A / A / A

Chapter II Exceptional situations

Date de publication
Rony Brauman

Medical doctor, specialized in tropical medicine and epidemiology. Involved in humanitarian action since 1977, he has been on numerous missions, mainly in contexts of armed conflicts and IDP situations. President of Médecins sans Frontières from 1982 to1994, he also teaches at the Humanitarian and Conflict Response Institute (HCRI) and is a regular contributor to Alternatives Economiques. He has published several books and articles, including "Guerre humanitaires ? Mensonges et Intox" (Textuel, 2018), "La Médecine Humanitaire" (PUF, 2010), "Penser dans l'urgence" (Editions du Seuil, 2006) and "Utopies Sanitaires" (Editions Le Pommier, 2000).

The majority of armed conflicts in the last 50 years, except for the war in the former Yugoslavia, have been internal armed conflicts that have taken place in third-world countries. These present practical and organisational problems as well as political and ethical issues, which distinguish them from other situations in which humanitarian medicine is practised. From a medical and health perspective, these conflicts are characterised by three kinds of consequences: significant movements of population, either spontaneous or prompted by the situation, which can cause difficult ethical problems for those involved in humanitarian efforts; a disruption in the existing healthcare infrastructure, which is always in poor condition and often deprived of local personnel for security reasons; and finally the potential outbreak of epidemics, either of infectious diseases or malnutrition.


Surgery is the most emblematic form of medical practice in the field, though it is not the most frequent. It has a central role in the common representations of conflicts, although it is primarily used in urban warfare, where the wounded are within a short distance of healthcare facilities. It presents its own specific technical, organisational and political problems.

On a political level, the location of the surgical centre must be carefully considered. The advantage for the warring groups of having access to surgical facilities means that their location is extremely important. Ease of access for the wounded, the primary selection criterion for doctors, is not only a question of topography but also requires the agreement of military officials, which means there must be immediate, regular and transparent contact with them.

The leaders of the various warring parties must make a commitment to respect medical teams, convoys and treatment centres, but this cannot be guaranteed in a context in which uncontrolled armed groups are operating. As a result, it is sometimes necessary to organised armed protection, intended to dissuade potential looters, the alternative being not to get involved at all. Humanitarian players are divided about this way of operating, which, in spite of the objections on principle and practical difficulties it raises, has been used on several occasions. In any event, and contrary to a widely accepted idea, protection of this kind is provided more effectively by local guards than by foreign forces. As they are not familiar with either the area, the language or local customs, foreign forces have enormous difficulty assessing the risks and distinguishing the various armed groups from each other. Given their uncertainty when faced with a real or apparent risk, soldiers tend to make excessive use of their weapons, triggering a cycle of violence and removing any possibility of moving through the area later, when often, a discussion would be sufficient to overcome the obstacle without becoming involved in confrontation. This is why the majority of humanitarian organisations working in situations of armed conflict, though not all, refuse any protection provided by foreign forces, whether from UN peacekeepers or national contingents. Nonetheless, even NGOs that refuse to have their activities protected in this way have occasionally made use of escorts when evacuating from critical areas. The urgency of removing teams considered to be facing a significant threat meant compromising on principles otherwise critical to ensure the continuity of action.

In medical terms, operating techniques depend on the equipment and trained staff available and are therefore not dictated purely by clinical considerations. Furthermore, these techniques vary from one surgical school of thought to another, and the short periods of time for which surgeons are generally sent out by humanitarian organisations makes it even more difficult to provide the necessary continuity of care. Surgery requires special equipment – surgical instruments, sterilisation and anaesthetic equipment, lighting and metal furniture – and therefore also needs appropriate facilities. Water and electricity are used in abundance. The logistical problems of set-up, maintenance and supply need to be tackled from the outset in order to avoid serious consequences for patients. It is worth noting that the degree of specialisation of young surgeons, who are trained to work with highly sophisticated technical facilities (primarily in terms of imaging and biology), and consequently have little inclination to get back to grips with clinical skills and push the boundaries of their area of competence, does not lend itself well to practising surgery in these conditions. Older surgeons, with a more general range of skills and a more clinically-based focus, are generally better equipped to practise in unstable situations.

This difficulty is mitigated, however, by the existence of a growing number of competent surgeons in the countries of the South, who are often keen to join foreign humanitarian teams, where they find conditions that enable them to stay in their own countries and practise their profession. Whilst in the 1980s, it was rare for local healthcare professionals to be involved, it has since become the rule and today’s humanitarian medical and surgical teams are most often made up of a majority of people from the country where they are operating, supplemented by expatriate colleagues as necessary. Among the problems frequently observed are post-operative care, in particular functional rehabilitation for trauma patients, which is essential to ensure their recovery, and specialist surgery, such as neurosurgery and maxillofacial surgery, gaps which can usefully be filled by medical NGOs.

The speed of bringing in the wounded and the quality of the conditions in which they are transported to the surgical centre are essential factors in their immediate survival. Ambulances and specialist personnel are the exception rather than the rule in this type of situation, which is why preoperative mortality is generally high, particularly given that the wounded often arrive in waves, in the wake of fighting or attacks. In these conditions, it is important to triage patients so that those cases that are both most urgent and most operable are treated as a priority. A head and brain injury where the chances of recovery are very slim, for example, will be given a lower priority than an open fracture of the leg, which is less urgent but more operable, with a good chance of a successful outcome. There are several medical classifications in use, developed by military surgeons during the First World War.For a detailed description, see in particular Alain Larcan, “Triage”, in Pierre Huguenard (dir.), Catastrophes, De la stratégie d’intervention à la prise en charge médicale, Encyclopédie Médico-Chirurgicale, 1996.These have evolved with medical advances and the changing contexts in which conflict takes place, but the logic behind them has remained the same. This is based on distinguishing between immediate vital emergencies (primarily respiratory distress and circulatory distress), vital emergencies that can be delayed for a few hours (such as open fractures and wounds to the base of limbs) and finally, non-urgent cases or those that are too complex.

In human terms, triage is a painful process for the medical teams, but it is a practical imperative that has to be applied whenever there is an overwhelming influx of large numbers of wounded people. It jars to some extent with the “compassionate” practice of medicine, which would mean, conversely, that those who were most seriously injured or were suffering the most would be operated on as a priority. Whilst for military doctors, it was a question of saving the largest possible number of “useful lives, namely the lives of combatants”,Alain Brossat, “Triage, égalisation, modernité”, in Passant ordinaire, no. 45-46, June-Sept. 2003. the principle of efficiency, although still in effect, now has a different focus. For humanitarian doctors, it is now about preserving life for its own sake, purely as human life. Although justified from a utilitarian point of view, this method of rationalising care still clashes with the logic of war, which does not place the same value on every life. The life of a combatant is worth more than that of an ordinary civilian and that of a leader more than that of an ordinary man. These differences in perspective can result in some stark confrontations, from which the humanitarians rarely emerge as the winners.

Some surgical indications put doctors in situations where they face similar difficulties. Amputation, for example, an ordinary practice in a war situation, was frequently rejected by young militiamen wounded in Mogadishu or Monrovia during the wars in Somalia (1991 onwards) and Liberia (1989-1998). The combatants believed that this assault on their physical integrity was undignified and they preferred to die with their limbs intact rather than survive but be mutilated. Some, who had arrived at the hospital unconscious, woke up to find they had lost a leg or an arm. They had been operated on as an emergency to prevent the inevitable spread of gangrene or because the limb was deemed impossible to save because of a significant loss of bone mass. Although medically based, amputation was socially unacceptable. The reactions of the wounded, which were sometimes seriously threatening to the surgeons concerned, reminded medical staff that they could not see themselves as the owners of their patients’ bodies and that a surgical operation cannot therefore be carried out, even in a vital emergency, without the consent of the person concerned, or their representative. The principle of efficiency cannot, in fact, be the ultima ratio for medical practice, which must do its utmost to take the patient’s view of their condition into account. To borrow Georges Canguilhem’s words, what interests the patient is not so much the function of a particular organ that is threatened by disease, but their “relationship to life”, i.e. the way they put up with their social environment.”Georges Canguilhem, “Le Normal et le pathologique”, PUF, Paris, 1966.In the situation described above, the obligation to help someone at risk comes up against a limiting factor, namely the desires of the person concerned.

Whilst most field surgery takes place in an urban environment, for the reasons explained previously, it is sometimes possible to treat the wounded in rural environments. In Darfur, for example, an agreement has been reached with the various armed groups to enable a mobile ICRC team to use light aircraft or cars to reach wounded people living in isolated villages. Operating in a tent put up specially for the purpose (although the temperature inside is often unbearable) or in a room in a house with as much dust as possible kept out, the surgeon carries out basic surgery, dressing wounds, extracting bullets or shrapnel and setting fractures. However rudimentary it may be and because there are no resources available for additional examinations, palliative surgery of this kind must be carried out by experienced surgeons. In addition to the valuable service it provides to the wounded, the other positive aspect of assistance of this kind is that it persuades the armed groups of the advantage they gain by allowing the medical teams to move around freely, and thus plays a part in keeping them safe.

When local conditions mean it is impossible to set up a surgical centre, either for security reasons or because the authorities will not allow it, it is sometimes still possible to set up a hospital on the borders of the country concerned. This is what was done by the ICRC on the Pakistani side of the Afghan border during the years of the Soviet War (1980-1989), and by MSF in Jordan and Iran to treat those wounded in the Iraq War (2003-…). To make these facilities accessible, it is necessary to set up a network to identify, select and transport the wounded to hospital. This therefore relies on close contacts inside the country, to ensure that the patients to be operated on are selected on the basis of medical need and not financial or political criteria, which is not the least of the difficulties involved.



During the 1970s, war and political violence in countries of the South resulted in large scale populations displacements. The Geneva Convention of 1951, the legal basis of the United Nations Office of the High Commissioner for Refugees (UNHCR) had been drafted and adopted to deal with refugees from Eastern Europe, which is why neither the USSR not any of the Communist countries were party to it. The Convention, which provides for legal status and the provision of assistance to refugees, has been gradually enlarged and is now universal. The distinctive feature in its development is the shift from an individual procedure of granting refugee status, which still applies in Europe, to a collective, prima facie procedure, under the terms of which whole groups can be granted refugee status. The change reflects the more limited resources of host nations and the larger scale of movements of refugees in the countries of the South as a result of war rather than political persecution. In practice, it means setting up camps, within which each refugee is also individually registered, giving them access to the aid systems developed by the United Nations and private humanitarian organisations. Between 1976 and 1982, the number of refugees identified in the world and assisted by the UNHCR increased from three to eleven million, driven by the fresh upsurge in armed conflict in the third world, and continued to grow in the 1990s.

The term “displaced persons” (Internally Displaced Persons or IDPs in the jargon of international aid) refers to people who have been forced to flee their usual homes as a result of war, like refugees, but who, unlike the latter, have not crossed an international border. In some sense, they are internal refugees. Because of the state of war in their country, they benefit – at least in theory – from the protection provided for in the Geneva Convention of 1949, which is distinct from the Convention of 1951 referred to above. The earlier Convention concerns populations affected by war in their own country, the later one those who have fled to a host country to escape violence or persecution. In the language of treaties, these legal distinctions convey significant differences on both a political and practical level.

