Trends in nutritional status in Egypt, 1985 to 2005^[4]

Nutritional status based on anthropometry

Low birth weight

Low birth weight (LBW) is defined as weight at birth of less than 2 500 g, irrespective of gestational age. Prematurity (infants born before 37 weeks gestation) and intrauterine growth retardation (IUGR) are the two main causes of LBW: most LBW in developing countries is the result of IUGR, while in industrialized countries it is caused by pre-term birth. Most of the studies of LBW in Egypt are small-scale or depend on hospital data. The first national study was carried out in 1995 to 1997 and revealed a LBW rate of 12.6 percent of live births (El-Sahn, 2004). This is lower than the average estimated rates of 18 percent for the East Mediterranean Region (EMR) and 15 percent for the Middle East and North Africa. Among countries in the EMR, Egypt’s LBW prevalence is lower than the estimated rates in Yemen (32 percent) and the Sudan (31 percent), but higher than those in the Syrian Arab Republic and Lebanon (6 percent) (UNICEF, 2003). It is also higher than the WHO estimate for Egypt (10 percent) and than most of those reported in other surveys (Figure 14). LBW is a significant public health problem in Egypt, and requires considerable attention (El-Sahn, 2004).

FIGURE 14
Trends in prevalence of LBW in Egypt, 1972 to 1997

Source: El-Sahn, 2004.

Children under five years of age - Undernutrition

WHO recommends that evaluations of nutritional status be based on three indices: weight-for-age, weight-for-height and height-for-age, using Z-scores in comparison with the reference population median (WHO, 1983; 1995). The reference population of well-nourished children used by most surveys is the international reference population of the National Center for Health Statistics (NCHS), accepted by WHO as the international growth reference.

Since 1978, Egypt has been experiencing decreasing trends in the prevalence of stunted, wasted and underweight children. Between 1978 and 2004, stunting decreased from 40 to 15.8 percent, and underweight from 20.6 to 18 percent. These time series trends should be regarded cautiously as the age range of the children is not consistent across all surveys. Using more recent data from EDHS surveys conducted in 1995 and 2003 (EIDHS, 2003) on children up to 59 months of age, there have been declines in stunting (from 29.8 to 15.6 percent), underweight (12.4 to 8.6 percent) and wasting (4.6 to 4.0 percent) (Figure 15). Boys have a slightly higher prevalence for all indicators. In EDHS 1995, the percentage stunted varied from 18 percent in urban governorates to 40 percent in rural Upper Egypt, while in EDHS 2000 it ranged from 9 percent in urban governorates to 27 percent in rural Upper Egypt. This urban/rural variation is mostly caused by variation in development indicators, particularly education. The general improvement in nutritional status can mainly be attributed to improving socio-economic status, which is reflected in improved health care and quality of food consumed. Nutrition education programmes directed at mothers are contributing factors and are discussed in a later section of this case study.

FIGURE 15
Trends in prevalence of undernutrition among children under five years of age, 1992 to 2003

Source: EIDHS, 2003.

Children under five years of age - Overnutrition

Shaheen, Hathout and Tawfik (2004) conducted a national survey to assess prevalence of obesity in children under five years of age. The survey covered a sample of nearly 4 154 children (2 165 males and 1 969 females) in eight governorates - Cairo, Gharbia, Quena, Beni-Suef, Beheira, Suez, Matrouh and El-Wadi El-Gadid - representing six geographic areas: metropolitan, Lower Egypt, Upper Egypt, coastal, canal and frontier. It revealed that almost 8 percent of preschool children were wasted, 3.6 percent were overweight and 2 percent were obese. Obesity was more prevalent in girls (2.6 percent) than boys (1.5 percent). Frontier governorates had the lowest proportions of obese preschool children, followed by the coastal and canal regions. Metropolitan governorates and Lower Egypt had higher proportions of obese preschool children. This could be explained by changes in dietary habits leading to more energy-dense fast food and beverages with high sugar content. Although overweight among children under five years of age cannot yet be considered a public health problem in Egypt at this time, the trend is clearly toward increasing prevalence.

