Treament and prevention of anemia with ferrous sulfate plus folic acid in children attending daycare centers in Goiânia , Goiás State , Brazil : a randomized controlled trial

The objective of this study was to assess the prevalence of anemia and the therapeutic and prophylactic response to ferrous sulfate and folic acid. A double-blind, randomized, controlled clinical trial was conducted with 196 children 6 to 24 months of age enrolled in municipal daycare centers in Goiânia, Goiás State, Brazil. The children were assigned to two treatment groups that received a daily dose (5 times a week) of either 4.2mg/kg/day of ferrous sulfate + folic acid (50μg) or 4.2mg/kg/day of ferrous sulfate + folic acid placebo. One of the prevention groups received 1.4mg/kg/day of ferrous sulfate + folic acid (50μg/day) and the other 1.4mg/kg/day of ferrous sulfate + folic acid placebo. Supplementation lasted approximately three months. Baseline anemia prevalence was 56.1% (95%CI: 48.963.1). After treatment, anemia prevalence in the folic acid group (14%) was lower than in the placebo group (34.9%) (p = 0.02). After prophylaxis in the non-anemic children, the incidence of anemia did not differ between the groups, but there was an increase in hemoglobin level in the folic acid group (p = 0.003). Iron plus folic acid was effective for the treatment of anemia and improvement of hemoglobin level in non-anemic children. Nutritional Epidemiology; Anemia; Ferrous Sulfate; Folic Acid; Child Day Care Centers Introduction Nutritional anemia can be due to iron, folic acid, or vitamin B12 deficiency 1. Iron is the main nutrient involved in nutritional anemia, followed by folic acid 2. Deoxyribonucleic acid (DNA) synthesis depends on the coenzyme folate for the biosynthesis of the nucleotide pyrimidine, which is required for cell division 3, and folate is thus one of the necessary nutrients for hematopoiesis 4. Iron deficiency anemia is the most common nutritional problem in the world 5, with a negative impact on mental, motor, cognitive, and immune development 6,7,8. An anemia prevalence greater than or equal to 40% is considered a serious public health problem 4, and in Brazil children 6 to 24 months of age constitute one of the most vulnerable groups 9,10. The World Health Organization and the Brazilian Ministry of Health recommend that infants receive exclusive breastfeeding until six months of life 11. It has been shown that absorption of iron from breast milk is reduced by the consumption of other foods or beverages, particularly before or after breatfeeding 12. With the introduction of supplementary feeding, infants require an additional source of iron to prevent iron deficiency anemia 4. Various studies have used various intermittent doses at different weekly frequencies for anemia prevention and treatment 13,14,15,16. Although the use of iron supplements is recommended for infants in areas with high prevalence ARTIGO ARTICLE Hadler MCCM et al. S260 Cad. Saúde Pública, Rio de Janeiro, 24 Sup 2:S259-S271, 2008 of anemia, their efficacy has not been totally elucidated 17. Some studies have suggested that other factors besides iron can limit the hemoglobin response and control of anemia, like vitamin A and other micronutrients or non-nutritional factors 13. A study of Mexican children showed that iron, vitamin A, vitamin B12, vitamin E, and riboflavin deficiencies were common at baseline. The authors observed that many children receiving iron supplements remained anemic (30%) after six months of treatment and at 12 months (31%), despite the eradication of iron deficiency 18. Thus, supplementation with one or more nutrients is one of the strategies to fight borderline micronutrient deficiency, since it can increase the effectiveness of programs to fight anemia. However, there is still limited information on interactions between micronutrients (when supplemented), based on pharmacological studies 19. It is important to assess the efficacy and effectiveness of iron supplementation compared to the combination of iron and folic acid in children younger than 24 months 20. In addition, the possibility of meeting the multiple needs for these micronutrients makes this an interesting strategy 21. The objectives of this study were to evaluate the prevalence of anemia in children from 6 to 24 months of age and the therapeutic and prophylactic response to ferrous sulfate plus folic acid on hemoglobin levels, in municipal daycare centers in Goiânia, Goiás State, Brazil.


