ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Tamil Nadu: ICDS with a Difference

Tamil Nadu: ICDS with a Difference

Tamil Nadu's noon meal programme has evolved via the successful intervention of twin pressures - political will that ensured budgetary provision for the programme and the widespread demand for it from below. Several nutrition-oriented programmes have now expanded to cover groups such as pregnant and nursing mothers, old age pensioners, widows and the destitute. The need is now to create a "nutrition literate" populace, i e, foster an awareness of non-food factors that in several ways influence behaviours and attitudes related to food.


ICDS with a Difference

Tamil Nadu’s noon meal programme has evolved via the successful intervention of twin pressures – political will that ensured budgetary provision for the programme and the widespread demand for it from below. Several nutrition-oriented programmes have now expanded to cover groups such as pregnant and nursing mothers, old age pensioners, widows and the destitute. The need is now to create a “nutrition literate” populace, i e, foster an awareness of non-food factors that in several ways influence behaviours and attitudes related to food.


ombating hunger and reducing the prevalence of malnutrition are a goal common to many national plans and is now part of the internationally accepted development goals, drawing from the Millennium Declaration. A specific target under the Millennium Development Goals is to “halve, between 1990 and 2015, the proportion of people who suffer from hunger”. From a human development perspective, good health and nutrition are inherently valuable, contributing to physical and cognitive development, facilitating educational attainment and overall wellbeing, and expanding human capabilities for present and future generations. Even from a more limited economic growth perspective, better nutrition decreases the incidence of non-communicable diseases and healthcare costs, reduces loss of worktime due to absenteeism and improves productivity.

While nutrition-oriented programmes operate in many countries, they often suffer from low coverage, resource limitations and inadequate focus on many critical non-food factors – protected water, sanitation, women’s status, political commitment, among others. Over 150 million children under age five in the developing world are underweight, reflecting the status of the population as a whole. While income poverty has declined steadily in most countries, malnutrition

Economic and Political Weekly August 26, 2006

has been harder to combat. In south Asia, hunger and malnutrition have been especially resistant to change and child malnutrition has tended to be high and persistent. Here the share of children under five who are moderately or severely underweight (almost 50 per cent) is much higher than the share of population living under a dollar a day (about 33 per cent). The incidence of undernutrition among children under five is as high as 47 per cent in India, 48 per cent in Bangladesh and 29 per cent in Sri Lanka [UNDP 2005].1 If micronutrient deficiencies, anaemia among women and adolescent girls and various forms of lifestyle-related adult malnutrition are added, the incidence of undernutrition is even larger. Thus, the problem of malnutrition is wider and more persistent than income poverty,2 indicating that poverty reduction by itself will not eliminate widespread malnutrition, and that direct interventions have to go beyond targeting the poor. The contrasting east Asian experience, and the diversity in south Asia itself, show that it is possible to combat malnutrition among children in a relatively short period.

It is in this context that Tamil Nadu’s Integrated Child Development Services (ICDS), and its broader noon meal programme (which is fully integrated with ICDS for children under six), present valuable experiences for the nutrition community and policy-makers.

Tamil Nadu Experience

Tamil Nadu (TN) is different from other Indian states. In India’s federal structure, it is possible for an individual state to have priorities somewhat different from the centre. Combating child hunger and malnutrition became political priorities in Tamil Nadu well before judicial intervention triggered responses at the centre. Today, virtually any child between the ages of 2 and 15 years is eligible for a daily hot lunch at the cost of the state. Most of the feeding takes place outside the home: in ICDS centres (‘anganwadis’) for children under six, and at school for older children.3 Other vulnerable groups like pregnant and nursing women, the destitute and pensioners are also covered. In addition, a range of complementary services are available for children under six and pregnant and nursing women through the ICDS, which is fully integrated with the noon meal programme.

