Canadian Educational Resources About Cannabis Use and Fertility

BMJ Open: first published as 10.1136/bmjopen-2020-045006 on 5 February 2021. Downloaded from

PEER REVIEW HISTORY

BMJ Open publishes all reviews undertaken for accepted manuscripts. Reviewers are asked to complete a checklist review form (http://bmjopen.bmj.com/site/about/resources/checklist.pdf) and are provided with free text boxes to elaborate on their assessment. These free text comments are reproduced below.

ARTICLE DETAILS

TITLE (PROVISIONAL) Canadian educational resources about cannabis use and fertility, pregnancy and breastfeeding: a scoping review protocol AUTHORS Bombay, Kira; Murphy, Malia; Denize, Kathryn; Sharif, Ayni; Sikora, Lindsey; Murray, Rebecca; Cobey, Kelly; Corsi, Daniel J.

VERSION 1 – REVIEW

REVIEWER Jon-Patrick Allem USC, USA REVIEW RETURNED 10-Oct-2020

GENERAL COMMENTS This study described the approach of a scoping review to identify and characterize all publicly available online educational resources and clinical guidelines providing information to the Canadian public and healthcare providers on the short- and long-term effects of cannabis use on fertility, pregnancy, and breastfeeding.

Could the authors please explain why searches will be limited to 1 minute? I didn’t understand the time constraint in this section of the study. http://bmjopen.bmj.com/

Could the authors please explain why this specific bibliometric information will be extracted? Some of this information at first glance did not seem relevant to answering the research question.

I’d like to see more information on the content analysis section. How will reliability be determined? How will saturation be determined?

How will dominant themes be determined? Why use an inductive on September 29, 2021 by guest. Protected copyright. approach? Did the authors check to see if there was a theory or framework or prior literature to pull from that may help in this area?

I’m concerned over the fact that social media was completely ignored. Why was this the case? The public often turns to social media platforms to engage in health-related conversations which includes seeking out information on cannabis. What the public comes across may be dubious (e.g., unsubstantiated health claims regarding cannabis) and impact public health See: https://ajph.aphapublications.org/doi/abs/10.2105/AJPH.2019.305461 This should at least be mentioned as an area of future research.

REVIEWER Melissa Sharp Royal College of Surgeons in Ireland REVIEW RETURNED 28-Oct-2020

GENERAL COMMENTS I commend the authors for their well thought out approach to BMJ Open: first published as 10.1136/bmjopen-2020-045006 on 5 February 2021. Downloaded from

mapping the available Canadian educational resources about cannabis use and fertility. The protocol appears to be well thought- out, thorough, and well written. I commend their registration on OSF, use of PRESS, and their proposed use of reporting guidelines for their results (PRISMA-ScR). I have a few clarification questions and suggestions below which I hope to offer to perhaps increase readability from the reader’s perspective.

Firstly, the conceptualization of ‘fertility’ is not clear throughout the paper. Until reading the supplemental file, I was not clear if fertility was encompassing of both male and female fertility, especially as it does not appear in the search strategy. It was only until I noticed ‘male parent health/sperm count’ in the supplemental materials that it was made abundantly clear that this was broader than initially assumed (it’s normally juxtaposed to pregnancy and breastfeeding). I would recommend rewording and clarifying this concept throughout the text, especially as the time frame for these effects may be different than for breastfeeding and during pregnancy.

Lastly, as a former neighbor of Ontario, I am keenly aware of its provincial status but as this is BMJ Open, a global journal, I would recommend explicitly stating this (page 4, line 10) prior to the statement ‘these findings are similar to those from other *number* Canadian provinces’ (page 4, line 14).