Introduction
Nutritional anemia can be due to iron, folic acid, or vitamin B12 deficiency 1 .Iron is the main nutrient involved in nutritional anemia, followed by folic acid 2 .Deoxyribonucleic acid (DNA) synthesis depends on the coenzyme folate for the biosynthesis of the nucleotide pyrimidine, which is required for cell division 3 , and folate is thus one of the necessary nutrients for hematopoiesis 4 .
Iron deficiency anemia is the most common nutritional problem in the world 5 , with a negative impact on mental, motor, cognitive, and immune development 6,7,8 .An anemia prevalence greater than or equal to 40% is considered a serious public health problem 4 , and in Brazil children 6 to 24 months of age constitute one of the most vulnerable groups 9,10 .
The World Health Organization and the Brazilian Ministry of Health recommend that infants receive exclusive breastfeeding until six months of life 11 .It has been shown that absorption of iron from breast milk is reduced by the consumption of other foods or beverages, particularly before or after breatfeeding 12 .With the introduction of supplementary feeding, infants require an additional source of iron to prevent iron deficiency anemia 4 .Various studies have used various intermittent doses at different weekly frequencies for anemia prevention and treatment 13,14,15,16 .Although the use of iron supplements is recommended for infants in areas with high prevalence ARTIGO ARTICLE Hadler MCCM et al.
Some studies have suggested that other factors besides iron can limit the hemoglobin response and control of anemia, like vitamin A and other micronutrients or non-nutritional factors 13 .A study of Mexican children showed that iron, vitamin A, vitamin B12, vitamin E, and riboflavin deficiencies were common at baseline.The authors observed that many children receiving iron supplements remained anemic (30%) after six months of treatment and at 12 months (31%), despite the eradication of iron deficiency 18 .Thus, supplementation with one or more nutrients is one of the strategies to fight borderline micronutrient deficiency, since it can increase the effectiveness of programs to fight anemia.However, there is still limited information on interactions between micronutrients (when supplemented), based on pharmacological studies 19 .It is important to assess the efficacy and effectiveness of iron supplementation compared to the combination of iron and folic acid in children younger than 24 months 20 .In addition, the possibility of meeting the multiple needs for these micronutrients makes this an interesting strategy 21 .
The objectives of this study were to evaluate the prevalence of anemia in children from 6 to 24 months of age and the therapeutic and prophylactic response to ferrous sulfate plus folic acid on hemoglobin levels, in municipal daycare centers in Goiânia, Goiás State, Brazil.

Sampling
In 2005, Goiânia had 67 municipal preschools, of which 53 included daycare centers with 447 children from 6 to 24 months of age.These are fulltime preschools where children participate in educational activities with teachers trained in pedagogy or teaching, along with preschool staff.The centers are operated by the Goiânia Municipal Department of Education.
Children were selected through probabilistic sampling, i.e., the number of children from 6 to 24 months of age was checked at the municipal preschools that had daycare centers, and the probability of the daycare center being picked was proportional to the number of children enrolled there.Twenty-five daycare centers were selected, with 268 children, using a probabilistic draw and the Excel random numbers table (Microsoft Corp., USA).After application of the exclusion criteria and one refusal, 219 children were finally selected.
• Population: universe of children from 6 to 24 months of age attending municipal preschools with daycare in the year 2005.
• Sample: included 219 children from 6 to 24 months of age, born at term, non-twins, with parental approval for participation in the study, and who were attending municipal daycare centers with more than four children each.Exclusion criteria were: children with special needs, low birth weight (< 2,500g), with growth-impairing heart diseases or neurological syndromes, with sicklecell anemia or sickle-cell trait, under treatment for anemia at the time of the first interview or screening performed by the pediatrician, and those no longer attending the daycare center.The clinical trial excluded children with hemoglobin ≥ 7 and < 8g/dL, adopting the cutoff for exclusion used by Tee et al. 22 .
To design the intervention with 40 children in each treatment arm, with the aim of detecting a minimum difference of 0.5g/dL with a standard deviation of 0.7 and significance of 0.05, the test power was 0.89.This sample size is sufficient to detect differences of the same magnitude between the groups.The SAMPSI procedure from Stata 7.0 (Stata Corp., College Station, USA) was used for calculation.
A safety margin of 24% was added to the initially predicted sample of 160 children to cover potential losses during the study.Of the 219 selected children, eight left the daycare center, four failed to appear for blood samples, and the parents of eight canceled the original informed consent.Thus, 199 children participated in the first blood draw.Of these, three did not perform the complete blood count, so the assessment of anemia prevalence used a sample of 196 children that participated in the subdivision of groups for the clinical trial (Figure 1).Since four children presented hemoglobin ≥ 7 and < 8.0g/dL, they were submitted to special treatment prescribed by the pediatrician and were therefore excluded.Thus, from the initial sample to the randomization for the clinical trial, a sample of 192 children was established, but eight of them left the daycare centers before beginning medication.Therefore, 184 children actually began treatment with ferrous sulfate, and 161 of them (87.5%)concluded the treatment (Figure 1).Thus, there was a loss of 12.5% among the children that began treatment, and a total sample loss of 19.1% (161/199) among those that appeared for the first blood sample until the end of the experiment.

