Consumption expenditure on food

The National Sample Survey Organization (NSSO) is a permanent survey organization that was set up in the Department of Statistics of the Government of India in 1950. NSSO has been carrying out five-yearly consumer expenditure surveys since 1972/1973, providing time series data in rural and urban areas of all India’s states. Household food consumption at the national and state levels is computed on the basis of data on household monthly per capita consumption expenditure (MPCE) in 12 MPCE classes (with expenditure ranging from less than US$5 to $30). NSSO surveys have excellent sampling design, large sample sizes, clearly stated estimation procedures and national coverage, but do not provide insight into the actual dietary intake of households or individuals or into the intra-family distribution of food.

FIGURE 7
Expenditure on cereals (percentages), 1972 to 2001

Source: NSSO, 2001.

FIGURE 8
Consumption of cereals in rural areas of India (kg/month), 1972 to 2001

Source: NSSO, 2001.

Cereals

Data from NSSO surveys from 1972 to 2000 show that expenditures on cereal in the lowest (LIG) and middle-income groups (MIG) declined. Among the highest income group (HIG), cereals accounted for a fairly low proportion of total expenditure, which has remained essentially unchanged over the last three decades (Figure 7). People in the lowest income group were consuming greater quantities of cereals, even though they were also spending a reduced proportion of their total expenditure on these (Figure 8). This is because there has been a reduction in the relative cost of cereals, especially those supplied through the public distribution system (PDS). There was a decline in household consumption of cereals in the middle-income group, while monthly cereal consumption in high-income rural households dropped from 26.2 kg in 1972/1973 (about 1kg/day) to 14.4 kg in 1999/2000. Data from diet surveys conducted by the National Nutrition Monitoring Bureau (NNMB, 1979 to 2002) show that the average intake of cereals in even the highest income group has never exceeded 400 g/day. It would therefore seem likely that the high cereal consumption among high-income rural households might be because cooked food is shared with guests, relatives and servants. The sharing of food with guests and servants has declined over the last two decades, which accounts for the steep reduction in cereal consumption in high-income group households. The simultaneous increase in cereal consumption in the lower-income group confirms this.

With wheat and rice available through the PDS, poorer segments of the population now use these as staple cereals. The consumption of coarse cereals rich in micronutrients and minerals has declined (Figure 9). The Tenth Five-Year Plan (Planning Commission, 2002) recommends that locally produced and procured coarse grains be made available through a targeted public distribution system (TPDS) at subsidized rates. This may substantially reduce the cost of subsidies without any decrease in the energy provided; an improved micronutrient intake from coarse cereal would be an added benefit. Such a measure would also improve targeting, as only the most needy are likely to buy these coarse grains.

FIGURE 9
Cereal consumption, 1972 to 2000

Source: NSSO, 2001.

Pulses

Between 1972 and 2001, there was a substantial increase in the proportion of expenditure spent on pulses in the lowest income group (Figure 10), but expenditure on pulses remained relatively unaltered in the middle and highest income groups. In spite of increased expenditure, household consumption of pulses declined in all income groups and in both urban and rural areas (Figure 11). Data from the NSSO 2000 survey show that middle and upper income groups spent more on milk and animal products, so their protein intakes were not adversely affected by the reduction in pulses. Pulses are still the major source of protein in the lowest income group. In order to ensure adequate protein intake for this group, it is therefore essential to increase the cultivation of a variety of pulses and legumes, so that they can be made available at affordable prices, perhaps through TPDS.

FIGURE 10
Expenditure on pulses (percentages of total), 1972 to 2001

Source: NSSO, 2001.

FIGURE 11
Consumption of pulses in rural areas (kg/month), 1972 to 2001

Source: NSSO, 2001.

