Does Powerlessness Explain Elevated HIV Risk Amongst Tajik Labor Migrants? An Ethnographic Study

Abstract

To investigate the role of powerlessness in elevated HIV risk among labor migrants, we compared internal verses external male migrant workers from Tajikistan using minimally structured interviews and focused field observations. The sample included 30 male labor migrants who traveled to work in Regar, Tajikistan (internal labor migrants), and 30 who traveled to work in Moscow, Russia (external labor migrants). Though powerlessness did not appear to account for whether labor migrants engaged in more HIV risk behaviors, the harsh living and working conditions of external labor migration impacted how the migrants manifested these HIV risks by amplifying group masculine norms and behaviors. Progress in preventing HIV infection amid the difficult social conditions of labor migration is contingent upon adequate conceptualization of how such conditions impact HIV risk behaviors

BACKGROUND

According to the United Nations Population Division (UNPD, 2009), in 2005 there were 191 million international migrants—representing about 3% of the world's population. Previous studies have found an association between the high mobility of migrants and an increased risk for HIV infection (Li et al., 2004; Liu et al., 2005; Ramjee & Gouws,2002; Thorne, Ferencic, Malyuta, Mimica, & Niemiec, 2010). Several factors have linked migration with increased HIV risk, including: separation from wives, family, and other support networks (Magaña, 1991); increased number of casual sex partners (Lurie et al., 2003); more money or higher socioeconomic status (Wardlow, 2007); lack of access to social and health services (Ku & Matani, 2001); limited knowledge about HIV/AIDS transmission and prevention (Fitzgerald, Chakraborty, Shah, Khuder, & Duggan, 2003); and hardship and hostility while in the receiving country (Weine, Bahromov, & Mirzoev, 2008). However, the particular social, economic, psychological, and cultural processes that might explain how labor migration leads to increased HIV risk have not been well studied.

With 80% of the country living below the poverty line and a GDP per capita of less than $290, the Central Asian Republic of Tajikistan is one of the poorest countries in the world (Godinho, 2005). Faced with difficult economic challenges resulting from a 5-year civil war that resulted in over 60,000 deaths and widespread unemployment, an estimated 700,000 (range 600,000 to 1,000,000) men leave Tajikistan each year in search of work, especially in Russia (Godinho, 2005; Mirzoev, Green, & Newell, 2007; Thorne et al., 2010). Most Tajik migrants in Russia are seasonal workers in Moscow, carrying goods in bazaars (arobakash), performing manual labor in construction sites, sweeping streets, and working in food services (Eurasianet, 2006). Tajik labor migrants may engage in high-risk sexual behaviors such as having multiple concurrent sexual relationships (Amirkhanian et al., 2010; Godinho, 2005; Morris & Kretzschmar, 1995; Weine et al., 2008). The migrants in Moscow also often use sex workers, and these sex workers have been documented to have an estimated HIV prevalence rate 30 to 120 times higher than that of the city's general population (Shakarishvili et al., 2005).

Eighty-five percent of migrants aged 20 to 49 years old are married (Godinho, 2005). The wives of migrants are at risk of being exposed to any illnesses that the migrants carry, including sexually transmitted infections such as HIV. Thus, labor migrants could serve as a bridging population and thus contribute to the acceleration of an epidemic in Tajikistan that has risen dramatically since 2000 (Hamers & Downs, 2003). In May 2011, 3051 HIV cases were registered in Tajikistan, of which 145 (10%) were migrant workers (Tajikistan HIV/AIDS Prevention Center, 2011). An increased number of HIV cases has also been observed among wives of male Tajik migrants following their return home (Rafiev, Mirzoev, Abbasova, & Lukyanov, 2006). The available statistics are widely believed to underestimate the true rate of HIV infection in Tajikistan.

In the field of HIV prevention, social psychology theory has been used to inform the design, implementation, and evaluation of behavioral interventions seeking to reduce HIV risk (Fishbein et al., 1997; Jemmott, Jemmott, & Fong,1998; Kamb et al., 1998). Theory can help explain key and often complex psychological and social determinants of risk and preventive behavior, such as attitudes, perceived norms, and self-efficacy, defined as the belief that one has the skills and abilities necessary to perform a behavior in a variety of circumstances (Fishbein, 2000). With regard to labor migrants, social psychology theory has been utilized to inform HIV prevention and behavior change educational programs for labor migration under adverse and complex social conditions. For example, Campbell (1997) used social identity theory (defined as a subtype of social psychology theory that focuses on the cognitive or motivational processes involved in self-categorization and group formation (Ashforth & Mael, 1989) to guide qualitative studies that described how the construction of masculine identities among South African male mineworkers rendered them especially vulnerable to HIV.