School-age children and adolescents (six to 18 years)

Moussa (1989) reported on the growth of school-age children using data obtained during the Health Examination Survey (HES) of the Health Profile of Egypt (HPE). The sample included at total of 6 004 school-age children (3 119 boys and 2 885 girls) aged six to 18 years from different governorates. The mean weight of boys aged six to eight years is just below the WHO reference mean; among boys of 11 to 18 years it deviates down to lie almost midway between the reference mean and 1 SD below it. The mean weight of girls is close to the reference mean at six years of age, deviates down until the age of 11 - when it is almost 1 SD below the standard mean - then improves and approaches the reference, reaching its closest point at 16 years of age before continuing below the reference mean until 18 years of age. The weights of girls are therefore better than those of boys in the six to 18 years age group. On the other hand, the curves representing mean height for boys and girls are both located below the reference mean, at close to -2SD. Boys show somewhat more relaxation in linear growth than girls, indicating chronic undernutrition.

Between 1998 and 2004, the prevalence of stunting among schoolchildren was essentially stable: 14.5 percent in 1998 and 13.2 percent in 2004. Wasting decreased from 7.8 to 4.1 percent, but underweight increased from 6.65 to 8.8 percent. Thus, the situation with regard to this age group does not seem to be improving noticeably. Underweight and stunting were more prevalent in rural than urban areas, but wasting was more prevalent in urban areas (Hassan et al., 1998). Prevalence of overweight and obesity was slightly higher among school-age than preschool children, so the problem appears to be emerging rapidly in children (Shaheen, Hathout and Tawfik, 2004).

Weight status based on percentile body mass index for age (PBMI) (WHO, 1995) and prevalence of obesity were studied among 6 190 adolescents aged 12 to 19 years (Shaheen, Hathout and Tawfik, 2004). Results revealed that the prevalence of overweight is twice that of obesity among both male and female adolescents. For males the figures are 10.6 percent overweight and 5.8 percent obesity, and for females they are 19.9 and 9.7, respectively. Obesity was observed to be more prevalent in metropolitan areas (Cairo). Results are illustrated in Table 4. The trend is for prevalence of underweight to decrease and overweight and obesity to increase among female adolescents. Clearly, overweight and obesity (combined) are prevalent among Egyptian adolescents of both sexes, and at least for girls the prevalence has increased in the last few years.

TABLE 4
Anthropometric data on adolescents

Adults

Several studies obtained information on the heights and weights of female adults as shown in Table 5. Underweight, defined as BMI < 18.5, which denotes chronic energy deficiency (WHO, 1995), is almost non-existent among the adult female population. Prevalence of chronic energy deficiency fell from 3.4 percent in 1995 (Moussa, El-Nehry and Abdel Galil, 1995) to 0.4 percent in 2004 (Hassan et al., 2004). However, there is a trend for increasing prevalence of overweight and obesity among female adults in Egypt. From the earliest available survey year (1995) to the latest (2004), overweight among women decreased from 31.3 to 26.9 percent, while obesity increased from 20.5 to 48.2 percent. Overweight and obesity prevalence among adults in Egypt is among the highest in the world, and there is evidence that the prevalence is still increasing, at least in women. There is a trend of increasing prevalence of overweight and obesity among female adults in Egypt (Figure 16). Between 1995 and the latest survey year 2004, overweight decreased from 31.3 to 26.9 percent among women, while, obesity has increased from 20.5 to 48.2 percent. Among urban women, data collected on large samples indicated that by 1998 the average BMI for women was in the obese range (30.08) (Galal, 2000). Furthermore, almost 5 percent of women were classified as "severely obese", i.e., BMI > 40. Men have been less studied, but 1998/1999 survey data indicated that 65.3 percent of urban Egyptian men and 34.1 percent of those in rural areas were overweight or obese. These trends can likely be attributed to changes in dietary habits towards higher consumption of energy-dense foods, together with a tendency to sedentary lifestyle particularly in urban areas. Continuous snacking between meals among housewives has been mentioned as another causal factor (Galal et al., 1987). Hussein, Moussa and Shaheen (1993) demonstrate that obesity is a problem in both privileged and underprivileged areas. However, the co-morbidities of obesity are more common among the privileged, which is attributed to the quality of diet. The main source of energy among the rich is animal fat (meat and pastries), while among the poor it is carbohydrate (bread and sugar in tea) and vegetable oil (fried vegetables and tubers). Efforts to reduce weight were more common among the rich than the poor (Hussein, Moussa and Shaheen, 1993).

