Comparison of the Efficacy and Acceptability of Three Types of Micronutrient Supplements in Ghana

Study design

We propose to conduct a randomized controlled study of infants recruited from Maternal and Child Health centers in the town of Koforidua, located in eastern Ghana. There are 8 such centers located in various parts of the town, each of which serves the mothers and children in the immediate vicinity. The vast majority of infants born in Koforidua are brought to the MCH centers beginning a few weeks after birth for monthly weight monitoring. A list of all infants attending the MCH centers who are approximately 5 mo of age, have no congenital abnormality or chronic disease, and are currently breastfed will be created during each clinic day. From these lists we will randomly select approximately 34 infants per month (during a recruitment period of 9 mo) who will be invited to participate in the intervention trial. These infants will then be randomly assigned to one of three groups:

1. "Sprinkles" group: Will receive micronutrient Sprinkles beginning at 6 mo of age.
2. "Foodlet" group: Will receive crushable multiple micronutrient tablet beginning at 6 mo of age.
3. "Nutributter" group: Will receive fat-based supplement (Nutributter) beginning at 6 mo of age

We have chosen not to include a negative control (placebo) group because there is a high rate of iron and vitamin A deficiency in this population, and we cannot leave children who are known to be anemic untreated during the 6-mo study period. Instead, we will assess a "Non-Intervention" (NI) group of infants at 12 mo of age, which will allow us to evaluate attained size and micronutrient status in the absence of any intervention. The NI group will be randomly selected from the subset of infants who were recorded on the MCH lists described above during the recruitment period for the intervention trial (and thus met the same selection criteria), but not randomly selected for the intervention trial. [We cannot take all of the infants who come to the MCH clinics each month because the field work would be unmanageable. Given the birth rate of 179/mo, there will be more than enough "left over" each month who won't be selected for the intervention, and from whom we can randomly select the non-intervention group once they reach 12 mo.] We plan to recruit approximately 9/mo for the NI group for a period of 9 months, starting 6 mo after we initiate recruitment for the intervention (i.e., when the "left-over" infants begin to reach 12 mo of age). The MCH lists will include contact information (address), so that we can find the NI mothers at 12 mo. When we contact them at 12 mo, we will confirm that their infants would have met the eligibility criteria for the intervention study (breastfed at 5 mo, and free of congenital abnormality, chronic disease or peanut allergy), and ask them to participate in the one-time assessment. At that time each NI family will be given compensation equivalent in value to the supplements provided to the intervention groups (such as a 6-mo supply of Nutributter).

Study site The study will be conducted in Koforidua, which has a population of about 87,000 and an average birth rate of 179/mo (between 1998 and 2001).

The University of Ghana will administer the project and provide scientific and technical support for the study. It will also provide office space for administrative tasks and laboratories for biochemical analyses of blood samples. UC Davis will provide supervision of the study through the Project Manager (Seth Adu-Afarwuah, a doctoral student at UC Davis), and will order and ship all of the necessary supplies and equipment.

Informed consent procedures: Approval for the study has been obtained from the UC Davis Human Subjects Review Committee and the University of Ghana (Noguchi Memorial Institute) Ethics Committee. Consent to conduct the study in the town will be obtained from the Koforidua Municipal Health Administration. Written consent to participate in the study will be obtained individually from the mothers of the infants. Consent forms explaining the purpose of the study, potential risks or discomfort, the rights of subjects and the benefits to the subjects will be available both in English and in the local language, Asante Twi. For illiterate mothers unable to read, the form will be read and explained to them in Asante Twi. To indicate their consent, mothers will be asked to either sign the consent form or thumb-print it.