ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Migration, Youth and HIV Risk: A Study of Young Men in Rural Jharkhand

This article analyses the sexual practices of young migrant and non-migrant men in Palamu district of Jharkhand in the context of hiv awareness. Based on a field study and using statistical analyses, the authors find that, contrary to popular stereotype, both migrant and non-migrant youth lack hiv awareness and practise "risky" sexual behaviour. They argue that hiv awareness policy must sensitise itself to the prevalence of "sexual networks" and concentrate advocacy drives in these areas.

REPRODUCTIVE HEALTHdecember 1, 2007 Economic & Political Weekly40Migration, Youth and HIV Risk: A Study of Young Men in Rural JharkhandMrinalika Dhapola, Mona Sharan, Bharat ShahThis article analyses the sexual practices of young migrant and non-migrant men in Palamu district of Jharkhand in the context ofHIV awareness. Based on a field study and using statistical analyses, the authors find that, contrary to popular stereotype, both migrant and non-migrant youth lack HIV awareness and practise “risky” sexual behaviour. They argue thatHIV awareness policy must sensitise itself to the prevalenceof “sexualnetworks” and concentrate advocacy drives in theseareas.It is increasingly recognised that young people and migrants are sub-populations that are at significant risk of contracting and transmittingHIV. Studies of youth suggest that a large percentage of young people practise risky sexual behaviour. Evi-dence also suggests that migrants are a group at risk of HIV as a result of their sexual behaviour at their destination points and on return to their place of origin. Understanding the sexual behaviour patterns of young mi-grants, an underserved yet high-risk group that serves as a bridge populationbetween high and low risk populations, will have an important bearing on the course of the epidemic. However, little is known about the situation of migrant youth in India, theirknowledge of transmission and prevention of HIV or the risk behaviour that makes migrants, in particular young migrants, vulnerable toHIV infection, and the policies and programmes to address them. This study seeks to understand the factors underlying the risk behaviour of young migrants through a com-parative study of HIV-related knowledge and sexual behaviour of young male migrants and non-migrants fromPalamu district, Jharkhand, a poorly developed setting in India,which has wit-nessed large-scale outmigration in recent years. Background and Literature ReviewOver 35 per cent of newHIV infections every year occur in the age group 15-24 years in India [UNAIDS 2006]. Socially and economi-cally disadvantaged persons, such as migrants and amongst them especially young people, are particularly vulnerable toHIV be-cause they are exposed to greater opportunities for risky sexual activity but have limited capacity to mobilise resources to cope with exposure to the risk of infection [Delor and Hubert 2000].Young people (15-24 years) are at a higher risk of contracting and transmittingHIV than other age groups while sexual activity among young people is on the rise, awareness of safe sex is lack-ing and few have access to information, skills, services and sup-port forHIV prevention. Many lack the power to make informed sexual and reproductive health choices. In India too, young people are particularly vulnerable to HIV, and research, though sparse, reveals that a larger number of youth than previously believed engage in premarital sex, including sex at an early age and high-risk behaviour [Jejeebhoy 1998]. Indeed, the growth of theHIV epidemic in India has coincided with the unprecedented rise in the number of people entering adolescence and youth; of an estimated population of 327 million young people aged 15-24 years, between 6,50,000 and 1.3 million are believed to be in-fected withHIV/AIDS [UNICEF, Joint UN Programme on HIV/AIDS andWHO 2002; United Nations Population Division 2001]. We are grateful to S Halli for valuable comments.Mrinalika Dhapola ( is programme manager, American India Foundation Trust; Mona Sharan ( is a health consultant with the World Bank and Bharat Shah ( associate director, ORG Centre for Social Research, New Delhi.
REPRODUCTIVE HEALTHEconomic & Political Weekly december 1, 200741At the same time, migration is also a major cause of the spread of the HIV virus [UNAIDS 2001]; it is a significant determinant of risky sexual behaviour, and studies have documented high-risk behaviours among migrants, including sex with multiple part-ners, sex with sex workers, same-sex relationships and low rates of condom use [Brockerhoff and Biddlecom 1999; Gras et al 2001; Wolffers et al 2002]. Studies of migration in India, though sparse, indicate that migrants are vulnerable toHIV, and migration is a major risk factor underlying the growing prevalence of HIV in the country. In a study conducted at anHIV-testing clinic in south India, for example, being a migrant or having a migrant spouse was a signi-ficantrisk factor for HIV-positive status [Solomon et al 2000]. Moreover, migration is a risk factor for HIV not only for internal migrants but also for migrants from neighbouring countries such as Nepal [Gurubacharya and Gurubacharya 2004]. That migration is significantly associated with the odds of ever having experienced sexual intercourse has been observed in studies conducted elsewhere in south Asia [see for example, Puri and Cleland 2006 for Nepal]. Few studies have explored the sexual behaviour and sexual knowledge of migrants in India. Available evidence shows that migrants are poorly informed about reproductive and sexual health, including HIV, and many practise risky behaviours at their place of destination and origin. In a study of sexualbehav-iour among migrant hotel workers to Pune city, Maharashtra (median age 22 years), respondents reported that first inter-course generally occurred after migration and was usually at the place of origin, with a girl or married woman, generally spontaneously. The large majority, 81 per cent, had not heard of sexually transmitted infections stis). Consistentcondom use was reported by less than half of those who were sexually experienced [Gujjarappa and Rasalpurkar 2004]. A study of riskbehaviours among migrant workers in Surat, India (aged 15-35 years) revealed that 15 per cent reported visiting sex workers and very few reported condom use [Chakraborty 2004]. Misperceptions about the transmission of HIV are common; a study of knowledge and