Campbell found that masculine identities and macho sexuality helped mineworkers to deal with the harsh day-to-day living and working conditions within the mines. She defined macho sexuality as a repertoire of insatiable sexuality, need for multiple sexual partners, and a manly desire for the pleasure of flesh-to-flesh sexual contact (Campbell,1997; Christianson, Lalos, Westman, & Johansson, 2007). Specifically, she delineated how difficult living and working conditions brought on a sense of “powerlessness” in the mineworkers, which accounted for their decreased self-efficacy regarding HIV protection. Campbell wrote, “for many migrants, deprived of key markers of masculinity on a day-to-day basis, frequent assertion of what are regarded as healthy and manly sexual urges could arguably serve to compensate for reduced opportunities for assertion of masculine identities in other contexts” (Campbell, 1997, p. 279).

Prior studies of women living in different contexts have examined the concept of powerlessness in sexual health promotion for HIV prevention. For example, Tawil, Verster, and O'Reilly (1995) argued that addressing the powerlessness of women to protect themselves against HIV requires an “enabling approach” that creates circumstances facilitating HIV risk reduction. For example, one strategy may be to focus on improving women's access to resources so that they may be more financially independent from their male partners (who may be reluctant to use condoms). Another study examined the prospects for community-based interventions to reduce HIV risk among female sex workers in Tamil Nadu, India, and suggested that these approaches were unlikely to succeed given the high degree of repression, taboo, social stigma, and fragmentation surrounding sex work in the city of Madras (Asthana & Oostvogels, 1996). This study concluded that individual level empowerment strategies (derived from stories of success among middle-class gay men in the West) were inadequate to address the complex social dimensions, such as economic structures, cultural traditions, gender roles, and societal institutions underlying HIV risk for sex workers in other countries. These studies underscore the importance of social and cultural contexts in understanding powerlessness and its possible role in HIV prevention. They also suggest that it is necessary to consider both actual and perceived powerlessness. Thus, for the purposes of this study, we focused on whether actual powerlessness (poor living and working conditions) and/or perceived powerlessness (low self-efficacy) was associated with sexual risk behaviors for HIV among labor migrants.

The possible link between difficult conditions, powerlessness, and HIV risk could in part explain the elevated HIV risk behaviors among external Tajik male labor migrants (Weine et al., 2008). Tajik migrants working in Moscow often speak of being “unprotected” (Weine et al., 2008). Unprotected means being without legal status, subject to arrests, imprisonment, beatings, and bribery, and without adequate housing, safe working conditions, or access to healthcare. These harsh living and working conditions caused us to hypothesize that Tajik male migrants could be practicing macho sexuality or something like it to compensate for reduced masculine role functioning while in Moscow. Thus, this study was an opportunity to assess the applicability of Campbell's claims regarding powerlessness based on social psychological theory.

RESEARCH QUESTIONS AND DESIGN

In accordance with International Organization for Migration (IOM, 2004) and United Nations Educational, Scientific and Cultural Organization (UNESCO, 2005) definitions, Tajik citizens who travel to work in Tajikistan are internal migrants and those who travel to work in Moscow are external migrants. To investigate the possible role of powerlessness in HIV risk for labor migrants, we took advantage of an opportunity to compare Tajik internal and external migrants. Specifically, for the internal sample, we focused on Tajik men who traveled to the Tajik city of Regar, where one of Central Asia's largest aluminum factories supplies many well-paid jobs and dormitories for the men to live in. Our research design was to compare internal and external migrants using ethnographic data and qualitative methods (Miles & Huberman, 1994). We hypothesized that external migrants in Moscow would experience powerlessness to a greater degree than internal migrants in Regar and that this would result in higher HIV risk behaviors. We sought to answer the following research questions:

1. How do social conditions differ between internal and external labor migrants? Do internal and external labor migrants differ in their experience of powerlessness?

2. Do internal and external labor migrants have similar sexual and HIV risk and protective behaviors?

3. If there are differences in HIV risk and protective behaviors, can those differences be explained by “powerlessness,” or is there some better explanation?

