NOTICE: This document contains correspondence generated during peer review and subsequent revisions but before transmittal to production for composition and copyediting:
• Comments from the reviewers and editors (email to author requesting revisions)
• Response from the author (cover letter submitted with revised manuscript)*
*The corresponding author has opted to make this information publicly available.
Personal or nonessential information may be redacted at the editor’s discretion.
Questions about these materials may be directed to the Obstetrics & Gynecology editorial office:
[email protected].
Date: Sep 16, 2022
To: "Monique Marie Hedderson"
From: "The Green Journal" [email protected] Subject: Your Submission ONG-22-1454
RE: Manuscript Number ONG-22-1454
Perinatal outcomes post-bariatric surgery compared with matched controls in a large integrated health care system Dear Dr. Hedderson:
Thank you for sending us your work for consideration for publication in Obstetrics & Gynecology. Your manuscript has been reviewed by the Editorial Board and by special expert referees. The Editors would like to invite you to submit a revised version for further consideration.
If you wish to revise your manuscript, please read the following comments submitted by the reviewers and Editors. Each point raised requires a response, by either revising your manuscript or making a clear argument as to why no revision is needed in the cover letter.
To facilitate our review, we prefer that the cover letter you submit with your revised manuscript include each reviewer and Editor comment below, followed by your response. That is, a point-by-point response is required to each of the EDITOR COMMENTS (if applicable), REVIEWER COMMENTS, STATISTICAL EDITOR COMMENTS (if applicable), and EDITORIAL OFFICE COMMENTS below. Your manuscript will be returned to you if a point-by-point response to each of these sections is not included.
The revised manuscript should indicate the position of all changes made. Please use the "track changes" feature in your document (do not use strikethrough or underline formatting).
Your submission will be maintained in active status for 21 days from the date of this letter. If we have not heard from you by Oct 07, 2022, we will assume you wish to withdraw the manuscript from further consideration.
REVIEWER COMMENTS:
Reviewer #1:
The authors present a case control study evaluating perinatal outcomes following bariatric surgery. The authors use a matched control group and find an increased risk of SGA, NICU admission, cesarean and PTB among patients who have undergone bariatric surgery. They note lower rates of GDM, preeclampsia and LGA. The study includes large numbers in a diverse cohort. Much of the data has been seen in prior studies as is noted by the authors.
Abstract:
Line 10- is this more correctly identified as a case-control study?
Introduction:
Line 70- how is the current study different from the recently published Kaiser study?
Line 92- Later the authors note an evaluation of time since bariatric surgery, please include the evaluation of the time since surgery in the methods (and the number of patients that this data was able to be obtained from).
Line 92- how were multiple pregnancies in this time period handled?
Line 102- How many patients had the surgery within one year of the pregnancy? Would this alter results when evaluating BMI or matching based on BMI if it was obtained up to 12 months prior to pregnancy.
Line 120- did the authors include both pre-existing HTN and HTN with super imposed preeclampsia?
Line 138- did the authors obtain data regarding primary versus repeat CD?
View Letter
1 of 6 10/17/2022, 11:51 AM
Line 140- again the pre-pregnancy weight may have been up to 1 year prior to pregnancy which can include the time prior to surgery.
Line 162- did the authors conduct a power analysis?
Results:
Line 171- was data obtained regarding indicated versus spontaneous preterm birth? It would be interesting to know the number of patients delivered for FGR?
Line 174- did the controls developing preeclampsia have higher rates of underlying HTN?
Line 175- again was this repeat or primary Cesareans?
Line 188- why was the time period of 2 years chosen for the sensitivity analysis?
Comment:
Line 240- How many patients were in the <12 month time period? How many of the SGA infants were among patients with more recent bariatric surgery?
Line 249- this is the first time the authors note a limitation on the sensitivity analysis. This limit should be noted in the methods section and the number evaluated in the sensitivity analysis included.
Reviewer #2:
Abstract:
1. The objective is clearly stated.
2. The primary outcomes were not mentioned.
3. Line 19-20: Mention age being a significant difference.
4. Results were reported using adjusted risk ratios and CI, which are helpful for clinical analyses.
Introduction:
1. Line 49: Please revise sentence, and clarify which group of women's obesity rates are being compared.
2. Line 51-56: Consider truncating.
3. Line 67: Clarify how your study differs from this study cited.
4. The objective of the study was clearly stated.
5. The primary outcomes were not mentioned.
Body:
1. Line 78- 86: Deidentify institution name.
2. Line 92: Consider explaining why the search query was limited to livebirths when the objective was to evaluate perinatal outcomes.
3. Line 97: Consider explaining why NDI was selected, as opposed to education or household income?
4. Line 113: Outcomes are clearly defined in the body of the text. Please include details of this in the abstract and introduction.
5. Line 115: "(EHR)" does not need to be repeated.
6. Line 138: Change EMR to EHR or include the full title.
7. Line 160-162: This is an important point. Consider including a truncated version of this sentence in the abstract.
8. This study did a great job with the statistical analysis of the study results. The statistical analysis plan was straight forward and appropriately detailed. Analysis of results included crude calculations and adjusted findings that can be applied clinically.
9. Line 209-214: Consider further expounding on the similarities and differences between the studies cited and this study given the mixed results.
10. Line 223-225: Does this limit the generalizability of the study results? Consider including in the study limitations.
11. Line 236: Remove the first "on" from "Further data on are needed on…"
12. Consider providing more information on research implications. The benefits of a prospective investigation of this population. Additionally, considering the implications of comparing post-bariatric surgery patients to matched controls (KPNC study) versus post-bariatric surgery patients to eligible candidates that did not undergo surgery (KPSC study).
