Chapter 3: The Nutrition of Refugees and Displaced Populations

Asia
Sub-Saharan Africa
Micronutrient malnutrition
Summary statistics
Conclusions
References

INTRODUCTION

This chapter is a summary of the ACC/SCN's Reports on the Nutrition Situation of Refugees and Displaced Populations which are published quarterly. Using information available from January 1995 through mid-1997, it is intended to update information that was presented in the Update on the Nutrition Situation, 1994 (ACC/SCN, 1994). Much of the background and explanatory information that was presented in the original chapter is not repeated here.

Assistance by the Office of the United Nations High Commissioner for Refugees (UNHCR) is provided to over 26 million people worldwide - either refugees, internally displaced people (IDPs) or returnees - representing one out of every 220 people in the world (UNHCR, 1996). The total number of refugees alone worldwide has continued to decrease, albeit marginally, due mainly to a reduction in the number of Afghan refugees in Pakistan, and to repatriation movements in the former Yugoslavia, the Benin/Ghana/Togo region, Kenya, and Mozambique region. Figure 8 illustrates the trends in numbers of refugees (both assisted and unassisted) over the last 25 years, based on data provided by UNHCR.

Figure 8.: Trend in the Global Number of Refugees 1960-1995

Figures on internally displaced people are far more difficult to obtain, but a variety of sources estimate that there are at least 30 million internally displaced people worldwide (UNHCR, 1996). In the situations involving IDPs covered by the Reports on the Nutrition Situation of Refugees and Displaced Populations, total numbers affected appear roughly stable during the time period covered in this chapter.

However, based on information available to the ACC/ SCN, it is apparent that although the total number of refugees and displaced people in Sub-Saharan Africa is decreasing, the number estimated to be at heightened nutritional risk, defined as high levels of wasting and/or mortality due often to factors like inaccessibility, has remained roughly constant. Furthermore, the sociopolitical situation in many countries in mid-1997, notably the Great Lakes region and Somalia, is very tense and could easily deteriorate, leading to population displacement, which in many cases may involve the second or third such displacement in as many years for a very large number of already destitute people. Trends in population numbers and associated nutritional risks in Sub-Saharan Africa are shown in Figure 9.

Figure 9.: Number (in millions) of Refugees and IDPs in Sub-Saharan Africa and Nutritional Risk over Time

Box 4: Indicators WASTING is defined as less than -2SDs, or sometimes 807. wt/ht by NCHS standards, usually in children of 6-59 months. For guidance in interpretation, prevalences of around 5-10% are usual in African populations in non-drought periods. We have taken more than 20% prevalence of wasting as undoubtedly high and indicating a serious situation; more than 40% is a severe crisis. SEVERE WASTING can be defined as below -3SDs (or about 70%). Any significant prevalence of severe wasting is unusual and indicates heightened risk. (When "wasting" and "severe wasting" are reported in the text, wasting includes severe - e.g. total percent less than -2SDs, not percent between -2SDs and -3SDs.) Data from 1993/4 show that the most efficient predictor of elevated mortality is a cut-off of 15% wasting (ACC/SCN, 1994, p81). Equivalent cut-offs to -2SDs and -3SDs of wt/ht for arm circumference are about 12.0 to 12.5 cms. and 11.0 to 11.5 cms, depending on age. BMI (wt/ht2) is a measure of energy deficiency in adults. We have taken BMI<18.5 as an indication of mild energy deficiency, and BMI<16 as an indication of severe energy deficiency (WHO, 1995). OEDEMA is the key clinical sign of kwashiorkor, a severe form of protein-energy malnutrition, carrying a very high mortality risk in young children. I+ should be diagnosed as pitting oedema, usually on the upper surface of the foot. Where oedema is noted in the text. it means kwashiorkor. Any prevalence detected is cause for concern. A CRUDE MORTALITY RATE in a normal population in a developed or developing country is around 10/1.000/year which is equivalent to 0.27/10,000/day (or 8/10,000/month). Mortality rates are given here as "times normal", i.e. as multiples of 0.27/10,000/day. [CDC has proposed that above 1/10,000/day is a very serious situation and above 2/10,000/day is an emergency out of control.] Under-five mortality rates (U5MR) are increasingly reported. The average U5MR for Sub-Saharan Africa in 1995 was 175/1,000 live births, equivalent to 1.4/10,000 children/day, and for South Asia the U5MR was 0.7/10,000/day (see UNICEF, 1997, p. 98). FOOD DISTRIBUTED is usually estimated as dietary energy made available, as an average figure in kcals/person/day. This divides the total food energy distributed by population irrespective of age/gender (kcals being derived from known composition of foods); note that this population estimate is often very uncertain. The adequacy of this average figure can be roughly assessed by comparison with the calculated average requirement for the population (although this ignores maldistribution), itself determined by four parameters: demographic composition, activity level to be supported, body weights of the population, and environmental temperature; an allowance for regaining body weight lost by prior malnutrition is sometimes included. Formulae and software given by James and Schofield (1990) allow calculation by these parameters, and results (Schofield and Mason, 1994) provide some guidance for interpreting adequacy of rations reported here. For a healthy population with a demographic composition typical of Africa, under normal nutritional conditions and an environmental temperature of 20° C, the average requirement is estimated at 1,950-2,210 kcals/person/day for light activity (1.55 BMR). Raised mortality is observed to be associated with kcal availability of less than 1500 kcals/person/day (ACC/SCN, 1994, p. 81). INDICATORS AND CUT-OFFS INDICATING SERIOUS PROBLEMS are levels of wasting above 20%, crude mortality rates in excess of 1/10,000/day (about four times normal - especially if still rising), and/or significant levels of micronutrient deficiency disease. Food rations significantly less than the average requirements, as described above, for a population wholly dependent on food aid would also indicate an emergency. REFERENCES: James W.P.T. and Schofield C. (1990) Human Energy Requirements. FAO/OUP. Schofield C. and Mason J. (1994) Evaluating Energy Adequacy of Rations Provided to Refugees and Displaced Persons. Paper prepared for Workshop on the Improvement of the Nutrition of Refugees and Displaced People in Africa, Machakos, Kenya, 5-7 December 1994. ACC/SCN, Geneva. ACC/SCN (1994) Update on the Nutrition Situation. 1994(p81). UNICEF (1997) State of the World's Children p.98. UNICEF, New York. WHO (1995) Bulletin of the World Health Organization, 1995, 73 (5): 673-680.

There are success stories to be noted, for example the successful completion of repatriation of Mozambican refugees, the gradual improvement of access to populations in need in Angola leading to a dramatic improvement in nutritional status, and the maintenance of levels of wasting less than those seen in the host population among Bhutanese refugees in Nepal. Another important example is the situation in the Great Lakes region, where humanitarian aid was delivered to over two million refugees and IDPs. More specifically, elevated levels of wasting and high crude mortality rates in the Goma camps in the former Zaire were rapidly brought under control and remained below rates seen in the local population throughout 1996. Efforts are underway to better understand how beneficiaries obtain food from all available sources, e.g. food aid, markets, trading, gardens, in order to better target the food aid available for maximum benefit.

The situations described in this chapter need to be viewed against a backdrop of decreasing global availability of food aid. While response to World Food Programme (WFP) appeals for food aid for emergency situations continues to be generous, there is a continuing downward trend in the resourcing for development projects. This could have longer-term implications for the ability of populations to avert or withstand future emergency events.

This chapter is divided into four main sections. First there are summaries of refugee or IDP situations which include information on some situations in Asia and in Sub-Saharan Africa. Within the first section, each country summary begins with background information and the numbers affected so as to give context. Details on nutrition and food security are then presented, often in chronological order, followed by concluding paragraphs). In the case of regional situations, the country summaries are preceded by a regional overview, and overall conclusions are drawn at the end of the country summaries for that region. Micronutrient malnutrition is discussed in the second section, and the third section offers some summary statistics. The concluding section aims to draw out trends based on the information presented and to highlight issues needing further consideration.

Asia

The estimated number of refugees in Asia decreased by 22% from the end of 1993 (5.8 million) to the end of 1995 (4.5 million). Numbers of internally displaced people are difficult to determine with any accuracy, but are at least 1.7 million. For the three situations included here (Afghanistan, Bhutanese refugees in Nepal, and refugees from Rakhine State, Myanmar in Bangladesh), the population in need of emergency programme support has decreased due to the attainment of self-sufficiency of many Afghan refugees in Pakistan, and the on-going repatriation of refugees from Bangladesh to Rakhine State, Myanmar. To give the context with which to interpret some of the nutritional data below, WHO estimates that regionally in South Asia, 17.3% of children under five years old (non-refugees) are wasted (weight/height <-2SD, see Box 4 on page 54).

Afghanistan Region

It has been almost 18 years since civil war broke out in Afghanistan, leading to the displacement of at least five million people. A great many are still displaced within Afghanistan, and over 2.5 million people are living as refugees in Iran and Pakistan. Many of these refugees have attained self-sufficiency in their host countries and do not require humanitarian assistance.

Afghanistan Fighting continued sporadically during 1994 in many parts of the country, although a brief lull in the conflict led to the return of large numbers of displaced people to Kabul in 1995. However, fighting resumed as the Taliban, an Islamic fundamentalist student group, waged an offensive against the government. This led to further population displacements from the capital, with many settling in the IDP camps in Jalalabad. The Taliban eventually took control of most of Afghanistan, including the capital, Kabul. Food supply lines to Kabul were periodically blocked, with prices for most food commodities increasing dramatically. As a result, the population of Kabul faced severe economic hardship, and many people were forced to sell off key assets in order to survive. Once again, the population of Kabul was forced into destitution merely to buy their daily food.

The attainment and maintenance of an adequate nutritional status in Kabul, particularly over the winter months (November to March), has been made possible in large part to the efforts of the international humanitarian community in providing aid, such as fuel, and wheat for bakery subsidy programmes. Nutritional data from the end of 1994 showed 28-40% wasting (using a QUAC stick which measures upper arm circumference for height). Although not directly comparable, survey results from November 1995 indicated a marked improvement in the situation, with only 6.2% malnutrition. A survey in May 1996 confirmed the improvement, finding 6.7% levels of wasting. No cases of oedema were noted.

There was some fluctuation in the nutritional situation in the camps for the displaced outside of Kabul. This may have been due to persistent problems in supplying sufficient quantities of clean water to the populations. For example, levels of wasting in New Hadda camp near Jalalabad were measured at 6.2% in December 1994. At that time, measles immunization coverage was only 34.7%. In March 1995, wasting was measured at 6.7%, and in September 1995 levels of wasting had risen to 11%. By December 1995, wasting was measured at 4.0% and the measles immunization coverage had risen to 79%, which was higher than coverage for the surrounding local population.

