Transforming delivery

How to choose the best interventions and delivery strategies to scale up nutrition?

Challenging dominance: identity politics in the Integrated Child Development Services (ICDS) Programme, India

By Shilpa Deshpande, PhD Candidate, Institute of Development Studies, University of Sussex

Ten years before I joined the ICDS as an anganwadi worker, my cousin mother-in-law used to work here. At that time, the division of the village population between anganwadi workers was such that lower caste households were served by my mother-in-law whereas only the higher caste households were served by Lata madam, this was her rule….so… my mother-in-law’s field area was scattered across the village. Then when I joined, Lata madam said that just like my mother-in-law I should be given the lower caste communities. I refused…. I said give me any part of the village but I want half and I want it along a continuous line then only will I be able to work. This led to a fight, which continued for several days.

Meena, Anganwadi worker

Meena* is a young, field worker of the Integrated Child Development Services (ICDS) programme in Mangaon* village of Sundar* block, Aurangabad district, Maharashtra. She has been recruited by the Sundar block* administration to cater to the health, nutrition and education needs of pregnant and nursing mothers, 0-6 year old children and adolescent girls in a population of about 650 persons through a small village based centre, the anganwadi. She is one of 3 anganwadi workers in the village and joined the programme in 2010. The other two anganwadi workers were recruited in the late 80s when the ICDS programme began in the block. Meena belongs to the Scheduled Castes†, whereas Lata*, is from the dominant Maratha caste. I met Meena in 2015 as part of ethnographic fieldwork that I conducted for my PhD on understanding the governance of the ICDS programme. In the interview extract quoted, she shares that because she belongs to the Scheduled Castes, the existing anganwadi worker from the dominant caste refused to allocate higher caste households to her and insisted that she focus only on the Scheduled Caste households.

Meena’s experience is rooted in the larger Maratha-Dalit politics of the Marathwada region of the state, of which Aurangabad district is a part. Marathas have traditionally been the dominant caste in the state and more so in this region. They constitute almost 40 percent of the population, are some of the largest landholders, often holding a majority of the land and are also politically dominant. Marathas have historically occupied a majority of seats in the Zilla Parishad (district government), state legislature as well as the cabinet! Besides this, Marathas also have a historically nurtured self-image as rulers and leaders. For Waghmore (2013) the last is the most significant and the key in understanding their aggression towards the Scheduled Castes or Dalits. In Mangaon village, where Meena lives and works, and which has always had a Maratha Sarpanch (head of the village self-governance committee) there have been at least two reported incidents of Marathas attacking Dalit neighbourhoods, killing and destroying property.

Social mapMaratha dominance in the ICDS programme is reflected in that a majority of anganwadi workers in Meena’s beat‡ are Maratha even though the anganwadi worker recruitment process is designed to favour candidates from the historically marginalised groups. Political patronage, caste and kinship networks work to benefit the Marathas. Maratha anganwadi workers, whom I studied, routinely discriminated against women and children from the Scheduled Castes and Scheduled Tribes, not visiting their areas and refusing to serve them. Because of their political connections, they were also able to violate rules with impunity, often remaining absent with little or no action taken.

There is existing research on how dominant castes such as the Marathas colonise the local state to corner resources and opportunities. But Meena’s experience is significant because it emphasises that the local state, in this case the ICDS, provides a stage for the performance of caste identities – for asserting, challenging and renegotiating identities. Meena, a slight young woman, with bright eyes and an infectious smile, took on Maratha dominance in the ICDS refusing to play by the informal rules and demanding an equal footing as a programme worker. She successfully solicited community and bureaucratic support to change the decades old practice of allocating households to ICDS workers by caste.