METHODS

Procedures and Sample

To respond to the public health risk of HIV in Tajikistan, we formed a collaboration between U.S.-based researchers from the University of Illinois at Chicago (UIC) and health professionals from the Ministry of Health of Tajikistan. The evidence presented here comes from ethnographic data collected by this collaboration between October 2005 and August 2007 through the support of the United States Civilian Research and Development Foundation. Data collection consisted of in-depth, qualitative interviews carried out jointly by a multidisciplinary, collaborative team of field workers that included an American psychiatrist, an American medical student, a Tajik physician and public health official, a Tajik journalist, and a Tajik educator. All team members participated in at least 20 hours of collaborative training that focused on recruitment and interview techniques and HIV prevention, migration, and sociocultural issues. To further enhance the cultural understanding among the research team, all members conducted readings, participated in group discussions with key informants, reviewed and discussed prior research on Tajik migrants, and conducted pilot interviews. Consent was obtained by using procedures approved by the institutional review board at the University of Illinois at Chicago as well as the Ministry of Health of the Republic of Tajikistan.

Minimally structured interviews and focused field observations were conducted with 30 married male internal migrants in Regar, Tajikistan, and 30 married male external migrants who worked in Moscow. All subjects were identified through purposive sampling at sites in Dushanbe or Regar.

The subjects for interviews in Dushanbe were identified with the help of staff members from the Republican HIV/AIDS Prevention Center. This site was selected because it is the location where the majority of Tajik migrant workers go in order to undergo HIV testing and counseling prior to leaving the country. Although there were several regional sites in the provinces of Gorno-Badakshan, Khatlon, and Sugd where migrants could receive an HIV test, the majority went to Dushanbe for testing because it is the main point of departure for flights and trains to Moscow. It is further worth noting that while Russian and Tajik policies aimed to have all labor migrants traveling to Russia tested for HIV, in practice, many migrants bypassed this requirement by entering Russia on tourist or student visas or by paying for certificates without being tested. The interviews with external migrants took place during the season when they resided in Tajikistan, which was typically from December to April. These interviews were conducted in a private room at the HIV/AIDS Prevention Center. Only a small number of interviews were conducted each day so as to reduce the likelihood that subjects could be identified by others as research subjects. Nonetheless, we informed subjects that there was a possibility that other migrants may find out that they were participating in our research project.

The subjects for interviews in Regar were identified with the help of a local physician who worked in the government health cabinet that provided health care services to the majority of the workers at the aluminum factory as well as to local sex workers. To protect confidentiality, most of the interviews in Regar took place in private rooms of the physician's clinic. Two interviews were conducted in a semiprivate tent of an outdoor restaurant near one of the factory worker's dormitories.

External migrant subjects were only included if they gave verbal and/or written consent and met the following criteria: (a) male migrant from Tajikistan between 18 and 59 years old; (2) married or engaged to a woman in Tajikistan; (c) seasonal or temporary migrant workers who return to Tajikistan at least once per year; and (d) risks (at least two of the following): (i) recent unprotected sexual contacts in Moscow; (ii) present or past sexually transmitted disease (STD); (iii) present or past tuberculosis (TB); (iv) present or past intravenous drug use; and (v) present or past alcohol abuse. The criteria for internal migrants were the same as those for external migrants with the exception of criterion “c” (which indicated that subjects must be “seasonal or temporary migrants employed at the aluminum factory who return to their home villages at least once per year”) and criteria “d, I” (which specified “recent unprotected sexual contact in Regar”). Of note, in the process of screening male migrants for risks, only three persons from the external migrant cohort of 33 potential participants, and one person from the internal migrant cohort of 31 were excluded for not meeting criteria.

We found important demographic distinctions between internal and external migrants (Table 1). Internal migrants were older and more highly educated and had more children. External migrants were younger and less educated and had fewer children. Many external migrants were not married because many Tajik men go to Moscow in order to earn enough money to start a family.

Table 1.

Demographic Characteristics of Internal and External Migrants

Characteristic	Internal migrants (n = 30)	External migrants (n = 30)
Mean age, yr (SD)	38.8 (11.44)	28.73 (8.17)
Education, n (%)
Secondary school	5 (17)	17 (57)
College degree	12 (40)	3 (10)
University degree	5 (17)	10 (33)
Technical degree	8 (26)	0 (0)
Region of origin	All 5 regions	All 5 regions
Marital status
Married, n (%)	30 (100)	26 (87)
Engaged, n (%)	0 (0)	4 (13)
Mean number of children, n (range)	2.93 (0–7)	2.23 (0–5)

All interviews were conducted in either Tajik or Russian. The reason for using Russian is that some of the internal migrants spoke Uzbek and Russian, but not Tajik. The interviewer in Dushanbe was a Tajik physician. The interviewers in Regar were either the Tajik physician (MB) or, for 14 of the 30 interviews, a trained American medical student assisted by a Tajik translator. The interviews lasted 30 to 90 minutes and were audio-taped and transcribed into English. The content of interviews focused on the following domains: work and living conditions in Regar or Moscow; family in Tajikistan; HIV risk and protective behaviors; HIV risk and protective knowledge and attitudes; spousal communication; and alcohol and drug use. For example, we asked: “What are your living arrangements like?” “What is your opinion about condoms?” “Do you talk to your wife about the girls you see here?”