On page 5, the paragraph is quite lengthy and a bit jumbled. I would recommend splitting this and perhaps reorganizing a few sentences in the beginning. Seeking info online --> questionable quality --> seeking sources elsewhere/anecdotal advice --> seeking info from doctors --> paragraph about specific issues with HCP discussions

On page 6 the framework steps are outlined but the step 3 header is http://bmjopen.bmj.com/ different further along in the text. I understand the need for the difference but they should be parallel in nature. I’d suggest study (resource) selection for both to reflect both the official steps and your adaptation.

One page 6, it is also noted that the search strategy was peer- reviewed using the PRESS guideline. Who conducted this review?

Another author or an external information specialist? on September 29, 2021 by guest. Protected copyright.

On page 7, regarding the webpage reviewers, is there any manner to ensure comprehensiveness or is there a rationale behind limiting the search to one minute?

Also, it is not mentioned until page 8 that ‘all documents will be screened independently by two team members.’ This information should come much earlier as it is currently unclear how the webpages are being identified and who is conducting the cited reference search.

Under step 4, if it is not too late, I would recommend also collecting the email of the corresponding author as you are already collecting their first and last name. This could tie into your dissemination plan (page 11) as I would assume that the authors may be interested in the results of your scoping review and could also update their educational resources accordingly.

BMJ Open: first published as 10.1136/bmjopen-2020-045006 on 5 February 2021. Downloaded from

Under step 5, the analysis is extremely vague (e.g., where relevant we will provide frequencies and proportions). While you may not have decided on every single item to report at this time and which will be most important, I think that you need to more carefully identify more concrete examples of metrics to report. Otherwise, this leaves things open to cherry-picking.

On page 11, it is noted that you will ‘identify evidence gaps’ but there is no prior mentioning of evidence mapping. Is this related to the information from the content analysis? How will you identify these gaps?

Lastly, I noticed in the search strategy that item 11 was or/1-9. I am not an information specialist but I just wanted to double-check that it shouldn’t be or/1-10. Also is there a reason why fertility is not mentioned in the search strategy given that it is one of your key areas of interest?

VERSION 1 – AUTHOR RESPONSE

Reviewer: 1

Jon-Patrick Allem, USC, USA

This study described the approach of a scoping review to identify and characterize all publicly available online educational resources and clinical guidelines providing information to the Canadian public and healthcare http://bmjopen.bmj.com/ providers on the short- and long-term effects of cannabis use on fertility, pregnancy, and breastfeeding.

1. Could the authors please explain why searches will be limited to 1 minute? I didn’t understand the time constraint in this section of the study. Author Response: We thank this reviewer for their thoughtful review of the submitted protocol. The rationale

for imposing a time-limit to the website search strategy was to encourage more efficient key word searching. on September 29, 2021 by guest. Protected copyright. However, we acknowledge that a time constraint, particularly one that is so short, may not be as effective as initially planned. We have revised our strategy accordingly:

• We have clarified that the search will be conducted by one individual, and validated by a second independent reviewer. • We have expanded on the search strategy that will be used in the internal search strategy applied to each parent organization website (added in new Appendix 3).

Revised Text (“Step 2: Identifying Relevant Literature, Page 7): “Eligible records within these organizations’ webpages will be identified through internal searches of the parent organization for key search terms (online supplementary appendix 3). Two independent reviewers will complete this work (KB, AS). One reviewer will conduct searches, and a second reviewer will independently validate the searches. Identification of eligible records will be documented using the screening tool provided in the supplemental materials (online supplementary appendix 4). Eligible records identified through the grey literature search will proceed directly to data extraction.” BMJ Open: first published as 10.1136/bmjopen-2020-045006 on 5 February 2021. Downloaded from

2. Could the authors please explain why this specific bibliometric information will be extracted? Some of this information at first glance did not seem relevant to answering the research question. Author Response: The proposed bibliometric characteristics will enable rapid identification and location of the resource (title, authors) and assessment of accessibility (format of dissemination, open access etc.). The recency of publication/dissemination (publication date) is relevant to contextualize the recommendations/information provided relative to changes in Canadian federal legislation regarding recreational cannabis production, sales and purchasing. Contribution from external parties (patient partners or organizations) is relevant to: (1) assess the frequency of patient/public-partner engagement on developing public-facing resources and the potential impact of such engagement on the included resources. (2) identify joint position-statements/recommendations across Canadian organizations and identify potentially relevant organization websites that our team may not have already searched.