Time trends in monthly per capita expenditure

MPCE on food and non-food items over the last three decades is shown in Figure 12. The proportion of total expenditure spent on foodstuffs has declined considerably over the last three decades - from 70.6 percent in 1972 to 55.3 percent in 1999/2000 - mainly because of the decline in cereal prices. Expenditures on pulses, vegetables, other foods and beverages increased. However, pulse and vegetable intakes among the poor remain low. There are massive urban-rural and inter-district/state differences in the costs of vegetables, milk, fish and meat. Therefore data on the quantities of these foodstuffs consumed by state or expenditure group are not available from NSSO surveys. India uses diet surveys for such information.

FIGURE 12
Percentage distribution of MPCE in rural areas, 1972 to 2001

Source: NSSO, 2001.

Nutrient intake computed from NSSO surveys

NSSO uses household expenditures on food to compute the energy, protein and fat intakes of the population. Over the last three decades, overall energy and protein consumption in rural areas has shown a small decline while remaining unaltered in urban areas. There have been increases in fat consumption in both rural and urban areas (Table 6).

TABLE 6
Average daily per capita nutrient intakes, 1972 to 2000

Source: NSSO, 2001.

TABLE 7
Average per capita calorie consumption by income group, 1972 to 1994

Source: NSSO, 2001.

Changes in energy consumption in different income groups in urban and rural areas are shown in Table 7; energy consumption has shown small increases in both the urban and rural poor and substantial declines among the urban and rural rich. As indicated earlier, data on household consumption expenditure in high-income groups include the food shared with guests and servants, so therefore have to be interpreted with caution. There are massive interstate differences in food expenditures.

Urban-rural differences

Data from the NSSO 55th round (1999/2000) on urban-rural differences in food expenditure are given in Figure 13. Among the urban and rural poor, most food expenditure was on cereals. Dietary diversification is seen mainly among middle- and high-income groups in both urban and rural areas; diversity is greater in urban areas, perhaps because of access to a wider variety of foodstuffs.

FIGURE 13
Expenditure on foodstuffs by income group

Source: NSSO, 2001.

Summary

Data from NSSO consumption expenditure surveys indicate the following:

• Expenditure on food has declined, mainly because of reductions in the cost of cereals.

• In spite of a steep decline in cost, cereal consumption has not increased, except in the lowest income group. This apparent paradox has been the subject of widespread debate among economists and nutritionists in India, and the consensus view is that it might be because cereal requirements are being met.

• In spite of increased expenditure on pulses, consumption among all segments has declined because of the soaring cost of pulses. Pulse consumption is very low among the poor.

• Rural populations consume more cereals, fewer pulses and less oil and fat compared with urban populations.

• Dietary diversification is seen mainly in middle- and high-income groups in urban and rural areas. Increasing incomes and wider availability of diverse foodstuffs accelerated this trend in the 1990s.

• There has been only a small increase in energy consumption among the poor in spite of steep declines in the cost of cereals. Among the middle- and high-income groups, energy consumption has declined.

• The energy consumption of the urban high-income group is associated with consumption of increased quantities of sugar, oil, milk and milk products and lower quantities of cereals. Changes in the energy density of the diet and sedentary lifestyles appear to be the major factors for the steep increase in obesity among this group.

Dietary intake data from nutrition surveys

Since 1975, the National Nutrition Monitoring Bureau (NNMB) has been providing data on dietary intake (by 24-hour dietary recall) and nutritional status (by anthropometry and nutritional deficiencies) for ten states of India (Kerala, Karnataka, Andhra, Tamil Nadu, Maharashtra, Orissa, Gujarat, Madhya Pradesh, West Bengal and Uttar Pradesh). NNMB is the only survey that provides data on intra-family food distribution and the dietary intake and nutritional status of all age groups. Proposals to expand the network to cover all states have not yet been implemented. A one-time district nutrition survey was carried out in the mid-1990s in order to obtain data on the dietary intake and nutritional status of individuals in other states. Both the NNMB 1994 survey and this one-time survey used the same methodology of data collection, using representative samples of households in every state. The combined data were reported as the India Nutrition Profile (INP) (DWCD, 1995/1996).

