can be through the introduction section, particularly on page 3-4, lines 106-149.

At the end of the introduction the specific study's aim is missing. please add

Response: We thank the reviewer this suggestion. We have added an explicit statement of the study aim at the end of the Introduction. This revision can be found on page 3, lines 136–140.

 

Clarify the measurement model and what is “latent” vs “observed”

Response: We thank the reviewer for this comment. We have clarified the specification of the measurement model by explicitly distinguishing between latent and observed variables in the Analysis section. In particular, we now specify which constructs were modeled as latent variables (motivational regulations, perceived behavioral control, and intention) and which variables were treated as observed indicators (adherence to the Mediterranean diet and covariates). These revisions can be found on page 5, lines 232–235.

 

Participants are “primarily” university students, and some non-students are included, but the manuscript lacks detail (e.g., proportions, SES proxies, living arrangement, region, employment). These variables can strongly influence MD adherence and perceived control in emerging adulthood.

 

Response: We thank the reviewer for this important comment. In the present study, participants were primarily university students, with a smaller proportion of non-students included. This information is now clarified in the Participants section (page 4, lines 153-160). The questionnaire collected basic sociodemographic variables (age, gender, university affiliation) but did not include more detailed indicators of socioeconomic status, living arrangements, or employment status. We have now explicitly acknowledged this limitation in the Discussion, noting that future research should include a broader set of contextual variables to better capture the complexity of dietary behavior in emerging adulthood. These clarifications have been added to the Limitations section (page 12, lines 462–466).

 

Consider adding age (and possibly gender) as covariates predicting intention/adherence, or justify why they were excluded.

Response: Thank you very much for this suggestion. We agree that age and gender represent important variables that may influence intention and adherence.

We would like to clarify that both age and gender were already included as covariates in the statistical model, as reported in Figure 1. However, we acknowledge that this information was not explicitly mentioned in the Methods section, and we apologize for this oversight.

We have now revised the text of the manuscript to clearly report the inclusion of age and gender as covariates in the analysis. This information has been added in the Analysis section (page 6, lines 245-246).

Briefly discuss overlap and why all regulations were retained simultaneously.

 

Consider a sensitivity analysis (e.g., combining autonomous regulations, or testing alternative SDT structure) or at minimum acknowledge potential multicollinearity effects.

 

Response: We thank the reviewer for these thoughtful comments. We agree that some motivational regulations that are theoretically adjacent along the Self-Determination Theory continuum show substantial empirical overlap, which is conceptually expected and reflects their proximity rather than redundancy. We have now explicitly discussed this overlap in the Discussion and clarified why all motivational regulations were retained simultaneously, in order to preserve conceptual specificity and to examine potentially distinct indirect pathways associated with different regulatory styles.

In addition, we now explicitly acknowledge that such overlap may raise concerns regarding multicollinearity and note that more parsimonious modeling approaches (e.g., higher-order factors or composite scores) or formal sensitivity analyses could be considered in future research. These clarifications have been added to the Discussion section (page 11, lines 403–415).

 

 

In Table 3, for external regulation → intention via PBC, the indirect effect CI appears to include zero (reported as -0.258 to 0.024), yet “Mediation = Yes”. If the CI truly crosses zero, mediation should not be claimed. Re-check Table 3 values, significance flags, and narrative interpretation in Results/Discussion accordingly.

Response: Thank you very much for this careful observation. We have re-checked the values reported in Table 3, the corresponding confidence intervals, and the significance flags. You are correct in noting that the reported confidence interval for the indirect effect of external regulation on intention via PBC (CI: −0.258 to 0.024) includes zero. The indication “Mediation = Yes” was therefore incorrect and resulted from a reporting error. We have now corrected Table 3 by removing the mediation flag for this pathway. In addition, we have revised the Results and Discussion sections to ensure that no mediation effect is claimed for this relationship. The corrections can be found in the revised Table 3 (page 9) and in the Results (page 9, line 330) and removed all these from the discussion.

Table 2: the correlation between intention and adherence is shown as “.30” without significance marker while others have * / **; please verify formatting and p-values.

Response: Thank you for pointing this out. We have carefully checked the formatting and the corresponding p-values in Table 2. The correlation between intention and adherence (r = .30) is statistically significant, and the absence of the significance marker was due to a formatting oversight. We have now corrected the table by adding the appropriate significance indicator, which is p < .01 (**). The revised version of Table 2 is reported in the manuscript (pages 8–9) with the correct notation. In addition, the table has been expanded to include descriptive statistics (means, standard deviations, and observed ranges).

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Significant matters that should be enhanced (methods/reporting - needed prior to acceptance) The measuring model (CFA) is not clearly stated. According to you, CFA was conducted to determine construct and discriminant validity, whereas the necessary CFA outputs are lacking in the manuscript so that readers can ascertain the quality of measurements. What is lacking (minimal expectations): Factor loading standardization of every item (with definite cut-off rule, e.g., 0.50 or more). The convergent validity (e.g., CR and AVE per construct, or at least average loadings + item reliability). Discriminant validity evidence (e.g. HTMT or FornellLarcker). The reason why this is important in your paper: In your correlation table, a number of motivation subscales are moderately-and-strongly correlating (about some range of 0.65-0.72). At such high levels of correlations, one will need to find some evidence of discriminant validity to make sure that constructs are not empirically redundant.

Such under-documented and under-justified motivation scale is shortened motivation scale (REBS). You had fewer items per subscale (3 items per subscale; 2 items amotivation). This is a substantial methodological determination since shortening may alter factor structure and weaken validity. What you must add: Specifically what items were kept (put the wording of the item in an Appendix). The reason why the items were chosen (e.g. the greatest loadings in earlier research, cultural adaptation limitations, pilot performances). Evidence that the abridged version still creates the desired multi-factor structure in your sample (CFA results and indices of reliability/validity of each factor).

The details of SEM estimations are not specific enough to Likert/ordinal data and missingness. Your variables have been calculated based on a Likert-type items and summed scores, but the manuscript fails to explain how the estimation was matched with the data properties. Please indicate (in the Statistical Analysis section): SEM version and software used. Type of estimator (e.g., ML, MLR, WLSMV) and their selection depending on the nature of variables and distribution. The treatment of non-normality (robust standard errors? bootstrapping?). The treatment of missing data (listwise deletion versus FIML versus multiple imputation). And the percentage missing.

There is an actual threat of common-method bias, which is not properly treated. All the constructs (motivation, intention, PBC, adherence) would be self-reported and measured in a single survey session. This inflates common variance and may bias path estimates. Elaboration Minimum improvement: include a special limitation with a short description of possible inflation. Greater improvement (preferable): include a plain diagnostic (e.g., marker variable approach or a factor sensitivity check on a latent method) where available. There should be modest explained variance in the main outcome through interpretation. You state that the model describes that the adherence is explained by R 2 = 16%. It is not insignificant, and yet not high. Adjustment that is needed in Discussion: Current adherence as affected by other determinants unmeasured by the model (e.g., food environment/access, cooking resources, time constraints, affordability, campus availability, social environment, cultural eating patterns). Do not use words like strongly predicted adherence when explained variance is small.

Minor problems (presentation/clarity — correct during revising) Issue in format/statistical notation in Tables. Table 2 indicates a correlation between adherence and intention as r = 0.30 though it is not significant as indicated even when smaller correlations are significant. Action: re-visit p-values and make sure that there is a consistent use of significance symbols and table footnotes on all cells. Terminology/spelling consistency. It contains a number of typos and irregular spelling/capitalization (e.g., Mediterrenean vs. Mediterranean).
QueMD scoring should be more justifiable and interpretable. According to your description, there are items that have 0 points with or without consumption. Such may diminish sensitivity and make interpretation more difficult in young adult sample. Action: explain the rationale behind the use of this scoring rule, and ensure that it aligns with the desired interpretation of the dietary adherence measure. Covariates and robustness. Your age report indicates that there is a correlation with adherence (r = 0.11). Although sex differences may not be significant when comparing both variables at once, it would be more credible to include the information that age and sex were covariated in SEM (or to provide the reason why not).

Comments on the Quality of English Language

The English can be mostly understood, yet it needs professional editing due to: consistency of spelling and terms (US vs UK; repeated typos), punctuation and structure of the sentences (there are some long and unclear sentences), regular application of technical terms (names of constructions and abbreviations). Readability and perceived rigor will be significantly enhanced with a careful edit.

Author Response

Significant matters that should be enhanced (methods/reporting - needed prior to acceptance).

The measuring model (CFA) is not clearly stated. According to you, CFA was conducted to determine construct and discriminant validity, whereas the necessary CFA outputs are lacking in the manuscript so that readers can ascertain the quality of measurements.

Response: We thank the reviewer for this important comment. We have substantially revised the manuscript to explicitly report the CFA results and to clarify the role of the measurement model in assessing construct and discriminant validity. In particular, we have added a dedicated Measurement Model section in the Results, where we now report the overall fit indices of the CFA (χ², χ²/df ratio, CFI, TLI, RMSEA, SRMR), a summary of the standardized factor loadings for all items, and the correlations among latent factors. We also clarify which constructs were modeled as latent variables in the Analysis sections.

These revisions can be found in the Measurement Model subsection of the Results (page 7, lines 276–297) and the analysis section (page 5, lines 232-235).

What is lacking (minimal expectations): Factor loading standardization of every item (with definite cut-off rule, e.g., 0.50 or more). The convergent validity (e.g., CR and AVE per construct, or at least average loadings + item reliability). 

Discriminant validity evidence (e.g. HTMT or FornellLarcker). The reason why this is important in your paper: In your correlation table, a number of motivation subscales are moderately-and-strongly correlating (about some range of 0.65-0.72). At such high levels of correlations, one will need to find some evidence of discriminant validity to make sure that constructs are not empirically redundant.

Thank you for this comment. We have revised the manuscript to provide a more complete and transparent evaluation of the measurement model.

Specifically, we now report the standardized factor loadings for each item (Appendix 1), adopting a clear cut-off criterion (λ ≥ .50). All loadings were statistically significant and generally exceeded this threshold, with one item of perceived behavioral control showing a slightly lower loading but retained based on theoretical relevance.

In addition, convergent validity was assessed using Average Variance Extracted (AVE) and Composite Reliability (CR) for each construct. AVE values ranged from .61 to .81, all exceeding the recommended minimum of .50, and CR values were above .70, indicating satisfactory convergent validity and internal consistency.

Finally, given the presence of moderate to strong correlations among theoretically adjacent motivational subscales (approximately r = .65–.72), discriminant validity was explicitly examined using the Fornell–Larcker criterion and the Heterotrait–Monotrait ratio (HTMT). Results showed that the square root of the AVE for each construct exceeded its correlations with all other latent variables, and all HTMT values were below the recommended threshold, indicating that the constructs are empirically distinguishable and not redundant.

Taken together, the standardized factor loadings, CR and AVE values, and the evidence from discriminant validity analyses indicate that the measurement model is psychometrically sound and appropriate for the aims of the study. These revisions can be found on page 7, lines 274–296.

Such under-documented and under-justified motivation scale is shortened motivation scale (REBS). You had fewer items per subscale (3 items per subscale; 2 items amotivation). This is a substantial methodological determination since shortening may alter factor structure and weaken validity. What you must add: Specifically what items were kept (put the wording of the item in an Appendix).

The reason why the items were chosen (e.g. the greatest loadings in earlier research, cultural adaptation limitations, pilot performances). 

Evidence that the abridged version still creates the desired multi-factor structure in your sample (CFA results and indices of reliability/validity of each factor).

 

Response: We thank the reviewers for this important methodological comment. We have revised the manuscript to provide a detailed justification of the item selection process and to document the psychometric adequacy of the abridged version used in the present study. Specifically, we now clarify that items were selected based on theoretical considerations and their conceptual representativeness of each regulatory style, as well as on their performance in terms of factor loadings reported in previous validation studies (in press). This strategy was adopted to retain the most informative items while reducing participant burden.

To enhance transparency, the full wording of all retained items for each motivational regulation has now been provided in an Appendix (Appendix 1), allowing readers to directly evaluate content coverage and construct representation.

Importantly, we now provide empirical evidence that the abridged version preserved the intended multi-factor structure in our sample. Confirmatory factor analysis supported the six-factor structure of motivational regulations, with satisfactory model fit indices, statistically significant standardized factor loadings, and adequate indicators of reliability and validity for each construct (Composite Reliability and Average Variance Extracted). These results indicate that the shortened REBS retained acceptable psychometric properties and is suitable for use within the present structural equation modeling framework.