A massive immunization campaign was carried out in Afghanistan in June-July 1996. Over five million children were immunised against polio, measles and DTP. At the same time vitamin A capsules were distributed. In addition, over two million women of child-bearing age were immunized against tetanus.

Areas under Taliban control have been under strict Islamic laws, many of which prevent women from working outside of the home, driving or attending school. This presented serious ethical and practical dilemmas for humanitarian agencies working in Afghanistan. For example, in many situations both international and local female agency staff were not allowed to work. In other cases, income-generating projects for women were stopped, and girls were not allowed to attend school. It was difficult for international agencies to work under these conditions, with their aid benefiting only a portion of the population, and certainly not reaching the most vulnerable, who were made more so by these legislative changes. A few of these restrictions have been subsequently relaxed, and in some cases women are being allowed to work.

Pakistan Despite the unstable security situation in Afghanistan, many refugees in Pakistan had repatriated by the end of 1995, leaving 860,000 refugees out of an initial population of 1.2 million. Many of those remaining were believed to be self-sufficient, and assistance is now only provided for targeted vulnerable groups such as children, the elderly, the disabled, families without an able-bodied adult male and new arrivals. Nutritional surveys in October 1995 showed a generally satisfactory nutritional situation for these refugees, with levels of wasting measured at 0.9-6.0% (below levels found in the general population).

Iran Most of the 1.4 million Afghan refugees in Iran are living and working among the local population. There are no current nutritional data on these refugees, but there is some concern that, as the general economic situation in Iran worsens, it could become more difficult for them to find work so that they may eventually become nutritionally vulnerable.

A major issue arising from experiences over the period under review involves ethical questions concerning working with a discriminatory government. Once the Taliban took control of most of Afghanistan, agencies were forced to deal with a government which discriminates openly against women. It has been difficult for agencies to balance the principles of neutrality and impartiality with equity. This was clearly a problem since the most vulnerable groups - those most in need of aid, in this case women - had restricted access to services. In many cases, aid programmes have continued and negotiations to improve the access of women to available food and health service programmes are on-going.

Rakhine State, Myanmar in Bangladesh

Between December 1991 and March 1992, approximately 250,000 people fled from Rakhine State (then called Arakan state) in Myanmar (then Burma) to Bangladesh. Repatriation began in 1994 and was planned to be completed by the end of 1995. The repatriation process proceeded more slowly than planned, and currently there are 21,000 refugees remaining in two camps in Bangladesh.

At the end of the 1995, there were about 50,000 refugees remained in Bangladesh. The nutritional status of this population had remained satisfactory. A survey in August 1995 showed levels of wasting of 9%. The general ration, which contained fortified blended food, has consistently provided over 2000 kcals/person/day. Despite this, low levels of vitamin B2 deficiency (seen as angular stomatitis) continued to be seen. Investigations into possible reasons for this are continuing.

By the end of 1996, the nutritional status of the smaller number of refugees remaining in Bangladesh showed a deterioration as compared with 1995 information. Wasting was over 15%. It was felt that this deterioration could possibly be attributed to that fact that 'healthier' refugees had already repatriated. Supplementary feeding programmes were initiated to address this problem.

Since the beginning of the repatriation efforts, questions have been raised about the voluntary nature of the process. A report in mid-1996 stated that:

'The repatriation of Rohingya refugees from Bangladesh cannot be considered fully voluntary...

Most Rohingya refugees in Bangladesh believe they have no choice but to return to Burma (Myanmar). Although refugees who may fear repatriation have ample opportunity to inform UNHCR that they do not wish to return to Burma (Myanmar), most do not see such opposition to repatriation as a viable alternative.' (USCR, 1997, p. 127).

Bhutanese Refugees in Nepal

There are estimated to be 92,000 Bhutanese refugees in Nepal. Most of these refugees arrived in the early 1990s, allegedly fleeing persecution in Bhutan. There are currently no plans for the repatriation of these refugees. Levels of wasting in the camps are low; a survey in July 1995 showed 5.7% wasting, and in October 1996, 2.4% (see Figure 10). These prevalences are below those found in the host community.

Cases of micronutrient malnutrition (scurvy, pellagra, beri-beri) were noted in both surveys, and although incidence rates were tow, new cases continued to be reported. Some possible explanations for persistently low levels of micronutrient deficiencies, despite the inclusion of fresh vegetables and a fortified blended food in the general ration, include: mis-diagnosis (especially of scurvy and beri-beri as the symptoms are somewhat vague and can easily be confused with other medical conditions), the sale or exchange of key food commodities containing micronutrients, and intra-household food distribution and consumption patterns which restrict access of some household members to essential micronutrient containing foods.

Figure 10.: Nutrition Survey Results for Bhutanese Refugees in Nepal over Time

Taken from: SCF(UK) (1997) Household Food Assessment of Khudunabari and Beldangi Refugee Camps, Jhapa District, South-East Nepal SCF(UK). London.

Sub-Saharan Africa

Introduction

In the middle of 1997, there were at least 11.8 million refugees and internally displaced people in Sub-Saharan Africa. Most of this population have been affected by nine national and regional emergency situations, the two largest being the Burundi/Rwanda (Great Lakes) region and Liberia/Sierra Leone, which together account for over half of the total refugees and internally displaced people. Descriptions of the nutrition situation for the refugees affected in the nine largest emergency situations are given in the sections below.

Approximately two million of these refugees and displaced people were considered to be at heightened risk of mortality in June 1997, on the basis of data showing elevated levels of mortality and/or wasting, lack of access to populations identified as being in need of aid, or the presence of micronutrient deficiency diseases. To give context, levels of wasting between 6.0-9.5% are estimated by WHO to be usual in Sub-Saharan Africa. The number of refugees and displaced people can vary quite dramatically during an emergency. One way to assess the magnitude of an emergency situation is to calculate the number of person-years over which a population have been affected, as in Table 20. The number of person-years at high risk are also calculated.

Table 20.: Refugee and Displaced Populations in Sub-Saharan Africa

Country	Number of Person-Years 1994-Jun 97	Number of Person-Years at high Risk 1994-Jun 97
Sudan	2,513,492	133,542
Rwanda/Burundi	3,403,048	1,288,479
Angola	1,903,792	143,779
Liberia/Sierra Leone	2,965,183	755,179
Mozambique Region	838,283	28,333
Somalia	614,500	346,458
Shaba, Zaire	565,000	188,833
Ethiopia	372,644	143,188
Kenya	207,788	54,738
TOTAL	13,383,730	3,082,529

Source: RNIS Reports #3-20

However, relief provision in these circumstances, i.e. Liberia, Sudan and Burundi, is often on an ad hoc basis as security permits. These situations have necessitated an adaptation of the usual strategies on the part of humanitarian agencies to a form of "hit and run" procedure where rapid assessments are immediately followed by food aid distributions which may provide as little as one month's supply of food.

Frequently, the end of a war does not guarantee that people can easily return home and re-build their lives. In Angola and Mozambique, the indiscriminate use of landmines during the conflict meant that much arable land could not be used until the time-consuming and costly process of mine clearance had been completed. Many people, therefore, remain dependent on food aid long after a conflict because it is too dangerous to farm in former conflict zones. In addition, the infrastructure of a country is often seriously damaged or destroyed, necessitating a phase of reconstruction, particularly of road and health systems.

In some circumstances refugees and IDPs may be afraid to return home out of fear of persecution. This was one of the stated reasons given as to why many Rwandan refugees remained in the former Zaire and Tanzania. However, an eruption of civil war in Eastern Zaire (now the Democratic Republic of Congo) during 1996 and an ultimatum by the Tanzanian government eventually led to a massive repatriation of Rwandan refugees. This experience raises issues around the concept and perception of 'voluntary' repatriation.

Angola

Following independence in 1975, a civil war gripped the country and continued for almost 20 years. A series of agreements concluded over those years, which were designed to end the conflict, finally resulted in national elections being held in 1992. However, the results of these elections were not accepted by UNITA rebels, and renewed fighting followed. The peace accord that was signed in November 1994 finally ended almost two decades of civil war. At the time of the signing of the peace accord, at least 3.2 million people were in need of emergency food aid in Angola, approximately 32% of Angola's population. In addition, it was estimated that 300,000 people were living as refugees, mainly in what was then Zaire and Zambia.

At the beginning of 1995, there were some 3.2 million people requiring humanitarian assistance in Angola. By the end of the year the number had decreased to 1.4 million, due to reasonable harvests as a result of provision of agricultural inputs and better access to land, and to improved food aid deliveries. However, only a small number of Angolan refugees have repatriated from the Democratic Republic of Congo and Zambia.

As peace was established, many areas became accessible to humanitarian agencies. In the course of 1995, enhanced security and the clearance of land-mines meant that more humanitarian aid could be transported by roads rather than by more expensive air transport, which was the main mode of transport at the height of the conflict. The experience and pattern of humanitarian aid activities were broadly similar throughout the country. Very high levels of wasting were identified in newly accessible areas and, if security permitted, the regular provision of emergency food aid and essential medical care almost always brought the situation rapidly under control.

Map 1.: Angola

For example, a survey in Cafunfu, Lunda Norte Province, in May of 1995 showed 29% wasting and/or oedema, with 20% severe wasting and/or oedema. Mortality rates were 20 times normal. A general ration programme began in mid-May, and a follow-up survey in July 1995 showed 2.8% wasting and/or oedema with 0.9% severe wasting and/or oedema. This marked improvement was believed to reflect a number of factors including: food aid provision, previous mortality of children under five and renewal of economic activities amongst the affected population.

In Mexico province, wasting levels were over 20% in April 1995. By the end of the year, food delivery capacity had improved, due largely to improved access. Nutrition surveys at the end of 1995 showed 6.0% wasting and/or oedema.

In many areas, such as Zaire Province and Menongue, a continued lack of basic health services was identified as a major factor contributing to persistently high levels of wasting. While substantive health interventions are now taking place in many areas (e.g., immunization campaigns in Malange), the urgent need to improve immunization coverage and general health service provision is still recognized in many parts of the country.

One legacy of the conflict has been the extraordinarily large number of land-mines throughout the country. This served to restrict movement of humanitarian aid and prevent the resumption of normal life as many farmers are unable to farm their fields, thereby prolonging the dependency on humanitarian aid. De-mining in Angola is a major undertaking, that is likely to continue for some time.

