Influence of alcohol on condom use pattern during non-spousal sexual encounter in male migrant workers in north IndiaSA Rizwan, S Kant, K Goswami, SK Rai, P Misra
Centre for Community Medicine, All India Institute of Medical Sciences, New Delhi, India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0022-3859.138752
Source of Support: None, Conflict of Interest: None
Context: Migrant workers constitute an important risk group for Human Immunodeficiency Virus/Acquired Immuno-Deficiency Syndrome transmission in India. Alcohol consumption before sexual intercourse has been postulated to influence condom use practices. This study aimed to assess this association with regard to non-spousal sexual encounters among male migrant workers in northern India. Materials and Methods: A cross-sectional facility-based survey was conducted in 2011. Male migrant workers aged ≥18 years, who were born outside Haryana, who had moved to the current location after 15 years of age,had worked in the current factory for at least 1 year, who were willing to participate and were able to give written, informed consent were included in the study. A consecutive sampling was performed. Descriptive, bivariate and multiple logistic regression analyses were carried out. Results: A total of 162 participants reported having experienced non-spousal sexual encounters in the last 1 year. The proportion of men who reported not having used a condom at their last non-spousal sexual encounter was 59.3%, and 78.4% of the men reported having consumed alcohol in the last 1 year. About 48.1% of men reported having consumed alcohol before their last non-spousal sexual encounter. Men who consumed alcohol were three times more likely to not use a condom at their last non-spousal sexual encounter (OR = 3.1, 95% CI: 1.5-6.4). This association persisted even after adjusting for relevant confounders. Conclusion: Alcohol consumption had a negative influence on condom use during non-spousal sexual encounter among male migrant workers. An integrated approach to promote condom use and reduce alcohol consumption among migrant men needs to be undertaken through targeted intervention strategies.
Keywords: Alcohol, condom use, HIV/AIDS risk behavior, India, male migrants, non-spousal sex
In India, the Human Immunodeficiency Virus/Acquired Immune Deficiency Syndrome (HIV/AIDS) epidemic has reached a stable trend among the general population, but the infection is still highly prevalent among core risk groups, of which male migrant workers form an important component.  There are several definitions of migrants considered in literature. According to the National AIDS Control Programme (NACO), migrants are defined as those who move from their place of origin in rural areas (source) to a town or city (destination) for work purposes.  However, several studies generally consider their own operational definitions to suit their study objectives. Migrants have the highest prevalence of HIV after the high-risk groups of female sex workers (FSW), men who have sex with men (MSM) and injecting drug users (IDUs).  Migrant workers also have diverse vulnerabilities.  They tend to be less educated, have poor HIV/AIDS knowledge and engage in risky unprotected sex. Alcohol consumption is a routine recreational activity for many migrants and expendable cash they earn through their work helps them to engage in recreational sex with FSWs. It has been observed that many men consume alcohol before they engage in sex with FSWs.  It is also generally accepted that alcohol-sex linkage has serious implications for health because, together, they increase vulnerability to HIV. Risky sexual behavior under the influence of alcohol has been especially noted in high-risk groups.  Some of the explanations given for this association include the disinhibitory effect of alcohol, the practice of heavy drinking among young risk takers and presence of licensed premises.  Alcohol and unsafe sex contribute to a significant proportion of DALYs lost globally.  There has been a significantly rising trend in the recorded consumption of alcohol in India, which can be considered to be due to the effects of modernization and media on young people. , Although some studies point to a positive association, others show no association.  It has been suggested that this is due to the fact that this relation is complex and conforms to a larger system of interconnected variables.  In this study, we aimed to study the association between alcohol consumption and non-use of condom at the last non-spousal sexual encounter among male migrant workers in an urban area in north India.
Ethics: Approval for the study was obtained from the Ethics Committee of the All India Institute of Medical Sciences, New Delhi. Written informed consent was obtained from all prospective participants.
Study design: This was a facility-based cross-sectional survey conducted in two factories located at the Ballabgarh block of Haryana. One of the sites was a steel bars-producing unit with approximately 1500 workers, located 2 km from a Primary Health Center (PHC), Chhainsa, and the other site was a turbine parts-manufacturing unit with approximately 2000 workers, located 4 km from the Comprehensive Rural Health Services Project (CRHSP) hospital, Ballabgarh.
Selection criteria: All l adult male migrants aged 18 years and above were selected. Data were collected during the months of June and July, 2011. Those born outside Haryana, who had moved to the current location after 15 years of age, who had worked in the current factory for at least 1 year, who were willing to participate and who were able to give valid consent were eligible for inclusion.
