Association between the Degree of Processing of Consumed Foods and Sleep Quality in Adolescents

Round 1

Reviewer 1 Report

The results are  expected, but this study provided an evidence using Pittsburgh Sleep Quality Index for evaluation of sleep quality through considerable interview procedures.

there are still questions needed to be addressed:

1. This study protocol approval can be moved to the first paragraph of the method section. Figure 1. Directed acyclic graph (GAD) is not study's result and can be removed from the main text and use as supplement figure.

2. why choose stats of  Poisson regression for association tests ? is there any reference? please cite.

3. if authors could describe what does it mean by the NOVA classification and therefore could further discuss what kinds of food were the most risks for adolescents, to increase the interest to the readers.

4. Line 23: in the association results also indicated the calories such as  (44.3–81,8% of total calories) according to Table 3. however, the contribution of energy to the sleep quality issues was not clearly presented and discussed (only Line 232-239). 

5. GAD mentioned coffee consumption and it also can significant change sleep quality pharmacologically. Can coffee consumption factor information be listed in the result table including the drinking time and number of cups? instead of merging into the NOVA classes.

Author Response

Dear editors and reviewers,We appreciate the contributions that you sent. The requested changes were made. The modified text is highlighted in yellow in the manuscript and described below:

REVIEWER I

The results are expected, but this study provided an evidence using Pittsburgh Sleep Quality Index for evaluation of sleep quality through considerable interview procedures.

There are still questions needed to be addressed:

This study protocol approval can be moved to the first paragraph of the method section.

Response: This correction was done (Lines 66-68).

Figure 1. Directed acyclic graph (GAD) is not study's result and can be removed from the main text and use as supplement figure.

Response: This correction was done (Line 148).

Why choose stats of Poisson regression for association tests? Is there any reference? Please cite.

Response: We decided to use Poisson regression, because in cross-sectional studies in which the dependent variable has an intermediate (≥ 10% and <50%) and high (>50%) prevalence, this regression has been shown to be more accurate than the logistic regression (since the odds ratio tends to overestimate associations for results with intermediate and high prevalence). We included the following reference in the manuscript as a justification (Lines 167-168).

Ref.: Coutinho, L.M.S.; Scazufca, M.; Menezes, P.R. Methods for estimating prevalence ratios in cross- sectional studies. Rev. Saude Publica 2008, 42, 992–998.

If authors could describe what does it mean by the NOVA classification and therefore could further discuss what kinds of food were the most risks for adolescents, to increase the interest to the readers.

Response: The NOVA classification was described in the introduction (Lines 36-46) and in the discussion it was commented which groups and foods presented the greatest risks to health and sleep quality (lines 255-262).

Line 23: in the association results also indicated the calories such as (44.3–81,8% of total calories) according to Table 3. however, the contribution of energy to the sleep quality issues was not clearly presented and discussed (only Lines 232-239).

Response: The results included information on energy contributions from quartiles that had an association with poor sleep quality (Lines 197-198, 208, 210,211), which was better commented in the discussion (Line 272-276).

GAD mentioned coffee consumption and it also can significant change sleep quality pharmacologically. Can coffee consumption factor information be listed in the result table including the drinking time and number of cups? instead of merging into the NOVA classes.

Response: This correction was done in Table 1 (lines 177, 178), which was better commented in the discussion (Lines 260-262).

REVIEWER II 

This article presents the association between food consumption, classified by its level of processing, and sleep quality in 18-19 year-old adolescents from the city of São Luís, Maranhão, Brazil; showing in a clear and pleasant way that there is a relationship between diet and sleep quality. The size of the n is reasonable and the data collection instruments have been previously validated for the Brazilian population. Therefore, this study presents robust results that may serve as a starting point for future research that explores this association in depth or that may even determine if there is causality.

Nevertheless, I believe that there are some points that should be reviewed/corrected to improve the quality of this manuscript, which I detail below: Line 167: Go to the next page. It is not advisable to leave the titles of the manuscript out of the body of the text, on separate pages.

Response: This correction was done (Line 174).

Line 170: Mention that the predominant population was composed of adolescents WITHOUT separated/divorced parents (51.5%), while Table 1 shows a predominance of adolescents WITH separated/divorced parents. The information provided in this line and/or in Table 1 should be reviewed and corrected.