Although considerable reconstruction and development of infrastructure in Angola is taking place, especially with respect to land-mine clearance and rebuilding the health system, in mid-1997 there were signs of a deterioration in the security situation in some regions of the country and efforts are now being made to contain the situation.

Burundi/Rwanda (Great Lakes) Region

The death of the Rwandan president in April 1994 signalled the start of a genocide that left at least 500,000 dead and led to the displacement of up to five million Rwandans, either internally or as refugees to neighbouring Burundi, the former Zaire and Tanzania. The generally successful provision of emergency aid to this enormous number of people, particularly given the scale and speed of the influx, represents a substantial achievement for the international community.

One reason for this success was the regionalization of emergency food aid provision. This meant that food aid was made available to the region, and was allocated within the region as needs arose. This gave the humanitarian community considerable flexibility in responding rapidly to needs.

Continued insecurity during 1995 and into 1997 in Burundi forced approximately 300,000 Burundis to seek refuge in the Democratic Republic of Congo and Tanzania, and led to repatriation of the approximately 200,000 Rwandan refugees in the country. During this period, a further 200,000 people have been estimated to be internally displaced in Burundi at any given time, although many were only displaced for short periods. Continuing insecurity throughout 1996 and into 1997 led to the government's formation of 'regroupment camps'. The army grouped approximately 200,000 civilians into these camps in order to better protect them from continuing violence. Reports indicated that the health and nutrition situation of these populations was poor.

In the first six months of 1996, the situation for the refugee population appeared to deteriorate. The governments of the former Zaire and Tanzania grew frustrated with a lack of progress in the repatriation of Rwandan refugees. In the case of the former Zaire, this led to restrictions on economic activities of refugees, and restrictions of services provided in many camps to only those considered essential for the refugees. In Tanzania, there were also restrictions on refugees' activities, although the measures taken did not appear to be as drastic as those in the former Zaire. The security situation in Burundi also deteriorated in the first half of 1996, despite international efforts to stop the fighting. This insecurity led to further population displacements within Burundi and also prevented most aid agencies from working, particularly in the northern provinces.

Map 2.: Great Lakes Region

Map taken from ReliefWeb.

Further population movements were seen in Eastern Zaire in early 1997 as the rebel forces of the Alliance of Democratic Forces for the Liberation of Congo (ADFL) led by Laurent-Desire Kabila, swept across the country and eventually took power in May 1997. The country was immediately renamed the Democratic Republic of Congo (DRC). Hundreds of thousands of Zairians were internally displaced, and the refugees remaining in the country moved from one makeshift camp to another, with barely time for humanitarian aid to arrive between moves. At the time of writing, there were estimated to be 200,000 refugees unaccounted for in Eastern DRC.

Democratic Republic of Congo (DRC) The nutritional status of refugees in the former Zaire improved in 1995-96 so that levels of wasting in the camps were about the same or lower than what would be expected in the host population. The security situation during 1995 was relatively calm, so that food aid deliveries were unaffected by security incidents. However, logistical factors and political considerations often determined that food aid deliveries were incomplete. Figure 11 shows levels of wasting in the camps by year.

In the last six months of 1996, ethnic conflict between Tutsi rebels and the army in Eastern DRC, which had been "simmering" at a low level, erupted. This escalation was possibly fanned by the presence of many Hutu militia in the camps. The fighting spread rapidly and led to the dispersement of the refugee camps, first in the Goma region, then in Bukavu and Uvira. In addition, hundreds of thousands of Zairians were internally displaced and a small number of Zairians crossed the border into Rwanda (see section below on Rwanda for details).

Figure 11.: Levels of Wasting in Camps for Rwandan Refugees, in the former Zaire

Massive numbers of Rwandan refugees began returning home at the end of October 1996, although many had spent several weeks in the bush prior to crossing the border. By December, it was estimated that almost a million refugees had returned. There remained, at the end of 1996, an estimated 300,000 Burundi and Rwandan refugees in Eastern DRC, and an unknown number of internally displaced Zairians. Access to these populations was very limited so that there was mounting concern over the health and nutrition situation of this population, with anecdotal reports of rising malnutrition and mortality.

The Alliance of Democratic Forces for the Liberation of Congo (ADFL) swept across the country in early 1997, arriving in Kinshasa in May 1997 and taking control. The advance westward of Kabila's forces meant that refugees were continually forced to move from one makeshift camp to another, with barely enough time between moves for the delivery of humanitarian aid. Mortality rates were measured at ten times normal in these camps, and cholera outbreaks were reported.

By mid-1997, many refugees and IDPs had either repatriated or returned home. It is estimated, however, that there are up to 200,000 refugees remaining in the eastern part of the country.

In Goma, DRC, the nutrition situation for the approximately 730,000 Rwandan refugees was stable throughout 1995 and the first half of 1996. Levels of wasting were below what would be expected in the host population and varied from 1.6-3.5% (see box on page 54), despite a ration which provided on average 1500 kcals/person/day, but sometimes as little as 800 kcals/person/day.

There were approximately 27,000 new arrivals from Burundi at Uvira, DRC, in the first half of 1996, bringing the total number of refugees to 220,000. These new arrivals were fleeing intensified insecurity in Burundi. Surveys in April 1996 showed that wasting varied from 2.3-11.0% amongst this population.

There were also approximately 300,000 refugees in Bukavu, DRC, whose nutritional status remained stable and adequate throughout 1995 and the first half of 1996. Levels of wasting in May 1996 were 0.9-7.0%.

The adequate nutritional status of these refugees was thought to be maintained due to a variety of factors, including a thriving market system and economic activities. Many humanitarian agencies were, in fact, unaware of the degree of self-sufficiency, as indicated by events in early 1995 when tow general ration deliveries in the Zairian refugee camps prompted agencies to declare a state of impending emergency. Yet, although the level of rattans distributed remained taw for many subsequent months, the nutritional status of Rwandan refugees in DRC did not deteriorate.

Another factor which contributed to improved nutritional status in the Zairian refugee camps was the changes made to general rattan distribution systems towards the end of 1994 and during 1995. At the start of the emergency in July 1994, general rattan distributions were organized through heads of communes and resulted in open discrimination against various sections of the community. High levels of wasting for the first four months of the emergency were partly attributed to inequitable distribution systems. However, systems gradually moved towards "lower levels" of distribution, e.g. from heads of sectors grouping together many families to cellules grouping together a few families and then eventually to families, so that inequity became far less pronounced.

Furthermore, several 'novel' and apparently successful approaches to general ration distribution were tried by different implementing NGOs. For example, several distribution points were established in a camp based on family size. This enabled each site to distribute equal quantities of food to the beneficiaries. This was felt to reduce opportunities for unfair distributions.

However, a recurrent problem was a shortage of firewood for cooking, which hindered the refugees' ability to prepare food. The government-imposed ban on economic activities alarmed many agency personnel concerned about the refugees' ability to supplement inadequate general rations.

The conflict in Eastern DRC and massive repatriation meant that the refugees remaining in the country created makeshift camps. In early 1997, refugees were regularly displaced from these camps as rebel forces moved westward, with little if any humanitarian relief provided between moves. By April 1997, mortality rates of 11-20/10,000/day (20 to 40 times normal) were recorded. These people, who were relatively healthy when the camps were dispersed at the end of 1996, were reportedly dying of malnutrition and cholera. Difficulties in supplying these camps may be related to political factors as well as logistical considerations. For example, it has been suggested that denial of access was part of the ADFL military strategy.

This situation highlights a few points worth considering further. Levels of wasting in the refugees camps were less than 10% and reflected in part the marked degree of self-sufficiency attained by large numbers of refugees. This was not widely understood by aid agencies at the time, and would have been useful to know. The need and importance of innovative approaches to food distributions, along with a demonstrated utility of improved information about the food economy of a refugee population, were also illustrated in this situation. The political obstacles to providing adequate relief aid to Rwandan refugees after camp dispersal need to be brought into the open so that ways of dealing with this type of problem in the future can be found.

Rwanda The overall security situation in Rwanda during 1995-96 was generally calm. Despite this, there were security incidents reported throughout the year, particularly in the west where there were frequent incursions from the Goma camps. A particularly serious security incident occurred with the forced closure of Kibeho camp (a camp initially set up in the French protection area 'Zone Turquoise' for IDPs in Rwanda), when many thousands of lives were reported lost in the resulting confusion and panic.

Approximately 800,000 people in Rwanda were thought to be nutritionally vulnerable throughout 1995. Fighting in Eastern DRC, and the subsequent repatriation of over one million refugees swelled the numbers considered to be vulnerable. Food aid deliveries in 1995 were relatively unimpeded, and nutritional indicators pointed to a stable situation comparable to that seen before the war. For example, a nutritional survey in Kigali in September 1995 showed 5.0% wasting while another survey at prefecture level showed 2.9% wasting in May 1995.

Ethnic fighting in Eastern DRC precipitated the return of over one million refugees from DRC between October and December 1996. Food aid was distributed to these new returnees at the commune level. Concern was expressed over the ability of the country to reabsorb such a large number of returnees, who needed homes and land to farm, in a short time period. However, the previous influx of returnees during 1995 to mid-1996 had been relatively easily assimilated. A variety of factors were said to have assisted this process, including the government policy of avoiding formation of camps, improved agricultural production allowing for an increased population, and NGO support of health programmes.

This fighting also led to a small influx of approximately 13,000 Zairian refugees, who are housed in Umubano Camp. Having spent some time in the bush before arriving in Rwanda, many of the new arrivals were malnourished. A survey in Umubano Camp in April 1996 (then called Petite Barrière Camp) showed 17% wasting with 7.4% severe wasting. The general ration was set at 1980 kcals/person/day for this camp. By August 1996, levels of wasting were 10.2%. It was felt that levels of wasting remained high due to the malnourished state of new arrivals.

A relatively good harvest in January 1997 was considered to be inadequate to cover the needs of the increased population in Rwanda (see Figure 12). In addition, January-February is generally when crops are planted, so that new arrivals were largely unable to plant and harvest any crops. It was therefore likely that emergency food assistance would be needed for at least the first six months of 1997, and that provision of farming inputs would be necessary for the newly returned population.

Figure 12.: Estimated Population in Rwanda from January 1996 Projected to December 1997

There are an increasing number of reports of insecurity in western prefectures of Rwanda which are, at least in some cases, leading to the suspension of humanitarian activities. At a time when efforts should be focused on activities to support the reintegration of a large number of returnees, this apparent upsurge in violence gives cause for concern.