Study procedure: Face-to-face interviews were conducted using a pre-tested semi-structured questionnaire. The questionnaire used in this study was constructed using a number of sources such as the National Family Health Survey-3 and the District Level Household and Facility Survey-3. The pre testing was done in a sample of 20 migrant workers who were not from the study area. The questions were put to the participants in Hindi. An attempt was made to assess the necessity and understandability of the questions. Those questions that were not properly understood were rephrased and those questions that seemed inappropriate were removed. Because of operational constraints, a random sample could not be obtained. Instead, workers attending the factory clinic were approached to take part in a consecutive manner. In order to maintain data quality, not more than 20 interviews were conducted in a day. Although not as robust as a random sample, consecutive sampling is an accepted methodology widely used in HIV Sentinel Surveillance activities in India.  Basic demographic characters of those who were eligible but refused to participate were collected.
Calculation of knowledge and attitude score: A knowledge score was calculated by taking into account 22 variables. The total score range was 0-22, and was classified based on the proportion of the maximum possible score obtained into three categories as ≥80% (comprehensive), 50-79% (moderate) and ≤49% (poor). "Intention to use condoms" was measured on a four-item five-point Likert scale. The total score range was 4-20 (higher the score, more unfavorable the attitude).
Calculation of mobility index: The mobility index for each participant was calculated by dividing the number of places migrated by the total duration of migration. Larger the index, the more mobile the person was, i.e., migration to greater number of places in a shorter duration.
Because the current research question was a secondary objective of an ongoing study on HIV risk behavior among migrant workers, the analysis was restricted to men who reported having been involved in non-spousal sex in the last 1 year. All analyses were performed using Stata/IC 11.1 (StataCorp LP, College Station, TX, USA). Descriptive statistics were used and data expressed as proportions, means, standard deviations (SD) and 95% confidence intervals (CI) wherever applicable. An adjusted analyses was performed by testing (using the Mantel-Haenszel test and the Mantel-Haenszel Odds Ratio) the association between alcohol consumption and non-use of condom at the last non-spousal sexual encounter across different strata of confounding variables. The confounders tested were age, income, marital status, migration characteristics, HIV/AIDS knowledge, age at first sex and history of self-reported STI symptoms. For simultaneous control of confounding variables, we performed backward stepwise multivariable logistic regression and adjusted ORs were presented with 95% CI. A P-value of less than 0.05 was considered to be statistically significant.
A total of 767 men were approached for participation, of whom 755 consented and were interviewed. Of the 755 men, 162 participants reported having experienced non-spousal sexual encounter in the last 1 year and were included for analysis. The mean (SD) age of these participants was 21.4 (6.0) years, and 54% were more than 25 years old. The mean (SD) salary received in the month prior to the month of interview was Rs. 5517 (2300), and 82.7% of them had received less than or equal to Rs. 6000. Literate (ability to read and write in atleast one language) participants constituted 96% of the sample and 52.5% were currently married [Table 1].
Bihar was the place of origin for 45.7% of the participants. The mean (SD) age at migration was 20.1 (3.3) years, and 68.9% of the participants had migrated for work-related purposes before the age of 21 years. The mean (SD) number of places migrated for work-related purposes was 4.1 (2.9), with majority of the participants having migrated to two to five places in their lifetime. The mean (SD) total duration of migration was 7.1 (5.0) years, and 50% of participants had completed at least 6 years since migration. The mean (SD) mobility index was 0.7 (0.3), and 65.4% had an index greater than 0.5. The mean (SD) duration of stay in Haryana was 2.5 (2.0) years, and 69.8% had completed 1-2 years of stay in Haryana. The mean (SD) number of home visits made by the participants in the last 1 year was 3.5 (2.1), and 57.4% had gone home for less than or equal to three times in the last 1 year [Table 1].
HIV/AIDS knowledge, alcohol consumption, risk behavior and self-reported symptoms of sexually transmitted infection (STI)
The mean (SD) HIV/AIDS knowledge score was 15.4 (6.1), and only 40% had comprehensive HIV/AIDS knowledge [Table 2]. The mean (SD) age at first sexual intercourse was 17.1 (1.3) years, and 86.4% had their first sexual encounter before the age of 18 years. About 59.3% of the men reported not having used a condom at their last non-spousal sexual encounter. About 78.4% of the men reported having consumed alcohol in the last 1 year. About 48.1% of the men reported having consumed alcohol before their last non-spousal sexual encounter [Table 2]. About 50% of men reported having at least one symptom of STI in the last 1 year (57%-burning micturition, 4.3%-having a genital ulcer, 1.2%-urethral discharge, 6.2%-inguinal bubos). The mean (SD) "intention to use condoms" score was 9.9 (3.3).
