Immigrants are more risk of HIV in their host country than back at home

This article is more than 14 years old. Click here for more recent articles on this topic

Mathematical modelling based on the Dutch HIV epidemic suggests that, in that country at least, heterosexual immigrants from Africa and the Caribbean stand more risk of acquiring HIV in the Netherlands than they do in their home country.

This is because when migrants enter host countries, their sexual networks tend to be concentrated amongst people of their own background and there is very little mixing with the host country population.

Maria Xiridou of the Netherlands National Institute of Public Health and the Environment told delegates that surveillance had already found that immigrants in Europe tended to have higher HIV prevalence than people in their home country. Was this a characteristic of the type of person who moved abroad, or was it due to risks encountered once they had moved?


mathematical models

A range of complex mathematical techniques which aim to simulate a sequence of likely future events, in order to estimate the impact of a health intervention or the spread of an infection.

Xiridou devised a mathematical model that used immigration and HIV surveillance to calculate the proportion of immigrants who arrived already HIV-positive, the proportion who became infected on trips back home, and the proportion who acquired it in the Netherlands.

HIV incidence among heterosexuals in the Netherlands in 2010 was estimated at 1 infection in 47,000 people a year, but was 1 in every 1170 Africans and 1 in every 4600 Caribbeans. Because incidence is higher in immigrants, 72% of heterosexuals acquiring HIV in the Netherlands are migrants.

Interestingly, infections seemed to be concentrated even more amongst partners of migrants, with 81% of partners of migrants who acquire HIV being migrants themselves. This is because, in the Netherlands at least, sexual mixing between the native and immigrant populations is rare. Eighty per cent of heterosexual contacts reported by African men were with African women and 77% of  encounters among people from the Caribbean were with others from the Caribbean. Dutch native men reported that a mere 0.3% of sexual contact was with an African or Caribbean woman.   

She found that 29.7% of HIV infections recorded in the Netherlands among the three groups in her model (African, Caribbean and native Dutch) were due to infections acquired abroad, the vast majority of them acquired before ever entering the country, with 78% of those infections being amongst Africans.

The remaining 70.3% of infections were acquired in the Netherlands. But because ethnic mixing was rare, native Dutch people represented only 21% of infections acquired in the Netherlands while Africans represented 32% of them and people from the Caribbean 18%. Because of their smaller population, therefore, immigrant groups were very considerably more vulnerable to acquiring HIV in the Netherlands than were Dutch natives.

Because of this, changes in immigration policy would have large effect on HIV incidence in the immigrant population, but virtually no effect on infections in native Dutch people. A purely theoretical ban on HIV-positive immigrants would cut new HIV infections in already-arrived immigrants by 30%. A more humane ‘test and immediately treat’ policy for new immigrants would be more effective; the HIV incidence rate amongst the total immigrant population would fall from 65 cases a year in 100,000 people to only 10 cases.

Xiridou commented that her model reinforced anecdotal evidence that the proportion of HIV infections amongst immigrants that were acquired in the host country were increasing, and emphasised that the HIV ‘threat’ from immigration was largely confined to immigrants themselves.

Further information

View abstract and slides from this session on the official conference website.


Xiridou M et al. Changes in patterns of migration barely influence the heterosexual HIV epidemic in Europe. Eighteenth International AIDS Conference, Vienna, abstract WEAC0104. 2010.