The Relationship between Dietary Intake and Adiposity in South African Female Adolescents: A Systematic Review
Round 1
Reviewer 1 Report
1. Line 105 - should "specific calories" refer to "specific dietary energy sources" or similar? Please update for clarity and sense.
2. Table 1 - please check Table 1 - content in the first rows appears to be descriptive Results. This summary table is not labelled and presumably should appear in the Results section rather than the Methods? Content of cells appears to be truncated/incomplete (e.g. Results for reference 13. Please check and update this Table for accuracy and sense.
3. Methods/Results - authors have not followed many elements of international standards for reporting their systematic review. Please consider PRISMA guidelines in developing this manuscript further (http://www.prisma-statement.org/).
4. Results - you should present a PRISMA diagram or similar to summarise your literature search.
5. Results - your Results section does not appear to contain text that summarises the main findings of your systematic review.
6. Discussion - from the available content in Table 1, it appears that authors must take much more careful consideration in interpretation of the meaning of their findings. (Weak?) Positive correlations of dietary factors with age-adjusted BMI do not necessarily relate to increased risk of overweight or obesity. There is no context considered as to the populations in question here, with some studies seemingly focusing on populations where undernutrition is a challenge.
7. Discussion (and Results) - authors have not considered the approaches to initial dietary data collection here. This is important when considering the totality of available evidence and challenges that there might be for future studies.
8. Discussion - the authors have not highlighted limits in scientific data or how these findings could be used to develop approaches to manage body weight from a public health perspective in the target population. Again, context of the included studies is important here. At the moment, the Discussion is brief, generic and does not centre on the findings of the current systematic review.
Author Response
Response to Reviewer 1 Comments
Point 1: Line 105 - should "specific calories" refer to "specific dietary energy sources" or similar? Please update for clarity and sense..
Response 1:
The definition has been edited.
Nutrient intake or dietary intake or dietary patterns are used to refer to the combinations, amounts, the variety, and frequency of different foods, beverage, and nutrients habitually consumed by an individual [12].
Point 2: Table 1 - please check Table 1 - content in the first rows appears to be descriptive Results..
Response 2:
The first row has been edited.
The text in the row has been removed and placed back where it should appear, under 2.4. The text might have shifted to the table when the submitted version was transferred to the MDPI manuscript version.
Point 3: This summary table is not labelled and presumably should appear in the Results section rather than the Methods?
Response 3
The title has been fixed and placed above the table where it should appear.
The table has been removed from the “Methods and material” section and moved to the “Results” section.
Table 1. South African studies investigating the association between dietary intake and adiposity in female adolescents
Point 4: Content of cells appears to be truncated/incomplete (e.g. Results for reference 13. Please check and update this Table for accuracy and sense.
Response 4
The results for reference 13 (changed to reference 14) have be updated.
A positive and significant association was observed between BMI-for-age and animal driven nutrients, characterized by high positive loadings of nutrients from animal derived sources.
No significant associations were observed between BMI-for-age Z scores and vitamins, fibre, vegetable oil nutrients, and plant derived nutrients.
Point 5: Methods/Results - authors have not followed many elements of international standards for reporting their systematic review. Please consider PRISMA guidelines in developing this manuscript further (http://www.prisma-statement.org/)..
Response 5
- Results
A total of 56 studies were retrieved through the database searches and they reported using a PRISMA flow chart (Figure 1) [14]. The study titles and abstracts were screened, and after removing duplicates and studies deemed not eligible from the records 7 of the studies were deemed eligible for inclusion.
|
|||||||||||
Figure 1: Study selection Flow diagram
Table 1 presents the seven studies that met the inclusion criteria. These studies were published between 2004 and 2021 [14, 15, 16, 17, 18, 19, 20]. The number of female adolescent participants varied across the studies, ranging from 156 to 3,490. Most of the studies focused on participants in the early adolescence stage, typically aged 11 to 15 years [14, 15, 16, 17, 18, 19, 20]. However, two studies included participants in the middle and late stages of adolescence, aged 16 to 21 years [15, 16, 18].
