Objective. To determine the behavioral and developmental effects of preventing iron-deficiency anemia in infancy.
Methods. Healthy full-term Chilean infants who were free of iron-deficiency anemia at 6 months were assigned to high- or low-iron groups or to high- or no-added-iron groups. Behavioral/developmental outcomes at 12 months of age included overall mental and motor test scores and specific measures of motor functioning, cognitive processing, and behavior. There were no differences between high- and low-iron groups in the prevalence of iron-deficiency anemia or behavioral/developmental outcome, and they were combined to form an iron-supplemented group (n = 1123) for comparison with the no-added-iron group (n = 534).
Results. At 12 months, iron-deficiency anemia was present in 3.1% and 22.6% of the supplemented and unsupplemented groups, respectively. The groups differed in specific behavioral/developmental outcomes but not on global test scores. Infants who did not receive supplemental iron processed information slower. They were less likely to show positive affect, interact socially, or check their caregivers’ reactions. A smaller proportion of them resisted giving up toys and test materials, and more could not be soothed by words or objects when upset. They crawled somewhat later and were more likely to be tremulous.
Conclusions. The results suggest that unsupplemented infants responded less positively to the physical and social environment. The observed differences seem to be congruent with current understanding of the effects of iron deficiency on the developing brain. The study shows that healthy full-term infants may receive developmental and behavioral benefits from iron supplementation in the first year of life.
Iron-deficiency anemia affects an estimated 20% to 25% of infants worldwide, with a higher proportion having iron deficiency without anemia.1–3 Despite the high worldwide prevalence, many countries, both industrialized and developing, have not made routine iron supplementation for healthy term infants a priority. The expense and effort of prevention or screening and theoretical concerns about iron overload or interference with immunity, absorption of other trace minerals, or breastfeeding have been issues.4–6 A basic question, however, has concerned functional consequences of iron deficiency, and effects on infant behavior and development have been at the core of the debate.
Evidence for a cause-effect relationship between poorer behavioral/developmental outcome and early iron deficiency remains equivocal.7 A recent comprehensive review8 detailed the research briefly summarized here. Case-control studies contrasting infants with iron-deficiency anemia with a comparison group generally find lower mental and motor test scores and other behavioral differences. There has been little or no evidence of lower developmental test scores among infants with iron deficiency that is not severe enough to cause anemia. Results after iron therapy vary. Most studies report persisting differences, even long term. Only 1 showed complete correction of test score differences in infancy.9 These results suggest either that iron-deficiency anemia in infancy has some effects that cannot readily be corrected with treatment, as in the animal model,10 or that other factors are the cause of poorer behavior and development.
A few studies of behavior and development have provided iron prophylactically to some infants but not others. No consistent pattern of results has emerged from the 3 published preventive trials in full-term healthy infants,11–13 and some methodologic issues have become apparent. The period of iron supplementation has varied, and no study has ascertained that infants were free of iron-deficiency anemia before entering the trial. Studies have focused on global developmental scores, which neither predict later functioning14 nor assess specific processes that might be affected by iron deficiency during early development. Furthermore, large samples are required to detect effects, because there is yet little indication that overall development is poorer in iron deficiency without anemia. A recent editorial concluded that the “jury is still out” and “large trials of both iron supplementation in infants and iron treatment in children with iron deficiency aneamia are urgently needed.”15
The purpose of this study was to determine the behavioral and developmental effects of preventing iron-deficiency anemia in healthy full-term infants. We predicted that iron supplementation effective in reducing iron-deficiency anemia would also result in better behavioral and developmental outcome. The study was designed so that the iron-supplemented group corresponded to the recommendations of the American Academy of Pediatrics (breastfeeding and use of supplemental iron or iron-fortified formula until 12 months of age). In a preliminary analysis of developmental test scores only, there were no differences between infants who did or did not receive additional iron.16 When all outcomes were examined in this final analysis, the iron-supplemented group performed better in every domain except global test scores.