Methodological Approach

Our study utilized a grounded theory approach to qualitative research (Charmaz, 2006). The initial study questions were refined through an iterative process of data collection and analysis that followed standardized qualitative methods (Miles & Huberman, 1994) and used ATLAS.ti, a qualitative data analysis program (Muhr, 2004). After establishing coder reliability, labels were attached line-by-line to segments of transcripts in order to form codes. Codes were systematically compared with one another and similar codes were classified into concepts, categories, or processes (Bryant & Charmaz, 2010). Coders also wrote memos or extended notes, which further explained key constructs or identified gaps in understandings. Through this approach of comparing codes, writing memos, and developing concepts, categories, and processes, a grounded theoretical model emerged (Miles & Huberman, 1994). This report summarizes the key components of that model. To ensure the trustworthiness of our findings, codes and concepts were reviewed by the entire research team at multiple points during and after data collection and data analysis to enable checking for other explanations and contrary evidence. Additionally, as part of the iterative process of research, when gaps or questions emerged, we addressed them by collecting new data, conducting further data analyses, and consulting key informants.

RESULTS

The grounded theory model that was derived through this investigation of Tajik male labor migrants represents the findings that external migrants described being “unprotected” due to harsh living and working conditions in Moscow. However, both external and internal migrants reported regularly having unprotected sex with sex workers when away from home. Thus, the model demonstrated that powerlessness did not directly explain whether or not migrants regularly had unprotected sex with sex workers. Yet it did demonstrate how the difficult social environment in Moscow was associated with a somewhat different manifestation of HIV risk behaviors. In the face of harsh conditions, external migrants’ actions were more highly determined by masculine group norms and behaviors (concurrent sexual partners, peer pressure to share sex workers, parties with heavy alcohol use, and shared reluctance to use condoms). In contrast, internal migrants were more likely to individually solicit sex workers and to use condoms based on each man's own sense of a woman's cleanliness.

Table 2 summarizes the differences between internal and external migrants, and Figure 1 presents the grounded theory model derived through qualitative analysis of ethnographic data to explain these differences. In the sections that follow, the elements of this model are explained in greater detail with illustrative quotations from both internal and external migrants.

Table 2.

Differences Between Internal and External Migrants

	Internal migrants (n=30)	External migrants (n=30)
Living and working conditions	Registered workers in one’s own country Live in dormitories with 1–3 others per room Earn $50–$400/month Alcohol in moderation	Unprotected by law Unsafe living conditions Migrants join social networks in diaspora Earn $500–$700/month Heavy alcohol use during parties
HIV/AIDS knowledge	Variable HIV knowledge Does not regard HIV as a big risk in Tajikistan Testing is available, but not used	Variable HIV knowledge Does not regard HIV as a big risk in Russia Compulsory HIV testing
Sexual behavior	Choice is selecting sex workers Mama Rosa provides girls Alcohol use separate from sexual experiences Men expected to have sex when away from wives/home	Sharing sex workers Do not use condoms when intoxicated Men expect to have sex when away from wives/home
Condom use	No pleasure with condoms No difficulty obtaining condoms Inconsistent condom use	Do not feel anything with condoms Difficulty obtaining condoms due to fear of police Unprotected sex, despite knowing that sex without condoms is dangerous

Figure 1.

Grounded theory model representing similarities and differences between internal and external migrants.

Living and Working Conditions

The external migrants working in the bazaars lived in overcrowded, unsanitary, and poorly heated dormitories or apartments. One migrant said, “Fifty people live in one dormitory. The conditions are terrible, there is no ventilation system and it is very difficult to breathe. We do not even have a bed.” External migrants lived together in order to save money and because they felt safer when in large numbers. Another said, “Ten people were living in one room as it was slightly bigger. We all manage to live together. As you know, it is difficult to find rooms for rent because migrants are not trusted. Also, it is expensive for one or two people to rent a place in Moscow.” Most migrants were working in Moscow illegally, and they were subject to frequent arrests, imprisonment, bribes, beatings from police, and unprovoked attacks from skinheads and ultranationalists. One migrant said, “If Russians want, they can kill you, beat you and say whatever they want. A Tajik does not have the right to say a simple word. Tajiks are unprotected.”

