Abstract
While human mobility has been implicated in fueling the HIV epidemic in South Africa, the link between migration and HIV has not been systematically reviewed and quantified. We conducted a systematic review of the role of migration in HIV risk acquisition and sexual behaviour based on 29 studies published between 2000 and 2017. Furthermore, we performed a meta-analysis of the association between migration and HIV risk acquisition in four of the studies that used HIV incidence as an outcome measure. The systematic review results show that HIV acquisition and risky sexual behavior were more prevalent among both male and female migrants compared to their non-migrant counterparts. The meta-analysis results demonstrate that migration was significantly associated with increased HIV acquisition risk (aOR = 1.69, 95% CI 1.33–2.14; I2 = 35.0%). There is an urgent need for effective combination HIV prevention strategies (comprising biomedical, behavioral and structural interventions) that target migrant populations.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
Migrants are among the latest sub-groups to be classified as an HIV ‘key population’, being acknowledged as epidemiologically carrying increased risk, vulnerability and burden of infection due to a combination of biological, socio-economic and structural factors [1]. While HIV prevalence is 18.9% among the general population in South Africa, its concentration is characteristically higher in specific locations and sub-populations that are reported to practice high-risk behaviours [2].Some rural communities, such as in KwaZulu-Natal (KZN) Province, are important socio-geographic settings for HIV outcomes, where adult migration rates are as high as 26% and HIV prevalence ~ 40% [3]. The exceptionally high prevalence of internal migration in South Africa (i.e. up to 42%) is rooted in apartheid era policies that were aimed at ensuring the supply of rural African male laborers in urban and industrial centers, such as Johannesburg [4, 5].
Existing literature on the link between migration and HIV discuss the importance of the social and political-economic context of risk, with migration and mobility playing a central role in the rapid spread of the epidemic in South Africa [6]. Historically, migration was a male domain that entailed the positive selection of prime-aged men (i.e. unmarried, or if married, often migrated without their wives and families), who left their rural communities for work [7,8,9]. Under these arrangements, migration fueled the epidemic [10], and facilitated temporal disengagement from the home region’s sexual networks, thereby creating a demand for new sexual networks in the urban destination, where there is greater occurrence of risky sexual behavior that has been linked to HIV transmission [11,12,13]. In a Gauteng Province study, migrant men were more likely to engage in risky sexual behaviours, such as having unprotected sex (83%) and casual sex partners (93%), than non-migrant men [14]. More recently after 2000, it has been demonstrated that a large increase in internal female migration [15, 16], often associated with more frequent trips over shorter distances than the male migration [9, 10], may contribute to the persistent HIV epidemic. Women tend to migrate for purposes of employment and often access seasonal and low-paying informal jobs (e.g. domestic work, informal selling and trade) [17], being thus at risk of turning to transactional sex, either as a primary/secondary livelihood strategy [18, 19].
Despite the evidence indicated above, little is known about the pathways linking migration and HIV transmission in sub-Saharan Africa [20]. The link between labour migration and HIV is illustrated by Weine and Kashuba [21], who showed that, across diverse global locations (i.e. Africa, the Americas, Europe, South East Asia and Western Pacific), increased HIV risk was associated with sexual practice determinants, such as low condom use, multiple partnering and lack of HIV knowledge. However, they have not reported on female migrants engaging in commercial sex work (i.e. with a large number of partners) and other mobile populations groups, such as refugees and/or asylum seekers. Moreover, their review featured results from only seven of the 97 South African studies that address HIV risk among labour migrants despite evidence that the country contains more than 19% of the world’s HIV-positive population [1]. This potentially underrepresented data from the largest epidemic in the world and region with higher than average prevalence of migration.
This paper therefore summarized the South African literature since there is no evidence of systematic review reports on migration and HIV risk. It attempts to address this issue through a collation of the demographic and behavioral health-science literature, and examines the relationship between measures of migration, HIV risk behaviours and infection. Given that HIV prevention strategies are tailored to maximize condom use, and that voluntary counselling and testing services for migrant populations are unavailable [22], the objective of this review is to more fully understand the complex relationship between mobility, risky sexual practices and HIV acquisition. This necessitated examining the various definitions of migration, and summarizing the key findings, predictors of HIV infection, risky sexual behavior and biological outcomes of HIV in the studies included. Our undertaking here may provide important evidence to support the development of effective prevention programs for this vulnerable population.
Methods
Study Eligibility
The review spans work published from 2000 to 2017, with the selection following the Population, Intervention, Comparative group, Outcomes, Study design (PICOS) criteria for systematic reviews [23], the inclusion criteria being empirical studies conducted in South Africa examining HIV and migration. Specifically, those reporting data on HIV risk factors, prevalent and incident HIV infection linked to migration.
Literature Searches
Systematic literature searches were performed on five academic databases, namely: PubMed Central, Sage Publications, Google Scholar, Web of Science and J-STOR, which identified over 4800 articles on South Africa. Title and abstract searches were conducted on relevant English studies between 2000 and 2017. This 17 year span marks an important period of HIV viral evolution [24] due to the improved ART coverage, and was associated with increased life expectancy, and a likely lowering of HIV risk perception in key populations [25].
