DOI: 10.1542/peds.2010-1659; originally published online March 14, 2011; 2011;127;e927Pediatrics

and Chunming ChenSuying Chang, Li Wang, Yuying Wang, Inge D. Brouwer, Frans J. Kok, Betsy Lozoff

Preschool-Aged Chinese ChildrenIron-Deficiency Anemia in Infancy and Social Emotional Development in

  

  http://pediatrics.aappublications.org/content/127/4/e927.full.html

located on the World Wide Web at: The online version of this article, along with updated information and services, is

 

of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2011 by the American Academy published, and trademarked by the American Academy of Pediatrics, 141 Northwest Pointpublication, it has been published continuously since 1948. PEDIATRICS is owned, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

at N/A on February 16, 2012pediatrics.aappublications.orgDownloaded from

Iron-Deficiency Anemia in Infancy and Social EmotionalDevelopment in Preschool-Aged Chinese Children

WHAT’S KNOWN ON THIS SUBJECT: Preschool follow-up studiesof children who had chronic, severe iron deficiency in infancyhave revealed that these children have poorer cognitive andmotor development and lower levels of alertness, physicalactivity, positive affect, and verbalization than children with goodiron status in infancy.

WHAT THIS STUDY ADDS: This study was an investigation ofwhether nonanemic preschool-aged children who had chroniciron-deficiency anemia in infancy would show altered affect andbehavior compared to children who were nonanemic throughoutinfancy.

abstractOBJECTIVE: We aimed to compare affect and behavior of 3 groups ofnonanemic 4-year-old children: children with iron-deficiency anemia(IDA) in infancy whose anemia was not corrected before 24 months(chronic IDA) (n� 27); children with IDA in infancy whose anemia wascorrected before 24months (corrected IDA) (n� 70); and childrenwhowere nonanemic in infancy and at 24 months (n� 64).

METHOD: Mother and child dyads were invited to a local clinic room.Children’s social referencing, wariness, frustration-tolerance behav-ior, and affect were observed during a set of situations encountered inthe laboratory, including free play, stranger approach, novel toy, anddelay of gratification. The whole procedure was videotaped. The chil-dren’s affective and behavioral displays were coded by using a time-sampling (5-second segments) code scheme.Iron status of childrenwas determined on the basis of hemoglobin concentration measuredwith the cyanomethemoglobin method in blood samples obtained byfingerstick in infancy and at the ages of 24 months and 4 years.

RESULTS: Children who had chronic IDA in infancy displayed less pos-itive affect, less frustration tolerance, more passive behavior, andmore physical self-soothing in the stranger approach and delay ofgratification. In contrast, the behavior and affect of children whoseanemia was corrected before the age of 24 months were comparableto those of children who were nonanemic throughout infancy.

CONCLUSION: The results point to the potential benefits of preventingiron deficiency in infancy and treating it before it becomes chronic orsevere. Pediatrics 2011;127:e927–e933

AUTHORS: Suying Chang, MS,a,b Li Wang, PhD,c YuyingWang, PhD,a Inge D. Brouwer, PhD,b Frans J. Kok, PhD,b

Betsy Lozoff, MD,d and Chunming Chen, MSa

aChinese Center for Disease Control and Prevention, Beijing,China; bDivision of Human Nutrition, Wageningen University,Wageningen, Netherlands; cPsychology Department, PekingUniversity, Beijing, China; and dCenter for Human Growth andDevelopment, University of Michigan, Ann Arbor, Michigan

KEY WORDSiron-deficiency anemia, infants, social-emotional developmental,preschool-aged children

ABBREVIATIONSID—iron deficiencyIDA—iron-deficiency anemiaWAZ—weight-for-age z scoreOR—odds ratio

www.pediatrics.org/cgi/doi/10.1542/peds.2010-1659

doi:10.1542/peds.2010-1659

Accepted for publication Jan 5, 2011

Address correspondence to Chunming Chen, MS, ChineseCenters for Disease Control and Prevention, 27 Nan Wei Rd,Beijing 100050, China; e-mail: chencm@ilsichina.org.Suying Chang, MS, UNICEF Office for China, 12 Sanlitun Road,Beijing 100600, China; e-mail: schang@UNICEF.org orchangsuying@hotmail.com

PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).

