Degree of Food Processing (NOVA Classification) and Blood Pressure in Women with Overweight and Obesity

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The subject of this study, the relationship between the degree of food processing and blood pressure values, is novel and of great interest to readers of this journal.

Although the introduction is well elaborated and based on an ample and updated bibliography, it could be expanded a little to better focus the subject of the article.

The methodology is correct and allows the research to be reproduced in the event that a researcher deems it appropriate. The objectives are clear.

The results allow achieving the proposed objectives. These results are very well supported by tables and figures that facilitate their follow-up and understanding.

The discussion is very complete and well-founded and allows comparison of the results obtained by the authors.

The bibliography is very up to date since the obsolescence index (median age of citations) is very low at 3 years.

Author Response

Please see the attached file

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

A cross-sectional study was conducted to determine the relationship between the degree of food processing and blood pressure among 85 adult women in Brazil. The authors hypothesized that “women with obesity would have worse food consumption, characterized by higher consumption of ultra processed foods (UPFs) and lower consumption of unprocessed or minimally processed foods (MPFs), contributing to increased blood pressure levels”.  Food consumption was determined with a food frequency questionnaire (FFQ) and then the degree of processing was determined. Yet, the details for the former were sparse and FFQ does not provide details for food preparation. In this study, hypertension was low as only 7% (6 of 85) of the women in the study and 18% of the women with obesity (6 of 33) were hypertensive. Contrary to the hypothesis, there was no difference in unprocessed/minimally processed food (MPF), processed food (PF) or ultra-processed food (UPF) or cooking ingredients between women classified as normal weight and obese (figure 1). Associations between the degrees of food processing (MPF, PF, UPF, cooking ingredients) and systolic and diastolic (SBP and DBP) were tested. After adjusting for age, medications and pathology; consumption of fewer MPF was associated with higher DBP but not SBP (table 4). Given the low prevalence of hypertension in this study and similar consumption of MPF, PF, UPF between women who were normal weight and had obesity, several questions remain.

 

1.    Add details to the methods. 

a.    Study design: add the inclusion criteria and how the women were recruited for the study. Also, were there any additional exclusion criteria aside from “Pregnant women and women who hadpathologies that could affect their nutritional status and/or consumption were excluded from thestudy”?

b.    While the discussion stated that use of the FFQ was a limitation, add a greater description for how the foods were classified according to the degree of food processing with the NOVA classification. For example, the manuscript does not address the NOVA classification (it is only in the title), how many people classified each women’s food consumption and what was the inter-rater reliability? These are important details since the NOVA classification has been highly scrutinized recently.

c.     Line 125: The methods state that “This procedure was repeated on two different days.”. Add more details for how many days between the measurements and the rationale for repeating the measurements. 

d.    When was the FFQ completed relative to the blood pressure measurements?

e.    Add to the methods the criteria for the sub-categories for each parameter in table 2. For example, state how the categories were determined for income, alcoholic beverage, physical activity, waist circumference, HWR rating, % BF rating and blood pressure.

 

2.    Clarify details for the statistical analysis and be consistent across the text, figures and tables. Section 2.5 states “The relationships among the degree of food processing, anthropometric measures and bloodpressure were analyzed using the partial correlation, with adjustment for age, medications andpathologies, and simple linear regression.” Specify which specific statistical tests were used for each endpoint. For example, this section states partial linear regression while the abstract states “spearman,partial correlations and simple linear regression” and the legend for table 2 states “Pearson's chi-square test”. 

3.    Table 2. 

a.    It is unclear what the P-values in this table represent. Clarify which specific statistical tests were conducted for each endpoint. Also, the parameters at the bottom of the table have significant p-values despite that it is unclear if there were differences between categories of obesity for each parameter. 

b.    The data from tables 2 and 3 could be combined into table 2 since there are no references in the results section to table 3. 

4.    Text in sections 3.1 contradict results in table 2. For example: Lines 152-154. There are statistically significant P-values for waist circumference, waist-hip ratio, % body fat rating and blood pressure. Yet the text states “Regarding clinical conditions no differences were found among the groups, withexception of the blood pressure which the group with obesity presented 13% of the women withPrehypertension or Hypertension.” “the groups were similar in terms of age” – this should reference table 3 not table 2.

5.    Lines 184-187. Report the association between body weight versus blood pressure to support the statement “suggesting an association between the excess of weight and the levels of blood pressure(Table 2).”

6.    Expand the discussion to address generalizability of the results for prevalence of hypertension among women and women with obesity: The discussion (lines 271-294) focuses on the relationship between weight and blood pressure but does not address how the demographics of this cross-sectional study compare to other studies that include women from Brazil. Thus, it is unclear how the results from the present study can be reconciled with the literature or if the results from this study align with prior studies among women in Brazil based on the low prevalence of hypertension among the sample of 85 women and also among the women who had obesity. 

7.    Discussion (line 295) “In this study, women with obesity presented a lower consumption of MPF,suggesting that this pattern of consumption could be related with obesity.” The results from Figure 1 contradict this statement since there was no differences between in MPF between women who were normal weight and had obesity. 

8.    Discussion (line 309) “An association was found between UPF and DBP, however when we adjusteddata for confounding factors this association seems to disappear. The reason for these results may beexplained by the age increasing and the presence of comorbidities”. This statement would be supported by reporting the association between age and degree of food processing in this study. 

Comments on the Quality of English Language

1. Editing needed for clarity. For example, line 150, 154: Re-phrase statements to clarify the meaning of . “The alcoholic beverages were related for 38.5% of the women.” “Most part of the women had noclinical condition previous diagnosis (58.8%).” Line 157: “…18.8% were in use of medication forcardiometabolic diseases (table 2).”

Author Response

Please see the attached file

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

1.      The text “Nevertheless, women with obesity presented lower consumption of MPF when compared to overweight women, suggesting that a lower intake of MPF could be associated with weight gain.” and Figure 2 are contradictory. Since there was no difference in MPF between normal weight versus overweight or normal weight versus overweight, the data do not support the statement that MPF contributes to weight gain. Moreover, these data do not suggest  “that a lower intake of MPF could be associated with weight gain.” In line 331. 

2.     The text “Nevertheless, women with obesity presented lower consumption of MPF when compared to overweight women, suggesting that a lower intake of MPF could be associated with weight gain.” and Figure 2 are contradictory. Since there was no difference in MPF between normal weight versus overweight or normal weight versus overweight, the data do not support the statement that MPF contributes to weight gain. Moreover, these data do not suggest  “that a lower intake of MPF could be associated with weight gain.” In line 331.

3.     The edits to section 2.2 and 2.3 clarified that physical activity, smoking and consumption of alcoholic beverage were investigated and self-declared by participants” Yet, in table 2 there is a binary response for alcoholic beverages and physical activity (yes or no). Please clarify the threshold values for yes and no similar to the sentence for BMI. For example, it is unclear if no equals zero alcoholic beverages in an unknown time period and physical activity equals 0 minutes of moderate to vigorous physical activity per week. This information could be added the table legend.

 

Minor: 

1.      Table 2 states “HWR rating” – correct this to WHR rating.

2.      Editing needed. For example, line 198-201 “It is important to highlight that although the group with obesity presented higher values for the BF, all groups  presented this parameter above what is considered normal, indicating excess body fat…”

Comments on the Quality of English Language

Editing needed throughout the manuscript. 

Author Response

please check the attached fille 

Author Response File: Author Response.pdf