The study was initially designed to be a double-blind, randomized, controlled trial comparing the behavioral and developmental effects of iron supplementation and no-added iron. However, unforeseen circumstances related to funding and secular changes in infant feeding affected the design such that the study could not be the randomized, controlled trial planned.17 When the study started, many Chilean infants were weaned from the breast by 6 months. Therefore, infant formula was the vehicle for supplementation, but, to avoid interference with breastfeeding, we planned on enrolling only those infants who had started to receive some cow milk or formula by 6 months of age. To conduct the study in the face of a 25% budget cut, we sought to have infant formula donated; Abbott-Ross Laboratories generously agreed. Because no-iron formula was no longer made, the study started with a low-iron condition instead of the no-added-iron condition originally planned. Study infants were randomly assigned to high- or low-iron formula (12 mg/L or an average of 2.3 mg/L, respectively). Part way through the study, we made the unexpected observation that the amount of iron in the low-iron formula was sufficient to prevent iron-deficiency anemia, although the infants’ iron status was not as good as those on high-iron formula.18 We also observed that breastfeeding had increased in the community as a result in part of a highly effective national campaign to encourage breastfeeding.
In mid-1994, the study was modified to enroll qualifying infants even if they had not started any bottle-feeding and to replace the low-iron with a no-added-iron condition. Thus, there were changes in enrollment criteria and supplementation vehicles. These changes in study design are diagrammed in Fig 1. To increase the size of the no-added-iron group rapidly and include more infants who were taking little or no formula/cow milk, while still allocating infants to high-iron formula, infants who were consuming at least 250 mL/d cow milk or formula were randomly assigned in a 1-to-3 ratio to high-iron formula or unmodified cow milk plus multivitamins without iron; infants who were taking <250 mL/d (“exclusively breastfed”) were randomly assigned in a 1-to-2 ratio to a liquid multivitamin preparation with or without iron.
These changes in study design meant that it was not a straightforward randomized, controlled trial. Instead, a complex design emerged with 6 groups varying in entrance criteria and supplementation procedures, with n’s in comparable conditions too small to have adequate power for causal inference. We therefore approached statistical analysis in a way that would include data from all infants and best approximate the study’s original purpose of assessing the behavioral and developmental effects of preventing iron-deficiency anemia in healthy full-term infants. Specifically, because preliminary analysis showed no differences between high- or low-iron groups in developmental/behavioral outcome at 12 months (see below), they were combined to form an iron-supplemented group for comparison with the no-added-iron group.
Developmental test scores from case-control studies provided the best data available to estimate the sample size required in a preventive trial. Case-control studies indicated that developmental test scores of infants with iron-deficiency anemia average approximately 10 points lower than scores of infants with better iron status. If the prevalence of iron-deficiency anemia in an unsupplemented population were 20% to 25%, as is often the case in otherwise healthy, well-nourished infants, then iron supplementation could conceivably prevent lower scores in 1 of every 4 or 5 infants. Considering that test scores for this 20% to 25% would otherwise be approximately 10 points lower, the mean for the supplemented group overall would be 2 to 2.5 points higher than the unsupplemented group. These considerations guided the study, which was originally designed to have sufficient power to detect a group difference of 2 points, requiring a total sample of 3200 infants. Enrollment was stopped when the sample size was sufficient to have 90% power to detect a 2.5-point difference in developmental test scores. A total of 1657 infants completed the study.
The study was conducted between September 1991 and August 1996 in 4 contiguous working-class communities on the outskirts of Santiago, Chile. Parasites causing blood loss, malaria, hemoglobinopathies, and high lead levels were almost nonexistent. All but 8 infants were initially breastfed. Clinics distributed unmodified powdered milk as part of a legally required and highly effective program for preventing generalized undernutrition. Screening infants for anemia was not a regular part of pediatric care, and routine iron supplementation was not the policy in Chile at the time.
Infants who received their usual health care at the community clinics were carefully screened to enroll healthy infants who did not have iron-deficiency anemia at 6 months (Table 1). The following entrance criteria were used: birth weight ≥3.0 kg, singleton term birth, routine vaginal delivery, no major congenital anomalies, no major perinatal complications, no phototherapy, no hospitalization for longer than 5 days, no chronic illness, and no iron therapy. The 3.0-kg birth weight cutoff was used because some clinics had a preexisting program providing iron to infants who weighed <3 kg. Exclusion criteria were residence outside the neighborhoods; another infant <12 months in the household; infant in child care; illiterate or psychotic caregiver or no stable caregiver available to accompany the child for appointments; and, until mid-1994, “exclusive” breastfeeding, defined as <250 mL/d cow milk or formula. Refusal/dropout before group assignment totaled 6.0%. Attrition after group assignment was 7.8%. There were no differences between those who did or did not complete the study in infant characteristics (birth weight, gestational age, sex, growth, and temperament), family characteristics (household size, father absence, parental education, maternal depressed mood, and Home Observation for Measurement of the Environment score), or group assignment.