These revisions are reported in the Instrument section, the Measurement Model subsection of the Results (page 5, lines 192-204), and in Appendix.

The details of SEM estimations are not specific enough to Likert/ordinal data and missingness. Your variables have been calculated based on a Likert-type items and summed scores, but the manuscript fails to explain how the estimation was matched with the data properties. Please indicate (in the Statistical Analysis section): SEM version and software used. Type of estimator (e.g., ML, MLR, WLSMV) and their selection depending on the nature of variables and distribution. The treatment of non-normality (robust standard errors? bootstrapping?). The treatment of missing data (listwise deletion versus FIML versus multiple imputation). And the percentage missing.

Response: We thank the reviewer for this comment. The requested information regarding SEM estimation and data properties is now explicitly reported in the Statistical Analysis section. All analyses were conducted using Mplus version 8, and models were estimated using maximum likelihood estimation. This estimator was considered appropriate because analyses were based on composite scores derived from Likert-type items and the distributional properties of the observed variables were within acceptable ranges. With regard to missing data, the dataset contained no missing values; therefore, no missing data handling or imputation procedures were required (0% missingness). These details are reported in the Statistical Analysis section (page 6, lines 236 -242).

There is an actual threat of common-method bias, which is not properly treated. All the constructs (motivation, intention, PBC, adherence) would be self-reported and measured in a single survey session. This inflates common variance and may bias path estimates. Elaboration Minimum improvement: include a special limitation with a short description of possible inflation. 

Greater improvement (preferable): include a plain diagnostic (e.g., marker variable approach or a factor sensitivity check on a latent method) where available. 

Response: We thank the reviewer for raising this important methodological concern. We agree that the exclusive reliance on self-reported measures collected within a single survey session may introduce a risk of common-method bias, potentially inflating shared variance among constructs and biasing parameter estimates. In response, we have explicitly acknowledged this issue by adding a dedicated limitation to the Discussion, where we briefly describe the potential inflation of associations due to common-method variance and encourage future studies to address this issue using multi-method designs or formal diagnostic approaches. These changes have been implemented in the Limitations section (page 12-13, lines 450–458).

 

There should be modest explained variance in the main outcome through interpretation. You state that the model describes that the adherence is explained by R 2 = 16%. It is not insignificant, and yet not high. Adjustment that is needed in Discussion: Current adherence as affected by other determinants unmeasured by the model (e.g., food environment/access, cooking resources, time constraints, affordability, campus availability, social environment, cultural eating patterns). Do not use words like strongly predicted adherence when explained variance is small.

Response: Thank you for this important and thoughtful comment. We agree that an explained variance of R² = 16% represents a modest, albeit meaningful, proportion of variance in adherence and should be interpreted accordingly. We have revised the Discussion to provide a more balanced and cautious interpretation of the predictive power of the model. In particular, we now explicitly acknowledge that current adherence is likely influenced by several additional determinants that were not included in the present model, such as food environment and access, cooking resources, time constraints, affordability, campus availability, social environment, and cultural eating patterns. (page10, lines 348 - 362).
Moreover, we have revised the wording throughout the Discussion to avoid overstatements (e.g., expressions implying strong prediction of adherence) and to more accurately reflect the modest explained variance of the model. These changes have been implemented in the Discussion section (throughout).

Minor problems (presentation/clarity — correct during revising) Issue in format/statistical notation in Tables. Table 2 indicates a correlation between adherence and intention as r = 0.30 though it is not significant as indicated even when smaller correlations are significant. Action: re-visit p-values and make sure that there is a consistent use of significance symbols and table footnotes on all cells. 

Response: Thank you for this comment, also raised by reviewer 1. We therefore conducted a systematic check of all p-values and statistical notations across all tables. As part of this revision, we corrected the missing significance marker for the correlation between intention and adherence in Table 2 and ensured that all correlations are now reported with consistent significance symbols and table footnotes. The revised tables are included in the updated manuscript (page 7, table 2).

 

Terminology/spelling consistency. It contains a number of typos and irregular spelling/capitalization (e.g., Mediterrenean vs. Mediterranean).

Response: Thank you for pointing this out. We have carefully proofread the entire manuscript to ensure consistency in terminology, spelling, and capitalization.

In particular, we corrected all instances of inconsistent spelling (e.g., “Mediterrenean” to “Mediterranean”, page 7, table 2) and standardized terminology throughout the text. We also revised capitalization and formatting to ensure consistency across sections, tables, and figures. The revised version of the manuscript reflects these corrections.

QueMD scoring should be more justifiable and interpretable. According to your description, there are items that have 0 points with or without consumption. Such may diminish sensitivity and make interpretation more difficult in young adult sample. Action: explain the rationale behind the use of this scoring rule, and ensure that it aligns with the desired interpretation of the dietary adherence measure. 

Covariates and robustness. Your age report indicates that there is a correlation with adherence (r = 0.11). Although sex differences may not be significant when comparing both variables at once, it would be more credible to include the information that age and sex were covariated in SEM (or to provide the reason why not).

Response: Thank you for this important comment. We agree that, given the observed correlation between age and adherence (r = .11), it is important to clarify the role of age and sex in the structural model. As suggested, we have now explicitly reported that both age and sex were included as covariates in the SEM analyses. This was reported in Figure 1. However, we acknowledge that this information was not explicitly mentioned in the text, and we apologize for this oversight. This information has been added to the Analyses section (page 4, line 180).

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you for the opportunity to read and evaluate your manuscript. The topic you address – motivational and socio-cognitive processes underlying adherence to the Mediterranean diet in young adults – is highly relevant for Healthcareand broadly aligned with its scope on prevention, health promotion and lifestyle behaviours. The use of an integrated Self-Determination Theory (SDT) and Theory of Planned Behavior (TPB) framework, together with a validated measure of Mediterranean diet adherence, represents a solid conceptual starting point.

Overall, I found the study promising and potentially suitable for publication after a substantial revision. In its current form, the manuscript would benefit from clearer positioning within the existing literature, a more transparent description of the methods (particularly the measurement model), and a stronger articulation of the implications for healthcare practice and policy. Below I detail my main comments and suggestions, which I hope you will find constructive.

1. General contribution and framing

Your core message – that both socio-cognitive variables (perceived behavioural control, intention) and motivational regulations (especially intrinsic motivation) are associated with adherence to the Mediterranean diet in emerging adults – is clear and interesting. The focus on young adults is timely, because this is a critical developmental window in which eating patterns consolidate and can have long-term consequences for physical and mental health.

At the same time, the theoretical model you adopt is not completely new. In recent years, several studies have integrated SDT and TPB to predict diet-related behaviours and, more specifically, adherence to the Mediterranean diet in Italian samples. Your manuscript already cites some of this work, but the introduction and discussion would benefit from a more explicit and nuanced comparison with these closely related studies. Rather than presenting your model as filling a completely unexplored gap, I would encourage you to frame it as an important extension and refinement in a specific population (emerging adults, university students) with some distinctive methodological choices (e.g., use of QueMD, full set of SDT regulations, SEM approach).

A more precise statement of what is genuinely novel here (e.g., the specific age group, the pattern of direct versus indirect effects of certain regulations, the use of a detailed Mediterranean diet index in a fully integrated SDT–TPB model) will make your contribution clearer and more compelling.

2. Introduction and positioning within the literature

The introduction is generally well written and provides a good overview of the health impact of obesity and the benefits of the Mediterranean diet. The sections on SDT and TPB are informative. However, in view of Healthcare’s readership and the density of existing work in this area, a few improvements would strengthen this section:

• When you present SDT and TPB, it would help to move quickly from a general description to a more focused rationale for why these theories are particularly suitable for understanding Mediterranean diet adherence in emerging adults. For example, you might emphasise developmental aspects (identity exploration, increasing autonomy, changes in social norms and living arrangements) that make motivational regulations and perceived control especially relevant in this age range.

• Regarding the literature on SDT–TPB integration and Mediterranean diet, the current treatment feels somewhat brief. I would encourage you to summarise more systematically what previous integrated models have already shown (e.g., the central role of PBC and autonomous motivation, typical ranges of explained variance for intention and behaviour, contexts/settings) and then situate your study as a logical next step in this trajectory.

• The claim that “no studies have integrated these theories to investigate the whole pattern of Mediterranean diet among young people” should be softened and better qualified. Given the existence of work on Mediterranean diet using integrated SDT–TPB models in adult or mixed-age samples, and TPB-based work on healthy eating in youngsters, you might reformulate this along the lines of: “To our knowledge, no previous study has tested an integrated SDT–TPB model of Mediterranean diet adherence specifically in a sample focused on emerging adults, using a detailed Mediterranean diet index.” This more cautious phrasing reduces the risk of overstating novelty.

Finally, because Healthcare is a healthcare-oriented journal rather than a purely psychological one, you might consider adding a short paragraph that explicitly links your conceptual model to the kinds of interventions or services where it might be applied (university health services, primary care, public health campaigns targeting young adults, etc.). This will help readers see more clearly why this psychological model matters for healthcare practice.

3. Conceptual clarity of the theoretical model

A central point that needs clarification concerns the operationalisation of TPB in your study. In the introduction you describe the full TPB (attitudes, subjective norms, perceived behavioural control, intention), but in the methods and model you include only PBC and intention as TPB-derived constructs.

This is not necessarily a problem in itself – many studies use partial or adapted TPB models – but it needs to be stated explicitly and coherently. At present, the reader may have the impression that the full TPB has been implemented, when in fact it is a reduced version focusing on what you consider the most relevant socio-cognitive variables.

I would therefore recommend:

• Clearly stating, in the introduction and methods, that your model incorporates selected key constructs from TPB(PBC and intention), and briefly justifying this choice (e.g., empirical evidence for the centrality of PBC and intention, constraints on questionnaire length, focus on controllability and volition rather than attitudes/norms, etc.).

• Adjusting the wording wherever you currently refer to “TPB constructs” in a generic way, so that it is clear to the reader which parts of the TPB framework were actually measured.

This will improve theoretical transparency and prevent misunderstandings.

4. Methods: sample, instruments and measurement model

The methods section is a crucial area where more detail is needed for the manuscript to meet Healthcare’s standards of transparency and reproducibility.

Sample and recruitment

You report that 365 young adults (with a balanced gender distribution and a mean age around 21.7 years) participated, mostly university students. This is an adequate sample for SEM, but more information is needed:

• Please indicate how participants were recruited (e.g., via university courses, email lists, social media), over which time period, and in which geographical area.

• It would be helpful to state any inclusion/exclusion criteria (e.g., age range, enrolment status, presence of chronic diseases that might affect diet) and whether any cases were excluded from the analyses (and why).

• Since your sample is essentially a convenience sample of university students, it would be important to acknowledge this explicitly in the methods (not only in the limitations) and to consider whether age and gender were examined as covariates or potential moderators in your model. If you decided not to include them, a brief justification would be welcome.

Instruments

The choice of instruments is appropriate, but the description of their adaptation and psychometric properties in your sample is somewhat limited.

• For the regulation of eating behaviours scale (REBS), you mention using a shortened version (three items per regulation, two for amotivation). Readers need to understand how this short form was created. Were the items selected based on previous validation work, theoretical considerations, or new factor analyses? Providing at least one example item per subscale would also make the constructs more concrete.

• You report internal consistency (Cronbach’s alpha) for the REBS subscales, which is useful, but you state that you tested construct validity using confirmatory factor analysis (CFA) and then do not provide any results for the CFA. Given that your structural equation model relies on these latent constructs, it is essential to report at least the main fit indices of the measurement model and a summary of factor loadings.

• For the TPB-derived measures (PBC and intention), please specify how many items were used per construct, the response scale, and example items. Internal consistency estimates should be reported as well.

• The QueMD is an appropriate choice for measuring adherence to the Mediterranean diet, and your brief description of the scoring system is helpful. However, I would encourage you to add:

  • the mean and standard deviation of the total score,

  • the observed range in your sample,

  • and, if relevant, the proportion of participants in categories of low, intermediate and high adherence according to established cut-offs.
    This will allow readers to judge the level of adherence in your sample and to compare it with other studies using QueMD or similar indices.

Measurement model and CFA

Since your main analytic approach is SEM, more complete reporting of the measurement model is necessary:

• Please report the results of the CFA (either in text or in a table): model fit indices, factor loadings, and correlations between latent factors.

• Some of the motivational regulations (e.g., integrated and identified regulation) are theoretically close and likely empirically correlated. If you observed very high correlations among some of these factors, it would be important to discuss whether a higher-order factor solution (autonomous vs controlled motivation) or composite scores might be more parsimonious, even if you ultimately decide to keep the more fine-grained structure.

