Eight sub-sessions
Chair: Helen Harris-Fry, London School of Hygiene and Tropical Medicine (LSHTM)
Speakers and presentations
Daniel Sellen, University of Toronto
Integrated Nutrition-Specific and Nutrition-Sensitive Maternal and Child Health Program Design in Asia and Africa: Experience from a Multi-Country Baseline Study
Recording
Elly Obwato, Kenya Medical Research Institute
Lessons from the Implementation of Shamba Maisha: A Multi-Sectoral Agricultural and Finance Intervention Randomized Trial to Improve Food Security, Nutrition, HIV, and Child Health in Western Kenya
Slides/Recording
Heather Ohly, University of Central Lancashire
A Randomised Controlled Trial to Examine the Effectiveness of Biofortified Wheat as a Strategy to Reduce Zinc Deficiency in Pakistan
Slides
Khumbo Mhango, LUANAR
Effect of Nutrition Education and Milk Processing on Milk Consumption and Nutritional Status of Under-Five Children among Dairy Farming Households in Dedza District of Malawi
Slides / Recording
Marco Santacroce, IFRPI
Some Good Food Today or Some Advice on Better Food Choices for Tomorrow?
Slides / Recording
Michael Diressie, IFPRI
On the Radio: Using Serial Mini-Dramas to Promote Orange Sweet Potato in Uganda
Slides/Recording
Stella Nordhagen, Helen Keller International
Is Anything Still Growing? Sustainability of an Urban Nutrition-Sensitive Agriculture Project
Slides/Recording
Aulo Gelli
Community-Based Early Childhood Development Center Platform Promoting Diversified Diets and Food Production Improves the Adequacy of Nutrient Intake of Pre-School Children in Malawi: A Cluster Randomized Trial
Recording
Q&A
Recording 1 / Recording 2
Integrated Nutrition-Specific and Nutrition-Sensitive Maternal and Child Health Program Design in Asia and Africa: Experience from a Multi-Country Baseline Study
Daniel Sellen, University of Toronto
Introduction
Malnutrition continues to be a major public health problem, particularly in South Asia and sub-Saharan Africa. The most recent Joint Child Malnutrition Estimates indicated insufficient progress in achieving Global Nutrition Targets 2025 and Sustainable Development Goals 2030. To accelerate the progress, integrated nutrition-specific and nutrition-sensitive interventions at scale are required to address both direct and underlying causes of malnutrition. A baseline assessment of the ENRICH (Enhancing Nutrition Services to Improve Maternal and Child Health in Africa and Asia) project examined direct and underlying causes of malnutrition in four countries that helped modify the nutrition-specific and nutrition-sensitive program design.
Methods
Our analysis is based on data from four large cross-sectional household surveys (HHS) conducted in targeted rural areas of Bangladesh, Pakistan, Kenya, and Tanzania as part of impact evaluation of the ENRICH project being implemented by World Vision Canada. Although a quasi-experimental mixed method design was used for overall impact evaluation, and non-program comparison areas were chosen from comparable communities outside the program areas, this analyses only included data from program areas. The sampling strategy used a proportional stratified three-stage cluster selection approach with villages as the primary sampling unit. The target age groups for HHS are children 0-59.9 months distributed across three age strata: 0-5.9 months, 6-23.9 months, and 24 to 59.9 months, selected randomly from specific clusters. The sample size included in the analyses was 884 in Bangladesh, 936 in Pakistan, 858 in Kenya, and 936 in Tanzania. Mothers were interviewed at their homes using pretested and locally-adapted structured questionnaires consistent across all four countries. Digital data collection through the Open-Data Kit platform using android tablets was done. Data were imported into SAS v9.4 for analysis (SAS Institute Inc., Cary, NC, USA), and PROC SURVEYFREQ with the CLUSTER option used to account for clustering.