All the seven studies [14, 15, 16, 17, 18, 19, 20] used body mass index (BMI)-for-age as a proxy or indicator for overweight or obesity among adolescents. In addition to BMI-for-age, the two studies conducted by Kruger and colleagues [18, 20] also used triceps skinfold thickness (TSF) and sum of skinfold thicknesses (SST) to further determine overweight/obesity in adolescents.
Regarding dietary intakes, four studies covered this aspect. One study employed a quantitative food frequency questionnaire (QFFQ) [15], while the remaining three studies utilized a structured 24-hour recall questionnaire [19, 20, 21]. Pisa and colleagues explored the relationship between dietary patterns, sociodemographic variables, lifestyle factors, and body mass index among rural black South African adolescents [15]. Kruger conducted two studies—one investigating the causes of overweight and obesity among schoolchildren in the North West Province of South Africa, and the other examining the differences in body composition between stunted and non-stunted adolescent girls [21]. Napier and Oldewage-Theron investigated the dietary intake and nutritional status of adolescent girls and young women in Durban, KwaZulu Natal (KZN), South Africa [20].
Sedibe et al. and Feeley et al. employed interviewer-assisted questionnaires to gather information on food choices, eating practices, and dietary habits of the adolescents [16, 17]. Sedibe's study focused on exploring the differences and similarities in dietary habits and eating practices between younger and older, rural and urban South African adolescents [16]. Feeley examined the relationship between dietary habits, changes in socioeconomic status, and body composition among a cohort of adolescent South Africans [17]. Debeila, Modjadji, and Madiba utilized a food frequency questionnaire (FFQ) to investigate the prevalence of overweight/obesity and associated factors among adolescents in rural high schools [18].
Point 6: Results - you should present a PRISMA diagram or similar to summarise your literature search
Response 6
Figure 1: Study selection Flow diagram
Point 7: Results - your Results section does not appear to contain text that summarises the main findings of your systematic review.
Response 7
Results
A total of 56 studies were retrieved through the database searches and they reported using a PRISMA flow chart (Figure 1) [14]. The study titles and abstracts were screened, and after removing duplicates and studies deemed not eligible from the records 7 of the studies were deemed eligible for inclusion.
|
|||
Figure 1: Study selection Flow diagram
Table 1 presents the seven studies that met the inclusion criteria. These studies were published between 2004 and 2021 [14, 15, 16, 17, 18, 19, 20]. The number of female adolescent participants varied across the studies, ranging from 156 to 3,490. Most of the studies focused on participants in the early adolescence stage, typically aged 11 to 15 years [14, 15, 16, 17, 18, 19, 20]. However, two studies included participants in the middle and late stages of adolescence, aged 16 to 21 years [15, 16, 18].
All the seven studies [14, 15, 16, 17, 18, 19, 20] used body mass index (BMI)-for-age as a proxy or indicator for overweight or obesity among adolescents. In addition to BMI-for-age, the two studies conducted by Kruger and colleagues [18, 20] also used triceps skinfold thickness (TSF) and sum of skinfold thicknesses (SST) to further determine overweight/obesity in adolescents.
Regarding dietary intakes, four studies covered this aspect. One study employed a quantitative food frequency questionnaire (QFFQ) [15], while the remaining three studies utilized a structured 24-hour recall questionnaire [19, 20, 21]. Pisa and colleagues explored the relationship between dietary patterns, sociodemographic variables, lifestyle factors, and body mass index among rural black South African adolescents [15]. Kruger conducted two studies—one investigating the causes of overweight and obesity among schoolchildren in the North West Province of South Africa, and the other examining the differences in body composition between stunted and non-stunted adolescent girls [21]. Napier and Oldewage-Theron investigated the dietary intake and nutritional status of adolescent girls and young women in Durban, KwaZulu Natal (KZN), South Africa [20].
