This was a randomized controlled trial involving two intervention groups: Weekly and Cyclical Groups based on the type of supplementation.
The design of this study followed the ethical requirements established by Resolution 196/96 of the National Health Council of the Ministry of Health. The Research Ethics Committee of the Medicine School in São José do Rio Preto, Brazil (process # 0708/2006) approved the study, which was conducted from January 2007 to July 2008, in Bady Bassitt, a town in the State of São Paulo, Brazil.
The town of Bady Bassit was chosen as the setting for this study because of the possibility of enrolling the entire under five-year-old population attending all four government daycare centers in the town and because of the relatively similar background of all the children.
The subjects (252) were 24- to 59-month-old children from four local public daycare centers. All children attended the daycare centers from 7:00 a.m. to 5:00 p.m. and received the same type of food, which was provided by a central kitchen run by the local council. A typical daily diet at school included bread and butter or cheese, milk, biscuits and fruit for breakfast, rice, beans, meat (minced beef, chicken, liver or fish), vegetables and a fruit or jelly for lunch and bread and butter or cheese, biscuits and fruit juice before going home.
Two weeks prior to iron supplementation, all the children in the study had parasitological tests with stool (feces) analysis. Those who presented with infections at the beginning of the intervention, reported any infection within two weeks before the supplementation, presented with hereditary anemia or those who were already taking medication containing ferrous sulfate were excluded from the study.
Thus, of the 252 children who were attending the daycare centers, only 131 parents consented to permit their children participate in the study. Of these 11 were excluded, six due to infections, four were already taking medicines containing ferrous sulfate and one had hereditary anemia.
Hence, iron supplementation was started in 110 anemic and non-anemic children. The children were randomly assigned (table of randomly distributed numbers) to one of two intervention groups with 55 subjects each. Sample size calculations indicated that with ≥ 33 children per group, it would be possible to detect hemoglobin changes with a power of 0.8 and p-value = 0.05, considering a prevalence of anemia of around 40% and a 50% reduction.
During the study, 11 children left the daycare centers and the study. Thus, the data discussed in this paper is on 99 children that received all the iron doses. Iron supplementation was administered to the children on empty stomachs after they had been playing for a short time so that they better accepted the medication by staff of the daycare centers who also recorded compliance to treatment.
Codes were used to identify bottles of unflavored elemental Fe (ferrous sulfate) oral solution (30 mL) in both groups. The Fundação Remédio Popular, a laboratory kept by the government of São Paulo State, provided the medication, with one drop containing 1 mg elemental iron.
Children in the Weekly Group received a single 30 mg oral dose (30 drops – 1.5 mL) ferrous sulfate using a 2-mL disposable syringe every Wednesday during the ten months of the study. Children in the Cyclical Group received a daily dose of 30 mg of elemental Fe on 20 consecutive school days (one month) repeated after four months without medication.
If a child in the Weekly Group missed a dose due to absence from the daycare center, the missed dose was administered on the day of the child’s return to the daycare center. Children in the Cyclical Group who were absent during the supplementation period continued to receive iron upon their return to the daycare center until all 20 scheduled doses had been administered. Therefore, the duration of supplementation (10 months) was the same for both groups and all children received the same number of doses (40).
In order to compare the groups, anthropometric measurements were taken of all children prior to the intervention including weight (weighing scales: scale 50 g), height (unstretchable measuring tape: scale 0.1 cm) and age. Measurements were made by a single trained nurse. Moreover, venous blood samples were drawn for laboratory tests to quantify the serum hemoglobin of each child at the start and at the end of the experimental period. Hemoglobin concentrations were calculated using a Coulter STKS ® device (Coulter Electronics, Inc. Hialeah, Florida U.S.A.) which uses the cyanomethemoglobin method of analysis .
If a child still presented with anemia at the end of the study, he/she was referred for specialist treatment. Anemia in this age range means hemoglobin concentration levels lower than 11.0 g/dL .
At the end of the experimental period, the four staff members of the four daycare centers who had administered the medication answered the following question: “In your opinion, which of the two treatments was simpler to administer and why?”
The t-test for paired samples was used to compare means between groups, and the normal approximation test for two ratios or the Fisher’s exact test, as recommended, were used to compare event probability in both groups. Statistical analysis was performed using the Minitab computer program version 12.22 (Minitab Inc®) and the adopted level of significance was 5% (p-value < 0.05).