• It would also be useful to mention how you handled missing data, which estimator you used (e.g., ML, robust ML), and whether you checked distributional assumptions for the items/composites.

Providing this information will greatly increase the credibility of your SEM results.

5. Results and interpretation

The main findings of the structural model are clearly summarised: perceived behavioural control and intention both predict adherence to the Mediterranean diet; intrinsic motivation also shows a direct association with behaviour; and integrated and external regulation influence intention indirectly through PBC.

I have a few suggestions to improve the presentation and interpretation:

• Consider including a table of descriptive statistics (means, standard deviations, ranges, Cronbach’s alpha) for all the key variables (regulations, PBC, intention, QueMD score). This will make it easier for readers to interpret the magnitude of effects and to replicate your work.

• When discussing the model, it may help to explicitly state that your model explains about 40% of the variance in intention but only about 16% of the variance in behaviour. This latter value is typical for complex behaviours but is relatively modest. Reflecting briefly on what other factors (e.g., environmental constraints, food availability, habits, time and financial resources) might account for the remaining variance would enrich the discussion and reinforce your argument for multifaceted interventions.

• Some of the non-significant paths are theoretically interesting, especially the lack of effects for identified and introjected regulation. It would be valuable to discuss these null findings in more detail, for example considering whether in emerging adults health-related diet patterns are more strongly driven by intrinsic enjoyment and fully integrated values than by partially internalised or guilt-driven motives.

• For the indirect effects, it would be ideal to report bootstrapped confidence intervals as well as point estimates, and to clearly indicate in the text which mediation paths are statistically significant.

Finally, I would suggest being cautious with causal language. Given the cross-sectional design, it is safer to describe your findings as consistent with a theoretical causal model rather than as demonstrating causal effects.

6. Discussion, practical implications and conclusions

The discussion rightly emphasises the central role of perceived behavioural control and intrinsic motivation in promoting adherence to the Mediterranean diet. The general alignment with previous SDT–TPB integration work is well noted.

However, for Healthcare readers, the discussion could be strengthened in two main ways:

1. Deeper engagement with the literature and quantitative comparisons
   It would be helpful to compare your levels of explained variance and pattern of significant pathways more explicitly with previous integrated models (both on general healthy eating and on Mediterranean diet), highlighting where your results converge and where they diverge. This will make your specific contribution more visible.

2. More concrete practical implications
   At present, the implications are quite general. I encourage you to translate your findings into more tangible recommendations for interventions in healthcare or educational settings, for example:

   • how autonomy-supportive communication and counselling might be implemented in university health services or primary care;

   • how programmes could be designed to enhance PBC by developing practical skills (meal planning, cooking, navigating canteen/restaurant choices);

   • how screening of motivational profiles could help tailor interventions for different subgroups of young adults.

The conclusions are concise and in line with the results, but could end with a slightly sharper message about what practitioners, educators and policymakers should take away from your work.

7. Limitations

You acknowledge some important limitations (cross-sectional design, self-report measures, convenience sample), which is commendable. You might consider expanding slightly on:

• the restricted generalisability of findings beyond university students and the need for replication in more diverse groups of young adults;

• the lack of objective indicators of diet or health status (e.g., BMI, biomarkers);

• the potential bias associated with self-reported intake and common method variance.

This will show that you are fully aware of the boundaries of your conclusions.

8. Language and presentation

The manuscript is generally well written, and the English is more than adequate to understand your work. Nevertheless, a light language edit would be beneficial to:

• correct a few minor typographical errors (e.g., “Mediterrenean” → “Mediterranean”) and occasional awkward phrasings;

• ensure consistency between British and American spelling (e.g., “behaviour” vs “behavior”);

• streamline some longer sentences in the Introduction and Discussion to improve readability.

I also recommend:

• defining all abbreviations at first use (e.g., World Health Organization (WHO), Theory of Planned Behavior (TPB), Self-Determination Theory (SDT), perceived behavioural control (PBC), Mediterranean diet (MD)), and

• ensuring that the list of abbreviations at the end is complete and consistent with the text.

Finally, please double-check your reference list for minor inaccuracies (including DOIs) and consider adding any key recent works that are currently missing but highly relevant to your model.

In summary, this is a relevant and promising manuscript with a solid theoretical basis and appropriate analytic approach. By clarifying the theoretical framing (especially the partial use of TPB), providing more detailed information on the measurement model and sample, strengthening the comparison with closely related studies and expanding the practical implications for healthcare, you can substantially increase the impact and clarity of the work. I hope these comments are helpful as you revise the manuscript.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

Overall, the quality of English is good and the manuscript is perfectly understandable throughout. However, a light but careful language revision would improve clarity, flow and consistency, and would help the paper meet the stylistic standards of Healthcare.

More specifically:

• There are a few minor typographical errors (e.g., “Mediterrenean” instead of “Mediterranean” and occasional small slips in plural/singular or articles) that should be corrected.

• The manuscript alternates between British and American spelling (e.g., “behaviour” vs. “behavior”, “organisation” vs. “organization”). It would be preferable to choose one variant and use it consistently across the text.

• Some sentences in the Introduction and Discussion are rather long and dense, with multiple clauses and several concepts embedded in a single period. These could be broken into shorter sentences or slightly rephrased to improve readability and make the key messages more immediately accessible to readers.

• A few expressions could be made more precise or natural (for example, replacing generic formulations such as “in the light of” or “in the frame of” with more standard academic phrasing). This is not a major issue, but small adjustments would further polish the style.

• Please make sure that all abbreviations (e.g., WHO, TPB, SDT, PBC, MD) are spelled out in full at their first occurrence in the main text and used consistently thereafter. This will also help international readers outside the psychological field.

• Punctuation is generally correct, but a careful final check of commas and conjunctions in complex sentences would help avoid occasional ambiguities and enhance the logical flow.

In summary, the English language does not require a major overhaul, but I would recommend a light professional copy-edit or a thorough internal language revision to correct minor errors, harmonise spelling, simplify a few complex sentences, and ensure full consistency of terminology and abbreviations. This would significantly enhance the overall clarity and presentation of your work.

Author Response

Thank you for the opportunity to read and evaluate your manuscript. The topic you address – motivational and socio-cognitive processes underlying adherence to the Mediterranean diet in young adults – is highly relevant for Healthcareand broadly aligned with its scope on prevention, health promotion and lifestyle behaviours. The use of an integrated Self-Determination Theory (SDT) and Theory of Planned Behavior (TPB) framework, together with a validated measure of Mediterranean diet adherence, represents a solid conceptual starting point.

Overall, I found the study promising and potentially suitable for publication after a substantial revision. In its current form, the manuscript would benefit from clearer positioning within the existing literature, a more transparent description of the methods (particularly the measurement model), and a stronger articulation of the implications for healthcare practice and policy. Below I detail my main comments and suggestions, which I hope you will find constructive.

1. General contribution and framing

Your core message – that both socio-cognitive variables (perceived behavioural control, intention) and motivational regulations (especially intrinsic motivation) are associated with adherence to the Mediterranean diet in emerging adults – is clear and interesting. The focus on young adults is timely, because this is a critical developmental window in which eating patterns consolidate and can have long-term consequences for physical and mental health.

At the same time, the theoretical model you adopt is not completely new. In recent years, several studies have integrated SDT and TPB to predict diet-related behaviours and, more specifically, adherence to the Mediterranean diet in Italian samples. Your manuscript already cites some of this work, but the introduction and discussion would benefit from a more explicit and nuanced comparison with these closely related studies. Rather than presenting your model as filling a completely unexplored gap, I would encourage you to frame it as an important extension and refinement in a specific population (emerging adults, university students) with some distinctive methodological choices (e.g., use of QueMD, full set of SDT regulations, SEM approach).

A more precise statement of what is genuinely novel here (e.g., the specific age group, the pattern of direct versus indirect effects of certain regulations, the use of a detailed Mediterranean diet index in a fully integrated SDT–TPB model) will make your contribution clearer and more compelling.

 

Response: We thank the reviewer for this thoughtful and constructive comment. We agree that, although the integration of Self-Determination Theory and the Theory of Planned Behavior has been applied in previous studies to diet-related behaviors and Mediterranean diet adherence, our model should be framed more clearly as an extension and refinement of existing work rather than as addressing a completely unexplored gap.

 

In response, we have revised both the Introduction and the Discussion to provide a more explicit and nuanced comparison with prior SDT–TPB studies, particularly those conducted in Italian samples. We now emphasize the specific elements that characterize the contribution of the present study, including the focus on emerging adults as a distinct developmental group, the use of a detailed Mediterranean diet index (QueMD) to capture adherence to the overall dietary pattern, the inclusion of the full set of motivational regulations, and the application of a comprehensive SEM framework to test both direct and indirect pathways.

 

We have also clarified the aspects in which our findings extend previous evidence, such as the pattern of direct and indirect effects of specific motivational regulations and the joint role of socio-cognitive and motivational processes in this age group. These revisions aim to more precisely articulate the novelty and relevance of the study while situating it within the existing literature. The relevant changes have been implemented in the Introduction and Discussion sections (page 3, lines 106-140).

 

2. Introduction and positioning within the literature

The introduction is generally well written and provides a good overview of the health impact of obesity and the benefits of the Mediterranean diet. The sections on SDT and TPB are informative. However, in view of Healthcare’s readership and the density of existing work in this area, a few improvements would strengthen this section:

When you present SDT and TPB, it would help to move quickly from a general description to a more focused rationale for why these theories are particularly suitable for understanding Mediterranean diet adherence in emerging adults. For example, you might emphasise developmental aspects (identity exploration, increasing autonomy, changes in social norms and living arrangements) that make motivational regulations and perceived control especially relevant in this age range.

Response: We appreciate this helpful suggestion. We have revised the Introduction by adding a paragraph that more explicitly links Self-Determination Theory and the Theory of Planned Behavior to the developmental characteristics of emerging adulthood, highlighting why these frameworks are particularly suitable for understanding Mediterranean diet adherence in this age group.

Specifically, we now emphasize key aspects such as identity exploration, increasing autonomy, changes in social norms, and transitions in living arrangements, which make motivational regulation and perceived behavioral control especially relevant for dietary behavior among young adults. This paragraph can be found on page 3, lines 106–115.

• Regarding the literature on SDT–TPB integration and Mediterranean diet, the current treatment feels somewhat brief. I would encourage you to summarise more systematically what previous integrated models have already shown (e.g., the central role of PBC and autonomous motivation, typical ranges of explained variance for intention and behaviour, contexts/settings) and then situate your study as a logical next step in this trajectory.

Response: We appreciate the reviewer’s suggestion. We have therefore revised the Introduction to provide a more structured overview of previous integrated models, highlighting their main findings, including the central role of perceived behavioral control and autonomous motivation, the typical ranges of explained variance for intention and behavior, and the contexts in which these models have been applied. We then position the present study as a logical next step in this line of research, extending previous evidence to adherence to the Mediterranean diet among young adults. These revisions can be found on page 3, lines 116-127.

The claim that “no studies have integrated these theories to investigate the whole pattern of Mediterranean diet among young people” should be softened and better qualified. Given the existence of work on Mediterranean diet using integrated SDT–TPB models in adult or mixed-age samples, and TPB-based work on healthy eating in youngsters, you might reformulate this along the lines of: “To our knowledge, no previous study has tested an integrated SDT–TPB model of Mediterranean diet adherence specifically in a sample focused on emerging adults, using a detailed Mediterranean diet index.” This more cautious phrasing reduces the risk of overstating novelty.

Response: We appreciate the reviewer’s careful observation regarding the formulation of our claim. We have reformulated the statement in line with the reviewer’s suggestion, clarifying that, to our knowledge, no previous study has tested an integrated SDT–TPB model of Mediterranean diet adherence specifically in a sample focused on emerging adults and using a detailed Mediterranean diet index. This revision can be found on pages 3-4, lines 134–136.

 

Finally, because Healthcare is a healthcare-oriented journal rather than a purely psychological one, you might consider adding a short paragraph that explicitly links your conceptual model to the kinds of interventions or services where it might be applied (university health services, primary care, public health campaigns targeting young adults, etc.). This will help readers see more clearly why this psychological model matters for healthcare practice.

Response: We thank the reviewer for this valuable suggestion. We have added a paragraph that explicitly links the proposed conceptual model to potential applications in healthcare settings and interventions aimed at promoting adherence to the Mediterranean diet among young adults, such as university health services, primary care, and public health campaigns. This revision can be found on page 4, lines 140–149.