Sedibe et al. and Feeley et al. employed interviewer-assisted questionnaires to gather information on food choices, eating practices, and dietary habits of the adolescents [16, 17]. Sedibe's study focused on exploring the differences and similarities in dietary habits and eating practices between younger and older, rural and urban South African adolescents [16]. Feeley examined the relationship between dietary habits, changes in socioeconomic status, and body composition among a cohort of adolescent South Africans [17]. Debeila, Modjadji, and Madiba utilized a food frequency questionnaire (FFQ) to investigate the prevalence of overweight/obesity and associated factors among adolescents in rural high schools [18].
Point 8: Discussion - from the available content in Table 1, it appears that authors must take much more careful consideration in interpretation of the meaning of their findings. (Weak?) Positive correlations of dietary factors with age-adjusted BMI do not necessarily relate to increased risk of overweight or obesity. There is no context considered as to the populations in question here, with some studies seemingly focusing on populations where undernutrition is a challenge.
Response 8
While the studies included in the review make an important contribution in determining the relationship between diet and adiposity in female South African adolescents, evidence from studies that have a more wholistic approach that have used a variety of study designs, and have used different tools may be required to establish if a definitive relationship exists [53]. In addition, a nationwide research is essential for providing a national perspective, that considers the population differences that exist in a country like South Africa in terms of socio-economic status.
Because there is no single dietary assessment method that is able to measure intake perfectly [54], the FFQ and the 24-hour recall methods were used to assess usual diet in the reported studies, these are known to provide detailed food intake data. However, because of self-reporting and measurement inaccuracies, these food intake assessment approaches have inherent limitations that cannot be ignored [55]. Biochemical markers used in combination with a number of dietary intake assessment methods could be a more effective measure as they are a more accurate measures than using one dietary assessment [54, 55].
In the reported research studies, weight and height measurements were used to calculate the BMI (the most used indicator of obesity) and to classify learners. It is important to note that BMI disregards other contributing factors such as sex, ethnicity, muscle mass and body fat distribution [56]. Our review only focused on dietary practices and not physical inactivity which is one of the main factors that contribute to the incidence of obesity.
Point 9: Discussion (and Results) - authors have not considered the approaches to initial dietary data collection here. This is important when considering the totality of available evidence and challenges that there might be for future studies.
Response 9
While the studies included in the review make an important contribution in determining the relationship between diet and adiposity in female South African adolescents, evidence from studies that have a more wholistic approach that have used a variety of study designs, and have used different tools may be required to establish if a definitive relationship exists [53]. In addition, a nationwide research is essential for providing a national perspective, that considers the population differences that exist in a country like South Africa in terms of socio-economic status.
Because there is no single dietary assessment method that is able to measure intake perfectly [54], the FFQ and the 24-hour recall methods were used to assess usual diet in the reported studies, these are known to provide detailed food intake data. However, because of self-reporting and measurement inaccuracies, these food intake assessment approaches have inherent limitations that cannot be ignored [55]. Biochemical markers used in combination with a number of dietary intake assessment methods could be a more effective measure as they are a more accurate measures than using one dietary assessment [54, 55].
In the reported research studies, weight and height measurements were used to calculate the BMI (the most used indicator of obesity) and to classify learners. It is important to note that BMI disregards other contributing factors such as sex, ethnicity, muscle mass and body fat distribution [56]. Our review only focused on dietary practices and not physical inactivity which is one of the main factors that contribute to the incidence of obesity.
Point 10: 8. Discussion - the authors have not highlighted limits in scientific data or how these findings could be used to develop approaches to manage body weight from a public health perspective in the target population. Again, context of the included studies is important here. At the moment, the Discussion is brief, generic and does not centre on the findings of the current systematic review.
Response 9
While the studies included in the review make an important contribution in determining the relationship between diet and adiposity in female South African adolescents, evidence from studies that have a more wholistic approach that have used a variety of study designs, and have used different tools may be required to establish if a definitive relationship exists [53]. In addition, a nationwide research is essential for providing a national perspective, that considers the population differences that exist in a country like South Africa in terms of socio-economic status.
