TThe manuscript by Reznik et al shows the findings of a prospective before-and-after observational multicenter study conducted to assess the influence of an automated insulin delivery system on patient-related outcomes in France.

The results show an association with improved outcomes in adults and, for hypoglycemia perception, in both adults and adolescents.

This reviewer found the results of the study of potential interest to the readership of the journal. However, the Authors need to address some relevant issues/comments:

• As observational studies are intrinsically associated with some relevant biases, the researchers are kindly requested to follow the STROBE statement in the communication of this type of studies. Please, include the STROBE checklist as supplementary material, and also a study flow-chart.

• The Authors describe that the primary outcome was two-fold, i.e. quality of life and diabetes distress. Please, provide a description of the study sample calculation, and the details on which was the outcome variable used to calculate this.

• Please, explain the reason for using the description of server hypoglycemia, instead of using the usual definition.

• Definition for all the study variables should be provided (e.g. diabetic nephropathy, retinopathy, etc.). The authors should follow the principle that any independent researcher should be able to reproduce the study with all the information provided in the paper.

• Please, provide insulin doses in units per kg per day.

• The description of statistical methods is a poor one. Surprisingly, they describe the results of logistic regression modelling without providing all the details and how this was conducted (no mention under the section on statistical analysis).

• Please, avoid the use of impact or effect in the concluding statements. An observational study can only demonstrate association between an intervention and any outcome variable.

No additional comments