Speakers:
Paul Spiegel, Senior HIV Technical Advisor & Global Coordinator, HIV Unit, UNHCR
Dieudonne Yiweza, Senior HIV Regional Coordinator, Central Africa, HIV Unit, UNHCR

Discussant:

Tom Ellman, Medecins Sans Frontier, UK

Chair:

David Booth, Research Fellow, ODI

Paul Spiegel: HIV Epidemiology among conflict-affected and displaced populations

1. Spiegel said that he would discuss the epidemiology of his research findings and dispel some myths about the relationship between conflict and HIV prevalence. Dieudonne's presentation would, he said, be more orientated towards policy implications.

2. Spiegel described the received wisdom as maintaining that:
- Conflict increases HIV transmission.
- Refugees have a higher HIV prevalence and therefore transmit HIV to surrounding host populations.

Main findings

3. Spiegel's studies compared HIV prevalence in Africa amongst refugees to that of refugee host communities and also to the prevalence of those who remained in the country of origin.

4. With caveats about data, he concluded that, in general:

- Countries that have experienced major conflict have a lower HIV prevalence than surrounding countries that have not experienced conflict.
- When HIV prevalence data for refugees is compared to that of surrounding populations, the prevalence for refugees is either similar or lower to that of host populations - not higher as has often been claimed.
- HIV prevalence gradually increases amongst refugees who come from an area of low HIV prevalence and move into areas of higher HIV prevalence.
- In countries where conflict started when HIV prevalence was very low, it remains low, even during conflict.
- In countries where conflict started when HIV prevalence was high, prevalence has stayed about the same throughout conflict, or it has decreased dramatically.
- Contrary to popular belief, war has not resulted in an increase of HIV prevalence.

Key factors determining data

- What are prevalence rates in the areas of origin?
- What is the HIV prevalence of the surrounding host population?
- What's the level of interaction between displaced populations and surrounding populations?
- What is the type and location of the displaced population's environment (e.g. urban vs. camp)?
- What phase is the emergency at?
- How long have refugees been living in camps?
- How long has the conflict been running?

Decreased risk due to conflict is overlooked

5. Most of the literature concentrates on how conflict increases risk (behavioural change, gender-based violence/transactional sex and reductions in resources and services). Usually overlooked is decreased risk from reductions in mobility, reductions in accessibility, reductions in urbanisation,

On Behavioural Surveillance Surveys (BSS)

6. Though behavioural change is deemed to be crucial, Spiegel said that evaluations of 31 BSSs in 14 countries revealed them to be of very poor quality.

7. Spiegel described making recommendations for improvements (such as capturing the interactions between refugees and host communities) and shared findings of new surveys.

8. Comparing, for example, knowledge and risk behaviour for southern Sudanese refugees in Kakuma camp in Kenya with that of the surrounding population and with southern Sudanese populations in Sudan, the camp populations proved to be far better informed and more likely to use condoms.

Implications

- Advocacy is necessary to correct the view that refugees have a higher HIV prevalence.
- This doesn't mean funds can be diverted from refugees.
- Refugees are particularly vulnerable and at risk because they are often living in close proximity to a surrounding community with a higher HIV prevalence.
- Repatriation is problematic because, having interacted with host communities with high prevalence, returning refugees may have a higher prevalence than those who have never left and may therefore spread HIV on return.
- Refugees may have a higher HIV prevalence on return, but it should also be acknowledged that their knowledge and behaviour might be better than that of people who have remained in-country.
- Returning refugees should be used as a resource and thought of just as a population that might spread HIV.

Dieudonne Yiweza: HIV/AIDS among conflict-affected and displaced populations - Programmatic and operational challenges

9. Yiweza addressed himself to the practical challenges of HIV/AIDS interventions.

10. He began by discussing the displacement cycle whereby a conflict-affected population may either become IDPs or cross an international border to become refugees. Having been displaced, these populations are surrounded by, and interact with, a host community. They also interact with armed forces, peacekeepers, aid workers and sex workers. When possible, displaced populations return to their original homes or to other locations. Both when displaced populations are amongst host communities and when they return home or relocate, they are at particular risk and therefore suitable HIV/AIDS interventions are required.

Yiweza argued that responses should work towards the following objectives:

- The affected population should be able to live in dignity, live free from discrimination and should have their human rights respected.
- HIV transmission should be reduced
- Access to prevention interventions should be linked with care and treatment programmes.
- Conflict-affected populations should not undergo mandatory testing or experience persecution or discrimination on the basis of HIV status.
- There must be multisectoral coordination between and amongst everyone working to provide relief and everyone working on HIV interventions All sectors - water and sanitation, food distribution etc. - must take HIV into consideration. The Inter-Agency Standing Committee Guidelines for HIV/AIDS in emergency situations are a useful reference point.
- The affected population must be part of the coordination - both the displaced population or refugees and the host community.

Approaches to implementation must be phased and hierarchical

- Given that there exist multiple needs and not everything can be done simultaneously, a phased approach means sequencing and prioritising.
- The minimal essential response includes ensuring: that blood is safe, that people with sexually transmitted infections are treated, and that condoms are available, promoted and used. Material for education, communication and information must be developed in the language of the affected population and must be culturally sensitive.
- When these activities are running correctly, more comprehensive interventions may be possible that link prevention with care and treatment. These include voluntary counselling and testing, prevention of mother to child transmission, antiretroviral therapy. There also needs to be thought as to the availability of these interventions/services to the surrounding population and in the areas of return and repatriation.

