Comparison of In-Person and Virtual Implementations of an Obesity Prevention and Culinary Nutrition Education Program for Family Care Providers
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsGoodman et al 2024, present a study using relatively small numbers to investigate the difference between a nutrition education program over a 4 week period (Face to face) compared to a self-paced virtual program. Both presenting and educating on the same topics such as knife skills, reduction of salt intake and preparation of alternative protein based foodstuffs. I approached this paper with scepticism as the variables to be considered in such a study are vast. However, it has been presented in a straight forward, transparent way with an in depth discussion of the limitations that need to be considered when interpreting such data. This paper is well written, clear and transparent and given the current environment of communicable diseases. Moreover, the benefits of this application for people who are rural and remote or regional populations, means this study could have a big impact.
Author Response
Comment 1. Goodman et al 2024, present a study using relatively small numbers to investigate the difference between a nutrition education program over a 4 week period (Face to face) compared to a self-paced virtual program. Both presenting and educating on the same topics such as knife skills, reduction of salt intake and preparation of alternative protein based foodstuffs. I approached this paper with scepticism as the variables to be considered in such a study are vast. However, it has been presented in a straight forward, transparent way with an in depth discussion of the limitations that need to be considered when interpreting such data. This paper is well written, clear and transparent and given the current environment of communicable diseases. Moreover, the benefits of this application for people who are rural and remote or regional populations, means this study could have a big impact.
Response 1. Thank you for your time to review our manuscript.
Reviewer 2 Report
Comments and Suggestions for AuthorsThis study is an exploratory quasi-experimental research comparing confidence in culinary skills and familiarity with food assistance programs among family care providers after implementing the 'Start Strong' program in both in-person and virtual formats. Review results indicate the following corrections are needed.
1. Abstract: It is necessary to provide the age range of study participants.
2. Instruction: Scholarly evidence must be provided to categorize the 'Start Strong' program as an obesity prevention culinary nutrition education program. Additionally, theoretical justification is required to explain the relationship between two dependent variables, namely Confidence in knife skills and Familiarity with food assistance programs, and Obesity Prevention. It should be confirmed how providing cooking to participants after family care providers received the program for four weeks affected the participants' obesity. It must be clarified what correlation exists between knife skills and obesity.
3. Materials and Methods: Eligibility criteria for participant selection and exclusion, appropriate sample size calculation, etc., should be provided for quasi-experimental research. There is a difference in the experimental period of the experimental group (September 2019) and the control group (fall 2021 to winter 2022), which may result in performance bias. Therefore, a theoretical explanation is required. Furthermore, reliability of survey tools should be demonstrated.
4. Results: Homogeneity testing for baseline is required for comparison between the two groups. Within-group program pre- and post-comparisons should be presented. Ultimately, the effect size of the 'Start Strong' program should be presented.
5. Discussion: Future measures should be described for dropouts.
6. Conclusion: While there is no data comparing urban and rural areas among the demographic characteristics, the authors state that the online version is "particularly suitable for rural areas where childhood obesity rates are higher than in urban areas." Clarification on this statement is required.
7. References: These sentences should all be changed to match the journal's style.
Author Response
This study is an exploratory quasi-experimental research comparing confidence in culinary skills and familiarity with food assistance programs among family care providers after implementing the 'Start Strong' program in both in-person and virtual formats. Review results indicate the following corrections are needed.
Comment 1. Abstract: It is necessary to provide the age range of study participants.
Response 1. We added the age of our participants to our abstract, lines 15-17, and made additional edits to remain within the abstract word count:
“The in-person program (n=12, mean age 45 years, September 2019) took place at community locations. The virtual program (n=27, mean age 41 years, Fall 2021–Winter 2022) used online learning and videoconferencing.”
Comment 2. Instruction: Scholarly evidence must be provided to categorize the 'Start Strong' program as an obesity prevention culinary nutrition education program. Additionally, theoretical justification is required to explain the relationship between two dependent variables, namely Confidence in knife skills and Familiarity with food assistance programs, and Obesity Prevention.
Response 2. We added additional details clarifying Start Strong’s designation as an obesity prevention program. Specifically, Start Strong aimed to prevent obesity by improving the early childhood food environment in child care settings and at home. The culinary nutrition education component for child care providers (which covered confidence in cooking skills) supported a nutritious food environment in child care centers. Culinary education is one avenue for intervention for obesity prevention and preliminary evidence shows such interventions can reduce child and adolescent body mass index. Regarding the home food environment, child care providers received education on food assistance programs to share with families in their centers. There is evidence that participation in such programs can support a healthy weight for children from low-income families by increasing access to nutritious foods.