Based on an initial scoping of the mobility and HIV-risk literature, the following query terms and phrases were included: HIV, AIDS, migrant, migrants, labour migrants, truck drivers, mobility and migration. Studies indexing the concept of mobility were identified by using the truncation ‘migrat*’ (exclusive to Web of Science and PubMed Central), returning results from multiple keyword variations such as; ‘migration’, ‘migrancy’, ‘migrate’ and ‘migrant’. Building on this, the subject heading ‘HIV risk behaviour’ was added to the search, including terms synonymous with sexual behaviour, with implications for HIV transmission “sexual risk behaviour”, “transmission risk”, “multiple sexual partnerships”, “AIDS risk behaviour” etc. We also explicitly examined incident infections by including common keywords, such as; “HIV infection”, “HIV incidence”, “new infections” “risk of HIV acquisition” etc. as detailed in Table 1. This demonstrated a composite filter process for our first round attempt at identifying the literature linking how aspects of migration intersect with HIV risk factors and outcomes.
Other study retrieval methods, while remaining secondary to the main searches above, were conducted to obtain additional literature, including hand-searches. This technique is a purposive page-by-page examination of relevant journal issues, specific studies and conference proceedings [23]. Moreover, the ‘cited by’ reference-searching feature compliant with Web of Science (i.e. linking journal articles to other studies in which they have been cited) was used.
Table 1 details the final search query terms pertaining to the independent/explanatory variable ‘migration’ and the outcome of interest/dependent variable ‘HIV risk’, which, when combined with Boolean operators AND/OR within each concept (labour migration and HIV risk), identified an initial range of relevant study records. The database searches began in January 2017, with electronic, hand searching (i.e. searching through content of journal articles to identify relevant studies) and examining the results generated from the initial search, and concluded on 30 August, 2018. Notably, the database search (PubMed Central) returned 4887 hits, while 11 studies were obtained through hand searching. A total of 4898 records were retrieved, and after multiple iterations and search query refinement, title and abstract screening was performed on 71 full-text records that were extracted to Mendeley (v1.16.2. reference manager), based on their relevance to the scope of the review. However, a further 42 articles were discarded after failing to meet the inclusion criteria, which resulted in 29 studies being included, as presented in Fig. 1.
Data Extraction
The data from the 29 empirical studies (published between 2000 and 2017) selected for inclusion, after successive rounds of filtering, were retrieved and entered onto a Microsoft Excel workbook. These studies comprised four longitudinal, 16 cross-sectional, five qualitative and five mixed-methods studies. Two independent reviewers, [i.e. a content and a method (systematic review) expert] conducted title-screening to verify which reports should be included in the final review. Queries regarding inclusion were resolved through consultation with the review team members. The Mixed Method Appraisal Tool (MMAT) was used to generate a score to measure the methodological quality of the studies and evidence reported to establish the risk of bias. Once consensus over the validity of study selection was reached, the following were explored: defining migration, identifying key conceptual frameworks and predictors of HIV infection, risky sexual behaviour and biological outcomes of HIV.
Meta-analysis
Data were analyzed using STATA version 15 (StataCorp, College Station, Texas, USA), with estimates of HIV acquisition risk and 95% confidence intervals (CI) of exposure-outcome pairs being presented on forest plots. These data were obtained from the studies that reported hazard ratio and 95% CI. I2 statistics were used to measure the percentage of observed total variation across the studies due to heterogeneity rather than by sampling error [26, 27]. We made a prior decision to use a random-effects model for combined effects (given our assumption about anticipated variation among the studies), and presumed that the studies were conducted independently. In terms of potential publication bias arising from small-study effects, we did not construct funnel plots or test them for asymmetry, given inadequate number of studies [28] (it is recommended to have a minimum of 10 studies). Due to the large variation in their sample sizes i.e. large population-based research versus small studies, we conducted a sensitivity analysis to examine the influence of each study on the overall meta-analysis summary estimate by omitting each study in turn.
Results
Defining Migration
Table 2 shows that while all of the studies offered definitions of migration, there were important differences among them. These included dimensions of mobility, including selectivity to migration (i.e. whether individuals had distinct characteristics suiting them to migrate), and migration stream (that migrants had a common origin and destination). Such studies used conventional demographic definitions of migration, such as ‘internal migration’, ‘temporary labour migration’ and ‘circular migration’, referring to migration within South Africa, where persons keep their contact with a rural homestead. Definitions further included immigration, in that migrant were referred to as: ‘foreign migrants’ or ‘external migrants’, ‘cross-border migrants’ and ‘transnational migration’ [29,30,31,32,33].
Nine of the 29 studies were conducted in demographic surveillance sites (DSS), either the Africa Centre Demographic Information System (ACDIS) or the Agincourt Demographic and Health Surveillance Site (ADHSS), respectively located in rural communities in KZN and Limpopo Provinces. They examine forms of mobility involving a change of residency, denoting migration events [34, 35]. Using residence history data, the DSS studies define in- and out-migration on the basis of residence change that consists of crossing the boundaries of the designated surveillance area [10], thusly classifying internal or external migrants. Conversely, other studies used the census definition that internal migrants cross geo-political boundaries, usually magisterial districts or provinces [14]. Migration for the purposes of labour or ‘occupational mobility’ featured prominently in the studies sampling mine workers [14, 36,37,38,39,40], truck drivers [41, 42] and migrant women who engage in sex-work in the destination and/or are mobile sex workers [29, 31, 32, 43].