3. Conceptual clarity of the theoretical model

A central point that needs clarification concerns the operationalisation of TPB in your study. In the introduction you describe the full TPB (attitudes, subjective norms, perceived behavioural control, intention), but in the methods and model you include only PBC and intention as TPB-derived constructs.

This is not necessarily a problem in itself – many studies use partial or adapted TPB models – but it needs to be stated explicitly and coherently. At present, the reader may have the impression that the full TPB has been implemented, when in fact it is a reduced version focusing on what you consider the most relevant socio-cognitive variables.

I would therefore recommend:

• Clearly stating, in the introduction and methods, that your model incorporates selected key constructs from TPB (PBC and intention), and briefly justifying this choice (e.g., empirical evidence for the centrality of PBC and intention, constraints on questionnaire length, focus on controllability and volition rather than attitudes/norms, etc.).

Response: We thank the reviewer for this valuable comment. We have specified, in both the Introduction and the Methods sections, that the present model focuses on perceived behavioral control and intention as the TPB-derived constructs. We have also provided a brief justification for this choice, drawing on empirical evidence supporting the central role of these variables, as well as on the focus of the study on controllability and volitional processes.

• Adjusting the wording wherever you currently refer to “TPB constructs” in a generic way, so that it is clear to the reader which parts of the TPB framework were actually measured.

This will improve theoretical transparency and prevent misunderstandings.

Response: We have clarified, throughout the manuscript, that the present study selectively incorporates perceived behavioral control and behavioral intention as the Theory of Planned Behavior–derived constructs. This clarification has been explicitly introduced in the Introduction (page 3, lines 128–132) and in the Methods section (page 5, line 195). We have also briefly justified this choice by referring to empirical evidence supporting the central role of these variables in predicting healthy eating behaviors, as well as to the focus of the study on controllability and volitional processes. 

4. Methods: sample, instruments and measurement model

The methods section is a crucial area where more detail is needed for the manuscript to meet Healthcare’s standards of transparency and reproducibility.

Sample and recruitment

You report that 365 young adults (with a balanced gender distribution and a mean age around 21.7 years) participated, mostly university students. This is an adequate sample for SEM, but more information is needed:

• Please indicate how participants were recruited (e.g., via university courses, email lists, social media), over which time period, and in which geographical area.
• It would be helpful to state any inclusion/exclusion criteria (e.g., age range, enrolment status, presence of chronic diseases that might affect diet) and whether any cases were excluded from the analyses (and why).
• Since your sample is essentially a convenience sample of university students, it would be important to acknowledge this explicitly in the methods (not only in the limitations) and to consider whether age and gender were examined as covariates or potential moderators in your model. If you decided not to include them, a brief justification would be welcome.

Response: We thank the reviewer for this important comment. We have revised the Methods section to provide a more detailed description of the recruitment procedure, sampling strategy, and participant characteristics.

In particular, we now explicitly state that the sample represents a convenience sample of young adults, and we report the recruitment channels, data collection period, and geographical area. We also clarify the inclusion criteria and explicitly acknowledge that no additional exclusion criteria were applied, as information on medical conditions potentially affecting dietary behavior was not collected. These changes have been implemented in the Participants and Procedure sections (page 3, line 153 - 159).

 

 

Instruments

The choice of instruments is appropriate, but the description of their adaptation and psychometric properties in your sample is somewhat limited.

• For the regulation of eating behaviours scale (REBS), you mention using a shortened version (three items per regulation, two for amotivation). Readers need to understand how this short form was created. Were the items selected based on previous validation work, theoretical considerations, or new factor analyses? Providing at least one example item per subscale would also make the constructs more concrete.

Response: Thank you for this important comment. We agree that a clearer description of the shortened version of the Regulation of Eating Behaviours Scale (REBS) is necessary to ensure transparency and replicability. We have therefore revised the Instrument section to explicitly describe the rationale and procedure used to create the short form. In particular, we now clarify that the items were selected based on theoretical considerations and their conceptual representativeness of each regulatory style, following previous validation work on the REBS. Moreover, we now provide one example item for each subscale to make the constructs more concrete and facilitate readers’ understanding. These revisions have been implemented in the Instrument section (page 3-4, lines 195-206). We have also added a table in the Appendix reporting the full wording of all items retained in the confirmatory factor analysis, together with their standardized factor loadings.

• You report internal consistency (Cronbach’s alpha) for the REBS subscales, which is useful, but you state that you tested construct validity using confirmatory factor analysis (CFA) and then do not provide any results for the CFA. Given that your structural equation model relies on these latent constructs, it is essential to report at least the main fit indices of the measurement model and a summary of factor loadings.

Response: We appreciate this helpful comment. We agree that, given the central role of the latent constructs in the structural equation model, it is essential to report the main results of the measurement model. This issue has been also raised by both reviewers 1 and 2. We have therefore revised the manuscript to include a dedicated paragraph reporting the main CFA results. In particular, we now provide: The overall fit indices of the measurement model (χ², CFI, TLI, RMSEA, SRMR); A summary of the standardized factor loadings for all items, together with the adopted cut-off criterion (λ ≥ .50). We are confident that these additions allow readers to evaluate the adequacy of the measurement model and the quality of the latent constructs used in the SEM. The CFA results are now reported in the Results section (page 7, line 275 - 297).

 

• For the TPB-derived measures (PBC and intention), please specify how many items were used per construct, the response scale, and example items. Internal consistency estimates should be reported as well.
• Response: We appreciate the reviewer’s careful attention to the description of the TPB-derived measures. We would like to clarify that the number of items, response scale, and internal consistency estimates were already reported in the original manuscript. However, we agree that this information could benefit from greater clarity and visibility. Finally, we have now provided example items for each scale in order to improve clarity. These revisions have been implemented in the Instrument section (page 4/5 lines 190 – 206). Reviewer may also found in the Appendix the full wording of all items, together with their standardized factor loadings.

 

• The QueMD is an appropriate choice for measuring adherence to the Mediterranean diet, and your brief description of the scoring system is helpful. However, I would encourage you to add:
• the mean and standard deviation of the total score.
• the observed range in your sample.
• and, if relevant, the proportion of participants in categories of low, intermediate and high adherence according to established cut-offs.
• This will allow readers to judge the level of adherence in your sample and to compare it with other studies using QueMD or similar indices.

Response: We thank the reviewer for this helpful suggestion. We have expanded Table 2 to report the mean and standard deviation of the total QueMD score, as well as the observed range of scores in the sample (pp. 8–9). These descriptive statistics have also been added for the other study variables. With regard to adherence categories, we note that the QueMD does not provide universally established cut-off values for classifying low, intermediate, and high adherence. For this reason, we limited our reporting to continuous descriptive statistics, in line with previous applications of this index. In addition, the manuscript includes a table reporting the mean and standard deviation for each individual daily consumption item assessed by the QueMD, providing a more detailed description of dietary patterns in the sample.

 

Measurement model and CFA

Since your main analytic approach is SEM, more complete reporting of the measurement model is necessary:

• Please report the results of the CFA (either in text or in a table): model fit indices, factor loadings, and correlations between latent factors.

Response: Specifically, we have added a dedicated Measurement Model section reporting the results of the confirmatory factor analysis (CFA). This section now includes:

(a) the overall model fit indices (χ², CFI, TLI, RMSEA, and SRMR);

(b) the standardized factor loadings, together with the correlations among the latent factors; and

(c) an explicit evaluation of convergent validity (Composite Reliability and Average Variance Extracted) and discriminant validity (Fornell–Larcker criterion and HTMT).

For transparency and ease of consultation, the full set of item wording and standardized factor loadings is reported in Appendix 1, while the main CFA results are presented in the Measurement Model section of the Results (page 7, lines 274–296).

 

• Some of the motivational regulations (e.g., integrated and identified regulation) are theoretically close and likely empirically correlated. If you observed very high correlations among some of these factors, it would be important to discuss whether a higher-order factor solution (autonomous vs controlled motivation) or composite scores might be more parsimonious, even if you ultimately decide to keep the more fine-grained structure.

Response: We appreciate the reviewer’s thoughtful observation regarding the conceptual proximity and potential empirical overlap among some motivational regulations. As expected within the Self-Determination Theory continuum, we observed strong correlations between theoretically adjacent regulations, particularly between integrated and identified regulation. We have now acknowledged this pattern in the Discussion and explicitly considered the potential value of more parsimonious modeling approaches, such as higher-order factors representing autonomous versus controlled motivation or the use of composite scores. At the same time, we explain our decision to retain the more fine-grained structure in the present model in order to preserve conceptual specificity and to explore potentially distinct pathways associated with different forms of regulation. This discussion has been added to the Discussion section (page 11, lines 393 - 401).

 

• It would also be useful to mention how you handled missing data, which estimator you used (e.g., ML, robust ML), and whether you checked distributional assumptions for the items/composites.

Providing this information will greatly increase the credibility of your SEM results.

 

Response: We appreciate the reviewer’s suggestion to further clarify the analytical procedure. We have now revised the Analyses section to explicitly report how missing data were handled, the estimator used, and the assessment of distributional assumptions.

Specifically, we clarify that the dataset contained no missing values, and therefore no imputation or missing data handling procedures were required. We also report the estimator used for the SEM analyses and indicate that distributional assumptions were examined prior to model estimation.

These additions have been implemented in the Analyses section (page 5, lines 239-242).

 

5. Results and interpretation

The main findings of the structural model are clearly summarised: perceived behavioural control and intention both predict adherence to the Mediterranean diet; intrinsic motivation also shows a direct association with behaviour; and integrated and external regulation influence intention indirectly through PBC.

I have a few suggestions to improve the presentation and interpretation:

• Consider including a table of descriptive statistics (means, standard deviations, ranges, Cronbach’s alpha) for all the key variables (regulations, PBC, intention, QueMD score). This will make it easier for readers to interpret the magnitude of effects and to replicate your work. Response: We thank the reviewer for this helpful suggestion. To improve transparency and facilitate the interpretation and replication of the findings, we have expanded Table 2 to report descriptive statistics for all key variables included in the study. Specifically, the table now presents means, standard deviations, observed ranges, and intercorrelations for the motivational regulations, perceived behavioral control, intention, and the QueMD score. Cronbach’s alpha coefficients for the study measures are reported in the text in the Measures section (page 5, lines 193–206). Together, this information provides a comprehensive descriptive overview of the key variables. These revisions are reported in Table 2 (Results section, page 9, lines 328–329).

 

When discussing the model, it may help to explicitly state that your model explains about 40% of the variance in intention but only about 16% of the variance in behaviour. This latter value is typical for complex behaviours but is relatively modest. Reflecting briefly on what other factors (e.g., environmental constraints, food availability, habits, time and financial resources) might account for the remaining variance would enrich the discussion and reinforce your argument for multifaceted interventions.

Response: We thank the reviewer for this helpful comment. We now explicitly report that the model explains approximately 40% of the variance in intention and about 16% of the variance in adherence behavior, and we clarify that this level of explained variance is typical for complex health-related behaviors. We also briefly discuss other unmeasured determinants (e.g., environmental constraints, food availability, habits, time and financial resources) that may account for the remaining variance and highlight the need for multifaceted interventions. These clarifications can be found in the Discussion section (page 10, lines 348–362)

• Some of the non-significant paths are theoretically interesting, especially the lack of effects for identified and introjected regulation. It would be valuable to discuss these null findings in more detail, for example considering whether in emerging adults health-related diet patterns are more strongly driven by intrinsic enjoyment and fully integrated values than by partially internalised or guilt-driven motives.

Response: We thank the reviewer for this thoughtful comment. We agree that the non-significant paths, particularly for identified and introjected regulation, are theoretically informative. We have therefore expanded the Discussion to more explicitly address these null findings. In the revised text, we suggest that, in emerging adulthood, adherence to a complex dietary pattern such as the Mediterranean diet may be more strongly supported by intrinsic enjoyment and fully integrated values, rather than by partially internalized or guilt-driven motives. We also discuss how identified and introjected regulation may be more relevant for initiating specific or short-term dietary changes, rather than for sustaining a broader and more demanding dietary pattern. These additions can be found in the Discussion section (page 11, lines 405–414).

• For the indirect effects, it would be ideal to report bootstrapped confidence intervals as well as point estimates, and to clearly indicate in the text which mediation paths are statistically significant.

Response: We thank the reviewer for this suggestion. This request has already been addressed in the revised manuscript. For all indirect effects, we now report standardized point estimates together with 95% bias-corrected bootstrap confidence intervals based on 5,000 resamples. In addition, the text and Table 3 clearly indicate which mediation paths are statistically significant, based on whether the confidence intervals exclude zero. This information is reported in the Results section and summarized in Table 3.

 

Finally, I would suggest being cautious with causal language. Given the cross-sectional design, it is safer to describe your findings as consistent with a theoretical causal model rather than as demonstrating causal effects.

 

Response: We thank the reviewer for this important remark. We agree that, given the cross-sectional design of the study, causal language should be avoided. We have therefore carefully revised the manuscript to ensure a more cautious interpretation of the findings, consistently framing the results as being consistent with the proposed theoretical model rather than as evidence of causal effects. In particular, we replaced causal expressions (e.g., “predicts”, “affects”) with more appropriate associative language where necessary, especially in the Results and Discussion sections. These revisions have been implemented throughout the manuscript.

 

6. Discussion, practical implications and conclusions

The discussion rightly emphasises the central role of perceived behavioural control and intrinsic motivation in promoting adherence to the Mediterranean diet. The general alignment with previous SDT–TPB integration work is well noted.

However, for Healthcare readers, the discussion could be strengthened in two main ways:

1. Deeper engagement with the literature and quantitative comparisons
   It would be helpful to compare your levels of explained variance and pattern of significant pathways more explicitly with previous integrated models (both on general healthy eating and on Mediterranean diet), highlighting where your results converge and where they diverge. This will make your specific contribution more visible.

Response: We thank the reviewer for this helpful suggestion. We have revised the Discussion to engage more deeply with the existing literature on integrated SDT–TPB models of dietary behavior. In particular, we now provide a more explicit comparison between the levels of explained variance and the pattern of significant and non-significant pathways observed in the present study and those reported in previous research on healthy eating and Mediterranean diet adherence. This comparison highlights both points of convergence (e.g., the central role of perceived behavioral control and autonomous motivation) and points of divergence, thereby clarifying the specific contribution of our findings in an emerging adult sample. These additions can be found in the Discussion section (page 10, lines 352–359).

 

1. More concrete practical implications
   At present, the implications are quite general. I encourage you to translate your findings into more tangible recommendations for interventions in healthcare or educational settings, for example:
• how autonomy-supportive communication and counselling might be implemented in university health services or primary care;
• how programmes could be designed to enhance PBC by developing practical skills (meal planning, cooking, navigating canteen/restaurant choices);
• how screening of motivational profiles could help tailor interventions for different subgroups of young adults.

Response: We thank the reviewer for this insightful comment. We have revised the Discussion to translate the study findings into more concrete and actionable implications for healthcare and educational settings. In the added paragraph, we now explicitly discuss how autonomy-supportive communication and counseling may be implemented within university health services and primary care, how interventions can be designed to enhance perceived behavioral control by developing practical skills (e.g., meal planning, cooking, navigating food choices in canteens or restaurants), and how the assessment of motivational profiles could be used to tailor interventions for different subgroups of young adults. These additions are presented in the Discussion section (page 12, lines 430–441).

 

The conclusions are concise and in line with the results, but could end with a slightly sharper message about what practitioners, educators and policymakers should take away from your work.

7. Limitations

You acknowledge some important limitations (cross-sectional design, self-report measures, convenience sample), which is commendable. You might consider expanding slightly on:

• the restricted generalisability of findings beyond university students and the need for replication in more diverse groups of young adults;
• the lack of objective indicators of diet or health status (e.g., BMI, biomarkers);
• the potential bias associated with self-reported intake and common method variance.

This will show that you are fully aware of the boundaries of your conclusions.

 

Response: We thank the reviewer for this insightful comment. In response, we have expanded the Limitations section to more explicitly acknowledge the boundaries of our findings. In particular, we now discuss the restricted generalisability of the results beyond university students and emphasize the need for replication in more diverse groups of young adults. We also note the absence of objective indicators of diet or health status (e.g., BMI or biomarkers) and elaborate on the potential biases associated with self-reported dietary intake and common method variance due to the single-source, single-time-point assessment. These additions are reported in the Limitations section (page 12, lines 449-458 and 467-470).

8. Language and presentation

The manuscript is generally well written, and the English is more than adequate to understand your work. Nevertheless, a light language edit would be beneficial to:

• correct a few minor typographical errors (e.g., “Mediterrenean” → “Mediterranean”) and occasional awkward phrasings;

 

Response: We thank the reviewer for this comment. We have carefully revised the manuscript to correct minor typographical errors (e.g., “Mediterrenean” → “Mediterranean”) and to improve clarity and fluency where wording was awkward or unclear. These corrections have been implemented throughout the manuscript, including tables and figure captions.

• ensure consistency between British and American spelling (e.g., “behaviour” vs “behavior”); marika

 

Response: We thank the reviewer for pointing this out. To ensure consistency, we have systematically revised the manuscript to adopt American English spelling throughout. In particular, all instances of British spelling (e.g., “behaviour”) have been replaced with American spelling (e.g., “behavior”) across the entire manuscript.

• streamline some longer sentences in the Introduction and Discussion to improve readability.

Response: We thank the reviewer for this suggestion. We have revised the Introduction and Discussion to streamline several longer or more complex sentences in order to improve readability and clarity for the reader. In doing so, we aimed to preserve the original meaning while enhancing the flow and accessibility of the text. These revisions have been implemented throughout the Introduction and Discussion sections

I also recommend:

• defining all abbreviations at first use (e.g., World Health Organization (WHO), Theory of Planned Behavior (TPB), Self-Determination Theory (SDT), perceived behavioural control (PBC), Mediterranean diet (MD)

Response: We thank the reviewer for this helpful suggestion. We have carefully revised the manuscript to ensure that all abbreviations are clearly defined at their first occurrence (e.g., World Health Organization, Theory of Planned Behavior, Self-Determination Theory, perceived behavioral control, Mediterranean diet). This revision has been implemented consistently throughout the manuscript.

 

• ensuring that the list of abbreviations at the end is complete and consistent with the text.

Response: We thank the reviewer for this suggestion. We have carefully reviewed and updated the list of abbreviations to ensure that it is complete and fully consistent with the abbreviations used in the main text. This revision has been implemented in the Abbreviations section at the end of the manuscript.

 

Finally, please double-check your reference list for minor inaccuracies (including DOIs) and consider adding any key recent works that are currently missing but highly relevant to your model.

Response: We thank the reviewer for this helpful suggestion. We have carefully reviewed the entire reference list to correct minor inaccuracies and to verify the accuracy of bibliographic details, including DOIs. In addition, we have updated the references by adding key recent works that are highly relevant to the integrated SDT–TPB framework adopted in the present study. These additions and corrections have been implemented in the revised manuscript (on page 13-14, lines 506-509, page 14, lines 516-519 and lines 526-528; page 15, lines 563-565).

 

In summary, this is a relevant and promising manuscript with a solid theoretical basis and appropriate analytic approach. By clarifying the theoretical framing (especially the partial use of TPB), providing more detailed information on the measurement model and sample, strengthening the comparison with closely related studies and expanding the practical implications for healthcare, you can substantially increase the impact and clarity of the work. I hope these comments are helpful as you revise the manuscript.

Comments on the Quality of English Language

Overall, the quality of English is good and the manuscript is perfectly understandable throughout. However, a light but careful language revision would improve clarity, flow and consistency, and would help the paper meet the stylistic standards of Healthcare.

More specifically:

• There are a few minor typographical errors (e.g., “Mediterrenean” instead of “Mediterranean” and occasional small slips in plural/singular or articles) that should be corrected.

 

Response: Thank you for pointing this out. We have carefully revised the manuscript to correct minor typographical errors (e.g., “Mediterrenean” → “Mediterranean”), as well as small inconsistencies in plural/singular forms and article usage. In addition, we performed a light but careful language revision to improve clarity, flow, and consistency throughout the manuscript, in line with the stylistic standards of Healthcare.

 

• The manuscript alternates between British and American spelling (e.g., “behaviour” vs. “behavior”, “organisation” vs. “organization”). It would be preferable to choose one variant and use it consistently across the text.

Response: Thank you for pointing this out. We have systematically revised the manuscript to ensure consistent use of American English spelling throughout. All instances of British spelling (e.g., “behaviour”, “organisation”) have been corrected to their American English equivalents (e.g., “behavior”, “organization”) across the entire text, including tables and the list of abbreviations.

 

Some sentences in the Introduction and Discussion are rather long and dense, with multiple clauses and several concepts embedded in a single period. These could be broken into shorter sentences or slightly rephrased to improve readability and make the key messages more immediately accessible to readers.

 

Response: Thank you for this suggestion. We have revised the Introduction and Discussion to improve readability by simplifying and restructuring several long or dense sentences. Where appropriate, complex sentences have been broken into shorter ones and slightly rephrased to make the key messages clearer and more immediately accessible to readers. These revisions have been implemented throughout the manuscript.

• A few expressions could be made more precise or natural (for example, replacing generic formulations such as “in the light of” or “in the frame of” with more standard academic phrasing). This is not a major issue, but small adjustments would further polish the style. Marika

Response: Thank you for this suggestion. We have carefully revised the manuscript to improve the precision and naturalness of the language, replacing generic or non-standard expressions (e.g., “in the light of”, “in the frame of”) with more conventional academic phrasing where appropriate. These refinements have been implemented throughout the manuscript to further polish the style.

 

• Please make sure that all abbreviations (e.g., WHO, TPB, SDT, PBC, MD) are spelled out in full at their first occurrence in the main text and used consistently thereafter. This will also help international readers outside the psychological field.

Response: Thank you for this helpful suggestion. We have carefully revised the manuscript to ensure that all abbreviations (e.g., WHO, TPB, SDT, PBC, MD) are spelled out in full at their first occurrence in the main text and used consistently thereafter. This revision has been applied throughout the manuscript to improve clarity for an international readership.

• Punctuation is generally correct, but a careful final check of commas and conjunctions in complex sentences would help avoid occasional ambiguities and enhance the logical flow.

 

Response: Thank you for this suggestion. We have conducted a careful final review of punctuation throughout the manuscript, with particular attention to the use of commas and conjunctions in more complex sentences, in order to improve clarity and logical flow and to avoid potential ambiguities.

In summary, the English language does not require a major overhaul, but I would recommend a light professional copy-edit or a thorough internal language revision to correct minor errors, harmonise spelling, simplify a few complex sentences, and ensure full consistency of terminology and abbreviations. This would significantly enhance the overall clarity and presentation of your work.

 

Response: We thank the reviewer for this overall assessment. In response, we carried out a careful internal language revision of the manuscript, aimed at correcting minor errors, harmonizing spelling, simplifying some complex sentences, and ensuring full consistency in terminology and abbreviations. This light but thorough revision was intended to enhance clarity, coherence, and overall presentation, in line with the stylistic standards of Healthcare

 

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Satisfied with changes provided

Author Response

We thank the reviewer for their positive evaluation and are pleased that the changes made adequately addressed their comments.

Reviewer 2 Report

Comments and Suggestions for Authors

Your reviewer comments are methodologically strong and on target. They focus on the real acceptance-critical weaknesses: incomplete measurement-model reporting, unclear SEM estimation details, risk of common-method bias, and over-interpretation of modest explained variance. These are the right priorities. For the second revision, the authors should now do the following—cleanly, explicitly, and with full transparency.

 

1) Measurement Model (CFA): make it fully judgeable The manuscript must report the CFA in a way that allows readers to assess measurement quality without guessing. At minimum: Provide global fit indices (χ², χ²/df, CFI, TLI, RMSEA with CI if available, SRMR). Provide standardized loadings for every item (a complete table; not “summaries”). State a clear cut-off rule (e.g., λ ≥ .50) and explain how exceptions were treated. Provide factor correlations among latent constructs (so redundancy can be judged). If any items fall below the cut-off, the authors must justify retention (theory + sensitivity check).

2) Convergent validity: report CR/AVE per construct The paper should not rely on Cronbach’s alpha alone or general statements. They must present: Composite Reliability (CR) and Average Variance Extracted (AVE) for each construct. A short statement interpreting whether AVE ≥ .50 and CR ≥ .70 (or their chosen thresholds), consistently applied.

 

3) Discriminant validity: mandatory because correlations are high Given moderate-to-strong correlations between motivational subscales, discriminant validity cannot be “implied.” The authors must: Report HTMT (with a stated threshold, e.g., < .85 or < .90—choose one and justify). Report Fornell–Larcker (√AVE vs inter-construct correlations). If any pair fails, they should address it directly: merging constructs, re-specifying the model, or presenting a defensible theoretical argument with robustness checks.