We added these details to lines 45-84, as well as several citations.
“Start Strong was a culinary nutrition education program that aimed to prevent obesity by increasing family care providers’ ability to prepare nutritious foods for children and by improving their knowledge of federal food assistance programs [4]. Culinary skills have a role in obesity prevention through their association with increased intake of fruits, vegetables, whole grains, fiber, and vitamins and minerals supportive of a healthy weight [6, 7]. There is some evidence to show that children with caregiver cooking skill and preparation of nutritious food is related to improved child diet quality and lower risk of obesity, and promising evidence that culinary interventions for adults and children can reduce child and adolescent body mass index [7-9]. Generally, community-based culinary obesity prevention programs include cooking skills such as knife skills, meal planning, healthy substitutions, portion size guidance, and techniques for reducing sugar and salt [10-12]. While there is little research examining the role of individual cooking skills in obesity prevention, knife skills tend to be a consistent feature across programs, as they can facilitate saving time, money, and preparation of nutritious foods, potentially reducing reliance on calorie dense convenience foods [6, 13]. To identify additional program components relevant to this population, focus groups with 19 rural family care providers and a statewide survey of ECE providers assessing unmet training needs informed curriculum development [4]. Providers indicated areas they were willing to change and areas in which they needed more education, including learning cooking techniques to use whole grains, adding flavor without salt, and tips to save cost and time [4, 6].
Start Strong curriculum was designed for family care providers in low-income areas that qualified for Tier 1 reimbursement through the Child and Adult Care Food Program (CACFP) [4]. CACFP is a federal program that reimburses the cost of meals to child care centers; centers qualifying for Tier 1 reimbursement provide care to families whose incomes are within 185% of federal poverty guidelines [15]. Obesity disproportionately affects children and families from low-income households, which may have reduced access to nutritious foods [16]. To address this disparity, families within this income guideline qualify for federal food assistance programs, including the Women, Infants and Children (WIC) program, Supplemental Nutrition Assistance Program (SNAP), and free and reduced priced school meals. Child participation in WIC, SNAP, and free and reduced price meals may improve diet quality and reduce obesity prevalence over time [17-19]. Further, the WIC program supports healthy weight in early childhood through additional growth monitoring and nutrition education [20]. Thus, by training providers how to prepare nutritious foods and share information about food assistance programs with families, Start Strong aimed to prevent obesity by improving the early childhood food environment in child care settings and at home [4]. Finally, the Learning Task Model informed Start Strong curriculum design [4]. The Learning Task Model involves learning new information anchored in prior knowledge, practicing new skills, sharing ideas with other providers, and goal setting [9].”
Nelson SA, Corbin MA, Nickols-Richardson SM. A call for culinary skills education in childhood obesity-prevention interventions: current status and peer influences. Journal of the Academy of Nutrition and Dietetics. 2013;113(8):1031-1036.
Dimple, D., & Ramesh, G. (2023). Cooking and Its Impact on Childhood Obesity: A Systematic Review. J Nutr Educ Behav, 55(9), 677–688. https://doi.org/10.1016/j.jneb.2023.06.004
Tani Y, Isumi A, Doi S, Fujiwara T. Associations of Caregiver Cooking Skills with Child Dietary Behaviors and Weight Status: Results from the A-CHILD Study. Nutrients. 2021 Dec 18;13(12):4549. doi: 10.3390/nu13124549. PMID: 34960100; PMCID: PMC8704868.
Kramer RF, Coutinho AJ, Vaeth E, Christiansen K, Suratkar S, Gittelsohn J. Healthier home food preparation methods and youth and caregiver psychosocial factors are associated with lower BMI in African American youth. J Nutr. 2012 May;142(5):948-54. doi: 10.3945/jn.111.156380. Epub 2012 Mar 28. PMID: 22457390.
Franzen-Castle L, Colby SE, Kattelmann KK, Olfert MD, Mathews DR, Yerxa K, Baker B, Krehbiel M, Lehrke T, Wilson K, Flanagan SM, Ford A, Aguirre T, White AA. Development of the iCook 4-H Curriculum for Youth and Adults: Cooking, Eating, and Playing Together for Childhood Obesity Prevention. J Nutr Educ Behav. 2019 Mar;51(3S):S60-S68. doi: 10.1016/j.jneb.2018.11.006. PMID: 30851862.
Fals, A. M., & Schnell, C. (2019). Innovative, multi-level nutrition education program within a hospital-based childhood obesity prevention and treatment program.
