BMJ - Decision on Manuscript ID BMJ.2017.03 7504.R1 Body: 11-May-2017 Dear Dr. Tillmann Manuscript ID BMJ.2017.037504.R1 entitled "Education and coronary heart disease: a Mendelian randomization study" Thank you for sending us your paper. We sent it for external peer review and discussed it at our manuscript committee meeting. We recognise its potential importance and relevance to general medical readers, but I am afraid that we have not yet been able to reach a final decision on it because several important aspects of the work still need clarifying. We hope very much that you will be willing and able to revise your paper as explained below in the report from the reviewers, so that we will be in a better position to understand your study and decide whether the BMJ is the right journal for it. We are looking forward to reading the revised version and, we hope, reaching a decision. Yours sincerely, dr. Wim Weber European editor, The BMJ wweber@bmj.com *** PLEASE NOTE: This is a two-step process. After clicking on the link, you will be directed to a webpage to confirm. *** https://mc.manuscriptcentral.com/bmj?url_mask=0edf61 d2534f4032bc30cc3b8fdb25d0 First, please revise your paper to respond to all of the comments by the reviewers. Their reports are available at the end of this letter, below.
In your response please provide, point by point, your replies to the comments made by the reviewers and the editors, explaining how you have dealt with them in the paper. ** Comments from the external peer reviewers** Reviewer: 1 Recommendation: Comments: I appreciate the authors' comprehensive responses and their incorporating in the revised manuscript. I think the revised manuscript provides much greater balance about the limitations and assumptions of this study. My remaining suggestion is to incorporate some statement about assumptions (mainly pleiotropy) in the abstract, would could easily be interpreted as suggestive of evidence with as strong of a causal interpretation as a randomized trial. Additional Questions: Please enter your name: Stephen Gilman Job Title: Investigator and Acting Branch Chief Institution: NICHD Reimbursement for attending a symposium?: No A fee for speaking?: No A fee for organising education?: No Funds for research?: No Funds for a member of staff?: No Fees for consulting?: No Have you in the past five years been employed by an organisation that may in any way gain or lose financially from the publication of this paper?: No
Do you hold any stocks or shares in an organisation that may in any way gain or lose financially from the publication of this paper?: No If you have any competing interests <A HREF='http://www.bmj.com/about-bmj/resourcesauthors/forms-policies-and-checklists/declarationcompeting-interests'target='_new'> (please see BMJ policy) </a>please declare them here: Reviewer: 2 Recommendation: Comments: Thank you for modifying your interpretation of the data, which now I find more balanced. I still feel there are a few outstanding points to address in this important paper. 1) Somewhat unsurprisingly, the most difficult remaining issue with this paper is pleiotropy. The authors have now more fully described this potential limitation and I think that in general it reads well. However, I would still argue that a description of the InSIDE assumption and its potential violation would be helpful to the average BMJ reader. I find their first three arguments for lack of pleiotropy to be reasonably convincing, however, I do not agree with the fourth, which is stated as Fourth, despite gaps in our understanding of the biological mechanisms through which these 162 SNPs influence education, they are disproportionately found in genomic regions that regulate brain development, they are enriched for biological pathways involved in neural development, and they are preferentially expressed in neural tissue.(16) The authors are likely aware that many of the pathways that lead to obesity and its associated risk factors likely act through neurological pathways. Thus, if a SNP acts upon a neural pathway, that does not mean it does not act on a pathway relevant to CHD. 2) I have looked at the referenced twin studies (but have not read them in detail). In my understanding, the Lungborg paper did not clearly find a twin-difference effect
on self-reported health or number of chronic conditions when using years of schooling as the exposure (Table 2). In the Danish twin study the intrapair twin effects were close to, or included the null. These authors suggested, the overall results of the study seem most compatible with an effect of early environmental factors in explaining the educational inequality in mortality, although the effect of education did not disappear entirely in the intrapair analyses. Last, it was not apparent to me that the Visscher paper looked at education effects, rather it appears that they have looked at intelligence. Thus, I would suggest that the authors interpret the twin studies more cautiously. Minor Comments There is no comprehensive MR method that I m aware of. All MR methods have advantages and disadvantages. Perhaps the authors might consider slightly changing the following sentence: No prior studies have applied the comprehensive Mendelian randomization method. Additional Questions: Please enter your name: Brent Richards Job Title: Associate Professor Institution: McGill University Reimbursement for attending a symposium?: No A fee for speaking?: No A fee for organising education?: No Funds for research?: No Funds for a member of staff?: No Fees for consulting?: No Have you in the past five years been employed by an organisation that may in any way gain or lose financially from the publication of this paper?: No Do you hold any stocks or shares in an organisation that