 

4) Shortened REBS: document the scale like a methods paper Using 2–3 items per subscale is a major methodological decision and must be treated as such. The authors should: List the exact item wording retained for each subscale in an Appendix (not just item numbers). Explain why these items were selected (content representativeness + prior psychometric evidence + any cultural/pilot constraints). Demonstrate that the shortened version still fits a multi-factor structure in this sample (CFA fit + loadings + CR/AVE + discriminant validity). Explicitly discuss the trade-off: reduced burden vs potential validity loss.

 

5) SEM estimation and ordinal/Likert handling: specify, justify, align Right now, the analysis section needs to be estimator-appropriate and transparent. The authors must explicitly state: Software and version used for SEM. Estimator (ML / MLR / WLSMV) and a concise justification tied to the data properties (ordinal vs composite scores; distribution; sample size). How non-normality was handled (robust SEs, bootstrapping, etc.). Missing data: percent missing, pattern, and method (FIML/listwise/multiple imputation). If truly 0%, state it explicitly.

 

6) Common-method bias: treat as a real threat, not a throwaway line A limitation paragraph is the minimum, but it should be written precisely: Explain why same-source, same-time measurement may inflate associations. Recommend concrete design remedies for future studies (time separation, multi-source outcomes, objective indicators where possible). If feasible within the current dataset, a simple diagnostic (marker variable or method-factor sensitivity) would strengthen credibility.

7) Explained variance (R² = 16%): fix the interpretation tone R² = 16% is meaningful but modest. The Discussion must: Avoid phrases implying strong prediction. Clearly acknowledge that adherence is likely influenced by unmeasured determinants (food environment, affordability, cooking resources, time constraints, campus availability, social and cultural context). Present findings as partial explanation, not a comprehensive model.

8) Tables: eliminate statistical notation inconsistencies All correlation tables must have: Correct p-values and consistent significance markers. Clear footnotes defining symbols and thresholds. A final systematic cross-check so no cell is mis-flagged.

9) Terminology/spelling: do a full consistency pass Typos and inconsistent spelling/capitalization reduce perceived rigor. The authors should run a full proofreading pass, including tables/figures.

10) QueMD scoring: clarify rationale and interpretability The scoring rule needs to be defensible and interpretable: Explain the rationale for any items that score 0 regardless of consumption. Clarify what the total score means in this population. If the scoring reduces sensitivity for young adults, the authors should either justify why this is acceptable or consider a sensitivity/alternative scoring check.

11) Covariates/robustness: be explicit and consistent If age correlates with adherence, the model should control for age and sex (or justify not doing so). This must be stated consistently in both: The text (analysis section), and The model figure and/or SEM output summary. Bottom line: The second revision should aim for a “fully auditable” measurement and analysis pipeline: complete CFA reporting, explicit validity evidence, estimator/missingness transparency, careful bias acknowledgement, and conservative interpretation of R². This is what will move the manuscript from “interesting but under-documented” to “publishable and defensible.”

Comments on the Quality of English Language

The English can be mostly understood, yet it needs professional editing due to: consistency of spelling and terms (US vs UK; repeated typos), punctuation and structure of the sentences (there are some long and unclear sentences), regular application of technical terms (names of constructions and abbreviations). Readability and perceived rigor will be significantly enhanced with a careful edit.

Author Response

Your reviewer comments are methodologically strong and on target. They focus on the real acceptance-critical weaknesses: incomplete measurement-model reporting, unclear SEM estimation details, risk of common-method bias, and over-interpretation of modest explained variance. These are the right priorities. For the second revision, the authors should now do the following—cleanly, explicitly, and with full transparency.

Response: Thank you for this careful and constructive assessment. We appreciate the reviewer’s focus on acceptance-critical methodological issues and have addressed all points in the second revision with explicit, transparent, and systematic revisions throughout the manuscript.

 

• Measurement Model (CFA): make it fully judgeable The manuscript must report the CFA in a way that allows readers to assess measurement quality without guessing. At minimum: Provide global fit indices (χ², χ²/df, CFI, TLI, RMSEA with CI if available, SRMR). Provide standardized loadings for every item (a complete table; not “summaries”). State a clear cut-off rule (e.g., λ ≥ .50) and explain how exceptions were treated. Provide factor correlations among latent constructs (so redundancy can be judged). If any items fall below the cut-off, the authors must justify retention (theory + sensitivity check).

Response: Thank you for this comment. The manuscript has been revised to ensure that the measurement model is fully transparent and judgeable. Global fit indices for the CFA (χ², χ²/df, CFI, TLI, RMSEA with CI and SRMR) are reported in the Results section (page 6, lines 271–272). The complete set of standardized factor loadings for all items is reported in Appendix A, while summary information is provided in the main text. We clarify that the previous reference to Table X was inaccurate, and the manuscript has been corrected accordingly. A clear cut-off criterion for factor loadings (λ ≥ .50) is explicitly stated. One item of Perceived Behavioral Control showed a slightly lower loading (λ = .46) but was retained due to its theoretical relevance, as justified in the revised manuscript (page 7, lines 273–276). As a sensitivity check, all analyses were replicated after excluding this item. The results remained substantially unchanged, indicating the robustness of the findings. This information has been added to the manuscript (page 7, lines 275-276). Factor correlations among latent constructs estimated from the CFA measurement model are now reported in Appendix B, allowing assessment of potential construct redundancy. This has also been referred to in the text (page 7, lines 286-287)

 

• Convergent validity: report CR/AVE per construct The paper should not rely on Cronbach’s alpha alone or general statements. They must present: Composite Reliability (CR) and Average Variance Extracted (AVE) for each construct. A short statement interpreting whether AVE ≥ .50 and CR ≥ .70 (or their chosen thresholds), consistently applied.

Response: Thank you for this comment. Convergent validity has been assessed and reported by Composite Reliability (CR) and Average Variance Extracted (AVE) for each construct. CR and AVE values for all constructs are reported in Appendix C.

Following commonly accepted criteria, convergent validity was evaluated using thresholds of CR ≥ .70 and AVE ≥ .50. Overall, the results indicate satisfactory internal consistency reliability and adequate convergent validity across constructs. This information has been added to the manuscript (page 7, lines 284-288).

 

• Discriminant validity: mandatory because correlations are high Given moderate-to-strong correlations between motivational subscales, discriminant validity cannot be “implied.” The authors must: Report HTMT (with a stated threshold, e.g., < .85 or < .90—choose one and justify). Report Fornell–Larcker (√AVE vs inter-construct correlations). If any pair fails, they should address it directly: merging constructs, re-specifying the model, or presenting a defensible theoretical argument with robustness checks.

Response: Thank you for this comment. Discriminant validity is now explicitly documented in the manuscript. Specifically, the Results section reports that the square root of the AVE for each construct exceeds its correlations with all other constructs (Fornell–Larcker criterion) and that all Heterotrait–Monotrait (HTMT) values are below the recommended threshold of .85. The complete Fornell–Larcker matrix and HTMT values are reported in Appendix D (Tables D1 and D2), allowing direct evaluation of discriminant validity despite the moderate-to-strong correlations among motivational subscales. The manuscript has been revised accordingly (page 7, lines 288–295). 

• Shortened REBS: document the scale like a methods paper Using 2–3 items per subscale is a major methodological decision and must be treated as such. The authors should: List the exact item wording retained for each subscale in an Appendix (not just item numbers). Explain why these items were selected (content representativeness + prior psychometric evidence + any cultural/pilot constraints). Demonstrate that the shortened version still fits a multi-factor structure in this sample (CFA fit + loadings + CR/AVE + discriminant validity). Explicitly discuss the trade-off: reduced burden vs potential validity loss.

Response: The use of a shortened version of the REBS was treated as a deliberate measurement decision and is now documented in greater detail. First, the exact wording of all items retained for each subscale is reported in full in Appendix A. It could be that this appendix was not attached in the revised version of the manuscript and we apologize for this inconvenient.

Second, we have expanded the methodological rationale for item selection. Items were selected to ensure content representativeness of each motivational regulation, based on prior psychometric evidence reported in the original validation studies of the REBS, as well as their clarity and appropriateness for the present cultural and research context. Where relevant, considerations related to respondent burden and feasibility in the target sample are also explicitly acknowledged.

Third, the shortened version retains a theoretically coherent multi-factor structure in the present sample. Specifically, results from the CFA, standardized factor loadings, Composite Reliability (CR), Average Variance Extracted (AVE), and discriminant validity analyses (Fornell–Larcker criterion and HTMT) jointly support the adequacy of the measurement model. These results are reported in the Results section and in the Appendices.

Finally, the manuscript now explicitly discusses the trade-off inherent in using a shortened scale, namely the balance between reduced respondent burden and the potential risk of diminished content coverage. We clarify that, in the present study, this trade-off was addressed by combining theory-driven item selection with empirical validation within the sample. The manuscript has been revised accordingly (page 4, lines 179-190).

 

• SEM estimation and ordinal/Likert handling: specify, justify, align Right now, the analysis section needs to be estimator-appropriate and transparent. The authors must explicitly state: Software and version used for SEM. Estimator (ML / MLR / WLSMV) and a concise justification tied to the data properties (ordinal vs composite scores; distribution; sample size). How non-normality was handled (robust SEs, bootstrapping, etc.). Missing data: percent missing, pattern, and method (FIML/listwise/multiple imputation). If truly 0%, state it explicitly.

Response: Thank you for this comment. In the version revised after the first round of reviews, the software and version used, the estimator (ML), and information regarding missing data (there were no missing data) were already reported (page 6, lines 251–252). In response to this suggestion, we have now expanded this section by providing a clearer justification of the estimator in relation to the data properties (page 6, lines 256–261), thereby improving the transparency of the SEM estimation procedure. 

• Common-method bias: treat as a real threat, not a throwaway line A limitation paragraph is the minimum, but it should be written precisely: Explain why same-source, same-time measurement may inflate associations. Recommend concrete design remedies for future studies (time separation, multi-source outcomes, objective indicators where possible). If feasible within the current dataset, a simple diagnostic (marker variable or method-factor sensitivity) would strengthen credibility.

Response: Thank you for this important comment. We agree that common-method bias represents a potential threat when variables are assessed using the same source and at the same time point. Accordingly, we have revised the limitations section to more explicitly explain how same-source, same-time measurement may inflate observed associations (page 12, lines 452-456). In addition, we have expanded the discussion by outlining concrete design strategies that future studies could adopt to mitigate this issue, such as temporal separation of measurements, the inclusion of multi-source outcomes, and the use of objective indicators where feasible (page 12, lines 456-460). With regard to diagnostic tests, the current dataset does not include a suitable marker variable, nor was the study design intended to support a method-factor approach. For this reason, additional statistical diagnostics could not be meaningfully implemented.

• Explained variance (R² = 16%): fix the interpretation tone R² = 16% is meaningful but modest. The Discussion must: Avoid phrases implying strong prediction. Clearly acknowledge that adherence is likely influenced by unmeasured determinants (food environment, affordability, cooking resources, time constraints, campus availability, social and cultural context). Present findings as partial explanation, not a comprehensive model.

 

Response: Thank you for this comment. We agree that the explained variance in adherence (R² = 16%) should be interpreted as meaningful but modest. While this aspect was already acknowledged in the limitations section (page 10, lines 363-370), we have further revised the Discussion to ensure a consistently cautious interpretation of the findings. In particular, we have avoided language that may imply strong or comprehensive prediction and more explicitly framed the proposed model as offering a partial explanation of adherence to the Mediterranean diet. We now emphasize throughout the Discussion that a substantial proportion of variance is likely attributable to contextual and structural determinants not included in the model, such as the food environment, affordability, access to cooking resources, time constraints, campus availability, and broader social and cultural factors.

 

• Tables: eliminate statistical notation inconsistencies All correlation tables must have: Correct p-values and consistent significance markers. Clear footnotes defining symbols and thresholds. A final systematic cross-check so no cell is mis-flagged.

Response: Thank you for this comment. We have carefully revised all correlation tables to ensure full consistency in statistical notation. Specifically, p-values and significance markers have been checked and corrected where necessary, footnotes have been standardized to clearly define all symbols and significance thresholds, and a final systematic cross-check was conducted to verify that no cells were mis-flagged. The tables in the revised manuscript now use consistent and transparent notation throughout.

 

• Terminology/spelling: do a full consistency pass Typos and inconsistent spelling/capitalization reduce perceived rigor. The authors should run a full proofreading pass, including tables/figures.

Response: We have conducted a thorough proofreading of the entire manuscript, including the main text, tables, and figures, to ensure consistency in terminology, spelling, and capitalization throughout.