Response: This correction was done (Line 177). 

Lines 180-181: Go to next page. Whenever possible, separating a table into two pages should be avoided. The reader of your manuscript will appreciate this.

Response: This correction was done (Line 181).

Lines 197-198: Same comments as for lines 180-181.

Response: This correction was done.

Lines 213-231: Since the content of these paragraphs focuses on the strengths and limitations of the study, I consider it appropriate to close the "Results" section with these notes, along with recommendations for future research. These changes would allow us to focus on the most relevant findings of this research, respecting the structure expected in this section.

Response: The strengths (Lines 234-237) and limitations (Lines 219-233) of this study was included in the final part of the results and contributions were made for future research (Lines 238, 239).

Lines 232-244: These paragraphs should be enriched bibliographically. The results addressed in them correspond to the backbone of this section and of the research; however, the findings are contrasted only with two studies of the same line as the present work -except for the studies that have been included to explain the possible causes of such results.

Response: New studies were added to support the results found (Lines 255-266). Line 249: It would sound more natural to replace "We can cite some of these nutrients: tryptophan, isoflavones..." with "Some of these nutrients are tryptophan, isoflavones...".
Response: This change was made (Line 281).

Lines 289-291: I suggest restructuring these lines so that the "form" is more correct, but the substance remains: "However, because this is a cross-sectional study, it has not been possible to assess whether there is causality between the consumption of this type of food and the quality of sleep."

Response: This correction was done (319-320). Line 316: The name of the institution must be in full, followed by its acronym in parentheses: Instituto Brasileiro de Geografia e Estatística (IBGE).

Response: This correction was done (Line 344).

Author Response File: Author Response.docx

Reviewer 2 Report

This article presents the association between food consumption, classified by its level of processing, and sleep quality in 18-19 year-old adolescents from the city of São Luís, Maranhão, Brazil; showing in a clear and pleasant way that there is a relationship between diet and sleep quality. The size of the n is reasonable and the data collection instruments have been previously validated for the Brazilian population. Therefore, this study presents robust results that may serve as a starting point for future research that explores this association in depth or that may even determine if there is causality.
Nevertheless, I believe that there are some points that should be reviewed/corrected to improve the quality of this manuscript, which I detail below:Line 167: Go to the next page. It is not advisable to leave the titles of the manuscript out of the body of the text, on separate pages.
Line 170: Mention that the predominant population was composed of adolescents WITHOUT separated/divorced parents (51.5%), while Table 1 shows a predominance of adolescents WITH separated/divorced parents. The information provided in this line and/or in Table 1 should be reviewed and corrected.
Lines 180-181: Go to next page. Whenever possible, separating a table into two pages should be avoided. The reader of your manuscript will appreciate this.
Lines 197-198: Same comments as for lines 180-181.
Lines 213-231: Since the content of these paragraphs focuses on the strengths and limitations of the study, I consider it appropriate to close the "Results" section with these notes, along with recommendations for future research. These changes would allow us to focus on the most relevant findings of this research, respecting the structure expected in this section.
Lines 232-244: These paragraphs should be enriched bibliographically. The results addressed in them correspond to the backbone of this section and of the research; however, the findings are contrasted only with two studies of the same line as the present work -except for the studies that have been included to explain the possible causes of such results.
Line 249: It would sound more natural to replace "We can cite some of these nutrients: tryptophan, isoflavones..." with "Some of these nutrients are tryptophan, isoflavones...".
Lines 289-291: I suggest restructuring these lines so that the "form" is more correct, but the substance remains: "However, because this is a cross-sectional study, it has not been possible to assess whether there is causality between the consumption of this type of food and the quality of sleep."
Line 316: The name of the institution must be in full, followed by its acronym in parentheses: Instituto Brasileiro de Geografia e Estatística (IBGE).

Author Response

Dear editors and reviewers,We appreciate the contributions that you sent. The requested changes were made. The modified text is highlighted in yellow in the manuscript and described below:

REVIEWER I

The results are expected, but this study provided an evidence using Pittsburgh Sleep Quality Index for evaluation of sleep quality through considerable interview procedures.

There are still questions needed to be addressed:

This study protocol approval can be moved to the first paragraph of the method section.

Response: This correction was done (Lines 66-68).

Figure 1. Directed acyclic graph (GAD) is not study's result and can be removed from the main text and use as supplement figure.