Most of the Tajiks who worked in construction reported that they lived in a barak or wagon. A barak is an old building within the construction site that would eventually be demolished. While construction was going on, the firm owners offered these old buildings to house the Tajik migrants. Wagons were one-room units including both a kitchen and sleeping area that were also located within the construction sites. The Tajik men reported that there have been many cases of fires in the wagons where Tajik migrants were burned and killed. Those working in construction were also subject to workplace accidents at the construction site for which they could not get medically treated in Moscow. As a consequence, external migrants lived with the daily fear that they would get seriously injured or die. One said, “I saw myself in the skin of a slave in Moscow. We are unprotected.”

The external migrants were seasonal migrants who typically departed for Moscow in April and returned home in December. Most external migrants reported being part of social networks organized around their extended family or their village. In the harsh conditions of life in Moscow, the external migrants were even more dependent upon these networks than they would be if living in Tajikistan. For example, one migrant said, “We listen to our village leaders. They keep us united, which is very important for us today. We are far from our homes and need support from each other. Nobody can survive alone in this terrible society. You have to belong to a network.” These networks helped migrants through sharing information, providing social support, and problem solving. For example, one workgroup leader said, “During the gatherings at work or when we have a party, I can talk to the migrants. If somebody gives information to people there, they can share this information with others.”

All the internal migrants interviewed in Regar worked at the aluminum factory and lived in a nearby dormitory or apartment. As part of their work contracts, most internal migrants were offered discounted rent at the factory-owned dormitories. One internal migrant described his living arrangements, “There are two people in each room. I live with my coworker and friend.” Another migrant described the dorms as being affordable and acceptable: “We have two rooms for three people … if the dormitory belongs to the factory—it costs five or six somoni per month … many people who live here are working in the factory.” In the city of Regar, the internal migrants felt at home and did not face any persecution from police or locals.

The internal migrants spoke positively about the work in the factory. “There is no work in the place where I live. You are only able to work in the farm. Here at the factory, I have good work.” Many migrants said they came to the factory in Regar because they had technical skills or degrees that could earn them higher wages than in their home communities. One said, “I decided to come to Regar. The payment at the factory is better than anywhere else. The factory is paying me good money and I am feeding my children.” The work at the aluminum factory was done in shifts with brigades of 20 to 30 workers. For each shift, the work was divided among the different members of the work brigade: “My work consists of melting metal … some workers clean, some produce locks … there are different tasks within one shift.” Typically, each shift lasted 15 days. Many internal migrants returned home for 15 days after working two shifts (for a total of 1 month).

Internal and external migrants also differed in their patterns of alcohol use. External migrants reported drinking alcohol in large parties held in their apartments or living quarters. One migrant explained, “Usually we invite girls on Sundays when we arrange good parties. We used to have parties occasionally on birthdays or during national holidays. We sit together and drink different alcoholic drinks, have fun and later dance with girls who we bring in from the street.” In contrast, most internal migrants reported either not drinking alcohol or drinking in relative moderation. None of the internal migrants interviewed reported getting intoxicated in conjunction with or during sexual experiences. When one internal migrant was asked how often he consumes alcohol, he replied, “Two or three times per week, because I am bored.” Another said, “I drink vodka with friends … but not so much, just for pleasure.” Neither any external or internal migrants reported using any drugs.

Both external and internal migrants expressed the same kinds of beliefs regarding the role of men in male-dominated and Muslim-influenced Tajik society. They spoke about respect for elders, honoring village traditions, and protecting their family. A strong man, they described, is someone who is able to be a breadwinner and decision-maker for his family. One notable difference reported by internal and external migrants with respect to beliefs about men is that external migrants had more difficulty embodying these beliefs about men's roles given their separation from their wives and children and their being a part of an impoverished minority group in an alien society.

HIV/AIDS Knowledge

Knowledge of HIV/AIDS was variable and vague among the external migrants. One said, “No I don't know [much] about HIV/AIDS. I have no good information about it, really. I know that it is transmitted through girls who are dealing with sex.” Another said, “I read in a magazine that using condoms is very useful for preventing HIV/AIDS and it is an easy way to protect yourself when you having sex with sex workers or other women. We have a proverb in Tajik language that says: be careful of incidents and solve problems before they happen. Therefore having enough knowledge about AIDS and using condoms primarily are very good ways for migrants to prevent themselves from being infected by HIV.”

Like external migrants, internal migrants also had variable and vague knowledge about HIV/AIDS. When asked how a person could become infected by HIV, one migrant said, “Truly speaking, I don't know.” While some internal migrants were not sure of how HIV was transmitted, others demonstrated a greater degree of knowledge. One migrant explained, “If a migrant looks after himself, and takes care of himself, does not use dirty syringes, and refrains from sex … I think these are some of the ways to prevent HIV.”