INP provides data on the dietary intake and nutritional status of all age groups, in all states and in both urban and rural areas. Both the NNMB and INP surveys used 24-hour dietary recall to assess food intake. The amounts consumed were compared with the RDAs for India drawn up in 1989 by the Indian Council of Medical Research (ICMR, 1989).

Household food intake obtained by 24-hour dietary recall is used to compute the average intakes of household members expressed as consumption units (CUs) per day (NNMB, 1981). The CUs for different age and gender groups were worked out from the basis of the energy consumption of an average adult male doing sedentary work being 1 CU (Box 4). The reference man is between 20 and 39 years of age, weighs 60 kg and is physically fit and moderately active. The reference woman is between 20 and 39 years of age, weighs 50 kg and is moderately active.

Box 4. Consumption units

Source: NNMB, 1981.

Nutrient intake is computed using the Nutritive value of Indian foods (NIN, 2004), first published by the National Institute of Nutrition (NIN) in 1971 and updated many times since then. Analysis of the iron content of foodstuffs using recent techniques shows that the iron available is only about 50 percent of the values previously reported; hence, values for iron content have been revised in the latest edition.

Food intake in urban and rural areas

Data from the NNMB and INP surveys show that average intakes of cereals in the mid-1990s were near the RDAs, but intakes of pulses, vegetables and fruits were low (Table 8). There are significant differences in food intake among states. Reported intakes of foodstuffs are higher in INP than in NNMB data, probably because there are higher dietary intakes - especially of cereals and pulses - in states not included in the NNMB survey, but covered by INP. Dietary intake was higher in some states with high per capita income (Punjab), but not others (Maharashtra), which suggests that greater per capita income is not always associated with higher dietary intake. Data from both NNMB and INP show that cereal intakes were higher in some of the poor states (Orissa in NNMB, Uttar Pradesh in INP), perhaps because most of the population of these states work as manual labourers and require high cereal intakes. NSSO (1975 to 2001) consumer expenditure surveys show similar interstate differences. Consumption of cereals is higher in rural areas, while that of pulses, milk and milk products, fruits and fat and oils is higher in urban areas.

TABLE 8
Food intakes in rural and urban areas (g/CUs per day)

Sample sizes: NNMB, rural - 1975-1979, 33 048; 1996-1997, 14 391; 2000-2001, 30 968. Urban slums - 1975-1980, 32 500; 1993-1994, 5 447. INP - 46 457.

Sources: NNMB; INP.

Time trends in food intake

Data on time trends in food intake in rural areas and urban slums in nine states are available from NNMB surveys (Table 8). These data show that there has been some decline in cereal consumption in both urban and rural areas over the last three decades. There has also been a substantial decline in the cost of cereals and an improvement in their availability. The decline is therefore not due to economic constraints. Over the same period, there has also been a decline in the consumption of pulses, which are a major source of protein in Indian diets. This is partly attributable to soaring costs and the inability of poor people to purchase them in adequate quantities, in spite of higher expenditure on pulses.

Although India’s milk output has increased massively, there has not been any improvement in the per capita consumption of milk. Consumption of vegetables and fruits also continues to be very low. In rural areas, there has not been any significant increase in the per capita consumption of fats and oils and of sugar and jaggery. However in urban areas - even among slum dwellers - there has been an increase in oil consumption and some increase in sugar consumption. Data from NNMB surveys suggest that dietary intake has not undergone any major shift towards increased consumption of fat and oils, sugar and processed food, and there has been no increase in energy intake. These data are confirmed by the consumer expenditure on food items reported in NSSO.

Nutrient intake

INP provides data on nutrient intake in the urban and rural areas of all states (Table 9). The nutrient intakes reported in INP are higher than those in NNMB because of higher intakes in states not covered by NNMB. At the aggregate national level, total energy intake was less than 2 300 kcal/CUs/day in the mid-1990s. There are substantial interstate differences in energy and other nutrient intakes.