Thang CK, Guerrero AD, Garell CL, Leader JK, Lee E, Ziehl K, Carpenter CL, Boyce S, Slusser W. Impact of a Teaching Kitchen Curriculum for Health Professional Trainees in Nutrition Knowledge, Confidence, and Skills to Advance Obesity Prevention and Management in Clinical Practice. Nutrients. 2023; 15(19):4240. https://doi.org/10.3390/nu15194240
Condrasky MD, Hegler M. How culinary nutrition can save the health of a nation. Journal of extension. 2010;48(2):1-6.
Nanney MS, Larowe TL, Davey C, Frost N, Arcan C, O’Meara J. Obesity Prevention in Early Child Care Settings. Health Education & Behavior. 2017;44(1):23-31. doi:10.1177/1090198116643912
Hamilton WL. Reimbursement tiering in the CACFP: summary report to congress on the family child care homes legislative changes study. Economic Research Service, US Department of Agriculture; 2002.
Traore, Stanislav Seydou, Yacong Bo, Guangning Kou, and Quanjun Lyu. "Socioeconomic inequality in overweight/obesity among US children: NHANES 2001 to 2018." Frontiers in Pediatrics 11 (2023): 1082558.
Daepp, M. I., Gortmaker, S. L., Wang, Y. C., Long, M. W., & Kenney, E. L. (2019). WIC food package changes: trends in childhood obesity prevalence. Pediatrics, 143(5).
Kenney, E. L., Barrett, J. L., Bleich, S. N., Ward, Z. J., Cradock, A. L., & Gortmaker, S. L. (2020). Impact Of The Healthy, Hunger-Free Kids Act On Obesity Trends: Study examines impact of the Healthy, Hunger-Free Kids Act of 2010 on childhood obesity trends. Health Affairs, 39(7), 1122-1129.
Hudak, K. M., & Racine, E. F. (2021). Do additional SNAP benefits matter for child weight?: Evidence from the 2009 benefit increase. Economics & Human Biology, 41, 100966.
Carlson, S., & Neuberger, Z. (2021). WIC works: addressing the nutrition and health needs of low-income families for more than four decades. Center on Budget and Policy Priorities.
Comment 3. It should be confirmed how providing cooking to participants after family care providers received the program for four weeks affected the participants' obesity.
Response 3. Participants in this program were family child care providers and we do not have any child-level data, including child age or weight status. Thus, we were unable to determine the effect of this program on the weight status of children cared for by the family care providers in our study and added this limitation of our study in lines 358-363:
“Finally, since we do not have data on child growth, we were unable to determine whether Start Strong had an affect on the body mass index of children cared for by the family care providers who participated in the training. Future longitudinal research should examine changes in child growth over time following a culinary obesity prevention intervention, as well as identify specific, essential cooking skills (e.g., knife skills) supportive of a healthy weight.”
Comment 4. It must be clarified what correlation exists between knife skills and obesity.
Response 4. Previous literature on culinary obesity prevention programs tends to examine the effect of overall program participation on weight without looking at the effect of individual cooking skills. Thus, although knife skills are generally included across culinary programs, there is insufficient evidence to connect this specific skill to obesity. We provided an overview of the literature in this area in lines 47-59:
“Culinary skills have a role in obesity prevention through their association with increased intake of fruits, vegetables, whole grains, fiber, and vitamins and minerals supportive of a healthy weight [6, 7]. There is some evidence to show that children with caregiver cooking skill and preparation of nutritious food is related to improved child diet quality and lower risk of obesity, and promising evidence that culinary interventions for adults and children can reduce child and adolescent body mass index [7-9]. Generally, community-based culinary obesity prevention programs include cooking skills such as knife skills, meal planning, healthy substitutions, portion size guidance, and techniques for reducing sugar and salt [10-12]. While there is little research examining the role of individual cooking skills in obesity prevention, knife skills tend to be a consistent feature across programs, as they can facilitate saving time, money, and preparation of nutritious foods, potentially reducing reliance on calorie dense convenience foods [6, 13].”
We agree that future research should examine specific cooking skills in relation to obesity risk, and added this recommendation to our limitations section, lines 361-363:
“Future longitudinal research should examine changes in child growth over time following a culinary obesity prevention intervention, as well as identify specific, essential cooking skills (e.g., knife skills) supportive of a healthy weight.”
Comment 5. Materials and Methods: Eligibility criteria for participant selection and exclusion, appropriate sample size calculation, etc., should be provided for quasi-experimental research.
Response 5. Eligible participants were licensed child care providers from Minnesota qualified for the Child and Adult Care Food Program (CACFP). CACFP sponsors received information about Start Strong and sent the information to child care providers. We added additional details regarding the eligibility criteria of participants in this study in lines 117-118 of the Materials and Methods section:
“Eligible participants were licensed family care providers from the state of Minnesota who self-selected into the study after receiving an email about the study.”