Burundi Continuous and widespread insecurity in Burundi has led to population displacements within the country, and temporary evacuation of aid personnel was commonplace in 1995. The security situation continued to deteriorate in 1996, with attacks, looting and fighting reported, especially in the northern provinces. In some instances, the attacks appeared to have been deliberately aimed at aid workers. A tragic example was the murder of three ICRC workers on their way to deliver water in June 1996. This heightened insecurity led to approximately 130,000 people leaving the country, with 100,000 seeking refuge in Tanzania and at least 30,000 in DRC.

It was estimated that at any one time up to 200,000 people were internally displaced by the conflict during 1995. There were also about 200,000 Rwandan refugees in Burundi in 1995, most of whom repatriated by early 1996. Ethnic fighting in Eastern DRC in 1996 led to the return of at least 46,000 Burundi refugees, many of whom arrived in an appalling nutritional state. A screening carried out at Gatumba transit camp in Burundi, showed 18.2% wasting across all age groups; 17% of those under five years old were wasted.

Regroupment camps were set up in some of the more insecure areas to protect civilian populations and housed up to 200,000 people. Conditions were extremely poor in many of these camps, and reports were of high levels of malnutrition and disease among their populations. For example, wasting was measured at 16% in one area, with over 3% severe wasting. The government intends to send people home from the camps as soon as security allows; there were reports of people leaving spontaneously in June 1997.

A rapid, flexible approach to food aid delivery was therefore necessary, given this fluid security situation. WFP developed a system of rapid assessment followed by food aid delivery, which was often determined to be needed only for a short period of time. Nutritional surveys were very difficult to carry out under these conditions; those that were possible frequently showed high levels of wasting. For example, a survey at the end of 1995 in four camps for the internally displaced found levels of wasting ranging from 15-18%. Cholera was also present in Burundi throughout 1995.

International economic sanctions imposed on Burundi after a bloodless coup d'état in June 1996 included fuel and food aid, and posed serious threats to the continuation of emergency humanitarian activities. By the end of 1996, some restrictions had been removed but supplies, particularly of fuel, remained limited. Continued sanctions in combination with a security situation that severely restricted the capacity of agencies to work, led to a continuing deterioration of the nutritional and health situation in many areas of Burundi continued to deteriorate in the latter part of 1996.

Tanzania In early 1995, there were estimated to be 630,000 Rwandese and Burundi refugees in Tanzania. This number increased throughout 1995 and into 1996 due to the continuous arrival of Burundi refugees. During the second half of 1996, the growing influx of Burundi refugees and the lack of progress on the repatriation of Rwandese refugees led to growing tensions. In mid-1996, the Government of Tanzania decreed that refugees could no longer cultivate land or conduct economic activities beyond a four kilometre radius of the camps. The government also issued an ultimatum that all Rwandan refugees must return before the end of 1996. This ultimatum was accompanied by the dispatch of Tanzanian troops so some 400,000 Rwanda refugees returned home at the end of December 1996. However, insecurity in Democratic Republic of Congo (then Zaire) led to a new influx of refugees so that by mid-1997 there were a total of 344,000 refugees from Burundi and DRC in Tanzania.

The health and nutritional status of the refugees in Tanzania remained adequate and stable during 1995 and the first half of 1996. Even though general ration deliveries were periodically inadequate during 1995, nutritional surveys continued to show levels of wasting under 5% in the Ngara and Karagwe camps for Rwandan refugees, below levels seen in the local population. This reflected the fact that refugees were able to carry out farming and other economic activities to supplement this ration. In late 1995, food distribution systems changed in the camps from family to communal level distribution based on community participation. The system was considered to be more successful than previous ones as there was greater transparency and equity, and refugee families did not have to spend so long in distribution queues. Furthermore, agencies spent less time implementing food distribution.

Despite restrictions on economic and farming activities, levels of wasting still remained low during 1996. However, cases of vitamin B2 deficiency, seen as angular stomatitis, were identified, especially in the Ngara camps (estimated population 490,000). No further investigations were undertaken regarding the presence of angular stomatitis because this population returned to Rwanda soon after the reports. Throughout this period, water availability remained a problem, particularly affecting the Karagwe camps during the dry season.

The insecurity that accompanied Kabila's takeover of the Democratic Republic of Congo led to the influx into Tanzania of approximately 100,000 refugees in early 1997.

Indicators available for this population point to a generally adequate health and nutrition situation, despite some difficulties with food provision due to poor road conditions. For example, levels of wasting in May 1997 varied from 1.8-7.2%. Organized repatriation of these refugees began before the end of 1997.

Water provision and logistic issues were problematic in the Tanzania camps through the period under review. Despite these problems, the health and nutritional status of the refugees was adequate and stable.

In conclusion, the humanitarian relief operations mounted in the wake of this regional crisis were on a scale rarely, if ever, seen before and were largely successful.

A number of some issues requiring further consideration emerged. It became clear that innovative approaches to food distributions, as seen in Goma and in Tanzania, greatly improved access to rations for a number of families. Information on the food economy in camps would also be useful and would allow the humanitarian community to better target rations and set ration levels. Improvement in these areas depends to a large extent on knowledge of the community and its organization in the camps.

The concept of voluntary repatriation has been questioned here in two instances, ft is difficult to say that repatriation was voluntary for Rwandan refugees in Tanzania and Eastern DRC, on the other hand it is not clear to what extent these refugees, particularly those in Eastern DRC, were kept in the camps against their will. Furthermore, obstacles to humanitarian aid provision put up by the ADFL, which were politically motivated, were in direct conflict with humanitarian goals, and ways of dealing with this situation must be sought for future conflicts.

Ethiopia

In early 1995, there were about 259,000 Somali refugees and 51,000 Sudanese refugees in Ethiopia. The number of Somali refugees increased through 1995 and 1996 due to fighting in north-west Somalia, but in 1997, some small-scale repatriation meant that the total number began to decrease. The number of Sudanese refugees fleeing conflict in Southern Sudan increased gradually from 51,000 in 1995 to 54,000 in mid-1997.

The number of Somali refugees in Ethiopia increased in the early months of 1995 due to fighting in Somalia. Once this influx was over, numbers were stable throughout the remainder of 1995-96. Early in 1995, crude mortality rates were low at 0.2/10,000/day in the camps.

Surveys conducted in March 1995 in some camps in the Jijiga area showed levels of wasting from 12% to 17.3% (see Figure 13). Although these levels are high, they represent an improvement over surveys conducted in 1994 when wasting was approximately 20%. Surveys carried out in July 1995 in the camps not included in the March surveys showed levels of wasting varying from 7.5% to 13.3%. It was felt that these somewhat elevated levels could be due to the continued arrival of malnourished individuals from Somalia.

Surveys conducted in the camps for Somali refugees in the east in May 1996 showed an extremely worrying situation. The nutritional situation had deteriorated in all the camps since the previous set of surveys, and levels of wasting varied from 15.2%-21.1%. The highest levels of malnutrition were seen in Kebri Beyah (estimated population 10,000) and Derwanaji (estimated population 43,000) at 20.5% and 21.1%, respectively. In other camps (i.e. Rabasso and Camaboker), levels of wasting have doubled since surveys were conducted in July 1995. Crude mortality rates were reported to be 1-2/10,000/day (3-7 times normal).

Figure 13.: Levels of Wasting in Somali Refugee Camps in Ethiopia over Time

*includes Hartesheik, Kebre Beyah, Denwonaji, Teferiber.

Map 3.: Ethiopia

Although an increase in levels of malnutrition is often seen during the 'lean' season, comparison with surveys carried out in July 1995 indicated a marked deterioration. A household food economy assessment conducted by Save the Children Fund (UK) towards the end of 1996 in Kebri Beyah refugee camp provided information on the food security of this refugee population. The assessment found that since food distributions had been irregular, providing less than 100% of caloric needs over the year, each family had to develop income-generating strategies in order to survive. As most families lacked any capital for starting up a business, most of these activities were small-scale and opportunistic, e.g. selling firewood and making charcoal.

The main sources of food for this population are the general ration and food purchased with wages earned by working for local people, which together account for 88-96% of total caloric intake. Another source of food is a blanket feeding programme which provides approximately 6% of an average household's food needs for each child registered in the programme. This programme also appears to have resulted in a reduction in the elevated rates of malnutrition observed in May 1996.

There is a serious water shortage in the camps. Water has been trucked into many of the camps for years, and the 3-4 litres/person/day available is far below the 20 litres/person/day recommended as a minimum. Water shortage has been linked to an outbreak of diarrhoeal disease in the camps in January and February of 1996. Other factors which may be leading to elevated levels of wasting include a lack of ration cards for new arrivals (particularly in Rabasso and Daror camps) and increases in cereal prices due to the devaluation of the Somali and Somaliland currencies.

The numbers of Sudanese refugees increased slowly over the two-year period. The nutritional situation in these camps was variable in early 1995, with levels of wasting from 5.4% to 15.7%. Similar rates of wasting were seen in a survey conducted in July 1995.

In mid-1996, the nutritional situation for the Sudanese refugees in the west appeared to be stable and adequate. Levels of wasting in the camps varied between 6% and 8% with almost no severe wasting. This population is situated in an area where there is greater opportunity for self-reliance than is the case for the Somali refugees in the east. The Sudanese refugees are able to supplement their rations with some limited crop cultivation, some livestock, wild foods and in some cases fishing. However, water is reportedly a major concern of the refugees in Fugnido camp, and existing boreholes and broken hand pumps need to be repaired.

Surveys in early 1997 showed a deterioration in the situation, with levels of wasting of 17%. In response to this situation, supplementary feeding programmes were established to assist children and other vulnerable groups, and a review of the monitoring system is underway.

In spite of the fact that these are relatively settled populations, particularly in the case of the Somali refugees, levels of wasting measured are similar to those seen in the initial phases of an emergency. These indicators continue to describe a very serious situation.

Kenya

In early 1995, there were some 232,000 Somali, Sudanese and Ethiopian refugees in Kenya. Over the period under review, this total number decreased to 173,000 due to repatriation of Somali and Ethiopian refugees and permitted the closure of some camps. The number of Sudanese refugees increased due to continued fighting in that country.