TABLE 9
Nutrient intakes in rural and urban areas (g/CUs/day)

Sample sizes: NNMB, rural - 1975-1979, 33 048; 1996-1997, 14 391; 2000-2001, 30 968. Urban slums - 1975-1980, 32 500; 1993-1994, 5 447. INP - 46 457.
¹ Changes in intake from 1975-1979 to 2000-2001.
² Changes in intake from 1975-1979 to 1993-1994.
³ Method of estimation different.
* Data not available.
Sources: NNMB, 1979; 2002; INP, 1996.

Time trends in nutrient intakes

Data on time trends in nutrient intakes are available from NNMB surveys (Table 9). These data show there has been a small decline in energy intake over the last three decades (Figure 14). There has also been some decline in the intakes of most nutrients in both urban and rural areas. The percentage of total energy intake derived from carbohydrates has declined and there has been some increase in the percentage dietary energy from fats (Figure 15). In spite of this, the proportion of dietary energy from fat remains less than 15 percent. These aggregate measures mask large disparities between the intakes of urban and rural populations and among different socio-economic groups. In India, the dietary intake of iron has always been low. The steep decline in iron intake reported in the last NNMB survey can be attributed to different estimation methods, which showed that absorbable iron was 50 percent less than it was in earlier surveys.

FIGURE 14
Time trends in energy intake, 1975 to 2001

Source: NNMB reports.

FIGURE 15
Macronutrient intake in rural areas as percentage of total energy, 1979 to 2001

Source: NNMB reports.

Urban-rural differences in nutrient intakes

Energy intake is lower in urban areas (Table 9), in spite of higher intake of fats and oils, because of lower cereal consumption. Data from NNMB surveys suggest that the consumption of all nutrients is lower in urban slums than in rural areas. INP, which covered most states, did not show any significant differences in nutrient intake between urban and rural areas. Interstate differences in nutrient consumption and the fact that NNMB data were available only on urban slums are two of the factors responsible for the apparent differences between NNMB and INP survey data.

Source of dietary energy

Data on total energy intakes and percentages of energy intake from fat, carbohydrate and protein for different age groups, as reported by NNMB and INP, are given in Table 10. Carbohydrates remain the major source of energy in Indian diets. There has been some reduction in the percentage of total energy intake from carbohydrates and some increase in the percentage from fats over the past three decades.

TABLE 10
Sources of dietary energy

Sample sizes: NNMB - 1975-1979, 33 048; 1996-1997, 14 391; 2000-2001, 22 945. INP - 46 457.
¹ No gender disaggregation of data before ten years of age.
² Data not available.
³ No gender disaggregation of data after 18 years of age.
Sources: NNMB, 1979; 2002; INP, 1996.

Dietary diversity

The second National Family Health Survey (NFHS-2) (IIPS, 1998/1999) collected data on frequency of consumption of various types of foods (daily, weekly or occasionally) to assess dietary diversity among 90 000 married women aged 15 to 49 years living in 26 states. The survey did not include the quantities of intake. Data from the survey are presented in Tables 11 and 12. Adult women in India consume cereals every day; their diets tend to be monotonous and there is very little dietary diversity. Fruits are eaten daily by only 8 percent of women, and once a week by only one-third. Almost one-third of women never eat chicken, meat or fish, and very few (only 6 percent) eat these foods every day. Eggs are consumed even less frequently than chicken, meat or fish.

TABLE 11
Women’s frequency of consumption of selected foods

Source: IIPS, 1998/1999.

There were substantial differences in food consumption patterns according to background characteristics (Table 12). Age does not play an important role in women’s consumption patterns, but women in urban areas are more likely than those in rural areas to include every type of food in their diet, particularly fruits and milk or curd. Illiterate women have less varied diets than literate women, and seldom eat fruits. Poverty has a strong negative effect on dietary diversity. Women from households in the low socio-economic group are less likely than others to eat items from each type of food group listed, and their diets are particularly deficient in fruits and milk or curd. There are substantial interstate differences in the consumption of different types of food.

TABLE 12
Women’s food consumption (percentages of survey population)

Source: IIPS, 1998/1999.

