Micronutrient deficiencies

Goitre caused by iodine deficiency, blindness by vitamin A deficiency (VAD) and anaemia by iron and folate deficiency are major public health problems in India. Over the last three decades, there has been a steep decline in keratomalacia caused by severe VAD, but no decline in the prevalence of anaemia caused by iron and folic acid deficiency; the declines in VAD and iodine deficiency disorders (IDDs) have been very slow. Data from NNMB surveys, IIPS and DLHS provide valuable insights for assessing the progress achieved in combating these deficiencies, help to formulate future interventions and provide baseline information for assessing the impacts of future interventions.

Anaemia

In India, the prevalence of anaemia is high because of:

• low dietary intake of iron (less than 20 mg/day, according to NNMB 2000) and folate (less than 70 mg/day);
• poor bioavailability of iron (only 3 to 4 percent) in the phytate fibre-rich Indian diet;

• chronic blood loss caused by infections such as malaria and hookworm infestations.

Data from DLHS (all states 1 100 households/district; Ministry of Family and Health Welfare, 2002/2003) and the NNMB survey (from eight states, NNMB, 2002) show that prevalence of anaemia is very high (ranging from 80 to more than 90 percent) in preschool children, pregnant and lactating women and adolescent girls (Figure 41). Criteria used for assessing anaemia in DLHS are given in Table 19.

FIGURE 41
Prevalence of anaemia (percentage)

Source: Ministry of Family and Health Welfare, 2002/2003.

TABLE 19
Anaemia measurement criteria used in DLHS (g/dl)

Source: Ministry of Family and Health Welfare, 2002/2003.

Moderate and severe anaemia is seen even among upper-income group families. There are interstate differences in prevalence, which are probably attributable to differences in dietary intake and access to health care.

Anaemia is associated with increased susceptibility to infections, reduced work capacity and poor concentration. Anaemia remains a major cause of maternal mortality in India, accounting for more than 20 percent of all maternal deaths. In response to the low dietary intake of iron and folate, the high prevalence of anaemia and its adverse health consequences, India was the first developing country to adopt a National Nutritional Anaemia Prophylaxis Programme to prevent anaemia among pregnant women and children. Screening for anaemia and iron-folate therapy in appropriate doses have been essential components of antenatal and paediatric care for the last three decades, but coverage of these programmes is very low. As a result, very high rates of anaemia in pregnant women persist, and the impacts of severe anaemia on birth weight and maternal mortality remain unaltered. Anaemia continues to be a major problem affecting all segments of the population, and there has been no substantial decline in the adverse health consequences associated with it.

Strategies for the prevention, detection and management of anaemia in the Tenth Five-Year Plan

The total Indian population of more than 1 billion people will have to double their iron and folate intakes and sustain these new levels life long. The major intervention strategies required for the prevention and management of anaemia are:

• dietary diversification to include iron- and folate-rich foods and food items that promote iron absorption;
• health and nutrition education to improve the consumption of iron- and folate-rich foodstuffs;
• food fortification, especially the introduction of iron and iodine-fortified salt at affordable costs;
• screening for early detection of anaemia among vulnerable groups (such as children and pregnant women);

• management of anaemia according to its severity/chronicity, the physiological status of the individual and the time available for treatment.

The Tenth Five-Year Plan (Planning Commission, 2002) has set the goal of reducing the prevalence of anaemia by 25 percent and of moderate/severe anaemia by 50 percent, by 2007.

Vitamin A deficiency

Vitamin A is an important micronutrient for maintaining normal growth, regulating cellular proliferation and differentiation, controlling development, and maintaining visual and reproductive functions. Diet surveys show that intakes of vitamin A are significantly lower than the recommended daily allowance in all groups, and that they have not increased over the decades (NNMB, 1979 to 2002). IIPS, Ministry of Family and Health Welfare (1998/1999) and DLHS surveys show that coverage of the Massive Dose Vitamin A Programme has been poor (Figure 42). However, over the years there has been a steep decline in severe forms of VAD in children; blindness caused by VAD is now very rare. All the large national surveys (NNMB, 2002; ICMR, 2004a; NNMB, 2001) have clearly shown that the prevalence of clinical VAD in children under five years of age in India is currently less than 1 percent (Figure 43). The decline in VAD in children appears to be caused by better access to health care and a consequent reduction in the severity and duration of common childhood morbidity to infections, especially measles.

