Registration of protocol

1. Did the authors provide evidence that the protocol for their systematic review was registered (e.g., with PROSPERO)?

Rationale and objective

2. Does the introduction clearly describe the rationale in the context of what is already known?

3. Does the introductionprovide an explicit statement of questions being addressed with reference to participants, interventions, comparisons, outcomes, and study design?

Search strategy

Item 8 in the PRISMA checklist states: "Present full electronic search strategy for at least one database, including any limits used, such that it could be repeated."

4. Would it be possible for another person to replicate the electronic search strategy for at least one of the databases used by the authors?

Flow diagram

5. From the flow diagram, are you able to determine why some studies that were included inthe box labelled "Total Studies Included" were not included in the meta-analysis (last box in the diagram)?

Descriptive information about each study in relation to selection bias, information bias and confounding

Table 2 in the supplementary information provided data/information extracted from each of the studies that were included in the review of asthma. An important purpose of this table is to enable the reader to assess each study's risk of bias due to selection bias, information bias and confounding.

6. What key items should be extracted from each of the study reports (articles) and included in Supplementary table 2 to enable the reader to assess the risk of selection bias in each of the studies? Was each item included in Supplementary table 2?

You can ignore possible selection bias due to inclusion of prevalent cases for the cross-sectional studies, since this would be an issue common to them all.

Please use the table below for your answers. You will see it asks you to list items separately for each of the three types of studies included in the review. Add extra rows as necessary. Don't assume that the number of rows indicates how many items you need to mention!

If an item was only partially included (i.e. some but not all of the relevant information was included, state what was missing).

Please do not format the table in landscape mode. It makes marking in Turnitin very difficult.

Marks will be allocated on a scale

6 marks - outstanding, all key points covered
5 marks - excellent, only minor problems
4 marks - very good, a few problems
3 marks - good, several problems
2 marks - poor, many problems
1 mark - very poor
0 marks - not answered

7. What key items should be extracted from each of the study reports (articles) and included in Supplementary table2 to enable the reader to assess the risk of information bias in each of the studies? Was each item included in Supplementary table 2?

Please make a copy of the table at the end of this document for your answers. As for selection bias, you will need to list items separately for each of the three types of studies included in the review.

Please use the table below for your answers. You will see it asks you to list items separately for each of the three types of studies included in the review. If you feel that the information is the same for more than one type of study (e.g. cohort and cross-sectional studies), you can amalgamate the relevant rows. If you do that, justify why you did it.

Add extra rows as necessary. Don't assume that the number of rows indicates how many items you need to mention!

If an item was only partially included (i.e. some but not all of the relevant information was included, state what was missing).

8. What information is necessary for a reader to assess the risk of confounding? Is this provided in Supplementary table 2?

Assessing risk of bias of individual studies & overall quality

9. What tool did the authors use to assess risk of bias (or quality) of individual studies?

10. What tool did the authors use to assess the overall quality of studies for each association?

Results for asthma

In the blue box on page 38, the authors state "There is low grade quality evidence that longer duration of breastfeeding is associated with a reduced risk of asthma in children aged 5-18 years."

11. In your own words, describe the relevant results. (Don't describe subgroup analyses or any of the results in Table 2)

12. On page 41, the authors stated "There was some evidence of publication bias from the funnel plot and Egger's test (p = 0.018), with a propensity for publishing small studies showing protective effects (Fig. S3)". Look at supplementary figure 3. Discuss why the authors reached this conclusion. (You can ignore the test and p-value.)

13. Table 2 shows the results of an analysis designed to explore reasons for heterogeneity in the associations between asthma and any type of breast feeding. The analysis is called meta-regression. The pooled odds ratios for cohort and cross-sectional studies are quite different.

a) Does that make you feel more or less comfortable about including cross-sectional studies in the overall meta-analysis?
b) How does the finding of different odds ratios for cohort and cross-sectional studies relate to the following sentence from the abstract? "Some of the protective effect of breastfeeding for asthma may be related to recall bias in studies of lesser methodological quality."

Bradford Hill guidelines for assessing causality

We will consider the outcome asthma at 5-18 years of age only. You can ignore the other outcomes for the questions about the Bradford Hill guidelines.

Unless specified otherwise, you should consider all three categorisations of breast feeding discussed in the paper.

Consistency of the association

14. How consistent is the evidence that breast feeding is associated with reduced risk of asthma?

Temporality

15. How strong is the evidence that the exposure (breast feeding) precedes asthma at 5-18 years of age?

Strength of the association

16. How strong do you consider the associations between breast feeding and asthma at 5-18 years of age?

Dose-response relationship

17. Does the paper provide evidence on whether there is a dose-response relationship between breast feeding and risk of asthma?

Biological plausibility

18. Do the authors attempt to make a case for the biological mechanisms by whichbreast feeding might affect risk of asthma? [we just want to know whether you think they have provided an argument that might convince an expert]

Evidence to action

19. Based only on this systematic review and your responses to the previous questions, do you think the evidence that breast feeding reduces the risk of asthma is of sufficient quality to take public health action? If yes, why? If not, why not? Please write a paragraph of between 50 and 100 words summarising the evidence and stating your conclusion.