Hematology and Supplementation
Infants were screened to prevent those with iron-deficiency anemia from entering the study. Fingerstick hemoglobin levels were determined at 5 to 6 months by HemoCue (Leo Diagnostics, Helsingborg, Sweden). A venipuncture to document iron status (7–10 mL of blood) was promptly performed for HemoCue values <103 g/L. Anemia at 6 months was defined as a venous hemoglobin ≤100 g/L. Iron deficiency was defined as 2 of 3 abnormal iron measures (mean cell volume <70 fl, erythrocyte protoporphyrin >100 μg/dL red blood cells [1.77 μmol/L], serum ferritin <12 μg/L). Infants with iron deficiency anemia at 6 months were treated with oral iron and did not enter the trial.
Formula/milk was provided in powdered form in identical cans and vitamins in identical bottles, distinguished only by labels with different numbers (several for each condition), which were placed by laboratory personnel who had no contact with families or fieldworkers. At clinic visits, personnel used lists of predetermined, randomly generated numbers to give participating infants the next available milk/vitamin number appropriate for the infant’s feeding method. Consumption was verified at weekly home visits and monthly clinic appointments.
Infants received monthly pediatric check-ups and growth measurements. Project physicians were instructed to make clinical judgments as if infants were not in a study. Thus, if doctors were concerned that an infant might be anemic, then they were to request a blood test and/or treat the infant. No infant was eliminated for this reason. All infants received a venipuncture at 12 months for determination of iron status. The criterion for iron deficiency was the same as at 6 months, but the cutoff for anemia at 12 months was a hemoglobin level <110 g/L. Infants with iron-deficiency anemia were treated with oral iron. Blood lead levels were measured in the last 331 study infants. The mean lead level was 7.8 ± 0.2 μg/dL, with no statistically significant differences between groups and no significant negative correlation with developmental/behavioral outcomes.
Continuity of pediatric care, monthly check-ups, and testing for anemia before and after the trial provided infants with considerably closer monitoring than they would otherwise have received. Study participants were given formula or milk, vitamins, pediatric care, study tests and evaluations, and transportation free of charge. The study was approved by the appropriate Institutional Review Boards of the University of Michigan, University of Chile, and the National Institutes of Health Office of Protection From Research Risks.
Behavioral and Developmental Assessments
For characterizing infant behavior and development at group assignment, the Fagan Test of Infant Intelligence14 and a measure of temperament19 were administered. Toward the end of the study, these measures could be obtained only for a randomly selected 10% because of lack of funds. N’s for the Fagan test and temperament measure were 1039 and 1357, respectively. Overall developmental outcomes at 12 months were the Mental Developmental Index and Psychomotor Developmental Index of the Bayley Scales of Infant Development.20 The Fagan test provided specific cognitive outcome measures at 12 months. Behavioral outcome measures were derived from the Behavior Rating Scale.21 Data on the age of crawling and walking were collected prospectively during weekly home visits between 6 and 12 months until the latter part of the study when this labor-intensive aspect could no longer be supported. Motor milestone data were available for 967 infants.
Information on family background included household composition, parental education and occupation, other indicators of socioeconomic status,22 maternal depressed mood,23 maternal IQ,24 life stresses,25 stimulation in the home,26 and so forth. Data on parental education were available for the entire sample, but toward the end of the study, some of these measures could be obtained only for a randomly selected 10%. Even for measures subjected to sampling, n’s averaged well over 1000 (range: 995–1379). There were no significant differences in family background between infants with and without complete background data. Nutritionists conducted the socioeconomic evaluation in the home after study criteria were confirmed. Weekly measures of feeding were obtained by fieldworkers. Psychologists conducted other assessments of the mother and the family and all infant evaluations. Personnel were specially trained and standardized for performing each measurement (≥80% interrater reliability).
Statistical analyses followed the general linear model for continuous variables, using multiple regression to control for background factors. Analysis of categorical variables used the χ2 test and logistic regression. The effect of missing background data on analysis of outcomes controlling for family factors was minimized by 2 data analytic strategies. In regression analyses, pairwise deletion of missing data was used in the covariance matrix to use all available data. For analyses that did not allow this strategy, we maximized relevant information by constructing a composite of family background data using factor analysis weighting and available-data mean combination, dividing by the number of items available. An α level of 0.05 was used in tests of statistical significance.