FIGURE 42
Coverage of the Massive Dose Vitamin A Programme by state

Sources: Ministry of Family and Health Welfare, 1998/1999; 2002/2003; IIPS, 1998/1999.

FIGURE 43
Prevalence of Bitot’s spots among children aged one to five years (percentages)

Sources: NNMB, 2002; ICMR, 2004a; NNMB, 2001.

Strategies for managing VAD in the Tenth Five-Year Plan

Clinical VAD often coexists with other micronutrient deficiencies; hence there is a need for broad-based dietary diversification programmes aimed at improving the overall micronutrient status of the population. In addition, the ongoing Massive Dose Vitamin A Programme in children aged nine to 36 months will be continued and its implementation strengthened.

Goals for the Tenth Plan

• Achieve universal coverage for each of the five doses of vitamin A.

• Reduce prevalence of night blindness to less than 1 percent, and that of Bitot’s spots to less than 0.5 percent, in children between six months and six years of age.

• Eliminate VAD as a public health problem.

Iodine deficiency disorders

Iodine deficiency disorders (IDDs) have been recognized as a public health problem in India since the 1920s. IDD is caused by a lack of iodine in water, soil and foodstuffs, and it affects all socio-economic groups in defined geographic areas. Surveys carried out by central and state health directorates, ICMR and medical colleges have shown that no Indian territory is free from the problem of IDD. An estimated 167 million people are at risk of IDD - 54 million of whom have goitre while more than 8 million have neurological handicaps. Universal use of iodized salt is a simple, inexpensive method of preventing IDD.

Ongoing interventions to reduce IDD

The Government of India launched the National Goitre Control Programme (NGCP) in 1962. Initially, the programme aimed to provide iodized salt to the well-recognized sub-Himalayan "goitre belt". However, the erratic availability of the salt, the availability of cheaper non-iodized salt and a lack of awareness regarding the need to use iodized salt meant that there was no substantial reduction in IDD. It was then decided to introduce universal iodization of all the salt used for human consumption. This was implemented in a phased manner from 1986, and major efforts were made to increase the production of and access to iodized salt (Salt Department, 2003/2004). In August 1992, the NGCP was renamed the National Iodine Deficiency Disorders Control Programme (NIDDCP) and took into its ambit control of the entire spectrum of IDD. India became the second largest producer of iodized salt in the world, after China. In 1997, the central government banned the storage and sale of non-iodized salt, but lifted the ban in October 2000 because "matters of public health should be left to informed choice and not enforced".

NNMB 2002 data on prevalence rates of goitre in six- to 12-year-old children are shown in Figure 44. The relatively high prevalence of goitre in these non-endemic states is a source of concern. Data from DLHS (Ministry of Family and Health Welfare, 2002/2003), which undertook spot tests of iodization in the salt consumed in 3 05 106 households, are presented in Figure 45. There has been some decline in the consumption of iodized salt since the ban on using non-iodized salt was lifted.

FIGURE 44
Prevalence of goitre in children aged six to 12 years, by state

Source: NNMB, 2002.

FIGURE 45
Percentages of households consuming iodized salt, by state

Source: Ministry of Family and Health Welfare, 2002/2003.

Strategies for the prevention of IDDs in the Tenth Five-Year Plan

On 25 June 2005, the Union Minister for Health and Family Welfare announced the decision of the Government of India to reimpose the ban on sales of non-iodized salt for human consumption. It is expected that this announcement will ensure universal access to iodized salt, such that the goals set in the Tenth Five-Year Plan can be achieved.

Goals for the Tenth Plan

• Achieve universal access to iodized salt.
• Generate data on iodized salt consumption by district.

• Reduce the prevalence of IDD in India to less than 10 percent by 2010.

Prevalence of non-communicable diseases

Soon after independence, India established systems for assessing per capita income, purchasing power, poverty, undernutrition and micronutrient deficiencies. Data from these surveys were used to assess interstate differences and time trends. A similar system for tracking overnutrition and the risk of non-communicable diseases (NCDs) was not established until the 1990s, and even now the coverage of this is not as extensive as that of other surveys. In view of this, for documenting time trends in prevalence of NCDs related to overnutrition, India has to depend on research studies carried out in different parts of the country. The differences in methodology of data collection, criteria used for case definition and parameters reported make it difficult to make comparisons among studies and to draw conclusions regarding time trends. However, from the existing data, it is clear that there has been an increase in prevalence rates of diabetes, hypertension and CVD over the last two decades, especially in affluent urban segments of the population. Prevalence of these diseases is lower in poorer segments and in rural areas, but case fatality rates may be higher in these areas because of poor access to health care.