People living in displaced persons’ camps are thus in direct contact with the belligerents and under the administrative authority of the government and the army, who are, of necessity, party to the conflict which resulted in the population being displaced. People living in refugee camps are in a different political situation, since they are under the legal authority of the host country. Host governments are more than often happy to delegate their responsibilities to the UNHCR, which then calls on the NGOs to carry out assistance operations on the ground.

Humanitarian workers therefore a priori have greater freedom of action in refugee camps under the authority of the UNHCR, than in displaced persons’ camps managed more directly by government authorities. All camps, whatever the origin of the population living in them, are nonetheless political areas at the same time as they are humanitarian sanctuaries, because the war that caused them in the first place does not stop at their doors. Those involved in humanitarian work need to be aware of this, or risk unknowingly lending support to certain belligerents. The refugee camps set up in Eastern Zaire/ Congo after the genocide of Rwandan Tutsis provide a particularly striking example of this.

In July 1994, a mass of over one million people formed in just a few days in Kivu Province, on the border with Rwanda. Their fear was well founded, because the “law of suspects” and killings of civilians it resulted in had become the rule in the country. The UNHCR, NGOs and numerous governments quickly mobilised to organise relief, made all the more urgent by the cholera epidemic that broke out in the days following the arrival of the refugees.

In a chaotic atmosphere, characterised by violence, death, and the presence of the international media and relief workers from all kinds of organisations, a huge assistance operation was mounted. Any attempt to distinguish refugees from murderers was suspended, with the necessity of treating the sick and protecting public health taking precedence over everything else. All were victims of the epidemic and equally in need of assistance which, in order to be effective, needed to be targeted at the whole of the population in Kivu. But what was true of a fatal epidemic did not automatically apply to assistance in general, because criminals on the run are not supposed to benefit from the protection of the United Nations or international humanitarian systems.

As a result, once the epidemic was over, the main NGOs present in the camps protested and appealed to the United Nations Security Council, asking it to send in an international police force to separate the perpetrators of genocide from the main body of the refugee population. The deployment of UN peacekeepers was discussed by the Security Council on several occasions at the request of the Secretary-General, Boutros Boutros-Ghali, but came to nothing. Meanwhile, people had returned to work in the feeding centres, dispensaries, hospitals and warehouses storing food and relief supplies, whilst the government’s political and military apparatus in place during the genocide was reorganising itself, under the protection of the aid system and with access to its resources.

Those responsible for the genocide and those directly involved in it, who supervised the refugees and took control of the camps, were now officially recognised as the operational contacts for the aid organisations. It meant that they recruited several thousand “refugees” employed as technical and administrative assistants for the NGOs, managing warehouses, drawing up lists of those who should benefit from the aid and organising distribution. In other words, the aid system, which had turned a deaf ear and a blind eye to the political issues inherent in the situation, was providing a criminal regime with the symbolic and material resources it needed to establish and extend its control over a population of hundreds of thousands of people. Military attacks were quickly launched against Rwanda, in addition to the countless acts of often bloody violence committed inside the camps against those who stood up to those in charge. A new war was being planned, financed amongst other things by international aid.

A minority of NGOs refused to continue their work under such conditions and withdrew, explaining their position publicly. The majority of their sister organisations took the decision to remain until the end, because of the “humanitarian crisis”, in the name of neutrality and the overwhelming necessity of staying with the victims. The war that broke out in Eastern Zaire in November 1997, when the Rwandan army attacked the camps, led to a general dispersal. According to an assessment by MSF, almost 200,000 people, hunted down by the Rwandan soldiers, died of hunger, disease and violence over the following months. Again, the humanitarian organisations allowed themselves to be used as instruments, this time by the enemies of those whom they had been assisting until then. Thousands of people were slaughtered on the spot, as the result of the presence of “liaison officers” accompanying the humanitarian teams and thus being able to locate people who were on the run. Those who had been spared were repatriated either willingly or forcibly by the UNHCR to Rwanda, where some of them were imprisoned and others either taken to displaced persons’ camps for a time, or sent back to their villages.

The intensity and radical nature of the violence of this situation were undoubtedly exceptional, but the players involved and humanitarian practices used were those normally found in refugee camps. Standard resources and techniques, which will be discussed later, were methodically applied by experienced personnel. Epidemiological analyses, at least during the three years of “stability” in the camps, only served to confirm the technical relevance of the actions carried out. Such operational success is not unambiguous, however, nor without its darker sides.

Under pressure from their institutional donors, who were urging them to act, but also pushed from within by the conviction that action is always preferable to standing aside, the humanitarian organisations confused the objective of mutual assistance, which is their raison d’être, with the practical system that makes it a reality. Humanitarian organisations, like other institutions, have their own needs, which they identify too often with those of their supposed “beneficiaries”. Such disregard may not have any consequences in circumstances where there is less of a political burden on them. Even those who find themselves supporting criminal policies in certain situations can engage in beneficial work elsewhere. One key point still holds true: whilst everything suggests that humanitarian aid, and therefore the medical activities that form a substantial part of it, equates to the deployment of logistics resources and technical savoir-faire, it is essential to understand the context in which these take place if one wants to preserve its meaning.


Not all refugee camps are under the control of criminal groups similar to those just referred to, far from it, but they are always characterised by rivalry between competing groups for the control of populations and material resources. This is a reality that the aid organisations have had to come to terms with since the phenomenon of the refugee camp as a humanitarian territory first appeared. This occurred in the late 1970s, in the Cambodian camps in Thailand. The novelty of the situation, in both political and medical terms, merits some attention.

Following the overthrow of Pol Pot’s regime in January 1979 by the Vietnamese army, several tens of thousands of Cambodians found themselves stranded in the “no man’s land” that separates Thailand from Cambodia. Having come to the end of a forced march in the forest lasting several months under pressure from the Vietnamese forces, during which their only nourishment was what they could find along the way, the majority of the refugees were in a severe state of malnutrition. Thousands of them were suffering from malaria, skin ulcers, various forms of parasitosis and digestive infections. Many died in the days following their arrival, whilst the aid system was struggling to get up and running. Initially, the Thai government refused access to a militarily sensitive area, forcing teams from abroad to remain in the capital, Bangkok. Only the few NGOs and the Red Cross association already on the spot were able to start providing a limited amount of care, very much below the level of what was needed. In order to carry out their task properly in such conditions, they would have needed medical equipment and drugs, storage facilities and transport resources from the outset, but none of them was equipped to respond to an emergency of this kind.

Under pressure from the international media and Western diplomats, the Thai government agreed to open up the region to foreign relief workers. The situation gradually improved, but the waiting was fatal for those in the most severe condition. New groups of Cambodians continued to arrive over the course of the following weeks, bringing the number of refugees to 300,000, and were settled in various camps set up along the Khmer-Thai border.

An unusual situation emerged from the Cambodian invasion of Vietnam, which was supported by governments allied with the USSR and condemned by others, in particular Cambodia’s anti-Soviet neighbours, including Thailand and China. The new Cambodian power base installed in Phnom Penh by the Vietnamese was not recognised by the United Nations Security Council, which implied continuing international recognition of the Khmer Rouge regime. As far as international law was concerned, the border camps thus became the headquarters of the legal State controlled by Pol Pot. As a result, the Khmer Rouge necessarily became the natural point of contact for the United Nations and Western countries. The Khmer Rouge troops had dragged tens of thousands of people into their collapse, using them as slave labour in their strongholds and during their movements. These were the people who formed the first core of refugees, which later grew to incorporate other civilian populations who were mobilised and used by other political movements, led by Prince Sihanouk and the former Prime Minister Son San. Each of these parties, drawn together in an anti-Vietnamese nationalist front, administered a portion of the border area under the tight control of the Thai army, which viewed these forces as a shield to protect it from the feared Vietnamese army.

This displacement of populations led to the largest humanitarian mobilisation, private and UN organisations combined, since the Biafran War. The extensive media coverage of the situation, which occurred just four years after the end of the Vietnam War, and the gravity of the initial condition of the refugees, triggered an unprecedented influx of private donations and public finance. The aid operation, which was coordinated by a United Nations agency created specifically to deal with the situation (UNBRO: United Nations Border Relief Operation), was to last for 12 years, until the refugees were repatriated to Cambodia in 1992. With the significant financial resources available to them and a considerable degree of latitude in their actions, NGOs practising medicine transformed their working methods. Mobile field hospitals, therapeutic feeding centres, monitoring systems for transmissible diseases, water and sanitation techniques, vehicle fleets, workshops and radio communications resources became everyday tools for medical relief operations in the 1980s, and would go on to be used in major crisis situations. Both in terms of the material resources used and the general way relief was organised, the Khmer-Thai border, from 1980 onwards, was the main place where humanitarian medical practice was redeveloped.

Even in Cambodia, most of the country and the majority of the population found themselves under the authority of an unrecognised government. As a result, the UN aid agencies were unable to establish a presence there and Western diplomacies had no desire to. These conditions hampered any increase in international aid to a country hit by one of the most violent tyrannies the world has ever known. The images of tens of thousands of people arriving in Thailand in an appalling state had hit the headlines on television and in the press, putting across the idea that the country was ravaged by famine. Multiple appeals were quickly launched by private solidarity organisations, UN agencies and governments to respond to the emergency. The government in Phnom Penh used this groundswell of emergency aid as a way of emerging from isolation and engaging in relationships with other States on an equal footing in an effort to gain recognition de facto, if not de jure. It was inflexible in its demands to control the aid, however, and refused, apart from a few rare exceptions, to allow organisations working in Cambodia also to operate in the camps in the border zone in Thai territory. In this highly polarised situation, the humanitarian movement was divided between those who thought the priority was to work inside Cambodia in order to provide relief to the population as a whole, and those who refused to operate under the control of the regime, which they accused of diverting aid for its own benefit and that of the Vietnamese army. This division remained throughout the 1980s, until the political agreement that put an end to the conflict and allowed the return of the refugees, in 1992.

The political events behind the displacements of populations therefore raise difficult problems for humanitarian workers, who have to focus their efforts on directing their aid according to need, and avoid getting caught up in the power struggles that continue to exist within the camps. Whilst it is not imaginable to escape totally from being used as a political instrument, it is often possible to create a degree of freedom that allows those involved in aid to carry out effective relief operations in accordance with the main humanitarian principle, namely impartiality. To achieve this, humanitarian medicine requires experienced management on the ground – as well as at the headquarters of the organisations – able to negotiate with the belligerents on the basis of their recognised rights and obligations under the Geneva Conventions. If they are blocked, protest and even public denunciation in the local and international press are useful tools, insofar as few political powers are insensitive to their public image.