Neither internal nor external migrants considered HIV/AIDS to be a major problem for Russia, Tajikistan, or other Central Asian countries. An external migrant reported, “AIDS is a big problem in Africa but not here.” An internal migrant said, “I think there are some HIV positives but it is not a large problem. If it was problem there would be more news [about this issue].” Another internal migrant stated, ” … but I was never infected by any of them, thank God. I heard that there is also a dangerous disease called AIDS and it is transferred by sexual contact if someone doesn't use condoms. But I think that this disease is not common in Tajikistan.”

Internal migrants had positive opinions about HIV testing but rarely reported ever getting an HIV test themselves. One migrant said, “I only know that there is a place in Dushanbe where one can get a HIV test, but I did not go there. I have never been tested. But I think that every migrant should have a HIV test.” None of the migrants interviewed reported testing positive for HIV.

Sexual Behavior

Most external migrants in Moscow reported regularly having sex with sex workers. In order to reduce the risk of being caught by the police, a leader of a group of migrants often arranged for a sex worker to meet the group back at their living quarters. Sex workers may be women needing food, vodka, or drugs (these women are called sarifak in Tajik orbumzh in Russian), or less commonly a woman from the tochkas or another sex worker operation (tochkas are huts or stands where relatively higher-class sex workers wait for clients, keep warm, and talk with other sex workers). One migrant described how the cost of a sex worker was split between many migrants: “One girl cost $200, and there were 20 guys in our wagon, so she agreed to sleep with us for a night. Then each of us paid her $10.” Another external migrant explained, “You know when a person is drunk and women are present, they obviously will have sex. During that night we had to exchange our women.”

The external migrants described how heavy alcohol use impaired their judgment and reduced condom use. One migrant said, “We drink vodka and want girls. When we are drunk we never use condoms.” Another said, “When I am drunk, I can't tell the difference between A from B … then how can I use condoms?” In contrast, internal migrants drank alcohol in moderation and none reported drinking during sexual contacts, including with sex workers.

Unlike their external counterparts who often had to share women, internal migrants demonstrated autonomy in choosing where and when to engage in sexual contacts with sex workers in Regar. One factory worker described clearly the different choices of sex workers that existed for migrants: “There are three types of sex workers—sex workers who stay in cafes and restaurants. These are cheap sex workers. Another group of sex workers are more expensive and work independently. You call them and negotiate the price over the phone. The third group are the sex workers who work through Mama Rosa [Mama Rosa is the name given to madams who negotiate with clients for the services of sex workers]. I always choose sex partners through Mama Rosa. I think this is safer because Mama Rosa is responsible for the health of the sex workers.”

Internal migrants in Regar often engaged sex workers through telephone networks, for example: “My friend has the phone number of a Mama Rosa and she provides us girls for 20 or 30 somoni for two or three hours of sex.” Sometimes frequent clients were identified and were the recipient of calls from women soliciting sex work: “If you have money, they will offer to go with you somewhere. They call us many times.”

Both external and internal migrants expressed the belief that men were expected to have sex when away from their wives. One internal migrant said that sex with women when away from wives was natural: “Something is inside a man that he needs to be with a woman. It is natural!” Another migrant claimed that the lack of regular sex caused referred physical pain: “My wife is far away. That's why I need to have sex. If I haven't had sex for many days I can feel pain in my back.” An external migrant explained that it was part of the Tajik tradition of masculinity to have sex with other women when away from home: “It is in our tradition that a man keeps his sexual activities secret and it is a sign of a real man, mardi haqiqi [masculinity]. My wife and other women know that when their husbands are away in Russia they obviously will have sex.”

Condom Use

Both internal and external migrants reported that condoms reduced pleasure during sex. One external migrant said, “A person has to feel that he is having sex, but with condoms you do not feel anything.” Similarly, one internal migrant thought that condoms reduced pleasure because they were not natural. He said, “I don't need them. You cannot take pleasure with condoms. It is not natural and isn't pleasurable for me.”

External migrants cited several additional reasons for their unwillingness to use condoms. Some reported being afraid of being caught by the police while out buying condoms: “I do not have time to go out and buy condoms. It is dangerous and the police might catch you.” Another said, “We really could not go to a shop or drug store to buy condoms, because policemen would definitely catch us, take a lot of money, or deport us.” Other external migrants said that they sometimes did not speak the same language as the sex workers, many of whom were migrants from other countries, including Vietnam and Korea. In contrast, internal migrants did not report any fear of police and did not have difficulties in obtaining condoms or communicating with the sex workers. One internal migrant commented on the wide availability of condoms in Regar, “In every kiosk there are condoms, and I always buy them myself. Everywhere you can find condoms in the drug stores.”