Copyright © 2011 by the American Academy of Pediatrics

FINANCIAL DISCLOSURE: The authors have indicated they haveno financial relationships relevant to this article to disclose.

ARTICLES

PEDIATRICS Volume 127, Number 4, April 2011 e927 at N/A on February 16, 2012pediatrics.aappublications.orgDownloaded from

Iron deficiency (ID) is considered to bethe most common nutritional disorderin the world, and infants are particu-larly at risk.1 A survey of Chinese chil-dren performed in 2001 showed theprevalence of ID and iron-deficiencyanemia (IDA) to be 65.5% and 20.8%,respectively, among infants youngerthan 12 months and 43.7% and 7.8%,respectively, among children aged 12to 36 months.2

Results of numerous studies havedemonstrated that infants with IDAhave lower cognitive and motor testscores than infants with good iron sta-tus.3–8 Follow-up studies have revealedpersistent lower scores among chil-dren who had IDA in infancy eventhough they received iron treatment asyoung children.9–11 Although mentaland motor findings have generally re-ceived the most attention, alterationsin the social/emotional domain alsohave been consistently observed in in-fants with IDA. Compared with nonane-mic infants, infants with IDA vocalizeless and are more likely to be fearful,hesitant/wary, unhappy, inactive, eas-ily fatigued, have less endurance, andbe in close contact with mothers.3,9

Such behavioral and affective altera-tions have been interpreted as evi-dence of “functional isolation.”12 Ac-cording to the functional-isolationhypothesis, nutritional deficienciescontribute to changes in infants’ af-fect, activity, and attention that leadthem to seek less stimulation fromtheir physical and social environ-ments.13 In response to infants’ behav-ior, caregivers may offer less stimula-tion. Over time the alterations in childand caregiver behavior interfere withthe child’s normal acquisition of envi-ronmental information and adverselyaffect the child’s development.14 Pre-school follow-up studies revealedthat children with chronic, severe IDin infancy had poorer cognitive andmotor development,15–18 as well as

lower levels of alertness, physical ac-tivity, positive affect, and verbalizationthan children with good iron status ininfancy.19,20

In this study children were exposed tomore stressful laboratory contexts toelicit behaviors that would reflect thelong-term social and emotional im-pacts of IDA in infancy. On the basis ofthe functional-isolation hypothesis andpast research, we hypothesized thataffective and behavior differences as-sociated with IDA during infancy wouldbe long lasting. Specifically, we pre-dicted that nonanemic preschool-agedchildren who had chronic IDA in in-fancy would show altered affect andbehavior compared with children whowere nonanemic throughout infancy.

SUBJECTS AND METHODS

This study focused on affect and behav-ior at the follow-up assessment per-formed at the age of 4 years in childrenwho were participating in a longitudi-nal complementary food-supplementproject. The aim of the original projectwas to evaluate the effects of a comple-mentary food supplement among in-fants and young children in 5 poorcounties of Gansu province in thenorthwestern part of China during2001–2003. In each county, 3 villageswere randomly chosen. With the helpof village doctors, we enrolled all chil-dren aged 4 to 12 months in the vil-lages until the number of study partic-ipants reached 200 per county. Thesechildren were recruited as the experi-mental group. Using the same sam-pling methods, we selected another100 children per county in the nearbyvillages as the control group. One thou-sand children in the experimentalgroup received a sachet containing 10g of nutrient-dense complementaryfood supplement to be added to theirnormal diet. The composition of thesupplement was as follows: 6 mg ofiron, 4.1mg of zinc, 385mg calcium, 0.2

mg of vitamin B2, 7.0 �g of vitamin D,and 3.8 g of protein. The total energy ofa sachet was 167 kJ. Five hundred chil-dren in the control group received asachet of rice flour with added vegeta-ble oil to match the total energy of theexperimental supplement. All childrenwere given supplements up to the ageof 24 months, and all were given vita-min A capsules every 6 months.21

Weight and height measurementswere taken every 3 months, and hemo-globin concentration was tested every6 months. The number of children whoattended the 24-month-old assess-ment in the experimental and controlgroups was 642 and 352, respectively.