sexual risk behaviour among migrants in Rajasthan indicated that 33 per cent believed that a person could get infected through mosquito bites. The study also reported risky behaviours among migrants; 4.1 per cent of married men reported a sexual partner other than their regular partner; 22 per cent had had a sexual encounter in the past 12 months, and 9 per cent reported engaging in sex with a sex worker at the place of destination. While awareness of condoms washigh,lessthan8 per cent of married men and 5 per cent of unmarried men re-ported using a condom for the last sexual act. Twelve per cent reported having had anSTI, and 29 per cent had used a local remedy for treatment [Singh et al 2004]. Study SettingThe study was conducted in Palamu district, one of 22 districts in Jharkhand state. Jharkhand, carved out of the state of Bihar in 2000, is a relatively poorly developed state, scoring lower than the all-India average on several indicators. According to census data [RGI 2001], only 24 per cent of households in Jharkhand had electricity (compared to 56 per cent all-India), 20 per cent had a supply of drinking water (39 per cent all-India) and 11 per cent had a toilet facility (all-India 18 per cent). Moreover, the majority of households (59 per cent) reported a low standard of living (all-India 36 per cent) [IIPS andORC Macro 2000]. The state has a large rural population (80 per cent). The majority of the people are socially disadvantaged: 15 per cent of households belonged to scheduled castes, 32 per cent to scheduled tribes and 39 per cent to other backward castes [IIPS and ORC Macro 2000]. Exposure to radio, television or other types of media and knowledge of HIV is low in Jharkhand; almost three-quarters of women in the state were not regularly exposed to the media, and nearly nine out of 10 women had not heard of HIV (all-India 60 per cent) [IIPSandORC Macro 2000].Employment Jharkhand has few employment opportunities for youth, and it witnesses large-scale outmigration as well as migration within the state. According to the 2001 Census, of a total population of 26.9 million, 6,16,160 persons from Jharkhand migrated to other states [RGI 2001a]. Of these, 2,91,093 were male and 3,25,067 were female (higher rates of migration among women is attributed to marriage). The largest proportion of out-migrants from Jharkhand migrated to the neighbouring states of West Bengal (25.1 per cent), Bihar (17.8 per cent) and Uttar Pradesh (11.1 per cent). A smaller proportion moved to Delhi (10.6 per cent) and Orissa (7.6 per cent). Migration within Jharkhand was largely from one rural area to another; 75.8 per cent of all intra-state migrants (i e, 12,15,941 persons) were rural to ru-ral migrants [RGI 2001a]. A study on migration in 12 villages of Jharkhand reports that one-third of households were found to have at least one member who had migrated; moreover, socially disadvantaged people had a greater tendency to migrate: 15 per cent ofscheduled castes and tribes migrated compared to only 8 per cent of upper castes [Dayal and Karan 2003]. Seasonal mi-gration, also referred to as survival migration, was high among the landless in the study: 80 per cent of those who were landless migrated for short periods. Palamu has been identified as among the most backward states in India ( The district, with a total population of 20,98,359 persons [RGI 2001b], reports poor deve-lopment and low levels of urbanisation. According to the 2001 Census, only 6 per cent of the district’s population is urbanised [RGI 2001c]. Most villages in the district have a scarce supply of drinking water and electricity, and poor availability of health and education services. The literacy rate is low [45 per cent; RGI 2001b]. The district has a high concentration of scheduled caste and tribe population; 25.7 per cent and 18.7 per cent respectively [RGI 2001d]. Villages in the district are generally located at a considerable distance from the district headquarters and are poorly connected by road. Access to towns is particularly difficult during the monsoon season. Agriculture is the primary source of liveli-hoodinthearea but is not a significant source of income for wage labourers as they have to pay huge commissions to contrac-tors who employ them to work on the farms. Lack of employment
REPRODUCTIVE HEALTHdecember 1, 2007 Economic & Political Weekly42opportunities and low agricultural productivity compel people to migrate to both urban and rural areas in search of alternative sources of employment [ORG 2005]. In Palamu, intra-state migra-tion is generally to rural areas; 57.2 per cent of the total intra-state migrants (27,059) migrated to other rural areas in the state. In addition, a total of 93,471 females and 13,872 males migrated from one setting to another within the district.1MethodologyData for the study has been drawn from a study of unmarried mi-grant and non-migrant young men in Palamu district. The study comprised both a survey of migrant and non-migrant youth and in-depth interviews with migrants, non-migrants and health- care providers. For purposes of this study, a migrant youth was defined as an unmarried male aged 15-24 years who had migrated to urban or rural areas in search of employment, had not relocated to the place of destination, did not live with his family at the place of destination for more than a month, had visited his place of origin at least once over the past year and was located at the place of ori-gin for at least five years. A non-migrant youth was defined as anunmarriedmale aged 15-24 years who had never migrated to an urban or rural area and had stayed in the place of origin for at least five years (see Appendix for operational definitions of variables). The survey comprised a sample of 504 migrants, including 403 young men and 101 young women. In addition, 458 young non-migrants, includ-ing 308 men and 150 women, were interviewed. In this paper, we focus on data obtained from the sample of 403 male migrant youth and 308 male non-migrant youth. We note that as the study was implemented in the areas of origin of young migrants, the sample may not represent those migrants who remained in their destination areas at the time of the survey. Efforts were made however, to ensure that the survey captured as many migrants as possible by timing its implemen-tation with the period during which migrants typically returned to Palamu for the festival season. The sampling procedure involved an identification of villages that had a sufficient number of outmigrants to attain the