Diet Optimization for Sustainability: INDIGOO, an Innovative Multilevel Model Combining Individual and Population Objectives

Round 1

Reviewer 1 Report

This manuscript details the use of a dietary optimisation model considering nutrition, environmental impact, cost, and deviation from current diet. The model is used to optimise a collection of observed diets to achieve a 30% reduction in population GHG emissions while respecting constraints on the other considerations.

The manuscript and the work it describes are excellent in most respects. I have some general comments on particular aspects which should be straightforward to address. The manuscript does suffer from some grammatical errors throughout, but these do not compromise the intended meaning - I would suggest a (further) proof-read by a native speaker.

Comments:

1. Isocaloric diets: my understanding is that the diets were optimised to be isocaloric to the original diets. Is this appropriate, given that there were likely many individuals who were showed large excess energy intakes in the survey data? Would it not have been more appropriate to optimise diets to within an acceptable energy range? Perhaps the authors could clarify this choice in the manuscript. Relates to lines 16, 127-128

2. % change in diets. The calculation of the % change in diets was not clearly explained (for example the "74% vs 69%" stated in the abstract). These % change values between the original and optimised diets feature throughout the paper, but their calculation should be more clearly explained. Is this the sum of the % change in all food items? In which case 69% is not large; however, at first glance, a diet that is 69% different to my current diet makes me think that two thirds of the foods I eat have disappeared. I suggest removing these percentages from the abstract (since there is not the space to explain them) and more clearly explaining their calculation in the methods. Also relates to L182-183, 246, figure 3

3. Classification of socio-occupational status: relates to L88-92, but also **. I do not believe that your classification of occupations as "low/intermediate/high/inactive" is entirely appropriate. Were these classifications made based on income data, or simply on job title? If the former, you should rename as "low-income" etc. If the latter, then this is a very subjective measure and will not achieve any valuable insights, and should be removed from the manuscript.

I also think the example job titles should be removed:  for example, "housewives/househusbands" is a rather dated title, and many such individuals would object to being described as "inactive".

Likewise, your use of "middle social class" (L375) is very outdated and should be removed.

4. L27: remove "almost"

5. Supplementary material S3: the written description of the objective function is very brief, and not very instructive to readers without a mathematical background. Suggest expanding this section with a more detailed verbal explanation of how the objective function was structured.

6. What were the first-limiting nutrients in the optimised diets? In my experience, it is usually the micronutrients, often calcium. I think the readers could add value to the results/discussion by covering this.

7. Figure 1 and elsewhere: I note the maximal limits set on milk and fish, due to the high amount of milk included in optimised diets otherwise. Perhaps in the discussion the authors could elaborate on what the diets looked like without these limits - were large increases in milk and fish seen in all diets?

8. Figure 2: y axis title would more correctly be "Change in dietary GHGE (%)"

9. Precision of environmental impact data: I think it is worth noting in the discussion the limitations of environmental impact data. GHGE, air acidification, etc. are all highly dependent on food production system; there is no single value for all food items of the same type. Poore and Nemecek (10.1126/science.aaq0216) would be a good reference that demonstrates this variation. I do not think this variation would qualitatively change your conclusions, but it is worth noting that it exists.

Author Response

Thank you for your comments. Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear Editor 

The manuscript has been reviewed following points need to be addressed for the consideration of the manuscript

The abstract should contain key findings of research with conclusive lines

Material and methods should be descriptive 

Recent studies should be added to the result and discussion

Author Response

Thank you for your revision. Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The author added all the suggested comment manuscript can be accepted in its current form