Because we recruited a convenience sample and used measures designed for this pilot program evaluation, we did not estimate a sample size to detect a desired effect size prior to program implementation. We added this limitation to lines 350-358.
“These results are from a convenience sample recruited with the assistance of CACFP sponsors and specifically reached child care providers in rural Minnesota. Further, individuals in our sample chose to participate in this voluntary training opportunity and may have had more interest in nutrition than providers who did not choose to participate. Our sampling method limits the generalizability of our findings to other locations and among family care providers who identify differently, and in combination with our study measures, did not allow for a priori calculation of sample size to detect a desired treatment effect prior to program implementation.”
Comment 6. There is a difference in the experimental period of the experimental group (September 2019) and the control group (fall 2021 to winter 2022), which may result in performance bias. Therefore, a theoretical explanation is required.
Response 6. Lines 342-349 of our limitations section notes that were unable to account for the different time points during which data collection occurred:
“The opportunity to compare in-person and virtual implementation of the program arose due changes made following the COVID-19 pandemic. As a result, data for the in-person and virtual implementations of Start Strong were collected at different time points and were taught by different class facilitators. The small sample size further limits our ability to account for potential confounding variables in our analysis. These factors limit the conclusions that can be made regarding differences and similarities between the implementation modalities.”
Comment 7. Furthermore, reliability of survey tools should be demonstrated.
Response 7. We an additional line about the internal consistency of our survey tools, to our results section, lines 199-201.
“Survey items showed good internal consistency pre-program (Chronbach’s alpha ranging from 0.74-0.92) and post-program (Chronbach’s alpha ranging from 0.87-0.95).”
Survey items were designed specifically to evaluate this pilot program and thus have not been formally validated. We discuss that these survey tools may not be suitable to make rigorous comparisons in our limitations section, lines 340-342:
“The surveys used to assess outcomes were not designed to make rigorous comparisons, and future programs may benefit from measuring outcomes using standardized measures.”
Comment 8. Results: Homogeneity testing for baseline is required for comparison between the two groups. Within-group program pre- and post-comparisons should be presented.
Response 8. We tested for baseline differences between groups in demographic characteristics and baseline outcome measures (cooking skill confidence, familiarity with food assistance programs, and likelihood of sharing food assistance programs) using t-tests for continuous variables and Fisher’s exact test for categorical variables. We found no significant differences between groups in demographic characteristics. We found that familiarity with SNAP (p<.01) was the only difference between groups on outcome measures.
We note this testing in the Methods and Materials section, lines 181-185:
“Baseline differences between in-person and virtual groups were compared using independent samples t-tests for continuous variables and Fisher’s exact test for categorical variables. After examining baseline characteristics between the in-person and virtual groups, there were no statistically different baseline demographic characteristics between groups.”
Our findings are described in the results on lines 203-205:
“The only statistically significant difference at baseline was familiarity with SNAP, which was significantly lower among the in-person group relative to the virtual group (p < .01).”
Comment 9: Ultimately, the effect size of the 'Start Strong' program should be presented.
Response 9: We revised Table 3 to include the effect size for each of our within group comparisons. Table 3 begins at line 207 of the manuscript.