The National Cancer Registry Programme (NCRP) (ICMR, 1983) established cancer registries based on hospitals and populations in the mid-1980s, and generates data on time trends and regional differences in cancer incidence, prevalence and mortality. Data from NCRP show that India has the lowest cancer rates in the world, although it also has relatively high rates of tobacco use (nearly half of the cancers in men are tobacco-related). In spite of increasing longevity, there has not been any increase in overall cancer incidence over the last two decades. However, there have been changes in the incidences of cancer in specific sites, for example, a decrease in prevalence of cervical cancer and an increase in breast cancer.

As NCDs are emerging as major public health problems in India, ICMR undertook an assessment of the disease burden of these diseases in 2004 using the DISMOD II model (ICMR, 2004b). The major data sources utilized for this exercise were medical certifications of causes of disease, a survey of causes of death (rural), cancer registry data, and review of 180 published articles, ten published reports, five unpublished reports and one personal communication dealing with diabetes, hypertension, IHD, stroke and cancers. The ICMR assessment provides national-level estimates of the disease burden of NCDs in the first five years of the new millennium.

This section of the case study reviews the available data on time trends in prevalence of hypertension, diabetes, IHD, stroke and cancers over the last two decades; ICMR estimates of the disease burdens of NCDs; and the relationship between nutritional status and NCD.

Diabetes and impaired glucose tolerance

Community-based studies on prevalence of diabetes in urban and rural areas have been conducted in all regions of the country (Figure 46); all these studies show that there has been progressive increase in prevalence of diabetes in both urban and rural areas over the last three decades.

Data from the Chennai on time trends in prevalence of diabetes and impaired glucose tolerance (IGT) in urban and rural urban populations (Figures 47 and 48) show that both have increased at escalating rates in urban and rural areas (A. Ramachandran, 2005). Potential factors associated with the higher prevalence of diabetes in urban areas are shown in Figure 49.

FIGURE 46
Prevalence of diabetes, 1971 to 1998

FIGURE 47
Increasing prevalence of diabetes and IGT in urban southern India, 1989 and 2003

Source: A. Ramachandran, 2005.

FIGURE 48
Temporal changes in prevalence of diabetes and IGT in urban southern India

Source: A. Ramachandran, 2005.

FIGURE 49
Factors associated with the prevalence of diabetes and IGT in urban areas

In 2000, the Diabetes Epidemiology Study group initiated a multicentre community-based study using the stratified random sampling method to assess the prevalence of diabetes and IGT in Bangalore, Chennai, Mumbai, Delhi, Kolkata and Hyderabad. The oral glucose tolerance test was carried out on 11 216 people (5 288 men and 5 928 women) aged 20 years and over in a representative sample drawn from all socio-economic strata. Information on socio-economic status, physical activity and anthropometric data were collected (National Urban Diabetes Survey, 2001). Age-standardized prevalence rates of diabetes and IGT are shown in Figure 50. Diabetes and IGT increase progressively with age (Figure 51), and subjects under 40 years of age have higher prevalence of IGT than diabetes (12.8 percent versus 4.6 percent, p < 0.0001). Diabetes is not usually listed as a predisposing cause of death in death certificates in India; data from hospital-based studies suggest that major causes of death in patients with diabetes are infections, renal failure, IHD and stroke.

Summary results of ICMR’s estimates of the disease burden of diabetes in 1998 and 2004 are presented in Table 20. The number of cases increased from 58.35 million in 1998 to 66.58 million in 2004 (37.73 million in urban and 28.85 million in rural areas). By 2004, diabetes accounted for 100 000 deaths a year, and is responsible for 1.15 million years of life lost (YLLs) to disease and 2.26 million disability-adjusted life years (DALYs) (ICMR, 2004b)

FIGURE 50
Prevalence rates of diabetes and IGT in India’s urban population

Source: National Urban Diabetes Survey, 2001.

