Alcohol consumption patterns among men who have sex with men in major cities of Myanmar: A cross-sectional study

Background: Alcohol consumption patterns vary widely across the regions of the world. Although previous studies have focused on the sexual risk behaviours among men who have sex with men (MSM), studies regarding binge alcohol drinking among MSM in Myanmar are scarce. Methods: A cross-sectional study was conducted to identify the alcohol consumption patterns among MSM aged over 18 years in two major cities of Myanmar where the MSM population is higher than other regions. Purposive sampling was applied and sampling was made through Myanmar MSM network. Face-to-face interviews were conducted using a structured questionnaire. Patterns of alcohol consumption were described as frequency/percentage and mean/median as appropriate. Bivariate analysis was also done to find out the association between types of MSM and binge drinking. Results: A total of 256 MSM included in the study (mean age, 27.33±7.7 years). Of 256 participants, 225 MSM had the experience of alcohol consumption in their lifetime (225/256, 87.9%). Among ever drinkers, 152 MSM consumed alcohol within three months (152/225, 67.6%). Regarding beer consumption, the highest proportions of MSM from both groups (42.8%, 36.8%) consumed 1-3 times per week. Overall, 57.2% of young MSM and 41.2% of adult MSM consumed beer together with their friends. Nearly 34% of young MSM and nearly 38% of adult MMS consumed beer at gatherings of friends. At different time periods, higher proportions of Thange (partners of MSM) had experienced of binge drinking than apwint (open) and apone (hidden) (p<0.05). Conclusions: The current study identified the alcohol consumption patterns in terms of type, amount, frequency at different time periods among MSM in major cities of Myanmar. It is suggested to develop and implement alcohol control policy for MSM since the proportion of current drinkers as well as binge drinking higher among these groups.


Introduction
Globally, alcohol is widely consumed as a beverage and for recreation and socialization. Alcohol consumption patterns vary widely across the regions, ranging from daily heavy drinking to occasional hazardous drinking (Ennett et al., 2016;World_Health_Organization, 2018). Excessive alcohol use and chronic alcohol binging are associated with high morbidity and mortality (Stockings et al., 2016). About 40% of global population aged over 15 years had consumed alcohol while 2.3 billion of them are current drinkers. Worldwide, 44.8% of total recorded alcohol is consumed in the form of spirits. The second most consumed type of beverage is beer (34.3%) followed by wine (11.7%) (World_Health_Organization, 2018).
Previous studies have highlighted that hazardous alcohol drinking was common among men who have sex with men (MSM), ranging from incidences of 14% to 52%, and it was also associated with unsafe sexual behaviours ( Furthermore, a significant association was noted between hazardous alcohol drinking and sexually transmitted infections, including HIV. According to a study conducted in China, 14.4% of MSM reported hazardous drinking and 16.8% reported binge drinking. Hazardous or binge drinkers were associated with various risky sexual behaviours such as have multiple partnerships, pay for sex, and have condomless insertive anal intercourse (Liu et al., 2016). In a study in Peru, 45% of MSM and transgender women had an alcohol use disorder. Higher incidence of condomless anal intercourse was seen among participants with alcohol use disorder (AUD). However, AUD positivity was not associated with either condomless anal intercourse or recent STI/HIV infection (Herrera et al., 2016). Another study in US has documented the heavy alcohol use patterns and correlates in a diverse sample of MSM using respondentdriven sampling. It was found that prevalence of RDS adjusted weekly drinking was 24.9% and weekly binge drinking was 19.3%. Independent correlates of hazardous alcohol consumption were identified as being moderately or extremely interested in reducing alcohol use; ever receiving alcohol treatment; using ecstasy; reporting syphilis diagnosis; and having more than five male partners (Santos et al., 2018).
In Myanmar, HIV is concentrated among key affected populations like MSM, whereas HIV prevalence was over 10% among them (UNAIDS, 2015). Alcohol drinking was common among MSM, as it was with other men. Although previous studies have focused on risky sexual behaviours among MSM, few studies assessed hazardous alcohol drinking among MSM. According to a previous study in Myanmar, there was 50% of lifetime drinkers and 20% of current drinker among general men (Oo et al., 2015). Similarly, in Myanmar, we know of no studies to have reported on alcohol consumption among MSM. Therefore, current study was conducted to identify the alcohol consumption patterns and binge drinking at different time periods among men who have sex with men in Myanmar.

Study design, population and area
A cross-sectional study was conducted among MSM those aged more than 18 years in Yangon and Mandalay, major cities of Myanmar where MSM population is higher than other regions during June and July 2020. We have revised this version according to the expert opinions of the reviewers: • The prevalence of alcohol drinking among men in Myanmar has been described in the Introduction.
• We have listed other places where recruitment was carried out in the Data collection section.
• We have updated the spelling of "Tha Nge" to "Thange" to be consistent with the terms recognised by the National AIDS Program.
• "Whisky" was added as an alcohol beverage to the Operational definitions section.
• Statistical analysis statement was added to the Data management and analysis section.
• We have updated our conclusion in the Discussion and recommendation section.