As well as displacements of populations, conflicts result in disruption of existing health services. The supply of care services decreases whilst demand increases, because of the deterioration in general living conditions. The incidence of infectious diseases increases, acute malnutrition becomes more widespread and epidemics appear. The vast majority of famines in the last 150 years are closely linked to armed conflicts, sometimes because of blockades or looting of crops and livestock and often because of people being uprooted and the extreme difficulty displaced populations face in meeting their needs. The correlation is not quite so clear for epidemics, which can occur independently of any conflict, but many of them do develop in these conditions, driven by the collapse of monitoring and control structures. Infectious diseases and parasitic infections have also increased sharply in war-ridden African countries.

The presence of foreign medical teams is a way of responding, at least partially, to these new needs, which local health structures are unable to cope with. It also offers the advantage, because these teams are not involved in the conflict, of making hospitals and dispensaries “neutral” areas, thus partially distancing them from the influence of the belligerents. It is frequent for local medical teams to be able to remain in place as soon as foreign teams recognised as humanitarian organisations are working alongside them.

Outside of urban warfare, the main problems are infectious diseases and parasitic infections. Most conflicts involve violence and confrontations that fluctuate on a local level in rural environments, where there are no means of transport available for the wounded to get to hospital on time and in decent conditions. War wounds represent thus only a small proportion of medical needs. Only urban wars produce a large number of wounded people who can be operated on.

Humanitarian medicine in conflict settings is therefore most often health care to people displaced by war. It is generally practised in rural areas that are difficult to access. Based on the experience gained from the 1970s and 1980s onwards, a period of intense growth in the number of people displaced by war and an increase in the number of refugee camps, there has been a gradual development of know-how, resulting in operational techniques in which epidemiology and logistics play an important role.

The camps on the Khmer-Thai border referred to earlier, as well as those in Somalia following the Ogaden War, were where many NGOs and UN agencies developed their first operational systems. Over the years, with increasing experience and resources, standardised lists of drugs and equipment were drawn up, with everything packed into kits and accompanied by a set of instructions for use. Clinical and treatment guides were published, covering the experience of “precarious situations” that academic textbooks ignored, along with instructions on water purification. This way of operating met the needs of the teams on the ground, who were faced with tasks for which they were not necessarily prepared. In addition, it was a way of simplifying the supply process and monitoring operations, by standardising the physical resources used. In medical terms, more specifically, its aim was to harmonise practices amongst carers, which were highly diverse as a result of the relatively short period expatriates tended to stay for, which was rarely longer than a year. That said, the kits and in fact, the whole system – in particular the size and composition of teams – need to be nuanced depending on the circumstances. The potential presence of other players in the medical field and different geographical situations or sanitary conditions mean they have to be adjusted on a case-by-case basis. In addition, the point at which refugees arrive and are settled in, i.e. the first few weeks after the exodus, is different from the following period, when they are established, which can last for years.


During the emergency phase, which may be marked by a high mortality rate, the operation relies on an initial assessment : this must be carried out by an experienced team, which is fully aware of the main problems and difficulties in such situations as well as the capacity to respond of the organisation that is sending them out. Having made contact with the local and national authorities, the exploratory team must, as far as possible, make every effort to answer four essential questions on the history of the displacement, the structure of the population, the resources available and the main public health problems.

The history covers recent sanitary and nutritional conditions in the home region, as well as how people have travelled to the resettlement area. It should provide a way of determining the overall state of health of the population on arrival and identifying or confirming the reasons for the exodus.

Understanding the size and breakdown by age of the population is necessary for setting up epidemiological monitoring and calculating overall needs in terms of food and general supplies (shelters, blankets, cooking utensils, etc.). When it is impossible to register refugees on arrival, as is the case with a mass exodus, epidemiologists can produce a schematic map of the settlement area and use sampling to determine the average number of people in each shelter. An assessment of this kind may not be possible where the political authorities consider the number of refugees as being solely within their area of jurisdiction. These restrictions are generally based on economic reasons, insofar as it is to a host country’s advantage to report figures that are higher than the actual numbers in order to increase the level of international aid subsidies. In this case, the volume of general assistance (food, water, shelters) is based on the official figures. Medical assistance, which is not based on groups but on individuals, ends up adjusting to the reality on the ground after an initial period of feeling its way. The problem is that, in these conditions, calculating the initial mortality rate and then monitoring it, which is important for assessing how well the relief effort is performing, is difficult and subject to a high margin of error. This point will be addressed in more detail below.

Assessing material resources includes accommodation conditions as well as supplies of food and water. Housing needs to be organised in a way that avoids concentrations likely to foster the development of epidemics: measles, meningitis, typhus, infectious diarrhoea and cholera are more frequent and more severe where the population density is high. The choice of a settlement site for displaced persons or refugees is generally down to the national authorities, sometimes with support from the UNHCR. For human and practical reasons, it is preferable to opt for “small” camps (fewer than 10,000 people), but most often the government authorities’ economic and political agendas hold sway, resulting in the creation of large concentrations of refugees that can contain more than 100,000 people. The food ration must provide 2,100 kcal per day according to WHO standards and contain essential nutrients. Water and sanitation are two of the immediate priorities. It is estimated that five litres of water per person, per day are the minimum that should be provided from the outset and that ideally, 15 to 20 litres per person per day are necessary after that. Local water sources can be captured, with water being pumped from a nearby waterway and routed through specially installed pipes, or brought in by tankers. Treatment centres consume much more (up to 100 litres/day/person for a maternity or surgical unit). Digging latrines (one for every 100 people at the beginning, then one for every 20 as soon as possible) and collecting waste need to be organised right from the outset. Public health doctors, logistics experts and sanitation technicians are therefore indispensable for ensuring these tasks are completed correctly.


There are traditionally five major pathologies in groups of refugee populations in tropical environments: measles, diarrhoea, acute respiratory infections, malaria and malnutrition. Mass vaccination and setting up oral rehydration centres and treatment dispensaries are used to provide a solution to the first four. Again, these require tried-and-tested equipment and methods. Nutritional diseases, whether they are the cause or consequence of the displacement, need to be assessed specifically through surveys. The number of people who can be treated in feeding centres, which are automatically opened in refugee camps, depends on this assessment. In recent years, echoing an increasingly frequent process in the industrialised countries, humanitarian organisations have included psychological treatment for mental problems and disturbances caused by violence and in particular, though not only, rape, as part of their care services.

Epidemiological monitoring needs to be set up very quickly in order to enable the medical team to adapt to changing needs and act as soon as an epidemic appears. Three different methods are used. First, monitoring the consultation records, secondly survey-type investigations, and finally regular “home visiting” carried out by local teams trained to look for specific symptoms (diarrhoea, fever, dehydration and malnutrition). The gross daily mortality rate is the most useful indicator, with the gravity threshold set at two deaths per 10,000 people per day. This needs to be measured regularly, which is made possible by home visiting and cemetery inspections (where these exist). The assessment will depend on whether the trend is up or down, with the change in mortality to rates comparable to those in the home or host country marking the start of the chronic phase.


Once the emergency phase is over, long-term epidemiological monitoring needs to avoid falling into the pitfall of becoming simply a routine exercise in order to remain sensitive to any possible warning signs: the mass of information gathered can make it difficult to establish a hierarchical order of events. In any event, whilst the presence of medical staff remains essential, it needs to be reduced during the chronic phase, during which the problems and responses change, not in terms of substance but in terms of scale.

In addition to the expatriate teams, these various tasks involve recruiting local medical, paramedical, logistics and administrative staff. On average, there are ten local employees for one expatriate. Recruiting them often poses some serious problems: it is important to avoid unhealthy competition in terms of salaries with local medical organisations and ensure that no particular group – ethnic, political or national – is favoured over another. Imbalances of this kind can cause dangerous levels of tension in an environment where resources are scarce. Breaking down the work between the various groups involved (NGOs, United Nations agencies and local public authorities) is carried out, not without difficulty, by ad hoc coordination bodies, where information is exchanged. The different experiences and working practices that come into contact with each other within these committees can be a source of friction and conflict, not only because needs assessment and responding to crises are far from being exact sciences, but also because the different aid institutions operate to different timescales and under very different constraints.


The return of refugees to their own country is one of the first signs of the end of hostilities. Almost all the refugee camps that appeared in the 1970s and 1980s disappeared again in the 1990s, some of the refugees having blended into the population of the host country, and others returning home.

Sometimes camps have been dismantled within a few months, or even a few weeks, once the conditions for return have been met, namely a political agreement that puts an end to the conflict in the home country. The UNHCR is frequently required to organise an orderly, voluntary repatriation of the refugees by the governments concerned – those in the home and host countries – as well as donors. In these circumstances, the symbolic issue is always a significant one: the ideal for the local political authorities and the United Nations would always be to show tangible evidence of the situation being normalised after a generally long period of violence and chaos. In 1992-93 in Cambodia, Mozambique and El Salvador, for example, convoys of lorries and buses and sometimes aircraft and trains were organised, full of families returning to the country as part of an orderly repatriation operation. These were covered by the (local) media, confirming to the rest of the population that a supposedly new era had dawned. Humanitarian medical teams were mobilised to accompany the convoys, take care of the sick and deal with any health-related incidents, although the risks were minimal. Their presence was part of the desired symbolism, to which they willingly lent their support.

In practice, however, only a proportion, and sometimes a very small proportion, of the population concerned uses the channels that have been carefully prepared for it, with refugees most often deciding for themselves when and how they will return, and avoiding the aid system altogether. Alongside the organised repatriation, for example, the immense Mozambican camps referred to earlier, which were home to hundreds of thousands of people, simply melted away in the space of a few days, whilst the United Nations and NGOs were still working on preparing to transport them from Malawi.

Branding the situation in the home country is sometimes a very delicate matter. There can be situations of ongoing conflict and local insecurity, somewhere between all-out war and real peace. A well-founded fear of being subjected to violence is, in principle, a sufficient reason for obtaining and keeping refugee status. But in some cases political constraints takes precedence over the law and on several occasions in its history the UNHCR has embarked on repatriation operations that were voluntary in name only. This was the case of the Ugandans who were forced back into Zaire in 1986, the Rohyingas from Burma, who were expelled from Bangladesh in 1992, and the Rwandans who were repatriated from Zaire to Rwanda in 1997. On each occasion, NGOs made strenuous efforts, often unsuccessfully, to oppose the expulsions by alerting the press and sometimes by asking diplomatic missions with local influence to intervene with the authorities.


During the 1980s, French medical NGOs organised clandestine missions into countries at war to which their governments had refused them access. Whilst they were clandestine, their activities were not secret as such. On the contrary, the NGOs concerned publicised their presence in these countries, in regions controlled by guerrillas. This type of action made a significant contribution to the international reputation of French medical NGOs, which distinguished themselves in Afghanistan in particular.

The work carried out in rebel areas (in Afghanistan, Eritrea, Chad and Angola) was limited in medical terms because of the logistical difficulties peculiar to this type of terrain. Largely dependent on guerrilla supply chains, it was carried out in precarious and sometimes highly dangerous conditions, in particular in Afghanistan, because of the scale and relief of the country, which meant embarking on long expeditions at high altitude, exposed to attacks by the Red Army, before arriving at the destination. The medical teams remained in isolation there throughout the winter. In these conditions, it was impossible to offer anything more than basic care, dispensed in improvised rural health centres or modest bush hospitals. The wounded were sent there on occasion, however, in the event of bombardments or nearby armed confrontations, and the doctors then had to turn themselves into surgeons, with the few instruments they had available to them.