…. I started asking the people of the village if they had a problem. I said when you admit your child in the first standard and the Zilla Parishad (District Government) teacher comes do you ask them which (social) category they are from? Should we do this, should children be placed according to the (social) category of the teacher? They said no, there is no such problem, let the children learn, on the contrary your education is better than hers so let the children be with you….I got my survey area nearly 4-5 months (after joining)… I asked Madam (the Supervisor) to come and show me my area…..then I was able to start work. (Meena)

However, Meena’s challenge to caste roles did not go unanswered. Over the next several months Lata tried in different ways to appropriate her labour and resources reproducing the material and cultural basis of her social power. One day after a particularly nasty and bitter argument, Lata attacked her saying:

…. you demand half of my survey, you tell me what to do….. its gone to your head, you should never have been recruited, stop this, working by the rules… Chambhar, what did you think just because you move with the Marathas…. (Meena)

Meena filed a case against Lata under The Scheduled Castes and Tribes (Prevention of Atrocities) Act of 1989 – social power was countered formally by recourse to state provided legal instruments. However, Meena had to subsequently withdraw the case as Lata threatened to file a case of sexual harassment against her husband.

Meena’s experience underscores the social embeddedness of state functionaries highlighting the ways in which the state becomes an extended arena for struggles of dominance and power. These struggles in turn shape the implementation of state programmes such as the ICDS and their governance.


Waghmore, S., 2013. Civility against Caste: Dalit Politics and Citizenship in Western India. SAGE Publications India Pvt. Ltd, New Delhi.


[*] The names of persons, villages and the block have been anonymised to maintain confidentiality.

[†] Caste is a religiously sanctioned form of social stratification such that the higher castes are considered socially pure and the lower castes polluting. The Scheduled Castes are historically marginalised groups identified by the Indian Constitution for state support including affirmative action.

[‡] A beat is a set of 20-25 anganwadis under a single Supervisor.

Pastoral Community Platforms as channels for Behavioural Change for Nutrition

New research just been published by Transform Nutrition and Vétérinaires Sans Frontières Suisse with preliminary findings of a study on Pastoral Community Platforms as Channels for Behavioural Change for Improved Nutrition (BCIN) The research was designed to explore the potential of using pastoral-community platforms for channeling maternal, infant and young-child nutrition messages to community members. We have also published some useful case studies which provide an insight into the perceptions of the participants of the study who live in the study area.

Bangladesh : Delivering for Success at Scale

Conference on Delivering for Success at Scale 7- 8 February 2017, BRAC Auditorium, Mohakhali, Dhaka, Bangladesh

This conference built on the longstanding partnership between BRAC, ICDDR,B and the Institute of Development Studies (IDS) to explore the role of knowledge in the conceptualisation, design, delivery and management of development programmes and policies. Bangladesh has been a lead player in the use and integration of research into development decision-making, notably in relation to health systems, nutrition, social protection, agriculture and poverty. Evidence will continue to be important in strategic development planning by the Government of Bangladesh among others.

The conference included examples of how high quality research can be embedded within programmes, showcased research findings, discussed the challenges of integrating research with programmes, highlighted successful examples and discussed future needs. The UK Department for International Development funded Transform Nutrition and Future Health Systems presented relevant research.

Read an article about the conference in the Bangladesh Financial Express ‘On health and nutrition’. We also published a blog by Rittika Brahmachari, IIHMR Pathways to scaling up interventions: reflections from delivering for success at scale 

For more information contact Aazia Hossain

What factors influence community nutrition workers in performing their jobs? Preliminary findings from Bihar, India

by Aparna John, PhD candidate, Institute of Development Studies at University of Sussex

Child undernutrition rates in India are among the highest in the world (Raykar, 2015). Despite the decline in undernutrition indicators such as stunting (low height for age), underweight (low weight for age) and wasting (low weight for height), India is home to 40 million stunted and 17 million wasted children. The Government of India has invested in efforts to reduce childhood undernutrition and improve maternal and child health through one of its flagship programmes — the Integrated Child Development Services (ICDS) Scheme. ICDS, the world’s largest community nutrition programme, is delivered by 1.34 million village based female workers, Anganwadi workers (AWWs).