Study design
The current study was part of a larger project called Nutritional Anemia: Prevention and Treatment in Early Childhood, which aimed to assess the causality of anemia and diet and to compare different micronutrient supplements in the prevention and treatment of anemia in early childhood.The study done in Goiânia from April 2005 to March 2006 was therapeutic, according to the Medical Research Council (MRC), since it involved both treatment and prevention of diseases 23 .
This was a controlled, randomized, prospective, double-blind clinical trial with 192 children from 6 to 24 months of age.The design was lon-gitudinal, and the research unit was children in this age bracket on the date the first blood sample was drawn.
After informing the children's parents/guardians and obtaining written informed consent in duplicate, the first interview was held for sample selection and screening by the pediatrician to determine whether the child was eligible to participate in the study.
The study collected socioeconomic and cultural data and information on the parents; re-

S262
Cad. Saúde Pública, Rio de Janeiro, 24 Sup 2:S259-S271, 2008 alizes food and anthropometric surveys, blood counts and blood chemistry, stool tests, and fecal occult blood.Clinical examination of the participating children, using a previously tested form with modifications 14 and clinical charts.The interviewers were students, interns, and scientific initiation or extension scholarship students from the undergraduate course in nutrition, as well as a nutritionist and faculty members from the School of Nutrition, Federal University in Goiás (UFG), all previously trained.
Parents were informed that the blood tests posed no risk of contamination or infection, and that they could cancel their consent at any time, without harm the care provided to their children.

Intervention
For the intervention, the daycare centers were divided into four groups.Groups A and C received ferrous sulfate at different concentrations plus folic acid placebo for the prevention and treatment of anemia, and Groups B and D received ferrous sulfate plus folic acid.
• Group A: daily treatment with 4.2mg/kg/day of ferrous sulfate + folic acid placebo.
• Group C: daily prophylaxis with 1.4mg/kg/day of ferrous sulfate + folic acid placebo.
Ferrous sulfate drops were used with 25mg of elemental iron per mL.The dose administered to each child was calculated and prescribed by the pediatrician responsible for the children's follow-up, prevention, and treatment, based on the child's body weight as measured on the day of the first blood sample.Anemic children were treated with 4.2mg/kg/day of ferrous sulfate, with or without folic acid, administered five days a week (Monday to Friday), the equivalent of a mean daily treatment dose of 3mg/kg/day 4 .Non-anemic children (Hb ≥ 11g/dL) received 1.4mg/kg/day of ferrous sulfate five days a week, corresponding to a daily dose of 1mg/kg/day.This is the dose recommended by the Brazilian Society of Pediatrics since 1995, beginning with the introduction of food supplementation at 24 months of age 24 , and by the American Academy of Pediatrics for term infants with birth weight > 2,500g 25 .
Folic acid supplementation of 50µg/day was based on Stoltzfus & Dreyfuss 20 .
The pediatrician provided one copy of the test results to the mothers.The number of drops to be administered was written on a medical prescription that was kept in the daycare center, and these amounts were also recorded on the medication sheet.
The teachers, school staff, and nutrition interns were trained to administer the medication and record the data on the forms.The interns also acted as supervisors, and every two to three days, besides administering the medication (ferrous sulfate and then folic acid drops), they checked the forms to see whether the teachers or school staff had properly administered and recorded the medication on the other days of the week.On the days without supervision, the control was also done by means of telephone calls.
The children in each group of daycare centers received a median of 67 doses of medication.At the end of treatment, they had new blood counts and blood chemistry tests.Children that presented anemia and/or iron deficiency at the second examination received ferrous sulfate for two more months, furnished free of cost.Parents were instructed to bring their children to the nearest health clinic for a follow-up examination.