3. I’d like to see more information on the content analysis section. How will reliability be determined? How will saturation be determined? How will dominant themes be determined? Why use an inductive approach? Did the authors check to see if there was a theory or framework or prior literature to pull from that may help in this area? Author Response: We agree with the reviewer, that inductive content analysis is not well-aligned with the approach we have outlined in the protocol. We have revised our analysis plan to address the concerns from both peer-reviewers, and have added more detail on how our findings will be summarized and reported.

Revised text (Step 4: Charting the data, Page 10): “From the included resources, we will also

chart the following characteristics: target population (general public, healthcare providers), the language of publication (English, French, both), and the scope of the information presented on cannabis use (i.e. related to fertility, pregnancy or breastfeeding), and recommendations conveyed as it relates to cannabis use and http://bmjopen.bmj.com/ fertility, pregnancy, or breastfeeding. Accessibility of online resources (webpages, PDF documents or similar) will scored as reviewer perception of how easy it was to find the resource from the parent website homepage. Readability of online resources will be evaluated using the Simple Measure of Gobbledygook to determine the reading level.”

Revised text (Step 5: Collating, summarizing and reporting the results, Pages 10-11): “We will present on September 29, 2021 by guest. Protected copyright. our results in graphic and tabular formats, with key characteristics, concepts and themes summarized and explained in keeping with scoping review methodologies. We will collate results distinctly for resources targeting healthcare providers versus the lay public and comment on consistency in messaging between these resources. Key characteristics will include bibliometric details, level of jurisdiction of the authoring organization (local, provincial, federal), language and format of publication, use of cannabis-related terminology, readability and aesthetic features. Key concepts will include the type of cannabis use addressed (medicinal, recreational), potential health impacts/outcomes discussed (i.e., related to male or female fertility, pregnancy and the developing child, breastmilk and the breastfeeding infant). Extracted themes will focus on the recommendations made to the reader, the language used to make those recommendations (with quotes provided as examples), and evidence gaps cited by the included resources. A sub-analysis of resources specific to short-term versus longer-term (e.g., neurodevelopmental) outcomes of exposed infants will also be presented. We will synthesize dominant concepts and themes in the form of a visual figure, and consolidate the identified resources into a searchable database. We will adhere to the PRISMA-ScR checklist to report the results of this scoping review.”

BMJ Open: first published as 10.1136/bmjopen-2020-045006 on 5 February 2021. Downloaded from

4. I’m concerned over the fact that social media was completely ignored. Why was this the case? The public often turns to social media platforms to engage in health-related conversations which includes seeking out information on cannabis. What the public comes across may be dubious (e.g., unsubstantiated health claims regarding cannabis) and impact public health See: https://ajph.aphapublications.org/doi/abs/10.2105/AJPH.2019.305461 This should at least be mentioned as an area of future research. Author Response: Thank you very much for this comment. We agree with the reviewer that social media platforms are increasingly being consumed as a source of health-related information. Given social media's importance, we plan to pursue a social media analysis separately after the proposed project is underway or nearing completion.

At this reviewer’s recommendation, we have acknowledged the omission of social media in our search strategy and have highlighted it as an area of future research.

Revised text (Strengths and limitations of this study, Page 3): “The review will not include an analysis of social media content.”