Findings and Interpretations
Maternal dietary diversity was lower in Pakistan (14%) and Tanzania (17%) in comparison to Bangladesh (24%) and Kenya (28%). Reported iron-folic acid consumption during previous pregnancy was very low in Pakistan (1%), Kenya (4%), and Tanzania (11%) relative to Bangladesh (34%). Exclusive breastfeeding rates were 71, 63, 43, and 73 percent in Bangladesh, Pakistan, Kenya, and Tanzania, respectively. Minimum dietary diversity among children 6-23.9 months was lowest in Pakistan (19%) and highest in Bangladesh (39%). Consumption of micronutrient powder or biofortified crops among children 6-23.9 months were almost non-existent.
Severe household food insecurity was more prevalent in Tanzania (32%) and Kenya (23%) in comparison to Bangladesh (11%) and Pakistan (16%). Improved toilet facilities were very low in Tanzania (16.4%) and Kenya (12%) relative to Bangladesh (37%) and Pakistan (47%). Access to better handwashing practices was less than 50 percent in all four countries. Mothers’ knowledge of child feeding, care, and nutrition during pregnancy was inadequate in all countries. Household heads were mostly male (ranged from 85-95%); economic empowerment of mothers was very low in Bangladesh (8%) in comparison to Pakistan (25%), Tanzania (34%), and Kenya (51%); these reflect male dominance and gender inequality in household decision-making and purchasing power.
Conclusions
Both direct and underlying factors are prevalent in four ENRICH countries, likely contributing to the poor nutritional status. Interventions only addressing direct causes (nutrition-specific interventions) are not sufficient to tackle malnutrition; projects should also focus on nutrition-sensitive interventions to address underlying causes through a multi-sectoral approach.
In addition to nutrition-specific interventions such as promotion of breastfeeding and complementary feeding, dietary diversity among women and children, community-based growth monitoring and promotion to ensure timely identification and treatment of nutritional deficiencies, the ENRICH project is also designed to deliver nutrition-sensitive interventions using the delivery platform of HarvestPlus in production of agroecologically appropriate nutrient-dense local vegetables and fruits, and promotion of biofortified crops, and small animals. Partnership with Nutrition International is supporting distribution of micronutrient powder among children 6-23 months. The project considered capacity investments in health system strengthening to improve access to, and utilization of, quality health and nutrition information and services. In Tanzania, health services planning and budgeting needs and challenges of the Regional and District Health Management Teams are being addressed in partnership with the Canadian Society for International Health. Local-level advocacy through Citizen Voice and Action is also included in program design to enhance social protection and accountability.
Sources of funding: Funding for this project was provided by the Global Affairs Canada (GAC) through Partnerships for Strengthening Maternal, Newborn and Child Health (PS-MNCH) Initiative
Lessons from the Implementation of Shamba Maisha: A Multi-Sectoral Agricultural and Finance Intervention Randomized Trial to Improve Food Security, Nutrition, HIV, and Child Health in Western Kenya
Elly Obwato, Kenya Medical Research Institute
Introduction
Multisectoral health interventions have significant challenges with implementation and scale-up due to several factors. Each sector has unique interests, incentives, and regulations, which may obstruct collaboration and governance of a project. We conducted an informal evaluation among the partners conducting a multisectoral agricultural intervention trial to better understand challenges, and to propose solutions to improve the conduct of future research at the intersection of agriculture, nutrition, and health.
Methods
Shamba Maisha (NCT03170986 & NCT02815579), a cluster-randomized controlled trial (RCT), uses a multisectoral agricultural and microfinance intervention to improve food security, household wealth, empowerment, HIV clinical outcomes, and childhood growth and development among HIV-infected farmers and their children. The trial is designed to test the hypothesis that the intervention leads to improved HIV and childhood health outcomes through addressing the dual epidemics of food insecurity and poverty. The transdisciplinary research team includes the following disciplines: HIV medicine, pediatrics, epidemiology, nutrition, statistics, social science, agriculture, economics, health economics, banking and finance, and non-profit social enterprise development. The RCT is underway at 16 health facilities randomized 1:1 to intervention and control groups in western Kenya. HIV-infected patients older than 18 years, on antiretroviral therapy, with moderate/severe food insecurity and/or body mass index less than 18.5, and access to land and surface water are eligible for enrolment. The intervention includes: 1) a farming commodity loan (~$175) including a micro-irrigation pump, seeds, and fertilizer; and 2) structured training in sustainable agricultural practices and financial literacy. We have conducted continuous internal consultation, team reflection, and dialogue during the design and implementation phases of Shamba Maisha.
