Most malnutrition, characterised by growth faltering, occurs during the first 1000 days of life, from conception to a child’s second birthday¹; and its prevention is critical to avoid premature mortality and irreversible damage to development and human potential^2,3,4. Interventions to promote breastfeeding and appropriate complimentary feeding (CF) have been recognised as two of the most effective strategies for improving growth and development^5,6. The World Health Organization (WHO) recommends that approaches to reducing malnutrition feature locally available and produced foods and traditional foods wherever possible, and rely on introduced products, fortified food or supplements only if they address critical nutrient gaps⁷. Livelihoods are among the many pathways linking agriculture and nutrition and of particular significance is the high proportion of women who work in agriculture in many places where malnutrition persists most acutely. Therefore, emphasis is increasingly being placed on improving the nutritional status of women and children in agriculture projects, especially women who are pregnant or breastfeeding and children aged less than two years⁸.

 

Efforts to make agriculture more nutrition sensitive often ask how the nutritional impact of a planned or existing agricultural intervention can be enhanced. The ability to impact nutrition outcomes can often be taken for granted as it is assumed that increased agricultural growth and improved rural incomes lead to consumption of nutritious foods and improved nutritional status of vulnerable members of the population. To address malnutrition effectively in the short and medium term, agricultural programs should apply frameworks that consider the multiple pathways through which agriculture can improve nutrition outcomes; including production, post-harvest, income and consumption. This process requires context-specific evidence on the nutritional needs of populations and opportunities for intervention. In response to this, researchers have developed a tool called Optifood. Applying linear optimisation, Optifood identifies the nutrient gaps in diets of vulnerable populations and models the potential for meeting nutrient requirements using food-based and other approaches. As such, the tool can identify pathways for impacting nutrition outcomes through types of agricultural production, food processing and income utilisation as well as through tailored strategies to complement production.

 

Optifood is a software tool developed by the World Health Organisation (WHO) in collaboration with the London School of Hygiene and Tropical Medicine (LSHTM) and the Food and Nutrition Technical Assistance III Project (FANTA). The tool analyses data on the cost and nutritional content of locally-available foods as well as the dietary patterns and nutrient requirements of individual target groups to model the nutritionally best, lowest cost diets based on local consumption and food supply. Optifood identifies the potential quality and content of local diets, nutrients for which adequacy may be difficult to meet and the best food sources of nutrients. This analysis can be used to predict whether improved production and consumption of particular agricultural products, introduction of new foods to the local food system or increased family income could lead to improvements in local diets and enhance the nutritional impact of an agricultural program⁹.  

 

Optifood is unique in that it analyses diets at the individual target-group level, allowing the user to focus on priority groups within a population, such as women and children during the first 1000 days. In the same way the tool could be used, for instance, to focus on children targeted by a school garden project, adult male migrant farm workers, elderly household members in agricultural areas or infant recipients of a supplementary feeding program.

 

The set up for an Optifood analysis requires dietary intake data from the target group of interest. Market surveys have also been used to collect information on food costs. Finally, locally-relevant values for the nutrient composition of foods consumed in the dietary recall and nutrient requirements of the population are required.

 

Given the relative paucity of dietary intake databases globally^10,11, as well as the time and cost involved in collecting new data, the possibility of deriving Optifood data inputs from the routinely collected, representative datasets that do exist in many developing countries, such as Household Consumption and Expenditure Surveys (HCES), is being explored .

 

Optifood has been used in nutrition program design in a number of countries, to develop food-based recommendations (FBRs) that would fill or come as close as possible to filling nutrient gaps for women and children. These FBRs are context-specific and focus on feasible and acceptable consumption of local foods, with targeted fortified products and supplements added only if nutrient requirements cannot be met using a food-based approach alone. Increasingly, Optifood is also being applied to assist decision-making in agricultural programs.