Six aspects of the TN case are of particular interest. One, political will exerted pressure from above, resulting in public policy attention backed by budgets to combat hunger and malnutrition. TN is perhaps the only Indian state where this issue has entered the political discourse and remained there consistently for over four decades. Two, the programme has become very popular, catalysing pressure from below and making it difficult, say, for an anganwadi to remain closed without immediate enquiry. Three, while political will led to pressure from below, in turn, contributed to retention of political will over time, regardless of the party in power. Four, while visible public feeding centres have the potential to woo voters in a democracy, if well managed, visibility itself contributes to maintaining quality through pressure on local staff and higher officials. The interests of nutritionally vulnerable groups in Tamil Nadu have coincided with prospects for electoral spoils [Rajivan 2004]. Five, near-universalisation contributed to de facto child rights to nutrition. Since malnutrition is not confined to poor households, universal coverage was more effective than targeted schemes. Six, near-universalisation paved the way to other positive achievements also. In particular, well managed ICDS centres produce school – ready six-year olds with relatively well developed social and cognitive skills that contribute to better education later on. They also contribute to social equity through common dining, especially in rural India’s caste-class conscious context.

Thus, the TN case may be described as a “sandwich”: a unique combination of pressure from above through political will, and from below through public expectations, with a wide network of feeding centres in between. With a population of over 62 million in 2002, TN is larger than many developed and developing countries like the UK (60 million), the Republic of Korea (48 million), Nepal (28 million), Malaysia (24 million) or Sri Lanka (20 million). TN’s experience can provide useful pointers about what works and what does not, including the needed resources of funds, personnel and time, not only to other states in India, but also to other similarly placed developing countries that wish to address child malnutrition and promote child development.

More than 70,000 feeding centres operate every day in the state, providing a hot mid-day meal to around 8 million persons.

Of these, over 30,000 centres are exclusively for children under six (the most elusive age group), mostly under the ICDS.4 Here a wider range of services is provided for over a million under-six children – including complementary nutritious feed to pre-schoolers, pregnant and nursing women; growth monitoring; health services; pre-school education; and communication. Around 1.5 million young children and women benefit every day. An impressive physical infrastructure is in place as well as extensive staff. Over time the government of Tamil Nadu (GoTN) successfully leveraged the massive network of centres created under different schemes with varying objectives and priorities (the noon meal programme, the Tamil Nadu Integrated Nutrition Project (TINP) and ICDS), and consolidated staff, infrastructure and administrative capacity to focus on nutrition security. Nutrition has come to occupy a permanent place on the state’s political agenda, with de facto child rights to food having become operational. GoTN is the only government that has an official policy for making the state malnutrition free, i e, reducing malnutrition to such a level that it is no longer a public health issue.5

Evolution of Direct Nutrition Investments

Evolution from hunger to nutrition: Tamil Nadu has a long history of direct nutrition investments through provision of food to children outside the home. This started with the aim of combating “classroom hunger” for school children, but over the years it became integrated with the larger goal of addressing malnutrition and promoting overall child development [Rajivan 2004]. School meals through public contributions were introduced in 1956, and soon evolved into joint publicprivate funding. Donor assistance was introduced in 1961, helping to expand coverage. But school feeding initially faced many operational problems due to inadequate staffing, infrastructure, training and support. For instance, headmasters were in charge and teachers and children were involved in cooking, resulting in serious diversion of teaching-learning hours. Lack of staff and poor storage also led to wastage of ingredients and questionable hygiene. Central kitchens changed operations dramatically, with pre-packed food packets delivered to schools. While this improved hygiene,

Economic and Political Weekly August 26, 2006 standardisation and teaching-learning, it was more expensive and reduced the number of feeding days.

In parallel, with financial support from the central government, ICDS was introduced in 1976 on a small scale (in just three blocks), covering children under six for the first time and combining supplementary nutrition with other child development services. Shortly after ICDS began, in 1980, the TINP was introduced on a pilot basis in another block, with funding and technical support from the World Bank. TINP focused on an even younger age group: children under 36 months of age. These early ages are the hardest to reach

– being very young, children under six do not assemble every day at a place like a school and need an adult to bring them to a facility. Yet, from a nutrition perspective, this is efficient as interventions reach the most important age group. The preschool age, especially below 36 months, is a nutritionally most vulnerable period when rapid growth takes place. Deprivations in this period can cause irreversible damage which later feeding cannot fully compensate.