|
|
In-Person (N=12) |
|
|||||
|
Confidence with:a |
Pre-program |
Post-program |
Difference |
Effect Size |
P-valueb |
|
|
|
Using a chef knife |
4.08 (.79) |
4.25 (1.22) |
.17 (1.27) |
.13 |
.66 |
|
|
|
Cutting vegetables |
4.50 (.52) |
4.50 (1.17) |
0 (1.04) |
0 |
1 |
|
|
|
Cutting fruit |
4.42 (.52) |
4.33 (1.23) |
.09 (1.08) |
.08 |
.80 |
|
|
|
Preparing whole grains |
3.42 (1.31) |
4.50 (1.17) |
1.08 (1.31) |
.82 |
.02 |
|
|
|
Using beans and low cost protein sources |
3.25 (1.29) |
4.33 (1.15) |
1.08 (1.38) |
.78 |
.02 |
|
|
|
Planning menus |
3.92 (1.00) |
4.25 (1.22) |
.33 (1.37) |
.24 |
.42 |
|
|
|
Using cooking techniques to reduce salt |
3.00 (.74) |
4.33 (1.15) |
1.33 (.89) |
.01 |
<.01 |
|
|
|
Familiarity with food assistance programs:c |
|
||||||
|
WIC1 |
3.17 (.83) |
3.67 (.49) |
.50 (.80) |
.63 |
.05 |
|
|
|
SNAP2 |
1.67 (.65) |
3.33 (.49) |
1.66 (.98) |
1.69 |
<.01 |
|
|
|
Free and reduced priced school meals |
2.83 (1.19) |
3.67 (.49) |
.84 (1.03) |
.82 |
.02 |
|
|
|
Likelihood of sharing information about food assistance programs with families:c |
|
||||||
|
WIC |
4.42 (.67) |
4.83 (.40) |
.41 (.79) |
.52 |
.10 |
|
|
|
SNAP |
3.83 (1.27) |
4.83 (.40) |
1.0 (1.04) |
.96 |
<.01 |
|
|
|
Free and reduced priced school meals |
4.33 (.89) |
4.75 (.45) |
.42 (.79) |
.53 |
.10 |
|
|
|
|
|
|
Virtual (N=27) |
||||
|
Confidence with:a |
Pre-program |
Post-program |
Difference |
Effect Size |
P-valueb |
|
|
|
Using a chef knife |
4.19 (1.04) |
4.78 (.42) |
.59 (1.01) |
.58 |
.01 |
|
|
|
Cutting vegetables |
4.41 (.80) |
4.78 (.42) |
.37 (.88) |
.42 |
.04 |
|
|
|
Cutting fruit |
4.52 (.58) |
4.81 (.39) |
.29 (.67) |
.43 |
.03 |
|
|
|
Preparing whole grains |
4.19 (.74) |
4.52 (.64) |
.33 (1.04) |
.32 |
.11 |
|
|
|
Using beans and low cost protein sources |
3.63 (.97) |
4.22 (.70) |
.59 (1.01) |
.58 |
.01 |
|
|
|
Planning menus |
3.85 (1.06) |
4.52 (.70) |
.67 (1.33) |
.50 |
.02 |
|
|
|
Using cooking techniques to reduce salt |
3.63 (1.00) |
4.59 (.50) |
.96 (1.16) |
.83 |
<.01 |
|
|
|
Familiarity with food assistance programs:c |
|
||||||
|
WIC1 |
3.11 (.93) |
3.63 (.49) |
.52 (.80) |
.65 |
<.01 |
|
|
|
SNAP2 |
2.63 (.84) |
3.41 (.64) |
.78 (.93) |
.84 |
<.01 |
|
|
|
Free and reduced priced school meals |
3.07 (1.04) |
3.63 (.49) |
.56 (.89) |
.63 |
<.01 |
|
|
|
Likelihood of sharing information about food assistance programs with families:c |
|
||||||
|
WIC |
4.26 (1.06) |
4.74 (.45) |
.48 (1.12) |
.43 |
.03 |
|
|
|
SNAP |
3.96 (1.22) |
4.70 (.54) |
.74 (1.35) |
.55 |
.01 |
|
|
|
Free and reduced priced school meals |
4.26 (1.06) |
4.78 (.42) |
.51 (1.09) |
.47 |
.02 |
|
|
aResponse options ranged from 1-5.
bP-values were determined using paired samples t-tests with significance set at p=0.05. Values in bold indicate statistically significant within group changes.
cResponse options ranged from 1-4.
1WIC = Women, Infants, and Children Program
2SNAP = Supplemental Nutrition Assistance Program
Comment 10. Discussion: Future measures should be described for dropouts.
Response 10. We observed a greater loss of participants in our virtual group (12 out of 52 did not log into the online learning platform) compared to our in-person group (1 participant who registered did not complete the program). We addressed the potential technology barrier that may have contributed to the loss of the 12 participants in our virtual group in our discussion section, lines 301-305.
“A portion of virtual participants interested in participating in Start Strong did not complete the program because they did not log into the online learning system (23%). As use of videoconferencing and online learning platforms becomes increasingly popular, additional research is needed to address technology barriers and literacy, which might prevent some populations from accessing information in online formats.”
Comment 11. Conclusion: While there is no data comparing urban and rural areas among the demographic characteristics, the authors state that the online version is "particularly suitable for rural areas where childhood obesity rates are higher than in urban areas." Clarification on this statement is required.
Response 11. We rephrased this sentence to clarify that the online program could be accessible across a wider geographic area, such as rural communities, who are disproportionately affected by obesity and may benefit from program participation.
Lines 368-372: “Notably, by eliminating the need to meet at a central location, the online version is capable of reaching audiences in wider geographic areas, particularly in rural communities, where rates of obesity in children are higher relative to urban areas.”
Comment 12. References: These sentences should all be changed to match the journal's style.
Response 12. We revised references to comply with the journal’s style.