FIGURE 51
Prevalence rates of type-2 diabetes and IGT in India’s urban population

Source: National Urban Diabetes Survey, 2001.

TABLE 20
Projections of disease burden of diabetes, 1998 and 2004

Source: ICMR, 2004b.

A WHO burden of disease study carried out in 2000 estimated that 2.7 million DALYs are attributable to diabetes; ICMR estimates for 2004 correspond closely to this estimate (ICMR, 2004).

Hypertension

Hypertension is probably the most common NCD, and is the most common factor responsible for IHD and cerebrovascular accidents. In the early 1970s, the reported prevalence of hypertension was low, ranging from 2 to 5 percent of the adult population. However, over the years rates have increased and currently range from 5 to 15 percent in urban adults. Yagnik (1998) showed that some Indian people are prone to developing hypertension from early childhood. Gopinath et al. (1994) investigated 10 200 Delhi schoolchildren (5 709 males and 4 506 females) aged five to 14 years and showed that hypertension existed even among this age category. Prevalence of hypertension increases with age, BMI, parental history of hypertension, and diabetes. A community-based study of hypertension (systolic BP > 140 and diastolic BP more than 85) in 6 543 people aged 15 to 25 years in Delhi in 1985 to 1987 showed overall prevalence of hypertension was 3.9 per 1 000 population (Reddy, 1998; Table 21).

TABLE 21
Hypertension rates by age and gender (thousands)

Sample size: 6 543.
PR = prevalence rate per 1 000, SE = standard error.
Source: Gopinath et al., 1994.

Results from some of the major community-based studies on hypertension over the last two decades are shown in Figures 52 and 53. There have been clear increases in prevalence of hypertension among men and women living in urban and rural areas. Prevalence is lower in rural than in urban areas.

ICMR undertook an assessment of the burden of disease of hypertension (systolic BP > 140 mmHg and/or diastolic BP > 90 mmHg), based on studies carried out between 1995 and 2002 in the urban and rural areas of different regions. Meta-analysis of the data indicated that prevalence of hypertension was 157.4 per thousand at the national level (ICMR, 2004b).

FIGURE 52
Prevalence of hypertension (SBP > 140/DBP > 90), 1959 to 2005

Sources and sample size: Padmavathy et al, 1959: 1642;Gopinath et al, 1990: 6372(Males), 7351 (Females); Kutty et al, 1993: 1130; (Females); Gupta et al., 1994: 1982 (rural males), 1166(rural females); Gupta et al, 1995: 1415 (males) 797(females); Gopinath etal.5998 (Urban males), 7136 (Urban females), 616 (Rural males), 1116 (Rural females), Singh et al, 1998: 3714; Chadha et al., 1998: 13134 (urban), 1982 (rural); Gupta et al.,2000: 1415(urban males), 797 (urban females), 1982 (rural males), 1166(rural females); Misra et al., 2001: 532; Mohan et al., 2001: 1175; Gupta et al., 2002: 550 (urban males), 573 (urban females); Ahlawat et al, 2002: 937; Reddy et al, 2002: 3307; Hazarika et al, 2004: 3180; Prabhakaran et al., 2005: 2122.

FIGURE 53
Percentage prevalence of hypertension (SBP > 160/DBP > 90), 1990 to 1998

Sources and sample size: Gopinath et al, 1990: 6372(Males), 7351 (Females); Kutty et al, 1993: 1130; (Females); Gupta et al., 1994: 1982 (rural males), 1166(rural females); Gupta et al, 1995: 1415 (males) 797(females); Gopinath etal.5998 (Urban males), 7136 (Urban females), 616 (Rural males), 1116 (Rural females), Singh et al, 1998: 3714.

Health consequences of hypertension

ICMR estimated the risk ratios for NCD that are associated with hypertension; 16 percent of IHD, 21 percent of peripheral vascular disease, 24 percent of acute myocardial infarctions and 29 percent of strokes can be attributed to hypertension (ICMR, 2004b). ICMR also computed the risks of NCDs that are attributable to diabetes and hypertension (Figure 54). Because hypertension and diabetes often coexist, the actual risk of various NCDs when both are present may be higher than the risk for either individually.

FIGURE 54
Risks of NCD that are attributable to diabetes and hypertension

Source: ICMR, 2004b.