In other countries, such as Angola, Eritrea, El Salvador and Chad, which were less difficult to access, rotating volunteers was less hazardous, though not without its perils. In all of them, the presence of foreign medical teams was of significant symbolic importance. It could not be interpreted as international recognition of the rebel movements that were hosting them, on the one hand because the humanitarian organisations concerned were purely private ventures, and on the other because humanitarian law clearly dissociates humanitarian aid from political recognition. Nonetheless, the presence of foreigners in rebel areas increased the prestige of the leaders who had welcomed them amongst the population, contributed to breaking down their isolation and created a de facto relationship between distant societies. French humanitarian organisations, in particular the medical organisations, thus played a role in warning and putting forward criticism of the Soviet War in Afghanistan, condemning the devastation caused. In doing so, they contributed to breaking down the isolation of the resistance and armed opposition movements that were little known by the press or viewed negatively because of their somewhat questionable alliances. This was certainly the case for the UNITA rebels in Angola, who were supported by South Africa in their war against the pro-Soviet regime in power in Luanda, and who were stigmatised as supporters of apartheid as a result. The action of humanitarian organisations in the part of the country controlled by this group of combatants served its political interests by partially “opening it up”, particularly in the eyes of its own population. This is no doubt why its leaders were happy to make room for humanitarian medicine, even at the cost of numerous logistical difficulties. Political reasoning is not the same as humanitarian reasoning, but it can contribute to it by strengthening it indirectly.

Clandestine operations gradually disappeared with the end of the Cold War. Until then they had been justified, in some cases, by the fact of certain countries at war being closed to humanitarian NGOs. When a situation of this kind arose, it needed a neighbouring country to agree to let the relief teams transit through its territory to get round the problem. This is what Pakistan did for Afghanistan, for example, Zaire for Angola, Sudan for Eritrea and Honduras for El Salvador, serving as corridors but also more or less openly accepted sanctuaries for the guerrilla movements who set up bases and even official representations there. As mentioned earlier, the refugee camps were under the control of parties associated with the guerrillas, which increased their room for manoeuvre. In the new political order that emerged after the collapse of the USSR, this kind of political configuration no longer existed and support for armed rebel groups was no longer officially accepted even though, of course, they continue to exist in more discreet ways. Furthermore and above all, it is exceptional for guerrillas to control large areas with a significant civilian population in a stable manner. Both these conditions were essential for moving teams and their equipment into rebel areas and settling them there. Such conditions rarely exist in contemporary armed conflict situations, which are in fact less common than they were in the 1980s and 1990s, which explains why clandestine humanitarian missions are no longer common practice.


The disappearance of clandestine missions is therefore related to a change in the nature of war, characterised by the fragmentation of guerrilla groups, which is itself the result of the withdrawal of the major powers that previously supported them. One consequence of this change, during the 1990s, was that the number of displaced persons grew significantly whilst the number of refugees fell. In addition to the reasons referred to above, it was the transfer of assistance from outside to inside countries at war, in the context of restrictive policies in respect of hosting refugees, that lay behind this phenomenon. Until then, refugees had fled countries at war led by pro-Communist regimes, in staggering proportions. Afghanistan, Ethiopia, the three countries in Indochina, Angola and Mozambique were the source of the vast majority of exoduses that ended up in their neighbours’ border regions. With the end of the Cold War, several of these conflicts gradually drew to a close and most of the refugees returned to their respective countries. Whilst not all of these wars ended, and others began, the international strategy on movements of populations changed. Refugees were and are no longer living witnesses, who could be exploited for political ends, of the failure of Communism and the superiority of democracy, those who had “voted with their feet” against tyrannical regimes. In this new context, on the contrary, they were and still are perceived as a burden.

For all that, assistance from the United Nations did not end but was reoriented. Under pressure from national governments, the UNHCR saw its mandate extended to the management of displaced populations within their countries of origin. It was a matter of containing movements and re-internalising conflicts by trying to stop fleeing civilians, offering them assistance in situ and establishing both a civilian and military international presence. This new mission frequently took place in the context of peacekeeping operations. With the disappearance of bipolarity, taking with it the end of the practice of the automatic veto in the United Nations Security Council, the number of military contingents under a UN flag increased. Between 1990 and 1995, 45,000 soldiers were deployed, the same number as during the 45 previous years. This trend has continued – over 120,000 UN peacekeepers were present in various war zones in 2008 – but is now expected to slow down because of the very high costs of such deployments.



Humanitarian organisations were called on to cooperate closely with peacekeeping forces by the UN Secretary-General from the outset. The “Agenda for Peace”, presented by Boutros Boutros-Ghali in 1992, sees NGOs as a lever for achieving peace and aims to incorporate them into a broader United Nations scheme, alongside diplomats, soldiers and economic players. From the point of view of private humanitarian organisations, however, this is not an acceptable position, because it leads to their being placed under the authority of a party to the conflict and consequently forces them to abandon their position as external to the belligerents, which is an essential condition for delivering aid impartially. The UN’s plans for restoring peace are inevitably made up of political alliances, which necessarily exclude some parties, leading UN troops to take part in the conflict or organise blockades in order to force rebel groups to surrender. This is why, contrary to the wishes of the UN Secretary-General and widespread superficial impressions, the humanitarian approach cannot be the same as that of the peacekeepers, however legitimate the latter may be. Moreover, in the context of the new political situation that has emerged since the fall of the Berlin Wall, the work of humanitarian organisations has extended to other countries that until now were partially or totally inaccessible, and are now open. This expansion was supported and accelerated by the significant level of funding made available to NGOs by the European Union, which set up an agency to finance humanitarian aid in 1990 (ECHO, European Commission Humanitarian Office), which immediately became – and has remained – the leading donor for humanitarian aid in the world. The new economics of aid are characterised by an increasing number of participants of all kinds and the scale of their deployments, in a context in which conflict is increasingly fragmented as a result of the withdrawal of the traditional “sponsors” of guerrillas, namely the USSR and the United States during the Cold War. Humanitarian organisations have had to refine and broaden their areas of competence, in both operational and technical, as well as political terms.


In this significantly more complex political environment, the question of the security of humanitarian teams arises. 2006 was the bloodiest in the history of NGOs, with the attacks several of them suffered in Darfur and the killing of 17 employees of Action Contre la Faim in Sri Lanka. The number of serious acts of violence directed against relief workers has almost doubled in ten years. In total, according to a survey published in 2006,“Providing aid in insecure environments: trends in policy and operations”, Abby Stoddard, Adele Harmer and Katherine Haver, Humanitarian Policy Group, Briefing Paper 24, www.odihpn.org.the only serious study of the subject, 408 separate acts of violence had been perpetrated, causing 947 victims, 434 of whom died, since 1997. The survey also shows that it is local people who pay the highest price (80%), a useful reminder of the scale of their involvement in the relief teams. To be interpreted properly, however, these raw figures need to be related to the whole population of field actors. Their number also doubled over the same period, so that in relative terms, the situation did not in fact worsen but stabilised, in the context of a very significant expansion of the activities of the aid organisations described above.

Contrary to what is often heard, in reality the humanitarian “profession” is statistically no more risky than being a lumberjack, metal worker or pilot, all occupations with a comparable or higher mortality rate. Alarmist statements on the growing dangers of humanitarian assistance are far from well-founded. But these dangers do exist and should not, on the other hand, be hidden in the overall figures. Whilst they may not threaten aid as a whole, they can sometimes lead to its being significantly restricted in relation to need, by limiting movements and the areas covered, and thus preventing action from being taken in certain places and at certain times. This has been the case for several years in Iraq and certain regions in Sudan and Afghanistan, as well as in Chechnya, the Central African Republic and Somalia.

“Humanitarian corridors” and other “safe areas”, or in other words, the militarisation of humanitarian aid, are not responses but traps, except for certain specific situations where it is feasible to put in place some kind of international protectorate, as was the case in Kosovo and East Timor, in particular. The population targeted by relief efforts in fact, is generally too sparse to be protected by foreign forces. Furthermore and above all, a military shield means humanitarian workers are seen as complicit with the soldiers, who are in turn seen as invaders. NGOs then become “legitimate” political targets.

Far from limiting the danger, in fact, the militarisation of aid increases it. To work in such circumstances, humanitarian organisations have no choice but to try, through dialogue and the example of their own behaviour, to persuade the belligerants of their usefulness in human terms and their political neutrality. Experience shows that it is when they are responding to vital needs that humanitarian organisations have the power to negotiate and, consequently, room for manoeuvre that allows them to align their actions as closely as possible with the principles that underpin it. In other words, it is the services they offer, rather than the values they convey, that give them space in which to work and a relatively safe environment in which to operate. The fact remains, though, that in countries where foreign troops are fighting to defend a regime, as in Iraq, Afghanistan or Somalia, international humanitarian NGOs can hardly establish the minimum safety conditions they need to be able to work.


From the point of view of medical practice, working in displaced persons’ camps differs very little from working in refugee camps, as described above. That said, because the displaced persons are in their own country, public-sector medical facilities, where these exist, are part of the aid system, and the medical and healthcare facilities set up by humanitarian organisations in the camps themselves are only a part of it. NGOs therefore work in closer cooperation with the national health authorities than they generally do in the refugee camps. It then becomes a question of bolstering local medical teams with staff from the NGO, providing missing equipment and renovating or even extending premises where necessary.

Cooperation of this kind raises its own particular problems in terms of sharing responsibilities and changing medical practices and ways of doing things on both sides. The high turn over amongst NGO care teams, whose members rarely stay more than six to twelve months in the field, sometimes less, does not lend itself well to embedding shared practices and relationships of trust with local contacts. The difficulties in the relationship are often made worse by the luxurious resources – at least by national standards – that NGOs enjoy, and which can be a source of irritation, or even humiliation, for their national counterparts. These problems are structural and are part of the operation itself. Even if the excesses of certain participants could be eliminated (and further progress needs to be made in this area), the contrast in terms of resources would remain and expatriate humanitarians would still be people who were only there temporarily. Whilst minimising these difficulties is desirable, doing away with them altogether is impossible.

There are other difficulties as well, from the opposite point of view. For example, the treatment protocols for the main diseases, which are set by the Ministry of Health in the country concerned, can sometimes cause difficult ethical dilemmas. What should be done, for example, when the public authorities impose the use of an ineffective treatment, such as chloroquine for malaria? Such questions are sometimes a matter of life and death. They arise because pathogens inevitably build up resistance to current treatments, which therefore need to be updated; this in turn means introducing new compounds, which are expensive. NGOs generally have access to them, but are not authorised to depart from the national protocol. Solutions are generally found to problems of this kind, insofar as the health authorities agree to change their instructions. This takes time, however, and a doctor facing a serious outbreak of malaria does not operate on the same timescale as a discussion with a director of health. The specific issues and challenges faced by humanitarian medicine in displaced persons’ or refugee camps make it a very special field. Dealing with these problems and overcoming them means understanding them in the first place. This is why NGO’s have gradually developed analytical capacities that look beyond purely practical questions, using the services and work done by researchers in social and political sciences.