In 2004, children recruited in the orig-inal study were evaluated at 4 years ofage. Weight, height, and hemoglobinconcentration were measured, andbehavior and development were as-sessed. A total of 404 children at-tended the 4-year follow-up assess-ment. All methods were reviewedand approved by the ethics reviewcommittee of the China Center forDisease Control, and informed verbalconsent was obtained from all par-ents. Details of the original surveyand follow-up study have been pub-lished elsewhere.21

The present study focused on childrenwho attended the baseline survey, 24-month assessment, and 4-year follow-up: 232 children met this inclusion cri-terion. We excluded 71 children whowere anemic at the 4-year follow-up as-sessment, because preschool-agedchildren with IDA may show altered af-fect and behavior.22 Data for 161 chil-dren are reported here. They did notdiffer in background characteristicsfrom children with incomplete data.None of children involved attendedpreschool or day care and �97% ofthe mothers were farmers.

The children were categorized intogroups as follows: 27 children whowere anemic before 12 months, still

e928 CHANG et al at N/A on February 16, 2012pediatrics.aappublications.orgDownloaded from

anemic at the 24-month assessment,and nonanemic at the 4-year follow-upwere in the chronic-IDA group. Seventychildren who were anemic before 12months and nonanemic at the 24-month and 4-year follow-up assess-ments were in the corrected-IDAgroup. Sixty-four children who werenonanemic at all time points were inthe nonanemic group.

Data Collection

On the basis of cross-cultural researchby Chen and colleagues23 and the as-sessment battery of Goldsmith et al(H. H. Goldsmith, PhD, J. Reilly, PhD,K. S. Lemery, PhD, S. Longley, PhD, andA. Prescott, The Laboratory Tempera-ment Assessment Battery: PreschoolVersion, unpublished manuscript,1999), we designed a set of laboratorycontexts in which to measure chil-dren’s social referencing, wariness/inhibition, frustration tolerance, andaffect. Mother and child were invitedto a clinic room and spent 10 minutesin free play. During this period, we pro-vided some toys and puzzle tasks. Afterfree play, a male stranger came in, satin front of the child, and engaged thechild in a scripted conversation for5 minutes (stranger approach). Thestranger then left after telling the childthat he would show something specialto him or her. The stranger came backwith a box, put the box on the tablewhere the child was seated, and thenremoved the cover of the box. A specialtoy popped out, and the strangershowed an expression of astonish-ment (novel toy). In the last episode,the experimenter came in and gave thechild a snack, asking the child not totouch or eat it until the experimentercame back again (delay of gratifica-tion). The experimenter returned in1 minute. The whole procedure wasvideotaped. The mother was presentthroughout the process and was in-formed that she should not give anyinstructions to the child.

Behavior Coding

We coded child affective display usinga time-sampling (5-second interval)coding scheme. Positive affect wascoded for clear verbal and nonverbalexpressions of joy and pleasure inmother-child interactions, such aslaughing, smiling, singing happily,jumping with joy, or saying “I am sohappy!” Negative affect was coded forthe presence of anger, sadness, or cry-ing. Unengaged affect was coded forquiet and uninvolved behavior. Socialreferencing was coded as looking to-ward the mother or talking to themother during the conversation withthe stranger or before touching thenovel toy or the snack during thefrustration situation. Latency to touchthe novel toy, whether being in proxim-ity (within arm’s length) to themother during the stranger approach,whether response to the stranger, andthe time spent with passive behaviorsin the delay of gratification were codedto index wariness or inhibition behav-ior. Passive behavior was defined asdoing nothing, only looking at thesnack. Latency to touch the snack andphysical self-soothing in the snack-delay situation were also coded. Phys-ical self-soothing included thumbsucking, cloth or hair twisting, and us-ing soft or familiar objects, presum-ably for comfort or security.