Study Hypotheses and Key Messages
The studies reviewed diverse aspects of the migration-HIV link, which we summarized accordingly the various study objectives and key findings that indicated the mechanisms affecting this relationship (Table 2). Cohort prospective studies capturing individual sexual histories highlighted the impact of time-variant factors, such as income Adjusted Hazard Ratio (AHR) 0.93 (s.e. 0.0279) p = 0.022 wealthier compared to poor and sex AHR 1.83 (s.e. 0.3616) p = 0.002, on risk of HIV acquisition [44]. Other studies highlighted the socio-economic differences between the male migrants and non-migrants that were associated with risky sexual practices, such as having more than two lifetime sex partners [4, 39]. The main reasons why foreign migrant women who engage in sex work were implicated in HIV infection and risky behaviour relates to their very high exposure to structural and sexual violence, including poor access to local medical care and support service inequalities [31, 33]. In three studies, non-migrating spouses of male labour migrants were included, with the hypothesis of a bi-directional stream of HIV transmission being emphasized [42, 45, 46].
Risky Sexual Behaviour
An array of self-reported sexual behaviour practices in both men and women in migration situations have implications for HIV infection (Table 3). Disaggregated by gender, migrant men, mainly those residing in single-sex settings, were more likely to engage in risky sexual behaviours, such as unprotected and anal sex, multiple partnering and visiting commercial sex workers. One study found that migrant men were 1.5 times more likely to have multiple sex partners compared to non-migrant men (95% CI 1.03–2.20) [47]. Other factors (i.e. attitudes) for increased HIV vulnerability included limited knowledge of HIV transmission and low self-assessment for HIV risk prevention [30].
Reported Condom Use
Migrants were less likely to report condom use compared to non-migrants in the included studies. This lower usage was more prominent when results were disaggregated by migrant origin, and whether before or after a behaviour intervention. Non-migrant women compared to migrant women were more likely to report condom use (OR 1.88, 95% CI 1.02–3.45) [39]. One study shows that male migrants compared to non-migrants were less likely to report condom use with regular sexual partner (10.9% vs. 23.7%, p = 0.04). Foreign migrant women who engage in sex work in South Africa were less likely to use condoms compared to local migrant women for the same activities (90.6% vs. 94.5%, p = 0.08) (see Table 4). Similarly, among South African women, migrants in sex work compared to non-migrants were less likely to report condom use with clients (94.5% vs. 89.1%, p = 0.02) in the same study [43]. In 2000, a study comparing the prevalence of risky sexual practices among mineworkers between a pre- and post-HIV prevention campaign noted that condom use with spouse was 18% before and 26% after, and with other partners, 60% before and 67% after [36].
Multiple Partnering
Many studies reported higher rates of multiple partnering among migrants compared to non-migrants. Older, married, migrant men, sexually active young men under the age of 35 years, younger women, and resident women married to migrant men reported having additional sexual partners other than their spouse or main sexual partner [10, 33, 37, 45, 46, 48, 49]. On the other hand, being a non-migrant, shorter distance migration and returning home regularly was not associated with multiple partnering. Migrant men returning home frequently had their odds for having multiple sexual partners reduced by 53% compared to those who did not (OR 0.47, p = 0.024). Migrants working close to home were less likely to have multiple partnerships compared to those working far away (OR 0.66, p = 0.071) [46]. One study showed that female migrants had additional partners ranging between two and 11 in the past 6 months [31]. In a mining township in Carletonville, migrant women were four times more likely to have two or more partners compared to non-migrant women (OR 4.18, 95% CI 2.25–7.76, p < 0.001). Additional partners, other than the regular ones, were more likely to be reported by migrant women compared to non-migrants (OR 2.38, 95% CI 1.60–3.54) [39]. A study enrolling migrant men from KZN found that those who migrate had a mean lifetime number of sexual partners of 18.2 compared to 13.4, p = 0.0001 among those who do not migrate [38]. Employed migrant men compared to those unemployed were more likely to have multiple sexual partners (50% Vs 38%). Gender specific comparisons by migration status and prevalence of multiple sexual partnerships showed increased odds for men (AOR 1.51, 95% CI 1.03–2.20) compared to women (AOR 1.05, 95% CI 0.53–2.07) [47].
Visiting Sex Workers/Engaging in Sex Work
Most studies in Table 4 reported that male migrants compared to non-migrants had a heightened likelihood of being clients of sex workers. In addition, female migration was frequently identified as a predictor and/or outcome of sex-work, this being an important risk factor for mobile women [33, 50]. In two studies conducted over a decade apart, i.e. 2002 and 2014, a high proportion of long-distance truck drivers reported frequently using the services of sex workers (37% and 30%) [41, 42]. One qualitative study reported that mineworkers often visited commercial sex workers in single-sex hostels in Mpumalanga and Northwest Provinces mining towns to cope with long detachment from spouses or main partners. Recent migrants were significantly more likely to be sex workers in South Africa shortly before the 2010 Soccer World Cup compared to recent migrants after the event (2.4% vs. 0.4%, p = 0.011) [43]. Foreign female migrants were 2.3 times more likely to be sex workers in South Africa compared to non-mobile local women (95% CI 1.5–3.7). Internal migrants were however 1.5 times more likely to be sex workers compared to non-migrating women [42]. Other qualitative studies on women in migration situations highlighted that unemployed migrant young women were at risk of engaging in transactional and commercial sex-work as a survival strategy [48]. Compared to non-migrants, the majority of migrants were more likely to have had received cash for casual sex in the past year (53% vs. 30%) [14].