Map 4.: Kenya

In mid-1994, the levels of wasting in the three camps for Somali refugees in the Dadaab area of Kenya were low and varied from about 5.3% to 8.9% (see Figure 14). However, surveys in March 1995 showed an apparent decline in the nutritional status of this population, when wasting was measured at 15.3% in Ifo camp. It was suggested that one reason for this decline might be that even though the officially allocated ration was 2400 kcals/ person/day, distributions to non-registered refugees determined that receipts were likely to be closer to 1800 kcals/person/day.

Figure 14.: Levels of Wasting In Dadaab Area Camps for Somali Refugees in Kenya over Time

*Hagadera Camp not included in this group

Surveys conducted in January 1997 confirmed a declining nutritional status of this population, leading to what was described as a nutritional emergency. For example, a recent survey in Ifo camp showed 33.3% wasting, with 6.7% severe wasting. The general ration provided approximately 1850 kcals/person/day in November 1996 and 1700 kcals/person/day in December 1996. The under-five mortality rate was 5/10,000/day in January 1997 (5 times normal). Coverage of the therapeutic feeding programme was only 63%. Similar situations were described in Hagedara and Dagahaley camps.

Problems with the food supply and distribution systems in the camps have been identified as primary factors contributing to the increased levels of wasting. Since June 1996, there have been problems with the supply of beans and the average number of kcals/person/day supplied in the general ration has been less than 1700. The general ration has not contained blended foods or sugar, despite recommendations made following an assessment mission in October 1996 to include these foods. An increase in the incidence of diarrhoeal disease and malaria may also be influencing nutritional status. Furthermore, some cholera cases were confirmed at the end of 1996.

Scurvy appears to be a seasonal problem in the Dadaab area camps in the September-December period. Indeed, cases were noted in 1996 during this period. Nevertheless, the international community has not, in the past, acted to prevent these predictable outbreaks. Instead, curative action is taken after the outbreaks occur.

Indicators in Sudanese camps in early 1995 were normal, with a CMR of 0.17/10,000/day and an under-five mortality rate of 0.67/10,000/day. However, an assessment of Kakuma camp, where 32,000 refugees live, revealed a very high rate of severe anaemia among boys aged 8-20 years. Some possible explanations for this were that most of this population were unaccompanied minors with little linkage to household economies in the camp. Furthermore, the average per capita kilocalorie requirement for this age group (2200 kcals/person/day) was not being met by the 1900 kcals/person/day provided by the ration. The ration was also deficient in bioavailable iron and vitamin C, which enhances iron absorption.

A high proportion of this population was known to be selling off some of their rations in order to purchase non-food items. This is most likely to be due to a lack of those items among distributed commodities. High levels of intestinal worm infestation may also have been contributing to this unusual pattern of anaemia. In response to this problem, a school feeding programme including CSB was recommended.

It should be noted that there are serious difficulties in supplying the Kakuma and Dadaab camps with food aid. These camps are very isolated and road conditions, particularly during the rainy season, are not good. In addition, the availability of some foods locally, for example fresh vegetables or meat and fish, is poor. However, indicators from the camps, including high rates of wasting and micronutrient malnutrition, point to a very serious situation, which should not be seen in a stable refugee population.

Liberia/Sierra Leone Region

Fighting in Liberia broke out in 1989 and continued virtually unabated until 1994. The conflict between rebel factions continued throughout 1995, but a peace accord signed in August 1995 led to an improvement in the security situation and a gradual opening up of areas in the country to humanitarian aid. A Council of State was installed as the governing body, and ECOMOG, a West African peace-keeping force, began to be deployed outside the capital city, Monrovia. For most of 1995, the city was relatively calm. However, factional fighting broke out there in April 1996, displacing almost 300,000 people and leading to the evacuation of almost all international aid personnel. The security situation since then has gradually improved, with disarmament beginning in November 1996, raising hopes for a more durable peace. The election of Charles Taylor as president in what has been termed fair and transparent elections are further strengthening hopes for a lasting calm in the country.

In Sierra Leone, democratic elections took place in March 1996, calming years of rebel activity. The installation of a democratically elected government and subsequent peace talks with rebels led to a marked decrease in insecurity and rendered most needy populations accessible to humanitarian relief. However, the improving situation changed abruptly with the military takeover of the government; insecurity led to disruptions to and in many cases cessation of humanitarian relief by June 1997.

By then, there were some two million people affected by the conflicts in Liberia and Sierra Leone, some of whom still remained inaccessible or only periodically accessible to humanitarian aid.

Liberia Almost two million people in Liberia were estimated to require humanitarian assistance throughout 1995 and 1996. Many of them were inaccessible for extended periods of time due to insecurity. The number of people requiring aid began to decrease in 1997 as the disarmament process took hold and following Charles Taylor's election as president.

As the villages in the Liberian countryside became accessible to humanitarian organizations, extremely high levels of wasting were often seen. For example, in Lower Bong and Upper Margibi, wasting was almost 20% and oedema was measured at 37%. A follow-up survey was conducted after the implementation of general ration and selective feeding programmes. Wasting and/or oedema had decreased to 6.4%.

A similar situation prevailed in Tubmanburg, a village which was cut off from humanitarian aid for almost eight months. Once access was gained, a catastrophic situation was seen. Wasting levels of almost 40% were measured, with 32% severe wasting. Mortality rates were 40 times normal and under-five mortality rates were 50 times normal (see Figure 15). After one month of emergency aid deliveries, the crude mortality rate had decreased to 5.4/10,000/day. White this marked an extreme improvement, mortality rates were still ten times the normal rate in mid-October 1996.

In other newly accessible areas, levels of wasting were not as high as expected. For example, in Buchanan in July 1995, wasting was measured at 8.9% and in Harbel Unification Town in April 1995,7.4%.

Figure 15.: Crude Mortality Rates and Under-five Mortality Rates in Tubmanburg, Liberia between 1 June 1996 and 10 October 1996

Data from 'Demographic and Nutritional Assessment Tubmanburg Bomi County, Liberia'
Médecins sans Frontières and EPICENTRE, October 1996.

However, the destruction of infrastructure and equipment caused by this resurgence of violence in the capital had a longer-term impact on the ability of the humanitarian aid community to work. A survey in Monrovia in July 1996 revealed an alarming situation. Wasting among the displaced and resident population was measured at 21.2% and 13.3% respectively. Measles immunization coverage was tow at about 60%. A follow-up survey in February 1997 showed a much improved situation. Wasting among the displaced populations had decreased to 13.2%, with 0.9% severe wasting, and among the resident population wasting was measured at 6.1%.

Humanitarian agency policy with regard to emergency general ration provision in Liberia has been to phase out this form of support by introducing the targeting of general rations. However, this policy, which effectively began in 1992, has been criticized on the basis that lack of socio-economic data led to indiscriminate ration reductions, without taking into account the differential capacity of populations to employ coping strategies. In addition, it was not always possible to deliver the targeted ration due to insecurity.

Insecurity and looting also led to the decision to replace rice distribution, a highly valued commodity often looted by the military factions, with buglur wheat, a commodity less likely to be looted. This required an extensive campaign to prepare the populations, as this commodity was unfamiliar to them.

Security at distribution points, nevertheless, remained a problem, and beneficiaries stated their preference for supplementary feeding programmes as opposed to general distribution. They felt they would not be put at so great a risk if they participated in these programmes.

However, there has been criticism over the implementation of selective feeding programmes. There is agreement that these were necessary during the early phase of the emergency when levels of wasting were extremely high. However, feeding programmes were believed to be inefficient and costly in the context of a frequently inadequate general ration, leading to high numbers of re-admissions and default. One view is that resources would have been more effectively allocated to support the general ration programmes in order to prevent the need for selective feeding.

Sierra Leone Rebel activity in the Sierra Leone countryside throughout 1995 led to large-scale displacement and rendered many areas periodically inaccessible to humanitarian relief agencies. The estimated number of displaced people requiring food aid assistance increased from 490,000 to 730,000 during the year. There were estimated to be a further 800,000 people internally displaced who were not in need of emergency assistance. By the end of 1996, the security situation had improved to the point where some people had begun to return home, leaving approximately 600,000 people dependent on emergency food aid.

Elections were successfully held early in 1996, and the installation of a new government and subsequent peace talks gave rise to hopes for a lasting peace. Rebel activity decreased markedly, and almost all areas of the country were accessible to humanitarian relief. However, a military coup ousted the elected president in May 1997, and periodic outbreaks of violence once again led to population displacements and at the same time restricted humanitarian activities. Information prior to the coup d'etat is presented here.

Once areas that had been inaccessible opened up, high levels of wasting and mortality were most often found. In Kenema Town, wasting in August 1995 was measured at 21% with 8% severe wasting. Mortality rates were eight times the normal. In camps around Kenema, wasting was 37% and severe wasting was 11%. Mortality rates were 17 times normal. A follow-up survey three months later showed 6.8% wasting in Kenema, and 5.6% wasting in the camps. This rapid improvement was felt to be largely due to general ration distributions.

As rebel forces were increasingly pushed to the west of the country, some alarming situations were uncovered. For example, a group of 500 people who had apparently been held captive for up to four years and forced into slave labour by rebel fighters were discovered in mid-1996 in Blama, near Kenema. Among this population, estimates of levels of adult malnutrition were 25%. This group was moved to a camp near Kenema, where preparations had been made for their arrival. It was believed that there were at least a further 1500 similarly affected people hiding in the bush, whose nutritional status was likely to be catastrophic. It was also believed that this situation was not unique and that there were many other brutalized captive populations throughout the country who would emerge from the bush in the coming months.

Insecurity following a coup d'état in May 1997 meant that many areas were once again inaccessible, and it is likely that this lack of access is resulting in a decline in the nutritional status of many people in Sierra-Leone.

Cote d'Ivoire There are approximately 305,000 Liberian refugees in Cote d'Ivoire. Repatriation plans are in place, but no time frame has been established for this programme. There are no new nutritional data on this population, but it is believed that the adequate nutritional status of this population shown in a 1994 survey (wasting of 8%) has not changed.

Guinea There are also about 600,000 Liberian and Sierra Leonean refugees in Guinea. These refugees are also expected to return home as soon as security allows.

The Sierra Leonean refugees are in the Forecariah Prefecture, and a survey conducted in 1995 showed 8.2% wasting. Since the ration distributed provided only 1300 kcals/person/day at the time, it was felt this population had attained some degree of self-sufficiency. The recent upsurge in violence in the country has led to the arrival of approximately 10,000 new refugees.

New refugees continued to arrive in Guinea from Liberia in 1995, many of whom arrived in a dire nutritional state. A survey carried out in early 1995 showed 30% wasting among new arrivals; wasting among those who had been in Guinea was 4%. Another survey in August 1996 showed levels of wasting from 1.1 -3.2%.