The introduction of the noon meal scheme in 1982 was a watershed which marked the beginning of a continuous expansion to the pre-school ages, moving towards universalisation, provision of dedicated staff and infrastructure, systematic training, and budgeting for recurring and nonrecurring expenses. Without waiting for the GoI led expansion of the ICDS, or the World Bank led TINP, GoTN proceeded to expand the noon meal programme which was fully under its control, eventually combining it with ICDS and integrating all feeding schemes for children under six. After 1982 the state has never looked back.

While hunger was well understood, initially the concept of nutrition was not. Pressure from below to provide nutrition security did not exist the way it did for the provision of food to combat immediate hunger. Consequently, the shift of focus from hunger to malnutrition did not happen easily or automatically. The nutritionoriented schemes (the ICDS and TINP pilots) and the hunger-oriented noon meal scheme expanded and operated in parallel. To begin with, nutrition experts undertaking evaluation and providing advice on nutrition schemes remained separate from political decision-making in the social welfare sector. But after about three decades of experimenting with different models, linkages were consciously forged between what was electorally attractive (visible, publicly funded feeding schemes) and what was technically recommended (multi-sectoral nutrition schemes), and different programmes eventually became fully integrated.

While administrators and scientists initially had strong reservations about the noon meal scheme because of its logistical and financial requirements, sheer political will saw it through the initial years. Today, the nutrition-cum-child-development infrastructure is so well entrenched in Tamil Nadu that it has become a de facto entitlement, not seen in other Indian states. Coverage: Coverage, in terms of both children and adults, has expanded steadily over the years. Pregnant and nursing women are eligible under the ICDS. Old age pensioners, the disabled, widows and destitute also avail of the daily lunch as the marginal cost of adding them to the already large list of eligible persons was found to be relatively small. Both rural and urban areas are covered. Attendance records indicate that nearly 1.7 million persons (1.3 million children aged 2-6 years and 0.4 million adults) benefit from ICDS services including the daily noon meal.6 If the noon meal for school children is also accounted for, then the state feeds a population of over 8 million (including 7.6 million children) almost everyday through a network of over 71,000 centres.

As mentioned earlier, virtually any child aged between 2 and 15 years is now eligible for a noon meal at the cost of the state. Though there used to be an income criterion, in practice, any willing child in the eligible age group has been allowed to participate. This was done for socio-cultural and practical reasons – it would have been hard to feed some children and deny others based on some BPL list when all children watch the cooking and serving of food. In practice, it is children from poorer households who predominate in the ICDS centres and in publicly funded schools. One estimate suggests that around 33 per cent of pre-schoolers actually avail the benefits. Infrastructure and staffing: The importance of good infrastructure with separate staff was not lost to policy-makers, especially given the earlier experience with school feeding. Noon meal premises had to be suitable for children to assemble, play and rest, for storing food commodities and records, and for cooking, with adequate water supply facilities. It was not always easy to find suitable buildings. Most centres operate in panchayat union buildings or other public buildings, specially constructed feeding centres, or rent-free private buildings. Where no suitable public building existed, a small rent of Rs 50 per month covered the use of private buildings. School centres are run in the school buildings. Centres are equipped with cooking utensils, plates and tumblers, play equipment, weighing scales and mats for children to sleepon. For pre-school education charts, toys, etc, are provided. Local communities also contribute – e g, by providing cooking gas to replace firewood, additional toys, kitchen gardens, etc. Former noon meal programme centres have become available for ICDS and vice versa, facilitating integration. The programme has one of the most extensive infrastructure network of centres, equipped and fully staffed.