Casual Sex
Several studies in Table 4 reported increased casual sex in both migrant men and women in comparison to those who do not migrate. Casual contact was defined as any sex with a partner not considered to be a steady partner, such as a casual associate, friend etc. Migrant mineworkers in the Carletonville and Khutsong male hostels were more likely to report having casual encounters in the past compared to non-migrant men (53% vs. 32%). In the same study, migrant women compared to non-migrant women were more likely to reporting having casual sex in the past 12 months (37% vs. 17%) [14]. In another study, significant differences in the proportions and odds for casual sex by gender and migration status were observed, with migrant men compared to non-migrant men being more likely to report having at least one casual partner in the past year (20% vs. 6%, p = 0.02). Similarly, migrant women compared to non-migrant women were 1.6 times more likely to have had casual sex in the past year (95% CI 1.03–2.53) p < 0.05 [35].
Men Who Have Sex with Men
Three studies in Table 4 included men who have sex with men (MSM). While data in these studies do not have comparative endpoints and measures quantifying the extent of the risk of being a labor migrant, they descriptively highlight the prevalence of unprotected sex with other men in masculine same-sex workplaces and hostels (e.g. mines and factories), whether consensual or ‘forced’. A high number of long distance truck drivers in KZN (42%, n = 92 of N = 220) were reported to have sex with other men [41]. Based on recollections of former migrant miners, one qualitative study found that men working in mine in Impalahoek, Mpumalanga Province, regularly had unprotected sex with younger colleagues in exchange for money, protection and preferential treatment at work [37].
HIV Infection Rates
This section presents the odds ratio and hazard rate effect sizes of HIV prevalence and risk acquisition by migration status. We further present the meta-analysis results based on reported adjusted hazard ratios in Figs. 2 and 3.
HIV Prevalence
The data in Table 5 are prevalent HIV estimates for both migrants and non-migrants. In a study in rural KZN, female migrants had more than double the odds of infection (AOR = 2.55; 95% CI 2.07–3.13) compared to male non-migrants. Moreover, female migrants also had a 48% higher odds of being HIV-positive than female non-migrants (AOR 1.48, 95% CI 1.23–1.77) [51]. An earlier study found that migrant women were more likely to be HIV positive than their non-migrant counterparts (51% vs. 39%; p = 0.002) [14]. Another study showed that migrant women were 1.5 times more likely to be infected by HIV than non-migrant women (OR 1.52, 95% CI 1.01–2.28, p = 0.046) [39], while 25.9% of migrant men compared 12.7% of non-migrant men were infected with HIV (OR 2.4, 95% CI 1.1–5.3; p = 0.029) [38]. Migrant men compared to partners of migrant men were more likely to be infected by STI’s, including HIV (OR 1.54, 95% CI 1.00–2.37; p = 0.049) [39]. HIV infection among non-residents compared to residents was AOR 1.8 (95% CI 1.3–2.4) for men and AOR 1.5 (95% CI 1.2–2.0) for women in a rural area in KZN [52]. More recently, in the same DSS, non-residents compared to residents in the study area had higher odds of infection: AOR 1.19 (95% CI 1.07–1.33) for men and AOR 1.8 (1.10–1.26) for women [53]. Other results by duration of mobility associated with higher HIV prevalence i.e. spending more time travelling per-month, increased the odds of HIV infection by 1.5 times compared to spending less time travelling per-month (95% CI 0.9–2.3) [42]. Truck drivers were four times more likely to be HIV positive compared to the men of the same age in the general population AOR (4.26 (95% CI 24–28) [42].
Hazard of Acquiring HIV
Four of the studies with multivariate statistics for incident infections are presented in Table 6. For instance, male migrants compared to non-migrants were more likely to acquire HIV (AHR 2.54, 95% CI 1.67–3.85; p = 0.0001) and female migrants compared to non-migrants were more likely to acquire HIV (AHR 1.57, 95% CI 1.14–2.17; p = 0.006) [3]. In KZN Province, non-migration reduced the risk of HIV acquisition by over 50% (AOR 0.4988, 95% CI 1.23–3.24; p = 0.005) [44].
Meta-analysis Results
To examine the overall estimate of the effect of migration on increasing the risk of acquiring HIV, we statistically summarized the six estimates from four studies, as shown in Table 6. According to our meta-analysis (Fig. 3), exposure to migration was associated with a 69% increase in the HIV acquisition risk (95% CI 1.33–2.14). We found low to moderate heterogeneity in the combined effect size, (I2 = 35.0%, p = 0.17). Our sensitivity analysis, which was conducted by omitting each study in turn, did not alter the significant relationship between migration and risk of HIV acquisition. Our study also suggests that the removal of small studies, such as Hargreaves et al. [47], did not alter the significance of our overall findings.