REVISED
Sampling and sample size Purposive sampling was applied and sampling of the MSMs was made through the Myanmar MSM network. Identification of the places for recruitment of the possible participants was made after discussion with the focal persons from the networks of MSM. There might be bias resulting from applying purposive sampling; however, attempts to reduce the bias were made by providing thorough explanations to the focal person to recruit different type of MSM from different sources, such as drop-in-centre and beauty parlours.
By considering the estimated proportion of MSM who are current drinkers as 20% (Oo et al., 2015), to achieve a 95% confidence level and an error of 5%, the minimum required sample size becomes 246 (Wayne, 1995).

Data collection
Firstly, a structured questionnaire was developed in English by reviewing the literature (see Extended data; Htut et al., 2020b). Then, translation was done into Myanmar language and back translation was carried out into English by a translator who was expert in both languages and had experience of translation regarding questionnaire used in MSM related research. Training of the interviewers was done at Department of Medical Research and pre-test was done at a non-study township in Yangon Region. After receiving ethical approval, at the venues where MSMs usually gather like drop-in-centers of International Non-governmental Organizations, beauty parlors, famous beautician, moat and office of community based organization eligible participants were contacted and invited to participate in the study. After getting the informed consent, data collection was done by face-to-face interview. Strict adherence to ethical principles were ensured throughout the data collection period in order to maintain the confidentiality of the information of the study participants.

Operational definitions
According to the local terminology in Myanmar, three groups of MSM were included in the study: Apwint or open type, Apone or hidden type and Thange (National_AIDS_Program, 2019). Apwint or open MSM are defined as individuals born biological male but who openly express themselves femininely by dress and/or social interactions. Apone or hidden MSM are defined as individuals born biological males who may also want to express themselves femininely but may not disclose this behaviour to all segments of their social networks. Tha Nge are defined as having a masculine outward appearance but have sex with men".
Types of alcohol beverages consumed by MSM included Beer, Wine, Whisky and Rum.

Data management and analysis
Data entry was carried out with EpiData version 3.1 and data analysis was done with SPSS version 21. Exploratory data analysis was done to check the errors, consistencies and missing values. The number of standard drinks was calculated by volume of container in liters multiplied by the percentage of alcohol volume multiplied by 0.789 (the specific gravity volume of ethyl alcohol). Binge drinking was defined as five or more standard drinks for men in a sitting or within two hours. Descriptive statistics were shown according to the data obtained from the assessment. Patterns of alcohol consumption were described as frequency/percentage and mean/median as appropriate. Bivariate analysis using chi-squared was also done to find out the association between types of MSM and binge drinking. Level of statistical significance was set as p value of <0.05.

Ethical considerations
The proposal was submitted to Institutional Review Board, Department of Medical Research, Myanmar (Ethics/DMR/2020/036). Written informed consent was taken from the participants after thorough explanation about the objectives of the study.
Confidentiality and anonymity of the information were strictly ensured. All answer sheets and data reports were kept in locked cabinet.         There was a limitation in the present study that should be acknowledged. Alcohol consumption patterns were self-reported and their behaviours could not be validated with other methods, such as observation. However, we tried to overcome this limitation by carefully explaining the objectives of the study to allow participants to answer with accurate responses.

Results
The development and implementation of an alcohol control policy for MSM should be considered, since over half of them were current drinkers (within one month). Between one-third and three-quarters of them had binge drinking at different time periods, which could lead to adverse health and social consequences. Emphasis should be done more on the partners of MSM as higher proportion of them had practice of binge drinking. It is good idea to study alcohol consumption which is very popular among men who have sex with men (MSM) but it is not clearly described why the authors want to choose target population of this group as well as aim of this study. We suggest that the prevalence of alcohol drinking among men in Myanmar should be stated in general so the prevalence in this group will be higher or not can be determined.

1.
The study design is suitable and sample size is adequate. We recommend that age range should be included in abstract. (18-57 yrs) We aware that the authors used 15-24 yrs in (Table -1 ) but young MSM should be 18-24yrs (Table -1) because the study was done over 18 yrs of age in this study.

2.
The authors tried to explore the data as much as possible but it is difficult to get exact data of alcohol consumption among MSM. Out of 256 participants, 225 participants consumed alcohol in their lifetime, 89.7%, but the authors did not specify about others who were more than 18 yrs of age or not. (Never drinking alcohol in their lifetime?) These data only indicates alcohol drinking but no impact. We recommend that if the authors can correlate the data with some points like the prevalence of sexually transmitted disease among them, it will be more scientific and reproducible. We also doubtful about the lowest salary among MSM is only 1,5000 MMK per month. If the salary is too low, it will affect the alcohol consumption.

3.
Justify for using bivariate analysis of Binge drinking according to the category of MSM at different time periods. It is not relevant with Table -4 results in discussion: "over one third to nearly two third among different types of MSM reported binge drinking within three months".

4.
As we mentioned above, it is difficult to guarantee for full reproducibility. It will need to be more scientific study and outcome for reliable data. But, we understood that there would be some constraints because of COVID-19 pandemic.

5.
We suggest that there should be some outcome for conclusion for alcohol consumption among MSM. After describing the results, the authors could not discuss properly about binge drinking among different category of MSM in different time periods and it does not highlight how different time periods of binge drinking have an influence on different category of MSM. So it was difficult to get conclusion from the results. It is not a good conclusion. 6.

Is the work clearly and accurately presented and does it cite the current literature? Partly
Is the study design appropriate and is the work technically sound? Yes