Though most external migrants could correctly identify condoms as protective against HIV and other sexually transmitted infections (STIs), many continued to have unprotected sex. One external migrant said, “Everybody says that sex without condoms is dangerous, but nobody changes their behavior. Even I have sex with prostitutes and don't use condoms. Knowing is different from doing.”

Internal migrants reported making more independent and deliberate choices regarding whom to use condoms with. One migrant spoke of being able to detect whether or not a woman could have an STI: “If she is young—and you are suspicious, you can use condoms. If you have no suspicion, or you know her very well, you don't need to use condoms.” Many internal migrants used condoms based on their sense of a partner's cleanliness or style of dress. One migrant said, “Sometimes, I used condoms with sex workers but never with my girlfriend. I just look at the girl I am going to have sex with, and if she is clean I don't use condoms. If I see that she is not clean or not well dressed, I use condoms.” Another said, “You can know that from their clothes or their cleanliness. I ask them if they have any diseases. But with sex workers I mostly use condoms.”

DISCUSSION

This study aimed to explore the relationship between difficult living and working conditions, sense of powerlessness, and HIV risk and protective behaviors among Tajik external and internal labor migrants. The ethnographic findings revealed that due to differences in the social environments of the receiving country, external migrants experienced more harsh living and working conditions that internal migrants did not, including having no legal status; being outsiders in an unfamiliar city; arrests and beatings from police and attacks from skinheads and ultranationalists; overcrowded and unsanitary housing; and lack of access to health care. Tajik external migrants regarded themselves as “unprotected” and unable to change this situation, which is similar to what Campbell (1997) described as powerlessness among South African mine workers. In contrast, Tajik internal migrants described their social and work environments in more positive ways. However, external migrants’ powerlessness did not appear to be associated with elevated HIV risk, as both external and internal migrants from Tajikistan reported engaging in HIV risk behaviors including (a) concurrent sex partners through the use of sex workers when working away from home; (b) vague and variable knowledge about HIV risk and protection; and (c) reluctance to use condoms. Other possible explanations of the high-risk sexual behaviors for external and internal migrants alike should be considered, such as simply that pleasure seeking trumps self-protective behaviors.

This study found other important differences in social environment that appeared were to impact how HIV risk behaviors were manifested among migrants (Figure 1). Most external migrants belonged to social networks defined by their extended family or village in Tajikistan. They lived in close quarters with other Tajik migrants who provided much needed support amid adverse conditions. These close living arrangements also reinforced group masculine norms and behaviors such as heavy alcohol use, concurrent sex with multiple partners, sharing of sex workers, and reluctance to use condoms. In contrast, internal migrants lived more independently, exercised more autonomy in selecting the type of sex worker (café, street, or Mama Rosa), had little to moderate alcohol use, and used condoms based on their sense of a woman's cleanliness. However, our findings indicated that both internal and external migrants were at elevated risk for HIV because both groups had unprotected sex with multiple concurrent partners.

Our study has important implications for building theory about migrants and HIV risk that involves powerlessness. Campbell's contextual formulation of powerlessness was helpful in explaining the interpretative repertoires by which South African mineworkers shaped their own social and sexual identities (Campbell, 1997, 2000). She described how the mine workers experienced powerlessness and a lack of control in a variety of contexts, from mine accidents to health problems (Campbell, 1997). While their concept of powerlessness was an appropriate starting point for this investigation, and did resonate with the Tajik external migrants’ sense of being “unprotected,” it did not appear to be as useful in the context of Tajik labor migration as an explanation for HIV risk.

We do not want to trivialize the experience of powerlessness among labor migrants in South Africa or Russia. However, our results point toward limitations in theorizing powerlessness as a psychological factor that could become removed from social context. Rather, we find advantages in viewing powerlessness through an ecological lens, both as being strongly linked to social context, and because powerlessness is one of several multilevel consequences of men living and working in groups amid difficult living and working conditions.

Previous studies of migrants and marital HIV risk in rural Mexico also utilized an ecological approach (Hirsch et al.,2007). The authors argued that in the social spaces of migration, male migrants established and maintained relationships with other male migrants through demonstrating assertive, competent, and sexually independent masculinity with respect to one another (e.g., married men frequented brothels, bars, or other semiprivate spaces confident that their actions will not be revealed to their wives because of an unspoken agreement that men will cover for one another). This model may help to explain why Tajik external migrants living in shared apartments engaged in the exchange of sexual partners during drunken parties. It was not only because it was less expensive, but also because this practice strengthened the men's relationships with one another. In contrast, internal migrants did not share women during sex because they had better defined roles in the Tajik society and did not need to resort to affirming their bonds to other men through such sexual behaviors.