This regional crisis has highlighted some issues for further consideration. The inaccessibility of population groups has been evident in many instances and is an obvious obstacle to the provision of adequate humanitarian aid. The problem of inadequate general ration provision is also highlighted, and the efficacy of selective feeding programmes, particularly in the absence of adequate general rations, is also questioned.

Shaba, Democratic Republic of Congo

Ethnic violence in the Shaba region of the former Zaire in 1992 led to the displacement of about 600,000 people to the Kasai region further north. During the migration, many people stopped in villages along the route north, while others settled permanently on these sites. Since the end of 1995, there has been little further displacement from the Shaba region.

Survey information shows varying degrees of self-sufficiency have been attained by this population. Wasting and mortality rates among the displaced populations, and to a lesser extent the resident populations hosting them, were extraordinarily high in 1993 and 1995. However, the nutritional situation in many areas where displaced people are now staying has since improved. For example, surveys in Likasi at the end of 1995 showed 5.5% wasting and in Lupata wasting was measured at 9.9%.

In contrast, the nutritional situation in Mwene Ditu remains critical. Throughout 1993-1994 levels of wasting in the town and the camps for the displaced were consistently over 25%. In October 1995, wasting among the resident population was measured at almost 18%. In the camps for the displaced, wasting was measured at 43% with almost 10% severe wasting. These extremely high levels of wasting were thought to reflect a variety of factors. For example, many of the displaced people were not farmers, therefore they had difficulty finding work in these rural surroundings. Furthermore, there was little household income available to spend on health care so that untreated childhood illnesses often led to malnutrition. The survey was conducted during the pre-harvest season. It is likely that the establishment of income generating projects for both the resident and displaced populations of Mwene Ditu would mitigate some of the worst effects of this chronic emergency.

Mozambique

The signing of a peace accord in October 1992 was a definitive step in ending a 16-year civil war in Mozambique. During the war, over 1.6 million people fled the country as refugees, and at least 3.4 million people were internally displaced. Once it became apparent that the peace would last, refugees began returning home, and by the end of 1995 the repatriation operation was completed. Due largely to a series of successful harvests amongst recent returnees to Mozambique, the estimated number of people requiring food aid has decreased from 1.6 million to 70,000 at the end of 1996. This level of aid was required until the harvest in April 1997, when it was felt that returnees had reached self-sufficiency (see Figure 16).

Levels of wasting reported in 1995 varied from 1.9-5.0% between July-August 1995, although there appeared to be a slight upward trend, based on information available in 1996 when levels of wasting were reported to vary between 11% and 13.2%.

Figure 16.: Number of People Requiring Emergency Food Aid in Mozambique over Time (excluding flood victims)

In October 1995, an outbreak of pellagra occurred, centred mainly in the Mutarara district of Tete Province. A fortified blended food (CSB) was requested by WFP for distribution among the vulnerable population. Once the CSB arrived, administrative difficulties meant that it could not be imported or distributed for three months. Subsequent distributions appeared to lead to a decline in the number of cases of pellagra seen. These outbreaks are likely to be seasonal, and nutrition education programmes are being arranged to address the probable root causes of the outbreaks.

Logistics and health care provision were two major constraints to the continued improvement of the nutrition and health status of populations in Mozambique. For example, there were some areas, such as Gaza province, where a lack of logistic capacity undermined planned general ration distributions. In these areas, improvement of roads and storage capacity allowed for some stockpiling of foods for vulnerable populations. Reports also regularly underscored the need to strengthen health service provision, particularly for malaria control and measles immunization, and this remains a priority after the emergency aid programmes have ended. In addition, the de-mining process in Mozambique is on-going and, as more land is cleared, the food security situation of the country will most likely continue to improve.

Somalia

A civil war erupted in Somalia in 1991 with the overthrow of the military rulers, and insecurity has persisted in varying degrees of severity since that time. A UN peacekeeping force was established for a time, and it was feared that the end of its mission would lead to all-out interclan warfare. Although this was not the case, periodic and escalating insecurity continued. The death of General Aideed in August 1996 did not lead to the reduction in the conflict that many had hoped for. Indeed, the security situation sharply deteriorated in many areas.

Throughout 1995 and into 1997, food security continued to be adversely affected by variable harvests, and periodic insecurity affected economic activities and agricultural production. For example, following the capture of Baidoa by General Aideed in November 1995, some 20,000 people were displaced, leading to an estimated 67% reduction in food production. Dramatic food price rises due partly to the closure of Mogadishu port have also exacerbated economic hardship for this population.

Throughout 1995 and up to August 1996, an estimated 840,000 people received food aid (600,000 returnees and 240,000 internally displaced people). By August 1996, the numbers estimated to require emergency food provision had dropped to 150,000 people. Populations most affected by food insecurity appeared to be returnees and IDPs, particularly in Mogadishu, Kismayo and Juba valley. By June 1997, the number of people estimated to require emergency assistance had risen to 688,000.

Map 5: Somalia

The security situation remained tense in Somalia throughout most of 1995-97, making humanitarian relief work difficult to carry out. In most cases, however, relief activities were not prevented. Nutritional and health data available for the two-year period illustrate very poor conditions. For example, a survey in June 1995 in Mogadishu showed 24.9% wasting among children 6-59 months old. As a result, feeding centres were set up throughout the city and subsequent surveys/screenings showed a much improved situation, with some areas reporting levels of wasting as low as 2.1-7%. Another nutritional survey in Kismayo town in July 1995 found 17.8% wasting in children under five years old.

A cholera epidemic which broke out in October 1995 illustrated that humanitarian agencies were able to continue working effectively, even under difficult circumstances. Medicines were initially in short supply and, although it took several months, full supply and distribution was possible by June 1996.

In the Juba region, a combination of poor rainfall and subsequent crop failure and increasing insecurity led to population movements toward the cities (e.g., Kismayo) and toward the Kenyan border. A nutritional survey in Bulla Huwain, near the Kenyan and Ethiopian border in October 1996 found 37% wasting and 10% severe wasting. Most of the population were nomadic pastoralists, many of whom were displaced from other parts of Somalia. Logistical difficulties remained the major constraint on relief activities in the Juba valley. A combination of poor roads and insecurity hampered the delivery of humanitarian aid, which was increasingly needed.

Figure 17: July Harvest Cereal Production in Somalia

From: Food Security Assessment Unit, Gu Harvest Assessment 1 Aug 1997.

There are signs of a deteriorating nutritional situation in much of the country. Numerous maternal health centres are reporting increasing numbers of wasted children among new admissions. For example, in Kismayo, Lower Juba, MCH centres are finding an increase in the percentage of children admitted with malnutrition for the fourth month in a row. In Hiran region a screening exercise showed an increased level of malnutrition in January and February.

This apparent decline in nutritional status is likely to be exacerbated by a number of factors. Prices of grains, have continued to rise since and cereal stocks are reportedly depleted, particularly in the south. The price of sorghum has quadrupled and is beyond the purchasing power of large sections of the population. Conversely, prices of livestock have declined in rural areas due to increased sales as a result of shortages of water and fodder.

There are reports of population movements, mainly from Bay and Bakool regions towards the Juba valley, Mogadishu and Kenya. The nutritional situation is reported to be alarming in the recently established settlements for the displaced in Baidoa town. Food and water shortages have also resulted in a deterioration in the health situation of the population, with cases of cholera and tuberculosis on the increase.

In summary, the purchasing power of much of Somalia's population has been eroded and assets have been sold off, leading to widespread chronic food insecurity, and in many cases destitution.

It is likely that programmes such as food for work, subsidized food sales and income-generating projects would improve food security for this population.

Sudan

The current phase of Sudan's civil war has persisted for 11 years, pitting the military forces of the government against the Sudan People's Liberation Army (SPLA). The fighting has been concentrated in the southern region, and has led to huge population displacements. At least two million people are thought to have migrated north, a large section of whom live in poverty in and around Khartoum. There are estimated to be almost two million people affected in the south, many of whom have been displaced more than once.

'The protracted war has created a cycle of displacement and malnourishment among the civilian population, has flooded neighbouring countries with refugees, created enormous squatter and displaced camps, and has caused a chronic humanitarian emergency that has remained one of the UN's most expensive operations.' (USCR 1995. p. 76)

Khartoum There are some data available on the health and nutrition situation for the approximately 120,000 displaced people living in camps around Khartoum. Many of these people have been forced to relocate farther away from Khartoum as former camps have been razed to the ground. Reports indicate under-five mortality rates of 4/10,000/day (13 times normal). A major cause of death is malnutrition, and other causes include diarrhoea, lower respiratory tract infection and malaria. Clinical signs of vitamin A deficiency are also reported.

Map 6: Sudan

In general, the health and nutrition status of this internally displaced population is deteriorating, and this is felt to be largely attributable to increased poverty resulting in decreased access to food. Likely causes are frequent forced population movements as a result of camp demolitions and relocations, and camps being moved farther away from income-earning opportunities and relocated where minimal services can be provided. Provision of water and sanitation, food, selective feeding and curative health care has been increasingly poor.

At the end of 1996, there were hopeful signs that the situation might improve when greater access to the camps was granted. Somewhat improved access allowed humanitarian agencies to begin to address high levels of wasting, and several therapeutic and supplementary feeding centres have now been opened. Priority interventions will include improving the supply of clean water and essential drugs, and establishing nutritional surveillance systems.

Southern Sudan Fighting continues in Southern Sudan, and at least 1.7 million people in the region are affected by the conflict. A ceasefire went into effect in 1996, but the security situation remains volatile. Restrictions on Operation Lifeline Sudan (OLS) activities have made relief work difficult to carry out and many areas remain 'off limits'. Denial of flight clearance for transport of emergency food aid and lack of funds to cover transport and monitoring costs are among the major problems faced by OLS. Population displacement is a regular feature of life in Southern Sudan and, in recent years, there has been a switch from livestock to subsistence agriculture. This has meant that large sections of the population have become dependent on food aid when even small drops in production due to drought or pest attack occur.

When access has been granted so that nutritional surveys could be carried out, variable situations have been found. For example, surveys made in March 1996 in the Attar region (Upper Nile/Jonglei) showed 25% wasting with 5.8% severe wasting; in Mangalatore Camp wasting was measured at 16%. There has been a shift in policy in southern Sudan away from emergency relief provision and towards a more development-oriented approach. This move is intended to avoid creating dependency and to encourage greater self-sufficiency. In spite of the fact that levels of wasting were as high as in the early stages of the emergency (i.e. often 20% or higher, see Figure 18), general rations were targeted to certain groups or on a seasonal basis. There is some concern, therefore, that levels of wasting that would have triggered immediate food aid intervention in the early years of OLS are now no longer doing so.