Each centre has its own staff for cooking, cleaning and pre-school education, both for the ICDS (all women) and school meals (both sexes). Integration of different schemes under the ICDS led to considerable variation in staffing at the centre level.7 The need for staff rationalisation based on work load, though well understood, was difficult as it involved sensitive problems of relocation and retrenchment. Moreover, the numbers are large – around 1,00,000 women, who themselves are now an effective pressure group, long perceived as overworked and underpaid. Full integration took place more easily at the village level, with no overlap or wastage, but took time to fall into place at the district and state levels. For a while a district could have both ICDS and TINP offices and staff, functioning independently. At the state headquarters two independent programme heads led to different guidelines, and variations in service delivery based not on variations in local needs, but due to the different programmes. A full merger was achieved in 2003 with operations under a single management through a single set of personnel. Food composition and its costs: Hot rice cooked with dhal, soyabean flour, vegetables, oil and condiments prepared daily on the spot is served to all participants. Periodically new ideas and recipes were tried out including a weekly egg, occasional ‘uppama’, sweets on special occasions, etc. The scale of rations is higher for older children and adults. Minimum calorieprotein values are prefixed to provide

Economic and Political Weekly August 26, 2006

one-third of total daily requirements. Pregnant and nursing women and children aged two or less are provided with a nutrientdense complementary feed – ‘sathu maavu’

– that also provides a predetermined nutrient composition per 100 grams. School children eat their noon meal at school. The other categories, i e, pre-schoolers, pensioners and pregnant and nursing women eat at ICDS centres (anganwadis).

The cost of ingredients of the basic meal ranges between Rs 1.38 per head per day (for the youngest age group) and Rs 2.32 per person per day (for adults), excluding administrative overheads, but including transport charges. Thrice a week the addition of potato, bengal-gram and green-gram dhals, one item each on alternate days, adds around Rs 0.19 per head with corresponding calorie-protein addition. The sathu maavu consists of a mix of roasted cereals, gram, jaggery and a vitamin premix based on a prescribed protein-calorie scale and costs about the same – around Rs 1.61 per 100 grams per head [Rajivan 2004].

The sheer logistics of managing these operations are mind-boggling – procurement, delivery of grains minimising crisscross movements to the vast network of village centres and urban slums, ensuring stock adequacy, regular replenishments, cooking, etc. Staff capacity at different levels built on the basis of experience and training helped. The scheme has grown and stabilised over time.

The noon meal is better described as a substitute rather than a supplement as it is in lieu of a home meal. Yet there is some additionality from higher quantity and quality as compared with what the children get at home. For the poorest, more likely to come hungry to the anganwadi or school, the additionality is the greatest.

Regardless of the party in power, successive governments have consistently maintained or increased the level of budget support to nutrition.8 The programme has a first call on the state’s resources. Expenditure on many of the complementary social services (public health, water and sanitation, labour and employment, education, etc) has also followed the same pattern, sometimes growing faster than expenditure on nutrition programmes, and supplementing them by addressing many of the non-nutrition factors important for nutrition security. Quality – Tamil Nadu is different: Tamil Nadu’s experience with ICDS indicates that the sandwich situation, with pressure from above and below (a combination of political will and high expectations from participating households), not only ensures budget support to stated policies but also puts in place accountability. As discussed in the editorial introduction to this collection, this has contributed to higher quality of services compared with most other states, especially the northern states. For instance, ICDS centres in Tamil Nadu are open longer, those below three attend more regularly, pre-school education is prevalent in most centres, infrastructure is much better, and workers are regularly paid.9

Weight-for-age data indicate a trend reduction in malnutrition among participating rural children – a joint outcome of nutrition interventions and overall development.10 Data for the period 1983 to 1999 among former TINP participants show a steady decline in severe malnutrition (grades III and IV) to almost zero – from a little over 12 per cent to half of one per cent. The proportion of moderately malnourished (grade II) children also declined, from over 35 per cent to under 10 per cent, while the percentage of children in the “normal” category went up from under 20 to over 50. This is a significant achievement.

Other achievements of ICDS in Tamil Nadu include promoting social equity due to regular common dining, washing, sleep and play (in spite of occasional incidents), and building skills among preschoolers for school readiness. Primary schoolteachers tend to prefer children who have had a few years in anganwadis as these are seen to build social and cognitive skills among children. In fact there are complaints from some ICDS staff that, in the interest of school enrolment targets, primary schoolteachers sometimes entice ICDS children to join regular school prematurely! Second-generation issues: It is not as if Tamil Nadu’s ICDS does not face problems. The network of ICDS centres faces second-generation issues, such as poor access in remote locations, staff motivation and training issues, and limited participation of local bodies and communities. In recent years, a trend towards private nurseries has created a further mismatch between demand and supply in some locations.