Reviewer 3 Report
Comments and Suggestions for AuthorsThis manuscript studied the effects of their nutrition/cooking education program to prevent obesity designed for rural family care providers in low-income areas. It showed a clear difference between before and after: They were more confident about cooking after the event, particularly in-person participants, than the virtual participants. The results will still be preliminary since the n is not large enough. However, this can be a great model for places with a significant diabetes population.
My primary concern is the absence of IRB approval for this study, which is a fundamental requirement of your university's guidelines. This approval is crucial to ensure the ethical conduct of research.
hrp-103_-_investigator_manual_google_doc.pdf (umn.edu)
In this document, in IRB exemption, it says,
The following are examples of activities that are likely not to be human research:
"Program Evaluation/Quality Assurance Review/Quality Improvement Projects: The activity is limited to program evaluation, quality assurance, or quality improvement activities designed specifically to evaluate, assure, or improve performance within a department, classroom, or hospital setting. There is no intent to alter or control the evaluation for research purpose."
It's important to note that while IRB approval may not be necessary for program improvement activities, it becomes mandatory when the data is intended for research publication. This distinction is crucial for understanding the regulatory requirements for your study.
I did not find any English concerns and would recommend this manuscript as is in Diabetics after the confirmation of the evidence of IRB approval "not required" because it contains helpful information for all communities; in particular, there are low-income areas that fight against obesity.
Author Response
This manuscript studied the effects of their nutrition/cooking education program to prevent obesity designed for rural family care providers in low-income areas. It showed a clear difference between before and after: They were more confident about cooking after the event, particularly in-person participants, than the virtual participants. The results will still be preliminary since the n is not large enough. However, this can be a great model for places with a significant diabetes population.
My primary concern is the absence of IRB approval for this study, which is a fundamental requirement of your university's guidelines. This approval is crucial to ensure the ethical conduct of research.
hrp-103_-_investigator_manual_google_doc.pdf (umn.edu)
In this document, in IRB exemption, it says,
The following are examples of activities that are likely not to be human research:
"Program Evaluation/Quality Assurance Review/Quality Improvement Projects: The activity is limited to program evaluation, quality assurance, or quality improvement activities designed specifically to evaluate, assure, or improve performance within a department, classroom, or hospital setting. There is no intent to alter or control the evaluation for research purpose."
It's important to note that while IRB approval may not be necessary for program improvement activities, it becomes mandatory when the data is intended for research publication. This distinction is crucial for understanding the regulatory requirements for your study.
I did not find any English concerns and would recommend this manuscript as is in Diabetics after the confirmation of the evidence of IRB approval "not required" because it contains helpful information for all communities; in particular, there are low-income areas that fight against obesity.
Response 1. Thank you for your comment and recommendation. Many program evaluation studies are routinely published in peer-reviewed journals, particularly in more applied fields such as public health nutrition in which many interventions are implemented prior to any studies on them being conducted. Program evaluation allows for interventions being implemented in practice to be evaluated. Such evaluations are valuable for many reasons including increasing understanding of whether current interventions are improving health as intended and helping researchers stay up-to-date with programs currently in practice to help reduce the gap between research and practice. We added additional details about our IRB approval in lines 133-136:
“A Determination Form was completed for the University of Minnesota Institutional Review Board (IRB, project identification number STUDY00006219). Because this project evaluated a training program (i.e., program evaluation), the IRB determined it was exempt from full IRB review on 16 April 2019.”
Reviewer 4 Report
Comments and Suggestions for AuthorsThank you for the possibility to read this interesting manuscript. This manuscript is, for the most part, easy to follow and clearly written. I believe this paper makes an important contribution to the field.
I have some suggestions for improving the paper. Please see my comments below.
Title:
The title appears to be quite long but informative though.
Introduction:
Line 46: What means “Tier 1 reimbursement”? Could it be briefly explained? It is not familiar to readers outside the United States.
Methods:
Line 115: On line 115 you use the abbreviation IRB. I suppose it refers to Institutional Review Board (line 114). Maybe you could add the abbreviation after the whole name of the Board on line 114.
In general: Are you able to tell more precisely what was the participation rate to both In-Person and Virtual groups? In the case of In-Person program how many family care providers were invited and how many refused to participate? How many licensed child care providers there are and were they all contacted via email? How representative sample you had in your study? Can you elaborate on this? If not maybe it should be considered in discussion? Is it possible that those who participated already in the beginning were different from those who did not participate? Could they have been more interested in nutrition/food preparing and could their background knowledge about e.g. food assistance programs be better than those who did not participate in program?
Tables:
-Tables are easy to interpret and understand. However, some suggestions to improve them below.