13. The retrospective nature of the study should be mentioned as a limitation.
Conclusion:
1. Summary of study was clearly stated.
Figures and Tables:
1. Appropriate.
View Letter
2 of 6 10/17/2022, 11:51 AM
References:
1. Appropriate.
Checklist:
1. STROBE not included.
STATISTICAL EDITOR COMMENTS:
General: The basic method used was first to match the cases ~ 1:5 with controls, but then to assess crude RR and RRs adjusted for multiple variables that were employed in the matching process.
Should refine the matching process (suggest propensity score matching). Since the total size of the singleton data base was > 260,000 vs a case cohort of ~1600 bariatric surgery cases, that should be quite doable. Then, compare outcomes for the matched groups. If desired, could also do multivariable regression analysis using all available controls and adjust for relevant variables/characteristics.
Also, should include a flow diagram of the database.
Need to include a Table showing the baseline characteristics of all patients included in the analysis and either enumerate all missing data or show in flow diagram how many were excluded at each step due to missing data.
Table 1: See above comments, but typically the format is a comparison of standardized mean differences (SDM) before and after matching, rather than evaluation of p-values.
EDITORIAL OFFICE COMMENTS:
1. If your article is accepted, the journal will publish a copy of this revision letter and your point-by-point responses as supplemental digital content to the published article online. You may opt out by writing separately to the Editorial Office at [email protected], and only the revision letter will be posted.
2. When you submit your revised manuscript, please make the following edits to ensure your submission contains the required information that was previously omitted for the initial double-blind peer review:
* Funding information (ie, grant numbers or industry support statements) should be disclosed on the title page and at the end of the abstract. For industry-sponsored studies, describe on the title page how the funder was or was not involved in the study.
* Include clinical trial registration numbers, PROSPERO registration numbers, or URLs at the end of the abstract (if applicable).
* Name the IRB or Ethics Committee institution in the Methods section (if applicable).
* Add any information about the specific location of the study (ie, city, state, or country), if necessary for context.
3. Obstetrics & Gynecology's Copyright Transfer Agreement (CTA) must be completed by all authors. When you uploaded your manuscript, each coauthor received an email with the subject, "Please verify your authorship for a submission to Obstetrics & Gynecology." Please ask your coauthor(s) to complete this form, and confirm the disclosures listed in their CTA are included on the manuscript's title page. If they did not receive the email, they should check their spam/junk folder.
Requests to resend the CTA may be sent to [email protected].
4. For studies that report on the topic of race or include it as a variable, authors must provide an explanation in the manuscript of who classified individuals' race, ethnicity, or both, the classifications used, and whether the options were defined by the investigator or the participant. In addition, describe the reasons that race and ethnicity were assessed in the Methods section and/or in table footnotes. Race and ethnicity must have been collected in a formal or validated way. If it was not, it should be omitted. Authors must enumerate all missing data regarding race and ethnicity as in some cases missing data may comprise a high enough proportion that it compromises statistical precision and bias of analyses by race.
Use "Black" and "White" (capitalized) when used to refer to racial categories.
List racial and ethnic categories in tables in alphabetic order. Do not use "Other" as a category; use "None of the above"
instead.
Please refer to "Reporting Race and Ethnicity in Obstetrics & Gynecology" at https://edmgr.ovid.com/ong/accounts /Race_and_Ethnicity.pdf.
View Letter
6 10/17/2022, 11:51 AM
5. ACOG uses person-first language. Please review your submission to make sure to center the person before anything else. Examples include: "People with disabilities" or "women with disabilities" instead of "disabled people" or "disabled women"; "patients with HIV" or "women with HIV" instead of "HIV-positive patients" or "HIV-positive women"; and "people who are blind" or "women who are blind" instead of "blind people" or "blind women."
6. The journal follows ACOG's Statement of Policy on Inclusive Language (https://www.acog.org/clinical-information /policy-and-position-statements/statements-of-policy/2022/inclusive-language). When possible, please avoid using gendered descriptors in your manuscript. Instead of "women" and "females," consider using the following: "individuals;"
"patients;" "participants;" "people" (not "persons"); "women and transgender men;" "women and gender-expansive patients;" or "women and all those seeking gynecologic care.
7. Responsible reporting of research studies, which includes a complete, transparent, accurate and timely account of what was done and what was found during a research study, is an integral part of good research and publication practice and not an optional extra. Obstetrics & Gynecology supports initiatives aimed at improving the reporting of health research, and we ask authors to follow specific guidelines:
STROBE: observational studies
Include the appropriate checklist for your manuscript type upon submission, if applicable, and indicate in your cover letter which guideline you have followed. Please write or insert the page numbers where each item appears in the margin of the checklist. Further information and links to the checklists are available at www.equator-network.org/.
8. Standard obstetric and gynecology data definitions have been developed through the reVITALize initiative, which was convened by the American College of Obstetricians and Gynecologists and the members of the Women's Health Registry Alliance. Obstetrics & Gynecology has adopted the use of the reVITALize definitions. Please access the obstetric data definitions at https://www.acog.org/practice-management/health-it-and-clinical-informatics/revitalize-obstetrics-data- definitions and the gynecology data definitions at https://www.acog.org/practice-management/health-it-and-clinical- informatics/revitalize-gynecology-data-definitions. If use of the reVITALize definitions is problematic, please discuss this in your point-by-point response to this letter.
9. Make sure your manuscript meets the following word limit. The word limit includes the manuscript body text only (for example, the Introduction through the Discussion in Original Research manuscripts), and excludes the title page, précis, abstract, tables, boxes, and figure legends, reference list, and supplemental digital content. Figures are not included in the word count.
Original Research: 3,000 words
10. Specific rules govern the use of acknowledgments in the journal. Please review the following guidelines and edit your title page as needed:
* All financial support of the study must be acknowledged.
* Any and all manuscript preparation assistance, including but not limited to topic development, data collection, analysis, writing, or editorial assistance, must be disclosed in the acknowledgments. Such acknowledgments must identify the entities that provided and paid for this assistance, whether directly or indirectly.