The videotapes were coded by 4 Chi-nese graduate students in psychologywho were trained by the second au-thor. Interrater agreement averaged85% for child’s affect, social referenc-ing, and inhibited behavior on the ba-sis of 20% of the sample. Neither theexperimenters nor the coders wereaware of the children’s iron status.

Iron Status

Iron status was determined on the ba-sis of the concentration of hemoglobinin a fingerstick blood sample obtainedat age 4 to 12months, 24months, and 4

years. Hemoglobin concentration wasdetermined by the cyanomethemoglo-bin method. The cutoff for anemia wasa hemoglobin concentration of �110g/L at sea level. Because the study ar-eas were 1300 to 3000 m above sealevel, the hemoglobin cutoff was cor-rected according to the altitude ofeach county.24

Statistical Analysis

Independent sample t tests and �2

tests were used to assess differencesin background characteristics be-tween the groups. Affective and behav-ioral outcomes were compared be-tween the chronic-IDA-group and thenonanemic-group and the corrected-IDA-group and the nonanemic-group.For affective and behavioral outcomesthat were coded as continuous vari-ables (such as latency to touch thenovel toy), a general linear model wasusedwith control for confounding vari-ables. For the outcomes treated as cat-egorical data (such as percentage ofchildren who ate the snack), we per-formed logistic regressions to assessthe effect of the group status of iron onaffective and behavioral outcomes.Child gender was included as a covari-ate in all analyses, given that a greaterproportion of the children in thecorrected-IDA group were male. Thechild’s age and mother’s educationwere also considered as potential con-founding variables. Weight-for-age zscores (WAZ) were calculated by usingEpi Info (US Centers for Disease Con-trol) to determine the children’s nutri-tional status. WAZ did not correlatewith the outcome variables and there-fore was not included as a covariate inthe analysis. Statistical significancewas set at P � .05. All data were ana-lyzed with SAS 9.1 (SAS Institute, Cary,NC).

RESULTS

Descriptive characteristics for thechildren and their mothers are pre-

ARTICLES

PEDIATRICS Volume 127, Number 4, April 2011 e929 at N/A on February 16, 2012pediatrics.aappublications.orgDownloaded from

sented according to iron status group(Table 1). The proportion of male chil-dren in the corrected-IDA group washigher than that in the nonanemicgroup. The percentage of WAZs lessthan �2 in the corrected-IDA groupwas lower than that in the nonanemicgroup at baseline survey and 24-monthassessment. There were no other sta-tistically significant differences be-tween the chronic-IDA group and thenonanemic group or between thecorrected-IDA group and the nonane-mic group with regard to backgroundcharacteristics, such as child’s ageand growth and mothers’ age andeducation at the 4-year follow-upassessment.

Wariness and Inhibited Behavior

Children in the chronic-IDA groupspent more time in passive behaviors(P� .05) than those in the nonanemicgroup. There were no statistically sig-nificant group differences in the la-tency to touch the novel toy before orafter covariate control. The proportionof children in proximity to their moth-ers during the stranger approach andthe proportion of children who re-

sponded to the stranger’s questionsdid not differ between the 2 groups.With regard to social referencing, agreater proportion of children in thechronic-IDA group showed social refer-encing to their mother before talkingto the stranger and before and afterpicking up the novel toy, but the differ-ences were not significant (P � .05)(Table 2).

We did not find significant differencesin wariness/inhibited behaviors be-tween the children in the corrected-IDAgroup and nonanemic group.

