Despite clear evidence of the disastrous consequences of childhood nutritional deprivation in the short and long terms, nutritional health remains a low priority. Therefore, enhanced and rigorous actions are needed to deliver and scale up education and provision of complementary feeding interventions. In this review we have included trials that evaluated the disaggregated evidence of the impact of education on CF alone, and provision of CF with or without education (excluding those on food fortification and supplementary feeding) on growth and morbidity in children less than 2 years of age in LMICs.
Our review indicates that in food secure population, education on CF had a significant impact on linear growth as evident by significant increase in height gain and HAZ scores, and also significantly improved weight gain, however rates of stunting reduced non-significantly. Education in food insecure population also improved linear growth and weight gain as evident by significant increase in HAZ and WAZ scores and significant decrease in rates of stunting. We did not find any study on provision of complementary feeding (with or without education) from food secure population, however, from food insecure population the intervention improved HAZ and WAZ scores. Our review indicated that CF provision had no significant impacts on height or weight gain, while previous reviews [5, 9] suggested otherwise, an explanation for this could be that these reviews also included studies that provided the intervention for less than 6 months, Imdad et al. , on the other hand, also included studies on children with moderate malnutrition.
As part of causal chain, it is well recognized that educational interventions improve feeding practices which then lead to improved growth outcomes. The educational messages should lay emphasis on the importance of appropriate home prepared foods, hygiene and high energy foods and it is important to assess the recall of the messages by mothers once the messages are delivered and our review suggest a significant 62% increase in compliance with the imparted messages, reinforcing the importance of such intervention. Considerable variations were observed in the types of educational messages delivered and an attempt to assess the quality of educational messages and delivery strategies was difficult, but in general most of the studies delivered educational interventions of reasonably good quality with the appropriate use of charts, posters and booklets. The two studies that had the most impact on linear growth [15, 21], also provided clear messages regarding the use of affordable home-prepared animal source products which indicates that giving messages specifically promoting the use of nutrient-rich animal products may have an impact on growth. However, financial constraints limit the possibility of including adequate amount of animal products in the child's diet, particularly among food insecure populations. Thus, in food insecure populations these nutritional messages need to be combined with provision of adequate amounts of animal products. One option can be the use of protein-rich plant foods, however, most plant foods, especially staples, legumes, lentils and vegetables contain anti-nutrients which can reduce the bioavailability of micronutrients and interfere with digestion. These include phytate and alpha amylase. Processing is required in order to reduce the content of anti-nutrients such as phytate or addition of alpha amylase in order to increase the impact of plant foods. This is in turn associated with additional cost and required expertise.
Nutritional status has a strong and consistent relation to death from respiratory infections. Nutrition education and complementary feeding with or without education had a positive impact on reducing respiratory infections. A review by Rice et al.  reported that the risk of mortality from respiratory infections is increased by two folds to three folds if associated with anthropometric status. Respiratory infections are one of the leading killers of children in developing countries. Prevention of undernutrition can potentially have an indirect impact on reducing childhood mortality through respiratory infections.
There were a variety of complementary food(s) used as intervention in the included studies. Amongst these foods were maize, fortified fat based spread, food prepared from locally available raw ingredients, cereal and porridge. The scarcity of available studies and their heterogeneity as well as the variety in complementary feeding interventions makes it difficult to conclude one particular type of complementary feeding intervention as the most effective; moreover, the variation in the reported outcomes amongst studies makes it difficult to compare them.
In future, further trials are needed particularly from food insecure population in which interventions are consistent and standardized in terms of study design and quality, complementary food chosen, duration of intervention and should report consistent outcomes for growth and morbidity. However, the available evidence is sufficient to recommend that in food insecure populations, education should be accompanied with provision of affordable yet effective complementary food. Accelerated and concerted actions are required to deliver and scale up nutrition education and CF provision interventions that are cost-effective, feasible and effective in improving the nutritional status of children.