Implementation: realities, challenges and opportunities

- Conflict-affected populations often move to remote rural areas where facilities, government institutions and NGO services are absent or weak.
- The majority of these people are children, young people and women.
- The host population may have similar or greater needs to the displaced population or refugees. Both groups need to be kept in mind simultaneously at every stage in planning, executing and evaluating interventions.

- Country and/or district plans often overlook the needs of refugees or displaced populations - both in terms of development and HIV. Such populations are seen to be the responsibility of the 'international community'. Continuous advocacy is required at national, regional and global levels.
- Responses must be tailored according to the different challenges presented by urban vs. rural caseloads, scattered vs. concentrated groups and new vs. protected situations.
- Cross-border dimensions are crucial and challenging. Surrounding countries are often also affected by conflict. Sub-regional approaches are necessary to increase continuity of services, to ensure similar testing and treatment and to lower costs and increase efficiency. The Great Lakes Initiative on AIDS (GLIA) is one positive example of this.
- HIV must be given greater consideration before, during and after repatriation. Cross-border coordination is necessary, not least to prepare the area of return and to try and ensure ART continuation.
- Returning conflict-affected populations should be thought of as a resource. Upon their return, these populations can bring the knowledge that they acquired during asylum/displacement and use this knowledge to shape HIV interventions in their return areas.

Discussant: Tom Ellman

11. Ellman complimented the presentations. He noted that the presentations focussed on refugees and IDPs but stressed that we also need to think about people who do not flee - people who do not come under the advocacy umbrella of UNHCR and also about what the relevance of the information presented is for non-conflict environments.

In relation to Spiegel's presentation, Ellman noted that:

- Mobile populations such as migrant workers raise similar issues to those discussed in relation to conflict.

- The prospect that the Global Fund may disappear could result in serious disruption to HIV/AIDS programmes - much like disruptions caused by conflict.

- Ellman acknowledged that myths were a problem but raised the issue of whether the contexts examined were ones where it was almost inevitable that the myth would be wrong. HIV prevalence in Somalia or Sudan, for example, was very low to begin with so it is no surprise that rates are higher in the host populations of countries such as Rwanda or Uganda than for Somali or Sudanese refugees.

- Ellman noted the use of data from antenatal clinics and suggested that those who visit ANCs are self-selecting populations that may not be representative of a general population. Apparent drops post-conflict HIV prevalence may only be amongst those who are able to access ANCs.

- Because access to prevention and to care is generally so low across Africa perhaps it is only over the coming years as access becomes broader that we will see war's disastrous effects.

- On advocacy: It's clear that populations who move across borders are disproportionately affected by stigma. But it also may indeed by the case that, in some contexts, refugee populations will have a significantly higher HIV prevalence.

Addressing Yiweza's presentation, Ellman observed that:

- In relation to conflict, we need to be careful not to allow ourselves to have to win again the battles we fought in relation to HIV/AIDS in stable settings.

- The idea of hierarchies is problematic. It may be true that it is difficult to develop a comprehensive care response in conflict or a post-conflict setting. But prevention and treatment and community involvement need to go together as much, if not more, in conflict settings as in stable settings.

- The presentation highlighted the need for treatment to be given in ways that it can be continued during and after repatriation. Does it really matter whether treatment is sustainable? It is desirable but is it essential? Treating refugee populations can change attitudes and quality of life. Cessation of treatment need not necessarily damage a patient's future. Public health arguments have emphasised the risks of promoting resistance, but treatment can be stopped safely and preparing for stopping has to occur - both in conflict and in stable settings. It is wrong to suggest that treatment must be continued as a prerequisite for starting.

Discussion

12. The discussion raised specific questions about how and where data was gathered and focussed particularly on how to account for trends observed in northern Uganda. In addition, the following comments and queries were raised:

- If we are to reject the notion that conflict automatically raises HIV prevalence we must also be careful not to overcompensate by concluding that the opposite is true.

- It may be the case that HIV transmission reduces during conflict because people with HIV tend not to live as long during conflict so they will have less time to transmit also people with HIV people may have less opportunities for sexual intercourse during conflict because of opportunistic infections that they might have addressed had there been no conflict. [In response, Spiegel found this suggestion to be unconvincing, partly for reasons associated with mortality rates].

- It is important to have data that is disaggregated for age and gender.

- Often overlooked are those in the military and their families. Who is institutionally responsible for working with this important conflict-affected population?

- A further important issue concerns the demobilisation and reintegration of combatants and the associated risks for the spread of HIV.

- There is often a focus on women and children but we also need to know what happens to males of working age.

- In terms of practical advocacy, UNAIDS or UNHCR might do more to help agencies to include in their funding applications the HIV/AIDS dimensions of displacement and asylum.

- Have we seen populations not repatriating because treatment is not available on the other side of the border? [In response, it was noted that large-scale resistance to repatriation on grounds of treatment availability has not been identified].

- Spiegel picked up on how refugees and rape have been identified as being responsible for the spread of HIV in Rwanda but that these claims are poorly substantiated by evidence. The significance of rape during war for the spread of HIV has been exaggerated. Myths are recycled. Spiegel also said that HIV prevalence in the military is often exaggerated. Often soldiers are recruited from rural populations where prevalence is low.

- Ellman elaborated on the issue of stopping treatment. He noted that stopping is a difficult issue because it might seem confusing to patients to be given a strong message about adhering to treatment regimes and then to be told that it can be safe to stop treatment. There are ways of stopping treatment regimes that do not promote resistance and that aren't necessarily a disaster for the patient. It is very important to ensure that communities understand what course of action is being advised and why.

- Ellman also suggested that thought should be given to implications of the presenters' findings and data beyond using them to dispel myths.