Blinding and use of placebo
Blinding was applied to the folic acid.In this study the mothers (who were informed ahead of time) and the other participants did not know whether folic acid or folic acid placebo was being administered.The variable with the type of medication administered was only included after duplicate keying-in of the data and performing VALIDATE in Epi Info 6.04d (Centers for Disease Control and Prevention, Atlanta, USA).
The placebo and folic acid were manufactured by a pharmaceutical company according to the required standards, using identical bottles and packaging, distinguished only by the batch number, with no potential conflict of interests.A control card was used to deliver the medicines to the daycare centers, to avoid possible switching of batches.

Blood counts and blood biochemistry
At baseline and at the end of treatment, aliquots of 8 to 10mL of fasting venous blood were drawn with disposable syringes and needles, placed in three separate tubes for: (1) a complete blood count (CBC); (2) folic acid and vitamin B12 analysis, with the tube wrapped in aluminum foil; and (3) ferritin and C-reactive protein, the latter capable of increasing ferritin in the presence of infection 4,26 .
CBC was performed with electronic counting in Cell-Dyn 3200 SL (Abbott Diagnostics, Mount View, USA), obtaining hemoglobin levels.Children with hemoglobin < 11g/dL were considered anemic 4 .Severity of anemia was classified as: severe when Hb < 7.0g/dL; moderate with Hb ≥ 7.0 and < 9.0g/dL; and mild with Hb ≥ 9.0 and < 11g/dL 27 .
Serum ferritin was assessed by chemoluminescence in the AxSym system by Abbott Laboratories of Brazil Ltd., using the specific kit.
Serum C-reactive protein (CRP) was measured by turbidimetry in Konelab 60i (Wiener, Finland) at the Clinical Laboratory of the University Hospital of the Federal University in Goiás.Children with CRP < 0.50 were considered inflammation-free.
Serum folate and vitamin B12 were measured by chemoluminescence with the Bayer Advia Centaur System (Bayer Diagnostics, Tarrytown, USA) at the Hermes Pardini Institute in Belo Horizonte.The cutoff points for detecting folic acid deficiency were < 3.0ng/mL for high risk of deficiency, 3-5ng/mL for moderate risk, and > 5ng/mL for low risk 28 ; vitamin B12 levels from 120 to 200pg/mL were considered borderline and < 120pg/mL as B12 deficiency 29 .

Monthly weight gain
The children were weighted undressed at the time of the first blood sample on an electronic platform scale by Kratos (Kratos Cas, Taboão da Serra, Brazil) with a capacity of 150kg, accurate to 50g, with tare function.Weight was measured to obtain mean monthly weight gain, defined as the difference between current weight and birth weight, divided by the decimal age, expressed in months and days.

Statistical analysis
The coded data were entered in duplicate, performing VALIDATE in Epi Info 6.04d.Statistical analyses used Statistical Package for the Social Sciences version 10.0 (SPSS Inc., Chicago, USA).
Statistical analyses used the following tests: Kolmogorov-Smirnov to assess goodness-of-fit with normal distribution, chi-squared, Student's t, Fisher's exact, paired samples t, and Mann-Whitney tests, with a significance level of 0.05 or 5%.
Treatment effect was estimated as the absolute risk reduction, relative risk, and relative risk reduction.Absolute risk reduction is the difference in risk between the control group (CR) and treatment group (TR).Relative risk (RR) is the ratio between risk in the treated group and control group.Relative risk reduction expresses the percentage reduction in events in the treatment group (TR) as compared to controls (CR), expressed as [(CR-TR)/CR] x 100%.Meanwhile, the number needed to treat is the absolute risk reduction number, that is, it demonstrates the number of patients that need to be treated to avoid one case of anemia 30 .
Study protocol no.098/04 was approved by the Institutional Review Board at the University Hospital of the Federal University in Goiás (UFG) on December 10, 2004.