Moreover, in spite of a trend reduction in malnutrition observed among programme participants, malnutrition in the state has proved to be persistent. According to the second National Family Health Survey (NFHS II), over 36 per cent of children under three in Tamil Nadu are underweight, and one-third are stunted. Further, 23 per cent of new-borns are categorised as low birth weight (below 2.5 kg at birth). New batches of underweight babies further add to the probability of malnourishment in future. Adding other correlates, including social and lifestyle-related factors which are prevalent among the better-off sections of the population as well, reasons for policy concern continue.

The state’s focus has been more on managing malnutrition after it sets in. Perhaps, simultaneously, greater attention is needed for prevention (without discontinuing the management of prevailing malnutrition). While the state is committing resources to counter food deprivation, further improvements could depend on better addressing non-food factors.11 In particular, there is a need to facilitate better nutrition education for behavioural change, enabling individuals to make pro-nutrition choices – a knowledge-based approach.

Pointers for Future

Widening nutrition knowledge, and more generally, knowledge about child development and the human life cycle seems to be the missing ingredient in the case of Tamil Nadu, as elsewhere. Food can be “delivered” to beneficiaries; nutrition status cannot – it is an outcome of a number of factors, food and nonfood, including actions of people themselves that substantially influence healthnutrition outcomes.12 Nutrition-related knowledge cannot be taken for granted even among better-off sections of the population, underlining the relevance of a knowledge-based approach. This critical ingredient – nutrition literacy – costs far less but constitutes the strategic “software” required to enhance the effectiveness of per capita income growth and improved access to food.

Equally important is the ability to seek out information and to segregate accurate information from the inaccurate. As the media bombard consumers with slick advertisements for various convenience foods, human milk substitutes, pills and potions for growth and strength, it is important to develop a discriminating eye based on sound knowledge. There are also many inaccurate traditional beliefs that

Economic and Political Weekly August 26, 2006 inhibit consumption of low-cost nutritious food, and need to be countered.13 This is even more important for the poor, who might end up spending a high share of their limited resources on nutritionally inferior products. Hence, the importance of nutrition education, along with other nutrition inputs.

Recognising this, GoTN has taken the first step of moving towards a malnutrition free state, i e, outcomes in line with international norms so that malnutrition is no longer a public policy concern by 2020. Goals for indicator values that match the levels of the best performing countries in the world are incorporated in GoTN’s official nutrition policy of 2003, under which the government intends to reduce human malnutrition of all types, including subclinical deficiencies, to match international norms [Rajivan 2004].14 Indicators are comprehensive and include percentage of low birth weight; underweight and stunting (0-3 years); anaemia among adolescent girls, 0-3-year olds and pregnant women; average weight and height among adolescents; and body mass index among men and women. These need to be monitored on a regular basis. Setting up of a monitoring mechanism based on panel rather than cross-section data would also be useful, apart from being of great scientific interest.

Leveraging the state’s relatively well functioning network of centres, staffing and budgets, based on political will and public expectations, and combining it with a “knowledge approach” to child nutrition, are the suggested future directions for TN. In respect of other states, the GoI, and countries similarly placed, the TN experience underlines the utility of a “sandwich” situation – combining pressure from above and below for the public good. It also provides a body of experience which can facilitate more efficient operations on a shorter timeline, learning from some of the pioneering attempts in TN and also sharing experiences with others.




1 South Asia’s experience contrasts with the progress made in east Asia, where, for instance, the number of malnourished children declined from 24 to 10 million between 1995 and 2003. In 2003, the proportion of underweight children was around 10 per cent in China, 10 per cent in Thailand and 12 per cent in Malaysia (in the Philippines, it was

higher at 31 per cent). 2 See, for example, Rajivan (2004). 3 It is only in remote locations, or wherepopulation

density is very low, that “take-home rations” are provided instead of cooked meals.