The evidence on the ICDS suggests that it is performing sub-optimally—implementation of the interventions is often poor leading to low utilisation and low coverage (PEO, 2012). Considering AWWs are responsible for the delivery of the ICDS services, improving their performance is critical for improving the overall programme performance. However, the evidence base on what shapes AWW performance is scarce. What are the facilitators and barriers in their individual, programme, community, and organisational contexts that influence their motivation and performance? These questions are either unanswered or partly answered by studies looked at the overall ICDS implementation. This research lacuna was the key reason I chose these interesting yet unanswered questions for my doctoral study. In this blog post, I would like to present preliminary findings from the explorative qualitative study I conducted to answer these questions.

In the study, I defined performance as ‘AWWs making services accessible to the required number of beneficiaries with quality (i.e. adhering to guidelines)’ derived from the literature on Community Health Workers (CHWs) and health workers in general. I interviewed thirty AWWs from varied education backgrounds, caste status, and distance (Gram Panchayat distance from the block headquarters) from two administrative blocks of one of the southern districts of Bihar. Bihar is a state grappling with a high burden of undernutrition and demonstrating renewed efforts to improve its social welfare programmes in the last decade. As shown in Figure 1, AWWs deliver a varied set of services to different beneficiary groups. I analysed the data using the hybrid method of deductive and inductive thematic analysis.

Figure 1 Anganwadi worker in Bihar

Parth Sanyal/Save the Children

Parth Sanyal/Save the Children

  • Qualification: minimum 10th pass
  • Honorarium: Rs. 3000 per month
  • Main job responsibilities:
    • Pre-school education for forty  3-6 years
    • Supplementary nutrition
      • Daily lunch for forty 3-6years
      • Monthly dry rations to eight pregnant, eight lactating women and forty 6-36 months children
    • Counselling (individual and group)
    • Preparation for monthly Village Health and Nutrition Day to facilitate immunisation in coordination with ASHAs and Auxiliary Nurse Midwife (ANM) and growth monitoring of children and pregnant women
    • Identification and referral of Severely and Acutely Malnourished (SAM) children

Preliminary key findings from the study suggest that the following factors in the individual, programmatic, and community contexts influence an AWW’s motivation and performance:

• Individual:
Initial financial motive rooted in family reasons as the reason for continuing with the job: Whilst some AWWs actively chose to take up the job due to “moral” motives (the opportunity to positively impact their community), for many it was their families who were the driving force. Their families tended to be more focused on the “social” motive – the prestige within the community and access to social and bureaucratic networks that came with the job. Irrespective of the wider drivers leading AWWs to take up the job, there was a strong “finance” motive for workers and their families; the salary was a key source of household livelihood.

• Programmatic:
- Service preference of the beneficiary and AWW: The ICDS services can be broadly classified as product-oriented services and information-oriented services. Services that linked to products include food distribution, immunisation and pre-school. Services linked to information are individual and group counselling. AWWs perceive that beneficiaries have a clear preference for product-oriented services over information-oriented services. This product preference positively influences the attendance of these services compared to information-oriented services. Considering the pre-school attendance is influenced by food, even when the food is not available, AWWs provide the pre-school. This leads to another type of preference from the worker towards the pre-school service because of their self-identity as a teacher, as the majority of them could not be teachers and ended up being AWWs.
- Work environment factors: various work environment factors influence AWW’s motivation, job satisfaction, and performance. A summary of work environment factors and how they influence AWWs is given in Figure 1.

Figure 2 Influence of work environment factors on AWW performance


• Community:
- Caste dynamics between the AWW and community and discrimination towards the AWW emerged as a factor influencing the AWW motivation and performance. In the AWW context, when there are multiple caste groups in a community, and when the AWW is not able to satisfy any particular caste group due to the targeting of food and pre-school services, community dynamics explicitly negatively influence the delivery of services. In a few cases, physical violence was reported between the AWW and the community.