Findings and Interpretations
- Transdisciplinary agriculture, nutrition, and health (ANH) research requires a collective-impact framework that specifies common goals, shared measurements, and clear roles and responsibilities for all partners. Frequent communication, coherent implementation of memoranda of understanding, and establishment of a joint-intervention implementation team has improved overall uptake and performance of the intervention.
- Implementation of a finance intervention requires a nuanced understanding of credit, financial practices, and laws in the local context, with full participation of a lending institution open to implementing innovative practices. Working with the banking partner, we reduced the amount of the down payment, created additional incentives within groups to encourage loan repayment, and coordinated financial literacy training with the agricultural training to help farmers improve production and income generation.
- Multisectoral ANH research requires thinking and acting outside the normal global-health research paradigm with frequent innovation and problem solving throughout study implementation. To keep on track with the training targets and overall study timelines, we shifted management and supervision of the agricultural trainers to the local study manager to improve oversight and coordination of study activities, leading to greater uptake of the intervention.
Conclusions
Cross-sectoral research collaborations require mobilizing partners with differing goals, orientation, strategies, motivations, and definitions of success. Complex social challenges, such as those presented by HIV and food insecurity, require the development of a common agenda, matching set of priorities, shared measurements for success, development of a shared vocabulary, and constant communication in order to achieve the desired results.
A Randomised Controlled Trial to Examine the Effectiveness of Biofortified Wheat as a Strategy to Reduce Zinc Deficiency in Pakistan
Heather Ohly, University of Central Lancashire
Introduction
Zinc deficiency is a global problem, particularly in low and middle-income countries where access to animal-source foods is limited. In Pakistan, zinc deficiency affects over 40 percent of the adult female population. The consequences of zinc deficiency include stunted growth and development, increased susceptibility to infections, and complications during pregnancy and childbirth. Biofortification of staple crops has potential as a sustainable means of increasing population dietary zinc intake. The aim of this study is to examine the effect of consuming flour made from biofortified wheat on the zinc status of zinc-deficient women living in a rural community in Pakistan.
Methods
We are conducting a double blind, randomised controlled trial with crossover design, in a low resource setting in North West Pakistan. Fifty households were recruited into this study, which runs from October 2017 to February 2018. Each household includes a female aged 16-49 years who is neither pregnant nor breastfeeding, and not currently consuming nutritional supplements. These women are the primary study participants. Households were provided with flour milled from biofortified wheat grain (Zincol-2016/NR-421) or control wheat grain (Galaxy-2013). All households received control flour for an initial two-week baseline period. In the first eight-week intervention period, 25 households received biofortified flour (Group A) and 25 households received control flour (Group B). In the second eight-week intervention period, Group A and B crossed over and received control flour and biofortified flour, respectively. Tissue samples (blood, hair, and nails) are being collected from the women at five points: baseline, mid and end of period 1, mid and end of period 2. This study will compare established biomarkers of zinc status (plasma zinc concentration) with novel indicators, including markers of DNA damage and a new laser technique for measuring nail and hair zinc concentration. The trial protocol was prospectively registered: ISRCTN83678069.
Findings and Interpretations
The study findings will show whether or not consuming flour made from biofortified wheat has a beneficial effect on the zinc status of zinc-deficient women living in a rural community in North West Pakistan. With four weeks remaining to the end of the trial, the retention rate is 90 percent (45/50 households).
The biofortified wheat was developed by HarvestPlus using conventional plant breeding techniques. Its zinc concentration for this study was further increased using agronomic techniques – the application of zinc-enriched fertilisers. The zinc concentration of the grain was 62 mg/kg, compared to 23 mg/kg on average in traditional varieties of wheat.
