Targeted Food Distribution to Women and Children in Northern Afghanistan

By Regine Kopplow, Concern Worldwide

Regine Kopplow is a senior nutrition advisor with Concern, working in Afghanistan since April 2002. Previous field experiences include Namibia and Sierra Leone.

This article describes the impact of a food distribution programme targeting households of malnourished women and children in northern Afghanistan. The observations reflect the challenges of anthropometric measurement in this population, and also raise questions over the efficacy of short-term BP5 supplementation programmes. The nutritional vulnerability of Afghan women is also highlighted¹.

Concern Worldwide has been operational in northeast Afghanistan since 1998, following an intervention in response to the earthquakes that struck the northern part of the country. With a landscape of high mountains and narrow lush plains, north-eastern Afghanistan is an extremely remote area and access to many villages is only possible on foot or by donkey/horse. Poverty is a major issue. Water for irrigation is scarce, access to markets severely limited, and almost all the population are involved in subsistence agriculture.

Between 1999 and 2001, Concern assistance targeted rural water supply, health education, and shelter rehabilitation. In the second half of 2001, Concern began to assist and support IDP (internally displaced persons) settlements in Takhar province and extended its FFW (food for work) infrastructure projects to Baghlan province. By 2002, Concern began to adopt a longerterm strategy, reflected in the programming approach in Rustaq district, an area with a population of 167,455 people. Here, the many villages surrounding the relatively large market town of Rustaq are hampered by poor land consisting mainly of rock formations, with little arable potential. In Rustaq district, community based organisations were established, through which Concern began to address food security needs. The early FFW and FoodAC (food for asset creation) projects were replaced by community based agricultural activities such as seed distributions, seed banks and livestock vaccination, water projects to provide safe drinking water and irrigation systems, as well as infrastructure projects to construct roads, bridges and dams with a voluntary community component.

Nutrition and food security

Following three years of drought, a food security and nutritional assessment by Concern in December 2001 showed alarming results. Although the sample size was small (100 households in 7 villages), it indicated that there were pockets of great need that required immediate intervention to prevent death, displacement and destitution. Women seemed particularly affected. In the surveyed villages, 36.5% of the females measured, had a MUAC (mid-upper arm circumference) <215mm, indicating chronic severe malnutrition, and a further 27% had a MUAC <230mm, suggesting a high risk of becoming malnourished. In comparison, malnutrition rates (MUAC <125mm) for children aged 1 to 5 years were 10.1%. Meanwhile, the 2001 WFP VAM² report suggested that 60% of the population in Rustaq district were drought affected and required food assistance until the end of June 2002.

In response, Concern intervened to assist and support malnourished vulnerable families in Rustaq district. Aiming to meet immediate basic food needs, the project objectives³ included provision of:

• a balanced food basket for the malnourished to improve their nutritional status before/over winter
• fortified food (BP5) to malnourished children and severely malnourished women to improve their nutritional status before/over winter.

Targeting

One of the project villages, 300 families, 100% rain-fed agriculture on land 1700m above sea level, 15km away from the district capital with it's bazaar and health facilities. Saqawa. September 2002

The project operated in two main phases. Phase One (August to December 2002) targeted 12 villages, while Phase Two (December 2002 to April 2003) targeted a further 16 villages. The 28 targeted villages were located between 5 and 35km from Rustaq town. The average population was 122 households per village, of which 9% constituted vulnerable groups, e.g. elders, disabled, unaccompanied children, returnees or female-headed households. Overall, 24 of the 28 villages were wholly dependent on rain-fed agriculture.

In the villages, all women aged 15 years and older and children aged between 6 and 59 months were assessed using MUAC. In order to facilitate nutritional monitoring and determine programme impact, all registered beneficiaries were later assessed using Body Mass Index (BMI)⁴ for women and weight-for-height % of the median (WH) for children.

All households with at least one person fulfilling MUAC criteria (women £220mm and children £124mm) were issued with a ration card. In total, 2734 women and 2127 children under 5 years were screened and 1294 households qualified to receive a monthly dry food ration over a period of 5 months. Based on an average family size of 6 members and given many years of war and drought, the recent earthquake, exhausted coping mechanisms and empty food stores, the aim was to supply 100% of calorie need (2100kcal/day/person)^5,6. Monthly famil rations consisting of 45kg wheat, 45kg rice, 10kg dried beans and 5kg vegetable oil, provided an individual daily intake of 2156 kcal, 60.5g protein and 33.7g fat.

Boy on a donkey in Saqawa village, the normal transport facility, Saqawa. September 2002

Over a four-week period, BP5 biscuits (4 bars = 1000kcal) were distributed to selected children in Phase One and Two and women in Phase Two only, in addition to the monthly household ration. For children in Phase One, entry was based on MUAC only (?124mm) while Phase Two entry included weight-for-height criteria (MUAC ?124mm and < 80% WH). Entry for women in Phase Two was based on MUAC and BMI criteria (MUAC< 185, and/or BMI ?16⁷).