We previously reported that the high- and low-iron groups up to the 1994 change in enrollment criteria were comparable in background and the prevalence of iron-deficiency anemia at 12 months (2.8% vs 3.8%, n = 430 and 405, respectively) but differed in the proportion with iron deficiency overall (39% vs 20%).18 (This difference provides good evidence that the infants actually took the assigned formula and that the high-iron formula had the expected effect of improving iron status more than low-iron formula.) Because the high- and low-iron groups differed with respect to iron deficiency and the effects of iron deficiency without anemia are still unclear, a preliminary step in data analysis was to determine whether there were any differences in behavior and development. The groups were similar at baseline except for an isolated difference in temperament: infants in the low-iron group were considered to be slightly less adaptable. There were no statistically significant differences on any behavioral or developmental outcome measure at 12 months. Another preliminary step was to compare infants who were enrolled before and after the changes in study design of July 1994. This analysis was based on infants who were assigned to the high-iron condition, as this was the only group with enrollment before and after the changes. Other than anticipated differences in feeding, there were no differences in infant or family characteristics. Therefore, the high- and low-iron groups were combined for comparison with the no-added-iron group.
Final Study Groups
The final study groups, totaling 1657 infants, were similar in most family background and infant characteristics (Table 2). However, they differed in feeding as expected, with the most intense breastfeeding in the no-added-iron group. At the outset, infants in the no-added-iron group were advantaged in other ways as well. They weighed 50 g more at birth, were bigger at study entry, and had slightly higher screening hemoglobin levels. They were considered by their mothers to be slightly less difficult and unpredictable. Their mothers had somewhat fewer symptoms of depression, and their homes were slightly more stimulating. All background factors, including these differences, were controlled statistically.
In keeping with larger size at birth and 6 months, infants in the no-added-iron group remained larger at 12 months. Differences in gain in weight or length during the study did not reach statistical significance, but the increase in head circumference was significantly greater. There were dramatic differences in iron status. Iron-deficiency anemia was present in 3.1% and 22.6% of the iron and no-added-iron groups, respectively. Iron deficiency (with or without anemia) was observed in 26.5% and 51.3%, respectively. There was no indication of iron excess in iron-supplemented infants (see Table 3).
Behavioral and Developmental Outcome
Table 4 compares iron-supplemented and unsupplemented infants, showing adjusted means on outcome measures after control for the set of background variables in Table 2. There were no significant differences in mental or motor test scores at 12 months. On the Fagan test, there was a significant effect of iron supplementation on looking time (the time infants looked at pairs of novel and familiar pictures). Infants who did not receive iron looked longer, on average, controlling for looking time at 6 months in addition to background factors. There were no group differences in novelty preference.
There was a significant effect of iron supplementation on the age of crawling or creeping. Unsupplemented infants crawled a bit later, on average, than infants who received iron supplementation. Differences in the exact age at walking could not be determined, because fewer than half of the sample were walking at study conclusion.
On the Behavior Rating Scale, there were significant effects of supplementation on 2 of 4 factors generated by a factor analysis of the Chile data (Table 4). Analysis of individual scales in factors with overall differences revealed the following: higher proportions of unsupplemented infants were rated as showing no positive affect, no attempt to interact socially, and no reference to others’ reactions to test materials/no bids for help; more of them were rated as very “adaptable,” ie, readily relinquishing test materials and accepting new ones more than half the time. Of the 5 scales that did not fit into any factor, there were significant effects of supplementation on Soothability and Tremulousness. A higher proportion of the no-added-iron group could not be soothed by words or objects when distressed. Thus, more required physical comforting or could not be soothed by any means. There was no difference in the proportion showing upset or distress. More of the unsupplemented group was rated as tremulous occasionally or more often during the test.
All comparisons were repeated for the subset of study infants who were randomly assigned to high- or no-added-iron groups after the changes in study design of mid-1994 (n = 288 and 534, respectively). There were no differences in background characteristics except expected differences in breastfeeding as a result of the disproportionate assignment procedure (see “Overall Design”). The subsets were comparable in baseline temperament and novelty preference, but the high-iron group had longer looking times on the Fagan test on study entry at 6 months (2.00 ± 0.06 seconds vs 1.86 ± 0.04 seconds in the no-iron group, F(1379) = 4.08, P < .05). At 12 months, the hematology results showed the same marked differences in iron status reported above using the entire supplemented group. The pattern of differences in behavioral and developmental outcome was also similar to that reported for the entire cohort. Because of much smaller sample size, however, the only comparison that reached statistical significance was factor 4 of the Behavior Rating Scale (unadjusted mean = 0.07 in the high-iron group vs −0.13 in the no-added-iron group, P < .01 controlling for background factors).