The main outcome measures will be:
- Plasma zinc concentration, measured by inductively coupled plasma mass spectrometry,
- DNA fragmentation, measured using the comet assay, and
- Hair and nail zinc concentration, measured by laser induced breakdown spectroscopy.
One of the key challenges in measuring zinc status is the lack of a sensitive biomarker that is suitable for use in remote settings. Therefore, this study is innovative in the field of zinc biology as well as its implications for agriculture, nutrition, and health.
Conclusions
This study is the first randomised controlled trial to examine the effect of consuming flour made from biofortified wheat (developed using conventional plant breeding coupled with agronomic techniques) on the zinc status of women of child-bearing age. It will offer new insights into the potential for strategies involving the biofortification of staple crops such as wheat to reduce zinc deficiency in Pakistan and other countries. It will also evaluate the use of novel biomarkers of zinc status in low resource settings. Non-invasive measures of hair and nail zinc concentration may be particularly useful – if found to be sensitive and reliable biomarkers – for researchers operating in similar settings where cultural sensitivity is paramount. Therefore, the findings of this study will be of interest to nutritionists, public health specialists, plant and soil scientists, and policymakers.
Funding was awarded by the Biotechnology and Biological Sciences Research Council (BBSRC) Global Challenges Research Fund.
Effect of Nutrition Education and Milk Processing on Milk Consumption and Nutritional Status of Under-Five Children among Dairy Farming Households in Dedza District of Malawi
Khumbo Mhango, LUANAR
Introduction
Despite the role dairy production can play in mitigating impacts of under nutrition, milk consumption among dairy farming households in Malawi is poor and so is stunting status of children under-five. A study was therefore conducted to explore the effect of incorporating nutrition education and milk processing pathways within dairy production intervention and assess their effect on household milk products processing and milk and milk product consumption by children under-five and their nutritional status.
Methods
A quasi-experimental study involving 61 dairy farming households and 79 children aged 6 to 52 months was conducted in two milk bulking groups (intervention and control) in Dedza district, from April to December 2016. Nutrition education and milk products processing training were provided concurrently, four times for three days at two months intervals. Chi-square tests were done for proportions and t-test for means in SPSS. Standardized nutrition messages adapted from the Scaling Up Nutrition (SUN) 1000 days key messages for Malawi were given to child caregivers in both milk bulking groups. P-value ˂ 0.05 was considered significant.
Findings and Interpretations
Consumption of milk (≥250 ml) per day by children under-five increased from 3.3 percent to 48.3 percent at the intervention and reduced from 6.7 percent to 3.4 percent at the control group. Household milk products processing increased by 93.5 percent resulting in a subsequent increase (86.2%) in consumption of homemade dairy products by children under-five at the intervention group. Differences in household milk product processing and consumption by children under-five were statistically significant between intervention and control. Stunting dropped from 34.2 percent to 20.5 percent and increased from 31.6 percent to 35.1 percent at end-line for intervention and control respectively but not significant. Mean height difference increased from 0.6 to 1.5 cm in favor of intervention group. Proportion of children from both milk bulking groups combined that met the minimum dietary diversity increased by 10.6 percent at end-line from 80 percent at baseline.
Conclusions
Study findings reveal that nutrition education and milk product processing training has positive effects on daily milk quantity consumption, household milk product processing, and consequently, consumption by children under-five, hence improved nutrition status. Similarly, nutrition education on age-appropriate complementary feeding is vital in improving dietary intake for under-five children. Therefore, dairy production nutrition interventions should incorporate nutrition education and milk products processing to effectively contribute to improved nutritional status of children under-five.
Some Good Food Today or Some Advice on Better Food Choices for Tomorrow?
Marco Santacroce, IFRPI
Introduction
There is evidence suggesting that cash, voucher, and food transfers increase food consumption, improving the quantity and quality of food consumed by poor households. However, differences emerge in the types of food consumed, with food transfers leading to larger caloric intake and vouchers to greater dietary diversity. There is also evidence of “production to consumption” impact pathways for nutritious foods in ag-nutrition interventions, with positive associations between production diversity, dietary diversity, and essential micro-nutrient intake. However, nutrition education often fails to improve micronutrient intake above recommended levels. We assess and compare a social transfer to an ag-nutrition intervention in Malawi.