Revised text (Ethics and dissemination, Pages 11-12): “Generating a complete picture of currently available Canadian resources for the lay public and healthcare providers on the short- and long-term effects of cannabis use on fertility, pregnancy and breastfeeding is a necessary first step to improving these resources’ scope and quality. The proposed scoping review will identify and synthesize Canadian clinical guidelines and educational resources related to cannabis use and fertility, pregnancy, and breastfeeding. We will summarize the strengths and weaknesses of resources developed for healthcare providers and the public, consolidate cited evidence gaps, identify possible inconsistencies in messaging and make recommendations for improvement. Although we will not consider information distributed through social media platforms, social media is a primary source of health-related information for many consumers, which will be a subject of future work. Through this work, we will generate a comprehensive list and searchable http://bmjopen.bmj.com/ repository of currently available educational resources that can be made available for anyone seeking information on the safety of cannabis use before and during pregnancy and while breastfeeding.”

Reviewer: 2

Melissa Sharp, Institution and Country: Royal College of Surgeons in Ireland on September 29, 2021 by guest. Protected copyright.

I commend the authors for their well thought out approach to mapping the available Canadian educational resources about cannabis use and fertility. The protocol appears to be well thought-out, thorough, and well written. I commend their registration on OSF, use of PRESS, and their proposed use of reporting guidelines for their results (PRISMA-ScR). I have a few clarification questions and suggestions below which I hope to offer to perhaps increase readability from the reader’s perspective.

1. Firstly, the conceptualization of ‘fertility’ is not clear throughout the paper. Until reading the supplemental file, I was not clear if fertility was encompassing of both male and female fertility, especially as it does not appear in the search strategy. It was only until I noticed ‘male parent health/sperm count’ in the supplemental materials that it was made abundantly clear that this was broader than initially assumed (it’s normally juxtaposed to pregnancy and breastfeeding). I would recommend rewording and clarifying this concept throughout the text, especially as the time frame for these effects may be different than for breastfeeding and during pregnancy. BMJ Open: first published as 10.1136/bmjopen-2020-045006 on 5 February 2021. Downloaded from

Author Response: Thank you, we acknowledge that we were not clear on how the concept of fertility would be addressed in the scoping review. We have provided clarification throughout the protocol.

Revised text (Step 2: Identifying relevant literature, Page 7): “Target literature will include clinician guidelines and recommendations, and resources designed for non-clinicians seeking information on how cannabis may affect male or female fertility, pregnancy and the developing child and breast milk and the breastfeeding infant.”

Revised text (Step 3: Literature/Resource selection, Page 8): “We will include documents that meet our Population, Concept and Context of interest (Table 1).23 Specifically, we will include resources that target Canadian clinicians or the lay public (Population) and provides recommendations, guidance, or reports on the safety or impacts of cannabis use on male and female fertility, pregnancy and the developing child and breast milk and the breastfeeding infant (Concept). Included literature must be available in either English or French and developed by Canadian organizations (obstetrical societies or networks, government organizations, and public health agencies) (Context). If the literature directs users to external webpages/resources, this will be noted, and the linked webpage/resource will be screened for eligibility for inclusion in this review.”

Revised text (Table 1): “Concept: Documents developed by Canadian organizations (obstetrical societies, networks as well as government and public health agencies). We will include English and French language records. The timeframe will not be limited from 2010 – present.”

Revised text (Appendix 7. Data Extraction): “Effects of cannabis use: Does it speak to the effect on male fertility?” http://bmjopen.bmj.com/

2. Lastly, as a former neighbor of Ontario, I am keenly aware of its provincial status but as this is BMJ Open, a global journal, I would recommend explicitly stating this (page 4, line 10) prior to the statement ‘these findings are similar to those from other *number* Canadian provinces’ (page 4, line 14). Author Response: Thank you for this thoughtful suggestion. We agree and have clarified Ontario’s provincial status accordingly. Revised text (Introduction, Page 4): “In the province of Ontario, the

prevalence of cannabis use in pregnancy rose from 1.2% in 2012 to 1.8% in 2017, a relative increase of on September 29, 2021 by guest. Protected copyright. 61%.3 These findings are similar to those from other Canadian provinces, and….”