 

• QueMD scoring: clarify rationale and interpretability The scoring rule needs to be defensible and interpretable: Explain the rationale for any items that score 0 regardless of consumption. Clarify what the total score means in this population. If the scoring reduces sensitivity for young adults, the authors should either justify why this is acceptable or consider a sensitivity/alternative scoring check.

Response: Thank you for this comment. We have clarified both the rationale and the interpretability of the QueMD scoring procedure. Specifically, we now explain that items assigned a score of 0 refer to non-Mediterranean, highly processed foods (e.g., manufactured sweets and sugary beverages) and are included for descriptive purposes rather than contributing positively to the adherence score. In addition, we have clarified the meaning of the total QueMD score specifically in young adults, emphasizing that it reflects alignment with core Mediterranean diet components rather than fine-grained variability in dietary intake. We also explicitly state that this level of aggregation is consistent with the aims of the present study. Please find changes on page 5, lines 224-231.

• Covariates/robustness: be explicit and consistent If age correlates with adherence, the model should control for age and sex (or justify not doing so). This must be stated consistently in both: The text (analysis section), and The model figure and/or SEM output summary. Bottom line: The second revision should aim for a “fully auditable” measurement and analysis pipeline: complete CFA reporting, explicit validity evidence, estimator/missingness transparency, careful bias acknowledgement, and conservative interpretation of R². This is what will move the manuscript from “interesting but under-documented” to “publishable and defensible.”

Response: Thank you for this comment. As requested, the specification and reporting of covariates and robustness checks are now explicit and fully consistent across the manuscript. Age and sex were included as covariates in the SEM analyses, and this choice is clearly stated in the analysis section (page 6, lines 242-243), reflected in the model figure (figure 1, page 6), and reported in the SEM output summary (page 7, line 307-310).

More broadly, the revised manuscript now provides a fully transparent and auditable measurement and analysis pipeline. This includes complete CFA reporting, explicit validity evidence, clear documentation of the estimator and missing data information, a careful discussion of potential biases, and a conservative interpretation of the explained variance. We believe that these revisions substantially strengthen the methodological rigor and defensibility of the study.

 

Comments on the Quality of English Language

The English can be mostly understood, yet it needs professional editing due to: consistency of spelling and terms (US vs UK; repeated typos), punctuation and structure of the sentences (there are some long and unclear sentences), regular application of technical terms (names of constructions and abbreviations). Readability and perceived rigor will be significantly enhanced with a careful edit.

Response: Thank you for this comment. The manuscript has been carefully revised to improve clarity, consistency of spelling and terminology (including consistent use of US/UK conventions), sentence structure and punctuation, and the regular application of technical terms and abbreviations. We believe these revisions have enhanced the overall readability and rigor of the manuscript.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you very much for your careful and substantial revision of the manuscript. It is clear that you have engaged seriously with the previous review and have implemented many meaningful improvements at both the conceptual and methodological level. Compared to the earlier version, the revised manuscript clarifies the theoretical framing and the specific focus on emerging adulthood, provides a much more transparent and complete description of the methods (including recruitment procedures, context, and the psychometric treatment of the measures), introduces a dedicated section on the measurement model (CFA) with appropriate indices of convergent and discriminant validity, corrects the previous ambiguity in the interpretation of indirect effects (especially with regard to external regulation), significantly expands the Discussion and Limitations with a more nuanced engagement with the SDT–TPB literature and a clearer acknowledgement of the constraints of the design, and formulates more concrete implications for healthcare and public health practice. These changes collectively address the major concerns raised in the first round. At this stage, the remaining issues are mostly of a formal and presentational nature, together with a few small clarifications, and they appear compatible with a recommendation of acceptance after minor revision once they have been resolved.

 

The revised Introduction is notably stronger and more focused than in the previous version. You now situate the work explicitly within the context of emerging adulthood, highlighting this developmental period as a critical window for the consolidation of dietary habits and for the emergence of autonomy and identity-related processes. You articulate more clearly why Self-Determination Theory (SDT) and the Theory of Planned Behavior (TPB) are particularly suitable frameworks for understanding Mediterranean diet adherence in this age group, and you integrate and discuss in greater depth the existing literature on integrated SDT–TPB models applied to diet and health behaviour, including studies that are conceptually very close to your own work on Mediterranean diet and specific eating behaviours. In this revised framing, the novelty of your contribution is presented more accurately as incremental rather than as filling a completely unexplored gap. You state more explicitly that your study applies an integrated SDT–TPB model to Mediterranean diet adherence in a sample focused on emerging adults, using a detailed Mediterranean diet index, and you highlight this as the specific added value. It is important that the way you describe the TPB component remains fully consistent throughout the manuscript: the revised version already makes it more explicit that you focus on selected key constructs from TPB, namely perceived behavioural control and intention, rather than the full set of attitudes, subjective norms, perceived behavioural control and intention. Ensuring that this partial implementation of TPB is clearly and consistently described will prevent misunderstandings among readers.

 

The methodological description has also improved substantially. You now specify the recruitment procedures, the university context, the time window of data collection, and the inclusion criteria, and you make explicit that the sample is essentially a convenience sample of young adults, mostly university students. This is important both for transparency and for understanding the limits of generalisability. The description of the shortened regulation of eating behaviours scale (REBS) is much clearer than before: you indicate that you used three items per regulation and two for amotivation, explain the criteria for item selection, and provide example items for each regulation. This directly addresses the previous concerns about the use of an abbreviated version of a validated measure. You report internal consistencies for all subscales and, crucially, you now provide a separate section for the CFA / measurement model, including global fit indices, factor loadings, average variance extracted, composite reliabilities and HTMT values. This is a major step forward and aligns your psychometric reporting with contemporary standards. Moreover, you clarify the software (Mplus), the estimator (maximum likelihood), the absence of missing data, and the inclusion of age and gender as covariates in the structural model predicting Mediterranean diet adherence.

 

A few small methodological clarifications would still be useful for full transparency and replicability. It would be helpful to state explicitly how scale scores were computed (for example, whether you used the mean of the items for each subscale). This can be inferred from the ranges but should nonetheless be clearly described. In the section on the measurement model, you mention that the full set of standardized loadings is reported in “Table X”, yet in the current version such a table does not appear among the numbered tables. This seems to be an oversight introduced during revision. You will need either to remove the reference to this non-existent table and retain only the summary in the text, or to actually include the table with the next appropriate number and ensure that it is correctly referenced.

 

The presentation of the results is also more complete and reader-friendly than in the previous submission. Table 2 now includes means, standard deviations and ranges for all key variables, in addition to the correlations, which gives a much clearer picture of the data. The measurement model results are clearly separated from the structural model, with adequate justification of the factor structure and a thoughtful discussion of the high correlations between adjacent motivational regulations, which you appropriately interpret in the light of the SDT continuum. In the structural model, the main pattern of results is unchanged but now more carefully described: perceived behavioural control and intention predict adherence to the Mediterranean diet, intrinsic motivation shows a direct association with behaviour, and integrated regulation exerts an indirect effect through perceived behavioural control. Importantly, the previous inconsistency regarding external regulation has been resolved. In the earlier version, external regulation was reported as having a significant indirect effect despite confidence intervals including zero. In the revised manuscript you correctly indicate that the indirect effect for external regulation via perceived behavioural control is not statistically significant, and this correction is consistently reflected both in the text and in the mediation table.

 

Given that age and gender are now included as covariates in the structural equation model, it would be useful to state explicitly in the Results whether they show any significant association with Mediterranean diet adherence, even if such effects are small. A single sentence would be sufficient and would prevent readers from wondering about these paths. In addition, you report that the model explains around 40% of the variance in intention and about 16% of the variance in behaviour. You already discuss this in the Discussion, but emphasising in the Results section that the explained variance for behaviour is modest yet typical for complex health behaviours can help frame expectations appropriately and reinforce the need to consider additional contextual factors.

 

The Discussion has been considerably enriched in the revised version and compares favourably with the more concise discussion in the previous manuscript. You now engage more deeply with the existing SDT–TPB literature, systematically comparing your findings to those of related models on diet and health behaviours, and highlighting both convergences and specificities. You stress the central role of perceived behavioural control and intrinsic motivation, while carefully examining the non-significant effects for identified and introjected regulation and considering what these patterns might imply about the motivational dynamics of Mediterranean diet adherence in emerging adults. You explicitly address the relatively modest explained variance for behaviour and argue, appropriately, that additional contextual and environmental factors, such as food availability, economic constraints and time pressures, are likely to play a substantial role. The limitations section is now more complete and more sophisticated than in the previous version: you discuss the cross-sectional design, the reliance on self-report, the potential for social desirability and common-method bias, the lack of objective indicators such as BMI, and the restricted sociodemographic variability of the sample, underlining the need for replication in more diverse groups of young adults.

Of particular relevance for Healthcare, the practical implications are now articulated in a much more concrete way than in the earlier manuscript. You outline how autonomy-supportive communication and enhancement of perceived behavioural control could be implemented in real-world settings, including university health promotion programmes (through workshops on meal planning, choosing healthier options in canteens, and basic Mediterranean cooking skills), primary care (through brief counselling that links healthy eating to personal goals and values), and broader public health initiatives targeting emerging adults. This translation of your findings into tangible strategies for interventions in healthcare and educational settings represents an important added value and fits well with the journal’s orientation.

 

Finally, with respect to language and formal issues, the overall English in the revised version is clearly improved and fully adequate for publication, although a final light polishing would still be beneficial. It would be advisable to correct a few typographical and wording errors that appear to have remained or emerged in the revision, such as “World Heart Organization (WHO)” instead of “World Health Organization (WHO)” and “50.1%% women” instead of “50.1% women”. There are also occasional inconsistencies between British and American spelling (for example, behaviour/behavior), and choosing one variety and applying it consistently would improve stylistic coherence. Some sentences, especially in the Introduction and Discussion, remain quite long and dense; breaking a few of these into shorter, more focused sentences would further enhance readability without altering content. In addition, a careful final check of the reference list and DOIs is strongly recommended. In the previous version there were a few DOIs that appeared to be copied incorrectly or mismatched with the cited article, and it is important to verify that each DOI corresponds to the correct reference and that all references comply with the journal’s style.

 

In summary, the revised manuscript represents a substantial improvement over the initial submission and addresses the major conceptual and methodological concerns raised in the first review. The study now offers a clearer and more rigorously documented contribution to understanding how SDT- and TPB-based constructs relate to adherence to the Mediterranean diet in emerging adults, with a convincing discussion of implications for healthcare and public health practice. The remaining issues concern a missing or misreferenced table of factor loadings, minor methodological clarifications (such as scoring procedures and explicit mention of age and gender effects), a small set of typographical and stylistic corrections, and a careful final check of references and DOIs. Once these points have been addressed, the manuscript will, in my view, be in a very good position for acceptance in Healthcare.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

Overall, the quality of English is good and the manuscript is perfectly understandable throughout. However, a light but careful language revision would improve clarity, flow and consistency, and would help the paper meet the stylistic standards of Healthcare.

More specifically:

• There are a few minor typographical errors (e.g., “Mediterrenean” instead of “Mediterranean” and occasional small slips in plural/singular or articles) that should be corrected.

• The manuscript alternates between British and American spelling (e.g., “behaviour” vs. “behavior”, “organisation” vs. “organization”). It would be preferable to choose one variant and use it consistently across the text.

• Some sentences in the Introduction and Discussion are rather long and dense, with multiple clauses and several concepts embedded in a single period. These could be broken into shorter sentences or slightly rephrased to improve readability and make the key messages more immediately accessible to readers.

• A few expressions could be made more precise or natural (for example, replacing generic formulations such as “in the light of” or “in the frame of” with more standard academic phrasing). This is not a major issue, but small adjustments would further polish the style.

• Please make sure that all abbreviations (e.g., WHO, TPB, SDT, PBC, MD) are spelled out in full at their first occurrence in the main text and used consistently thereafter. This will also help international readers outside the psychological field.

• Punctuation is generally correct, but a careful final check of commas and conjunctions in complex sentences would help avoid occasional ambiguities and enhance the logical flow.

In summary, the English language does not require a major overhaul, but I would recommend a light professional copy-edit or a thorough internal language revision to correct minor errors, harmonise spelling, simplify a few complex sentences, and ensure full consistency of terminology and abbreviations. This would significantly enhance the overall clarity and presentation of your work.