Because there is no single dietary assessment method that is able to measure intake perfectly [54], the FFQ and the 24-hour recall methods were used to assess usual diet in the reported studies, these are known to provide detailed food intake data. However, because of self-reporting and measurement inaccuracies, these food intake assessment approaches have inherent limitations that cannot be ignored [55]. Biochemical markers used in combination with a number of dietary intake assessment methods could be a more effective measure as they are a more accurate measures than using one dietary assessment [54, 55].
In the reported research studies, weight and height measurements were used to calculate the BMI (the most used indicator of obesity) and to classify learners. It is important to note that BMI disregards other contributing factors such as sex, ethnicity, muscle mass and body fat distribution [56]. Our review only focused on dietary practices and not physical inactivity which is one of the main factors that contribute to the incidence of obesity.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The manuscript is a concise systematic review of studies examining the relationship between dietary and sociodemographic factors and the occurrence of excessive body weight in girls from the South African region. Comparing and systematizing the results of these observations could help to develop effective nutritional education preventing the occurrence of overweight and obesity in this region.
However, some details are worth elaborating. For example Body Mass Index and Waist Circumference and its indices are assessed in children and adolescents in relation to the appropriate growth chart, not reference values for adults.
Also, in subchapter 2.4. it is worth to include a scheme for identifying and selecting articles.
The table on pages 4-8 has no title and the first column has unnecessary text at the beginning. In addition, information about the presented research is repeated in many places. It needs to be improved and unified. This table should be part of the "Results" section, not the "Materials and methods" chapter.
Wouldn't it be worth discussing the methodology used in the presented studies and providing conclusions for further research in this area? In the discussion of the results, the authors could refer to the causes of the described dependencies.
In conclusion, it is worth adding practical tips that may support nutritional education for this group of young people in South Africa.
Author Response
|
12 August 2023
Dear Reviewers
I hope trust this letter finds you well. We are grateful for the comments and corrections you have provided on our manuscript. All comments have been incorporated into the manuscript and we hope that they meet your considerable approval. Below we explain and highlight how all comments and suggestions were addressed in the manuscript. All specific sections and text on which comments and suggestions were made in the manuscript is shown in yellow to show all modifications as requested by the reviewers. In the current letter, we indicate how all comments were addressed.
We thank you again for taking time to review this work.
Regards
Dr. N. Vilakazi
Response to Reviewer 1 Comments
Point 1: However, some details are worth elaborating. For example Body Mass Index and Waist Circumference and its indices are assessed in children and adolescents in relation to the appropriate growth chart, not reference values for adults..
Response 1:
The WHO Reference 2007 (5-19 years) BMI-for-age was used.
Obesity is a state of being excessively overweight classified according to WHO Reference 2007 (5-19 years) BMI-for-age where the +2 SD (equivalent to the 97th centile) coincides at 19 years with the adult’s cut-off of BMI=30 [kg/m2] [11].
Point 2: Also, in subchapter 2.4. it is worth to include a scheme for identifying and selecting articles.
Response 2:
2.4. Selection process and data extraction
Based on the defined selection criteria, the reviewing team conducted a thorough screening of titles and abstracts in the databases to identify eligible articles. Subsequently, the full publications of potentially relevant articles were obtained and independently assessed by three researchers. A collaborative group discussion was then held, during which each reviewer presented their findings for each article, and any uncertainties or discrepancies were resolved through consensus.
Once the articles were finalized for inclusion, data extraction and validity assessments were carried out independently by the same three reviewers. Any discrepancies that arose were addressed through interactive discussions and mutual agreement. The following key data points were extracted to address the main objective of the review: author(s), publication year, sample size (pertaining to the relevant outcome), description of the participants or population, study objective, outcomes measured, observed trends, statistical significance, factors associated with dietary and nutrient practices, adiposity outcomes, and factors that were adjusted in the analyses.
2.5. Data analysis
A descriptive narrative synthesis of the data was performed to produce results according to the outcome of interest of the studies that met the inclusion criteria.