In the national survey by Shaheen, Hathout and Tawfik (2004), a nationally representative sample of 19 021 adults (aged 20 years and over) was used to assess the prevalence of obesity in Egypt. Results of this study revealed that women had higher overall rates of obesity than men (48.2 and 18.7 percent, respectively), although men had higher rates of overweight (34.5 percent) than women (26.9 percent). Frontier governorates and Upper Egypt had the lowest proportions of overweight and obesity. However, Lower Egypt and the metropolitan region, followed by the canal and coastal regions, had the highest percentages of overweight and obesity. This is also documented by Moussa, El-Nehry and Abdel Galil (1995), who record the highest rates of overweight and obesity in Cairo (70 percent) and the lowest in Upper Egypt (39 percent). Urban areas had higher rates of overweight and obesity than rural ones.

According to Shaheen, Hathout and Tawfik (2004), the prevalence of overweight and obesity among adults aged 20 to 80 years differs according to age. The lowest proportions of overweight and obesity were among the 20 to 30 years age group (27.8 and 8.1 percent, respectively). Prevalence gradually increased with age to reach a peak between the ages of 50 and 60 years, when overweight and obesity among men were 37.2 and 29 percent, respectively, and among women 21.8 and 66.1 percent. After the age of 60 years, the prevalence of overweight and obesity decreased slightly among men; among women obesity decreased, but overweight increased, to reach 35.7 percent at 80 years of age. These findings reflect the behaviour of the population, as consumption of extra quantities of food is usual at younger ages, and low levels of physical activity increase with age.

Although overweight and obesity are still increasing in prevalence in Egypt, and the problem is now receiving attention owing to the global emergence of obesity as a public health problem, the phenomenon has been evident in Egypt for at least 20 years. The 1981 national food consumption survey included measurements of the mothers and fathers of sampled children and reported 63.1 percent of mothers and 14.5 percent of fathers were overweight or obese (i.e., > 110 percent of the standard weight at that time) (Galal, 2000). Thus recent trends toward urban living and an abundant food supply do not by themselves totally explain the phenomenon in Egypt. However, there are many physical and cultural barriers to a physically active lifestyle in Egypt, and there is significant opportunity for the development of effective health promotion programmes to encourage physical activity.

TABLE 5
Anthropometric data on adults

FIGURE 16
Trends in overweight and obesity among females, 1995 to 2004

* Women 15 to 49 years in EDHS surveys
+ Women ³ 20 years in Shaheen et al., 2004

Central obesity among adults

Alternative methods to the measurement of BMI are valuable in identifying individuals at increased risk from obesity-related illness owing to abdominal fat accumulation (WHO, 2000). A high waist-to-hip ratio (WHR) (³ 1 in men and ³ 0.85 in women) has been accepted as the clinical method of identifying patients with abdominal fat accumulation. Findings of Shaheen, Hathout and Tawfik (2004) demonstrate that abdominal obesity (high WHR) exists among 20.8 percent of men and 45.3 percent of women. There were significant differences (p = 0.000) in the prevalence of abdominal obesity between men and women in the total sample and between urban and rural areas (p = 0.000), with 41.6 percent of adults in urban and 20.4 percent in rural areas affected. There were also significant differences (p = 0.000) in prevalence of high WHR among different governorates of Egypt.

Recent evidence suggests that waist circumference alone - measured at the midpoint between the lower border of the rib cage and the iliac crest - provides a more practical correlate of abdominal fat distribution and associated ill health (WHO, 2000). This is an approximate index of intra-abdominal fat and total body fat. Furthermore, changes in waist circumference, mainly associated with overweight and class 1 obesity, reflect changes in risk factors for cardiovascular disease (CVD) and other forms of chronic disease. Waist circumferences ³ 88 cm in women and ³ 102 cm in men are considered above normal (WHO, 2000).