New difficulties and obstacles should not, however, mask the considerable progress that has been made in the aid sector. The loss of human life and the consequences in terms of disease of the war in Darfur, to take just one of the major crises of recent years, would have been much worse without the impressive machinery of international aid built up over the 1980s and 1990s, both by NGOs and the UN. It is also worth pointing out that it is the creation of the camps that allowed NGOs and the United Nations to organise aid for the victims of the war effectively. Thanks to the humanitarian system set up as the conflict progressed, these communities of displaced persons and refugees gave millions of people the opportunity to find temporary shelter, rebuild their strength and be protected from the ravages of epidemics and famine, which were the main causes of mortality in these conflicts.

Although precarious and sometimes stifling, the camps also provide new opportunities for socialisation, with better access to primary education for children or occupational apprenticeships, where distant social groups are able to mix more easily, and where often unexpected social practices develop. The cases of Rwanda and Cambodia have already been discussed to show that politics still holds sway in the camps, and in fact is extended and reconstituted. At the same time, new positions of authority are created, under the influence of the international aid system. Knowledge of a foreign language, for example, an essential skill for communicating with the aid system, can enable refugees or displaced persons to access posts outside any pre-existing hierarchy, which offer them both an income and access to aid resources. In an environment where resources are scarce and violence common, this ability represents on the one hand, a significant source of power and in some cases an invitation to abuse power, and on the other a reconfiguration of social relationships. The traditional authority of the group – the village, clan or region – is swept away by a new form of power as the result of the competition between different forms of legitimacy that undermines the order that had prevailed until that point. Children’s education, and girls’ education in particular, no matter how desirable it may be, is not the least profound of these transformations. The brutality of the social change brought on by war and rootlessness is accompanied by a disturbance in the social hierarchy and the allegiances that depend on it. This explains why it is illusory to seek to re-establish, at some point, the conditions that existed before the war, as humanitarian institutions such as the UN too frequently try to do.


The emergence of peace at the end of a conflict means dismantling some rural camps, generally on a gradual basis, the dispersion of their population signalling a return to a certain degree of security. But this is not the most common scenario: in addition to the fact that violence may continue even after the war is officially over, the camps are often located near towns or villages. Over time, “displaced persons” become urban migrants, the camps become neighbourhoods where children are born and grow up, and the habits and obligations of both day-to-day life and social and political structures change. Such changes, which are the structural consequences of war and the relief effort, are to a large extent irreversible. They do not affect all displaced populations in the same way, but make it impossible to envisage a complete return to the statu quo ante once the conflict is over.

As the years pass, villages therefore become small towns and small towns become larger ones. Once the war is over, the reasons why people were displaced and humanitarian organisations moved in disappear, but the reasons for keeping medical facilities in place remain, as the public authorities are generally not in a position to respond to healthcare needs.

At this point, medical NGOs face difficult choices, because of a number of different and competing conceptions of humanitarian aid and its limitations, possible demands from the health authorities and problems around resources and decision-making. It is not possible to provide an overall response to the questions raised by these transitional situations, except that the medical and healthcare systems set up during a conflict cannot remain in the new context. In any event, the material resources put in place, firstly designed to suit the gravity of the initial crisis and then the changing situation, are no longer appropriate to a situation that is closer to an ordinary health policy. Logistical resources, particularly in terms of transport and communications, but also staff, need to be cut back. The location of medical facilities, selected on the basis of ease of access and security issues, needs to be re-examined in the context of a map of healthcare facilities that is appropriate to new requirements and drawn up by the government authorities, all of which needs to be thought through on the basis that they will later reassume responsibility for them.

A humanitarian organisation that decides to keep up its medical activities at the end of the conflict that caused it to be there in the first place therefore needs to completely rethink its working methods. The new system then becomes part of a different humanitarian medical approach, one based on long-term structures, which is not designed to deal with crises or major emergencies, and which will be discussed in more detail below.


Earthquakes, hurricanes, floods and either seismic or climatic events are an important aspect of international aid and certainly one of the most spectacular. The scenes of devastation that result from them seize the imagination and provoke a unanimously emotional and compassionate response, much more than other violent events do. In spite of an abundance of specialist literature and widespread media coverage, however, the effects of natural disasters are not well understood, not only by the general public, which is not surprising, nor by the decision-makers concerned within governments and relief organisations, which is less easy to condone. It is difficult to find an area of public intervention where beliefs and interests conceal the reality to such an extent.

It is worth pointing out, first of all, that a natural disaster is only a disaster if it occurs in an inhabited area. Otherwise, it is a natural event with no human consequences, which is why what everyday language calls a natural disaster is properly speaking, a social and political as much as a natural event. Climatic catastrophes, which seem to have increased in frequency and intensity in recent years, are increasingly seen as a manifestation of global warming and therefore as a consequence of human activity. Is this the reality or does it express a sense of increased vulnerability? Are the data we have on past periods sufficient to conclude that global warming is the result of pollution? Historians are divided, but most climatologists would agree that they are.Cf L’Histoire no. 304, Dec. 2005.In any event, the major change is the fact that the regions affected are today not only much more densely populated than in the past but that the ways in which they are populated have changed. A earthquake hitting Kashmir in the 19th century would not have killed many people. In 2005, it killed tens of thousands. In this case, the increase in the population was only partially due to natural demographic growth. It was to a great extent the result of a deliberate strategy adopted by the Pakistani authorities in the context of their conflict with India.


Natural disasters are certainly playing an increasingly large role in international aid, which is why it is particularly useful to understand their practical consequences. The relief effort prompted by the tsunami that struck south-east Asia in December 2004 is a useful illustration of the common errors made in assessing the situation, which were magnified by the scale of this exceptional event.

The first error, which is systematically made in similar situations, was raising the alarm about epidemics. This immediately occupied a central place in the aid system. Senior figures from the WHO and the OCHA in fact predicted an imminent doubling of the number of deaths, because of there being corpses everywhere and the supposed source of infection these represented. As a consequence, it became a matter of urgency to bury the bodies in mass graves, set up a prevention and detection system for transmissible diseases and launch mass vaccination campaigns. Such efforts were, in fact, futile and a waste of resources, since epidemics have never been seen to break out in such circumstances, for reasons that have not been difficult to understand since Pasteur’s revolution.Nathalie Floret, Jean-François Viel, Frédéric Mauny, Bruno Hoen and Renaud Piarroux, “Negligible risks for infectious diseases after geophysical disasters”, Emerging Infectious diseases, www.cdc.gov/eid, vol.12, No. 4, April 2006.As Dr Claude de Ville de Goyet, a specialist in the epidemiology of disasters, wrote, “the bodies of victims of earthquakes or other natural disasters do not present a public health risk of cholera, typhoid or other plagues mentioned by misinformed medical doctors (…). In fact, the few occasional carriers of those communicable diseases who were the unfortunate victims of the disaster are a far lesser threat to the public than they were while alive.”Claude de Ville de Goyet, “Stop propagating disasters myths”, The Lancet, vol.356, August 26, 2000.This former head of disaster relief for the Pan American Health Organization adds that burying bodies hurriedly and anonymously is first of all an additional ordeal for the survivors, who are deprived of the possibility of honouring their dead, and very often a source of interminable legal and financial problems for their relatives, because of the lack of a death certificate. The same mistake was made when assessing the cyclone that devastated southern Burma in May 2008. Again, the risk of an epidemic was put forward by several Western NGOs and government ministries, but in a different political context. The initial refusal by the Burmese military junta to allow foreign relief to enter the country and its inertia in the face of the disaster were widely and justifiably criticised. The British Foreign Office and the French Ministry of Foreign Affairs then raised the necessity of a relief operation being imposed by force in order to contain the threat of cholera and other infectious diseases likely to cause as many victims as the cyclone itself. In addition to the risk of epidemics, it was said, there was the problem of hunger caused by the destruction of crops. Oxfam, for example, one of the largest international NGOs, with a solid reputation for the quality of its analyses, published a press release a few days after the disaster entitled: “Oxfam warns up to 1.5 million people in danger if aid efforts cannot reach cyclone victims. Risks of disease outbreak increasing daily as Myanmar stands on the edge of a public health catastrophe.”http://www.oxfam.org/fr/news/2008/pr080511_myanmar_cyclone_public_healthIn reality, although the number of deaths caused by the cyclone is unknown, all the observations made by members of NGOs and journalists in the following weeks came to the same conclusion: neighbourhood solidarity swung into action, there was no large-scale gathering of the population, and no epidemics broke out. The risk of a “second wave of deaths” used as a justification for imposing aid by force did not exist in Burma any more than it had in other countries that had been hit by a natural disaster in the past.

Emergency medical assistance for the wounded is also seen as a priority, even though the spectacle of disillusioned international surgical teams trying to make themselves useful is part of the ordinary landscape of the consequences of natural disasters. Waves of epidemics, masses of wounded people: it was as if a natural disaster were being confused with an armed conflict, which, in addition to causing death and injury, does indeed prompt the appearance of epidemic diseases. In reality, natural disasters and conflict differ in many ways from the point of view of their consequences. Conflicts cause three to five times as many injuries as deaths, whilst natural disasters generally cause more deaths than injuries, most of which tend to be minor and are relatively simple to treat. In addition, whilst wars last for years, with the consequences on the population and medical facilities already described, justifying the role of large numbers of expatriate medical teams, natural disasters affect a clearly demarcated area, for a very limited time. Outside of the region hit by the disaster, healthcare facilities and the rest of the social fabric remain intact, even within just a few minutes’ walking distance of the centre of the affected area.

In Sri Lanka, the tsunami swept along a strip of coastline between 100 and 300 metres wide, depending on the relief of the land. Beyond the area penetrated by the wave, the country operated as normal, which explains why a thousand local doctors and nurses could get to the area immediately to replace their lost colleagues and take over from hard-hit teams. Because they were not hampered by language problems or difficulties in adjusting to the environment, they were able to be operational immediately and treat the considerable influx of people coming into the hospitals. Whilst the number of injured people is smaller, and they are less seriously wounded than in a conflict situation, they are there nonetheless and the work done by the medical teams is highly valuable. They are always overwhelmed by demand in the immediate aftermath of the disaster and ad hoc external support may be necessary at this point.

