The FINDING-Food Intervention: A Mixed-Methods Feasibility Study Addressing Food Insecurity

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Really interesting study. 

I appreciate you might be limited by word count but if possible it would be useful to indicate in the abstract some additional details about data collection procedures, eg the sample size of the quant and qual stages

Appropriate literature review and clear objectives. As one of your objectives focuses on SNAP it might be useful to include some more detail on it in the literature review, for example reiterating the sentence from the abstract "Public and non-profit services in the U.S., such as the federally supported Supplemental Nutrition 12 Assistance Program (SNAP) and community food banks, provide food-related assistance to individuals who are at a high risk of experiencing food insecurity." And then following up with a line or two on SNAP and the registration process (as it is registration the obj considers).

page 4 lines 154-156- this detail is repeated from earlier in the paragraph, could avoid relisting the criteria by instead noting that a screener survey was used to screen according to aforementioned eligibility criteria.

Clear and detailed overview of methods.

2.6.2. were all questions asked to all respondents or was branching so that only those who answered affirmatively to food insecurity in the previous blocks of the survey progressed to answer the next block (as per the USDA classification methodology). Either way also note how food insecurity was defined - how many Q's did they need to answer affirmatively to, or if the USDA classifications were used include an explanation of these e.g. if respondents were classified as mild, moderate, severe. (Note: I see now that I have read the results how you have ranked/measured FS - perhaps include detail on this as it may be confusing for some, I am assuming from the tables/figure that the higher the number is the more food insecure they are (i.e. the more items they have answered affirmatively to), but when this is labelled as 'food security' it may be confusing as one could also assume the reverse that the higher the score the more food secure they are

The sentence noting that aim 3 was achieved may be misleading as on reading aim 3/hypothesis 3 it appears that a decrease in food insecurity in the intervention group was hypothesised to be related to aim 1 and 2 being achieved, i.e. that food insecurity would decrease related to increased uptake of food pantries and SNAP registration. There may be something here to consider around wording. 

 

This is not necessarily a comment that needs addressed in the manuscript but it was interesting to note in the table that a higher % in the control group had visited a food pantry before and had slightly higher FS scores.. it may not be worth including but could consider potentially referring to this / noting how the control and intervention groups were allocated, was it random or did you try to allocate individuals so that baseline of both groups would be roughly similar

 

lines 478-480 - although technically this is true, as per my above point I am not sure if you can claim that this study makes this contribution, as although the intervention group did have a decrease in food insecurity, considering the results of aim 1 and 2 this probably cannot be associated with the nudging and is instead assumed coincidence?

 

lines 526-528 , as above I am not sure this claim can be made, I understand I do not have all the details but from the data presented it seems the decrease in FI in the intervention group may be coincidence rather than related to the nudging? I am basing this on the text message examples noted earlier in the methods which were based on promoting eg the food pantry, rather than necessarily containing any information that would contribute to helping reduce food insecurity?

All in all a very interesting study, I enjoyed reading it. It makes a good contribution to the literature, and as you noted it would be interesting to see how results may differ with a larger sample (I would imagine with a larger sample you might see more significant results and that hypotheses/aims are more likely to be achieved). It would also be interesting to see more qual responses in a larger study re how they found the text messages.

Author Response

Really interesting study. 

I appreciate you might be limited by word count but if possible it would be useful to indicate in the abstract some additional details about data collection procedures, eg the sample size of the quant and qual stages

Response: The sample size for both the quant and qual portions of the study have been included in the abstract.

Appropriate literature review and clear objectives. As one of your objectives focuses on SNAP it might be useful to include some more detail on it in the literature review, for example reiterating the sentence from the abstract "Public and non-profit services in the U.S., such as the federally supported Supplemental Nutrition 12 Assistance Program (SNAP) and community food banks, provide food-related assistance to individuals who are at a high risk of experiencing food insecurity." And then following up with a line or two on SNAP and the registration process (as it is registration the obj considers).

Response: Additions to the introduction have been included that highlight how SNAP has been shown to alleviate food insecurity.

page 4 lines 154-156- this detail is repeated from earlier in the paragraph, could avoid relisting the criteria by instead noting that a screener survey was used to screen according to aforementioned eligibility criteria.