Clinical trial
Figure 1 shows the flowchart of children that participated in the clinical trial.Children with hemoglobin ≥ 7.0 and < 8.0g/dL (n = 4), i.e., 2% of the children, were excluded from the clinical trial, since they began treatment with higher doses of ferrous sulfate, administered twice a day.The attending pediatrician prescribed the medication for all the children participating in the study and those excluded from it.Thus, of 196 children with a baseline complete blood count, 192 participated in the clinical trial, and of these 106 were anemic (55.2%) and 86 were non-anemic (44.8%).

Characteristics of participants in the clinical trial
Table 1 shows the characteristics of the 192 participants in the four intervention groups and the families' socioeconomic conditions.Among the participants, 97.4% were from Goiânia and five (2.6%) from Aparecida de Goiânia.
The study compared the characteristics of anemic and non-anemic children that received ferrous sulfate plus folic acid with those that received ferrous sulfate plus folic acid placebo.No significant differences were observed in gender, decimal age in months, birth weight, per capita income (in times the minimum wage), baseline hemoglobin, baseline ferritin after excluding children with infection (CRP ≥ 0.5), housing conditions, and maternal schooling.The groups were thus homogeneous (Table 1).
Cad. Saúde Pública, Rio de Janeiro, 24 Sup 2:S259-S271, 2008 There was a statistically significant difference in relation to hemoglobin level and decimal age (months) at the time of the first blood count, when comparing anemic and non-anemic children (data not shown).
When evaluating whether the children that concluded treatment differed from those that dropped out, no significant difference was observed in serum hemoglobin, folic acid, vitamin B12, or ferritin in relation to gender, per capita family income, maternal schooling, mean monthly weight gain, and proportion of anemic subjects.Only decimal age was lower among treatment dropouts (Table 2).

Effects of the intervention
Table 3 shows the effects of treatment for anemia in the treatment and control groups with and without folic acid.
Among the anemic children that completed treatment with ferrous sulfate plus folic acid, final anemia prevalence was 14%.However, among the anemic children that received ferrous sulfate plus folic acid placebo, final anemia prevalence was 34.9%.Folic acid a protective factor for the reduction of anemia prevalence (p = 0.02), as shown by comparison to anemic children that received folic acid placebo (Table 3).
The incidence of anemia among children that received prophylactic iron plus folic acid was lower than those receiving iron plus folic acid placebo, but the difference was not statistically significant (Table 3).
Final prevalence of anemia in anemic and non-anemic children treated with ferrous sulfate plus folic acid was 10.1% (8/79), as compared to 23.2% (19/82) among those that received ferrous sulfate plus placebo (Table 3).The absolute risk reduction was 0.21.Thus, the group treated with active folic acid showed 21% less anemia than the group that received folic acid placebo.
The group treated with folic acid showed a relative risk of anemia of 0.40 (95%CI: 0.17-0.93)compared to the control group, showing that the intervention was highly beneficial.The risk of remaining anemic in the placebo group was 2.5 times that of the group that received folic acid.The odds ratio was 0.3, i.e., similar to the value obtained for relative risk.
Relative risk reduction was 60%, meaning that treatment with ferrous sulfate plus folic acid allowed a 60% reduction in anemia compared to the group treated with ferrous sulfate plus folic acid placebo.
The number needed to treat (NNT) with ferrous sulfate and folic acid to avoid anemia was five patients treated to avoid one case of anemia.
Pre and post-treatment hemoglobin levels showed a significant increase in anemic individuals, both for those that received folic acid and those receiving placebo.Among non-anemic individuals, the increase only occurred in the group that received folic acid (p = 0.003).However, there was no statistically significant difference when comparing hemoglobin levels between the active folic acid and placebo groups (Table 4).