* All persons who contributed to the work reported in the manuscript, but not sufficiently to be authors, must be acknowledged. Written permission must be obtained from all individuals named in the acknowledgments, as readers may infer their endorsement of the data and conclusions. Please note that your response in the journal's electronic author form verifies that permission has been obtained from all named persons.
* If all or part of the paper was presented at the Annual Clinical and Scientific Meeting of the American College of Obstetricians and Gynecologists or at any other organizational meeting, that presentation should be noted (include the exact dates and location of the meeting or indicate whether the meeting was held virtually).
* If your manuscript was uploaded to a preprint server prior to submitting your manuscript to Obstetrics & Gynecology, add the following statement to your title page: "Before submission to Obstetrics & Gynecology, this article was posted to a preprint server at: [URL]."
* Do not use only authors' initials in the acknowledgement or Financial Disclosure; spell out their names the way they appear in the byline.
11. Provide a short title of no more than 45 characters, including spaces, for use as a running foot. Do not start the running title with an abbreviation.
12. Be sure that each statement and any data in the abstract are also stated in the body of your manuscript, tables, or figures. Statements and data that appear in the abstract must also appear in the body text for consistency. Make sure there are no inconsistencies between the abstract and the manuscript, and that the abstract has a clear conclusion statement based on the results found in the manuscript.
In addition, the abstract length should follow journal guidelines. Please provide a word count.
Original Research: 300 words
View Letter
6 10/17/2022, 11:51 AM
13. Only standard abbreviations and acronyms are allowed. A selected list is available online at http://edmgr.ovid.com /ong/accounts/abbreviations.pdf. Abbreviations and acronyms cannot be used in the title or précis. Abbreviations and acronyms must be spelled out the first time they are used in the abstract and again in the body of the manuscript.
14. The journal does not use the virgule symbol (/) in sentences with words, except with ratios. Please rephrase your text to avoid using "and/or," or similar constructions throughout the text. You may retain this symbol if you are using it to express data or a measurement.
15. In your abstract, manuscript Results sections, and tables, the preferred citation should be in terms of an effect size, such as odds ratio or relative risk or the mean difference of a variable between two groups, expressed with appropriate confidence intervals. When such syntax is used, the P value has only secondary importance and often can be omitted or noted as footnotes in a Table format. Putting the results in the form of an effect size makes the result of the statistical test more clinically relevant and gives better context than citing P values alone.
Please standardize the presentation of your data throughout the manuscript submission. For P values, do not exceed three decimal places (for example, "P = .001").
Express all percentages to one decimal place (for example, 11.1%"). Do not use whole numbers for percentages.
16. Please review the journal's Table Checklist to make sure that your tables conform to journal style. The Table Checklist is available at http://edmgr.ovid.com/ong/accounts/table_checklist.pdf.
17. Please review examples of our current reference style at https://edmgr.ovid.com/ong/accounts/ifa_suppl_refstyle.pdf.
Include the digital object identifier (DOI) with any journal article references and an accessed date with website references.
Unpublished data, in-press items, personal communications, letters to the editor, theses, package inserts, submissions, meeting presentations, and abstracts may be included in the text but not in the formal reference list. Please cite them on the line in parentheses.
If you cite ACOG documents in your manuscript, be sure the references you are citing are still current and available. Check the Clinical Guidance page at https://www.acog.org/clinical (click on "Clinical Guidance" at the top). If the reference is still available on the site and isn't listed as "Withdrawn," it's still a current document. In most cases, if an ACOG document has been withdrawn, it should not be referenced in your manuscript.
Please make sure your references are numbered in order of appearance in the text.
18. Authors whose manuscripts have been accepted for publication have the option to pay an article processing charge and publish open access. With this choice, articles are made freely available online immediately upon publication. An
information sheet is available at http://links.lww.com/LWW-ES/A48. The cost for publishing an article as open access can be found at https://wkauthorservices.editage.com/open-access/hybrid.html.
If your article is accepted, you will receive an email from the Editorial Office asking you to choose a publication route (traditional or open access). Please keep an eye out for that future email and be sure to respond to it promptly.
***
If you choose to revise your manuscript, please submit your revision through Editorial Manager at
http://ong.editorialmanager.com. Your manuscript should be uploaded as a Microsoft Word document. Your revision's cover letter should include a point-by-point response to each of the received comments in this letter. Do not omit your responses to the EDITOR COMMENTS (if applicable), the REVIEWER COMMENTS, the STATISTICAL EDITOR COMMENTS (if applicable), or the EDITORIAL OFFICE COMMENTS.
If you submit a revision, we will assume that it has been developed in consultation with your coauthors and that each author has given approval to the final form of the revision.
Again, your manuscript will be maintained in active status for 21 days from the date of this letter. If we have not heard from you by Oct 07, 2022, we will assume you wish to withdraw the manuscript from further consideration.
Sincerely,
Dwight J. Rouse, MD, MSPH Deputy Editor, Obstetrics
__________________________________________________
In compliance with data protection regulations, you may request that we remove your personal registration details at any time. (Use the following URL: https://www.editorialmanager.com/ong/login.asp?a=r). Please contact the publication office if you have any questions.
View Letter
6 10/17/2022, 11:51 AM
October 7, 2022 Dear Editorial Board,
We are pleased to submit our responses to the reviewers’ comments for the manuscript titled Perinatal outcomes post-bariatric surgery compared with matched controls in a large integrated health care system.
This manuscript is submitted solely to Obstetrics and Gynecology. It is not under consideration for publication elsewhere and will not be submitted for consideration elsewhere except in case of a final negative decision made by the Editors. We followed the STROBE guidelines for observational studies.
The Kaiser Permanente Northern California Institutional Review Board determined this study to be exempt.