desired sample size, as documented by a previous mapping exercise of the area [ORG 2005]. A total sample of 32 villages was thus identi-fied. Household listing was carried out and eligible respondents were identified. A systematic random sampling method, with a random start, was used to select study respondents. Interviews were conducted in the homes of respondents by interviewers of the same sex. Questions asked during the survey pertained to migrants’ and non-migrants’ socio-economic status, knowledge of HIV transmission and prevention, sexual behaviour at the village of origin and perceptions of risk associated with sexual behaviour. During in-depth interviews, migrants were asked to further describe patterns of migration and sexual behaviour at the place of destination. Quantitative data analysis for the study provided descriptive statistics including means, frequencies and proportions for the variables of interest. Transcripts of the qualitative interviews were reviewed and corroborating evidence summarised, which is presented in this paper.Profile of RespondentsTable 1 presents a socio-demographic profile of respondents. Find-ings suggest that compared to non-migrants, migrants were older (19.0 and 16.8 years respectively) and significantly less educated: approximately 64 per cent of migrants compared to 44 per cent of non-migrants were uneducated or had received no more than a primary school education. The mean age of first migration was 16.1 years.A measure of socio-economic status was calculated by combining scores as-signed to household assets, and the index was further divided into three categories denoting low, middle or high socio-economic status. The socio-economic disadvantage of migrants was apparent by the finding that while 24.3 per cent of non-migrants belonged to the high socio-economic category, only 16.4 per cent of migrants were similarly well off. Most migrants belonged to the low (35.7 per cent) or middle (47.9 per cent) socio-economic status group. Lifestyle indicators, leisure activi-ties and drug and alcohol use werealsoexplored.Findingssug-gest that migrants had greater exposure to mass media than non-migrants (Table1). A larger proportion of migrants than non-migrants reported watching television during their spare time (31.8 per cent vs 26.3 per cent respectively),perhapsdueto greater access to television at the place of destination. Exposure to print media, as denoted by subscription to a newspaper, was simi-lar among both migrants and non-migrants (2.8 per cent and 3.2 percent respectively).Some migrants reported using drugs (2.2 per cent) and alcohol consumption (6.0 per cent). In contrast, no non-migrantreportedusing drugs and only two persons reported alcohol consumption.Migration and the Context of Sexual BehaviourThis study measures the context of sexual behaviour among mi-grants and non-migrants through three sets of factors: (a) aware-ness of HIV and the role of condoms in the prevention of HIV transmission; (b) discussion about HIV with a partner or friend; and (c) premarital sexual experiences, including lifetime and first sexual experience, sexual partners and condom use (for op-erational definitions of variables, see Appendix). Awareness ofHIV was measured by dichotomous variables indicating whether the respondent had ever heard of HIV or was aware of the role of condoms in preventing HIV transmission. Table 1: Socio-Demographic and Lifestyle Characteristics MigrantsNon-migrants N=403 % N=308 %Age (years) 15-17 111 27.5 214 69.5 18-20 184 45.7 76 24.7 21 and above 108 26.8 18 5.8Education Illiterate 87 21.6 24 7.8 Primary 170 42.2 113 36.6 Secondary and above 146 36.2 172 55.7Socio-economic status* Low 144 35.7 66 21.4 Middle 193 47.9 167 54.2 High 66 16.4 75 24.3Leisure activities Watchedtelevision 128 31.8 81 26.3 Subscribed to a newspaper 11 2.73 10 3.2 Druguse 9 2.2 0 0.0 Alcohol consumption 24 6.0 2 0.6* Based on an index of ownership of household assets (see Appendix for details).
REPRODUCTIVE HEALTHEconomic & Political Weekly december 1, 200743In addition, a categorical variable reflecting more general levels of awareness of HIV was created from the knowledge index. The knowledge index was obtained by summing correct re-sponses to questions about knowledge of HIV such as whether HIV can be transmitted during unprotected sex/sex without a condom including with a same sex partner, sex with a sex worker, sex with multiple partners, sex with anHIV-infected person, sex with a person with aSTI, sharing needles and blood transfusion. The prevalence of common misconceptions about transmission of HIV was also included in the index, for example, by shar-ing clothes and utensils, using common toilets and bathrooms, mosquito bites, touching an infected person and having sex with a healthy looking person. Other components included in the in-dex were related to knowledge about HIV prevention, including abstinence, reducing the number of sexual partners, remaining faithful to one partner, abstaining from unsafe same sex activity, not sharing needles and using a condom for every sexual act. The scores ranged from 0-20: 0 denoting “no knowledge”, 1-13 reflect-ing “low knowledge” and 14-20 indicating “high knowledge” (see Appendix for details of the knowledge index).Awareness Not Universal Findings indicate that awareness ofHIV is far from universal among migrants and non-migrants: for example about two in five respondents had not heard of HIV and three in five were unaware of the role of condoms inHIV prevention (see Table 2). Non-mi-grants were somewhat better aware ofHIV and its transmission routes than migrants. The differences in awareness are significant in the bivariate analysis for HIV awareness (Chi-square = 2.74, p<0.10), role of condoms (Chi-square= 5.34, p<0.05) and the knowledge index (t=1.99, p<0.05). Qualitative evidence suggests that even when there is basic awareness of HIV, several misperceptions abound. Both migrants and non-migrants were similarly misinformed as the following statements of informants illustrate. My girlfriend is married now but I will continue to have sex with her. There is no possibility that I can haveHIV. If my girlfriend is having sex with two men, there is no problem. Even if a woman has sex with four men, she will not have a problem. But if it happens daily, morning and evening, with several men, then there will be a problem (migrant male, 16 years).I know about HIV. We can get it if we touch or eat with an infected per-son. Also by getting bitten by a mosquito that has bitten someone