- Seasonal migration of low-income families (to brick kilns and back) creates changes to the fixed food distribution targets. This creates frictions between the AWW and the community and negatively influence her motivation and the delivery of food-related services.
Although factors influencing the AWW performance belong to different contexts, they are strongly interrelated. For example, this study considered ‘service preference of beneficiaries and AWWs’ as a programme factor; however, it is a good example of programmatic and community contexts interacting with each other and their dynamics influencing AWW performance (access of services). To conclude, the initial findings from the study suggests that any reform to improve the ICDS needs to be rooted in understanding of the self-identity of the AWWs and the complex configurational environment which she works.

Acknowledgement: This study was conducted as part of my doctoral work at the Institute of Development Studies funded by the Transform Nutrition doctoral fellowship. The field work for this study was was funded by POSHAN (Partnerships and Opportunities to Strengthen and Harmonize Actions for Nutrition in India) project led by International Food Policy Research Institute (IFPRI).

Should the Maternal Infant and Young Children Nutrition programme in Kenya be scaled up?

This briefing Social Return On Investment Assessment Of A Baby Friendly Community Intervention In Urban Poor Settings, Nairobi, Kenya by Transform Nutrition project partners African Population Health Research Center answers that question. It outlines the impact of the Maternal Infant and Young Children Nutrition project that aimed to improve the health and nutritional status of children and inform implementation of the government’s Baby Friendly Community Initiative. The Study is now also featured in June 2016 issue of Field Exchange.

Public-Private Partnerships and the Reduction of Undernutrition in Developing Countries

This Transform Nutrition discussion paper Public-Private Partnerships and the Reduction of Undernutrition in Developing Countries brings structure to the discussion of private-sector engagement in nutrition. It clarifies different models of engagement, reviews the evidence base on public-private partnerships (PPPs) for the reduction of undernutrition, and outlines some potential ways forward.

Assessing Bangladesh’s National Nutrition Services Program

By Masum Billah and Shams El Arifeen, ICDDR,B

Despite important economic progress and reaching lower middle-income status in 2015, Bangladesh remains one of the high under nutrition burden countries. This is not due to a lack of political will. In 2009, the Annual Program Review of the Health, Nutrition, and Population Sector Program of the World Bank recommended scaling up nutrition interventions in Bangladesh through mainstreaming critical nutrition interventions in health and family planning services. To achieve this goal, the country made nutrition a priority.

The National Nutrition Services (NNS) has since been pursuing a variety of key strategies and actions. In 2011, the operational plan (OP) of NNS under the current Health, Population and Nutrition Sector Development Programme (2011-16) was approved by the Government of Bangladesh which meant mainstreamed nutrition interventions should be implemented through the existing health system from July 2011 onwards.

But how well is this program working? A new report  Bangladesh National Nutrition Services: Assessment of Implementation Status, published by the World Bank presents the findings of a research study done by IFPRI and icddr,b, partly supported by Transform Nutrition, to assess the implementation of the Government of Bangladesh’s National Nutrition Services Program (NNS). Importantly it also identifies the achievements, identifies key bottlenecks that adversely impact these achievements, and highlights potential solutions to ensure smooth delivery of the program in the future.

The study
The overall objective of this study was to assess the effectiveness of the delivery of the different components of NNS. The report was written by a team from IFPRI and ICDDR,B, including Drs. Shams El Arifeen and Purnima Menon, who lead Transform Nutrition’s work on making health systems more nutrition-sensitive.  The study focused on assessing issues related to management and support services; training and capacity development; service delivery; monitoring and evaluation, and; exposure to interventions.  A mix of quantitative and qualitative methods were used to assess the quality of  service provision related to integration of nutrition with antenatal care services and sick child care in health facilities: The research also focused on elements of the enabling environment to support the roll-out of the NNS interventions, examining institutional arrangements and the policy environment at the national level.