These findings were previously presented in poster form at the Society for Maternal Fetal Medicine 42
ndAnnual Pregnancy Meeting, held virtually January 31 – February 5, 2022.
Please see our responses to the comments below.
REVIEWER COMMENTS:
Reviewer #1:
The authors present a case control study evaluating perinatal outcomes following bariatric surgery. The authors use a matched control group and find an increased risk of SGA, NICU admission, cesarean and PTB among patients who have undergone bariatric surgery. They note lower rates of GDM,
preeclampsia and LGA. The study includes large numbers in a diverse cohort. Much of the data has been seen in prior studies as is noted by the authors.
Abstract:
Line 10- is this more correctly identified as a case-control study?
In a case-control study design cases are identified based on the outcome. In this study we matched on the exposure of bariatric surgery not the outcome, therefore, we believe it is correctly described as a cohort study.
Introduction:
Line 70- how is the current study different from the recently published Kaiser study?
Differences in study design and findings for key variables are now discussed starting on page 14 (the first
paragraph under ‘Results in the Context of What is Known’) in the Discussion section. One important
difference is that the Getahun study compared patients who underwent bariatric surgery versus patients
who met eligibility criteria but did not undergo bariatric surgery; in contrast, we did not require our
controls to meet the eligibility criteria for bariatric surgery at baseline; instead, we matched on pre- pregnancy BMI. Thus, in our study we were comparing clinical outcomes among patients who underwent bariatric surgery to patients who started pregnancy at a similar BMI but did not have a history of
bariatric surgery. Therefore, the two studies answer a slightly different clinical question.
Line 92- Later the authors note an evaluation of time since bariatric surgery, please include the
evaluation of the time since surgery in the methods (and the number of patients that this data was able to be obtained from).
We obtained information on time since and type of surgery on n=681 pregnancies post-bariatric surgeries from either the clinical database of bariatric surgeries or electronic health record for bariatric surgeries performed at KPNC. We now added this information to the manuscript - see page 8, lines 147- 149.
Line 92- how were multiple pregnancies in this time period handled?
We included multiple pregnancies in the analysis and fit a generalized estimating equation model to account for potential correlations. We now describe that in the statistical analysis section (see page 11;
3
rdsentence of the first paragraph of the Statistical Analysis section). We also specify that the
pregnancies post-bariatric surgery group included 1,560 pregnancies among 1,334 unique patients and the matched control group included 7,383 pregnancies by 7,257 patients (see page 8; final sentence before ‘Matching Variables’ section).
Line 102- How many patients had the surgery within one year of the pregnancy? Would this alter results when evaluating BMI or matching based on BMI if it was obtained up to 12 months prior to pregnancy.
Among the 681 pregnancies with information on timing of surgery, n=121 had the surgery within one year of the pregnancy. Of the 121 pregnancies, 118 had a pre-pregnancy BMI that was measured after the surgery, only 3 had pre-pregnancy BMI that was measured before surgery. Therefore, we do not think this issue significantly impacted our results.
Line 120- did the authors include both pre-existing HTN and HTN with super imposed preeclampsia?
We appreciate the reviewer’s comment and we now updated our hypertensive disorder definitions. First, we moved pre-existing hypertension and pre-existing diabetes to Table 1 since they are baseline
characteristics rather than outcomes. We also now further adjust our adjusted models in Table 3 for chronic/pre-existing HTN and diabetes since they are baseline characteristics not matched on. We also updated our algorithm so that women with pre-existing HTN can also develop super imposed
preeclampsia - there were 214 (44 in the case group and 190 in the comparison group) and they are now included in the preeclampsia outcome in Tables 2-4.
Line 138- did the authors obtain data regarding primary versus repeat CD?
We appreciate this suggestion and we now added information on primary versus repeat cesarean
delivery to Table 2. We found that 59.7% of cesarean deliveries among pregnancies post bariatric surgery were repeat cesarean deliveries compared to 51.3% among the comparison group.
Line 140- again the pre-pregnancy weight may have been up to 1 year prior to pregnancy which can include the time prior to surgery.
See response above only where we found only 3 weights were obtained prior to surgery. Therefore, we
do not think this impacted our results.
Line 162- did the authors conduct a power analysis?
We did not conduct a power analysis. We are following the STROBE guidelines, which explicitly state that no post hoc power calculations should be performed in observational studies and that “From the point of view of the reader, confidence intervals indicate the statistical precision that was ultimately obtained.”
Vandenbroucke JP, von Elm E, Altman DG, Gøtzsche PC, Mulrow CD, Pocock SJ, Poole C, Schlesselman JJ, Egger M; STROBE Initiative. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. Int J Surg. 2014 Dec;12(12):1500-24. doi:
10.1016/j.ijsu.2014.07.014. Epub 2014 Jul 18. PMID: 25046751.
Results:
Line 171- was data obtained regarding indicated versus spontaneous preterm birth? It would be interesting to know the number of patients delivered for FGR?
We now added information on medically indicated and spontaneous preterm birth to Table 2.
Pregnancies post bariatric surgery were more likely to have medically indicated preterm births compared to matched controls. Unfortunately, the ICD coding of intrauterine growth restriction is not reliable and that is why we selected small for gestational age as a hard outcome reflective of intrauterine growth.
Line 174- did the controls developing preeclampsia have higher rates of underlying HTN?
The controls developing preeclampsia had a lower rate of underlying HTN.
116 cases developed pre-eclampsia, 44 had pre-existing HTN, the rate was 37.9%.
670 controls developed pre-eclampsia, 190 had pre-existing HTN; the rate was 28.4%
We now also include the baseline pre-existing HTN in Table 1 which was higher among post-bariatric pregnancies compared to matched controls.
Line 175- again was this repeat or primary Cesareans?
See response above.