Summary

During the past three decades there have been:

• reductions in energy intake from cereals, except among the poor; overall there has been a small decrease in total energy intake in both urban and rural areas;

• some increase in percentage of dietary energy derived from fat, and a reciprocal reduction in that derived from carbohydrate;

• some increase in consumption of fats and oils in urban populations, even in urban slums;

• increasing dietary diversity among upper-income groups in rural and - especially - urban areas;

• cereal-based and monotonous diets for the rural poor, with low micronutrient content;

• low iron intake, which coupled with the poor bioavailability of iron from Indian diets is responsible for a high prevalence of anaemia

Dietary intake and nutritional status in different age groups

As well as the dietary intake and nutritional status data collected by NNMB and INP, NFHS 1 and 2 (IIPS, 1992/1993; 1998/1999) provide state-level estimates of time trends in the nutritional status of women and preschool children in all major states during the 1990s. The District-Level Household Survey (DLHS) 2002/2003 (Ministry of Family and Health Welfare, 2004) provides district-level estimates on the nutritional status of preschool children. In addition, several smaller studies provide follow-up data on the nutritional status of specific groups over decades. This section reviews these data on time trends in the dietary intake and nutritional status of different age groups.

Time trends in anthropometric indices

Data from NNMB rural surveys of trends in weight, height, mid-arm circumference and triceps fat fold thickness in males and females of all age groups are shown in Figures 16 to 21. Even in the rural population, adult height has increased by about 4 cm. Increases in body weight have been greater, mainly due to fat deposition, as shown by rising fat fold thickness over this period. These affected all age groups, and especially women.

FIGURE 16
Trends in mean weights in rural males, 1975 to 2001

Source: NNMB reports.

FIGURE 17
Trends in mean heights in rural males, 1975 to 2001

Source: NNMB reports.

FIGURE 18
Trends in mean tricep fat fold thickness in rural males, 1975 to 2001

Source: NNMB reports.

FIGURE 19
Trends in mean weights in rural females, 1975 to 2001

Source: NNMB reports.

FIGURE 20
Trends in mean heights in rural females, 1975 to 2001

Source: NNMB reports.

FIGURE 21
Trends in mean tricep fat fold thickness in rural females, 1975 to 2001

Source: NNMB reports.

Data from NNMB surveys in urban slums are shown in Figures 22 to 25. Mean body weight, mid-upper arm circumference and fat fold thickness at triceps have increased in all age groups. Most of the body weight increase is due to increased fat, as shown by rising fat fold thickness.

FIGURE 22
Trends in mean weights in urban males, 1975 to 1994

Source: NNMB reports.

FIGURE 23
Trends in mean tricep fat fold thickness in urban males, 1975 to 1994

Source: NNMB reports.

FIGURE 24
Trends in mean weights in urban females, 1975 to 1994

Source: NNMB reports.

FIGURE 25
Trends in mean tricep fat fold thickness in urban females, 1975 to 1994

Source: NNMB reports.

Low birth weight

Nearly one-third of Indian infants weigh less than 2.5 kg at birth. Incidence of low birth weight (LBW) is highest among low-income groups (Table 13). There is clear correlation between birth weight and maternal body weight (Figure 26); low birth weight rate doubles when Hb levels fall below 8 gm/dl. Low birth weight incidence has remained unaltered over the last three decades (Figure 27) (NFI, 2004).

TABLE 13
Birth weight and socio-economic status

Source: Ramachandran, 1989.

FIGURE 26
Birth weight in relation to maternal body mass index

Source: Planning Commission, 2002.

FIGURE 27
Trends in birth weight, 1969 to 1998

Source: NFI, 2004.

Although there has been no decline in the prevalence of low birth weight, India has achieved a substantial decline in infant mortality (RGI, 2002). As more low-birth-weight newborns survive, there is growing concern regarding the relationship between low birth weight and poor growth during childhood and adolescence, as well as increased risk of chronic degenerative diseases in later life.