Key findings
The study indicated that NNS lacked specificity in choice of interventions and preventive service platforms resulting in attempts to deliver too many interventions with limited coverage and quality.  Frequent changes in leadership, coordination, capacity and workload-related challenges faced by the NNS have hampered the implementation of nutrition interventions.  The study also reported slow roll-out and poor coverage of training, especially among the frontline health workers. Although the quality of nutrition service during antenatal care, one of prioritized health service delivery contacts, was somewhat better; height and weight measurements, provision of appropriate nutrition counselling and screening for severe acute malnutrition during sick child management contacts were sub-optimal.  Critical challenges relating to service delivery at community and outreach level (preventive platforms), lack of systematic supportive supervision, poor recording and reporting of service statistics, and weak monitoring mechanisms have jeopardized the NNS’s ability to achieve adequate delivery of quality nutrition services.

Our recommendations
Our study found that the overall NNS effort is an ambitious, but valuable, approach to support nutrition actions through an existing health system with diverse platforms. The results indicate that although the maintenance of strong and stable leadership of NNS is an essential element to ensure integrated and well-coordinated service delivery, the current arrangement is unable to ensure effective implementation and coordination of NNS. So how can this be improved?

  • Focusing on some of the critical challenges related to leadership and coordination and then embedding a smaller set of priority interventions into well-matched health system delivery platforms is most likely to help achieve scale and impact.
  • Strategic investments in ensuring transparency, engaging available technical partners for monitoring and implementation support, and not shying away from other potential high coverage outreach platforms, such as some NGO platforms, also could prove fruitful.
  • The Government of Bangladesh, and the health system in particular, must lead the effort to deliver nutrition and the NNS is an important start towards this goal. Nonetheless, it is clear that development partners who have expressed a commitment to nutrition must coordinate their own activities both among themselves and with the government and provide the support that can deliver nutrition’s potential for Bangladesh.

Bangladesh is ranked 142 of 187 countries on the 2014 Human Development Index and 57 among 120 developing and transitioning countries on the 2014 Global Hunger Index.  There is still some way to go to to ensure the population of Bangladesh is well nourished and able to make the most of economic progress. icddr,b, IFPRI and Transform Nutrition are working to generate evidence on effectiveness of different  nutrition-specific and nutrition-sensitive interventions in Bangladesh to support strategic policy and intervention choices that can improve nutrition in Bangladesh.



From global mantra to local results: scaling up impact on nutrition

By Stuart Gillespie, Purnima Menon and Andrew Kennedy

 This blog is based on an article recently published in the ‘Global Nutrition Report 2015’.

Despite relative consensus on actions to improve nutrition globally, less is known on how to operationalize the right mix of actions - nutrition-specific and nutrition-sensitive - equitably, at scale, in different contexts. How to “put it all together” to maximize sustained impact?

In a recent review by the Transform Nutrition consortium of approaches to scaling up impact on nutrition, a number of elements repeatedly emerged as key factors. First, the start point for most successful large-scale programs (e.g. Progresa-Opportunidades in Mexico, Alive and Thrive in Bangladesh) was a discussion and ultimately a shared vision of what large-scale impact actually looks like — and not on any intervention per se. [Read more...]

How to scale up impact on nutrition

We know a lot more about what is driving malnutrition and we know more about the type of interventions that are needed to respond. And yet, we continue to struggle with the “how” questions. A new Transform Nutrition research brief written by Stuart Gillespie is out now Scaling up impact on nutrition: what will it take? which will help guide policymakers and programme managers when designing and planning for scaling up impact on nutrition.

Scaling Up Impact on Nutrition: Four Case Studies

Supplementary material for the paper: Gillespie, S, Menon, P., Kennedy, A (2015) Scaling up impact on nutrition: what will it take? Advances in Nutrition 6:440–51.

Scaling Up Impact Four Case Studies (pdf)

Case Study 1:        Scaling up a high quality IYCF counseling intervention in Bangladesh: Alive & Thrive(1,2)

Case Study 2:        Scaling up iron-folic acid supplementation in Nepal(3)

Case Study 3:        Scaling up social protection in Mexico: Progresa-Oportunidades

Case study 4          Scaling up homestead food production in Bangladesh