Table 1:
- - In table 1 you have superscript numeral 1 below the table but I can’t find the superscript number 1 in table itself? Maybe it will be clearer if you add superscript to the table after every program mentioned. i.e. CACFP1, SNAP2, WIC3 and explain them in that way in footnote.
- - Also WIC is now written without the word “program” in footnote. Please, add it to it.
Table 2:
- I think one decimal is enough to be used in this table.
Table 3:
- - As each table have to be understood without reference to the manuscript text, please spell it out in table 3 what means WIC and SNAP in footnotes.
- - I was also bit confused as the footnote letter b appeared in table before the letter a. Can it be changed so that they are in “alphabetical” order?
Results:
Line 190: instead of using the verb “significantly improved preparing”, should you use the form “significantly improved the confidence in preparing”. Isn’t that more precise? Check this out throughout the article.
Line 194: to be exact maybe the (p<.05) should be (p≤.05) as in the In-Person group the familiarity with the WIC p-value was 0.05.
Discussion:
Maybe you could in discussion take briefly into consideration why it is important to reduce salt intake already among children. The health effects are not related to obesity but reducing salt intake have other beneficial health effects already in children. Also, you had important result that the confidence to use beans and low cost protein sources improved. I think that is an important result when thinking about the discussion concerning the climate change and the need to change our diet more towards plant-based protein sources.
I hope that my observations will help you to further improve the already good manuscript.
Author Response
Thank you for the possibility to read this interesting manuscript. This manuscript is, for the most part, easy to follow and clearly written. I believe this paper makes an important contribution to the field.
I have some suggestions for improving the paper. Please see my comments below.
Comment 1. Title: The title appears to be quite long but informative though.
Response 1. We agree that the title is long. As indicated we prioritized being descriptive over length. For example, given the scope of the journal we felt it was important to note in the title that the program was an obesity prevention program, which added some length but provided meaning to the title. We feel this is aligned with the journal’s policy about unlimited length of articles to ensure that the article is as informative as possible.
Comment 2. Introduction: Line 46: What means “Tier 1 reimbursement”? Could it be briefly explained? It is not familiar to readers outside the United States.
Response 2. We added additional details describing Tier 1 reimbursement through the Child and Adult Food Care Program in lines 66-70.
“Start Strong curriculum was designed for family care providers in low-income areas that qualified for Tier 1 reimbursement through the Child and Adult Care Food Program (CACFP) [4]. CACFP is a federal program that reimburses the cost of meals to child care centers; centers qualifying for Tier 1 reimbursement provide care to families whose incomes are within 185% of federal poverty guidelines [15].”
Comment 3. Methods: Line 115: On line 115 you use the abbreviation IRB. I suppose it refers to Institutional Review Board (line 114). Maybe you could add the abbreviation after the whole name of the Board on line 114.
Response 3. We added this abbreviation, as well as additional details regarding our IRB approval on lines 133-136:
“A Determination Form was completed for the University of Minnesota Institutional Review Board (IRB, project identification number STUDY00006219). Because this project evaluated a training program (i.e., program evaluation), the IRB determined it was exempt from full IRB review on 16 April 2019.”
Comment 4. In general: Are you able to tell more precisely what was the participation rate to both In-Person and Virtual groups?
Response 4. We added participation rate details to the Materials and Methods section at lines 118-123 and lines 127-133:
“The in-person Start Strong program took place in September 2019 at a community location determined by Start Strong facilitators. Thirteen participants registered for the course and 12 participants completed all parts of the course (92% participation rate). All 12 participants completed the in-person program, provided complete pre- and post-program data, and were included in the analysis.”
“Of the 52 participants who initially expressed interest in the online intervention, 12 individuals did not log into the online learning platform to access the intervention materials and 40 completed the training (77% participation rate). However, of these 40 completers, 13 were excluded from analysis for providing either duplicate or incomplete data. The final analytical sample of virtual participants included 27 individuals who completed the training and provided complete pre- and post-program data.”
Comment 5. In the case of In-Person program how many family care providers were invited and how many refused to participate? How many licensed child care providers there are and were they all contacted via email?
Response 5. This study was an evaluation of an existing program and participants were recruited based on their participation in the Start Strong program. The program was made available to all family care providers eligible for the Child and Adult Care Food Program (CACFP) and advertised through email by CACFP sponsors. Thus, family care providers self-selected into the program and evaluation. There was not a refusal to participate beyond not completing the program or evaluation surveys. The provided incentives were for participating in the program, not completing data collection. Since four CACFP sponsors shared information about the study, we do not have specific information regarding how many family care providers ultimately received information about the program, but there are about ~5,000 CACFP family care providers in the state of Minnesota. Theoretically all these providers would have received an advertisement about the program. We added these details to the Recruitment section of Table 1, which starts on line 137.