Discussion
The risk of acquiring HIV increased by 69% (95% CI 1.33–2.14) in migrants compared to non-migrants, according to our meta-analysis results of four studies. These studies were conducted over eight years (2007–2015), a period covering two HIV treatment eras; one in which it barely existed, with an associated increase in prevalence and incidence, and the latter in which it was widely available and HIV incidence decreased in the general population. Despite universal ART coverage being achieved after 2010, the meta-analysis result suggests that HIV incidence due to migration remained high, as demonstrated in a cohort study in Uganda [20]. HIV prevalence was high in migrants compared to non-migrants, the largest difference being across genders, with female migrants having more than double the odds of infection (aOR = 2.55; 95% CI 2.07–3.13) compared to male non-migrants [17]. Another key finding was that sexual practices (i.e. multiple partnering, limited condom use, client to sex workers/engaging in sex work, casual sex and transactional sex) are important mechanisms shaping the risk of HIV infection among those who migrate, confirming an earlier observation [21]. We highlight the intersectionality of bio-demographic and structural factors in modifying the continuum of HIV risk in migrants at an ecological level in South Africa [48]. For instance, we found that unemployed, migrant young women had a higher risk of having additional sex partners and being involved in sex work or transactional sex compared to non-migrants.
Multiple partnering was frequently identified as high among migrants compared to non-migrants, with having additional sexual partners when separated from the main partner during a migration episode being significantly increased across gender and age group. In two studies, female migrants (34.4 years, mean) were 4.2 times more likely to have two or more lifetime sexual partners compared to non-migrants [39]. Male migrants (37.4 years, mean) had an average of 18.2 compared to 13.4 lifetime sexual partners in non-migrants compared to non-migrants [38], highlighting large differences by migration status. This age and gender clustering challenges the traditional male-infector model, which presumes that only prime-aged migrant men are vulnerable to HIV risk through their interaction with external partners and thereby infect their wives or regular partners. As suggested in Ferrand [53], easy generalizations, such as viewing separate and static characteristics of individuals as independent risk factors of HIV, are no longer plausible, instead, the interaction of social and biological influences (age, gender, nature of sexual relationships) conjointly increases the risk of infection. This study showed that both migrant men and women reported having a greater number of additional sexual partners outside their community of origin, but gender differences in the experience of risk emerged. Typically, men with multiple partners had better paying jobs, and were therefore more likely to afford additional partners and further expand their sexual networks while maintaining multi-local partnerships in comparison to those who do not migrate. In contrast, women, in poorly paying jobs, as is often the case with female migrant labour [17], send most of their earnings home i.e. two-thirds higher than men’s [54], and thus may rely on multiple sexual partners or engage in transactional sex as an additional source of income, as noted in some qualitative studies [18, 19]. Here we demonstrated that the finding for women did not necessarily conflict with that for men. This supports our initial hypothesis that employment is not protective for women, as livelihoods associated with migration often include commercial sex work and transactional sex.
It follows that further studies exploring the migrant employment continuum (i.e. looking for employment to already working) are needed to confirm the findings about the intersections of various correlates of migration and sexual behaviour outcomes. A synthesis of the sexual behaviour data reveals that women who migrate were associated with increased prevalence of HIV risk behaviours. Young (or older), employed (or unemployed) and migrating men and women reported having more multiple sex partners and transactional sex than non-migrants. Moreover, that some women who engaged in risky sexual contact were employed [48] reflects that economic insecurity (as conventionally interpreted), may not completely explain why they are disproportionately at greater risk of HIV infection. The literature likely explaining this highlights that material deprivation and consumerism instigates this sex-gift exchange, and influences young women’s high sexual risk taking [55, 56].
In three of the studies, adult men who did not migrate with their spouses/partners, returned home less frequently or migrated to the furthest destinations were associated with high additional sexual partnering [10, 42, 46]. In light of this result, it is possible to infer that increased risky sexual behaviour in married men is driven by extraordinarily long episodes of absence from home, i.e. mineworkers and long-distance truck drivers were frequently reported to have casual sexual partners.
There are two important observations emerging from this finding. Firstly, it indirectly affirms the evidence from other studies in rural South Africa showing that migrant men who work close to their home communities were less likely to report multiple sexual partners than those travelling to destinations further afield (returning once or twice a year), suggesting that short distance migration is a protective factor to sexual risk [17]. Secondly it gives credence to our examination of migration definitions in this study and forthcoming studies that may use mobile application technology to collect real-time data (i.e. on distance travelled and pattern of migration) to test more thoroughly the proposition that detailed measurement of migration (i.e. length of time away and destination correlates) is a key HIV independent risk factor.
However, these results may have limited explanatory value, due to a number of methodological challenges. For instance, many of the existing studies examining multiple partnering obtained no data on the characteristics of the sexual partnerships, such as the length of overlaps between and the type of sexual partners, information that is particularly important in determining transmission during concurrent partnerships. If characteristics of the partner or partnership are not collected, it is difficult to determine their epidemiological interpretation, given that some partnerships may be once-off sexual encounters. For instance, in one study reporting multiple sexual partnering in women, up to 11 sexual partners were reported, although it may not be known with certainty whether additional partners were encountered during on-going specific partnerships. It is therefore important that definitions of multiple partnering are standardized, particularly in quantitative studies, to avoid misreporting, given its implications for HIV risk. Similarly, it is difficult to accurately interpret results from studies measuring migration differently. Systematic attention to definitions would simplify cross-study comparison of multiple partnering results, and may render effective the need for clear messages aimed at reducing multiple partnerships, irrespective of whether those overlap in time [57].