Differences in breastfeeding were statistically controlled in all analyses shown in Table 4. However, the entire no-iron group had been enrolled without restrictions on feeding method. Therefore, as another secondary analysis, we compared outcome in infants in the no-added-iron group depending on whether they were primarily breast- or bottle-fed or mixed. There were no statistically significant differences in hematologic or behavioral or developmental outcome (data available on request).
The unanticipated changes in study design meant that the study was not the simple randomized, controlled trial originally planned. The study thus does not provide the strongest possible basis for causal inferences, and there continues to be an urgent need for large randomized, controlled trials. Nonetheless, the study goes beyond previous research methodologically in several ways. The sample size was considerably larger than any other study, and the behavioral and developmental measures were more comprehensive. Infants were checked to make sure that none had iron-deficiency anemia before entering the study, and baseline measures of behavior and development were obtained. In-depth assessments of family background, such as mother’s IQ, depression, life stress, and stimulation in the home, were available for >1000 infants. In these well-characterized groups, iron-deficiency anemia was experimentally prevented in some but not others.
The study’s most important results are the differences between infants who did or did not receive supplemental iron in looking time on the Fagan test and social/emotional functioning. In contrast to developmental test scores in the first year, which do not predict later functioning, the clinical significance of these findings may be clearer. The longer mean looking time in the no-added-iron group, combined with a lack of difference in novelty preference, indicates less efficient information processing. Processing efficiency is considered a fundamental cognitive property, and longer looking time in infancy predicts poorer overall cognitive functioning later on.27–30 There is also strong theoretical and empirical support that social interaction and social referencing are essential for normal infant development, both cognitive and emotional.31,32 Thus, the observed behavioral differences in some infants in the no-added-iron group may contribute to functional isolation or self-induced impoverishment, limiting their ability to seek and receive stimulation from the physical and social environment.33,34 Follow-up of study children at 5 and 10 years of age will determine whether differences resolve or persist.
Most of our specific findings require replication, because previous preventive trials in healthy, well-nourished infants have generally focused on global test scores. However, personal and social skills were most affected in 1 such trial,12 and earlier case-control studies have reported more iron-deficient infants to be solemn, wary, hesitant, and so forth,34 (also see review8). These findings are congruent with our results. The lack of differences in global measures is not surprising. With screening at study entry, no infant had iron-deficiency anemia for >6 months—a short time in which to have an impact on overall developmental test scores. Furthermore, lower test scores have not been consistently observed in other preventive trials,11–13 and most infants in case-control studies have been older.8
The assignment procedures produced groups generally comparable in family background, making it unlikely that poorer outcome in the unsupplemented group was attributable to environmental disadvantage. If anything, this group started the study with slight advantages (in growth, temperament, breastfeeding, initial hemoglobin level, maternal depression, and stimulation in the home), yet the group finished the study disadvantaged in cognitive processing, behavior, and motor function. Nonetheless, it should be noted that differences are at the group level and relatively small in magnitude.
All infants received vitamins, but infants on formula received additional micronutrients, relative to those receiving cow milk. Had the entire no-added-iron group been on cow milk, differences between cow milk and formula might have contributed to our results. However, the no-added-iron group included many predominantly breastfed infants, who also showed altered behavior and development. It thus is unlikely that the findings are explained primarily by a difference between cow milk and formula.
Although more studies of basic processes are needed, recent research on iron deficiency and the developing brain (and brain/behavior relations more generally) makes it possible to postulate plausible mechanisms for the observed effects. This body of research and our related speculations are briefly summarized to stimulate hypothesis-driven studies in the future. Iron is required for many relevant central nervous system processes, the most studied being myelination and dopaminergic functioning (see reviews36,37). It seems logical that systems that rapidly myelinate during the period of iron deficiency might be especially vulnerable. Recent research has provided direct evidence of this by demonstrating altered transmission in both the auditory and the visual systems in children who had iron-deficiency anemia in infancy.38–42 A delay in visual processing (at the level of the optic pathway or its intracerebral connections) could result in longer looking times on the Fagan test, because infants were expected to discriminate between novel and familiar pictures rapidly and repeatedly. A somewhat later age at crawling could also be congruent with a delay or disruption in myelination. However, we are not suggesting that a week difference in the age of crawling is clinically important.