“There were an estimated 5,474 licensed family care providers in the state of Minnesota during this study [26]. Although this number of family care providers may have received information about the study at both time points, the number who may have had interest in the in-person implementation of the program was limited by geographic location.”
Comment 6. How representative sample you had in your study? Can you elaborate on this? If not maybe it should be considered in discussion?
Response 6. Our sample was generally representative of family child care providers in the state of Minnesota. We added this detail on our results section, lines 195-199.
“All providers identified as Non-Hispanic White and 95% percent of participants identified as female, which is generally representative of family child care providers throughout the state of Minnesota (nearly all female, average age of 42 years, 94% identify as Non-Hispanic White) [28].”
Comment 7. Is it possible that those who participated already in the beginning were different from those who did not participate? Could they have been more interested in nutrition/food preparing and could their background knowledge about e.g. food assistance programs be better than those who did not participate in program?
Response 7. Yes, as is the case with many voluntary training opportunities for child care providers, those most interested in the topic of nutrition may be more likely to participate. We added this possibility to our limitations section in regarding limiting generalizability, lines 352-354.
“Further, individuals in our sample chose to participate in this voluntary training opportunity and may have had more interest in nutrition than providers who did not choose to participate.”
Comment 8. Tables: Tables are easy to interpret and understand. However, some suggestions to improve them below. Table 1: In table 1 you have superscript numeral 1 below the table but I can’t find the superscript number 1 in table itself? Maybe it will be clearer if you add superscript to the table after every program mentioned. i.e. CACFP1, SNAP2, WIC3 and explain them in that way in footnote. Also WIC is now written without the word “program” in footnote. Please, add it to it.
Response 8. We revised this table to ensure that superscripts within the table align with superscripts in the table footnote, which fully spell out program names.
Comment 9. Table 2: I think one decimal is enough to be used in this table.
Response 9. We agree with your suggestion to use one decimal point. We revised Table 2 and rounded numbers to one decimal point (Table 2 starts at line 206).
Comment 10. Table 3: As each table have to be understood without reference to the manuscript text, please spell it out in table 3 what means WIC and SNAP in footnotes. I was also bit confused as the footnote letter b appeared in table before the letter a. Can it be changed so that they are in “alphabetical” order?
Response 10. We agree with your suggested revisions to the table. We reorganized footnotes to appear in alphabetical order in the table and added footnotes to spell out program names for WIC and SNAP.
Comment 11. Results: Line 190: instead of using the verb “significantly improved preparing”, should you use the form “significantly improved the confidence in preparing”. Isn’t that more precise? Check this out throughout the article.
Response 11. We agree with your suggestion that confidence in preparing whole grains is more precise. We revised our manuscript throughout to ensure our language is consistent with the results of our paper.
Comment 12. Line 194: to be exact maybe the (p<.05) should be (p≤.05) as in the In-Person group the familiarity with the WIC p-value was 0.05.
Response 12. During the revision process, we noticed an error in our reporting of the familiarity with WIC p-value, which is p=.0527. Therefore, we removed indications from our manuscript that this result was statistically significant, including in Table 2 and our results section.
Comment 13. Discussion: Maybe you could in discussion take briefly into consideration why it is important to reduce salt intake already among children. The health effects are not related to obesity but reducing salt intake have other beneficial health effects already in children. Also, you had important result that the confidence to use beans and low cost protein sources improved. I think that is an important result when thinking about the discussion concerning the climate change and the need to change our diet more towards plant-based protein sources.
Response 13. We added additional information in our discussion to address the need for further research about the health effects of improving provider confidence in cooking skills, including using cooking techniques to reduce salt and use of beans and low-cost protein sources, in lines 283-293:
“Thus, further research is needed to examine how culinary interventions can support a healthy food environment in child care settings and their effect on children’s diet quality, taste preferences, and health outcomes through exposing care providers to a variety of culinary skills. For example, improving care provider confidence in using cooking techniques to reduce salt could be impactful as intake of sodium-dense foods in childhood is associated with hypertension that can persist into adulthood, and with increased risk of overweight and obesity [33, 34]. While beyond the scope of this study, the improvements in caregivers confidence preparing beans and low cost protein sources could be particularly relevant given current discussions regarding the environmental impact of food choices and plant-based protein options.”
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsThis manuscript has been appropriately revised according to the reviewer's comments. Thank you for your hard work.
Reviewer 3 Report
Comments and Suggestions for AuthorsNow I can see the number of IRB, and I recommend this article in Obesities.