It seems, however, that the medical problems caused by catastrophes are changing. The earthquake that struck Kashmir in November 2005, for example, killed 75,000 people and injured 40,000. In this case, a mass, long-term presence of foreign medical and surgical teams was essential in supplementing and broadening the care services available locally, which were submerged by the number of serious surgical cases. On a smaller scale, the earthquake that struck Java (Indonesia) in May 2006, led to similar consequences, as did the one in May 2008, in China. Several thousand injured people – and not several tens of thousands, as was announced in the immediate aftermath of the disaster – had to be operated on. It is worth noting that, in addition to the usual international participants, namely the industrialised countries of the North, other nations, in particular Singapore, Qatar and China immediately sent field hospitals to the area. The expansion and increasing number of dense urban areas with poor-quality buildings are undoubtedly the main reason why so many people were injured in Pakistan, Indonesia and China. In addition to these political and economic reasons, there was a lack of knowledge of past disasters amongst people who had only recently settled in the area, and had not therefore been able to organise their housing to take account of the risks.René Favier and Anne-Marie Granet-Abisset (dir.), Histoire et mémoire des risques naturels, éd. MSH-Alpes, Grenoble, 2000.On top of these new surgical needs are medical problems, which are themselves related to demographic and economic change. Industrial installations may be damaged, as was the case with the Shengdu earthquake in China (May 2008), and cause widespread poisoning that requires specific treatment. The risk of earthquakes certainly needs to be taken into account when selecting sites where hazardous substances are produced and handled. In a context of rapid and relatively unregulated industrialisation, however, the probability of industrial disasters in the wake of natural disasters, primarily earthquakes, is greater than it was in the past and is continuing to increase. The Chinese warning needs to be heeded by the various organisations operating in emergency situations as a significant precedent, and should prompt specific preparatory work, in particular producing an inventory of toxic substances, the possible ways they may be disseminated (air, water, food) and appropriate medical responses.

An analysis of comments from both journalists and relief workers highlights another element of the argument, as an additional but barely more valid justification that the previous ones, for the deployment of battalions of relief workers, namely that populations who had suffered such catastrophes would be plunged into a state of torpor that would leave them unable to take care of themselves. The fashion for the disputed notion of “post-traumatic stress disorder” has bolstered this belief to a significant extent and is again derived from representations of war.Richard Rechtman, “Du traumatisme à la victime, Une contruction psychiatrique de l’intolérable”, in Patrice Bourdelais et Didier Fassin (dir.), Les constructions de l’intolérable, Etudes d’anthropologie et d’histoire sur les frontières de l’espace moral, La Découverte, Paris, 2005.As with the points discussed above, the fact that reality has proved them to be false carries little weight with preconceived ideas. The depressing scenes broadcast relentlessly on television are certainly real but do not convey the situation as a whole, which is the opposite of staggering.

In all disasters, in fact, what is most striking is how local solidarity immediately comes to the fore: accommodation centres are set up, food is distributed, rubble cleared, those who have disappeared searched for, and so on. Some people are indeed in a state of shock, and some behave primitively or are indifferent to the situation, but these are isolated cases and most spontaneous reactions are to cooperate and provide mutual support. One survey carried out in Sri Lanka, for example, shows that all those caught up in the disaster who were questioned had been given a hot meal and shelter from the first night.Vladimir Najman, Économie de la survie des populations sri lankaises du district de Batticaloa après le tsunami de décembre 2004, < http://www.msf.fr/site/bibli.>It is not possible to draw general conclusions from such evidence, nor to decide as a result that external help is of no use, but the obstinate refusal to acknowledge the existence of mutual support at a local level and errors in assessing the consequences of natural disasters automatically lead to an overestimate of the necessity of urgent “turnkey” responses provided hastily and in inappropriate ways.

If there is one area, however, where the needs for immediate foreign help are generally pressing, it is logistics: re-establishing communications routes, setting up transport systems, evacuating isolated groups and clearing away rubble are all problems that local authorities, which are ill-equipped to deal with sudden events on a massive scale, cannot resolve on their own. In this case, military engineering resources – helicopters, civil engineering vehicles, heavy lorries and off-road vehicles amongst others – are all the more useful insofar as they can be put to use quickly and on a large scale. This is not, as we will see, a rescue operation but about reducing the chaos in material terms, a task for which armies are better equipped than any other organisation. The role of soldiers in this case is quite different to the one they play in peacekeeping operations, because on the one hand, it is not about using force and on the other, such operations are of necessity short-term. The presence of foreign soldiers does not therefore raise any difficulties in natural disaster situations, naturally subject to its being accepted by the authorities of the country, which was not the case with the Burmese cyclone referred to earlier.


It is worth recalling that a government’s offer to send assistance is always within a context of international relationships: “the diplomacy of disaster” is simply a circumstantial variant of ordinary diplomacy. When Iran was hit by an earthquake in June 1990, for example, the French government proposed to the national authorities that it should send in specialist teams, although the two countries had broken off diplomatic relations. In this case, emergency aid was used to give official recognition to the fact that discussions between Paris and Tehran has restarted, bringing into the open something that had until then been kept secret. More recently, the assistance offered by the President of Venezuela, Hugo Chavez, to the inhabitants of New Orleans who were victims of Hurricane Katrina in 2005 was based on the same logic, but the other way round. The proposal was seen by the US authorities as a public criticism of their failures and an act of hostility directed at the Bush administration, which was, in fact, the case. Between these two highly polarised situations there are more often intermediate positions. A few days after the 2005 tsunami, Condoleezza Rice, at a US Senate hearing prior to her appointment as Secretary of State, declared that “the tsunami was a wonderful opportunity to show not just the US government, but the heart of the American people,” adding: “And I think it has paid great dividends for us.” No doubt tactless rather than cynical, as she was criticised for being by members of the Senate, the declaration reflected above all a well-established political reality, which means that natural disasters provide useful opportunities for sending political messages. The dispatch of military ships to Burma following the 2008 cyclone, and the injunctions to open up the country that accompanied then, need to be understood in this context.


On a different level, NGOs and UN agencies also have their own “agendas”, in particular during situations where there is extensive media coverage or the pressure to act quickly is very strong. No doubt some of the errors described above are due to the pressure to intervene quickly brought about by the large number of alarming reports. Given that local relief teams can be up and running immediately, however, and can take care of most of the immediate needs, what is really urgent for foreign aid is not a rapid deployment but carrying out an assessment of what is lacking and responding in as well-coordinated a manner as possible. The priority for relief organisations therefore needs to be assessment, in spite of the pressure from public opinion and the media to be seen to be taking action. Tents, food, tankers and systems for providing clean drinking water, communications resources, heavy machinery for clearing rubble, medical equipment and drugs, surgical assistance and transport resources (helicopters and boats, if necessary) are the main components of emergency aid, in varying proportions depending on the situation. In this respect, the direct involvement of national governments via their armies’ logistical resources and civil engineering experts is essential.

The aid system needs to be adjusted to meet the specific needs of the situation, which vary from one context to another, but also the responses of the many other participants, although this can be a very difficult parameter to grasp. Only the authorities in the affected countries, supported where necessary by the United Nations, can coordinate the various forms of aid on the ground, but realism demands an acceptance of the fact that this cannot be done immediately.


The importance of epidemiology for assessing and monitoring medical assistance has already been discussed. Measuring the prevalence in a given population of malnutrition or malaria, the change in the number of cases of measles or meningitis following a vaccination campaign or the geographical location of digestive infections can be used to determine priorities and assess the impact of a healthcare strategy. With the increase in humanitarian action in refugees’ and displaced persons’ camps during the 1980s, the development of epidemiological surveillance systems, namely monitoring the main pathologies and mortality, gradually became an ordinary practice in medical and healthcare assistance.Francesco Checchi and Les Roberts, “Documenting Mortality in Crises: What Keeps US from Doing Better”, PLoS Med. 2008 July; 5(7): e146. This article includes an extensive bibliography on the subject.

Retrospective mortality surveys (RMS) are thus carried out on a regular basis for the purpose of assessing the level and the evolution of mortality. These are primarily used to determine whether an “emergency threshold” has been crossed and to improve the assistance provided by relief organisations by expanding or readjusting it as necessary. The emergency threshold is defined is twice the ordinary mortality rate in the region concerned. How it changes over time, which is measured weekly in acute periods, is as important to consider as its absolute value, which is generally imprecise, as we shall see.


Whilst RMS are most often carried out for practical purposes, they are also a communications tool, designed to attract international attention to a situation that has been neglected. As their know-how increased, NGOs started to produce figures whose function went beyond purely operational concerns, to get the problem concerned onto the political agenda. It is this issue, with its associated controversies and uncertainties, that we are going to address now. The example of Somalia is a useful illustration of the way the effects of such communications do not necessarily go the way the authors of the surveys were hoping for.

In March 1992, just over a year after war broke out in Somalia, teams from MSF and the ICRC observed a rapid rise in the number of people suffering from malnutrition. MSF then took the initiative of carrying out an RMS in order to establish a quantified baseline for assessing the situation, which until then had received very little press coverage. The survey showed that, within the population displaced by the war, a quarter of children under the age of five had died within the previous six months. A demographic “axe blow” of this kind confirmed the pressing need for a massive food aid operation, which both MSF and the ICRC had been calling for for several months and which the United Nations had put off for security reasons. The conclusions were presented at press conferences and parliamentary hearings, in particular before the European Commission, the European Parliament and the US Senate. The report was widely disseminated and its results picked up by the press, but did not prompt the response that had been hoped for, and things remained as they were.

Six months later, however, the Secretary-General of the United Nations, Boutros Boutros-Ghali, undertook to mobilise the Security Council to intervene militarily in Somalia. Meanwhile, over the summer, the international press had been reporting on what had become a devastating famine. Under media pressure, Boutros Boutros-Ghali tried to react by demonstrating the UN’s capacity to intervene, now that the impediments of the Cold War were no longer in play.For a detailed analysis of the UN operation in Somalia, see Roland Marchal, Les dégâts d’une improvisation, in Marie-Claude Smouts (dir.), L’ONU et la guerre, la diplomatie en kaki, Editions Complexe, 1994, Paris.For the Secretary-General of the United Nations, Somalia was to be a testing ground for armed interventions to impose and maintain peace in the new world order that had emerged since the collapse of the USSR. In order to persuade the members of the Security Council, he took the MSF figures – amongst others – and extrapolated them to the whole of the country’s population, whilst in fact, the abnormally high death rate only concerned those who had been displaced and the survey showed that the so-called resident populations, on the contrary, were only marginally affected. He explained the scale of the disaster as the result of looting and the misappropriation of almost all food supplies by the armed groups and “warlords”. This dramatisation of the situation, which was designed to increase pressure on the United States, so that it would send in troops to protect humanitarian convoys and civilian populations, achieved its goal with the launch of Operation “Restore Hope” in December 1992. The figures that the relief organisations had produced to speed up food supplies were in the end only used by those they were intended for to justify a military operation for a humanitarian purpose, the consequences of which were catastrophic.

Whether the distortion of the figures in this way was intentional or not, which is impossible to judge, does nothing to change the fact that the production of morbidity and mortality statistics by humanitarian organisations becomes a political exercise as soon as the results are released into the public domain. The question of the opportunistic use that certain players may make of them is not the main problem, if one accepts the principle of freedom to interpret facts. In the example above, therefore, Boutros Boutros-Ghali could be criticised for having manipulated the figures themselves, but not for having used them to justify his plan for military intervention in Somalia.

In practice, the real difficulty is determining the accuracy of the figures circulated, in addition to the fact that such figures should always be cited with a confidence interval, which in theory is an integral part of them. Experience shows that this is often far from the case: the number of victims is often influenced by political considerations or even reduced to the level of a propaganda instrument.