The nutrition team consisted of two female nurses, one female translator, two male nutrition workers, one driver and an international female nutritionist.

MUAC screening

MUAC cut-off points were identified after a thorough review of nutrition surveys conducted in Afghanistan by different agencies, and crosschecking during screenings at the start of the project. MUAC, BMI and WH (% of the median) criteria used in screening and nutritional surveillance, are outlined in table 1.

Village MUAC screening of women and children identified rates of malnutrition necessitating immediate intervention. From 2127 children screened, 12.9% were moderately malnourished and 2.3% severely malnourished (15.1% global acute malnutrition). Malnutrition rates were not evenly distributed in children, most notably,

• the prevalence of global acute malnutrition was highest in children aged 12 to 23 months (37.2%) followed by children aged 24 to 35 months (28.4%).
• under 48 months of age, girls had higher rates of global acute malnutrition than boys.
• of those who were classified as severely malnourished (n=48), 39% comprised infants under 1 year of age.

From a population of 2734 women screened, moderate malnutrition rates were particularly high (39.6%). The overall prevalence of severe malnutrition was 1.0%. Women aged between 20 to 29 years had the highest prevalence of both moderate (38%) and severe malnutrition (0.9%). Among malnourished women aged 20-29 years (MUAC £220mm), 17% were pregnant and 41.5% breastfeeding. Meanwhile, half (50%) of those aged 15-19 years were pregnant (26.9%) and/or breastfeeding (23.1%).

Weight-for-height (WH) and body mass index (BMI)

Amongst children with a MUAC £124, the vast majority (85%) had a WH ?80%. Similarly, over half (64%) of the women classified as malnourished using MUAC had a normal/overweight BMI (?18.5). Using BMI, only 31% of women registered in the programme were classified as moderately malnourished (BMI ?16<18.5)⁸. Conversely, the proportion classified as severely malnourished (BMI <16) increased. Compared to women with a normal BMI (?18.5 <25) and MUAC ?220, the severely malnourished group were older (39y versus 34y), taller (1.55m versus 1.52m) and had an average BMI of 14.98, and MUAC of 192mm (BMI 20.22, MUAC 209 in normal BMI group).

Impact of monthly food ration

After five months of food distribution, all Phase One beneficiaries were reassessed using BMI (women) or WH (children). Prevalence of moderate and severe acute malnutrition fell to acceptable levels amongst children (see table 2). However global acute malnutrition rates remained unacceptably high for women (28%), while severe malnutrition rates halved but remained elevated (2.4%), (see table 3). It should be noted that the statistical significance of these findings was not tested.

Impact of BP5 distribution

BP5 biscuits were distributed to 207 children and 37 women over a four week period. All women and children receiving BP5 were weighed before, weekly during, and 6 weeks after the intervention. During the distribution, the nutritional status of both women and children improved. However amongst children, nutritional status declined once supplementation ceased (see figure 1 ). Since different entrance criteria were used for the children in Phase One and Two, children entering Phase Two had a lower starting WH than those in Phase One. Children in Phase Two, with an average WH of 77.5%, showed a very rapid response to the 4 weeks of BP5 supplementation, reaching a higher WH than Phase One (starting WH 88.4%). However, six weeks after the last BP5 ration, the Phase Two WH had dropped to 86.9%.

Since only women in Phase Two received the BP5 supplement, it was possible to compare their improvement in nutritional status between Phase One and Two. With BP5 supplementation, the mean body weight increased by an average 2.7kg (36.6kg to 39.3kg) and the improvements were maintained 6 weeks following cessation of supplementation (see figure 2). However, the same gain was found in the group without BP5 supplementation (36.1kg to 38.8kg). It should be noted that participation in the final weight monitoring was only 50% for the non-BP5 supplemented group.

Women's nutritional vulnerability

The limited impact of the interventions on women can, at least in part, be explained by those longer-term factors underpinning nutritional vulnerability of Afghan women. The cultural and social consumption patterns within the household do not favour women in terms of dietary quantity or quality. Culturally, it is less acceptable for women to move out of the confines of the village than men. Consequently, there are fewer opportunities for women to improve the quality of their diet, through purchasing or consuming vitamin/protein rich foods during visits to the market, for example. Similarly, access to health systems is reduced by this restricted movement. There is also poorer access to health education at a time when there appears to be considerably less transfer of inter-generational knowledge, especially in the rural areas. Finally, pregnancy patterns suggest insufficient time for nutritional recovery between births.