Response: This correction was done (Line 148).

Why choose stats of Poisson regression for association tests? Is there any reference? Please cite.

Response: We decided to use Poisson regression, because in cross-sectional studies in which the dependent variable has an intermediate (≥ 10% and <50%) and high (>50%) prevalence, this regression has been shown to be more accurate than the logistic regression (since the odds ratio tends to overestimate associations for results with intermediate and high prevalence). We included the following reference in the manuscript as a justification (Lines 167-168).

Ref.: Coutinho, L.M.S.; Scazufca, M.; Menezes, P.R. Methods for estimating prevalence ratios in cross- sectional studies. Rev. Saude Publica 2008, 42, 992–998.

If authors could describe what does it mean by the NOVA classification and therefore could further discuss what kinds of food were the most risks for adolescents, to increase the interest to the readers.

Response: The NOVA classification was described in the introduction (Lines 36-46) and in the discussion it was commented which groups and foods presented the greatest risks to health and sleep quality (lines 255-262).

Line 23: in the association results also indicated the calories such as (44.3–81,8% of total calories) according to Table 3. however, the contribution of energy to the sleep quality issues was not clearly presented and discussed (only Lines 232-239).

Response: The results included information on energy contributions from quartiles that had an association with poor sleep quality (Lines 197-198, 208, 210,211), which was better commented in the discussion (Line 272-276).

GAD mentioned coffee consumption and it also can significant change sleep quality pharmacologically. Can coffee consumption factor information be listed in the result table including the drinking time and number of cups? instead of merging into the NOVA classes.

Response: This correction was done in Table 1 (lines 177, 178), which was better commented in the discussion (Lines 260-262).

REVIEWER II

This article presents the association between food consumption, classified by its level of processing, and sleep quality in 18-19 year-old adolescents from the city of São Luís, Maranhão, Brazil; showing in a clear and pleasant way that there is a relationship between diet and sleep quality. The size of the n is reasonable and the data collection instruments have been previously validated for the Brazilian population. Therefore, this study presents robust results that may serve as a starting point for future research that explores this association in depth or that may even determine if there is causality.

Nevertheless, I believe that there are some points that should be reviewed/corrected to improve the quality of this manuscript, which I detail below: Line 167: Go to the next page. It is not advisable to leave the titles of the manuscript out of the body of the text, on separate pages.

Response: This correction was done (Line 174).

Line 170: Mention that the predominant population was composed of adolescents WITHOUT separated/divorced parents (51.5%), while Table 1 shows a predominance of adolescents WITH separated/divorced parents. The information provided in this line and/or in Table 1 should be reviewed and corrected.

Response: This correction was done (Line 177).

Lines 180-181: Go to next page. Whenever possible, separating a table into two pages should be avoided. The reader of your manuscript will appreciate this.

Response: This correction was done (Line 181).

Lines 197-198: Same comments as for lines 180-181.

Response: This correction was done.

Lines 213-231: Since the content of these paragraphs focuses on the strengths and limitations of the study, I consider it appropriate to close the "Results" section with these notes, along with recommendations for future research. These changes would allow us to focus on the most relevant findings of this research, respecting the structure expected in this section.

Response: The strengths (Lines 234-237) and limitations (Lines 219-233) of this study was included in the final part of the results and contributions were made for future research (Lines 238, 239).

Lines 232-244: These paragraphs should be enriched bibliographically. The results addressed in them correspond to the backbone of this section and of the research; however, the findings are contrasted only with two studies of the same line as the present work -except for the studies that have been included to explain the possible causes of such results.

Response: New studies were added to support the results found (Lines 255-266). Line 249: It would sound more natural to replace "We can cite some of these nutrients: tryptophan, isoflavones..." with "Some of these nutrients are tryptophan, isoflavones...".
Response: This change was made (Line 281).

Lines 289-291: I suggest restructuring these lines so that the "form" is more correct, but the substance remains: "However, because this is a cross-sectional study, it has not been possible to assess whether there is causality between the consumption of this type of food and the quality of sleep."

Response: This correction was done (319-320). Line 316: The name of the institution must be in full, followed by its acronym in parentheses: Instituto Brasileiro de Geografia e Estatística (IBGE).

Response: This correction was done (Line 344).