Ischaemic heart disease

IHD, also known as coronary artery disease, is becoming an important cause of death in India. The findings of some of the major studies on prevalence of IHD in urban and rural areas in different parts of India are shown in Figure 55. Over the last three decades, there has been a progressive increase in prevalence of IHD, particularly during the last decade, especially in urban areas. Most of this increase is attributed to lifestyle changes, which have affected people in urban areas more than rural ones (ICMR, 2004b). For the purpose of ICMR’s meta-analysis of these studies, which were carried out during the 1990s, IHD was diagnosed on the basis of:

• history of documented angina or infarction and previously diagnosed CHD;
• affirmative response to the Rose questionnaire;

• electrocardiogram changes: Minnesota codes 1-1, 4-1, 5-9, 5-2 or 9-2.

FIGURE 55
Prevalence of IHD (percentages), 1959 to 2002

Age-specific prevalence rates of IHD among males and females were obtained by pooling the data of these five studies (separately for urban and rural areas), the results of which are given in Table 22. There is a steep increase in IHD prevalence in both sexes in the 40 to 50 years age group. Prevalence rates in women are similar to or higher than those in men.

TABLE 22

Age-specific prevalence derived from selected studies of IHD

PR = prevalence rate per thousand.
Source: ICMR, 2004b.

Indices of the burden of disease of IHD in India are presented in Table 23. Estimated prevalence rates are 64.4 per thousand in urban and 25.3 per thousand in rural populations. Projections of the burden of disease of IHD in India from 1998 to 2004 are given in Table 24. The number of IHD cases is estimated to have increased from 34.78 million in 1998 to about 39.43 million in 2004 (20.58 million in urban and 18.85 million in rural areas). In 2004, the total number of DALYs attributable to IHD was estimated to be 16 million (ICMR, 2004b).

TABLE 23
Indices of disease burden of IHD

Source: ICMR, 2004b.

TABLE 24
Projections of disease burden of IHD, 1998 and 2004

Source: ICMR, 2004b.

It is often assumed that IHD affects mainly the well-to-do. However, several studies suggest that poor people are vulnerable to IHD. A community-based cross-sectional survey looked at the prevalence of CHD and coronary risk factors in Rajasthan by educational level in 3 148 residents over 20 years of age (1 982 men and 1 166 women) in three villages (Gupta, Gupta and Ahluwalia, 1994). The prevalence of CHD (diagnosed by electocardiography) showed an inverse relation with education in both sexes; prevalence of coronary risk factors such as smoking and hypertension were higher among the uneducated. NSSO (1975 to 2000) surveys have documented higher prevalence of tobacco use among the poorer segments of the population (Figure 56). Lack of physical exercise and stress are common among the urban poor in sedentary jobs. It is therefore not surprising that there is high prevalence of hypertension and IHD among this segment. Results of some of the studies carried out in Delhi show that prevalence of hypertension and IHD is high among poorer segments of the population in urban areas. Some data indicate that untreated/poorly controlled severe hypertension and IHD were higher among low-income groups, perhaps because of poor access to health care; data also indicate that mortality rates associated with IHD are higher among the poor (Srinath Reddy, personal communication). It is therefore important to recognize that not only the urban affluent are at risk of hypertension and IHD in India. Programmes aimed at lifestyle modifications for all segments of the population are of critical importance for preventing IHD. Facilities for screening to detect IHD and for managing those with the disease also have to be built up.

FIGURE 56
Prevalence of alcohol and tobacco use in India, by income quintale

Source: NSSO, 1995/1996.

Stroke

WHO defines stroke as "rapidly developed clinical signs of focal disturbances of cerebral function, lasting more than 24 hours or leading to death, with no apparent cause other than vascular origin". The 24-hour threshold in the definition excludes transient ischaemic attacks. Stroke is the acute severe manifestation of cerebrovascular disease, and is one of the leading causes of mortality and morbidity in developed countries.

ICMR undertook a meta-analysis of stroke from well-designed studies with adequate sample sizes (Figure 57). The weighted average of stroke prevalence was 1.54 per thousand. Estimated prevalence of stroke is lower in India than in developed countries. However, it may increase proportionally with increasing longevity. The prevalence rates, stroke-specific mortality rates, case fatality rates, all-cause mortality rates and age distribution of population (1998) were inputs for a DISMOD analysis of stroke data.

FIGURE 57
Prevalence of stroke, 1985 to 2001

Source: ICMR, 2004b.