Under the Reproductive and Child Health Programme 1 and 2 (Ministry of Family and Health Welfare, 1998/1999; 2002), efforts are under way to provide effective antenatal care and reduce rates of low birth weight. Factors such as maternal height, which has a significant influence on birth weight, cannot be improved with short-term corrective interventions, but anaemia, pregnancy-induced hypertension and low maternal weight gain during pregnancy can be detected and treated. Effective management of these could result in substantial reductions in both pre-term births and the birth of small-for-date infants.

Growth during infancy and early childhood

Growth during infancy and childhood depends on birth weight, adequacy of infant feeding and absence of infection. Available data clearly indicate that exclusively breastfed infants thrive better during the first six months of life and have lower morbidity episodes (diarrhoea, respiratory tract infection and fever) than those receiving supplements in addition to breastmilk. In India, steps taken to protect and promote the practice of breastfeeding have been effective, and breastfeeding is now almost universal (Planning Commission, 2002). However, the message that exclusive breastfeeding up to six months followed by the gradual introduction of semi-solids is critical for the prevention of undernutrition in infancy has not been as effectively communicated. Data from NFHS 2 (IIPS, 1998/1998) indicate that although breastfeeding is nearly universal and the mean duration of lactation is more than two years, only 55.2 percent of infants up to three months of age receive exclusive breastfeeding. In spite of the emphasis on the need to introduce complementary food gradually, only 33.5 percent of infants in the six to nine months age group receive breastmilk and semi-solid food.

There are substantial interstate differences in exclusive breastfeeding and the timely introduction of semi-solid food (Figure 28). Early introduction of supplements is a major problem in states such as Delhi, Himachal Pradesh and Punjab, while late introduction is a problem in Bihar, Uttar Pradesh, Madhya Pradesh, Rajasthan and Orissa. Kerala fares well in terms of appropriate infant feeding practices, and this might be one of the reasons for the relatively low undernutrition rates in this state (IIPS, 1998/1999).

Early introduction of milk substitutes and late introduction of complementary food are associated with increased risk of undernutrition and infection. Faulty infant feeding practices are causing the prevalence of undernutrition to increase steeply with age, from 11.9 percent at less than six months to 58.5 percent in the 12 to 23 months age group (Figure 29). A major thrust of the Tenth Five-Year Plan is to prevent the onset of undernutrition in infancy and early childhood through nutrition education, so that by 2007 more than 80 percent of women breastfeed exclusively up to six months and the complementary feeding rate at six months goes up to 75 percent (IIPS, 1998/1999).

FIGURE 28
Infant feeding practices by state

Source: IIPS, 1998/1999.

FIGURE 29
Prevalence of undernutrition (weight for age less than -2 SD)

Source: IIPS, 1998/1999.

Time trends in the dietary intake and nutritional status of preschool children

Data from NNMB on energy intake and the prevalence of undernutrition in children under three years of age are shown in Figure 30. There has been a steady decline in undernutrition in children, even though the dietary intake has not shown a major change. The decline in undernutrition is most probably attributable to better access to health care and the effective management of infections.

FIGURE 30
Energy intake and undernutrition in children aged one to three years, 1979 to 2002

Source: NNMB reports.

Preschool children constitute one of the most nutritionally vulnerable segments of the population, and their nutritional status is considered to be a sensitive indicator of community health and nutrition. Their dietary intake has not improved substantially over the last two decades (Table 14).

TABLE 14
Average nutrient intakes among preschool children, 1975 to 1997

Source: NNMB, 2000.

Data on energy intake in children, adolescents and adults from NNMB 2000 are shown in Table 15. Mean energy consumption as a percentage of RDA is lowest among preschool children. Time trends in the intra-family distribution of food (Figure 31) indicate that although the proportion of families in which both adults and preschool children have adequate food has remained at about 30 percent over the last 20 years, the proportion of families with inadequate intake has decreased substantially. However, the proportion of families in which preschool children receive inadequate and adults adequate intakes has nearly doubled, even though the RDA for preschool children forms only a very small proportion (an average of 1 300 kcal/day) of the family’s total intake of about 11 000 kcal/day (assuming a family size of five). It therefore appears that poor young child feeding and care practices - and not poverty - is the factor responsible for inadequate dietary intake. The Tenth Five-Year Plan (Planning Commission, 2002) emphasizes the importance of health and nutrition education to ensure proper intra-family distribution of food, based on needs.