3. On page 5, the paragraph is quite lengthy and a bit jumbled. I would recommend splitting this and perhaps reorganizing a few sentences in the beginning. Seeking info online --> questionable quality --> seeking sources elsewhere/anecdotal advice --> seeking info from doctors --> paragraph about specific issues with HCP discussions Author Response: Thank you. We have taken this opportunity to revise the last paragraph of the introduction.

Revised text (Introduction, Page 5): "...North Americans are increasingly seeking out health-related information online, before or in place of, consulting a healthcare provider.14 Indeed, women have identified internet searches and anecdotal advice from family and friends as their primary sources of information about the consequences of perinatal cannabis use.15 Guidance from popular or generic online resources may not BMJ Open: first published as 10.1136/bmjopen-2020-045006 on 5 February 2021. Downloaded from

be evidence-based; however, women have expressed dissatisfaction with the extent and quality of online information.15 Furthermore, where obstetrical care providers may focus more on legal or social consequences of cannabis use than on possible health risks when providing counselling to patients,16 patients may be dissuaded from accurately disclosing their cannabis or other drug use habits.

There is a complementary need for supports for healthcare providers to enable patient counselling about cannabis use in the perinatal period. Perinatal healthcare providers cite the lack of information regarding the consequences of cannabis use during pregnancy, and express discomfort in presenting and discussing the evidence with patients.17 This is concerning, as women may perceive a lack of discussion or specific counselling about the risks of cannabis use from healthcare providers to indicate that cannabis does not represent a significant risk to their pregnancy or their developing child.15 Identification and synthesis of guidance on perinatal cannabis use provided by Canadian organizations is warranted to help consolidate the scope of the information supplied to Canadians and lay the foundation for recommendations aimed at improving existing resources and identifying evidence gaps for future research.”

4. On page 6 the framework steps are outlined but the step 3 header is different further along in the text. I understand the need for the difference but they should be parallel in nature. I’d suggest study (resource) selection for both to reflect both the official steps and your adaptation. Author Response: Thank you for drawing this discrepancy to our attention. We have revised the steps outlined on page 6, and have aligned the headings for each section/step accordingly.

Revised text (Methods and Analysis, Page 6): “The steps outlined by these frameworks are:

1. Identifying the research question 2. Identifying relevant literature/resources

3. Literature/resource selection http://bmjopen.bmj.com/ 4. Charting the data 5. Collating, summarizing and reporting the results

Revised section headings (Methods and Analysis): Step 1: Research question; Step 2:

Identifying relevant literature; Step 3: Literature/Resource selection; Step 4: Charting the on September 29, 2021 by guest. Protected copyright. data; Step 5: Collating, summarizing and reporting the results.

5. One page 6, it is also noted that the search strategy was peer-reviewed using the PRESS guideline. Who conducted this review? Another author or an external information specialist? Author Response: Thank you, the database search strategy was developed by a trained information specialist (L. Sikora, co-author), and per reviewed by an external information specialist (Amanda Hodgson). We have used this opportunity to clarify these details in the main text of the manuscript as requested by the reviewer. Amanda Hodgson is now included in the Acknowledgements section.

Revised text (Step 2: Identifying relevant literature, Page 6): Our search strategy will be developed by a trained information specialist (LS) and peer-reviewed by an external information specialist using the PRESS guideline.

BMJ Open: first published as 10.1136/bmjopen-2020-045006 on 5 February 2021. Downloaded from

Revised text (Acknowledgements, Page 12): “The authors would like to acknowledge Amanda Hodgson, Medical Librarian Information Specialist, for her peer review of the database search strategy.”