Line 188- why was the time period of 2 years chosen for the sensitivity analysis?
Expert recommendation holds to defer conception for 12-24 months following bariatric surgery to minimize fetal exposure to rapid parental weight loss and allow time for patients to achieve weight loss goals prior to pregnancy. We selected the time period of two years based upon this guidance. See lines 241-244 (page 12, final sentence before the ‘Results’ section) and references 23 and 24 (ACOG Practice Bulletin Number 105).
Comment:
Line 240- How many patients were in the <12 month time period? How many of the SGA infants were among patients with more recent bariatric surgery?
Among the 681 with information on timing of surgery, 121 had the surgery within one year of the
pregnancy, and n=19 (15.7%) delivered an SGA baby, which was higher than among patients who had
the surgery ≥ 12 months of the pregnancy where n=64 (11.5%) delivered a SGA baby. However, the p-
value of the difference is 0.405 was not statistically significant.
Line 249- this is the first time the authors note a limitation on the sensitivity analysis. This limit should be noted in the methods section and the number evaluated in the sensitivity analysis included.
We have now added to our methods section the number for which we had information on timing and type of surgery. See lines 239 – 244 on page 12.
Reviewer #2:
Abstract:
1. The objective is clearly stated.
2. The primary outcomes were not mentioned.
Now added to abstract
3. Line 19-20: Mention age being a significant difference.
This is now added to the abstract.
4. Results were reported using adjusted risk ratios and CI, which are helpful for clinical analyses.
Introduction:
1. Line 49: Please revise sentence, and clarify which group of women's obesity rates are being compared.
This is now revised and clarified, thank you 2. Line 51-56: Consider truncating.
Now truncated, thank you
3. Line 67: Clarify how your study differs from this study cited.
4. The objective of the study was clearly stated.
5. The primary outcomes were not mentioned.
Primary outcomes are now listed in the introduction.
Body:
1. Line 78- 86: Deidentify institution name.
Institution is now de-identified
2. Line 92: Consider explaining why the search query was limited to livebirths when the objective was to evaluate perinatal outcomes.
Due to outcomes of interest occurring primarily among livebirths.
3. Line 97: Consider explaining why NDI was selected, as opposed to education or household income?
We do not have individual level information on education and income so to adjust for socioeconomic
status we rely on census tract information. Neighborhood Deprivation Index (NDI) is calculated for each
Census tract in the U.S. using a factor analysis to identify key variables from 13 measures in the following
dimensions of socioeconomic (SES) status: wealth and income, education, occupation, and housing
conditions(Messer). It has been associated with several perinatal complications in past studies(Ncube,
O’Campo, Janevic)..
Messer, L.C., et al., The development of a standardized neighborhood deprivation index. J Urban Health, 2006. 83(6): p. 1041-62.
Ncube, C.N., et al., Association of neighborhood context with offspring risk of preterm birth and low birthweight: A systematic review and meta-analysis of population-based studies. Soc Sci Med, 2016. 153:
p. 156-64.
O'Campo, P., et al., Neighborhood deprivation and preterm birth among non-Hispanic Black and White women in eight geographic areas in the United States. Am J Epidemiol, 2008. 167(2): p. 155-63.
Janevic, T et al. “Neighborhood deprivation and adverse birth outcomes among diverse ethnic groups.”
Annals of epidemiology vol. 20,6 (2010): 445-51. doi:10.1016/j.annepidem.2010.02.010
Elo, Irma T et al. “Neighbourhood deprivation and small-for-gestational-age term births in the United States.” Paediatric and perinatal epidemiology vol. 23,1 (2009): 87-96. doi:10.1111/j.1365-
3016.2008.00991.x
4. Line 113: Outcomes are clearly defined in the body of the text. Please include details of this in the abstract and introduction.
Primary outcomes are now identified in the abstract and the introduction, thank you
5. Line 115: "(EHR)" does not need to be repeated.
This is now corrected, thank you
6. Line 138: Change EMR to EHR or include the full title.
This is now corrected, thank you
7. Line 160-162: This is an important point. Consider including a truncated version of this sentence in the abstract.
Now added to abstract, thank you
8. This study did a great job with the statistical analysis of the study results. The statistical analysis plan was straight forward and appropriately detailed. Analysis of results included crude calculations and adjusted findings that can be applied clinically.
9. Line 209-214: Consider further expounding on the similarities and differences between the studies cited and this study given the mixed results.
Differences and similarities are now further discussed (See page 15, 1
stparagraph; lines 291 - 310).
10. Line 223-225: Does this limit the generalizability of the study results? Consider including in the study limitations.
This is now included in our study limitations.
11. Line 236: Remove the first "on" from "Further data on are needed on…"
Now corrected, thank you
12. Consider providing more information on research implications. The benefits of a prospective investigation of this population. Additionally, considering the implications of comparing post-bariatric surgery patients to matched controls (KPNC study) versus post-bariatric surgery patients to eligible candidates that did not undergo surgery (KPSC study).
See the updated discussion of how our control group compares to the control group of the KPSC study (page 14, paragraph 1). We also now mention the following in the discussion.
Further prospective research is needed to evaluate whether nutritional, surgical, or patient specific factors can improve fetal growth and perinatal outcomes following bariatric surgery.
13. The retrospective nature of the study should be mentioned as a limitation.
This is now included as a limitation.
Conclusion:
1. Summary of study was clearly stated.
Figures and Tables:
1. Appropriate.
References:
1. Appropriate.
Checklist:
1. STROBE not included.
STATISTICAL EDITOR COMMENTS:
General: The basic method used was first to match the cases ~ 1:5 with controls, but then to assess crude RR and RRs adjusted for multiple variables that were employed in the matching process.