Results of Shaheen, Hathout and Tawfik (2004) revealed that there was correlation between waist circumference levels and overweight and class 1 obesity in the total sample and in the studied governorates (p = 0.000). Nearly three-quarters of those with class I obesity had central obesity (high waist circumference), except in Beni-Suef where the figure was even higher, at 85 percent. The majority of class 1 obese women (at least 80 percent) in the study governorates and the total sample had high waist circumference, while about one-third of overweight females had high waist circumference. The percentages were lower among men than women.

Stunting and obesity

Many studies document the relation between stunting (< -2 SD height for age) as an indicator for long-standing chronic malnutrition and overweight or obesity (> +2 SD weight for height) due to inadequate intervention programmes early in childhood. There is considerable evidence, mostly from developed countries, that intrauterine growth retardation is associated with an increased risk of coronary heart disease (CHD), stroke, diabetes and raised blood pressure.

Results of Shaheen, Hathout and Tawfik (2004) revealed that 2.3 percent of stunted male and 3.7 percent of stunted female preschool children (aged two to six years) were obese; this is nearly double the prevalence of obesity in children of normal height (1.2 percent of males and 2.1 percent of females). About 6 percent of stunted male and female preschool children were overweight, compared with 2.3 percent of male and 2.9 percent of female normal-height children. This finding also holds true for school-age children: 6.1 percent of stunted male and 5.7 percent of stunted female children were obese, compared with 1.9 and 3.4 percent, respectively, for male and female children of normal height. Nearly 6 percent of stunted male and 8 percent of stunted female school-age children were overweight, while the respective percentage among normal-height school-age children was 3.8 percent for both males and females. There were significant differences in the weight status of both preschool and school-age male and female children depending on their stature levels (p = 0.000).

There was no significant difference in the weight status of either male or female adolescents (12 to 18 years) between the different stature levels. About 5 percent of stunted male and 9.2 percent of stunted female adolescents were obese, compared with 5.3 and 8.5 percent of normal-height male and female adolescents, respectively.

Micronutrient status

Iron deficiency anaemia (IDA)

Table 6 presents a summary of results from different studies on IDA in Egypt. A national survey to assess vitamin A status (Moussa, El-Nehry and Abdel Galil, 1995) recorded haematocrit values for 1 623 preschool children (aged six months to six years) and 762 mothers. Preschool children and pregnant mothers with haematocrit values equal to or less than 33 (11 gm Hb/100 ml) were considered anaemic, as were non-pregnant mothers with haematocrit values equal to or less than 36 (12 gm Hb/100 ml) (WHO, 1989).

EDHS 2000 included direct measurement of haemoglobin levels in a sub-sample of half of all EDHS households for three groups: women aged 15 to 49 years who were or who had been married; children under six years of age; and boys and girls aged 11 to 19 years. Anaemia is classified as mild, moderate or severe depending on the concentration of haemoglobin in the blood. Mild anaemia corresponds to haemoglobin concentration levels of 10 to 10.9 g/dl for pregnant women and young children; 10 to 11.9 g/dl for non-pregnant women, girls aged 11 to 19 years and boys aged 11 to 13 years; and 10.0 to 12.9 g/dl for boys aged 14 to 19 years. For all the tested groups, moderate anaemia corresponds to levels of 7 to 9.9 g/dl, and severe anaemia to levels less than 7 gm/dl.

EDHS 2000 revealed that about three out of ten young children suffer from some degree of anaemia. This is similar to the level that was found among women. Some 11 percent of young children had moderate levels of anaemia, and less than 1.0 percent were classified as having severe anaemia. Children under two years of age were more likely to be anaemic than older children, and rural children were more likely to be anaemic than urban children (33 and 24 percent, respectively). The highest anaemia prevalence (38 percent) was among children aged six to 59 months in rural Upper Egypt and the Frontier governorates, and the lowest (23 percent) was in urban Lower Egypt.

IDA among mothers, whether pregnant, lactating or non-pregnant and non-lactating (NPNL) increased significantly between 1978 (Nassar et al., 1992) and 2000 (EDHS, 2000) (Figure 17).

TABLE 6
Prevalence of IDA in different age groups

¹ HCT < 33 percent.
² HCT < 36 percent.

FIGURE 17
Trends in prevalence of IDA among pregnant, lactating and NPNL women, 1978 to 2004

Sources: Nassar et al., 1992; Moussa et al., 1997; EDHS, 2000; El-Sayed et al., 2002; Hassan et al., 2004.

