Reply to Letter to the Editor by Robertson et al.

In their letter to the Editor, Robertson and colleagues state that biodiverse food solutions hold great potential for improving complementary feeding and oppose the use of lipid‐based nutrient supplements (LNS) for improving nutritional quality of diets of young children (De Pee, 2015).

The focus of my editorial was on solutions to improve complementary feeding as studied by the papers that were contained in the December 2015 supplement of Maternal and Child Nutrition. Collectively, these papers dealt with special nutritious foods or formulations that were designed to complement diets of young children, mainly aged 6–23 months, with specific (micro)nutrients. These solutions are required when the local diet, as accessed, prepared and consumed by a large proportion of the population, is unlikely to provide young children with sufficient amounts of all the nutrients they require. As some of the papers referenced by Robertson et al. also report, some nutrient needs are very difficult to meet, such as iron, zinc and calcium, by 6–11‐month‐old infants (Kuyper et al. 2014; Termote et al. 2012).

The authors' arguments are not supported by the publications they have cited. Bogard et al. (2015) and Kuyper et al. (2014) identified nutrient‐rich local foods, i.e. specific fish species and underutilized crops, respectively, that can contribute to meeting nutrient needs of young children, and Waswa et al. (2014) show that nutrition education can improve the diversity of complementary feeding diets. However, these papers have not shown, as claimed by Robertson et al. (2016), how to overcome some practical constraints of feeding young children that I listed in the editorial, including distance and frequency of market days, ability to keep foods cool and time to prepare and feed frequent, balanced meals. Johns and Powell discuss how food sovereignty and supporting biodiversity can improve diversity of diets at household level, but they do not present evidence of an impact on chronic undernutrition (i.e. stunting). Neither Parlesak et al. (2014) nor Shrimpton and Rokx (2016) assessed, or even discuss, whether LNS can contribute to increasing prevalence of childhood obesity and double burden of malnutrition. And last but not least, the paper on linear programming by Parlesak et al. (2014) fails to discuss whether daily addition of a combination of beef heart, beef liver, moringa leaves and/or small dried fish to young children's diet, which could meet their nutrient requirements, would be feasible.

It appears that the authors oppose efforts to ensure that children have access to nutrients required to grow, develop and remain healthy, when these include the use of specific fortified solutions, however small they are (i.e. micronutrient powder and LNS‐SQ are provided in quantities of 1–20 g/day, for 10–30 days/month). It should be noted that food fortification has been implemented for almost a century and was identified by the Copenhagen Consensus as having the highest benefit‐cost ratio for improving human development.

Robertson et al. (2016) also mix up the terminology and use of different fortified spreads. RUTF is a ready to use therapeutic food for the treatment of severe acute malnutrition (SAM), which is provided in amounts of 500–1000 kcal/day, whereas LNS refers to lipid‐based nutrient supplements that are added to diets (as small or medium quantity, i.e. 120 kcal/day (SQ) or 250 kcal/day (MQ)) to ensure that the diet contains adequate amounts of essential nutrients for prevention of undernutrition (LNS‐SQ) or wasting more in particular (LNS‐MQ) (De Pee et al. 2011). While LNS for improving complementary feeding is provided in amounts of 20 g/day, the authors approximate its cost by referring to quantities used for SAM treatment (i.e. almost 14 kg, 150 × 92 g). Their claim that LNS can displace breastmilk is not substantiated, and they did not acknowledge that a carefully conducted study showed otherwise, i.e. there was no difference of breastmilk intake among groups that received 0, 10, 20 or 40 g/day of LNS (Kumwenda et al. 2014).

To achieve Zero Hunger and the Sustainable Development Goals, efforts to improve child nutrition should be held against a high standard, i.e. it is not enough to make any improvement in nutrient intake; an adequate level should be reached. While I have always been of the opinion, like Robertson et al. (2016), that local solutions should be optimized to the extent possible, we should also critically assess whether that is sufficient to achieve an adequate nutrient intake among key target groups, and if not, do more. For children aged 6–23 months, this may include adding specific, complementary, nutritious commodities, while strategies for increasing biodiversity and dietary diversification should continue, also for the benefit of other target groups in the population.

de Pee, S. (2016) Reply to Letter to the Editor by Robertson et al. . Maternal & Child Nutrition, 12: 641–642. doi: 10.1111/mcn.12319.