This was the case in Kosovo in 1999 where, according to the governments of those countries that were most supportive of an armed intervention, Germany and the United Kingdom, 100,000 Kosovars had been massacred by the Serb militias. Such a level of targeted violence justified the accusation of genocide brought by Germany, along with the war that was supposed to end it. Later research, carried out after the NATO operation and under the authority of the United Nations, was to reduce this figure to under 10,000. Exhumations of most of the mass graves and surveys with families suggest that this is a more reliable estimate.According to ICT surveys, the death toll amounts to 2108 in the ethnic Albanian population, in addition to 4206 missing.

In both cases, Somalia and Kosovo, exaggerating the figures is based on the same logic: dramatise to intervene. Not, undoubtedly, because the conflicts would in any sense be different in kind depending on whether there were more or fewer deaths. Whilst they may contrast with each other, the assessments referred to above change little on a human or political level, as regards the gravity of the conflicts they are trying to record. The hypothesis under consideration here is that these exaggerations need to be understood in the context of competition for access to the public arena, in which escalation allows the authors to achieve an effect of revelation and denunciation liable to place them in a better media position than their rivals. In other words, it is not the figures themselves but the battle to impose the highest ones that gives those with the higher version a moral advantage over their opponents, who are generally accused, in a context of public controversy, of trying to trivialise a tragedy.


The political intention underpinning the assessment work carries even more weight in how it is done insofar as the techniques available to do so are of questionable reliability. The epidemiologists who have developed these methods suggest a prudent approach to interpreting the results that is only rarely observed in militant speeches.F. Checchi, L. Roberts, “Interpreting and using mortality data in humanitarian emergencies”, ODI, London, Network Paper no. 52, 2005.

There are four factors to take into account in terms of uncertainty, each of varying levels of importance. Firstly, the difficulty of determining whether a given sample is representative of the population being studied, which raises the problem of extrapolating figures gathered in a particular place and at a particular time to a much broader population and a longer period. Secondly, the size of the population to which the survey applies; even in the case of refugees, a “captive” population recorded with the United Nations Agency for Refugees (UNHCR), differences in the figures can be significant depending on the source, i.e. the government of the host country, the United Nations or an NGO. Estimating the number of displaced persons is even more difficult, because they cannot be clearly defined. Thirdly, the mortality baseline, which allows to calculate the excess mortality rate due to the conflict itself ; a question that becomes even more tricky insofar as the size of the population concerned is often unclear, because of a lack of credible demographic data. Fourthly, and to a lesser extent, the impossibility of knowing whether someone being questioned is exaggerating the number of deaths in their immediate family, or not. This may be to their advantage for various reasons, such as discrediting the enemy or receiving more aid, but it is not within the epidemiological surveyor’s remit to cross-reference the information gathered. Conversely, families that have been completely destroyed may be ignored because there are no survivors to report the facts. It should be noted that the first three factors of uncertainty are limited where the assessment is of situations involving displaced persons or refugees but are significant when it relates to a region, namely a population whose contours are impossible to define.


Two conflicts, which are still underway in 2009, illustrate the issues and difficulties of such assessments. The first is the armed rebellion in Darfur, which began in March 2003 with rebel attacks on government military bases, to which the Sudanese regime responded with blind repression. The second is the invasion of the Democratic Republic of the Congo (DRC) by the armed forces of Rwanda and Uganda in 1998, prompting a war that lasted until 2003 and was followed by a civil war and a series of massacres in the provinces in the east of the country.

The first 15 to 18 months of the war in Darfur were characterised by air bombardments of populations who were supposed to be sympathetic to the guerrillas and by mass atrocities committed by pro-government militias. Over two million people fleeing the terror gathered in the main urban centres in the province and over 200,000 in neighbouring Chad. The international aid provided by the United Nations and NGOs was rolled out gradually, primarily in the displaced persons’ and refugee camps. In October 2004, based on several RMS conducted during the summer, the WHO stated that 10,000 people were dying in Darfur every month. Based on this, and extrapolating the data to the whole of the period, the United Nations declared in March 2005 that the conflict – which had been underway for 18 months – had caused 180,000 deaths. The story does not end there, as this same method of “calculation” was applied by NGOs advocating an armed intervention to protect the civilian population, such as Save Darfur (USA) and Urgence Darfour (France). According to them, in 2007, the number of victims was already in excess of 400,000 and was continuing to grow at a rate of 10,000 a month, an estimate that was largely taken up in the international press, although articles had appeared as early as 2005, questioning these assumptions as risky at the very least.

Two kinds of bias need to be taken into account here. On the one hand, whilst the average estimated mortality rate for the first seven months of the conflict was within a range of 35 to 70,000, only the higher value was actually used. On the other, the mortality rates for the period of mass killing were applied for the whole duration of the war.

Although extrapolations and interpretations of this kind lead to a significant overestimate of mortality, the results of the surveys themselves did, in fact, allow more rigorous assessments to be made. Several studies compared these RMS with morbidity and nutrition studies by including historical and geographical variations in their calculations.D. Guha-Sapir & O. Degomme, Darfur: Counting the Deaths. Mortality Estimates from Multiple Survey Data, May 26, 2005, CRED.These calculated a mortality rate of between 5 and 6,000 a month during the first 15 to 18 months of the war, a quarter as the result of direct violence and three quarters as a result of exhaustion and disease. This reassessment, which does not, it must be said, in any way diminish the seriousness of the situation nor the responsibility of the Sudanese government, has the merit of reporting reality as accurately as possible, which is what aid organisations need in order to refine their programmes. The essential point for understanding the dynamics of the conflict lies elsewhere, however, in the change in the nature of the war from the autumn of 2005 onwards. From this point on, average monthly mortality was not measured in thousands but in dozens, according to the United Nations and various NGOs on the ground. The period of mass killings had thus come to an end, if only to be replaced by skirmishes, local atrocities and widespread insecurity. It was this transformation in the regime of violence, expressed by a shift in orders of magnitude, that was the focus of the controversy between those who supported military intervention and those who opposed it.

The overdramatisation of the conflict, represented by the former as a long series of massacres of civilians perpetrated against a background of the “indifference of the international community”, masked the reality of a large-scale humanitarian operation, the largest in history on the basis of the number of people involved. It is worth noting, in passing, that no attempt has ever been made to calculate the number of lives saved thanks to international aid. It is fair to say that such a calculation would also be questionable, even if the almost total disappearance of malaria thanks to access to care and a significant improvement in the rates of infant malnutrition are important parameters, amongst others.

In the Democratic Republic of the Congo (DRC), a series of mortality surveys carried out by the American NGO International Rescue Committee since 2000 concluded that 5.4 million people had died from the consequences of the war in the country, making it the “worst humanitarian crisis since the Second World War.”Mortality in the Democratic Republic of Congo, An Ongoing Crisis, http://www.theirc.org/special-report/congo-forgotten-crisis.html.According to the IRC, the abnormally high mortality rate of 45,000 people per month was 99% due to infectious or parasitic diseases and malnutrition. These figures have a certain descriptive value, for the reasons already cited, and there is no disputing their interest. Nonetheless, these too need to be carefully handled: how many people who died of malaria or malnutrition during the course of a conflict would have lived if there had not been a war? That is the calculation that needs to be done to assess excess mortality due to violence. As far as the IRC survey is concerned, the basis used was a crude mortality rate of 0.5 deaths per 10,000 people per day, although it is generally assumed that the figure in Africa is around 1 per 10,000 people per day. Excess mortality would have been lower if it had been calculated on the basis of the latter figure.

In any event, in addition to the uncertainties of assessing mortality related to conflict there are questions around previous mortality levels, although considerable resources were brought to bear on the problem: for their final survey, the IRC teams visited 14,000 homes in 10 of the DRC’s 11 provinces. In spite of the scale of the research, the figures cannot be extrapolated to the general population without multiple precautions similar to those described earlier in the case of Darfur. A comparison of mortality and nutrition surveys carried out by different players again shows significant variations of up to ten times or even more, within the same region.RNIS Volume 32 & 33,<www.unsystem.org/SCN/Publications/RNIS/rnis3233.PDF>.


Being aware of distortions of this kind does not call into question the reality of a political and human disaster, which would be no less serious even if “only” half the victims it caused were identified. First and foremost, it is a question of getting as close as possible to the facts of the situation, in order to provide a fair description as a basis for making appropriate decisions. It is also about trying to get away from a tendency to escalation and ‘outbidding’ that the organisations producing the figures can trigger, without necessarily intending to do so. Aside from the fact that it cannot help but induce fatigue in the general public and the media, although its aim is, in fact, to attract their attention, this approach forces those who rely on it to up the stakes ever higher to raise the alarm, at the risk of trivialising situations that are less striking in numerical terms even more. The reservations expressed about these results should therefore not be seen as a way of discrediting any attempt at quantification. The aim is simply to call for a critical verification of figures that are too often seen as objectively true once they are in circulation. Verifications of this kind are all the more important insofar as they can also highlight an underestimate of mortality and, where necessary, the need to increase or redirect the aid effort.

Some authors have proposed the creation of an international organisation specialising in epidemiological assessment in major crisis situations.Paul B. Spiegel, Mani Sheik, Bradley A. Woodruff, Gilbert Burnham, “The Accuracy of Mortality Reporting in Displaced Persons Camps During the Post-emergency Phase”, Disasters, Volume 25, Issue 2.The indisputable importance of access to reliable figures is a solid justification for this proposal, which international organisations should consider. Provided it is and remains outside of aid operations and adheres strictly to its terms of reference, an institution of this kind could be very useful. It could, for example, be based around an international network of researchers developed for this purpose and funded by public-sector donors who make a commitment not to seek to influence the research. This would not replace the surveys carried out by aid organisations but could undoubtedly reduce their number, and above all would provide a more neutral point of reference against which to compare their methods and results.



During the 1970s, specialized United Nations agencies embarked on preventive vaccination programmes for infectious diseases. Starting in 1974, the Expanded Program on Immunisation (EPI), set up by the WHO and Unicef in developing countries, helped to extend vaccination coverage significantly year-on-year:The EPI covers the six main illnesses that can be prevented by vaccination: whooping cough, tuberculosis, tetanus, polio, measles and diphtheria.Unicef estimates that vaccination rates have increased from less than 10% in the 1970s to almost 75% today, saving the lives of over 20 million children in the last two decades. These figures are estimates and should always be treated with caution, but they do reflect undeniable progress. Overall trends and averages, however, conceal significant disparities, with the acceleration of vaccination coverage in the 1980s slowing down and even decreasing in the following decade. Some African countries, again according to United Nations figures, have seen rates drop to below 30%. The deepening of inequalities that has occurred during this period and the reduction in public services resulting from the structural adjustment programmes demanded by the IMF, as well as armed conflict, are the main reasons for the decline.

