PEER REVIEW HISTORY BMJ Open publishes all

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) AUTHORS

Concurrent drug use among methadone maintenance patients in mountainous areas in Northern Vietnam: a cross-sectional study Xuan Bach, Tran; Boggiano, Victoria; Nguyen, Lan Huong; Nguyen, Long; Nguyen, Hung; Hoang, Canh; Thi Huong, Le; Tho, Tran; Le, Hai; Latkin, Carl; Vu, Minh Thuc

VERSION 1 – REVIEW

REVIEWER

REVIEW RETURNED GENERAL COMMENTS

Adam Searby Alfred Health, Melbourne, Australia RMIT University, Melbourne, Australia 18-Jan-2017 Thank you for the opportunity to review this manuscript, which provides good insight into methadone maintenance treatment in Vietnam. I would recommend that the manuscript undergoes review for both spelling and grammar, as there are a number of spelling and grammatical errors. In addition, some sentences are written in a clumsy fashion and would benefit from revision to strengthen the paper (an example is page 8, line 7). The study methodology seems sound, however I would like to see some expansion of the consent process, particularly given the study was conducted in the health service providing methadone maintenance itself. How were the participants approached? Were there ramifications for disclosure of illicit drug use? Given that many methadone maintenance programs exclude participants for illicit substance use, I feel this is an important aspect to describe. It may also go some way to explaining the rate of illicit substance use in comparison to other studies conducted elsewhere in the world. In regards to limitations, I feel that there should be some discussion of the reliability of self-report in this study. I understand there is much literature discussing self report, however is this reliable in the context of Vietnamese methadone treatment programs? Again, I feel this limitation could be addressed with a more thorough discussion of the consent process and the potential ramifications of refusal. Overall, this is a good study that provides some interesting discussion of methadone maintenance treatment in Vietnam. Although conducted in a single service, and therefore making generalisation difficult (as addressed in the limitations), the paper provides some valid discussion and conclusions. The conclusion indicates this, however should be reworded to reinforce these limitations and perhaps highlight that future research is needed to draw the conclusion that MMT be applied across the nation.

REVIEWER REVIEW RETURNED GENERAL COMMENTS

Tran, BX School of Public Health, University of Alberta, Canada 28-Mar-2017 Thank editors for inviting me to review a very interesting paper that can address the emerging issue in planning MMT program in Vietnam. Please find below my comments for the manuscript: General comments - This is an important study since it is very few studies examining the impact of MMT in mountainous areas, and people are questioning about its long-term impacts especially in difficult settings. - Grammar and typo errors should be addressed in the revised manuscript. Specific comments: - The title seems to not match the research question. Authors should clarify it - Given the primary focus of the paper is looking at concurrent drug use, I'd like to see more background details as to the current status of these domains in the context of Vietnam and the limitations. - How were variables chosen to be included in the initial model? Variables chosen to be included in the model are very limited in nature. I'm afraid that there are few of the many important factors that are missing in the model, which will have a significant impact on the overall finding. Few of the significant correlates of concurrent drug use found among PWUD in common drug treatment (i.e., MMP) are interrelationship, stigma, and AUD. Were any of these variables collected? Because these may significantly impact the overall findings. Also, if the participants are a mix of people with both status (HIV), how will the result differ between these two groups? How about methadone dose? Were these variables controlled for? -How was missing data handled? - In the results section, the text reports many numbers that are already reported within table. These redundancies should be eliminated. - What is the difference between this study and other studies conducted in mainland area? Is there any specific implication for patients in mountainous areas?

REVIEWER REVIEW RETURNED GENERAL COMMENTS

David C Marsh MD CCSAM Northern Ontario School of Medicine, Canada 16-May-2017 This is a well designed and written report of a cross-sectional analysis of methadone maintenance patients in a novel setting. The methods and results are well described and support the conclusions drawn. While the paper explicitly states that written informed consent was obtained from participants, it does not state whether the study was reviewed and approved by a Research Ethic Board. I have

suggested minor editorial changes to clarify some of the language used through the attached PDF. In addition, the authors should check the difference between Tables 1 and 3 in the way marital status data is described. I suspect the labels are reversed for living with partner and widowed in Table 1. Finally the description of "seduced by friend" might be misunderstood as a reason for relapse. Unless there is a sexual interaction intended to be described here, I suggest the authors revise the language to reflect patients having been encouraged by friends to use drugs. The reviewer also provided a marked copy with additional comments. Please contact the publisher for full details.

REVIEWER

REVIEW RETURNED

Xun Zhuang Nantong University China 23-Jun-2017

GENERAL COMMENTS

1,We can see that the Absract is divided into 7 parts（Objectives、 Setting、Participants、Primary and secondary outcome measures: 、Results、Conclusion、Strengths and limitations of this study） .We recommended that the Abstract should be divided into 4 parts(Objectives、Methods、Results、Conclusion). 2,P4L20 “wherein all eligible clients were invited to enroll” is wrong. The writer should change “wherein” into “where in”. 3, The particants is only included in male. The passage is lacking the information of female , so the passage lacks representation. 4,We can see from the passage that“The missing data were ignored when analyzing the data.” We recommend that the missing date can be delt with a more scientific method such as Missing value added. 5,In Table 1 and Table 3 we recommend that the Age group should be divided into ＜35,35-,≥50. 6,In Table 1，Table2 ，Table3,we recommend that the writer should add the total number of each column. 7、We recommend that in Table 2 the writer should do Chi-square test. 8、In Table4 we recommend that the writer should add the basic data for each item（absolute number）. 9、In the part of References，we recommend that the writer should use standard format，for example some references should limit the number of writers name，such as 17-21,23-25.

REVIEWER

Szu-Hsien Lee National Taiwan Normal University, Taiwan 23-Jul-2017

REVIEW RETURNED GENERAL COMMENTS

This article intended to investigate the association of concurrent drug use and mental health in patients in two MMT clinics using a cross-sectional survey. The topic is important since heroin users may use heroin and have mental problems while they are on MMT. However, there is a similar paper to this study used the same survey data recently accepted by the Substance Abuse Treatment, Prevention and Policy although the research question is a bit

different. Additionally, there are some suggestions for the authors. 1. In Abstract, the 3rd line of Objectives, ‘elicit’ should be ‘illicit’. 2. Please provide the IRB approval number if there is any. 3. Please provide more detailed information and reasons for logistic and tobit regressions because Tobit regression may not be familiar to readers. Please also specify what variables needed to be censored? 4. Some words need to be revised. Page 5 line 48 under the section of Statistical Analysis, T-test should be used to examine ‘continuous variable’ and chi-square should be used for ‘categorical variable’. It is not appropriate to say that chi-square is used for qualitative variables. 5. Please describe more precisely regarding missing data? What does it mean to ignore the missing data? (Page 5, line 54) 6. For Table 2, it is suggested to combine some small numbers in a cell so that chi-squares can be meaningfully performed. 7. Did this study collect patients’ daily methadone dosages? 8. For Table 4, it is suggested that *p