Author Response

Thank you very much for your careful and substantial revision of the manuscript. It is clear that you have engaged seriously with the previous review and have implemented many meaningful improvements at both the conceptual and methodological level. Compared to the earlier version, the revised manuscript clarifies the theoretical framing and the specific focus on emerging adulthood, provides a much more transparent and complete description of the methods (including recruitment procedures, context, and the psychometric treatment of the measures), introduces a dedicated section on the measurement model (CFA) with appropriate indices of convergent and discriminant validity, corrects the previous ambiguity in the interpretation of indirect effects (especially with regard to external regulation), significantly expands the Discussion and Limitations with a more nuanced engagement with the SDT–TPB literature and a clearer acknowledgement of the constraints of the design, and formulates more concrete implications for healthcare and public health practice. These changes collectively address the major concerns raised in the first round. At this stage, the remaining issues are mostly of a formal and presentational nature, together with a few small clarifications, and they appear compatible with a recommendation of acceptance after minor revision once they have been resolved.

 

Response: Thank you very much for this positive and encouraging evaluation. We appreciate your careful assessment and are pleased that the revisions have addressed the major concerns raised in the first round. The remaining minor points have been addressed accordingly in the revised manuscript (see below).

 

The revised Introduction is notably stronger and more focused than in the previous version. You now situate the work explicitly within the context of emerging adulthood, highlighting this developmental period as a critical window for the consolidation of dietary habits and for the emergence of autonomy and identity-related processes. You articulate more clearly why Self-Determination Theory (SDT) and the Theory of Planned Behavior (TPB) are particularly suitable frameworks for understanding Mediterranean diet adherence in this age group, and you integrate and discuss in greater depth the existing literature on integrated SDT–TPB models applied to diet and health behaviour, including studies that are conceptually very close to your own work on Mediterranean diet and specific eating behaviours. In this revised framing, the novelty of your contribution is presented more accurately as incremental rather than as filling a completely unexplored gap. You state more explicitly that your study applies an integrated SDT–TPB model to Mediterranean diet adherence in a sample focused on emerging adults, using a detailed Mediterranean diet index, and you highlight this as the specific added value. It is important that the way you describe the TPB component remains fully consistent throughout the manuscript: the revised version already makes it more explicit that you focus on selected key constructs from TPB, namely perceived behavioural control and intention, rather than the full set of attitudes, subjective norms, perceived behavioural control and intention. Ensuring that this partial implementation of TPB is clearly and consistently described will prevent misunderstandings among readers.

Response: Thank you for this positive and constructive feedback. We are pleased that the revised Introduction is clearer and more focused, and that the positioning of the study within emerging adulthood and within the integrated SDT–TPB framework is now more explicit. In line with your suggestion, we have ensured that the description of the TPB component is fully consistent throughout the manuscript, clearly specifying that the model focuses on selected key TPB constructs (perceived behavioral control and intention) rather than the full TPB framework.

 

 

The methodological description has also improved substantially. You now specify the recruitment procedures, the university context, the time window of data collection, and the inclusion criteria, and you make explicit that the sample is essentially a convenience sample of young adults, mostly university students. This is important both for transparency and for understanding the limits of generalisability. The description of the shortened regulation of eating behaviours scale (REBS) is much clearer than before: you indicate that you used three items per regulation and two for amotivation, explain the criteria for item selection, and provide example items for each regulation. This directly addresses the previous concerns about the use of an abbreviated version of a validated measure. You report internal consistencies for all subscales and, crucially, you now provide a separate section for the CFA / measurement model, including global fit indices, factor loadings, average variance extracted, composite reliabilities and HTMT values. This is a major step forward and aligns your psychometric reporting with contemporary standards. Moreover, you clarify the software (Mplus), the estimator (maximum likelihood), the absence of missing data, and the inclusion of age and gender as covariates in the structural model predicting Mediterranean diet adherence.

 

Response: Thank you for this positive feedback. We appreciate the reviewer’s careful evaluation of the methodological improvements and are pleased that the revisions have addressed the previous concerns regarding transparency, psychometric reporting, and model specification.

 

A few small methodological clarifications would still be useful for full transparency and replicability. It would be helpful to state explicitly how scale scores were computed (for example, whether you used the mean of the items for each subscale). This can be inferred from the ranges but should nonetheless be clearly described. In the section on the measurement model, you mention that the full set of standardized loadings is reported in “Table X”, yet in the current version such a table does not appear among the numbered tables. This seems to be an oversight introduced during revision. You will need either to remove the reference to this non-existent table and retain only the summary in the text, or to actually include the table with the next appropriate number and ensure that it is correctly referenced.

 

Response: Thank you for this comment. We have now explicitly clarified how scale scores were computed (i.e., by averaging the items for each subscale; page 7, lines 273-275). In addition, the reference to “Table X” was an oversight introduced during revision. The corresponding information has been moved to Appendix A, which has now been included, and the in-text reference has been corrected accordingly.

 

The presentation of the results is also more complete and reader-friendly than in the previous submission. Table 2 now includes means, standard deviations and ranges for all key variables, in addition to the correlations, which gives a much clearer picture of the data. The measurement model results are clearly separated from the structural model, with adequate justification of the factor structure and a thoughtful discussion of the high correlations between adjacent motivational regulations, which you appropriately interpret in the light of the SDT continuum. In the structural model, the main pattern of results is unchanged but now more carefully described: perceived behavioural control and intention predict adherence to the Mediterranean diet, intrinsic motivation shows a direct association with behaviour, and integrated regulation exerts an indirect effect through perceived behavioural control. Importantly, the previous inconsistency regarding external regulation has been resolved. In the earlier version, external regulation was reported as having a significant indirect effect despite confidence intervals including zero. In the revised manuscript you correctly indicate that the indirect effect for external regulation via perceived behavioural control is not statistically significant, and this correction is consistently reflected both in the text and in the mediation table.

 

Response: Thank you for this positive feedback. We appreciate the reviewer’s careful evaluation of the revised Results section and are pleased that the improvements in clarity, consistency, and interpretation have addressed the issues raised in the previous round.

 

Given that age and gender are now included as covariates in the structural equation model, it would be useful to state explicitly in the Results whether they show any significant association with Mediterranean diet adherence, even if such effects are small. A single sentence would be sufficient and would prevent readers from wondering about these paths. In addition, you report that the model explains around 40% of the variance in intention and about 16% of the variance in behaviour. You already discuss this in the Discussion, but emphasising in the Results section that the explained variance for behaviour is modest yet typical for complex health behaviours can help frame expectations appropriately and reinforce the need to consider additional contextual factors.

 

Response: Thank you for this comment. We have now explicitly reported in the Results whether age and gender show significant associations with Mediterranean diet adherence (page 7, lines 308-310). In addition, we have clarified in the Results section that the proportion of explained variance for behaviour is modest yet typical for complex health behaviours, thereby framing expectations more appropriately and reinforcing the relevance of additional contextual factors (page 7, lines 320-323).

 

The Discussion has been considerably enriched in the revised version and compares favourably with the more concise discussion in the previous manuscript. You now engage more deeply with the existing SDT–TPB literature, systematically comparing your findings to those of related models on diet and health behaviours, and highlighting both convergences and specificities. You stress the central role of perceived behavioural control and intrinsic motivation, while carefully examining the non-significant effects for identified and introjected regulation and considering what these patterns might imply about the motivational dynamics of Mediterranean diet adherence in emerging adults. You explicitly address the relatively modest explained variance for behaviour and argue, appropriately, that additional contextual and environmental factors, such as food availability, economic constraints and time pressures, are likely to play a substantial role. The limitations section is now more complete and more sophisticated than in the previous version: you discuss the cross-sectional design, the reliance on self-report, the potential for social desirability and common-method bias, the lack of objective indicators such as BMI, and the restricted sociodemographic variability of the sample, underlining the need for replication in more diverse groups of young adults.

 

Response: Thank you for this positive and thoughtful evaluation. We appreciate the reviewer’s careful reading of the revised Discussion and Limitations sections and are pleased that the revisions have strengthened the theoretical integration, interpretation of the findings, and acknowledgement of the study’s constraints.

Of particular relevance for Healthcare, the practical implications are now articulated in a much more concrete way than in the earlier manuscript. You outline how autonomy-supportive communication and enhancement of perceived behavioural control could be implemented in real-world settings, including university health promotion programmes (through workshops on meal planning, choosing healthier options in canteens, and basic Mediterranean cooking skills), primary care (through brief counselling that links healthy eating to personal goals and values), and broader public health initiatives targeting emerging adults. This translation of your findings into tangible strategies for interventions in healthcare and educational settings represents an important added value and fits well with the journal’s orientation.

 

Response: Thank you for this positive feedback. We appreciate the reviewer’s assessment of the strengthened practical implications and are pleased that the translation of our findings into concrete healthcare and educational strategies is considered a valuable contribution aligned with the journal’s aims.

 

Finally, with respect to language and formal issues, the overall English in the revised version is clearly improved and fully adequate for publication, although a final light polishing would still be beneficial. It would be advisable to correct a few typographical and wording errors that appear to have remained or emerged in the revision, such as “World Heart Organization (WHO)” instead of “World Health Organization (WHO)” and “50.1%% women” instead of “50.1% women”. There are also occasional inconsistencies between British and American spelling (for example, behaviour/behavior), and choosing one variety and applying it consistently would improve stylistic coherence. Some sentences, especially in the Introduction and Discussion, remain quite long and dense; breaking a few of these into shorter, more focused sentences would further enhance readability without altering content. In addition, a careful final check of the reference list and DOIs is strongly recommended. In the previous version there were a few DOIs that appeared to be copied incorrectly or mismatched with the cited article, and it is important to verify that each DOI corresponds to the correct reference and that all references comply with the journal’s style.

 

Response: Thank you for this comment. We have performed a final careful proofreading to correct remaining typographical and wording errors, ensure consistent use of spelling conventions, improve sentence clarity where needed, and verify the accuracy and formatting of all references and DOIs in accordance with the journal’s style.

 

In summary, the revised manuscript represents a substantial improvement over the initial submission and addresses the major conceptual and methodological concerns raised in the first review. The study now offers a clearer and more rigorously documented contribution to understanding how SDT- and TPB-based constructs relate to adherence to the Mediterranean diet in emerging adults, with a convincing discussion of implications for healthcare and public health practice. The remaining issues concern a missing or misreferenced table of factor loadings, minor methodological clarifications (such as scoring procedures and explicit mention of age and gender effects), a small set of typographical and stylistic corrections, and a careful final check of references and DOIs. Once these points have been addressed, the manuscript will, in my view, be in a very good position for acceptance in Healthcare.

 

Response: Thank you for this comprehensive and encouraging evaluation. We appreciate the reviewer’s assessment of the substantial improvements made to the manuscript. All remaining minor issues have been addressed accordingly, including the clarification of scoring procedures, explicit reporting of age and gender effects, correction of tables and references, and final typographical and stylistic revisions.

peer-review-53735018.v2.pdf

Comments on the Quality of English Language

Overall, the quality of English is good and the manuscript is perfectly understandable throughout. However, a light but careful language revision would improve clarity, flow and consistency, and would help the paper meet the stylistic standards of Healthcare.

More specifically:

• There are a few minor typographical errors (e.g., “Mediterrenean” instead of “Mediterranean” and occasional small slips in plural/singular or articles) that should be corrected.
• The manuscript alternates between British and American spelling (e.g., “behaviour” vs. “behavior”, “organisation” vs. “organization”). It would be preferable to choose one variant and use it consistently across the text.
• Some sentences in the Introduction and Discussion are rather long and dense, with multiple clauses and several concepts embedded in a single period. These could be broken into shorter sentences or slightly rephrased to improve readability and make the key messages more immediately accessible to readers.
• A few expressions could be made more precise or natural (for example, replacing generic formulations such as “in the light of” or “in the frame of” with more standard academic phrasing). This is not a major issue, but small adjustments would further polish the style.
• Please make sure that all abbreviations (e.g., WHO, TPB, SDT, PBC, MD) are spelled out in full at their first occurrence in the main text and used consistently thereafter. This will also help international readers outside the psychological field.
• Punctuation is generally correct, but a careful final check of commas and conjunctions in complex sentences would help avoid occasional ambiguities and enhance the logical flow.

In summary, the English language does not require a major overhaul, but I would recommend a light professional copy-edit or a thorough internal language revision to correct minor errors, harmonise spelling, simplify a few complex sentences, and ensure full consistency of terminology and abbreviations. This would significantly enhance the overall clarity and presentation of your work.

 

Response: Thank you for this suggestion. The manuscript has undergone a careful final internal language revision to correct minor errors, harmonise spelling, simplify complex sentences where appropriate, and ensure consistent use of terminology and abbreviations, thereby improving overall clarity and presentation.

Author Response File: Author Response.pdf