Methods
This study is based on three rounds of surveys from the Nutrition Embedded Evaluation Program Impact Evaluation (NEEPIE) study. NEEPIE is a cluster-randomized controlled trial (cRCT) of a communication-based behavioral change nutrition and agriculture intervention targeting preschool children aged 3-6 years. Data was collected at baseline in September 2015 (postharvest season), at midline in February 2016 (peak of the lean season), and at endline in September 2016 (12 months after baseline). Sixty food-insecure communities and 1,200 households were randomly selected in two stages of sampling within the Zomba district in southern Malawi. Households were assigned to one of two treatment arms. During the same period, following extreme flooding in the 2015 production season, the government of Malawi and development partners scaled up food transfers to affected communities. Fifteen percent of households sampled (175 households) in NEEPIE had received lean season food transfers. We employ quasi-experimental and experimental methods to compare short (6 month) and medium (12 month) term effects of lean season food transfers and a behavioral change intervention on food security outcomes. We evaluate food transfers by combining propensity-score-matching difference-in-difference (DID) and multilevel (mixed-effects) DID regression to estimate cRCT treatment effects.
Findings and Interpretations
We detect a positive NEEPIE impact at the household level on the consumption of various food groups (legumes, nuts, and seeds; meats; eggs; oils and fats; sweets) driving up household dietary diversity score (HDDS) at 12 months, but not at 6 months. More crops and food groups were produced at 6 and 12 months (cereals, roots and tubers, legumes, vegetables) and poverty, measured as $1.25 PPP pc/day, was eight percent lower among treated. At the child level, we observe lower rates of stunting (17%), higher HAZ scores (0.49), and greater food variety and dietary diversity scores (roots and tubers, vegetables, fruits, fish, milk).
Lean season food transfers have a strong impact at 6 months, but no impact at 12 months. Impact is detected at the household level on food expenditure (20% higher), lower poverty (12%), more crops and food groups being planted (cereals), and greater HDDS. At the child level we detect lower rates of wasting (8%) for children aged 6-24 months at baseline, and greater food variety and diversity scores (fish, milk, oils, sweets). We do not find compelling evidence suggesting that impact is concentrated in the poorest quintile of households. However, we find that targeting of transfers can be improved.
Conclusions
During the lean season, significant declines in food security, diet quality, and nutrition status were present. We find the communication-based behavioral change nutrition and agriculture intervention targeting preschool children aged 3-6 years had protective effects during a period of high food insecurity (2016 lean season). This suggests a potential role for similar interventions within social protection portfolios. Community-owned preschools can provide effective platforms to scale up nutrition interventions and change household behaviors related to food production and consumption. The positive impact is detected both in the short (6 month) and medium-term (12 month) for most outcomes, but impact magnitude (DID) tends to be stronger at 12 months.
The analysis undertaken on social transfers suggests that lean season food distribution had important short-term benefits (protective effects on food security) but failed to improve dietary and nutritional outcomes in the medium-term once distribution of social transfers ceased. However, there is strong evidence that these transfers can improve both the quantity and quality of food consumed when distributed. Targeting of transfers can be largely improved as 20 percent of households in the richest quintile received transfers.
On the Radio: Using Serial Mini-Dramas to Promote Orange Sweet Potato in Uganda
Michael Diressie, IFPRI
Introduction
Since 2012, HarvestPlus and its partners have been disseminating biofortified vitamin A-rich orange sweet potato (OSP) in Uganda. As part of its scaling strategy, HarvestPlus worked closely with Farm Radio International (FRI) to raise awareness about OSP. “My Children”, an interactive radio mini-drama, was aired for two seasons, and included all the elements of a drama (love, domestic strife, family, money, and power) as well as information about the nutritional benefits of and best agronomic practices for the crop.