The YLLs to stroke are 496.3 per 100 000, and the DALYs 597.6 per 100 000 (Table 25).Projections of the burden of disease of stroke in India for 1998 to 2004 are given in Table 26. In 2004, the total number of stroke cases in India was expected to be 1.64 million and the total number of DALYs attributable to stroke 6.37 million.

TABLE 25
Indices of disease burden of stroke

Source: ICMR, 2004b.

TABLE 26
Projections of disease burden of stroke, 1998 and 2004

Source: ICMR, 2004b.

Cancers

NCRP estimates that there are about 700 000 new cases of cancer a year and about 2 million cases of cancer in the country (ICMR, 1990 to 2005). Age-adjusted cancer incidence in India varies from 91.9 to 120.9 per 100 000 in urban males and from 108.7 to 134.8 per 100 000 in urban females. The cumulative incidence rates in selected population-based cancer registries in India are given in Table 27. Over all, cancer incidence in India is among the lowest in the world. Incidences of cancers reported in the urban cancer registries are similar to cancer incidences among Indians in Singapore and are substantially lower than cancer rates reported in other countries. Cancer epidemiologists have been exploring the protective role of the Indian diet - with its high fibre, phytate and spices, including turmeric - in the observed low prevalence of malignancies in India. Cancers associated with tobacco use account for 36 to 55 percent of all of cancers in men and for 10 to 16 percent of those in women. Anti-tobacco education and reduction of tobacco use can result in further substantial reductions in cancer rates in India. Data on time trends in prevalence of cancers (all sites) from the six population-based cancer registries are shown in Figure 58. It is obvious that, unlike CVD and diabetes, there has not been any increase in overall cancer prevalence over time.

TABLE 27
Cumulative incidence rate, cumulative risk and possibility of developing cancer at all sites

Source: ICMR, 1990 to 2005.

FIGURE 58
Trends in prevalence of cancer rates (per 100 000 population), 1990 to 1998

Source: ICMR, 1990 to 2005.

The Bombay cancer registry has population-based data on incidence of cancer from the 1960s to the present (Yeole, 2001). Analysis of time trends from the 1960s until 1999 confirms that although there have been massive changes in prevalence of some cancers (reductions in cervical cancer and increases in breast cancer) there has been no increase in overall prevalence of cancers over the last five decades.

ICMR estimates of the burden of disease of cancer (all sites), based on data from NCRP’s population-based cancer registries, are given in Table 28. The number of cases of cancer in 2004 is expected to be 820 000, and the total number of DALYs due to cancer in India is estimated at 5.9 million. This estimate is low compared with the 8.6 million DALYs estimated in the WHO burden of disease study (2000) (Figure 59). To obtain cancer disease burden estimates, ICMR used mortality rates obtained by pooling the data of all six population-based registries. However, if the cancer mortality rates reported in the Chennai registry (which are the highest reported) are used, the figures become comparable to those in the WHO study.

TABLE 28
Projections of disease burden of cancer

Source: ICMR, 2004b.

FIGURE 59
WHO and ICMR estimates/projections of disease burden of cancer, 1990 to 2004

Source: ICMR, 2004b.

Tobacco as a risk factor for NCDs in India

Data on tobacco use in India from the fiftieth NSSO survey (NSSO, 1975 to 2000) are shown in Figure 60. Prevalence rates of tobacco use are highest among urban males, followed by rural males. The countrywide prevalence of tobacco use (rural and urban) is 35.5 percent.

The risk ratios associated with tobacco use in NCDs are presented in Figure 61; 15 percent of IHD cases, 48 percent of acute myocardial infarction (AMI) and 22 percent of strokes are attributable to tobacco use, which is also the major factor responsible for cancers of the lung, mouth and oesophagus. A strategy for controlling the use of tobacco would therefore result in significant reductions of these NCDs.

FIGURE 60
Use of tobacco in various forms

Source: ICMR, 2004b.

FIGURE 61
Disease risk attributable to tobacco use

Source: ICMR, 2004b.