Point 3: The table on pages 4-8 has no title. It needs to be improved and unified
Response 3
The table title has been updated.
“Table 1. South African studies investigating the association between dietary intake and adiposity in female adolescents”.
Point 4: and the first column has unnecessary text at the beginning
Response 4
The unnecessary text has been removed and placed back where it should appear, under 2.4. The text might have shifted to the table when the submitted version was transferred to the MDPI manuscript version..
Point 5: In addition, information about the presented research is repeated in many places.
Response 5
The authors would like to have a bit more explanation for further clarity regarding this comment.
Point 6: This table should be part of the "Results" section, not the "Materials and methods" chapter..
Response 6
The table has been removed from the “Material and methods” and moved to the “Results” section.
Table 1. South African studies investigating the association between dietary intake and adiposity in female adolescents.
Point 7: Wouldn't it be worth discussing the methodology used in the presented studies and providing conclusions for further research in this area? In the discussion of the results, the authors could refer to the causes of the described dependencies..
Response 7
While the studies included in the review make an important contribution in determining the relationship between diet and adiposity in female South African adolescents, evidence from studies that have a more wholistic approach that have used a variety of study designs, and have used different tools may be required to establish if a definitive relationship exists [53]. In addition, a nationwide research is essential for providing a national perspective, that considers the population differences that exist in a country like South Africa in terms of socio-economic status.
Because there is no single dietary assessment method that is able to measure intake perfectly [54], the FFQ and the 24-hour recall methods were used to assess usual diet in the reported studies, these are known to provide detailed food intake data. However, because of self-reporting and measurement inaccuracies, these food intake assessment approaches have inherent limitations that cannot be ignored [55]. Biochemical markers used in combination with a number of dietary intake assessment methods could be a more effective measure as they are a more accurate measures than using one dietary assessment [54, 55].
In the reported research studies, weight and height measurements were used to calculate the BMI (the most used indicator of obesity) and to classify learners. It is important to note that BMI disregards other contributing factors such as sex, ethnicity, muscle mass and body fat distribution [56]. Our review only focused on dietary practices and not physical inactivity which is one of the main factors that contribute to the incidence of obesity.
Point 8: 8. In conclusion, it is worth adding practical tips that may support nutritional education for this group of young people in South Africa.
Response 8
The outcomes from the review add to the concerns regarding the high levels of obesity which is a major burden to the South African health system [57]. To address these concerns, population level interventions that are directed at reducing overweight and obesity rates in adolescents that target the school food environment are recommended [58]. The school environment is the ideal setting for preventative interventions for reducing adolescent obesity because people in the target ages (10 to 19 years) spend a significant amount of their time in school, where they consume up to 30% of their daily calories [59]. To combat hunger among children of school going age in the country the Department of Basic Education has been providing one nutritious meal daily at school through the school feeding program, the National School Nutrition Programme (NSNP) since 1994 [60]. Interventions targeted at obesity could be added to run alongside the school feeding program. Addressing obesity and overweight through the schools has the ability to provide interventions that are specifically targeted to learners from different socio-economic groupings. Interventions that use the school environment also have the added advantage of different the stakeholders such as teachers, food vendors, the school governing body, and parents who can be used directly or indirectly as agents of change [61].
Comprehensive family, school and community-based intervention programs, including education on diet intake and physical activity targeting adolescents should be strengthened. Again, evidence based mobile Health interventions to address overweight and obesity among adolescents can be beneficial [62].
Round 2
Reviewer 1 Report
While authors have made updates to the manuscript, there is no narrative response to reviewer comments to explain the changes made. Many of these do not seem to directly address the comment in-hand.
Specifically, I note that PRISMA guidelines exist related to all aspects of systematic review planning and presentation http://www.prisma-statement.org/documents/PRISMA_2020_checklist.pdf.
Author Response
Please see attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Thank you very much for taking note of the suggestion. In this form, the manuscript can be published.
Author Response
Please see attachment
Author Response File: Author Response.pdf