4 The non-ICDS child welfare centres are very few (around 718 covering around 30,000 children and 3,000 adults) and are restricted to urban areas where TN has introduced services but the GoI has yet to “officially sanction” ICDS. This is another indication of policy and programming in TN being ahead of national thinking.

5 An earlier 1993 draft document was thoroughly reviewed with technical inputs from UNICEF and a number of stakeholders from within and outside government. This led to the development of a policy document and its adoption in November 2003.

6 The population of children 0-6 years is estimated at 7.2 million in the 2001 Census for TN. The noon meal coverage starts at age 2 plus while other services including complementary nutrition for the below 2 (sathu maavu – a nutritious premix powder), health services, growth monitoring, nutrition education, etc, are separately available to all. Approximately one-third of the child population in relevant ages seem to access these services.

7 Integration was not an issue for the school centres and the 714 urban pre-school centres outside ICDS and TINP. Here there is a separate noon meal organiser, a cook and a helper in each centre.

8 State budget data from the finance department of the government of Tamil Nadu for the years 1992-93 to 2002-03 are presented in Rajivan (2004). For other years it can be compiled from the Statistical Handbooks of Tamil Nadu.

9 See particularly Table 6 in Jean Drèze’s editorial introduction.

10 Weight-for-age data among participating children over 17 years were obtained from the Office of the Project Coordinator, World Bank ICDS III, government of Tamil Nadu [Rajivan 2004].

11 Examples of some non-nutritional factors that impinge on nutritional status include: the synergistic relationship between infectious disease and nutritional status; poor personal hygiene that contributes to infections; poor sanitation and waste disposal; biological factors of parity and spacing and inadequate weight gain during pregnancy; inadequate knowledge and capacity to seek out information; and social factors like the relative neglect of females in food, care, health and education, early marriages of girls resulting in teen pregnancies, and inadequate adolescent care for females. All these also contribute to the birth of underweight children.

12 That is why immunisation is easier to universalise than good nutrition status – in the former case it is enough for people to be relatively passive participants in public health schemes, while in the latter case a much more active role is needed to complement government efforts.

13 Erroneous beliefs about nutrition are many. Some examples: soda and other carbonated drinks cure gas; peas and dhals are too “heavy” and so unsuitable for young children and the elderly; sprouts are suitable for the elderly, not the young; iron tablets in pregnancy can result in a dark-skinned child; pregnant women should not eat papaya as it causes abortion; tea is good for health; banana can lead to diarrhoea or colds.

14 Malnutrition free implies reaching international standards, not zero malnutrition, which is biologically impossible.


Rajivan, A K (2004): ‘Towards a Malnutrition Free Tamil Nadu: A Case Study’ in M S Swaminathan and Pedro Medrano (eds), Towards a Hunger Free India: From Vision to Action, EastWest Books, Madras.

UNDP (2005): ‘Millennium Development Goals Report’, New York.


EPW seeks the help of contributors for quicker processing of articles submitted for publication.

Soft/Hard Copies

Contributions of less than 4,000 words for the Commentary/Discussion sections can be sent by email to. There is no need to send hard copies of short articles. Articles longer than 4,000 words for Perspectives/Special Article sections, however, must be sent in hard and soft copy formats. Soft copies can be sent by email or on a CD/floppy. Please note that EPW will not be able to process articles longer than 4,000 words unless both soft and hard copies have been received.

Revision of Articles

Authors are requested to send us articles that are complete and up to date in all respects. They are requested not to send us revised versions – by email or post – unless there are major developments that require changes to the article first submitted. Since revisions pose difficulties in managing submissions and processing the correct version, EPW requests authors to send in articles that have been finalised and are complete in all respects. Revised versions will not be processed other than in exceptional circumstances.

(Please also see Notes to Contributors for additional details)

Economic and Political Weekly August 26, 2006

Dear reader,

To continue reading, become a subscriber.

Explore our attractive subscription offers.

Click here


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top