who is infected. I have had sex only once. I don’t think that one time sex can causeHIV infection (migrant male, 18 years).Living in dirty places and sharing a bath with someone can cause HIV (non-migrant male, 15 years).Migrants in Palamu district also view relationships with “good girls” in conceptual opposition to brothel sex; these relationships are associated with a degree of trust and safety. Similar findings are reported in other studies [Lyttleton and Amarapibal 2002].My friend got anSTI. He goes to a prostitute. That is the reason why he got the disease. This is what happens when you do wrong things (male migrant, 20 years).If a person goes to prostitute and does wrong things, then he will get HIV. She [the sex worker] has sex with so many people, she surely will haveHIV (migrant male, 23 years).A multivariate logistic regression analysis was carried out to evaluate the extent to which lack of awareness could be explained by migrant status, after controlling the effects of socio-demographic factors. An ordinal logistic regression model was fitted for one ordinal outcome, which was knowledge of HIV. De-mographic variables such as respondent’s age, socio-economic status and education were modelled as independent variables. Age was treated as a continuous variable. Socio-economic status and education were included as categorical variables. Media expo-sure, an independent variable, was measured by reported television watching during spare time and by having a newspaper subscrip-tion. Reported drug or alcohol use and migrant status were also included as independent variables in the multivariate models. The results of the multivariate analyses (Table 3, p 44)) indi-cate that once socio-demographic and other factors are control-led, migrants are about as likely as non-migrants to be aware of HIV and its transmission. Indeed, a host of socio-demographic factors are associated withHIV knowledge. For example, age was positively associated with awareness, suggesting, as expected, that older respondents were more likely than others to be aware of HIV know that condoms prevent HIV and have better know-ledge about HIV transmission and prevention.Likewise, education was positively associated with all know-ledge variables. Both primary and secondary education, relative to no education, significantly increased respondents’ odds of Table 2: Awareness of HIV and Premarital Sexual Experiences MigrantsNon-migrantsA Among all respondents N=403 % N=308 %Awareness of HIV Aware of HIV 225 55.8 191 62.0 Aware that condoms prevent transmission of HIV 133 33.0 128 41.4 HIV knowledge index* Noknowledge 179 44.4 118 38.2 Low knowledge 99 24.6 72 23.3 Highknowledge 125 31.0 119 38.5Communication about HIV Discussed HIV with a partner/friend 82 20.3 55 17.7Premarital sexual experiences Respondents who reported a sexual experience 121 30.0 45 14.6B Among those reporting sexual experience N=121 N=45 Had risky sex** 38 31.4 14 31.1 Place of sexual experience At place of destination 29 23.9 – At place of origin 97 80.1 – Age at first sex† <=15years 53 44.2 26 57.8 >15years 67 55.8 19 42.2First sexual partner Girlfriend 94 77.6 41 91.1 Femalerelative 13 10.7 2 4.4 Sex worker 5 4.1 0 0.0 Fiancée 2 1.6 0 0.0 Boyfriend 1 0.6 0 0.0 Other place 7 5.7 – –Timing of first sex by migration – – First sex at same age as migration 19 15.7 First sex after migration 55 45.4 – – First sex before migration 47 38.8 – –Place of first sex At place of destination 23 19.0 – – At place of origin 91 75.2 – – Other place 7 5.7 – –1) † One observation missing for age at first sex for the migrant sample. 2) * See Appendix for details about the HIV knowledge index.3)** Risky sex was defined as sex with sex workers, eunuchs, men or boys, or more than one female partner.
REPRODUCTIVE HEALTHdecember 1, 2007 Economic & Political Weekly44having a higher level of knowledge of HIV. Secondary school edu-cation in particular was strongly associated with awareness of HIV (HIV 6.88, p<0.001), knowledge that condoms prevent HIV (HIV 7.19, p<0.001) and knowledge of HIV prevention and trans-mission (HIV 7.48, p<0.001).Socio-economic status was significantly associated with all three knowledge outcomes. Respondents from the high socio-economic group were significantly more likely to be aware of HIV (OR 2.41, p<0.05), know that condoms prevent HIV (OR 1.97, p<0.05) and have good knowledge of HIV transmission and pre-vention (OR 1.83, p<0.05) relative to those from the lower socio-economic groups. Exposure to media was not associated with awareness of HIV or knowledge of the role of condoms in preventing HIV. Exposure to media increased the odds of having greater awareness of HIV relative to no knowledge (OR 1.33, p<0.10). Drug or alcohol use was unrelated to any of the knowledge variables. In short, awareness ofHIV pre-vention is low among both migrants and non-migrants, and was largely explained by age, education and socio-economic status of respondents. Respondents who are older, better educated and socio-economically ad-vantaged are likely to have access to information on HIV and the ability to comprehend HIV-related messages from various sources. Exposure to media may contribute to some improve-ment inHIV-related knowledge. Clearly, migration itself is not associ-ated with greater awareness of HIV, and there is no evidence that mi-grants, as a vulnerable population, have received more information about HIV than have non-migrant youth.Communication about HIVGiven the hypothesis that communication about sexual matters may protect young people from risky sexual outcomes, the study explored whether respondents had discussedHIV with a partner or friend – a key component of protective strategies – in the month preceding the survey. Findings reveal that communication on HIV with a partner or friend is rare: reported by 17-20 per cent of youth (Table 2, p 43). Migrant-non-migrant disparities are modest and insignificant (Chi-square = 0.69, p = 0.40). As before, multivariate analysis was used, with the independ-ent variables noted in Table 4 (p 45), to assess the extent to which communication about sexual matters is influenced by migrant status. The results indicate that migrants and non-migrants are as likely to have discussedHIV-related issues with a partner or friend. Among socio-demographic factors, age and secondary education (OR 3.43, p<0.05) were found to be positively associated with communication on HIV. Socio-economic status was unrelated to whether or not respondents communicated with friends or part-ners about HIV. Drug