Response: Thank you for this feedback. We’ve mostly retained the original sentence since it also mentions measures included in the screener that did not determine eligibility (i.e., past food pantry utilization).

Clear and detailed overview of methods.

2.6.2. were all questions asked to all respondents or was branching so that only those who answered affirmatively to food insecurity in the previous blocks of the survey progressed to answer the next block (as per the USDA classification methodology). Either way also note how food insecurity was defined - how many Q's did they need to answer affirmatively to, or if the USDA classifications were used include an explanation of these e.g. if respondents were classified as mild, moderate, severe. (Note: I see now that I have read the results how you have ranked/measured FS - perhaps include detail on this as it may be confusing for some, I am assuming from the tables/figure that the higher the number is the more food insecure they are (i.e. the more items they have answered affirmatively to), but when this is labelled as 'food security' it may be confusing as one could also assume the reverse that the higher the score the more food secure they are

Response: Thank you for this suggestion. We have added a paragraph in the Food Security subsection of the Study Measures section in the Methods that explains how the food security data was used to answer the study questions.

The sentence noting that aim 3 was achieved may be misleading as on reading aim 3/hypothesis 3 it appears that a decrease in food insecurity in the intervention group was hypothesised to be related to aim 1 and 2 being achieved, i.e. that food insecurity would decrease related to increased uptake of food pantries and SNAP registration. There may be something here to consider around wording. 

 Response: We appreciate this feedback. Our hope was that Aim 3 would be achieved alongside Aims 1 and 2, but our description of and hypothesis for Aim 3 did not require that Aims 1 and 2 be achieved in order for Aim 3 to be accomplished. Aim 3 simply mentioned the improvement of food security, which was detected.

This is not necessarily a comment that needs addressed in the manuscript but it was interesting to note in the table that a higher % in the control group had visited a food pantry before and had slightly higher FS scores.. it may not be worth including but could consider potentially referring to this / noting how the control and intervention groups were allocated, was it random or did you try to allocate individuals so that baseline of both groups would be roughly similar

 Response: It has been mentioned throughout the manuscript that participants were randomly assigned to study groups.

lines 478-480 - although technically this is true, as per my above point I am not sure if you can claim that this study makes this contribution, as although the intervention group did have a decrease in food insecurity, considering the results of aim 1 and 2 this probably cannot be associated with the nudging and is instead assumed coincidence?

 Response: Given the random assignment of participants to study groups, we cannot credit coincidence as the reason for the change in food security status since, in theory, the process of random assignment results in confounding variables being evenly distributed across study groups.

lines 526-528 , as above I am not sure this claim can be made, I understand I do not have all the details but from the data presented it seems the decrease in FI in the intervention group may be coincidence rather than related to the nudging? I am basing this on the text message examples noted earlier in the methods which were based on promoting eg the food pantry, rather than necessarily containing any information that would contribute to helping reduce food insecurity?

Response: Thank you for these important considerations. Please refer to my previous comments concerning random assignment of participants to study groups. In addition, it is plausible that, in the absence of significant changes in food pantry utilization and SNAP registration, unmeasured variables (i.e., knowledge of food assistance resources) were impacted by the informational nudges and contributed to the significant food security improvements.

All in all a very interesting study, I enjoyed reading it. It makes a good contribution to the literature, and as you noted it would be interesting to see how results may differ with a larger sample (I would imagine with a larger sample you might see more significant results and that hypotheses/aims are more likely to be achieved). It would also be interesting to see more qual responses in a larger study re how they found the text messages.

Reviewer 2 Report

Comments and Suggestions for Authors

Substantive comments/suggestions:

1. I propose to revise the "ultimate goal" of the research indicated in lines 521–522, because it cannot be treated as a generalization of the (specific) goals indicated in the introduction (lines 87–90), and it does not correspond to the main goal formulated in the introduction (lines 93– 95). Moreover, it is neither a cognitive nor an application goal, nor does it refer to the development of research method, so it cannot be the goal of a scientific study.

2. In the part where the results were discussed, it was indicated that the subsequent goals formulated in the introduction were not (lines 389–390, 400) or were (lines 415–416) achieved. I propose to revise this approach and refer to the hypotheses put forward in the introduction. Determining that a given hypothesis has/has not been confirmed will mean achieving the appropriate cognitive goal indicated in the introduction (lines 87–90).