Figure 18.: Levels of Wasting in Southern Sudan over Time (usually wt/ht <-2SD in children 6-59 months)

In 1997, successful advances made by the Sudanese People's Liberation Army (SPLA), along with rebel activity in northern Uganda and eastern Democratic Republic of Congo, have led to the spontaneous return of many refugees. Reports indicate that two of the largest Sudanese refugee camps, Koboko (with a population of 26,516) and Ikafe (with 55,162), had virtually emptied in March and that as many as 60,000 returnees arrived in Southern Sudan. A major area of returnee concentration is Yei, with estimates of up to 100,000 returnees and internally displaced people.

Red Sea State The food security situation in Sinkat and Tokar province in the Red Sea State markedly deteriorated in the four months towards the end of 1996, with food prices increasing dramatically. For example, the price of sorghum increased by 300% in the last six months of 1996. Price increases were largely due to a lack of rain and subsequent harvest failure. Livestock prices have been declining simultaneously as households have been selling animals. A recent survey showed a catastrophic nutritional situation for the approximately 240,000 people in the region, which has led to some population displacement.

Food security in the area has been declining for many years now as successive droughts have led to large numbers of livestock deaths which have significantly affected animal husbandry activities. Furthermore, employment opportunities and wages have been declining while the gradual reduction in food availability has affected traditional coping strategies such as community sharing.

The survey found levels of wasting among the displaced people around Sinkat at 47.8%, with 7.8% severe wasting. Results of a separate survey carried out among displaced people in Tokar are not yet available, but are likely to be as high as those seen in Sinkat. Surveys conducted in rural areas among the non-displaced populations near Sinkat showed 30.4% wasting, with 7.4% severe wasting. Micronutrient malnutrition was noted in both Sinkat and Tokar provinces. High levels of vitamin A deficiency and anaemia were seen. In addition, some cases of scurvy were noted as well as isolated cases of beriberi. Measles immunization coverage was low, at 30%.

A follow-up visit to the region after the survey revealed no change in the situation. Indeed, displacements to urban areas were continuing and increased population movements were expected. With the onset of winter rains, it was anticipated that the situation would further deteriorate.

Ethiopian and Eritrean Refugees Results from a census conducted in most refugee camps in Eastern Sudan in April 1996 showed that there are 15,000 assisted Ethiopian refugees and 133,000 assisted Eritrean refugees in Sudan. There are a further 270,000 unassisted refugees. There are also 4,400 assisted Chadian refugees in western Sudan. There are no new nutritional data on these refugee populations; the most recent survey information showed a situation that varied considerably between camps with wasting rates ranging from 7% to 15%. Repatriation of Ethiopian refugees from camps is almost completed, and the repatriation of refugees living outside the camps has begun. It is hoped that 100,000 Eritrean refugees will be repatriated in 1997.

Table 21.: Micronutrient Malnutrition

Deficiency	Year	Location	Public Health Problem*
Vitamin A (seen as night blindness)	1996	Camps for Displaced, Khartoum, Sudan	yes
	1996	Somalia (IDPs)	yes
	1996	Red Sea Hills, Sudan (IDPs and residents)	yes
	1994	Ethiopia - IDPs in Gode	yes
	1993	Ethiopia - IDPs in Gode	yes
Scurvy	1996	Red Sea Hills, Sudan (displaced and resident)	mild
	1996	Somali refugee camps, Kenya	moderate, but seasonal
	1996	Bhutanese refugee camps, Nepal	mild
	1995	Bhutanese refugee camps, Nepal	mild
	1994	Somali refugee camps, Kenya	moderate, but seasonal
	1994	Bhutanese refugee camps, Nepal	mild
	1994	Ethiopia - IDPs in Gode	mild
	1993	Ethiopia - IDPs in Gode	moderate
Beri-beri	1996	Bhutanese refugee camps, Nepal	mild
	1996	Red Sea Hills, Sudan (displaced and resident)	mild
	1995	Bhutanese refugee camps, Nepal	mild
	1994	Bhutanese refugee camps, Nepal	mild
Pellagra	1996	Mutarara district, Mozambique - returnees	mild
	1995	Mutarara district, Mozambique - returnees	mild
	1994	Bhutanese refugee camps, Nepal	mild
	1994	Camps for refugees from Rakhine State, Myanmar, in Bangladesh	mild

* See Annex 8 for guidance in interpretation.

Micronutrient malnutrition

Micronutrient deficiencies occur in refugee situations, frequently after a long period of residence in camps. Many of the deficiencies are also seen in the host population. However, prevention or treatment should normally be easier in a relatively controlled refugee situation where access to health care facilities is often better than in the host community. A major reason for micronutrient malnutrition is poor quality of the general ration. The food basket distributed often provides three commodities only, and fresh vegetables are rarely distributed. Efforts to prevent micronutrient malnutrition, therefore, focus on the distribution of a micronutrient-fortified blended food in the general ration, where populations are totally dependent on food aid and no fresh vegetables are included in the general ration. Vitamin A is also widely distributed.

Anaemia is generally widespread, particularly among young children and women of child-bearing age. However, marginalization of specific groups in camps can occasionally lead to unusually high rates of severe anaemia. This occurred recently in Kakuma camp in Kenya, where severe and life-threatening iron deficiency anaemia was found to be particularly prevalent amongst adolescent unaccompanied mates within the 8- to 20-year old age category.

Riboflavin deficiency, seen as angular stomatitis, is fairly widely seen in refugees camps. Effects of the deficiency on growth and well-being are not yet fully understood. However, presence of this deficiency is thought to be a general indicator of a possible more serious nutritional-problem, and should trigger further investigation into other possible deficiencies.

Data on other deficiency diseases, particularly beri-beri, scurvy, and vitamin A deficiencies, are summarised in Table 20. In some instances, these micronutrient deficiencies are seasonal in nature, such as scurvy which occurs in Somali camps in Kenya generally between September and December, or pellagra in Mozambique in the October-December period. In cases such as these, prevention seems relatively straightforward yet particularly in the case of the Kenyan camps, the operational agencies still appear to react to the occurrence of the deficiency rather than prevent it in the first place.

In other cases, micronutrient deficiencies seem to appear after a prolonged stay in a refugee camp, most likely indicating that sub-clinical deficiencies may have been present for a period of time and that food diversity has gradually diminished. This has been seen in Ethiopia and in some camps in Tanzania. These populations were dependent on food aid with little opportunity for cultivation and/or trading. Food aid rations were inadequate in micronutrients. Their diets were therefore consistently unvaried and lacking in micronutrients and gradually deteriorated due to a variety of economic and political factors.

A more perplexing situation has arisen in the camps for Bhutanese refugees in Nepal. This is a stable population. with a varied ration which includes fresh fruits and vegetables, along with a fortified blended food. Despite this, cases of scurvy, beri-beri and vitamin B2 deficiency (seen as angular stomatitis) continue to be reported. Possible explanations for this are discussed in the section on Asia.

Summary statistics

The following analysis is based on information about refugees and displaced people available to the ACC/SCN from September 1993 to July 1997. This information is provided by a number of NGOs and UN agencies involved in humanitarian relief and, while this database, which covers largely anthropometry, mortality and other indicators on the well-being of refugees and internally displaced people, provides an extensive body of information, it should not be considered as exhaustive. The following analysis may therefore not be extrapolated to all emergency situations and programmes.

Figure 19.: Reports of Kcals/person/day Received in Tanzania

Figure 20.: Reports of Kcals/person/day Received in Kenya

Each point represents a report on kcals distributed; the line represents assessed needs.

Figure 21.: Levels of Wasting (usually in children 6-59 months) Based on Anthropometric Survey Data Available to the ACC/SCN, Jan 1995-Jun 1997

There are other factors to consider in relation to food distributions. One is the timing of commodity distribution; it can happen that only one or two commodities are distributed over a certain time period and, while this might meet the caloric needs of a population, it is unlikely to meet their nutritional needs. There are, in turn, many possible reasons for this such as logistical and security problems that impede food deliveries. In addition, a lack of food aid pledges by government or delays in shipments are sometimes causes of reduced ration receipts. There is growing awareness that how food aid is delivered (i.e. the distribution system) has an impact on what is actually consumed by target beneficiaries, and that distribution within the family also plays a key role in target group distribution.

Levels of wasting, usually in children 6 months to 5 years old and generally expressed as weight/height <-2 Z scores below the median, are widely reported. In general, levels of wasting in a population under five that exceed 20% are considered to be very high and indicate a serious situation (see Box 4, p. 54).

Figure 22.: Crude Mortality Rates (CMR)

In Figure 21, each point represents the results of an anthropometric survey. Half of the reports available to the Refugee Nutrition Information System (RNIS) indicate levels of wasting that are over 10%. One in five of the reports indicates a serious situation demanding immediate intervention.

Worldwide, the crude mortality rate (CMR) is 0.25/10,000/day. A CMR of 1/10,000/day is considered to be alarming, and over 2/10,000/day a serious crisis.

Figure 23.: Wasting and Crude Mortality Rates (CMR) (log scale)

CMR

	<1	>1	Total
<10%	43	14	57
>10%	15	83	98
Total	58	97	155

sen=0.74
total=1.6
spec=0.85

CMR

	<1	>1	Total
<15%	50	31	81
>15%	8	66	74
Total	58	97	155

sen=.62
total=1.51
spec=.89

CMR

	<1	>1	Total
<20%	52	48	100
>20%	6	49	55
Total	58	97	155

sens=.52
total=1.41
spec=.89

Figure 22 illustrates the CMRs from Sub-Saharan Africa (excluding three points in Goma, Democratic Republic of Congo, of 51,40,34/10,000/day in August 1994 during a cholera epidemic). Over 30% of the records show a CMR of over 2/10,000/day, a level which is considered to indicate an emergency out of control.

Figure 23 plots, on a log scale, wasting and CMRs, again for information available on Sub-Saharan Africa. This figure shows the close association between wasting and mortality, which was also demonstrated in the earlier analysis on this dataset (ACC/SCN, 1994). The expected association is observed, and if we use the cut-off of CMR>1/10,000/day as indicative of a crisis, the best predictor of elevated mortality, based on this information, is 10% wasting (highest total sensitivity + specificity, and highest sensitivity i.e. true positives identified). This is shown in the calculations below Figure 23.