and alcohol use, likewise, were not related to communication on HIV. The findings clearly show that discussion about HIV with a partner or friend is unaffected by migrant status. Findings hint that communication or discussion on sexual matters is significantly more likely to occur among older and secondary schooled youth than among others.Premarital Sexual ExperiencesThe study explored young people’s sexual experiences through survey and qualitative data reflecting lifetime and first sexual experiences, multiple sexual partners and condom use. Ever Had Sex: Findings suggest that migrants are significantly more likely to have engaged in premaritalsexthan are non-migrants: 30 per cent migrants versus 15 per cent non-migrants report sexual experience (Chi-square = 23.17, p<0.001). However, migrants did not report high rates of sexual activity during migration, as was expected. Migrants in this study were more likely tohavehad sex at the place of origin – before and after first migration – than at the place of migration. Among mi-grants who had ever had sex, 29 (23.9 per cent) reported having sex at the place of destination and 97 (80.1 per cent) reported sex at the place of origin.Both migrants and non-migrants reported early sexual activity during interviews:I used to go to the river for a bath. She knew my timings and would wait for me. She threw a letter at me which said, “I love you”. She told me to come to her so I went up to her and hurriedly had sex. I was scared that someone would find out and Iwould get beaten up (migrant male, 18 years). I used to go to school with her and we became friends. After 2-3 months, we had sex. I must have been around 15 years old when we had sex for the first time. She was the same age as me as we were class-mates (migrant male, 20 years).When I was 15 years old, I had sexual relations with a girl. She is the sister of my brother’s wife. She forced me to have sex with her, threatening that she would start shouting and call the neighbours and create prob-lems for me if I didn’t (non-migrant male, 18 years).I had sex for the first time when I was in Class 9. I was 16 years old then. After coming back from school, we used to have sex in her house when her parents were in the field (non-migrant male, 17 years).At first we used to have fun talk. Then our friendship deepened. I told her that I wanted to have sex with her. She said that she had no prob-lem in having sex and told me she would come in the evening for it. I was 16 years old when we had sex; she was the same age as me (non-migrant male, 18 years).Age at first sex, first sexual partner and location of first sex: Of those who have ever engaged in sex, findings, though not significant, Table 3: Odds Ratios for Effects of Independent Variables on HIV Knowledge Indicators HIV Awareness Condom Use HIV Knowledge No vs Yes Prevents HIV No vs Low No vs Yes and HighAge 1.11**1.10**1.08** (1.03 - 1.20) (1.02 - 1.19) (1.01 - 1.15)Socio-economic status Low 1.00 1.00 1.00 Middle 1.31 1.28 1.21 (0.89 - 1.94) (0.83 - 1.97) (0.84 - 1.74) High 2.41** 1.97** 1.88** (1.42 - 4.09) (1.17 - 3.31) (1.19 - 2.95)Education Illiterate 1.00 1.00 1.00 Primary 2.23** 2.14** 2.34** (1.35 - 3.66) (1.11 - 4.13) (1.44 - 3.82) Secondary 6.88*** 7.19*** 7.48*** (4.01 - 11.78) (3.73 - 13.89) (4.46 - 12.55)Exposure to media No 1.00 1.00 1.00 Yes 1.27 1.22 1.33* (0.88 – 1.83) (0.85 - 1.76) (0.97 – 1.82)Drug/alcohol use No 1.00 1.00 1.00 Yes 1.37 1.25 1.75 (0.61 – 3.09) (0.54 - 2.85) (0.85 - 3.57)Migrant status No 1.00 1.00 1.00 Yes 0.88 0.76 0.83 (0.59 - 1.30) (0.51 - 1.14) (0.59 - 1.18) Observations 711 711 711* p<.10, **p<.05, ***p<.001. See Appendix for operational definition of variables.
REPRODUCTIVE HEALTHEconomic & Political Weekly december 1, 200745suggest that non-migrants rather than migrants reported early sex-ual initiation (before 15 years) (57 per cent non-migrants versus 44 per cent migrants, Chi-square= 2.42, p=0.119). The majority of respondents, both migrants (78 per cent) and non-migrants (91 per cent), reported that their first sexual partnerwasagirlfriend. Among migrants, 4 per cent reported visiting a sex worker during their first sexual encounter and 11 per cent had their first sexual experience with a female relative. No non-migrant reported visiting a sex worker for their first sexual encounter and 4 per cent experienced first sex with a female relative (Table 2, p 43). Among migrants, a larger proportion of first sexual experiences took place after first migration than before first migration (45 per cent com-pared to 38 per cent; see Table 2). Findings sug-gest that among migrants, the first sexual en-counter generally occurred in the village of origin. Seventy-five per cent of migrants who had had sex reported that their first sexual experience was at the place of origin, followed by 19 per cent who experienced first sex at the place of migration; 6 per cent reported experiencing first sex at a place other than the place of origin or migration (Table 2). Findings from Nepal similarly show that young migrants experienced sexual initia-tion in their home villages with a boyfriend or girlfriend; and a large proportion of sexual debuts took place in the village prior to migration [Puri and Cleland 2006].Sexual Partners: For the majority of young male migrants, sexual partners were girlfriends irre-spective of whether the sexual experience occurred at the place of origin or destination. For the 97 who reported sex at the place of origin, 82 reported that their partners were girlfriends; and of the 29 who reported sex at the place of destination similarly, 24 reported that the partner was a girlfriend. Of the 45 sexually experienced non-migrants, 15 had had sex with girlfriends. Young people did not commonly report experiences with sex workers, even among those reporting sex at the place of desti-nation. A few migrants had sex with a sex worker or eunuch: 10 in the place of origin and nine in the placeofdestination. Similarly, only a few non-migrants had sexual encounters with sex workers (3) and eunuchs (2).That sex among migrants typically took place with a girl-friend rather than a sex worker is supported by textual data, for example:I did not have any girlfriend over there [place of migration]. I have a girlfriend here. She is from the same village. We had a friendship for two years then our friendship turned into love. The first time we were together in a quiet place and we both wanted it, so we ended up having sex (migrant male, 22 years).My friends used to help me pass on