Inadequate iron and zinc intakes are major contributors to poor health in developing countries especially during infancy when supplementary feeding is prematurely introduced and breastfeeding is not sufficient to meet nutritional requirements. There is a growing consensus on the need to establish the efficacy of iron and zinc supplementation during infancy. A previous study in the same population showed that iron and zinc consumption were very low, 25% and 18% respectively of the recommended dietary allowance.
This present study was nested in an iron and zinc supplementation study conducted in Indramayu, Indonesia (IIZS) in conjunction with UNICEF and University Indonesia. The IIZS aimed to evaluate the effect of iron and zinc supplementation on infant growth and morbidity. This nested study aimed to assess the effect of dietary pattern, and initial haemoglobin on the infant's response to iron and zinc supplementation. The
design was a double-blind randomised controlled trial.
The IIZS sample included all healthy infants aged 5-7 months living in the 2 study sub-districts, still breastfeeding. For the nested study, an additional criteria that the mother allowed her infant to give the blood at start was applied. The IIZS study was conducted from December 1997 to January 1999 and the nested study from June 1998 to January 1999. Before the study, informed consent was collected. No families refused to join the study. The sample size for the IIZS was 800. The final 200 children recruited into the IIZS were used for the nested study (one mother refused to give her infant's blood, leaving 199). The children were randomly allocated to treatment groups (zinc 10 mg/day; iron 10 mg/day; zinc and iron 10 mg/day each; or placebo), given in 2 mL syrup 5 day a week for 6 months by trained village cadres. Prior to the study; infants were given a 100,000 IU dose of vitamin A. The original
sample size for the nested study was 50 per treatment group.
Anthropometry and blood (for Haemoglobin) assessments occurred at baseline and end of supplementation. Morbidity and food habits data were collected weekly throughout the supplementation period. The additional information gathered for the nested study's purposes included food history, food recall, and hair zinc. A food history was taken at the end of the supplementation period together with a one-day dietary recall. Nutrient intakes were estimated using Indonesia Food Composition Table and World Food 2. Hair samples were collected by interviewers 2 months after supplementation finished. The main outcomes assessed were haemoglobin (Hb), hair zinc, anthropometric indices, and morbidity (diarrhoea, pneumonia; measles; runny nose; liquid from ears).
The supplement effects were estimated by regression analysis using an intention-to-treat approach. The impact of dietary pattern on
supplement effects was assessed by evaluating interaction between dietary factors and supplement intake in these models.
In this nested study, the following overall supplementation effects were found. Iron had a significant positive effect on Hb. A consistent trend to benefit for all anthropometric indices was also found, but these were not significant. No significant effect of iron on morbidity was found. Zinc had a significant preventive effect on persistent diarrhoea, and was associated with a greater risk of measles, but it was not consistent across all models. In addition, zinc had a significant and consistent association with a greater occurrence of runny nose. Iron and zinc had a significant interaction in their effect on measles, giving a net reduction in risk.
The following interactions between dietary intake and supplementation effects were found consistently, indicating deferential benefit: for growth, between zinc
supplement and Available Iron (AI) on the difference in Weight for height at start and final period (WHZ31); zinc supplement and Available Zinc (AZ) on the difference in Mid Upper Arm Circumference at start and final period (MUAC31). For morbidity, zinc supplement and fruit; and vegetables and tubers on diarrhoea; zinc supplement and milk; and AI and iron supplement on measles; zinc supplement and animal product, and zinc supplement and vitamin on runny nose. The direction of causality for the food/morbidity results are not clear since the data were collected concurrently throughout the study.
In conclusion, iron had a beneficial effect on improvement of Hb; and the interaction of iron and zinc had a protective effect against measles. Zinc supplementation had a complex interaction with diet that needs further investigation. In analysis of the overall IIZS, trends in the nested study may become statistically significant in analyses with the larger sample size.