Number of doses received
When evaluating the number of doses or days of medication, we observed that the 25 th percentile, median, and 75 th percentile in the number of doses received were 66, 67, and 70, respectively in both the anemic (n = 86) and non-anemic group (n = 75).The Mann-Whitney test showed that the two groups did not differ as to the number of doses received (p = 0.71).
Comparing the anemic and non-anemic groups, there was also no difference in the number of doses received in the folic acid group as compared to the placebo group.
The number of doses or scheduled days of treatment was 67 days, so no children that participated in the second post-intervention blood sample showed low adherence (i.e., no children received fewer than 75% of the scheduled doses).

Serum folic acid in children in the clinical trial
Among the children with baseline anemia, serum folic acid levels did not differ between the groups.However, after the intervention the active folic acid group showed significantly higher folic acid levels than the placebo group (Table 5).Folic acid level increased significantly after supplementation with ferrous sulfate plus folic acid, but not with ferrous sulfate and placebo (Table 5).
Among children without baseline anemia, serum folic acid did not differ from the first to the second blood sample.Although there was an upward trend in the folic acid group, it was not statistically significant (p = 0.055, Table 5).

Discussion
The current study was part of a larger research project on the assessment of nutritional ane-mia in early childhood, conducted at municipal daycare centers in Goiânia.The principal focus was supplementation with two micronutrients, iron and folic acid, to prevent and treat anemia.
The municipal daycare centers contribute to the child's development and allow the mother to participate in the labor market, which in turn leads to greater availability of food in the home 31 .However, if the center is located far from the mother's workplace, it does not allow exclusive breastfeeding up to six months of age, considering that the official maternity leave in Brazil is only four months.
According to one study, children enrolled in a daycare center showed improved nutrient intake 32 , while another study showed inadequate food consumption among children attending a daycare center 31 .
The prevalence of anemia in this study (56.1%) among children 6 to 24 months of age in the 25 municipal daycare centers in Goiânia is similar that in other studies performed in the same city.Hadler et al. 33 observed a 60.9% anemia prevalence in 110 infants 6 to 12 months of age, while in a study in daycare centers with children 1 to 48 months of age the prevalence was 51% 34 .Other Brazilian studies have shown even higher prevalence than in the current study.In municipal daycare centers in the city of São Paulo, prevalence of anemia was 65.3% among children 6 to 12 months of age and 69% from 12 to 24 months 14 .In Cuiabá, the prevalence was 82% in children 6 to 24 months of age 35 .Analyzed jointly, these data confirm that children less than 24 months of age constitute one of the main risk groups for anemia 10,36,37 .
In the municipal daycare centers in this study, among anemic children less than 18 months of age, 17.9% (15/84) showed Hb < 9.0g/dL, while in the city of São Paulo similar hemoglobin levels were found in one-third of anemic children in this age bracket 15 .
Food fortification, supplementation, and nutritional education programs have been recommended to fight anemia 4 .In Brazil, Ruling RDC 344 of December 13, 2002 38 requires fortification of wheat flour and corn meal and corn flakes with 4.2mg of bioavailable iron and 150µg of folic acid per 100g of food product, as one of the measures to fight anemia and reduce the risk of neural tube defects and myelomeningocele.
Supplementation with ferrous sulfate and folic acid to treat anemia proved more effective than ferrous sulfate with placebo and was a protective factor against anemia, since the reduction in anemia prevalence was significantly greater.
Although the children did not present folic acid deficiency at baseline, administration of this micronutrient allowed a significant increase in serum folic acid and a greater and quicker reduction in anemia prevalence.This proves that folic acid can be combined with ferrous sulfate supplementation to treat anemia.
The Brazilian Ministry of Health created the National Iron Supplementation Program to prevent anemia in children from 6 to 18 months of age, using 25mg of elemental iron/week 40 .A scientific report by the Department of Nutrition of the Brazilian Pediatric Society in February 2007 40 contends that the above supplementation has still not been properly proven, especially in Brazil, and recommends, for prophylaxis of anemia, that when infants (born at term with adequate-for-gestational age birth weight) begin receiving complementary food (besides breastfeeding), they should receive 1mg of elemental iron/kg body weight/day unless they consume a minimum of 500mL/day of iron- enriched baby formula 40 .The existence of two different recommendations hinders the implementation of a supplementation program.Silva 16 compared different doses of ferrous sulfate for prevention of anemia in children 6 to 12 months of age and showed that the daily dose of 1mg/kg/day recommended by the Brazilian Pediatric Society 40 , which is the dose used in the current study, had better results than 2mg/ kg/day, or the weekly dose recommended by the Ministry of Health.
Definition of the recommended dose is also based on the child's iron nutritional status 4 .Anemic children receive a therapeutic dose, while non-anemic children receive prophylactic doses, like those administered in this and other studies in Brazil 4,16,35 .However, some studies have used the same amount of iron for both prevention and treatment of anemia 15,41,42 .
The results of the current intervention demonstrate that the administration of 4.2mg of ferrous sulfate/kg/day significantly increased the hemoglobin level in anemic children, both in groups receiving folic acid and those receiving folic acid placebo.The reduction in the prevalence of anemia was higher in those that also received folic acid, and although the increase in hemoglobin (1.56 ± 0.91g/dL) was higher than in those that received placebo (1.32 ± 1.2g/dL), the difference between the two groups was not statistically significant (p = 0.19).There are controversies as to the effect of iron and micronutrient supplementation on child morbidity 43,44 , including the non-recommendation of prophylaxis in areas with a high prevalence of malaria 44 .
Even some non-anemic children that received a prophylactic dose of ferrous sulfate presented anemia, in both the active folic acid and placebo groups.Other studies have also shown some anemia incidence in children receiving a prophylactic dose, using different types of iron in the medication 16,45 .
In the non-anemic children, the final anemia incidence did not differ between the groups, although there was an increase in hemoglobin among those receiving ferrous sulfate plus folic acid (p = 0.003), but not in those receiving placebo.Still, final hemoglobin did not differ between the two groups.