messages to her. She told me she liked me and gradually I started liking her too. We had sex at my home, at the farm and near the river. I felt happy that I could show her my true love. Recently, she got married (migrant male, 18 years).We were neighbours and I used to go to her place to watch television. Our friends used to help us exchange messages. When we got a chance, we would have sex, mostly at night. If I had married her it would have been good, because I love her. But now it is too late because she has got married (migrant male, 16 years). I met her when I was visiting my uncle. She was my cousin’s friend. When I went out with my cousin during the power cuts in the eve-nings, she would come with us. Then I started avoiding my cousin and would go alone to meet her. One day I took her to a park and had sex with her. I have only had sex with this girl; I don’t have such relations with any other girls (migrant male, 22 years). Condom Use: Findings show that although more migrants had heard of condoms than non-mi-grants (52.61 per cent vs 7.14 per cent, Chi-square= 163.42, p<0.001), among those who had ever had sex, only 10 per cent of migrants and no non-migranthad ever used a condom. Regular condom use was practically non-existent, with only four migrants reporting that they had used condoms with every sexual act. Among respondents who ever had sex, for example, 31 per cent, irrespec-tive of migrant status, reported either having sex with a sex worker or sex without a condom (Chi-square= 0.0013, p= 0.971).Young people reported engaging in un-protected sex with girls from the village because they were perceived to be completely “safe” [similar findings reported in Puri and Cleland 2006]. Even when young migrant workers are aware of forms of protection against HIV, condoms are not used because of the mispercep-tionthat village girls are clean and cannot be infected withSTIs/HIV. This is evident from the following quotes:I did not use condoms when having sex with my girlfriend because she was pure. I was also pure. I had not had sex before and it was her first experience too (male migrant, 18 years).People who go to sex workers need condoms. I have always had sex with girlfriends so there was no need to use a condom. I do not have sex with girls who are over 18 years of age or who are married. I know I am safe if she has not had sex with anyone else. All three girls I had sex with were virgins. I know by looking at a girl if she has had sex with some-one else or not (migrant male, 23 years).I only have sex with my girlfriend, whereas the sex worker has sex with 10 to 20 people every day. I don’t see any reason to use a condom (male migrant, 16 years).As seen from the qualitative data, inhibitions about condom use were widely prevalent, and in most cases sexual activity appeared to be spontaneous, which further limited the chances of condom use.I have seen condoms in the hospital when I had gone there for getting medicines. They were kept in an open box but I didn’t pick up any be-cause many people I knew were around. I didn’t want anyone to know I use them (migrant male, 20 years).I did not use condoms because they were not available nearby. Also, it was not planned when we would be having sex. It used to happen suddenly, whenever we felt like it, and condoms were not available then (migrant male, 22 years).Table 4: Odds Ratios for Effects of Independent Variables on Discussion about HIV DiscussedHIV No vs YesAge 1.15** (1.06-1.25)Socio-economic status Low Middle 1.40 (0.83-2.38) High 1.57 (0.84-2.93)Education Illiterate Primary 1.42 (0.66-3.03) Secondary 3.43** (1.63-7.22)Exposure to media No Yes 0.82 (0.53-1.25)Drug/alcohol use No Yes 0.54 (0.18-1.64)Migrant status No Yes 1.16 (0.73-1.85) Observations 7111) * p<.10, **p<.05, ***p<.001. 2) See Appendix for operational definition of variables.
REPRODUCTIVE HEALTHdecember 1, 2007 Economic & Political Weekly46Few people who live in towns know about condoms. Village people do not have much knowledge about it. Even if they know about it, they do not think of using it. Among young people, sexual intercoursehappens all of a sudden (healthcare provider, Daltongunj).As before, a multivariate analysis was conducted, using the independent variables noted in Table 5, to assess the effect of migrant status on sexual behaviour. The results indicate that migrant status is a key factor underly-ing sexual experience of youth. Migrants were significantly more likely than non-migrants to have ever had sex (OR 2.14, p<0.05), even after controlling the ef-fects of socio-demographic factors. Other key correlates of sexual expe-rience included exposure to media (OR 1.40, p<0.10), and drug or alcohol use (OR 2.53, p<0.05). Of note are findings suggesting that neither education nor age were significantly associated with the likelihood of having engaged in pre-marital sex.A multiple regression was conducted to explore the association between mi-grant status and risky sexual behaviour. The study defines risky sexual behav-iour as sex with multiple partners, sex workers, eunuchs, men or boys. The results indicate that among the sexu-ally experienced, migrant status was not associated with sex with multiple partners, sex workers, eunuchs, men or boys, and the likelihood of early sexual initiation. Migrants, in fact, were more likely to have first sex after 15 years of age as compared to non-migrants (OR 2.57, p<0.05). The corre-lates of age at sexual initiation suggest that migrant status, along with secondary education and media exposure, is significantly associated with delayed sexual initiation. In short, a larger number of migrants than non-migrants have engaged in premarital sex. However, migration is not associated with risky sex; among those who reported sexual experience, migrants did not differ from non-migrants in their likelihood of having sex with sex workers, eunuchs, male partners or multiple female partners. Moreover, migrants are more likely to engage in sexual activity at their place of origin than at their place of destination. Clearly, while migration increases the likelihood of sexual initiation, it does not increase the probability of engaging in risky sexual behaviours.DiscussionThe study findings suggest that youth in general in Palamu, ir-respective of migrant status, are poorly informed about HIV, but do engage in risky sexual practices. Beyond this, the study find-ings are in contrast to other studies that point to the tendency of migrants to engage in sexual activity and have risky and casual partner sexual