Should refine the matching process (suggest propensity score matching). Since the total size of the singleton data base was > 260,000 vs a case cohort of ~1600 bariatric surgery cases, that should be quite doable. Then, compare outcomes for the matched groups. If desired, could also do multivariable regression analysis using all available controls and adjust for relevant variables/characteristics.
We completely agree that using propensity scores is a good way to approach matching to control for
confounding.
At this point, it will be substantial work for us to take this approach as it entails starting our study and analyses from scratch. Our original study design and analytical approach are valid, and we believe that we will likely see comparable results if propensity score matching were employed. To summarize our approach, we matched on six potential confounders (pre-pregnancy body mass index, age, parity, year of delivery, race/ethnicity, and neighborhood deprivation index). We were able to match 5 unexposed pregnancies to one exposed pregnancy with exactly the same covariate values (known as k:1 exact matching) because the number of potential confounders under consideration was small and our pool of unexposed pregnancies was large. We performed matching using a nearest neighbor (or greedy) matching algorithm using the SAS gsmatch macro [Kosanke, et al.], which has been widely cited. Our approach aligns with guidance in the literature on matching implementation, which states that when
“there are many more control than treated individuals (e.g., more than 3 times as many), k:1 nearest neighbor matching without replacement is a good choice for its simplicity and good performance”
[Stuart]. Finally, based on the statistical literature, we adjusted for the variables used in the matching procedure to reduce bias [Sjölander, Greenland] and control for possible residual confounding.
We believe that using propensity score matching would not change the findings in this study. First, Table 1 demonstrates that the distribution of potential confounders is comparable between the two groups, with the exception of age where the p-value corresponding comparison of the two means is 0.02 (mean age (SD) in post-bariatric surgery group vs comparison group: 34.1 (4.6) vs 33.8 (4.4)). When we consider the standardized mean difference (SMD), which assesses effect sizes and is independent of sample size (which impacts statistical significance), the SMD for age is 0.07, indicating that age is balanced between the two groups (in the literature, SMD values < 0.10 indicate comparable baseline distributions
[Normand et al.]). Second, propensity score matching lends itself to settings where the number of potential confounders is very large where exact matching cannot feasibly be performed, since this approach conveniently matches exposed to unexposed individuals based on a single propensity score [Ho et al.; Stuart; Austin]; however, in our study there were a small number of potential confounders under consideration, which we were able to perform an exact match on. If we were to employ propensity score matching, we would be using the same 6 potential confounders that we used in our original matching approach to build the propensity score; there is no apparent gain in information. Moreover, a recent study comparing exact matching and propensity score matching with a small number of covariates (<10) using real data demonstrated that both approaches produced similar results with respect to estimation and inference [Burden].
Kosanke J, Bergstralh E. GMATCH SAS Macro. Mayo Clinic College of Medicine.
2007. http://bioinformaticstools.mayo.edu/research/gmatch/. Accessed 7 Feb 2019.
Sjölander A, Greenland S. Ignoring the matching variables in cohort studies–when is it valid and why?.
Statistics in medicine. 2013 Nov 30;32(27):4696-708.
Ho DE, Imai K, King G, Stuart EA. Matching as nonparametric preprocessing for reducing model dependence in parametric causal inference. Political analysis. 2007;15(3):199-236.
Stuart EA. Matching methods for causal inference: A review and a look forward. Statistical science: a review journal of the Institute of Mathematical Statistics. 2010 Feb 2;25(1):1.
Austin PC. An introduction to propensity score methods for reducing the effects of confounding in
observational studies. Multivariate behavioral research. 2011 May 31;46(3):399-424.
Normand SL, Landrum MB, Guadagnoli E, Ayanian JZ, Ryan TJ, Cleary PD, McNeil BJ. Validating recommendations for coronary angiography following acute myocardial infarction in the elderly: a matched analysis using propensity scores. Journal of clinical epidemiology. 2001 Apr 1;54(4):387-98.
Burden A, Roche N, Miglio C, Hillyer EV, Postma DS, Herings RM, Overbeek JA, Khalid JM, van Eickels D, Price DB. An evaluation of exact matching and propensity score methods as applied in a comparative effectiveness study of inhaled corticosteroids in asthma. Pragmat Obs Res. 2017 Mar 22;8:15-30. doi:
10.2147/POR.S122563.
Also, should include a flow diagram of the database.
We now added a figure 1 which is a flowchart of our cohort creation.
Need to include a Table showing the baseline characteristics of all patients included in the analysis and either enumerate all missing data or show in a flow diagram how many were excluded at each step due to missing data.
We wanted to balance the groups on confounders. See response above for more detail on how we matched and why we chose this method. Table 1 currently shows the baseline characteristics in the analytic cohort stratified by exposure group. We expect the distributions of the baseline characteristics for all patients included in the analysis (pooled) to align with the distributions of the baseline
characteristics of two exposure groups, which are comparable due to matching. Table 1 also includes information on missing data, which was minimal. When we matched cases to the comparison group, we included a category for missing; therefore, pregnancies with missing data were largely included but as mentioned, our missing data was less than 3% for matching factors of interest.
Table 1: See above comments, but typically the format is a comparison of standardized mean differences (SDM) before and after matching, rather than evaluation of p-values.
We have calculated the standardized mean differences, which are typically presented in propensity score matched analyses to demonstrate balance in covariates between the two exposure groups after
matching. All standardized mean differences for the matching variables are <0.10 (the typical threshold used to assess baseline balance). We have made note of this as a footnote to Table 1.
EDITORIAL OFFICE COMMENTS:
1. If your article is accepted, the journal will publish a copy of this revision letter and your point-by-point responses as supplemental digital content to the published article online. You may opt out by writing separately to the Editorial Office at [email protected], and only the revision letter will be posted.
This is fine.