A few studies (four) measured HIV incidence and 13 reported on prevalence. Unfortunately, with the shortage of data on new HIV diagnoses as opposed to prevalence, it is impossible to ascertain any causal relationship, as migrants may have been infected in the past when they had different sexual behaviours. This finding implies the need for more cohort studies on migrant HIV risk to rigorously estimate the effect of mobility on new infections. Accordingly, this undertaking is consistent with the UNAIDS current funding model and package of prevention strategies, which involves identifying main modes of HIV transmission, key affected populations, and core epidemiological trends to ensure a greater impact on reducing new infections [1].
Conclusion
Mobility is highly associated with increased prevalence of HIV risk behaviours, and confers up to 69% increase in the risk of HIV acquisition. Studies included in this review documented increased multiple sexual partnering, unprotected sexual intercourse, visiting sex workers and engaging in sex work in migrants compared to non-migrants. Escalation of this sexual behaviour and risk of HIV acquisition among migrants in comparison to non-migrants calls for increased reliance on the targeted and best-combination HIV prevention strategies. Our review found only four studies on migration and HIV risk acquisition, emphasizing the need for incidence data to establish if the timing of new infections corresponds with the migrant sexual behaviours above, and can predict future risk patterns. The implications of this study include monitoring and tracking key trends of the epidemic in migrants to evaluate country level success towards the UNAIDS’s focus on optimizing the reduction of new HIV infections. Effective combination HIV prevention strategies that target migrant populations are urgently required.
References
UNAIDS. The Gap Report. Geneva, Switzerland: Joint United Nations Programme on HIV/AIDS. http://files.unaids.org/en/media/unaids/contentassets/documents/unaidspublication/2014/UNAIDS_Gap_report_en.pdf. Accessed 4 Dec 2018.
Shisana O, Rehle T, Simbayi LC, Zuma K, Jooste S, Zungu N, Labadarios D, Onoya D. South African national HIV prevalence, incidence and behaviour survey, 2012. https://books.google.co.za/books?hl=en&lr=&id=LlZRZjX-D_gC&oi=fnd&pg=PR5&dq=migration+and+hiv+incidence&ots=SulWs79uK1&sig=RhCwlz2yKue16da7nyzk9Fe36j0#v=onepage&q=migrationandhivincidence&f=false. Accessed 04 Dec 2018.
Dobra A, Bärnighausen T, Vandormael A, Tanser F. Space-time migration patterns and risk of HIV acquisition in rural South Africa. AIDS. 2017;31(1):137–45.
Collinson MA, Tollman SM, Kahn K. Migration, settlement change and health in post-apartheid South Africa: Triangulating health and demographic surveillance with national census data. Scand J Public Heal. 2007;35(Suppl 69):77–84.
Hargrove J. Migration, mines and mores: the HIV epidemic in southern Africa. SA J Sci. 2008;104(1–2):53–61.
Khan MR, Patnaik P, Brown L, Nagot N, Salouka S, Weir SS. Mobility and HIV-Related Sexual Behavior in Burkina Faso. AIDS Behav. 2008;12(2):202–12.
Kwena ZA, Camlin CS, Shisanya CA, Mwanzo I, Bukusi EA. Short-term mobility and the risk of hiv infection among married couples in the fishing communities along Lake Victoria, Kenya. PLoS ONE. 2013;8(1):54523.
Lagarde E, van der Schim M, Enel C, Holmgren B, Dray-Spira R, Pison G, et al. Mobility and the spread of human immunodeficiency virus into rural areas of West Africa. Int J Epidemiol. 2003;32(5):744–52.
Deane K, Parkhurst J. Linking migration, mobility and HIV. Trop Med. 2010;15(12):1458–63.
Collinson MA. Striving against adversity: the dynamics of migration, health and poverty in rural South Africa. Glob Health Action. 2010;3(1):5080.
Mah TL, Halperin DT. Concurrent sexual partnerships and the HIV epidemics in Africa: evidence to move forward. AIDS Behav. 2010;14(1):11–6.
Saggurti N, Nair S, Malviya A, Decker MR, Silverman JG, Raj A. Male migration/mobility and HIV among married couples: cross-sectional analysis of nationally representative data from India. AIDS Behav. 2012;16(6):1649–58.
Ramjee G, Gouws E. Prevalence of HIV among truck drivers visiting sex workers in KwaZulu-Natal, South Africa. STDs. 2002;29(1):44–9.
Gilgen D, Williams BG, MacPhail C, Van Dam CJ, Campbell C, Ballard RC, Taljaard D. The natural history of HIV/AIDS in a major goldmining centre in South Africa: results of a biomedical and social survey. S Afr J Sci. 2001;97(9–10):387–92.
Posel D. Households and labour migration in post-Apartheid South Africa. J Stud Econ Econom. 2010;34(3):129–41.
Camlin CS, Snow RC, Hosegood V. Gendered patterns of migration in rural South Africa. Popul Space Place. 2014;20(6):528–51.
Camlin CS, Hosegood V, Newell ML, McGrath N, Bärnighausen T, Myer L. Gender, migration and HIV in rural KwaZulu-Natal, South Africa. PLoS ONE. 2010;5(7):e11539.
Giorgio M, Townsend L, Zembe Y, Guttmacher S, Kapadia F, Cheyip M, Mathews C. Social support, sexual violence, and transactional sex among female transnational migrants to South Africa. Am J Public Health. 2016;106(6):1123–9.
Bond V, Ngwenya F, Thomas A, Simuyaba M, Hoddinott G, Fidler S, et al. Spinning plates: livelihood mobility, household responsibility and anti-retroviral treatment in an urban Zambian community during the HPTN 071 (PopART) study. J Int AIDS Soc. 2018;21:e25117.
Olawore O, Tobian AAR, Kagaayi J, Bazaale JM, Nantume B, Kigozi G, et al. Migration and risk of HIV acquisition in Rakai, Uganda: a population-based cohort study. Lancet HIV. 2018;5:e181–9.
Weine SM, Kashuba AB. Labor migration and HIV risk: a systematic review of the literature. AIDS Behav. 2012;16(6):1605–21.
Iwuji C, Larmarange J, Ba M, Dreyer Ndip J, Herbst K, Scd M, et al. Universal test and treat and the HIV epidemic in rural South Africa: a phase 4, open-label, community cluster randomised trial. Lancet HIV. 2017;5:e116–25.
Moher D, Shamseer L, Clarke M, Ghersi D, Liberati A, Petticrew M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev. 2015;4(1):1.
Tanser F, Baernighausen T, Graspa E, Zaidi J, Newell M-L. High Coverage of ART Associated with decline in risk of HIV acquisition in rural KwaZulu-Natal. Science. 2013;339(6122):966–72.
Ramjee G, Naidoo S. The road ahead: working towards effective clinical translation of biomedical HIV prevention strategies. Future Virol. 2015;10(3):271–82.
Higgins J, Thompson S. Quantifying heterogeneity in a meta-analysis. Stat Med. 2002;21(11):1539–58.
Julian T, Higgins AP, Thompson SG, Deeks JJ. Measuring inconsistency in meta-analyses. BMJ. 2003;327(7414):557–60.
Anzures-Cabrera J, Higgins JPT. Graphical displays for meta-analysis: an overview with suggestions for practice. Res Synth Methods. 2010;1(1):66–80.
Richter M, Chersich MF, Vearey J, Sartorius B, Temmerman M, Luchters S. migration status, work conditions and health utilization of female sex workers in three South African cities. J Immigr Minor Health. 2012;16(1):7–17.
Munyewende P, Rispel LC, Harris B, Chersich M. Exploring perceptions of HIV risk and health service access among Zimbabwean migrant women in Johannesburg: a gap in health policy in South Africa? J Public Health Policy. 2011;32(1):S152–61.
Townsend L, Giorgio M, Zembe Y, Cheyip M. HIV prevalence and risk behaviours among foreign migrant women residing in Cape Town, South Africa. AIDS Behav. 2014;18(10):2020–9.
Walker R, Oliveira E. Contested spaces: exploring the intersections of migration, sex work and trafficking in South Africa. Grad J Soc Sci. 2015;11(2):129–53.
Semá Baltazar C, Horth R, Inguane C, Sathane I, César F, Ricardo H, et al. HIV prevalence and risk behaviors among Mozambicans working in South African Mines. AIDS Behav. 2015;19(1):59–67.
Vandormael A, Newell M-L, Bärnighausen T, Tanser F. Use of antiretroviral therapy in households and risk of HIV acquisition in rural KwaZulu-Natal, South Africa, 2004–2012: a prospective cohort study. Lancet Glob Heal. 2014;2(4):e209–15.
Collinson M, Wolff B, Tollman S, Kahn K. Trends in internal labour migration from the rural Limpopo Province, male risk behaviour, and implications for spread of HIV/AIDS in rural South Africa Council Programme on AIDS in Uganda. J Ethn Migr Stud. 2006;32(4):633–48.
Meekers D. Going underground and going after women: trends in sexual risk behaviour among gold miners in South Africa. Int J STD AIDS. 2000;11(1):21–6.
Niehaus I. Renegotiating masculinity in the Lowveld: Narratives of male-male sex in compounds, prisons and at home. 2002. https://www.researchgate.net/profile/Melissa_Steyn/publication/305807288_The_Prize_and_the_Price_Shaping_sexualities_in_South_Africa/links/57a22ab508aeef8f311ec874/The-Prize-and-the-Price-Shaping-sexualities-in-South-Africa.pdf#page=93. Accessed 4 Dec 2018.
Lurie MN, Williams BG, Zuma K, Mkaya-Mwambury D, Abdool Karim SS. The impact of migration on HIV-1 transmission in South Africa: a study of migrants and non-migrant men and their partners. Sex Transm Dis. 2003;30(2):149–56.
Zuma K, Gouws E, Williams B, Lurie M. Risk factors for HIV infection among women in Carletonville, South Africa: migration, demography and sexually transmitted diseases. Int J STD AIDS. 2003;14(12):814–7.
Zuma K, Lurie MN, Williams BG, Mkaya-Mwamburi D, Garnett GP, Sturm AW. Risk factors of sexually transmitted infections among migrant and non-migrant sexual partnerships from rural South Africa. Epidemiol Infect. 2005;133(3):421–8.
Ramjee G, Gouws E. Prevalence of HIV among truck drivers visiting sex workers in KwaZulu-Natal. Sex Transm Dis. 2002;29(1):44–9.
Delany-Moretlwe R, Bello B, Kinross P, Oliff M, Chersich M, Kleinschmidt I, et al. HIV prevalence and risk in long-distance truck drivers in South Africa: a national cross-sectional survey. Int J STD AIDS. 2014;25(6):428–38.
Richter M, Luchters S, Ndlovu D, Temmerman M, Chersich MF. Female sex work and international sport events—no major changes in demand or supply of paid sex during the 2010 Soccer World Cup: a cross-sectional study. BMC Public Health. 2012;12(1):763.
Bärnighausen T, Hosegood V, Timaeus IM, Newell M-L. The socioeconomic determinants of HIV incidence: evidence from a longitudinal, population-based study in rural South Africa. AIDS. 2007;21(7):S29.
Dladla AN, Hiner CA, Qwana E, Lurie M. Speaking to rural women: the sexual partnerships of rural South African women whose partners are migrants. Soc Transit. 2001;32(1):79–82.
Kahn K, Collinson M, Tollman S, Wolff B. Health consequences of migration: evidence from South Africa’s rural northeast (Agincourt). African Migr; 2003. pp. 4–7.
Hargreaves JR, Bonell CP, Morison LA, Kim JC, Phetla G, Porter JD, et al. Explaining continued high HIV prevalence in South Africa: socioeconomic factors, HIV incidence and sexual behaviour change among a rural cohort, 2001–2004. AIDS. 2007;21:S39–48.
Hunter M. The changing political economy of sex in South Africa: the significance of unemployment and inequalities to the scale of the AIDS pandemic. Soc Sci Med. 2007;64(3):689–700.
Giorgio M, Townsend L, Zembe Y, Cheyip M, Guttmacher S, Carter R, et al. HIV prevalence and risk factors among male foreign migrants in Cape Town, South Africa. AIDS Behav. 2017;21(3):949–61.
Slabbert M, Venter F, Gay C, Roelofsen C, Lalla-Edward S, Rees H. Sexual and reproductive health outcomes among female sex workers in Johannesburg and Pretoria, South Africa: recommendations for public health programmes. BMC Public Health. 2017;17(3):442.
Welz T, Hosegood V, Jaffar S, Bätzing-Feigenbaum J, Herbst K, Newell ML. Continued very high prevalence of HIV infection in rural KwaZulu-Natal, South Africa: a population-based longitudinal study. AIDS. 2007;21(11):1467–72.
McGrath N, Eaton JW, Newell ML, Hosegood V. Migration, sexual behaviour, and HIV risk: a general population cohort in rural South Africa. Lancet HIV. 2015;2(6):e252–9.
Ferrand R. Living with HIV and dying with AIDS. Lancet Infect Dis. 2014;14(1):804.
Posel D. Social dynamics how do households work? Migration, the household and remittance behaviour in South Africa. Soc Dyn. 2008;27(1):165–89.
Kamndaya M, Thomas L, Vearey J, Sartorius B, Kazembe L. Material deprivation affects high sexual risk behavior among young people in urban slums, South Africa. J Urban Heal. 2014;91(3):581–91.
Jewkes R, Dunkle K, Nduna M, Shai NJ. Transactional sex and HIV incidence in a cohort of young women in the stepping stones trial. J AIDS Clin Res. 2012;3:5.
Tanser F, Bärnighausen T, Hund L, Garnett GP, McGrath N, Newell M-L. Effect of concurrent sexual partnerships on rate of new HIV infections in a high-prevalence, rural South African population: a cohort study. Lancet. 2011;378(9787):247–55.
Hunter M. Love in the Time of AIDS. Bloomington, IN: Indiana University Press; 2010. Retrieved from http://www.jstor.org/stable/j.ctt16gzmmw. Accessed 11 Dec 2018.
Acknowledgements
This research was funded through a Medical Research Council flagship grant from the Republic of South Africa (MRC-RFA-UFSP-01-2013/UKZN HIVEPI). The study design, findings, interpretation of data and conclusions in this paper are those of the authors, and do not necessarily represent the official positions of MRC South Africa or any other surrogate organization.
Funding
This study was funded through a Medical Research Council flagship grant from the Republic of South Africa (MRC-RFA-UFSP-01-2013/UKZN HIVEPI).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no conflict of interest.
Ethical Approval
The Biomedical Research Ethics Committee of the University of KwaZulu-Natal (BREC) Durban, South Africa, gave full ethics approval for this study.
Informed Consent
This article does not contain any studies with human participants or animals performed by any of the authors
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Dzomba, A., Tomita, A., Govender, K. et al. Effects of Migration on Risky Sexual Behavior and HIV Acquisition in South Africa: A Systematic Review and Meta-analysis, 2000–2017. AIDS Behav 23, 1396–1430 (2019). https://doi.org/10.1007/s10461-018-2367-z
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10461-018-2367-z