 

Recently Optifood was used by researchers at LSHTM to inform the design of nutrition sensitive interventions (NSI) that were being embedded within an Ethiopian smallholder poultry production program. Building on previous successful use of household consumption and expenditures survey (HCES) data with Optifood in Guatemala and Tanzania, household-level HCES data from the 2012-2013 Ethiopian Socioeconomic Survey was converted to allow analysis of the diets of the program’s priority populations; pregnant women, lactating women and children under two years.

 

The Ethiopian analysis identified a number of nutrient gaps in diets based on local foods and habitual food consumption by these target groups. In particular, requirements for niacin, folate, calcium and vitamin A were difficult to meet using local foods for most target groups and Vitamin B12, zinc and iron requirements could not be met using any acceptable combination of local foods for women and children in most areas. The analysis revealed the best local food sources of multiple nutrients for each area in current diets. These included breast milk (for children), legumes, teff and other grains, milk and cheese, green leafy vegetables and other vegetables and, where available, red meat. Chicken meat and eggs were selected by the software as key nutrient sources only in diets where red meat was not available.

 

Preliminary field results from the regions of interest had suggested that chickens are generally used for income generation and consumption of chicken meat is low, mainly reserved for special festivities. Egg consumption on the other hand was shown to be more common, especially among those families who keep chickens. Given the focus of the program on increasing chicken production and consumption, the Optifood linear programming models were adjusted to test the potential impact of greater chicken meat and egg consumption on the nutrient intake of small children and pregnant and lactating women.

 

The analysis revealed that even if chicken or eggs were included in local diets, adequacy of all modelled nutrients could not be achieved using local foods without significant alterations to current dietary practices. Informed by the best nutrient sources, different combinations of local foods, including chicken meat and eggs, were modelled in Optifood to show the extent to which nutrient intake could be improved and the foods necessary for bringing about these improvements. The results showed that including vegetables - especially green leafy vegetables - and milk  in diets promoting chicken meat or eggs and staple grains would mean that more nutrient requirements could be met than through promoting chicken meat or egg consumption alone.

 

These analyses provided two key findings to inform program design;

1. Increased production of chickens would not necessarily have a nutritional impact on vulnerable women and children in the study area

2. The promotion of other local foods, including vegetables, beans, milk and staple grains could complement the consumption of chicken and eggs by these target groups and lead to improved nutrient adequacy

3. Some nutrient requirements for these populations will not be able to be filled with food-based approaches

 

In the HCES data, consumption of vegetables, especially green leafy vegetables was incredibly low, mirrored by poor access to these foods observed during field visits. Given that these foods were shown to be essential for meeting the requirements for some modelled nutrients, local production and access was prioritised during NSI design for this program. It was appreciated that simple, small-scale home gardens could be incorporated into agricultural extension activities as part of efforts to make this program more nutrition sensitive. Further Optifood use was planned to identify which vegetable and bean varieties in particular would have the greatest potential for nutrient impact on the diet. There were even opportunities acknowledged for preparing chicken manure to act as a home garden fertiliser, meaning that yards would be cleaned of chicken faeces, which could also help to reduce infection risk, which has negative implications for child nutrition.

 

Lastly, using the results from Optifood, a behaviour change communication strategy was envisaged that, in addition to promoting chicken and egg consumption, would communicate food-based recommendations and recipes incorporating the other identified nutrient-rich foods, strengthening the overall nutritional impact of the program.

The next step for the program in Ethiopia should be to test the feasibility and acceptability of putting the resulting food-based recommendations into practice in terms of taste, preference, cost, access and local production through qualitative short field trials. The results from this process, coupled with those from the Optifood analysis can be used to finalise the design of the NSIs, ensuring that they are realistic and context specific, increasing their ability to impact nutrition outcomes.

 

Optifood was designed to provide evidence for decision making in nutrition and agriculture programs. Through identifying nutrient gaps and possible solutions for these gaps, the program offers a novel approach to assist the design of targeted NSIs that can focus on production, processing, income generation and utilisation and consumption. Concurrently, the Optifood analysis findings can be fed into advocacy efforts, communicating the ability to meet nutrient requirements using local foods and the cost of providing nutritionally adequate diets.