Early iron deficiency also adversely affects dopaminergic neurotransmission.36,37 Among many other functions, dopamine is involved in broad systems influencing behavioral activation and behavioral inhibition and plays an important role in the degree to which individuals experience inherent reward.43 The behavioral differences in the no-added-iron group seem to indicate less positive responsiveness to the physical and social environment. Even the seemingly paradoxical finding that a greater proportion of the unsupplemented group was very “adaptable” could mean that the infants did not find toys/objects sufficiently interesting to resist relinquishing them. Social referencing (joint attention), which showed the biggest difference, is a specific process rapidly emerging in the latter part of the first year, the period of iron deficiency in this study, and thus might be particularly affected. Social referencing provides the scaffolding for early learning from the physical and social environment.32 Altered dopaminergic functioning is also a known factor in extraneous motor movements, such as tremor. Tremor, which has not previously been assessed in early iron deficiency, was observed in more of the no-added-iron group.
It is also possible that other nonspecific, subtle differences contributed to the observed alterations. Although oxygen-carrying capacity and energy metabolism have not been found to be affected until hemoglobin levels drop considerably lower than those observed in this study, even slight increases in feeling unwell or fatigability could be manifested in less positive affect, less engagement, slower information processing, and so forth. Regardless of the specific mechanisms for direct effects of iron deficiency on central nervous system development, research in animal models indicates that behaviors that limit input from the physical and social environment produce secondary effects on brain structure and function.44,45
Using specific functionally relevant measures in a large sample, we demonstrated effects of iron supplementation on information processing, social/emotional behavior, and motor function. The study’s specific results require replication in future randomized, controlled trials,46 and more research on underlying mechanisms is needed. It should also be noted that many infants throughout the world have poorer health and nutrition than infants in this study. Effects of iron supplementation might differ among infants in less optimal condition. These cautions notwithstanding, this study shows that healthy full-term infants may receive behavioral and developmental benefits from iron supplementation in the first year of life.
This study was supported by grants from the National Institutes of Health (HD14122 and HD33487). The low- and high-iron formula (Similac), powdered cow milk, and vitamins (Vidaylin with and without iron) were donated by Abbott-Ross Laboratories (Columbus, OH).
We are grateful to Marisol Cayazzo and other project personnel (physicians, psychologists, home visitors, nutritionists, secretaries, drivers, research assistants, and other support staff) for competence and dedication; to the project’s External Advisory Committee (Peter Dallman, Ernesto Pollitt, Fernando Viteri, and Ray Yip) for expert advice and continued counsel and support; to Edward Nelson for critical comments on study design; to Joseph Campos for thoughtful discussion about interpreting the results; to Michael Moffatt for a draft diagram of the changes in study design; and especially to the study families for commitment to the project.