relations at the place of migration [Lurie et al 2003; Wolffers et al 2002]. The findings of this study indicate that while migrants are more likely to have engaged in sexual rela-tions than are non-migrants, their sexual experiences have large-ly occurred in their areas of origin – both before and especially during regular visits home – mainly with girlfriends and relatives and not sex workers as often hypothesised. We note that some of these differences may be attributed to the fact that in contrast to other studies, our study focused only on young unmar-ried migrants. These findings, however, are not unique; another study [for ex-ample, Coast 2005] has also noted that migrants may not have sex in urban des-tination areas but can increase sexual activity when they return home. Unfor-tunately, HIV prevention programmes for migrants have typically neglected to address risks associated with migrants’ sexual behaviour at places of origin. Findings also suggest that access to greater material resources at the place of destination afforded by migration ap-pears to enable migrants’ greater sexual access to girlfriends and relatives when they return to their areas of origin as com-pared to non-migrant youth. Greater ac-cess to resources, for example, allows mi-grants to exchange material gifts for sex in non-commercial transactions with girlfriends in areas of origin; such trans-actions could mean that migrant workers may also engage in multi-partner sexual life as a symbol of economic and social success and as a form of entertainment [Chirwa 1997]. Similarly, it is likely, as other authors have speculated, that the enhanced social status of migrants also enables themto seek and receive sexual favours from women in their village of ori-gin more successfully than non-migrant men [see for example, Poudel et al 2004 for Nepal]. Finally, we argue that in settings in which notions of masculinity are profoundly intertwined with sexual prowess for men [Campbell 1997 for South Africa], the addi-tional resources at their disposal may enable migrants to display their prowess in their villages of origin [Chirwa 1997 for Malawi].Way ForwardIn summary, study findings suggest both young migrants and non-migrants engage in risky sexual behaviours and lack the knowledge to protect themselves from HIV. For example, they are poorly informed about safe sex, lack the skills to discussHIV-related behaviours with their partners or friends, and among those who are sexually experienced, both migrants and non-migrants tend to engage in risky sexual behaviours. Findings also indicate that as compared to non-migrants, young migrants from Palamu are more likely to be inadequately informed about the transmis-sion of HIV Indeed, they are at elevated risk of acquiring STIs, in-cluding HIV, for several reasons. First, they are more likely than Table 5: Odds Ratios forEffects of Independent Variables on Sexual Behaviour Indicators Ever Had Sex First Sex Had Risky Sex No vs Yes after Age 15† No vs Yes <=15 Years vs >15Years Age 1.05 0.96 (0.97 - 1.14) (0.83 - 1.11)Socio-economic status Low 1.00 1.00 1.00 Middle 1.31 1.60 0.82 (0.84 - 2.05) (0.72 - 3.54) (0.35 - 1.89)High 1.521.170.96 (0.87 - 2.66) (0.44 - 3.16) (0.35 - 2.64)Education Illiterate 1.00 1.00 1.00 Primary 1.07 1.51 0.65 (0.62 - 1.85) (0.56 - 4.05) (0.24 - 1.76)Secondary 0.942.77* 0.74 (0.53 - 1.68) (0.97 - 7.88) (0.26 - 2.10)Exposure to media No 1.00 1.00 1.00 Yes 1.40* 2.12** 1.47 (0.96 - 2.04) (1.07 - 4.21) (0.73 - 2.98)Drug/alcohol use No 1.00 1.00 1.00 Yes 2.53** 0.57 0.37 (1.22 - 5.25) (0.19 - 1.73) (0.09 - 1.45)Migrant status No 1.00 1.00 1.00 Yes 2.14** 2.57** 1.22 (1.37 - 3.34) (1.17 - 5.63) (0.53 - 2.82) Observations 7111651661) * p<.10,**p<.05,***p<.001.2) † Age omitted because of high correlation with age at first sex (0.61). 3) See Appendix for operational definition of variables.
REPRODUCTIVE HEALTHEconomic & Political Weekly december 1, 200747non-migrants to have engaged in premarital sexual relations. Second, migrants tend to be relatively less educated and of lower socio-economic status than non-migrants, implying perhaps that they have had fewer opportunities to learn about HIV and its transmission and to gain the skills to ensure safe sexual relations. Third, migrants are more likely than non-migrants to report such associated risk behaviours as drug and alcohol use. These findings raise several concerns. Above all, they suggest that the current thrust of programmes for migrants that focuses on their sexual networks in their places of destination and on sex worker contacts may not reach or be relevant to young migrants. Moreover, cultural mores and sanctions in India have so far restricted openness on matters concerning adolescent sexuality; andprogrammes on the prevention of HIV tend to be largely school-based and exclude young people, such as migrant youth, who are at higher riskofSTI and are generally not part of the formal education system.Our findings call for programmes that reach migrant youth, both at destination sites and especially in their areas of origin, that focus on interventions to empower them with information and services for protection from HIV. Moreover, research is need-ed on patterns of sexual behaviour of migrants during visits to their home village, which would have significant implications for formulating HIV policies and programme interventions.Appendix: Operational Definitions of Variables Dependent VariablesHIV awareness Heard of a disease called HIVAware that condoms prevent HIV Know that people can protect themselves from HIV by using a male condom during sexual intercourseHIV knowledge Index created by summing scores of 1 assigned to correct responses to questions about knowledge of HIV, including whether HIV can be transmitted during condom-less sex/unprotected sex/sex without a condom, sex with a sex worker, sex with multiple partners, sex with an HIV-infected person, sex with an STI-infected person, condomless same sex experiences/acts, sharing needles and blood transfusion. Lack of common misconceptions about transmission of HIV, for example, by sharing clothesand utensils, using common toilets and bathrooms, by mosquito bites, touching an infected person, sharing food with an infected person and having sex with ahealthy looking person. Knowledge about prevention of HIV including abstinence, reducing the number of partners, remaining faithful to one partner, abstaining from condomless homosexual acts, not sharing needles and using condoms was scored as follows: 0=no knowledge, 