Frustration Tolerance

Compared with children in the nonane-mic group, chronic-IDA group chil-dren’s latency to touch/pick up a snackwas significantly less (P � .05), andthey spent a longer time physicallyself-soothing (P� .01). There were nostatistically significant differences be-tween children in the corrected-IDAgroup and nonanemic group in theabove-mentioned behaviors. In boththe corrected-IDA group and chronic-IDA group a higher proportion of chil-dren ate the snack during the observa-

tion period compared with thenonanemic group (odds ratios [ORs]for children in the chronic-IDA group:5.50 [P � .05]; OR for children in thecorrected-IDA group: 2.36 [P� .05]).

Affect

We found significant differences inpositive affect between the chronic-IDAgroup and the nonanemic group dur-ing the delay-of-gratification task; chil-dren in the chronic-IDA group spentless time smiling (P � .05). Duringstranger approach, logistic analysisresults indicated that the odds of hav-ing a more positive affect were 76.2%lower for children in the chronic-IDAgroup compared with children in thenonanemic group (P � .05), and theproportion of childrenwho showed un-engaged affect was significantlyhigher in the chronic-IDA group (P �.05). In the observation of free play andnovel toy sessions, children in thechronic-IDA group averaged less timesmiling than those in the nonanemicgroup, but the difference did not reachstatistical significance.

The affective behaviors of children inthe corrected-IDA group were compa-rable to those of children in nonane-mic group.

Negative affect was not analyzed be-cause�2% of children cried or fussedduring the observation period.

DISCUSSION

Our findings support the prediction oflong-term social-emotional effects ofchronic IDA during infancy. We foundthat despite correction of IDA by 4years of age, children who had chronicIDA in infancy showed less positive af-fect and frustration tolerance andmore passive behavior, unengaged af-fect, and physical self-soothing thanchildren who were nonanemicthroughout infancy. Furthermore, wefound that children whose anemia wascorrected before 24monthswere com-

TABLE 1 Child and Family Characteristics According to Iron Groups

Nonanemic Group(n� 64)

Chronic-IDA Group(n� 27)

Corrected-IDA Group(n� 70)

Child characteristicsMale gender, % 51.6 66.7 77.1a

Age, mo 46.2� 3.1 46.6� 3.0 45.4� 3.3WAZ� �2, % (baseline survey) 9.4 3.7 1.4a

WAZ� �2, % (24-mo assessment) 9.8 14.8 7.1a

WAZ� �2, % (4-y follow-up) 4.7 7.4 1.4Hemoglobin, g/L6–12 mo 129.2� 7.0 110.2� 10.3a 110.0� 7.4a

24-mo assessment 134.0� 6.8 114.9� 4.3a 131.7� 7.44-y follow-up 135.4� 10.3 133.7� 10.6 138.0� 11.8Family characteristicsMother’s age, y 29.5� 3.0 30.7� 3.0 30.0� 3.7Mother’s education�5 y, % 44.4 52.1 58.5

Nonanemic group: nonanemic before age 12months, nonanemic at 24months, and nonanemic at 4 years; chronic-IDA group:anemic before age 12 months, persisting anemia at 24 months, nonanemic at 4 years; corrected-IDA group: anemic beforeage 12 months, nonanemic at 24 months, nonanemic at 4 years. Values are expressed as mean� SD or as percent (n) forcategorical variables. There were no statistically significant group differences in background characteristics according toresults of the independent t test for continuous variables and the �2 test for categorical variables, except for the proportionof boys and percentage of WAZs of less than�2 in the corrected-IDA group versus the nonanemic group. The proportion ofboys in the corrected-IDA group was higher than that in nonanemic group. The percentage of WAZs of less than�2 in thecorrected-IDA group was lower than that in the nonanemic group.a Significantly different from the nonanemic group (P� .05).

e930 CHANG et al at N/A on February 16, 2012pediatrics.aappublications.orgDownloaded from

parable in behavior to children whowere nonanemic throughout infancy.