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Responses are received and are address for each comment; however, I believed that the lines 272-276 is regarding the coffee consumption (cover letter indicates Lines 260-262)

The inserted information in the discussion clarify certain points, however, the rationales of each discussion paragraph cannot link to each other.

For example: Authors tried to focus on their main results as following:

Our main [Line 244-246] findings were that higher energy contributions from FMPF were associated with a lower prevalence of poor sleep quality; whereas high energy contributions from UPF were associated with a higher prevalence of poor sleep quality.

However, the followed discussions mixing those information and did not clearly expression their points.

Please try to re-organize (rewrite and make the paragraph more concisely) the first five paragraphs. For example: their first paragraph is to emphasize the main findings (participants consumed more FMPF therefore could have good quality of sleep; even though also consumed UPF). Second paragraph can point out the high prevalence of percentage like previous reference33 and then try to clarify several factors which will support the main findings that those participants did not have good quality of sleep may be due to high consumption (HOW HIGH?) of UPF and also higher consumptions (>50% participant drink more than once per day etc…..) of coffee, and also higher energy (please describe the data!  Is this also meant for high calories? What does it mean by AUF [Line260]) etc. So, authors can compare their data of FMPF to that (69.5%) of a reference more clearly.

Otherwise, it is hard to understand why to mention about homemade culinary preparations [Line 253] to the energy consumption. There is no such information in the other text of this manuscript.

Authors could try to subgroup the participants into good or bad sleep and to compare their differences in consumptions of FMPF, UPF or coffee etc. Then, this comparison will assist you to write above discussion and compared to data from the cited references.

Author Response

Dear Reviewer,

Thank you very much for your letter and for your careful revision of our paper. Please find enclosed our point by point answers to each of the reviewer’s comments. The amendments are highlighted in yellow in the annotated version of the manuscript. We hope that you will find this rerevised version suitable for publication.

Best regards

Maylla Bragança

REVIEWER I – second round

Responses are received and are address for each comment; however, I believed that the lines 272-276 is regarding the coffee consumption (cover letter indicates Lines 260-262).

Response: We apologize for the inconsistency in the lines referenced in the cover letter.

The inserted information in the discussion clarify certain points, however, the rationales of each discussion paragraph cannot link to each other. For example: Authors tried to focus on their main results as following:

Our main [Line 244-246] findings were that higher energy contributions from FMPF were associated with a lower prevalence of poor sleep quality; whereas high energy contributions from UPF were associated with a higher prevalence of poor sleep quality.

However, the followed discussions mixing those information and did not clearly expression their points.

Please try to re-organize (rewrite and make the paragraph more concisely) the first five paragraphs. For example: their first paragraph is to emphasize the main findings (participants consumed more FMPF therefore could have good quality of sleep; even though also consumed UPF). Second paragraph can point out the high prevalence of percentage like previous reference33 and then try to clarify several factors which will support the main findings that those participants did not have good quality of sleep may be due to high consumption (HOW HIGH?) of UPF and also higher consumptions (>50% participant drink more than once per day etc…..) of coffee, and also higher energy (please describe the data!  Is this also meant for high calories? What does it mean by AUF [Line260]) etc. So, authors can compare their data of FMPF to that (69.5%) of a reference more clearly.

Response: The first five paragraphs were rewritten considering other study’s main findings, such as the new information included in Results (Table 1 – lines 183-185. Comments of Table 1 – lines 179-181. Table 2 – 194-196. Comments of Table 2 - lines 190-192) which report the sociodemographic, lifestyle and nutrient consumption characteristics according to the quality of sleep (poor or good). Thus, in the first paragraph we considered other study’s main results (Lines 249-252) which were discussed in the following paragraphs (Lines 262-265, 271-274, 288-291).

Otherwise, it is hard to understand why to mention about homemade culinary preparations [Line 253] to the energy consumption. There is no such information in the other text of this manuscript.

Response: This part was deleted from the manuscript (Lines 260-262).

Authors could try to subgroup the participants into good or bad sleep and to compare their differences in consumptions of FMPF, UPF or coffee etc. Then, this comparison will assist you to write above discussion and compared to data from the cited references.

Response: This correction was made (Table 1 – lines 183-185. Comments of Table 1 – lines 179-181. Table 2 – 194-196. Comments of Table 2 – lines 190-192).

Author Response File: Author Response.docx