Factors responsible for the emerging problem of overnutrition

Data presented in the section on food and nutrient intake indicate that over the last three decades there has not been any significant change in the energy intake of the Indian population, except in affluent families, especially in urban areas; even in this segment, however, most of the increase in consumption of energy-dense fast foods is among adolescents and youth. It is therefore obvious that increase in dietary intakes of fats, oils and sugar is not a major factor in overnutrition in India. Over this period, there has been a progressive reduction in physical activity in all segments of population. Reduction in energy expenditure and unchanged dietary intake results in a positive energy balance, and could be a major factor responsible for the rising prevalence of overnutrition in adults in India. Available evidence to support this is reviewed in this section.

Cross-sectional studies undertaken among affluent housewives aged 30 to 60 years in Delhi show that their dietary intake remained unaltered, at between 2 100 and 2 300 kcal/day (Wasuja and Siddhu, 2003). In each age group, energy expenditure is lower than intake by about 50 to 75 kcal/day. This is associated with a weight gain of about 5 kg per decade (Table 29). The women were not making any conscious effort to increase physical activity or take up regular exercise. It is possible that a similar situation exists among men in these segments of population. A small but persistent positive energy balance accounts for the slow but steady weight gain in adults among affluent segments of the population.

TABLE 29
Energy intake and expenditure in affluent urban housewives

Source: Wasuja and Siddhu, 2003.

During the last three decades, there have been a progressive decline in the poverty ratio and a steep increase in per capita income. Economic improvement inevitably results in improved purchasing power, including the ability to purchase and consume higher-value food items. This, in turn, can lead to some increase in the energy intake from fats, sugar and refined carbohydrates, and reductions in the energy intake from complex carbohydrates and in dietary fibre. Simultaneously, there has also been a reduction in physical activity and perhaps an increase in work-related stress because of changes in occupation. This combination of factors might be responsible for some of the rapid increase in overnutrition and hypertension in segments of the population that are emerging from poverty. It would also apply to rural migrants who settle in urban areas.

It is well documented that Indians have higher body fat per BMI compared with Caucasians. Prevalence of abdominal obesity is higher in India. Both overnutrition and abdominal obesity are associated with increased risks of hypertension, diabetes and CVD.

It is however important to remember that the seeds of obesity in adult life are often sown decades earlier. The thrifty gene hypothesis proposes that populations who have faced energy scarcity over millennia may have evolved so that the majority have the thrifty gene, which conserves energy. If energy intake of people with this gene obtain adequate or excess energy intake, they lay down fat, develop abdominal obesity and insulin resistance - which may progress to diabetes - and incur risk of hypertension and CVD. Barker’s thrifty phenotype hypothesis puts the evolution of thriftiness into the intrauterine period; Indians with one-third low birth weight rate can be deemed to have acquired the risk of this metabolic syndrome before birth.

Yagnik and colleagues in Pune explored the relationship between low birth weight and glucose and insulin metabolism using the oral glucose tolerance test on 477 children born in KEM hospital, Pune (Yagnik, 1998). They found that Indian newborns weighed less because they had low muscle mass and small abdominal viscera. However, they also conserved their subcutaneous fat. At four years of age, plasma glucose and insulin concentrations 30 minutes after glucose administration were inversely related to birth weight (Table 30), and directly related to current weight and skin fold thickness. The relationship between glucose/insulin and birth weight was independent of current weight. Thus, poor intrauterine growth with relatively excess growth later was associated with metabolic endocrine abnormalities, which could lead to diabetes in adult life. Adolescent obesity is well-documented in both urban and rural areas and may be the stepping-stone to adult obesity.

TABLE 30
Birth weight, plasma glucose and insulin concentrations in four-year-old urban children

Source: Yagnik, 1998.

In a study in urban Delhi, Bhargava and colleagues found that low- and middle-income adults who were undernourished in infancy, childhood and adolescence were prone to develop overweight, abdominal obesity, hypertension and diabetes by the time they were 30 years of age (Bhargava et al., 2004) (Tables 31 and 32).

TABLE 31
Trends in nutritional status of the Delhi cohort

Source: Bhargava et al., 2004.

TABLE 32
Current status of the Delhi cohort

Source: Bhargava et al., 2004.

The lesson to be learned from these data is that it is never too early for Indians to start practising healthy lifestyle and dietary habits. Early detection and correction of undernutrition, until children attain appropriate weight-for-height is essential to promote linear growth. Adolescents and adults should ensure a balanced diet with no more than adequate energy intake. Exercise has to become part of the daily routine in order to promote muscle and bone health, as well as to prevent the development of adiposity in all age groups.