6. On page 7, regarding the webpage reviewers, is there any manner to ensure comprehensiveness or is there a rationale behind limiting the search to one minute? Author Response: We thank this reviewer for their thoughtful review of the submitted protocol. The rationale for imposing a time-limit to the website search strategy was to encourage more efficient key word searching. However, we acknowledge that a time constraint, particularly one that is so short, may not be as effective as initially planned. We have revised our strategy accordingly:

7. Also, it is not mentioned until page 8 that ‘all documents will be screened independently by two team members.’ This information should come much earlier as it is currently unclear how the webpages are being identified and who is conducting the cited reference search. Author Response: Thank you. We have reformatted our description of the Literature/Resource http://bmjopen.bmj.com/ identification and selection to address this reviewer’s concerns.

In “Step 2”, we have clarified our description of how the webpages will be identified. In "Step 3”, we now identify our two-reviewer approach to literature/resource screening earlier in this section.

Revised Text (Step 2: Identifying relevant literature, Page 7): “...We will also perform a grey literature on September 29, 2021 by guest. Protected copyright. search of Canadian organizations known to provide guidance on pregnancy and breastfeeding. Target organizations include national and independent Canadian obstetrical and perinatal societies and networks, and federal and provincial government and public health offices that provide cannabis recommendations. We have identified relevant organizations in consultation with stakeholders in our professional networks (online supplementary appendix 2). Regional or local resources specifically recommended by project stakeholders are included. Eligible records within these organizations’ webpages will be identified through internal searches of the parent organization for key search terms (online supplementary appendix 3). Two independent reviewers will complete this work (KB, AS). One reviewer will conduct searches, and a second reviewer will independently validate the searches. Identification of eligible records will be documented using the screening tool provided in the supplemental materials (online supplementary appendix 4). Eligible records identified through the grey literature search will proceed directly to data extraction.”

Revised Text (Step 3: Literature/resource selection, Page 8): “All documents obtained will be screened independently by two reviewers (KB, AS). We will use the liberal accelerated approach to screening for titles and abstracts of records retrieved through the database search, whereby all documents in conflict will proceed to full -text screening.22 Screening questions that will be applied to Titles and Abstracts are provided BMJ Open: first published as 10.1136/bmjopen-2020-045006 on 5 February 2021. Downloaded from

in the supplemental materials (online supplementary appendix 5). The full texts of all potentially relevant documents will be retrieved and independently reviewed by two team members for eligibility using the screening questions provided in the supplemental materials (online supplementary appendix 6). The two independent reviewers will discuss any discrepancies at full-text screening until a consensus is reached. If the reviewers are unable to resolve disagreements through discussion, a third reviewer will be consulted. In cases where the reviewers are unable to retrieve the full text of an article (in the case of clinical guidelines and recommendations), efforts to obtain the full text will be made via a request to the corresponding author or an interlibrary loan placed through our local library system. If efforts to retrieve the full text fail, the document will be excluded from the scoping review.”

8. Under step 4, if it is not too late, I would recommend also collecting the email of the corresponding author as you are already collecting their first and last name. This could tie into your dissemination plan (page 11) as I would assume that the authors may be interested in the results of your scoping review and could also update their educational resources accordingly. Author Response: Thank you for this suggestion. We agree that it may prove helpful to collect the email of the corresponding author. We have added this in the main text of the protocol and addressed the change in the data extraction form provided in the Supplementary Materials.

Revised Text (Step 4: Charting the Data, Page 9): “Bibliometric details will be extracted from the included resources. For clinical guidelines and recommendations published in peer-reviewed journals, extracted bibliometric information will include: name and email of the corresponding author;…”

Revised Text (Rows 7-9, Data Extraction, Appendix 7, Supplementary Materials): “Is this resource published in an academic journal?

…. If yes, who is the corresponding author of this resource? If there is more than one corresponding author http://bmjopen.bmj.com/ listed, extract the first listed.

….If yes, what is the email of the corresponding author? If there is more than one corresponding author listed, extract the email of the first listed.”