Alongside national governments and the United Nations, NGOs and private foundations have played a significant role, both in terms of running vaccination programmes and in technical advice, research and funding. This kind of cooperation is still ongoing, whilst the relative weight of private funding continues to increase because of the growing involvement of a small number of foundations with much bigger budgets than those of national governments, the Bill and Melinda Gates Foundation now being the largest. This is not a new phenomenon: the Rockefeller Foundation was a significant player in combating epidemics (tuberculosis in Europe, yellow fever in Latin America and cholera in India) from the 1920s onwards.


For all that, and although the involvement of these very wealthy institutions has enabled the launch of vital health research programmes and initiatives, the growing privatisation of public health around the world nonetheless raises some serious political problems. The facilities that are thus made available to national governments effectively make them dependent on external decisions, priorities that can vary in line with particular imperatives and opportunities that have nothing to do with their situation. In addition, the large number of players involved makes it difficult, if not impossible, to plan public health initiatives in the way that is necessary.

Alongside these reservations, it is worth noting that, for the Gates Foundation, which can mobilise a programme budget each year that is higher than the total budget of the WHO, the question of intellectual property, i.e. drugs, is expressed in terms that are difficult to reconcile with the demands of public health. The pharmaceutical and software industries, in fact, have a shared strategy for defending and extending the protection of the patents on their products but also, since the WTO agreements that came into effect in 1995, on their manufacturing processes. In their view, the very high cost of drugs simply reflects their development costs and essential investment in research. Any waiving of the intellectual property rights that protect new drugs for 20 years, or in other words, authorising copyright-free versions now would put therapeutic innovation at risk in the future. This is why the firms make donations of branded drugs, often in large quantities, whilst making significant efforts to stem the production of generic versions of recent drugs developing in emerging countries such as Brazil and India, which are both particularly active in this area.


Several epidemic diseases are affected by the debate on access to high-quality drugs at affordable prices. AIDS is the most emblematic, as the illness that epitomises the main issues at stake. The appearance in 1996 of antiretroviral combination therapies accelerated and broadened awareness of the subject, as a result of the change in prognosis for the illness thanks to new treatments. Not only did these significantly improve patients’ life expectancy, they also allowed them to resume a reasonable life, as those who received treatment quickly recovered their ability to work and enjoy a social life. As these drugs were protected by patents, however, they were inaccessible to people in poor countries even though that is where almost all new cases were found. Alongside the argument about protecting their intellectual property, the spokespersons for the pharmaceutical companies added the risk of the development of resistance to antiretrovirals (ARV) in order to oppose their being more widely distributed: patients who were for the most part illiterate, they said, would not be compliant, i.e. capable of correctly following a treatment regime that required drugs to taken daily at set times. Without compliance, resistance would spread, compromising the efficacy of ARVs in the near future.

Given the scale of the problem, with AIDS aggravating the economic and social problems of poor countries, various initiatives were nonetheless taken to ensure that these patients had access to treatment. In 1997, the UNMore specifically UN-AIDS, the programme set up in 1995, designed to coordinate the actions of its specialist agencies.set up the Drug Access Initiative in order to evaluate accessibility to combination therapies in three countries (Uganda, Ivory Coast and Chile); in the same year, France created the International Therapeutic Solidarity Fund to increase treatment rates; NGOs in the northern hemisphere, in particular Act-Up, Oxfam and Médecins Sans Frontières, and others in the south, notably the Treatment Action Campaign (TAC) in South Africa, embarked on an awareness-raising campaign designed to speed up access to these drugs; a “Global Fund” was created in 2001 by the United Nations to finance efforts to combat AIDS, tuberculosis (the new epidemic of which is linked to AIDS) and malaria, the three main “killer” epidemics in developing countries; and a funding programme (Pepfar, Presidential Emergency Fund For Aids Relief) was agreed by the United States in 2003.For a detailed analysis of the political stakes of the battle against AIDS, see Marc Dix-Neuf, “Au-delà de la santé publique: Les médicaments génériques entre perturbation et contrôle de la politique mondiale”, Revue française de science politique, vol.53, no. 2, Paris, April 2003.


This non-exhaustive list is clear evidence of the emergence onto the international stage of a public health question, which from then on would be seen as likely to produce serious imbalances that could threaten the world economy and international peace. Such, at least, was the main thrust of the CIA report published in 2000, on the distribution of infectious diseases around the world. A year later, an important legal battle was won in South Africa by a coalition of NGOs led by TAC and strongly supported by the press: 39 pharmaceuticals companies that had filed a complaint against the South African government to repeal the law allowing imports of generic anti-AIDS drugs from India had been forced to drop their case in the face of widespread criticism. Shortly afterwards, in November 2001, the WTO talks held in Doha, Qatar, recognised the primacy of public health imperatives over commercial interests. It must be said that the United States, despite having supported American firms during the Pretoria trial, had recently applied this principle unilaterally because of a public health emergency. Faced with a terrorist threat to trigger an anthrax epidemic, it had forced the firm Bayer, the only producer of an effective treatment against the fatal disease, to supply it with several million doses at the price of a generic product.

It seems likely that, in a context in which epidemics are seen as a global security issue, the combination of these two events had an influence on the WTO declaration recognising the principle of exception on health grounds. It should be noted that in practice, some pharmaceutical companies are successful in getting round the Doha agreements or at least contravene the spirit of them, using a strategy of bureaucratic and legal wrangling that makes them evermore complex to implement.


In spite of these difficulties, it will only have taken a few years for the number of patients receiving treatment for AIDS to have increased tenfold. In the meantime, the annual cost of ARV treatment fell from 10,000 dollars to less than 100, in the generic form reserved for poor countries. According to the Global Fund, two million people have been saved from premature death since the beginning of the decade, a number that increases by 100,000 every month. The time when using condoms was seen as the primary means of combating the disease in the developing world is long gone. Campaigns to promote safe sex are certainly still on the agenda, but getting patients into treatment without discriminating on the basis of their financial situation, because it is free, is now the priority.

For the vast majority of those who benefit from it, it means a return to an almost normal life, which is sufficient justification in itself for the effort it has taken. It seems as though there is another benefit to treatment, however: by reducing the level of the virus in the body until it is almost impossible to detect, the risk of contamination is thought to be reduced to a very significant extent. In other words, treating one group of people prevents others from becoming infected, which is a further reason to continue down this path. The dire predictions of the pharmaceutical industry that patients would fail to comply adequately with treatment regimes and the resulting spread of resistance have not proved correct.

Public international aid in its various forms funding, technical advice and research has been effective because it has supported the mobilisation of efforts in the countries that have been hardest hit, in particular in Africa, which is home to two-thirds of those affected and where resources are the most limited. But the race between the number of new cases and the number of patients undergoing treatment is far from being won: the latter represent at best 25% of the population that needs it. NGOs have also played a role in changing the situation, initially by campaigning for access to generic ARVs but also later, by taking on part of the burden of providing treatment and training, alongside the health ministries. AIDS was an overwhelming burden for healthcare facilities that were already run down in the early 1990s. Whilst it continues to take its toll, it has also driven change, a paradox that can be explained by the mobilisation of efforts around the problem. Humanitarian organisations will need to continue to support the system for many years to come, insofar as the weakness of African medical facilities makes it impossible to imagine that governments will be in a position to respond to such needs for a long time yet.

It is worth noting that the majority of generic ARVs are produced in India, which has witnessed the development of a pharmaceutical industry that is now capable of developing and manufacturing original therapeutic compounds, rather than just high-quality copies. This scientific and industrial success story, which is currently of little benefit to the majority of Indians, does not necessarily bode well for Africa, insofar as the new Indian manufacturers currently say they are in favour of a repeal of the laws limiting patentability in their country.

The momentum maintained by the NGOs, based on a sense of recently developed legitimacy in this area, needs to be kept up – as it is being – because pressure from them is still a necessity. Their action on the ground no doubt represents only a small proportion of patient care but it is a valuable proportion for those who benefit from it. Based on this experience, it is easier for them than for more cumbersome structures to implement new developments, with the approval of the national health authorities, in terms of treatment organisation, therapeutic protocols and clinical monitoring. They can also continue with the vital campaign of expanding the production of generics. Paediatric versions of ARVs, “second line” drugs that are effective against the resistant forms that are developing inexorably and faster diagnostic tests are some of the most urgent priorities in relation to AIDS.

The exceptionally rapid progress made in this area recalls the combination of a scientific advance (pasteurisation), philanthropic motivation (socially-oriented doctors) and political risk (the fear of a fall in the birth-rate) that prompted the child protection systems set up at the end of the 19th century. A similar pattern, with the addition of the actions of patients’ associations, emerged at the end of the 20th. AIDS is undoubtedly an emblematic illness insofar as it reveals not only the shortcomings but also the potential responsiveness of healthcare systems. Many that are active today were practically moribund a few years ago. For all that, the various initiatives prompted by the epidemic are not a model that could be easily transferred to other illnesses, unless they are seen as a serious threat, and therefore become a political issue.


The fear of illness, particularly epidemic illnesses, and as a consequence the general need for security felt by the populations facing such a risk, nonetheless represent a lever for action for NGOs campaigning for access to drugs. Resistant forms of tuberculosis are developing, defying the diagnostic therapeutic resources available; new viral diseases (including SARS and chikungunya), outbreaks of sleeping sickness and leishmaniasis, and epidemic forms of malaria have raised awareness of the risks affecting hundreds of millions of people in the world. Improving living conditions is certainly one of the main prerequisites for reducing the incidence of such illnesses and improving health security. It is also clear that drugs are only useful to the extent that there are good-quality health facilities, competent personnel and efficient administration available. The responsibility of governments and public development aid in this area is central, as NGOs can only operate at the margins. But it is no less true that the absence or shortage of drugs condemns healthcare systems to passivity and that their presence is a dynamic factor, as the progress made on AIDS shows. In this case, the persistent pressure of the NGO campaigns plays a more important role: only they can carry out awareness-raising campaigns to encourage pharmaceutical companies that are concerned about their public image to make their products accessible to the poorest populations.

Criticism of the abusive commercial practices of the pharmaceutical companies is the most visible aspect of the action of NGOs in this area, but it is not the only one. Aside from the confrontations, joint projects bringing together industrialists, research institutes, foundations and NGOs have emerged in recent years.See in particular Drugs for Neglected Diseases Initiative, <www.dndi.org>

Potential drugs are remaining at the stage of untested compounds, because the pharmaceutical companies or universities that have developed them have not wanted or been able to take them further because of the lack of a “viable market”. In some cases, drugs are available, but the forms in which they are administered are inappropriate to use in remote rural areas. Several groups, organised by NGOs or foundations, have been working for ten years or so to tackle these shortcomings, with promising results. New drugs for sleeping sickness and leishmaniasis, for example, are currently under development and a new combination therapy for malaria was launched on the African market in 2007. All of these are the result of cooperation between public research centres, health ministries, businesses and NGOs.

These new partnerships, which provide an opportunity to remove the production of certain drugs from the “laws of the market” are an important step in the history of humanitarian medicine. Not-for-profit privatisation of a public health sector is the consequence of the shortcomings of the countries concerned. It underlines the increasing role of those involved in humanitarian medical practice in relation to health security, in practice making their actions closer to that of national governments.