2. When you submit your revised manuscript, please make the following edits to ensure your submission
contains the required information that was previously omitted for the initial double-blind peer review:
* Funding information (ie, grant numbers or industry support statements) should be disclosed on the title page and at the end of the abstract. For industry-sponsored studies, describe on the title page how the funder was or was not involved in the study.
* Include clinical trial registration numbers, PROSPERO registration numbers, or URLs at the end of the abstract (if applicable).
* Name the IRB or Ethics Committee institution in the Methods section (if applicable).
* Add any information about the specific location of the study (ie, city, state, or country), if necessary for context.
This has been addressed.
3. Obstetrics & Gynecology's Copyright Transfer Agreement (CTA) must be completed by all authors.
When you uploaded your manuscript, each coauthor received an email with the subject, "Please verify your authorship for a submission to Obstetrics & Gynecology." Please ask your coauthor(s) to complete this form, and confirm the disclosures listed in their CTA are included on the manuscript's title page. If they did not receive the email, they should check their spam/junk folder. Requests to resend the CTA may be sent to [email protected].
This is done.
4. For studies that report on the topic of race or include it as a variable, authors must provide an explanation in the manuscript of who classified individuals' race, ethnicity, or both, the classifications used, and whether the options were defined by the investigator or the participant. In addition, describe the reasons that race and ethnicity were assessed in the Methods section and/or in table footnotes.
Race and ethnicity must have been collected in a formal or validated way. If it was not, it should be omitted. Authors must enumerate all missing data regarding race and ethnicity as in some cases missing data may comprise a high enough proportion that it compromises statistical precision and bias of analyses by race.
Use "Black" and "White" (capitalized) when used to refer to racial categories.
List racial and ethnic categories in tables in alphabetic order. Do not use "Other" as a category; use
"None of the above" instead.
This is done.
5. ACOG uses person-first language. Please review your submission to make sure to center the person before anything else. Examples include: "People with disabilities" or "women with disabilities" instead of
"disabled people" or "disabled women"; "patients with HIV" or "women with HIV" instead of "HIV- positive patients" or "HIV-positive women"; and "people who are blind" or "women who are blind"
instead of "blind people" or "blind women."
This is done.
6. The journal follows ACOG's Statement of Policy on Inclusive Language
(https://urldefense.com/v3/ https://www.acog.org/clinical-information/policy-and-position-
statements/statements-of-policy/2022/inclusive-language ;!!BZ50a36bapWJ!oPRfN0W65-HX2nTBAt-
E6qiTHGoYbyygNQva6527VR6 cEwTRw5Ptqg6CuykJyCJgmjrgHlG-zqiHHdehJP 7Pg$ ). When possible,
please avoid using gendered descriptors in your manuscript. Instead of "women" and "females,"
consider using the following: "individuals;" "patients;" "participants;" "people" (not "persons"); "women and transgender men;" "women and gender-expansive patients;" or "women and all those seeking gynecologic care.
This is done.
7. Responsible reporting of research studies, which includes a complete, transparent, accurate and timely account of what was done and what was found during a research study, is an integral part of good research and publication practice and not an optional extra. Obstetrics & Gynecology supports initiatives aimed at improving the reporting of health research, and we ask authors to follow specific guidelines:
STROBE: observational studies
Include the appropriate checklist for your manuscript type upon submission, if applicable, and indicate in your cover letter which guideline you have followed. Please write or insert the page numbers where each item appears in the margin of the checklist. Further information and links to the checklists are available at https://urldefense.com/v3/ http://www.equator-
network.org/ ;!!BZ50a36bapWJ!oPRfN0W65-HX2nTBAt-
E6qiTHGoYbyygNQva6527VR6 cEwTRw5Ptqg6CuykJyCJgmjrgHlG-zqiHHded33zz4s$ .
This is done.
8. Standard obstetric and gynecology data definitions have been developed through the reVITALize initiative, which was convened by the American College of Obstetricians and Gynecologists and the members of the Women's Health Registry Alliance. Obstetrics & Gynecology has adopted the use of the reVITALize definitions. Please access the obstetric data definitions at
https://urldefense.com/v3/ https://www.acog.org/practice-management/health-it-and-clinical- informatics/revitalize-obstetrics-data-definitions ;!!BZ50a36bapWJ!oPRfN0W65-HX2nTBAt- E6qiTHGoYbyygNQva6527VR6 cEwTRw5Ptqg6CuykJyCJgmjrgHlG-zqiHHde4DUukoI$ and the gynecology data definitions at https://urldefense.com/v3/ https://www.acog.org/practice- management/health-it-and-clinical-informatics/revitalize-gynecology-data-
definitions ;!!BZ50a36bapWJ!oPRfN0W65-HX2nTBAt-
E6qiTHGoYbyygNQva6527VR6 cEwTRw5Ptqg6CuykJyCJgmjrgHlG-zqiHHdeIZ9HhfI$ . If use of the reVITALize definitions is problematic, please discuss this in your point-by-point response to this letter.
We’ve reviewed the definitions.
9. Make sure your manuscript meets the following word limit. The word limit includes the manuscript body text only (for example, the Introduction through the Discussion in Original Research manuscripts), and excludes the title page, précis, abstract, tables, boxes, and figure legends, reference list, and supplemental digital content. Figures are not included in the word count.
Original Research: 3,000 words
10. Specific rules govern the use of acknowledgments in the journal. Please review the following
guidelines and edit your title page as needed:
* All financial support of the study must be acknowledged.
* Any and all manuscript preparation assistance, including but not limited to topic development, data collection, analysis, writing, or editorial assistance, must be disclosed in the acknowledgments.
Such acknowledgments must identify the entities that provided and paid for this assistance, whether directly or indirectly.
* All persons who contributed to the work reported in the manuscript, but not sufficiently to be authors, must be acknowledged. Written permission must be obtained from all individuals named in the acknowledgments, as readers may infer their endorsement of the data and conclusions. Please note that your response in the journal's electronic author form verifies that permission has been obtained from all named persons.