9. Under step 5, the analysis is extremely vague (e.g., where relevant we will provide frequencies and on September 29, 2021 by guest. Protected copyright. proportions). While you may not have decided on every single item to report at this time and which will be most important, I think that you need to more carefully identify more concrete examples of metrics to report. Otherwise, this leaves things open to cherry-picking. Author Response: Thank you. We agree and apologize for not having provided more details in the original submission. We have taken this opportunity to elaborate on how we plan to summarize and report on our findings. Revised text (Step 5: Collating, summarizing and reporting the results, Page 10-11): “…We will present our results in graphic and tabular formats, with key characteristics, concepts and themes summarized and explained in keeping with scoping review methodologies. We will collate results distinctly for resources targeting healthcare providers versus the lay public and comment on consistency in messaging between these resources. Key characteristics will include bibliometric details, level of jurisdiction of the authoring organization (local, provincial, federal), language and format of publication, use of cannabis-related terminology, readability and aesthetic features. Key concepts will include the type of cannabis use addressed (medicinal, recreational), potential health impacts/outcomes discussed (i.e., related to male or female fertility, pregnancy and the developing child, breastmilk and the breastfeeding infant). Extracted themes will focus on the recommendations made to the reader, the language used to make those recommendations (with quotes provided as examples), and evidence gaps cited by the included resources. A sub-analysis of resources BMJ Open: first published as 10.1136/bmjopen-2020-045006 on 5 February 2021. Downloaded from

specific to short-term versus longer-term (e.g., neurodevelopmental) outcomes of exposed infants will also be presented. We will synthesize dominant concepts and themes in the form of a visual figure, and consolidate the identified resources into a searchable database. We will adhere to the PRISMA-ScR checklist to report the results of this scoping review…”

10. On page 11, it is noted that you will ‘identify evidence gaps’ but there is no prior mentioning of evidence mapping. Is this related to the information from the content analysis? How will you identify these gaps? Author Response: Thank you. We have expanded upon our plans for summarizing and reporting the findings from this scoping review. Evidence gaps acknowledged by the authoring organizations that are related to the safety/impact of cannabis use on our outcomes of interest will be collected at the data extraction stage of this scoping review. We have clarified this in the main text of the protocol, and have added these details to our data extraction form provided in Appendix 7.

Revised text (Step 5: Collating, summarizing and reporting the results, Pages 10-11): “…We will present our results in graphic and tabular formats, with key characteristics, concepts and themes summarized and explained in keeping with scoping review methodologies. We will collate results distinctly for resources targeting healthcare providers versus the lay public and comment on consistency in messaging between these resources. Key characteristics will include bibliometric details, level of jurisdiction of the authoring organization (local, provincial, federal), language and format of publication, use of cannabis-related terminology, readability and aesthetic features. Key concepts will include the type of cannabis use addressed (medicinal, recreational), potential health impacts/outcomes discussed (i.e., related to male or female fertility, pregnancy and the developing child, breastmilk and the breastfeeding infant). Extracted themes will focus on the recommendations made to the reader, the language used to make those recommendations (with quotes provided as examples), and evidence gaps cited by the included resources. A sub-analysis of resources specific to short-term versus longer-term (e.g., neurodevelopmental) outcomes of exposed infants will also be presented. We will synthesize dominant concepts and themes in the form of a visual figure, and consolidate

the identified resources into a searchable database. We will adhere to the PRISMA-ScR checklist to report http://bmjopen.bmj.com/ the results of this scoping review…”

Revised text (Appendix 7): Cited evidence gaps

• Does this resource identify a lack of evidence, data or information regarding the effect or safety of on September 29, 2021 by guest. Protected copyright. cannabis on male or female fertility? • Does this resource identify a lack of evidence, data or information regarding the effect or safety of cannabis on pregnancy or the developing child? • Does this resource identify a lack of evidence, data or information regarding the effect or safety of cannabis on breastmilk or the breastfeeding infant?

