Conflict and changing patterns of migration from Africa: the impact on HIV services in London, UK

Abstract

Among those living with diagnosed HIV infection in the United Kingdom in 2003, 36% were described as black African, making them the second largest group [1]. There have been substantial changes, however, in the epidemiology of this group over the past 15 years. Africans have been coming to the UK for many years. Traditionally, emigration from these countries was largely motivated by education and employment opportunities; more recently increasing numbers have sought asylum from social and political upheaval [2]. According to Foreign and Commonwealth Office figures, there have been noticeable influxes of populations to the UK associated with conflicts in countries such as Eritrea and Somalia. This study sought to examine the impact of armed conflicts on the populations of Africans attending an HIV unit in London, UK, in an attempt to improve the response of services to this group. Data were collected retrospectively on all HIV-positive Africans (defined as individuals born in Africa) attending an HIV unit in central London between 1985 and 2003. Information on sex, self-assigned ethnicity, country of birth and year of HIV diagnosis was collected from the clinic database. Details of armed conflicts in Africa over the past 20 years were obtained from websites for the US Committee for Refugees and the UK Home Office. Dates of conflicts were correlated with trends in country of origin and year of HIV diagnosis of our clinic population. A total of 656 African individuals from 34 countries presented during the study period; 40% were men and 60% were women. Figure 1 shows the annual presentation patterns of African patients by country of origin, in conjunction with periods of social and political unrest. Ugandans accounted for 41% of Africans presenting before 1994, but only 23% subsequently; civil war in Uganda occurred early in the study period (1985–1989). Since 1999, Zimbabweans have accounted for the majority of new African HIV presentations (19%); possibly reflecting the high levels of internal violence experienced there since 1997. Similar correlations are evident in Congolese and Rwandan (highest rates in early 2000s) and Somalian and Ethiopian (highest rates in late 1990s) populations. Conversely, countries with a high prevalence of HIV but low levels of conflict, such as Botswana, do not demonstrate the variability in presentation over time seen in countries that underwent social upheaval, and are underrepresented in our cohort. Interestingly, the numbers from South Africa have increased since 1998, but the majority have been of white ethnic origin. A possible explanation for this trend may be that the end of apartheid has been a time of social upheaval for the white population, and has allowed greater freedom to travel for black South Africans.Fig. 1: Presentation to HIV services in the UK. Uganda,Rwanda,Ethiopia,Congo,Somalia,Zimbabwe,Botswana,South Africa.The epidemiology of newly diagnosed HIV-positive African patients at our centre is changing, with migration patterns being closely related to conflicts. Data from the Health Protection Agency support this view. Between 1988 and 1997, 20–40% of all new HIV infections diagnosed in the UK but acquired in Africa were probably acquired in Uganda. In 2001, however, 36% were from Zimbabwe. Health Protection Agency figures also show that the numbers of new diagnoses in the UK from central Africa have been rising gradually [1]. To some extent this may reflect changes in HIV prevalence in Africa over time, with the epidemic intensifying in the southeast of Africa in the past 10 years (UNAIDS data). Despite recent advances in HIV treatments, Africans with HIV continue to present late to services [1,3]. Late presentation to services and the underrepresentation of countries with a high HIV prevalence but lower levels of social disruption suggest that gaining access to HIV services is not a significant reason for migration. Limitations to this study include the possibility that the identified trends may be a reflection of changes in HIV prevalence in Africa, because data on international migration in Africa are incomplete [4]. The identified trends are purely descriptive and no statistical correlation can be made. The results may also be biased by the fact that delays between arrival in the UK and the diagnosis of HIV are likely to have occurred. Nevertheless, this study highlights the likely association between conflict and migration, and the consequent impact this has on health services worldwide. HIV service providers need to be aware of the changes in the social and political context of the people attending their clinic so that they are better able to respond to their needs. This is especially important in groups such as Africans, who find it difficult to access HIV services and present with more advanced disease as a consequence. Acknowledgements The authors would like to thank Johanna Baruah and Andrew Rider for data extraction. Sponsorship: Dr Burns is supported by a Wellcome Training fellowship (grant 066866/Z/02).