Methods
The purpose of this paper is to present key findings from a study that assessed the effectiveness of “My Children” in promoting OSP nutritional knowledge and influencing the production and consumption of OSP among program listeners. Cross-sectional survey data were collected in five districts in Uganda. The survey was implemented in two distinct communities in each district, one in which HarvestPlus was directly implementing the intervention and the FRI-sponsored radio program was broadcast (410 households, called H+ villages), and a second which was outside the reach of HarvestPlus’ direct intervention, but where the FRI-sponsored radio program was broadcast (367 households, called non-H+ villages).
The relevance of using interactive radio drama as an integral component of a scaling strategy of vitamin A biofortified Orange Sweet Potato (OSP) was assessed, by applying Propensity Score Matching to see the impact on knowledge, planting and consumption of OSP, and Dose-Response Function to demonstrate the range of episodes that contributed to OSP knowledge gain."
Findings and Interpretations
The radio drama had a significant positive impact on the knowledge of the overall sample. Listening to MC increased the OSP knowledge of the overall households by 7 percent. Looking at the village types separately, the OSP knowledge of the non-H+ villages was improved by 8 more percentage points than the H+ villages just because of MC, because, HarvestPlus has been already promoting OSP through its program, making the impact of the radio drama in the H+ villages less than those of the non-H+ villages.
The impact of MC on practices, namely growing and consuming OSP, is positive. Looking at the impacts disaggregated by the types of villages, the impacts on consumption are only significant for the H+ villages, because, consumption is something that requires prior familiarity and abundant availability of OSP. In the case of planting OSP, we see that MC has a positive impact on both types of villages.
In terms of the effect of the length of the radio program, it was good it was a series with more than 40 episodes. However, the results show the maximum number of episodes that would bring the desired change on knowledge was 44.
Conclusions
The results indicate that the radio drama “My Children” was effective at achieving the intended study objectives. It improved the overall knowledge of OSP, as well as the intention to grow and consume OSP, in rural households. Additionally, the program bridged gaps by creating awareness of and knowledge about OSP to areas where HarvestPlus had not been operating in until the time of broadcast. This indicates that the radio show complemented the efforts to spread OSP awareness and knowledge across the country. However, in areas where HarvestPlus was not operating, the radio drama had a weaker effect on the intention to grow and consume OSP. This might be because planting materials were unavailable in those communities, or because the radio program was not as intensive and visual as face-to-face demonstrations in H+ communities. Overall, it can be concluded that the radio program was effective and useful in complementing the efforts of delivery of biofortified crops as radio is one of the most widely used media. It helps to reach more people that would have not been reached otherwise.
Is Anything Still Growing? Sustainability of an Urban Nutrition-Sensitive Agriculture Project
Stella Nordhagen, Helen Keller International
Introduction
Despite considerable interest in nutrition-sensitive agriculture (NSA) as well as genuine recognition of the need for development projects to aim for sustainability, little is known about the sustainability of nutrition-sensitive agriculture approaches. Indeed, sustainability was identified as a key gap in a recent review. Similarly, despite major urban growth and high burdens of malnutrition among urban populations in developing countries, there is scant information about NSA programs that can reach urban consumers. This study helps to fill these gaps by assessing sustainability of an NSA program implemented in urban Senegal (2013-2016) and the factors facilitating sustainability.
Methods
The intervention consisted of support for horticulture and poultry production using compact ‘micro-gardens’ and henhouses as well as social behavior change communication on nutrition and hygiene topics and women’s empowerment activities. Impacts were expected to be achieved primarily through own consumption of home production, given the small scale. We define our main metric of sustainability as continued production of vegetables or chickens/eggs using the project-provided tables and henhouses, applying the project-promoted practices, and consumption of the associated produce. Secondarily, we consider nutrition/hygiene knowledge levels and practices compared to during the project.
Data come from a cross-sectional survey of a random sample of 262 former project participants (20 percent of the total) 18 months after the project’s end. To minimize acceptability bias (i.e., falsely reporting continued use of project-provided items to curry favor in the eyes of the implementers), the survey included observations at each household to assess the current existence and functionality of the microgardens and henhouses, verified by photographs. Quantitative data analysis was performed using STATA SE15. The survey results were used to inform questions and identify respondents for 20 semi-structured interviews with women who were particularly successful at continuing activities. The resulting transcripts were analyzed using nVivo.