Conclusions and recommendations
Anemia prevalence was high in under-two children attending municipal daycare centers in Goiânia, even with the use of iron and folic acid supplemented flours.
A supplementation program with ferrous sulfate and folic acid, as developed in this study, proved effective in the reduction of anemia prevalence in children and improved the hemoglobin levels in non-anemic children.Such a program is easy to implement and does not require that the mother administer the medication at home.The children require nutritional follow-up and monitoring of anemia, and this type of program can be developed at municipal daycare centers or preschools.

Contributors
All the authors participated in the study design.M. C. C. M. Hadler coordinated the research, supervised the fieldwork, and participated in the data collection, processing, and analysis and drafting of the article.D. M. Sigulem and M. F. C. Alves also contributed to the data analysis, consultancy, and drafting of the article.V. M. Torres participated in the data collection and analysis, prescribed the medication, and conducted the clinical evaluation and follow-up of the children.

Figure 1 Clinical
Figure 1 Clinical trial fl owchart.

Table 1
Baseline characteristics of subjects in each group before supplementation with iron plus folic acid versus iron plus folic acid placebo *.
* Excluding children with hemoglobin ≥ 7 and < 8g/dL; one child lived in a squatting settlement.All the children received a therapeutic or prophylactic dose of ferrous sulfate plus folic acid or placebo; ** χ 2 test; *** Mean ± standard deviation or median (P25 -P75); # Student's t test for independent samples; ## Excluding children with infection (PCR ≥ 0.5) from the ferritin analysis; ### Mann-Whitney test; § Fisher's exact test.

Table 2
Comparison of subjects that concluded treatment and dropouts *.

Table 3
Prevalence of anemia in children with baseline anemia and incidence in non-anemic children after therapeutic or prophylactic dose.

Table 4
Mean hemoglobin concentrations (g/dL) in anemic and non-anemic children before and after treatment with ferrous sulfate plus folic acid or folic acid placebo *.

Table 5
Serum folic acid (ng/mL) in anemic and non-anemic children that received ferrous sulfate plus folic acid or placebo in the pre and post-intervention periods *.
* Excluding children with hemoglobin ≥ 7 and < 8g/dL and those without hemoglobin values from the second blood sample; ** Student's t test for independent samples; *** Paired samples t test.