Conclusions, recommendations and lessons learnt

Girls in one of the project villages observing sceptically the screening process. Rustaq/north-east Afghanistan. March 2003

A discrepancy was found between estimated malnutrition levels using different indicators. Amongst children, the vast majority who had been classified as acutely malnourished using MUAC, did not fulfil weight-for-height criteria. Only one-third of the women deemed moderately malnourished using MUAC, were classified as such using BMI criteria. On the contrary, the prevalence of severe malnutrition in women was higher using BMI criteria. Women, overall, appeared 'unusually' shorter than men. While we have no clear explanation for these observations, contributing factors may include phenotype, as well as chronic malnutrition and many early pregnancies influencing female development and body stature. Also, the inclusion of adolescents (10-19 years as defined by WHO) in our comparison, and the lack of standard BMI cutoff points for this group, may influence interpretation of population-based figures. Ultimately, these findings highlight that MUAC cut-off points for malnutrition need to be further reviewed in Afghanistan. Further research may be needed on the Cormic Index¹¹ to obtain more reliable anthropometric reference data of Afghan adults.

BP5 is a compact energy rich food with easy storing, transporting, distribution and preparation character. The acceptability is high. However in our intervention, the full weight gains were not sustained in children. Amongst women, weight gains were sustained although these were similar to those who had not been supplemented. Though interpretation of responses amongst women is limited due to incomplete data, it would be valuable to investigate how, and the extent to which, the factors contributing to the long-term nutritional vulnerability of Afghan women impacted the effectiveness of BP5 distribution. The experience also illustrates the need for clear education and training on the rationale, use and constraints of BP5, for both the beneficiary and the community.

In some of the targeted villages, Concern has been implementing community-based interventions that are at risk of being undermined by free food distributions. The use of high-energy BP5 with a likely quick impact strengthens this tendency. Even when beneficiaries receive the necessary training to understand the concept of the intervention, non-beneficiaries only see the quick nutritional improvement. Furthermore, it is common among staff and beneficiaries to call the BP5's "biscuits". This engenders a sense that normal biscuits are appropriate foods for children and endorsed by clinics and NGO's. In many cases this led to severely malnourished children being exclusively fed with normal biscuits. In emergency situations, BP5 seems to be an appropriate supplementary food for shortterm interventions, showing a quick impact on malnourished children. However, to reduce the chances of deterioration following the intervention, it is vital both to comprehend and concurrently address the underlying causes of malnutrition in the intervention community.

Our findings highlight that nutrition surveys in Afghanistan should include women. Focusing exclusively on children, and even pregnant women in the third trimester does not reflect the entire dimension of malnutrition in the country. Considering the prevalence of early and frequent pregnancy and the potential implications for maternal and infant nutritional status, addressing the needs of Afghan women is all the more critical.

For further information, contact Regine Kopplow at email: Regine.Kopplow@gmx.de, or afghanistan@concern.ie

¹Emergency Complementary Food Supplies To Drought Affected Vulnerable Populations in Afghanistan, Ref: 605/11285/CON 08 02, 17 April 2002-30 April 2003, Concern Worldwide

²World Food Programme/Vulnerability Analysis Mapping

³The project had other key objectives and activities relating to health and nutrition education, and hygiene which are not included here but are detailed in the main report (see footnote 1)

⁴Body Mass Index (BMI) is calculated as weight (kg) divided by height squared (metres)

⁵This meets Sphere standard on food aid requirements.

⁶In effect, this meant that the project provided a general, rather than complementary, food ration.

⁷Slightly different BMI cut-off points were used for severe malnutrition in women for surveillance (BMI<16) compared to BP5 distribution (BMI?16). As various standards exist, a cut-off point was accepted at the beginning of the project, which was later revised as the project progressed.

⁸All pregnant women (12%) and those breastfeeding infants under six months (7.8%) were excluded from this comparative analysis. Thus, 219 women (19.8%) of the total group (n=1110) were not included. Of the remaining 891 eligible for measurement, 808 women were available and BMI calculated.

⁹MUAC prevalence rates refer to the total population, while weight-for-height figures refer to the sub-group of the population with MUAC ?124.

¹⁰BMI prevalence rates refer to the sub-group of the female population with a MUAC ? 220 and excludes women pregnant and/or lactating (infant under one year). See footnote 8.

¹¹The Cormic Index assesses the relative contribution of the trunk and legs to stature. It is calculated as the ratio of sitting height (SH) to standing height (H) SH/H and expressed as a percentage. A means of standardising BMI using Cormic Index has been proposed, for both males (BMI =0.78(SH/S)-18.43) and females (BMI=1.19(SH/H) - 40.34), and detailed in the RNIS supplement, Assessment of nutrition status in emergency affected populations, Collins S, Duffield A, Myatt M, July 2000. See online at http://www.validinternational.org/tbx/docs/ACF88.pdf