TABLE 15
Average energy intakes for children, adolescents and adults

Source: NNMB, 2000.

FIGURE 31
Comparison of adequate energy status of preschool children and adults, 1975 to 1997

Time trends in prevalence of undernutrition in preschool children

Over the last three decades, there has been a steep decline in the prevalence of moderate and severe undernutrition as assessed by weight-for-age and height-for-age (Figures 32 and 33), but very little change in the prevalence of wasting. In spite of the steep decline in the prevalence of stunting, the mean height of children has changed only very slightly. The increase in adult height has also been a modest 2 to 4 cm in three decades.

FIGURE 32
Trends in prevalence of undernutrition in children (percentages), 1975 to 1999

Sample sizes: NNBB - 1975-1980, 6 428; 1988-1990, 13 432; 1996-1997, 8 654; 2000-2001, 6 646. INP - 46 457. NFHS - 1992-1993, 25 584; 1998-1999, 24 600.

Sources: NNMB; INP; IIPS.

FIGURE 33
Prevalence of severe undernutrition in children (percentages), 1975 to 1999

Sample sizes: NNBB - 1975-1980, 6 428; 1988-1990, 13 432; 1996-1997, 8 654; 2000-2001, 6 646. INP - 46 457. NFHS - 1992-1993, 25 584; 1998-1999, 24 600.

Sources: NNMB; INP; IIPS.

Indian children are short compared with the National Center for Health Statistics (NCHS) norms; even when they have appropriate weight for height they are classified as undernourished according to these norms. The so-called South Asian paradox (high undernutrition rates but comparatively good health status) disappears when the body mass index (BMI)-for-age is the criterion for defining undernutrition. Early detection and correction are needed if wasting is to be reduced so that Indian children can achieve their growth potential. There are considerable interstate differences in the dietary intake and nutritional status of children (Figure 34). Although dietary intake is a major determinant of nutritional status in children, it is not the only one. Energy intake is low and undernutrition high in Uttar Pradesh, Bihar and Rajasthan. However, in spite of low energy intakes, the prevalence of undernutrition in Kerala and Tamil Nadu is low, probably because there is more equitable intra-family distribution of food based on needs, and better access to health care. The combination of high energy intakes and high undernutrition prevalence in Madhya Pradesh and Orissa is probably due to inequitable food distribution and poor access to health care (IIPS, 1998/1999).

FIGURE 34
Energy intake and undernutrition among children, by state

Sources: INP; IIPS, 1998/1999.

The nutritional status of poor children in Kerala is similar to that of rich children in Uttar Pradesh and Orissa (Figure 35). This is probably attributable to better access to health care and more equitable intra-family food distribution in Kerala than in Uttar Pradesh. These data clearly indicate that lack of access to health care is a major factor in undernutrition among preschool children. The decline in fertility and the reduction in family size may also have contributed to this because the prevalence of severe forms of undernutrition is higher in large families (IIPS, 1998/1988).

FIGURE 35
Nutritional status of children (weight-for-age) by income group and state

Sources: IIPS, 1992-93.

Poor dietary intake, poor care practices and poor access to health care are some of the major factors responsible for undernutrition and a high under-five mortality rate (U5MR). In most of the states where undernutrition is high (e.g., Orissa), U5MR is also high; in states where undernutrition is low (e.g., Kerala), U5MR is also low (Figure 36). There are exceptions to this, however; in Maharashtra U5MR is relatively low, in spite of relatively high undernutrition rates - this might be because access to health care is relatively good. In Punjab, in spite of high per capita income and dietary intake and good access to health care, both undernutrition and U5MR are relatively high. These data indicate the importance of health care in reducing both undernutrition and U5MR (IIPS, 1992/1993).