may in any way gain or lose financially from the publication of this paper?: No If you have any competing interests <A HREF='http://www.bmj.com/about-bmj/resourcesauthors/forms-policies-and-checklists/declarationcompeting-interests'target='_new'> (please see BMJ policy) </a>please declare them here: None Reviewer: 3 Recommendation: Comments: This is a valuable study that makes a substantial contribution to the field. I appreciated the sensitivity analyses and the supplemental materials to go into more depth about the methods employed and the results of the sensitivity analyses. I also enjoyed the very thorough discussion of the strengths and limitations. The authors brought up a lot of insightful points including, the potential difference in effects of education from genetic causes vs environmentally acquired changes (like through compulsory schooling laws) and whether raising education would have same effect for those at lower end versus higher end of the education spectrum. I also appreciated the recommendation that it may be possible for future studies to collect and investigate specific mechanisms, including examining established CHD risk factors as well as others less commonly investigated (improved health care services, better jobs, income, material conditions, social ranking..etc.) Some comments below: In the what is already known about the subject, the authors state, No prior studies have applied the comprehensive Mendelian randomization method (including sensitivity analyses of effects from genetic pleiotropy) in order to investigate how exposure to socioeconomic risk factors might causally change the risk of disease occurrence. I would recommend rewording the sentence to state that few studies have applied a Mendelian randomization method in order to investigate how exposure to socioeconomic risk factors might causally
change risk of disease occurrence and none have conducted sensitivity analyses to examine the potential influence of genetic pleitropy. The original sentence makes it less clear that the big addition of this study is the thorough sensitivity analyses to examine pleitropy because there have been a few studies using genetic instruments for SES to look at health outcomes. On pg 9, the authors describe the 162 SNPS from the meta-analysis of discovery and replication data sets. They also state there were an additional 72 SNPS associated with education from another discovery data set. The authors write they used the 162 SNPS (rather than the 72 SNPs) for the main analysis to maintain sufficient power. It was not clear why the 72 SNPS were not combined with the 162 SNPs for the main analysis. Wouldn t that maximize power? I appreciated the supplemental materials going over the methods for sensitivity analyses (Supplemental section 3.2.1) to investigate extent to which pleiotropy may bias the MR results. Perhaps the authors can add just a few sentences on pg. 10 (where the authors describe the MR analyses) to describe the rationale for the sensitivity analyses and the assumptions being relaxed just to provide readers a big picture understanding of the value of these sensitivity analyses. This is a minor point. The MR-Egger and MR-Egger+SIMEX have wider confidence intervals compared to the standard MR estimates. On pg. 13, the authors just noted that these estimate gave similar findings, but perhaps cam state more specifically that the estimates were less precise but consistent with the original findings. Additional Questions: Please enter your name: Thu Nguyen Job Title: Associate Specialist, Step 4 Institution: University of California, San Francisco Reimbursement for attending a symposium?: No A fee for speaking?: No
A fee for organising education?: No Funds for research?: No Funds for a member of staff?: No Fees for consulting?: No Have you in the past five years been employed by an organisation that may in any way gain or lose financially from the publication of this paper?: Do you hold any stocks or shares in an organisation that may in any way gain or lose financially from the publication of this paper?: No If you have any competing interests <A HREF='http://www.bmj.com/about-bmj/resourcesauthors/forms-policies-and-checklists/declarationcompeting-interests'target='_new'> (please see BMJ policy) </a>please declare them here: