Migration and age differences between male and female partners fuelling the HIV epidemic in South Africa

This article is more than 8 years old. Click here for more recent articles on this topic

Analyses from two large household surveys in KwaZulu-Natal, South Africa shed new light on the dynamics of HIV transmission in the South African province that is hardest hit by HIV. Adolescent girls and young women typically acquire HIV from men several years older than themselves, while older men usually acquire HIV from women of their own age. Men and women who migrate just 50km away from home are more likely to become HIV positive than those who stay in their home community.

The two recently published studies each use data from large household surveys in two different areas of KwaZulu-Natal. Age-disparate relationships and migration have already been identified as factors driving the HIV epidemic in southern Africa, but these sophisticated analyses clarify and reinforce the evidence for these factors. 

Age gaps between sexual partners – the study

The first study, by Tulio de Oliveira, Ayesha Kharsany and colleagues, is published online ahead of print in the Lancet HIV. The data come from a cross-sectional, random sample of households in one rural and one urban subdistrict of uMgungundlovu district. In each of the 77% of households that took part, one randomly selected household member between the ages of 15 and 49 completed questionnaires and provided a blood sample for HIV testing.

The researchers used phylogenetic analysis to identify transmission clusters – in other words, groups of individuals who had genetically similar HIV and who probably acquired HIV from each other. Phylogenetic analysis estimates how closely related the samples of HIV taken from different people are, in comparison with other samples of HIV.


phylogenetic analysis

The comparison of the genetic sequence of the virus in different individuals in order to determine the likelihood that two or more samples are related. This involves creating a hypothetical diagram (known as a phylogenetic tree) that estimates how closely related the samples of HIV taken from different individuals are. Phylogenetic analysis is not a reliable way to prove that one individual has infected another, but may identify transmission clusters, which can be useful for public health interventions.


Studies aim to give information that will be applicable to a large group of people (e.g. adults with diagnosed HIV in the UK). Because it is impractical to conduct a study with such a large group, only a sub-group (a sample) takes part in a study. This isn’t a problem as long as the characteristics of the sample are similar to those of the wider group (e.g. in terms of age, gender, CD4 count and years since diagnosis).


Expresses the risk that, during one very short moment in time, a person will experience an event, given that they have not already done so.

hazard ratio

Comparing one group with another, expresses differences in the risk of something happening. A hazard ratio above 1 means the risk is higher in the group of interest; a hazard ratio below 1 means the risk is lower. Similar to ‘relative risk’.

longitudinal study

A study in which information is collected on people over several weeks, months or years. People may be followed forward in time (a prospective study), or information may be collected on past events (a retrospective study).

Of note, individuals who were taking treatment and who had a low viral load were not included in the phylogenetic analysis as the technique is not reliable with viral loads below 1000 copies per ml. This actually has the advantage of excluding from the study people on treatment who are unlikely to transmit HIV – the sample for phylogenetic analysis is skewed to people with recent infections, many of them undiagnosed.

A total of 9812 individuals took part, of whom 3969 were HIV positive. Of these, 1589 had a viral load over 1000 copies per ml and had provided blood samples which were successfully sequenced for phylogenetic analysis.

Four hundred and sixty-nine of these samples were grouped into 202 transmission clusters, of which 90 clusters included at least one woman and one man. These 90 clusters were then analysed further. 

Age gaps between sexual partners – key findings

In the 90 clusters, there were 188 possible pairings between women and men. A comparison of all possible pairings across age and sex categories identified two particular patterns.

Firstly, 20% of pairings were between men aged 25-40 and women aged 15-25. Secondly, 31% of pairings were between men and women who were both aged 25-40.

In other words, men aged 25-40 are the primary source of HIV acquisition in adolescent girls and young women. Many of these men probably acquired HIV infection from women aged 25-40, the group with the highest prevalence of HIV. Over time, when the current group of adolescent girls and young women reach their thirties, they will have become the next group of women with high HIV prevalence, and will probably perpetuate the cycle of HIV transmission to men in their thirties – who will infect the next cohort of adolescent girls and young women. (See the graphic in the journal article illustrating the cycle of transmission).

Age differences between women and men in transmission clusters were large for younger women, but not for older women. Women aged 15-20 had male partners an average of 11.5 years older; women aged 21-25 had partners 7.0 years older; women aged 26-30 had partners 1.5 years older; women aged 31-35 had partners 1.7 years older.

Of men aged 25-40 who were linked to a younger woman, 40% were also linked to a woman in their own age group.

People who were in transmission clusters were frequently unaware of their infection, were not taking treatment and had a high viral load. 

Internal migration – the study

Labour migration is another key issue behind the HIV epidemic in South Africa, having a great impact on sexual partnerships between men and women. The second study, by Adrian Dobra, Frank Tanser and colleagues is published in the January edition of AIDS. Data come from a longitudinal cohort of all households in a predominantly rural area within the Umkhanyakude district of KwaZulu-Natal.

The community is characterised by frequent migration to other parts of South Africa, low marital rates, multiple sexual partnerships and high rates of undiagnosed HIV.

All households were visited each year (between 2004 and 2014) and the researchers interviewed and tested for HIV all household members over the age of 15 who agreed to take part. Information about migration has been carefully recorded throughout, noting when and where people moved to. When household members were absent, the researchers asked family and friends about their whereabouts.

The analysis focuses on 17,743 individuals who tested for HIV twice or more as part of the study – it can therefore look at the relationship between new HIV infections and migration.

Internal migration – key findings

Within the cohort analysed, around one in five men and women in their twenties migrated at least once, generally for periods of between 8 and 24 months.  A quarter of all migrations were to places within a 100 km range and key migration destinations were Richards Bay (55km away), Durban (205km) and Johannesburg/Pretoria (473km).

Migrations to nearby destinations were associated with increased risk, especially for men:

  • Men who migrated 40km in a year had a 50% increased risk of acquiring HIV (hazard ratio 1.5).
  • Men who migrated 169km in a year had a 75% increased risk.
  • Men who spent 44% of their time away from the home community had a 50% increased risk of acquiring HIV.

Women’s risk was associated with further and longer migrations than for men:

  • Women who migrated 109km in a year had a 50% increased risk of acquiring HIV.
  • Women who migrated 652km had a 75% increased risk.
  • Women who spent 90% of their time away from the home community had a 50% increased risk.

“Larger average migration distances per year and increased periods of residence outside the rural study community represent proxy for key risk factors of HIV acquisition such as increased number of sexual partners, increased likelihood of risky sexual behavior, detachment from family, friends, community, and social norms, increased vulnerability, or lower socioeconomic status,” the authors say.

At the 21st International AIDS Conference (AIDS 2016) in Durban this year, Frank Tanser – one of the authors of this study – said that phylogenetic analyses done in this district also highlighted the importance of migration. A number of new infections throughout the local area can be linked back to individuals living in ‘hot spot’ communities beside the highway. 

Especially as migrants are often diagnosed very late and have difficulties engaging with and being retained in medical care, the researchers say that HIV testing and treatment services in South Africa need to be better adapted to the needs of migrants.


de Oliviera T et al. Transmission networks and risk of HIV infection in KwaZulu-Natal, South Africa: a community-wide phylogenetic study. Lancet HIV, online ahead of print, 2016. (Full text available with free registration).

Dobra A et al. Space-time migration patterns and risk of HIV acquisition in rural South Africa. AIDS 31:137–145, 2017. (Full text freely available).