- ↵Florentino RF, Guirriec RM. Prevalence of nutritional anemia in infancy and childhood with emphasis on developing countries. In: Steckel A, ed. Iron Nutrition in Infancy and Childhood. New York, NY: Raven Press; 1984:61–74
- ↵Stoltzfus RJ. Defining iron-deficiency anemia in public health terms: a time for reflection. J Nutr.2001;131 :565S– 567S
- ↵Dallman P. Upper limits of iron in infant formulas. J Nutr.1989;119 :1852– 1855
- Yip R, Reeves JD, Lonnerdal B, Keen CL, Dallman PR. Does iron supplementation compromise zinc nutrition in healthy infants? Am J Clin Nutr.1985;42 :683– 687
- ↵Grantham-McGregor S, Ani C. A review of studies on the effect of iron deficiency on cognitive development in children. J Nutr.2001;131 :649S– 668S
- ↵Felt BT, Lozoff B. Brain iron and behavior of rats are not normalized by treatment of iron deficiency anemia during early development. J Nutr.1996;126 :693– 701
- ↵Williams J, Wolff A, Daly A, MacDonald A, Aukett A, Booth IW. Iron supplemented formula milk related to reduction in psychomotor decline in infants for inner city areas: randomised study. BMJ.1999;318 :693– 698
- ↵Morley R, Abbott R, Fairweather-Tait S, MacFayden U, Sterman MB. Iron fortified follow on formula from 9 to 18 months improves iron status but not development or growth: a randomised trial. Arch Dis Child.1999;81 :247– 252
- ↵Fagan JF, Singer LT. Infant recognition memory as a measure of intelligence. In: Lipsitt LP, ed. Advances in Infancy Research. Norwood, NJ: Ablex; 1983:31–78
- ↵Lozoff B, De Andraca I, Walter T, Pino P. Does preventing iron-deficiency anemia (IDA) improve developmental test scores? Pediatr Res.1996;39 :136A
- ↵Bayley N. Bayley Scales of Infant Development. New York, NY: The Psychological Corporation; 1969
- ↵Bayley N. Bayley Scales of Infant Development. 2nd ed. San Antonio, TX: The Psychological Corporation; 1993
- ↵Radloff L. The CES-D Scale: a self-report depression scale for research in the general population. Appl Psychol Meas.1977;1 :385– 401
- ↵Wechsler D. Manual for the Wechsler Adult Intelligence Scale. New York, NY: The Psychological Corporation; 1955
- ↵Weinraub M, Wolf B. Stress, social supports and parent-child interactions: similarities and differences in single-parent and two-parent families. In: Boukydis I, Zachariah CF, eds. Research on Support for Parents and Infants in the Postnatal Period. Norwood, NJ: Ablex; 1987:114–135
- ↵Caldwell BM, Bradley RH. Home Observation for Measurement of the Environment, Revised Edition. Little Rock, AK: University of Arkansas; 1984
- ↵Colombo J, Mitchell DW. Individual differences in early visual attention: fixation time and information processing. In: Colombo J, Fagen J, eds. Individual Differences in Infancy: Reliability, Stability, Prediction. Hillsdale, NJ: Erlbaum; 1990:193–227
- Detterman DK. What does reaction time tell us about intelligence? In: Vernon PA, ed. Speed of Information-Processing and Intelligence. Norwood, NJ: Ablex; 1987:177–200
- ↵Committee on Integrating the Science of Early Childhood Development. From Neurons To Neighborhoods: The Science of Early Childhood Development. Washington, DC: National Academy Press; 2000
- ↵Mundy P, Neal R. Neural plasticity, joint attention and a transactional social-orienting model of autism. In: Glidden LM, ed. International Review of Research in Mental Retardation. New York, NY: Academic Press; 2000:139–168
- ↵Wolf AW, Lozoff B. A clinically interpretable method for analyzing the Bayley Infant Behavior Record. J Pediatr Psychol.1985;10 :199– 214
- ↵Beard JL. Iron biology in immune function, muscle metabolism and neuronal functioning. J Nutr.2001;131 :568S– 580S
- ↵Roncagliolo M, Garrido M, Walter T, Peirano P, Lozoff B. Evidence of altered central nervous system development in infants with iron deficiency anemia at 6 mo: delayed maturation of auditory brain stem responses. Am J Clin Nutr.1998;68 :683– 690
- Li YY, Wang HM, Wang WG. The effect of iron deficiency anemia on the auditory brainstem response in infant. Natl Med J China.1994;74 :367– 369
- ↵Sarici SU, Okutan V, Dundaroz MR, et al. The effect of iron supplementation on visual-evoked potentials in infants with iron deficiency anemia. J Trop Pediatr.2001;47 :132– 135
- ↵Black JE, Jones TA, Nelson CA, Greenough WT. Neural plasticity and developing brain. In: Alessi NE, Coyle JT, Harrison SI, Eth S, eds. The Handbook of Child and Adolescent Psychiatry. New York, NY: John Wiley & Sons; 1998:31–53
- ↵Greenough WT, Black JE. Induction of brain structure by experience: substrates for cognitive development. In: Gunnar M, Nelson C, eds. Developmental Behavioral Neuroscience. The Minnesota Symposia on Child Psychology. Vol 24. Hillsdale, NJ: Lawrence Erlbaum; 1992:155–200
- ↵Martins S, Logan S, Gilbert R. Iron therapy for improving psychomotor development and cognitive function in children under the age of three with iron deficiency anaemia (Cochrane Review). In: The Cochrane Library, Issue 2, 2003. Oxford: Update Software
- Copyright © 2003 by the American Academy of Pediatrics