3. I propose to revise the titles of the subsections in part 2.6 (especially 2.6.1. and 2.6.2) to make them more relevant to the content contained therein. This part actually describes the content of the questionnaire.

4. I propose that you consider including section 5 in the summary section.

Editorial comments/suggestions:

1. I suggest removing the dot at the end of the article title.

2. I suggest not using abbreviations as keywords.

3. I suggest limiting the list of keywords only to those that, when read in isolation from the context, explain the topic of the article.

4. Errors in the formatting of the main section titles should be removed (some of them are written in capital letters).

5. I suggest using the notation with the symbol "%" instead of the word "percent".

6. I propose to improve the formatting of objects (figures and tables), in particular to standardize the type and size of the font inside the objects.

7. It is not advisable to place objects (figures, tables) at the end of a section (chapter, subchapter).

8. With reference to section 3.2. I propose to present in a table the numbers and percentages of affirmative/negative answers given to individual questions. This would be a more reader-friendly approach.

9. Bibliographic descriptions should be adapted to the editorial requirements of the journal.

Author Response

1. I propose to revise the "ultimate goal" of the research indicated in lines 521–522, because it cannot be treated as a generalization of the (specific) goals indicated in the introduction (lines 87–90), and it does not correspond to the main goal formulated in the introduction (lines 93– 95). Moreover, it is neither a cognitive nor an application goal, nor does it refer to the development of research method, so it cannot be the goal of a scientific study.

Response: We respectfully disagree that our ‘Ultimate Goal’, which was stated as “Promoting food security among individuals who were food insecure” could not be treated as a generalization of the goals in the introduction. Our three study aims were designed with the intention of promoting food security among individuals who were food insecure, which largely informed the development of our research method.

2. In the part where the results were discussed, it was indicated that the subsequent goals formulated in the introduction were not (lines 389–390, 400) or were (lines 415–416) achieved. I propose to revise this approach and refer to the hypotheses put forward in the introduction. Determining that a given hypothesis has/has not been confirmed will mean achieving the appropriate cognitive goal indicated in the introduction (lines 87–90).

Response: Thank you for this guidance. We have modified these sentences in the results to better connect them with the hypotheses mentioned in the introduction.

3. I propose to revise the titles of the subsections in part 2.6 (especially 2.6.1. and 2.6.2) to make them more relevant to the content contained therein. This part actually describes the content of the questionnaire.

Response: We have retained the original subsection titles, as 2.6.1. comprises all of the participant characteristics, and we’re uncertain how to otherwise describe this collection of variables. Also, subsection title 2.6.2. has been retained, as it pertains to food security.

4. I propose that you consider including section 5 in the summary section.

Response: Thank you for this guidance.

Editorial comments/suggestions:

1. I suggest removing the dot at the end of the article title.

Response: We have removed the period at the end of the title.

2. I suggest not using abbreviations as keywords.

Response: We’ve retained SNAP as a keyword, as many people in the field of food security research refer to “SNAP” in place of the longer “Supplement Food Assistance Program”.

3. I suggest limiting the list of keywords only to those that, when read in isolation from the context, explain the topic of the article.

Response: We have retained our keywords, and will remove those that editors specifically would like us to remove.

4. Errors in the formatting of the main section titles should be removed (some of them are written in capital letters).

Response: Please let us know how to proceed and we will address these errors.

5. I suggest using the notation with the symbol "%" instead of the word "percent".

Response: We have replaced “percent” with % symbols.

6. I propose to improve the formatting of objects (figures and tables), in particular to standardize the type and size of the font inside the objects.

Response: Thank you for this feedback.

7. It is not advisable to place objects (figures, tables) at the end of a section (chapter, subchapter).

Response: We have placed our tables and figures to follow a logical flow of what is being explained in our methods and results sections.

8. With reference to section 3.2. I propose to present in a table the numbers and percentages of affirmative/negative answers given to individual questions. This would be a more reader-friendly approach.

Response: We have decided against this approach, as answering the hypothesis required the summing of individual responses.

9. Bibliographic descriptions should be adapted to the editorial requirements of the journal.

Response: Please let us know how to proceed and we will address these errors.