1-13=low knowledge, 14-20=high knowledge. Communication about HIV Discussed HIV with a partner or friend in the past monthEver had sex Ever had sexual intercourseFirst sex after age 15 Had first sex after 15 years of ageRisky sex Had sex with sex workers, eunuchs, men or boys, or with more than one female partner Independent VariablesAge Respondent’s age in completed yearsSocio-economic status The socio-economic index was created by assigning the following scores to household assets: Type of house: 4 for ‘pucca’, 2 for semi-‘pucca’ and 1 for ‘kachha’ Fuel used for cooking: 4 for electricity, biogas or liquid petroleum gas, 2 for coal and 0 for other fuel. Ownership of land: 2 for yes, 0 for no. Ownership of vehicle: 4 each for tractor or four-wheel vehicle, 3 for three-wheel vehicle, 2 for bullock cart.Ownership of durable goods: 4 for water pump, 3 each for radio, television, VCR, cable, washing machine, thresher and others, 2 for fridge, 1 each for ‘charpai’, cot-bed or table. The scores were summed to obtain an index. The index was divided into three categories which denoted low, middle and high socio-economic status. The index scores range from 0-6 for low socio-economic status, 7-12 for middle socio-economic status and 13-31 for high socio-economic status.Education Three categories denoting illiterate, primary (primary,upper primary or non-formal education) and secondary and above (secondary, higher secondary, graduate and above graduation)Exposure to media Watched television in spare time or subscribed to a newspaperDrug/alcohol use Used drugs or alcohol in spare timeMigrant Person who migrated to urban or rural areas for employment, had not relocated to the place of destination, stayed without the family at the place of destination for more than a month, visited place of origin at least once in the past year and was located at the place of origin for at least five yearsNote 1 Calculated from Census of India, 2001 [RGI 2001a], migration data for Jharkhand.ReferencesBrockerhoff, M and A E Biddlecom (1999): ‘Migration, Sexual Behaviour and the Risk of HIV in Kenya’,Inter-national Migration Review,33(4), pp 833-56.Campbell, C (1997): ‘Migrancy, Masculine Identities and AIDS: The Psychosocial Context of HIV Transmission on the South African Gold Mines’, Social Science and Medicine, 45(2), pp 273-81.Chakraborty, S (2004): ‘Risk Behaviour to HIV/AIDS among Migrant Workers: A Case Study of Surat, India’, eJour-nal of the International AIDS Society, 1(1), pp D12429.Chirwa, W C (1997): ‘Migrant Labour, Sexual Networking and Multi-partnered Sex in Malawi’,Health Transition Review, 3(Suppl 7), pp 5-15.Coast, E (2005): ‘Sexual Behaviour and Perceptions of Risk: Male Rural-to-Urban Migrants to Tanzania’, presented at IUSSSP XXV International Population Conference, July 18-23, Tours, France.Dayal, H and A K Karan (2003): ‘Labour Migration from Jharkhand’,Labour and Development, 9(2), pp 223-58.Delor, F and M Hubert (2000): ‘Revisiting the Concept of ‘Vulnerability’,Social Science and Medicine, 50(11), pp 1557-70.Gras, M J, B H van Benthem, R A Coutinho et al (2001): ‘Determinants of High-risk Sexual Behaviour among Immigrant Groups in Amsterdam: Implications for Interventions’,Journal of Acquired Immune Deficiency Syndromes, 28(2),pp166-72.Gujjarappa, L and S Rasalpurkar (2004): ‘Sexual Behav-iour among Migrant Hotel Workers in Pune City, Ma-harashtra, India’,eJournal of the International AIDS Society, 1(1), pp TuPeC4816.Gurubacharya, D L and V L Gurubacharya (2004): ‘HIV Prevalence among Nepalese Migrant Workers Work-ing in Nepal and Indian Cities’,Journal of Nepal Medi-cal Association, 43, pp 178-81.International Institute for Population Sciences (IIPS and ORC Macro (2000):National Family Health Survey (NFHS-2), 1998-99; India, IIPS, Mumbai.Jejeebhoy, S J (1998): ‘Adolescent Sexual and Reproduc-tive Behaviour: A Review of the Evidence from India’, Social Science and Medicine, 46(10), pp 1275-90.Lurie, M N, B G Williams, K Zuma et al (2003): ‘The Im-pact of Migration on HIV-1 Transmission in South Africa: A Study of Migrant and Non-Migrant Men and Their Partners’,Sexually Transmitted Diseases, 30(2), pp 149-56.Lyttleton, C and A Amarapibal (2002): ‘Sister Cities and Easy Passage: HIV, Mobility and Economies of Desire in a Thai/Lao Border Zone’,Social Science and Medicine, 54(4), pp 505-18.Operations Research Group (ORG) (2005): ‘Sexual Behav-iour of Young Migrants in Palamu District, Jharkhand’, ORG Centre for Social Research,NewDelhi.Poudel, K C, M Jimba, J Okumura et al (2004): ‘Migrants' Risky Sexual Behaviours in India and at Home in Far Western Nepal’, Tropical Medicine and International Health,9(8), pp 897-903.Puri, M and J Cleland (2006): ‘Sexual Behaviour and Per-ceived Risk of HIV/AIDS among Young Migrant Facto-ry Workers in Nepal’, Journal of Adolescent Health, 38(3), pp 237-46.Registrar General, India (RGI) (2001a): Census of India 2001: Data Highlights, Migration Tables: D1, D1 (Ap-pendix), D2 and D3 Tables, RGI, New Delhi.– (2001b): ‘Provisional Population Totals: Jharkhand’, Census of India, 2001, Series-21, Paper 1 of 2001, RGI, New Delhi. – (2001c): ‘Provisional Population Totals: Jharkhand’, Census of India, 2001, Series-21, Paper 2 of 2001, Ru-ral-urban Distribution, RGI, New Delhi. – (2001d): ‘Primary Census Abstract’, Jharkhand (00-57-2001-Cen-CD), RGI, New Delhi.Singh, D, A Adrien, R Washington et al (2004): ‘Baseline Results from HAMARA, A Project on HIV/AIDS and Rural Migration in Rajasthan, India’,eJournal of the International AIDS Society, 1(1), D10789.Solomon, S, A Ganesh, M Ekstrand et al (2000): ‘High HIV Seropositivity at an Anonymous Testing Site in Chennai, India: Client Profile and Trends over Time’, AIDS and Behaviour, 4(1), pp 71-81.UNAIDS (2001): ‘Population Mobility and AIDS’, UNAIDS Technical Update, Geneva. – (2006): ‘HIV Epidemic in India’, available at accessed on April 20.UNICEF, Joint UN Programme on HIV/AIDS and WHO (2002): Young People and HIV/AIDS: Opportunity in Crisis, New York, Geneva.United Nations Population Division (2001): ‘World Popula-tion Prospects: The 2004 Revision’ and ‘World Urbani-sation Prospects: The 2003Revision’, available at accessed on June20.Wolffers, I, I Fernandez, S Verghis et al (2002): ‘Sexual Behaviour and Vulnerability of Migrant Workers for HIV Infection’, Culture, Health and Sexuality, 4(4), pp 459-73.

To read the full text Login

Get instant access

New 3 Month Subscription
to Digital Archives at

₹826for India

$50for overseas users


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top