Association between alcohol consumption and non-use of condom at last non-spousal sexual encounter
In the crude analysis, it was found that men who consumed alcohol were three times more likely to not use a condom at their last non-spousal sexual encounter (OR = 3.1, 95% CI: 1.5-6.4) [Table 3]. Bivariate analysis was performed to test the association by examining the crude ORs across different strata of confounding variables by applying the Mantel-Haenszel test and calculating the Mantel-Haenszel Odds Ratios. This analysis showed that the association was statistically significant even after adjusting for age, income, marital status, migration characteristics, HIV/AIDS knowledge, age at first sex and history of self-reported STI symptoms individually [Table 4]. We performed multiple logistic regression to adjust for selected confounding variables simultaneously and estimate the true association between alcohol consumption and non-use of condom at last non-spousal sexual encounter. The multivariable analysis showed that even after adjusting for confounders, the association remained statistically significant (aOR = 3.1, 95% CI: 1.1-9.0) [Table 5].
This study is unique because of the fact that although the association between alcohol use and risky sexual behavior has been examined previously, the present study used a population where both these behaviors are fairly prevalent as an example to explore this relationship. The findings of this study have further strengthened the notion that alcohol consumption has a negative influence on condom use during high-risk sex.
The study population was young, educated, was mostly unmarried or living alone if married and earned a modest livelihood. Most of them had migrated from the northern state of Bihar and many had migrated at a very young age and were in the early years of migration, fairly mobile, relatively new to Haryana and made frequent visits to their hometown. They had moderate knowledge about HIV/AIDS. Prevalence of condom use at last non-spousal sex encounter was low and that of alcohol consumption was high.
In this population of migrant workers, we found that alcohol consumption before sex significantly increased the odds of not using condoms during high-risk encounters such as sex with non-spousal partners. This association remained significant even after adjusting for possible confounding variables. This finding confirms the amplifying effect of alcohol on the HIV vulnerability of migrant workers who engage in high-risk sex. Similar results were reported by other studies. ,,, A World Health Organization (WHO) cross-cultural study done in eight countries reported that alcohol consumption posed a risk for acquisition of STI/HIV by encouraging risky sexual behavior. The study also found that situational factors like venue of drinking and sex-related demonstrations further increased the negative influence of alcohol. 
On the other hand, several studies have reported no association between the two variables. ,,, A study performed among young adults in the USA found that even though overall there was no association between the two, heavy drinking was associated with non-use of condoms.  Similar findings of association between amount of alcohol and risk behavior were reported in a study carried out among Swiss heterosexual men.  A study by Velez-Blasini suggested that personality and behavioral dimensions were more important than alcohol effects in determining sexual behavior.  These were however not studied by us. Other interdependent variables that may confound this association include partner type,  partner alcohol use,  type and amount of drink and baseline prevalence of condom use in high-risk sex. Our study did not measure some of these variables and hence could not study their effect on this association. The difference in findings between our study and other studies might be due to the differences in the population studied, the settings, confounders used for adjustment and study methodology. The type of partner involved in the non-spousal sex in our study was universally FSW. Hence, it would be reasonable to state that the alcohol consumption indeed had a negative influence on condom use during non-spousal sex among male migrant workers. This risk behavior might put their regular partners or spouses at risk as they might engage in unprotected sex with them.
Interventions planned to reduce this vulnerability must address the two issues simultaneously. One set of interventions should be aimed at promoting condom use and the other set aimed at reducing alcohol consumption. There is an existing platform known as Targeted Interventions promoted by the NACO that can be used to deliver these services in an integrated manner.  Easy availability of condoms and promotion of their use should be increased, especially among workers who are involved in non-spousal and recreational sex. Workplace policies that provide a facilitating environment should be developed for such factories where migrant workers are employed. Provision of alternate recreational facilities such as gymnasia, movie screening, sports and games can help relieve the stress of work and reduce involvement in risky behaviors. Interventions based on social theories of human behavior, such as the Theory of Planned Behaviour, Diffusion of Innovation Theory, Empowerment Model, Social Networks Theory and Theory of Structural and Environmental Factors have been suggested. , The WHO study had suggested that for India, the interventions should specifically focus on health and sex education emphasizing the role of psychoactive substance in risky sex and promotion of condom use among high-risk groups.  Alcohol-related policies in a country also influence this relationship. Although India has many alcohol regulatory mechanisms such as excise duty, restrictions on advertising, sale and sponsorship and minimum legal age limit for consumption in place, there has been no decrease in the trend of per capita consumption.  Widespread implementation of these policies and awareness campaigns among target populations can go a long way in reducing alcohol consumption. Novel mechanisms that combine IEC activities with legal sanctions need to be devised for better control of this problem, especially among core risk groups such as migrant workers.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]