* If all or part of the paper was presented at the Annual Clinical and Scientific Meeting of the American College of Obstetricians and Gynecologists or at any other organizational meeting, that presentation should be noted (include the exact dates and location of the meeting or indicate whether the meeting was held virtually).
* If your manuscript was uploaded to a preprint server prior to submitting your manuscript to Obstetrics & Gynecology, add the following statement to your title page: "Before submission to Obstetrics & Gynecology, this article was posted to a preprint server at: [URL]."
* Do not use only authors' initials in the acknowledgement or Financial Disclosure; spell out their names the way they appear in the byline.
This is done.
11. Provide a short title of no more than 45 characters, including spaces, for use as a running foot. Do not start the running title with an abbreviation.
This is done.
12. Be sure that each statement and any data in the abstract are also stated in the body of your manuscript, tables, or figures. Statements and data that appear in the abstract must also appear in the body text for consistency. Make sure there are no inconsistencies between the abstract and the manuscript, and that the abstract has a clear conclusion statement based on the results found in the manuscript.
In addition, the abstract length should follow journal guidelines. Please provide a word count.
Original Research: 300 words
This is done. The abstract word count is 289.
13. Only standard abbreviations and acronyms are allowed. A selected list is available online at
https://urldefense.com/v3/ http://edmgr.ovid.com/ong/accounts/abbreviations.pdf ;!!BZ50a36bap WJ!oPRfN0W65-HX2nTBAt-E6qiTHGoYbyygNQva6527VR6 cEwTRw5Ptqg6CuykJyCJgmjrgHlG-
zqiHHdeeyNFAOE$ . Abbreviations and acronyms cannot be used in the title or précis. Abbreviations and acronyms must be spelled out the first time they are used in the abstract and again in the body of the manuscript.
This is done.
14. The journal does not use the virgule symbol (/) in sentences with words, except with ratios. Please rephrase your text to avoid using "and/or," or similar constructions throughout the text. You may retain this symbol if you are using it to express data or a measurement.
This is done.
15. In your abstract, manuscript Results sections, and tables, the preferred citation should be in terms of an effect size, such as odds ratio or relative risk or the mean difference of a variable between two groups, expressed with appropriate confidence intervals. When such syntax is used, the P value has only secondary importance and often can be omitted or noted as footnotes in a Table format. Putting the results in the form of an effect size makes the result of the statistical test more clinically relevant and gives better context than citing P values alone.
Please standardize the presentation of your data throughout the manuscript submission. For P values, do not exceed three decimal places (for example, "P = .001").
Express all percentages to one decimal place (for example, 11.1%"). Do not use whole numbers for percentages.
This is done.
16. Please review the journal's Table Checklist to make sure that your tables conform to journal style.
The Table Checklist is available at
https://urldefense.com/v3/ http://edmgr.ovid.com/ong/accounts/table checklist.pdf ;!!BZ50a36ba pWJ!oPRfN0W65-HX2nTBAt-E6qiTHGoYbyygNQva6527VR6 cEwTRw5Ptqg6CuykJyCJgmjrgHlG-
zqiHHdeLy1mFNk$ .
This is done.
17. Please review examples of our current reference style at
https://urldefense.com/v3/ https://edmgr.ovid.com/ong/accounts/ifa suppl refstyle.pdf ;!!BZ50a3 6bapWJ!oPRfN0W65-HX2nTBAt-E6qiTHGoYbyygNQva6527VR6 cEwTRw5Ptqg6CuykJyCJgmjrgHlG- zqiHHde-vK9wc0$ . Include the digital object identifier (DOI) with any journal article references and an accessed date with website references.
Unpublished data, in-press items, personal communications, letters to the editor, theses, package inserts, submissions, meeting presentations, and abstracts may be included in the text but not in the formal reference list. Please cite them on the line in parentheses.
If you cite ACOG documents in your manuscript, be sure the references you are citing are still current and available. Check the Clinical Guidance page at
https://urldefense.com/v3/ https://www.acog.org/clinical ;!!BZ50a36bapWJ!oPRfN0W65-
HX2nTBAt-E6qiTHGoYbyygNQva6527VR6 cEwTRw5Ptqg6CuykJyCJgmjrgHlG-zqiHHde8lLcbu4$ (click on
"Clinical Guidance" at the top). If the reference is still available on the site and isn't listed as
"Withdrawn," it's still a current document. In most cases, if an ACOG document has been withdrawn, it should not be referenced in your manuscript.
Please make sure your references are numbered in order of appearance in the text.
This is done.
18. Authors whose manuscripts have been accepted for publication have the option to pay an article processing charge and publish open access. With this choice, articles are made freely available online immediately upon publication. An information sheet is available at
https://urldefense.com/v3/ http://links.lww.com/LWW-ES/A48 ;!!BZ50a36bapWJ!oPRfN0W65- HX2nTBAt-E6qiTHGoYbyygNQva6527VR6 cEwTRw5Ptqg6CuykJyCJgmjrgHlG-zqiHHde-j6kkXQ$ . The cost for publishing an article as open access can be found at
https://urldefense.com/v3/ https://wkauthorservices.editage.com/open- access/hybrid.html ;!!BZ50a36bapWJ!oPRfN0W65-HX2nTBAt-
E6qiTHGoYbyygNQva6527VR6 cEwTRw5Ptqg6CuykJyCJgmjrgHlG-zqiHHdeYAjjp9s$ .
If your article is accepted, you will receive an email from the Editorial Office asking you to choose a publication route (traditional or open access). Please keep an eye out for that future email and be sure to respond to it promptly.
This is fine.
Thank you for your consideration.
Sincerely, Marie Boller Fei Xu
Catherine Lee
Sneha Sridhar
Mara Greenberg
Monique Hedderson