 

The Optifood software is currently being finalised by LSHTM and the WHO and a new version will be released shortly. Future plans for the tool include a redesign of the software to allow even more rapid processing and analysis of both individual and household-level data to provide decision makers with the evidence they need for decision making within agricultural and other programs.

Optifood training workshops are currently taking place in India, Bangladesh and Guatemala to strengthen the capacity of nutrition and agricultural researchers and program planners from South Asia and Central America to generate evidence for decision-making in this area. To enquire about the possibility of hosting or attending a training workshop, please contact Frances Knight. Those interested in learning more about the tool are welcome to attend a short learning lab on Optifood that will be provided as part of the Agriculture, Nutrition and Health Academy Week in Addis Ababa in June.

By Frances Knight, Research Fellow in Nutrition at the London School of Hygiene & Tropical Medicine (LSHTM)

 

References

¹ Dewey, K. G., & Adu-afarwuah, S. (2008). Systematic review of the efficacy and effectiveness of complementary feeding interventions in developing countries. Maternal and Child Nutrition, 24–85.

²Adair, L. S., Fall, C. H. D., Osmond, C., Stein, A. D., Martorell, R., Ramirez-Zea, M., … Victora, C. G. (2013). Associations of linear growth and relative weight gain during early life with adult health and human capital in countries of low and middle income: findings from five birth cohort studies. Lancet, 382(9891), 525–34. doi:10.1016/S0140-6736(13)60103-8

³Victora, C., Adair, L., Fall, C., Hallal, P. C., Martorell, R., Richter, L., & Sachdev, H. S. (2008). Maternal and child undernutrition: consequences for adult health and human capital. Lancet, 371(9609), 340–57. doi:10.1016/S0140-6736(07)61692-4

⁴Victora, C., de Onis, M., Hallal, P., Blössner, M., & Shrimpton, R. (2010). Worldwide timing of growth faltering: revisiting implications for interventions. Pediatrics, 125, e473–e480.

⁵ Bhutta, Z. A., Ahmed, T., Black, R. E., Cousens, S., Dewey, K., Giugliani, E., … Shekar, M. (2008). What works? Interventions for maternal and child undernutrition and survival. Lancet, 371(9610), 417–40. doi:10.1016/S0140-6736(07)61693-6

⁶Jones, G., Steketee, R. W., Black, R. E., Bhutta, Z. A., & Morris, S. S. (2003). How many child deaths can we prevent this year? The Lancet, 362, 65–71.

⁷World Health Organisation. (2008). Indicators for assessing infant and young child feeding practices, Part 1: Definitions. Geneva.

⁸Turner, R., Hawkes, C., Waage, J., Ferguson, E., Haseen, F., Homans, H., … Shankar, B. (2013). Agriculture for improved nutrition: The current research landscape. Food and Nutrition Bulletin, 34(4), 369–377.

⁹Daelmans, B., Ferguson, E., Lutter, C. K., Singh, N., Pachón, H., Creed-Kanashiro, H., … Briend, A. (2013). Designing appropriate complementary feeding recommendations: tools for programmatic action. Maternal & Child Nutrition, 9 Suppl 2, 116–30. doi:10.1111/mcn.12083

¹⁰Bermudez, O. I., Lividini, K., Smitz, M., & Fiedler, J. L. (2014). Estimating micronutrient intakes from Household Consumption and Expenditures Surveys (HCES): An example from Bangladesh, 33(3), 208–213.

¹¹Neufield, L., & Tolentino, L. (2013). Nutritional Surveillance: Developing Countries. In B. Caballero, L. H. Allen, & A. Prentice (Eds.), Encyclopedia of Human Nutrition (3rd ed.). Oxford: Elsevier Ltd.