FIGURE 36
Prevalence of severe underweight and U5MR by state

Sources: IIPS, 1998/1999.

Nutritional status of affluent schoolchildren

Studies carried out by NFI in 1991 (NFI, 2004) show that the growth of affluent children up to six years of age is similar to the NCHS and WHO norm. Data from NFI studies in Delhi between 2000 and 2004 (NFI, 2004) show that while undernutrition is a problem among children from low-income groups (LIG) who are studying in government schools, overnutrition is the cause for concern among high-income-group (HIG) schoolchildren from six years of age (Figure 37).

FIGURE 37
Comparison of weight-for-age in Delhi schoolboys

Source: NFI, 2004.

Growth of adolescents from affluent urban families

The heights and weights of adolescent girls and boys from affluent income groups are comparable to NCHS norms (Table 16), and higher than those of adolescents surveyed by NNMB. NFI data on height and weight distribution (compared with NCHS norms) in Delhi schoolchildren from affluent families are shown in Figures 38 and 39. Even in these affluent segments of the population, some children are stunted (-2 SD height for age). There are overweight children in all classes and age groups. Among children over ten years of age there is a reduction in overweight because children of this age try to lose weight through exercise or skipping meals (NFI, 2004). However, the adolescents have inconsistent eating and exercise habits and tend therefore to have cyclical weight gain and loss, thereby incurring the health hazards associated with this pattern.

FIGURE 38
Height-for-age

TABLE 16
Growth of adolescents from urban affluent families

FIGURE 39
Weight-for-age

Nutritional status of adults

NNMB and INP data show that the prevalence of undernutrition in adults is higher in rural than urban areas (Table 17). Overnutrition is higher in urban areas. Over the last three decades there has been a progressive decline in undernutrition and some increase in overnutrition in both urban and rural areas. The prevalence rates of both under- and overnutrition are higher in women than men.

TABLE 17
Prevalence of under- and overnutrition among adults, 1975 to 2001

Sample sizes: NNMB - 1975-1979, 11 973; 1989-1990, 21 398; 1993-1994, 2 772; 1996-1997, 30 773; 2000-2001, 11 074. INP - 17 7841.

Sources: NNMB; INP.

Nutritional status of women

Data from NFHS-2 (IIPS, 1998/1999) indicate that the prevalence of undernutrition among women in urban areas is half that of rural areas (Table 18). Overnutrition is four times higher in urban than in rural areas. In women, as age increases, the prevalence of undernutrition declines while that of overnutrition increases.

TABLE 18
Prevalence of under- and overnutrition among women (15 to 45 years)

Sample size: 77 119.
Source: IIPS, 1998/1999.

Data from NFHS-2 show that although undernutrition continues to be high among women in poorer segments of the population, overnutrition and obesity are emerging as major problems in all states of India. There are substantial differences in the prevalence of under- and overnutrition among states, but all states have to prepare to detect and manage this dual nutrition problem in women (Figure 40).

FIGURE 40
Comparison of BMI in women, by state

Summary

Over the past three decades there have been:

• a very small (2 to 4 cm) increase in adult height;

• a significant increase in mean body weights, mostly owing to increased body fat as shown by increased fat fold thickness, which is greater in urban than rural areas.

In the absence of increased energy consumption, increased fat deposition is attributed to reduced physical activity. Very few studies have documented changes in physical activity patterns over the last three decades, but it is documented that over this period there have been:

• reduction in the number of people engaged in manual work;

• substantial improvement of mechanical aids in agriculture, industry and allied activities;

• improved access to water and fuel near households, in both urban and rural areas;

• improved availability of urban transport at affordable costs, resulting in fewer people walking or cycling to work, school or market;

• more mechanical aids that reduce physical activity during cooking and household tasks;

• TVs and computers in affluent urban households, which contribute to steep reductions in physical activity.

These lifestyle changes have led to reductions in energy requirements. Unchanged energy intakes combined with reduced energy requirements are associated with a positive energy balance and fat deposition.