Vitamin A deficiency

Values of plasma retinol were available for 1 577 preschool children (aged six to 71 months) in Moussa, El-Nehry and Abdel Galil, 1995 and Moussa et al., 1997. Results denoted that vitamin A deficiency (VAD) is a moderate sub-clinical public health problem in Egypt. In 2002, a national survey to assess the prevalence of vitamin A status after implementation of a vitamin A supplementation programme among children of nine and 18 months revealed that the prevalence of VAD among preschool children (six to 71 months) was 7.2 percent, implying that the vitamin A status of those children had improved.

Results of the survey to assess micronutrient deficiency among primary schoolchildren showed that a higher percentage of girls had low serum retinol levels (< 20 µg/dl) than boys.

A national survey for the determination of bone mass density among adolescents and adults in Egypt (Hassan, Abdel Galil and Moussa, 2004) showed that the prevalence of VAD among adolescents was higher in rural than urban areas. Results for adults showed a similar pattern. Findings are higher than those reported by Moussa, El-Nehry and Abdel Galil (1995) (Figure 22).

TABLE 7
Prevalence of VAD in different age groups

TABLE 8
Trends in prevalence of IDD in different age groups

Iodine deficiency disorders

The national survey conducted by NNI in collaboration with WHO (Hussein et al., 1992) found a high prevalence of iodine deficiency disorders (IDDs) manifested by a total goitre rate (TGR) of 6.5 percent. Older adolescents (aged 15 years and over) had a TGR of 7.8 percent, which was higher than the 6.4 percent among younger adolescents (aged 12 years and over). Children of primary school age were the least affected, with TGR of 5.2 percent. Moussa, El-Nehry and Abdel Galil (1995) also revealed a TGR of ³ 5 percent among 23.0 percent of mothers in rural and 20.0 percent in urban sites, and in 6.5 percent of children aged six to 71 months. Both surveys denoted the existence of a public health problem among different age groups, particularly older females. Using a cut-off level of < 100 mg/l for urinary iodine excretion, Hassan et al. (1998) reported that 30 percent of children in primary schools had iodine deficiency.

Based on serum TSH value < 0.39 ml U/l, the national survey of osteoporosis revealed that nearly 9 percent of adolescents and adults had iodine deficiency (Hassan, Abdel Galil and Moussa, 2004).

Zinc and other micronutrients

Results of the national survey of micronutrient deficiencies among primary schoolchildren (Hassan et al., 1998) showed an overall serum zinc deficiency rate of 15.5 percent, with higher prevalence among boys in rural areas (19.7 percent).

Hassan, Abdel Galil and Moussa’s (2004) national survey to determine bone mineral density among Egyptians found that 8.5 percent of adolescents and 5.6 percent of adults had low serum zinc levels, with no significant gender- or area-based differences.

Regarding serum selenium levels, 8.8 percent of primary schoolchildren had low levels (Hassan et al., 1998). Adolescents and adults suffered more from this problem, with low serum selenium rates of 26.0 and 25.0 percent, respectively. Females had higher prevalence than males, regardless of area of residence (Hassan, Abdel Galil and Moussa, 2004).

Calcium, phosphorus and vitamin D data reflect that the problem was mainly related to calcium, with more than 20 percent of adolescents and nearly one-third of adults having low serum calcium. Rural sites showed greater deficiency, particularly among adults, with no apparent gender-based differences. This was mainly related to the higher prevalence of obesity among females on the one hand, and of smoking among males on the other (Hassan, Abdel Galil and Moussa, 2004).

The burden of low serum magnesium (Mg) levels was found to be far higher among adolescents than adults; 22.0 percent of adolescents and 10 percent of adults had low serum Mg. This was mainly explained by the effect of growth on skeletal health and mineral metabolism (Hassan, Abdel Galil and Moussa, 2004) (Table 9).

TABLE 9
Percentage distribution of studied populations according to serum cut-off of micronutrients

¹ WHO/FAO, 1996 cut-off.
² Nelson, 1996 cut-off.
³ Cut-off of serum zinc is >70 µg/dl, and of serum selenium <7.5 µg/dl (Hassan Abdel Galil and Moussa, 2004).