11. Lastly, I noticed in the search strategy that item 11 was or/1-9. I am not an information specialist but I just wanted to double-check that it shouldn’t be or/1-10. Also is there a reason why fertility is not mentioned in the search strategy given that it is one of your key areas of interest? Author Response: Thank you for drawing this to our attention. Item 11 should read or/1-10 and the omission of key words and search terms related to fertility was an oversight. We have revised the search strategy accordingly.

Revised Text (Appendix 1. Medline (OVID) search strategy): BMJ Open: first published as 10.1136/bmjopen-2020-045006 on 5 February 2021. Downloaded from

Database: Ovid MEDLINE(R) ALL <1946 to November 25, 2020> Search Strategy:

------

1 exp pregnancy/ (902511) 2 exp pregnancy trimesters/ (41219) 3 pregnan*.tw. (514508) 4 prenatal care/ (28244) 5 postnatal care/ (5769) 6 perinatal care/ (4741) 7 Pregnant Women/ (8652) 8 (prenatal or pre-natal or postnatal or post-natal or perinatal or peri-natal or antenatal or ante-natal or postpartum or post-partum or parturition or puerperium or gestation or gestating or gestate).tw. (424162) 9 exp Postpartum Period/ (65844) 10 exp Breast Feeding/ (38308) 11 Milk, Human/ (19556) 12 (breastfeeding* or breast feeding* or lactation or lactating or breast milk or breastmilk or human milk or chest milk or chestmilk).tw. (99853) 13 exp Fertility/ (41401) 14 exp Fertility Agents/ (29614) 15 Fertility Preservation/ (2622) 16 Involuntary Fertility Control/ (1) 17 exp Infertility/ (66482) 18 exp Fertilization/ (23578) 19 (fertil* or infertil* or fecund* or subfecund* or steril* or subfertilit* or sub-fertilit*).tw. (307050) 20 (fertiliz* or conception* or conceiv*).tw. (142457) 21 or/1-20 (1564925) 22 Cannabis/ (9378) 23 exp "Marijuana Use"/ (5897)

24 exp cannabinoids/ or Medical marijuana/ or Marijuana abuse/ (20781) http://bmjopen.bmj.com/ 25 (cannabi* or marijuana* or marihuana* or bhang or bhangs or ganja or ganjas or hashish or hashishs or hash or hemp or hemps or ganjah or cannador or charas or kush or weed or dronabinol or cannabichromene or THC or CBD or nabilone or tetrahydrocannabivarin or ganka or marinol or cesamet or dexanabinol or sativex or "HU211" or "HU 211" or indica or syndros or cannabinoid or cannabinoids or cannabidiol or nabiximols or namisol).tw. (70279) 26 ((blunt or blunts) adj2 smok*).tw. (51) 27 or/22-26 (74600) 28 exp Canada/ (160911) on September 29, 2021 by guest. Protected copyright. 29 (canada or alberta or british columbia or yukon or northwest territories or nunavut or manitoba or saskatchewan or ontario or quebec or newfoundland or prince edward island or nova scotia or new brunswick).tw. (126209) 30 Canada.cp. (284097) 31 28 or 29 or 30 (443861) BMJ Open: first published as 10.1136/bmjopen-2020-045006 on 5 February 2021. Downloaded from 32 21 and 27 and 31 (94) 33 limit 32 to yr="2010 -Current" (61)”

***************************

VERSION 2 – REVIEW

REVIEWER Jon-Patrick Allem USC and USA REVIEW RETURNED 02-Dec-2020

GENERAL COMMENTS The authors did a fine job of responding to my initial review.

REVIEWER Melissa Sharp Royal College of Surgeons in Ireland, Ireland REVIEW RETURNED 20-Dec-2020

GENERAL COMMENTS Thank you for your detailed and thorough responses to the feedback from the other reviewer, myself, and the editor. I am satisfied with the changes made. Thank you for the clarifications and additional information.