Findings and Interpretations
Results show widely different levels of sustainability for the two project components. The vast majority of former participants (80 percent) no longer grew vegetables, and 72 percent had not even retained the microgardens provided by the project. There were low levels of production even among those producing, and almost none had expanded production or delved into marketing: production was used for home consumption, making modest but regular contributions, and for social functions, such as gifting.
In contrast, 75 percent of former project participants continued to raise chickens. Moreover, 20 percent had expanded production, with 29 percent using poultry rearing as a source of income—in some cases considerable amounts of income. Among those actively raising chickens, some had diversified into other breeds or types of birds: whereas the project provided laying hens, the majority (69 percent) of the active chicken-raisers were raising broilers, for meat production. Qualitative data suggested that gardening was seen as a ‘passion project’, whereas poultry could be a true livelihood—thus motivating greater continuation. Both gardening and chicken-rearing were used to improve household diets, but the more significant contribution came from poultry-rearing, given the larger scale and higher associated income. Most nutrition knowledge and practices remained at or near levels seen during the project.
Conclusions
The project was successful at increasing poultry rearing in a sustainable manner, whereas the urban approach to vegetable gardening proved unsustainable. Earning income through small-scale commercial chicken-rearing increased sharply after the end of the project and proved to be empowering to the women taking part, in addition to adding to household incomes. Some nutrition impacts, though slight, appear to have been sustained both through own consumption of poultry (and to a lesser extent, garden) products and increased income from chicken sales. However, the widespread shift in chicken breeds to broilers was associated with a decrease in egg production—and hence in egg consumption among young children, a key goal of the project. Moreover, the results show clearly how urban residents are embedded in a different food system context than rural residents, suffering from increased pressure on production space but benefiting from an easy outlet for food sale (and purchase) nearby. Over time, this market context shaped their choices related to agricultural production and sale. Given easy access to markets, the main long-term operational ‘pathway’ to improved nutrition in this population thus appeared to be increased income. Future urban NSA projects might consider heavier reliance on this pathway in project design.
Community-Based Early Childhood Development Center Platform Promoting Diversified Diets and Food Production Improves the Adequacy of Nutrient Intake of Pre-School Children in Malawi: A Cluster Randomized Trial
Aulo Gelli
Introduction
Young children in Malawi face widespread problems of low-quality diets and chronic malnutrition. This study aimed to evaluate the impact of providing a nutrition behavior change communication (BCC) intervention through community-based pre-schools on the nutrient adequacy of diets of young children in Malawi.
Methods
A longitudinal cluster-randomized design was implemented in 60 community-based childcare centers, including 1,248 children aged 3 to 6 years. Nutrient intakes were estimated using an interactive 24-hour recall. Impacts were assessed by difference-in-difference (DID) estimates, adjusted for geographic clustering, child age and gender.
Findings and Interpretations
Intervention groups were similar for most baseline characteristics. The attrition rate was low (7 percent over the 12-month period). Participation in the intervention was high (>90 percent enrolment and ~80 percent attendance rate during the five days prior to the survey). Improvements were significantly greater in the intervention group, compared to control group, for consumption of different food groups (cereals, vegetables, fruits, fish and seafoods, and sweets) and dietary diversity (DID: 0.36 percentage points [pp]). Significant impacts were observed for the probability of adequacy of several individual micronutrient intakes: vitamin A (DID: 10pp), vitamin C (16pp), riboflavin (10 pp), zinc (9pp), vitamin B12 (5pp), and for the mean probability of adequacy for the 11 nutrients considered (5pp). These impacts were mostly driven by effects on younger children (aged 3 to 4 years) and children living in poor households.
Conclusions
Using a pre-school platform to roll-out nutrition BCC is an effective strategy to improve the adequacy of micronutrient intake of pre-school children in rural, food-insecure settings.