Some reports include information on measles immunization coverage. Refugees and IDPs are considered to be at very high risk of measles outbreaks. All of the main risk factors (crowding, overcrowded sleeping quarters, possible undernutrition and particularly vitamin A deficiency, and possible low immunization coverage) are likely to occur. Information on immunization coverage available to the ACC/SCN shows tremendous variations, indicating that ensuring adequate immunization coverage during emergencies continues to be problematic⁶. Due to the highly contagious nature of measles, immunization must remain as one of the highest priorities in refugee situations.

⁶ Relevant guidelines on the topic include: Conduite à tenir en cas d'épidémie de rougeole'. MSF 1996, and WHO Guidelines for epidemic preparedness and response to measles'.

Conclusions

Lessons Learned

The overall nutritional situation in most areas covered in this chapter has improved over the period under review with some exceptions - e.g., Ethiopia and Kenya. The information presented illustrates the increasingly complex environment in which humanitarian agencies must work. Access to those in need is often not possible at all or is only possible sporadically, as in Burundi, Liberia and Southern Sudan. Agencies are therefore constrained to work only when security permits, resulting in less than optimal services being provided to beneficiaries. Frequently, humanitarian aid agency personnel and those they are trying to help are placed at great risk during programme implementation. In some instances, beneficiaries may prefer one type of lower priority service provision, e.g., selective feeding, over another programme, e.g., general rations, as it places them at less risk while participating in the programme.

When continual access to populations is possible, levels of wasting and mortality are generally brought under control rapidly. However, it should be pointed out that extremely high rates of malnutrition (i.e. over 40%) are being seen, even in 'stable' situations such as in Ethiopia, Kenya, Liberia, Somalia and Sudan.

The analysis at the end of this chapter bears out what was shown in the Update on the Nutrition Situation, 1994 (ACC/SCN, 1994). High levels of wasting are a good predictor of elevated mortality (i.e. CMR above 1/10,000/day). Based on the data presented in this chapter, levels of wasting above 10% indicate elevated mortality. Ensuring adequate immunization coverage during emergencies continues to be problematic. While security constraints are often a primary cause of inadequate coverage, a variety of institutional and technical factors may also play a role and should be reviewed in order to determine whether improvements can be effected for future emergency immunization programmes.

There have been undoubted improvements in general ration planning and modalities which help to prevent micronutrient malnutrition; however, cases are still being seen. There are still many research and operational issues which need to be explored in order to ensure more effective preventive strategies. One such initiative is a study being undertaken to determine the use and acceptability of fortified blended foods at the household level. Other initiatives are looking into the feasibility of fortifying cereals for distribution in the general ration.

Issues for Further Consideration

There are several other management and operational issues which have emerged over the past two years in the context of the emergency programmes described in this chapter, some of which are beginning to be addressed while others require further consideration.

One pressing issue is the need for humanitarian agencies to acquire better information about the food security of the beneficiary populations which they are assisting. The need for this was shown clearly in the Great Lakes region during 1995 and in Liberia and southern Sudan over the past few years. In the case of the Zairian refugee camps, the unplanned reduction in rations which occurred at the start of 1995 was met with outcries by the international aid community which feared widespread starvation and mortality if the general ration was not improved. No such improvements occurred yet the nutritional status of the refugee population remained stable and adequate. Subsequent food security assessments showed that this occurred because refugees had evolved coping strategies which permitted considerable supplementation of the general ration. These strategies only began to be threatened as the government imposed increasing restrictions on refugees' economic activities.

In Liberia, policies of phasing out general ration provision in order to promote self-reliance and avoid dependency were applied too generally and should have been better targeted on the basis of differential ability to adopt coping strategies. Once again, improved assessment of socio-economic circumstances would have led to better informed decisions. The need to rapidly acquire better information on the food economy and coping mechanisms of refugees is being partly addressed by the use of an assessment methodology developed by Save the Children Fund (UK). This methodology is being increasingly used after the initial, acute phase of an emergency and is becoming a regular precursor to assessment missions to set ration levels.

Issues also arise over the role and context for emergency selective feeding programmes. For example, while emergency selective feeding programmes were felt to be highly appropriate at the start of the Liberia emergency, when levels of wasting were extremely high in many parts of the country, the effectiveness of these programmes began to be questioned as time went on. General ration deliveries were frequently inadequate so that many children were re-admitted and default rates were high. It may well have been that resources devoted to 'curative' selective feeding programmes could have been better allocated to supporting the preventive general ration programme.

Another issue that emerged during the 1995-96 period was the utility of adopting a flexible approach to establishing food distribution systems. The experiences in the Great Lakes region during 1994, when commune-headed distribution systems were established in refugee and IDP camps in the former Zaire, Tanzania and Rwanda, showed how such systems could lead to highly inequitable general ration allocations with alarming effects on the nutritional status of camp populations. However, towards the end of 1994 and throughout 1995, agency distribution mechanisms were adapted and evolved into "fairer" systems which proved far more acceptable to beneficiaries. The lessons learned in establishing these systems need to be disseminated and institutionalized in order that agencies can adopt best practice in future emergencies.

In addition to the above issues, a set of ethical considerations also emerges from the experiences of the past two years. These ethical issues need to be examined and thought through ahead of future emergency programmes which may pose similar dilemmas. For example, reluctance shown by some governments to support the largely Hutu population in Eastern DRC manifested itself in a gradual reduction in food aid pledges and raises issues of human rights as well as challenging concepts of 'voluntary repatriation'. It is unclear to what extent moral judgements, regarding the acceptability of supporting a population with known involvement in the 1994 genocide, or foreign policies of 'encouraging' repatriation to Rwanda, were behind such behaviour. Government rationales appeared to be implicit in actions rather than explicitly stated as foreign policy aims or objectives.

Figure 24.: UNICEF Conceptual Framework of the Causes of Malnutrition

The concept of voluntary repatriation was challenged in Tanzania when the Rwandan refugees went home. The decision that the refugees should return was taken by the Government of Tanzania, and it was reinforced by a military presence as the deadline for repatriation approached. Similarly, there are questions around the voluntary nature of the repatriation of refugees from Bangladesh back to Rakhine State, Myanmar.

Another ethical issue arises with regard to international humanitarian agency support for incumbent governments such as the Taliban in Afghanistan, where such governments openly discriminate against, or exclude, sections of the population (women) who require assistance from external agencies.

A further emerging issue concerns the tendency of the humanitarian aid community during long-term emergency programmes to move away from provision of emergency aid towards more 'development-oriented' emergency aid support. This targeted support (i.e. seasonal rations, food-for-work) is instituted in the name of discouraging dependency and encouraging self-sufficiency. However, a possible offshoot of this policy has been a potential lowering of acceptable standards of nutrition in an effort to accommodate development thinking, with the effect that levels of malnutrition that would have triggered emergency interventions at the start of a crisis are now seen as somehow normal or acceptable. This phenomenon has been identified in Southern Sudan.

A final issue that needs to be addressed is the extent to which it is valid to compare the nutritional status of refugee and internally displaced populations with that of host populations in order to assess the efficacy of an emergency intervention. For example, while it is often true that a host population has some endemic form of micronutrient deficiency disease this does not necessarily mean that a certain prevalence is acceptable within the refugee or IDP population. Indeed, the increased risk run by the refugees due to factors such as crowded conditions and inadequate sanitary facilities, often associated with camp conditions, means that a compromised nutritional status would have more profound and far-reaching effects. This type of comparison is often implicitly or explicitly made in assessments of emergency-affected populations.

If we accept the concept that refugees have a right to the best possible nutrition, this comparison is not useful. It is more relevant to look at the underlying causes of malnutrition (i.e. food security, caring practices and adequate health care, as set out by the UNICEF model, reproduced as Figure 24) so that the causes may be addressed both among the refugee and host populations.

One way to move forward on this issue may be for the international aid community to define a minimum standard of humanitarian aid provision for emergency-affected populations, as is being undertaken by a project spear-headed by the Steering Committee on Humanitarian Response. In the food and nutrition sector this might entail, for example, promoting best practice in infant feeding or nutrition education for the use of blended foods, a commodity with which many refugees are unfamiliar.

It is apparent that policies need to be formulated on many of these issues. Clarification and wide acceptance of such policies can help us to move another step towards improving emergency response performance and the overall well-being of refugee and internally displaced populations.

References

ACC/SCN (1994/7) Refugee Nutrition Information System (RNIS) Reports #3-20. ACC/SCN, Geneva.

ACC/SCN (1994) Update on the Nutrition Situation, 1994. ACC/SCN, Geneva.

FAO/WFP (1997) Crop and Food Supply Assessment Mission to Rwanda, Rome.

FSAU (1997) Assessment in Somalia, Nairobi.

James WPT and Schofield C (1990) Human Energy Requirements. FAO/OUP.

SCF(UK) (1997) Household Food Economy Assessment of Khudunabari and Beldangi Refugee Camps Jhapa District, South-East Nepal, London.

Schofield C and Mason J (1994) Evaluating Energy Adequacy of Rations Provided to Refugees and Displaced Persons. Paper prepared for Workshop on the Improvement of the Nutrition of Refugees and Displaced People in Africa, Machakos, Kenya, 5-7 December 1994. ACC/SCN, Geneva.

UNHCR (1996) UNHCR by Numbers, Geneva.

UNHCR (1996) Populations of Concern to UNHCR: A Statistical Overview, Geneva.

UNICEF (1996) Atlas of South Asian Children and Women. Nepal.

UNICEF (1997) State of the World's Children. UNICEF, New York.

US Committee for Refugees (1995) World Refugee Survey. Washington DC.

US Committee for Refugees (1996) World Refugee Survey. Washington DC.

US Committee for Refugees (1997) World Refugee Survey. Washington DC.

WFP (1997) 1997 Estimated Food Needs and Shortfalls for WFP-assisted: Emergency and Protracted Relief Operations, Development Projects, Special Operations. World Food Programme Rome.

WHO (1993) The world-wide magnitude of protein-energy malnutrition: an overview from the WHO Global Database on Child Growth, Bulletin of the World Health Organization, 71(6)703-712.

WHO (1996) Management of Nutrition in Major Emergencies. (in preparation).

WHO (1995) Global Prevalence of Vitamin A Deficiency, Geneva.

WHO (1995) Bulletin of the World Health Organization, 1995,73(5): 673-680.

World Bank (1997) World Development Indicators, Washington, DC.