In addition to implications for theory building, our findings have practical implications for HIV prevention programs focused on labor migrants. First, it is worth noting that in response to the theory of powerlessness developed by Campbell in South African mines, the authors advocated for developing HIV prevention programs that focused on peer education, in part because of their ability to address group processes that may underlie risky sexual behaviors (Campbell, 1997). However, a 2-year study of a comprehensive community intervention in the mining community of Carletonville, South Africa—which included peer education, condom distribution, syndromic management of STIs, and presumptive STI treatment for sex workers—found little evidence for significant behavior change, and the prevalence of curable STI increased (Williams et al., 2003). The authors attributed the lack of effectiveness to (a) inadequate resources, (b) inconsistent messaging about the role of HIV in AIDS, (c) conflicts between stakeholders, and (d) a lack in urgency (Williams et al., 2003). It could also be that as a prevention model, the conceptualization of powerlessness that underlay the intervention was in some way inadequate as a basis for HIV prevention.

With respect to Tajik labor migrants, attempting to diminish their individual sense of powerlessness, which did not appear to determine their HIV risk-related behaviors, may not be the most productive approach. Rather, HIV preventive interventions should build awareness, knowledge, skills, relationships, supports, and confidence that would assist men in adjusting to the harsh living and working conditions of migrant life. In the Tajik external migrant context, this should include discouraging behaviors that increase risk such as such as unprotected sex with multiple partners (including sex workers) and heavy alcohol use, because they are not consistent with the man's role as a protector and provider of his family. Interventions should also consider encouraging more positive aspects of men's social identity (e.g., protectors of family and community) and social networks (e.g., sharing health information) that may be found amongst male labor migrants. Given migrants’ perceptions of the inaccessibility of condoms, HIV/AIDS prevention efforts should also focus on increasing condom availability, perhaps by distributing educational materials and condoms at points of contact between external migrants and sex workers in Moscow. Last, structural interventions are also called for, such as addressing migrants’ legal status and access to health care services in Russia.

There are several limitations of our study. First, this study used a purposive sampling approach and thus the sample was not representative of all Tajik labor migrants, internal or external. Second, recruitment was conducted from only one source each for internal and external migrants, which may have limited the range of subject available for possible recruitment. With respect to these concerns, future studies with labor migrants should consider using larger representative samples. Third, it is possible that due to concerns about confidentiality, some subjects may have tailored their statements about social norms to fit with what they believed we wanted to hear. Regarding this last limitation, we took many steps to encourage frank disclosure by assuring participants of confidentiality, exploring each topic at length and in-depth, and maintaining a nonjudgmental interview approach. Fourth, another possible limitation is that our analysis could have been biased toward some interpretations and away from others. To minimize this possibility, we explicitly considered alternative and discrepant explanations and sought feedback from key informants on study findings. Fifth, data collection was limited to one point in time and to one location on the trajectory of migrants’ lives. Future studies should consider longitudinal and multisite (including sending and receiving countries) methods so as to better examine psychosocial, cultural, and familial processes over time and in the multiple real-life contexts where HIV/AIDS risk and protection occur.

One final limitation was that our study was unable to elicit data regarding same-sex experiences among male migrants. We suspect that this was in part due to the highly stigmatized and hidden nature of such behaviors in Tajikistan. Low rates of same-sex behaviors have been reported from other HIV studies in Central Asia and Russia (Amirkhanian et al., 2010; Hamers & Downs, 2003); however, these figures may also underestimate the true prevalence as a result of the extreme social vulnerability and stigma that men who have sex with men experience in many Muslim countries. In light of these limitations in the field, a future study is needed to address awareness of men who have sex with men and HIV risk in a culturally sensitive way among countries in Central Asia with a Muslim majority population.

Further research is needed to design and evaluate preventive interventions to change the risk behavior of male labor migrants. One implication of our findings is that HIV preventive interventions with migrants should pay attention to the different patterns of migration (e.g., external vs. internal) as these introduce different contextual factors that must be considered for the design and implementation of interventions. Another implication of our study is that exposure to traumatic events should be investigated as a possible factor in impacting HIV risk and protective behaviors (Simoni & Ng, 2002). Our findings also suggest that successful HIV prevention programs with this population may also require interventions for the wives and other sexual contacts of migrants in Tajikistan and Moscow.