The present study usedmore extensiveobservations than previous studies toexamine the affect and behavior of for-mer IDA, nonanemic preschool-agedchildren. We performed a delay-of-gratification task to assess the effectsof IDA on preschoolers’ capacity forself-control in the face of temptation.25

Delay-of-gratification tasks have notpreviously been used in ID studies. Re-sults of research in other areas haveshown that when 4-year-old childrenwho are more successful at waiting indelay of gratification situations reachadolescence, they are more attentiveand better able to concentrate, and ex-hibit greater self-control and frustra-tion tolerance than their peers.25

We included novel toy and stranger ap-proach situations specifically to pur-sue earlier observations of wary, hesi-tant behavior in infants with IDA. Novel

toy and stranger approach paradigmshave been used to examine individualdifferences in preschoolers’ behaviorin reaction to novel situations.26 Waryor behaviorally inhibited children typi-cally display low approach behaviorand stay in close proximity to theirmothers when confronted with arange of novel stimuli, including peo-ple, objects, and situations.22 There isincreasing evidence that this wary/inhibited behavioral pattern is a riskfactor for future problems, such asanxiety, depression, and negative self-perceptions of competence.22,27 In ourstudy, affect and behavioral altera-tions seemed especially apparent withincreased novelty, unfamiliarity, orstress.28,29 We found little difference inaffect and behavior during free play ornew toy presentation, but affect andbehavioral differences became appar-ent in the stranger-approach sessionand delay-of-gratification task. We

found that children in the chronic-IDAgroup were less successful at waiting,and showed more physical self-soothing and passive behaviors, andless positive affect.

Lozoff et al found that preschoolerswith IDA showed less social lookingwith their mothers during a play ob-servation that involved familiar andunfamiliar toys.22 Our finding on in-creased social looking in the chronic-IDA group is not directly comparable tothe results of Lozoff et al because all ofthe children in our study werenonanemic.

Our results require replication inother samples and settings, especiallybecause of our small sample size.Nonetheless, our observations of be-havioral alterations in preschool chil-dren with chronic IDA in infancy sup-port the findings of other research. Astudy in France followed children’siron status and development quotient,

TABLE 2 Child Affect and Behavior According to Iron-Status Group

Outcome Variables NonanemicGroup (n� 64)

Chronic-IDAGroup (n� 27)

Corrected-IDAGroup (n� 70)

Chronic-IDA Group andNonanemic Group, OR

(95% CI)

Corrected-IDA Group andNonanemic Group, OR

(95% CI)

Social referenceReference mom before talk to stranger, % 50.0 62.9 42.9 1.662 (0.563–4.676) 0.585 (0.356–2.066)Reference mom before picking up the toy, % 46.9 61.5 44.9 2.97 (0.96–9.17) 1.33 (0.54–3.27)Reference mom after picking up the toy, % 68.4 81.3 54.5 2.78 (0.51–15.1) 1.05 (0.32–3.43)Wariness or inhibited behaviorChild being in proximity (within arm length) tomother during stranger approach, %

48.4 37.0 54.3 0.737 (0.262–2.074) 1.432 (0.596–3.39)

Latency to touch the novel toy, s 27.2� 11.4 22.9� 16.4 38.0� 12.4 NA NAResponse to the stranger 82.8 85.2 78.8 0.84 (0.195–3.637) 1.420 (0.182–1.797)Passive behaviors, s 14.8� 4.3 23.0� 6.6a 18.1� 4.2 NA NAFrustration toleranceLatency to touch/pick up snack, s 24.8� 7.9 7.8� 11.4a 21.1� 3.5 NA NA% of children who ate the snack 15.4 25.9 30.0 5.497 (1.004–30.097) 2.357 (1.012–5.492)Seeking comfort or social support, s 17.2� 6.2 8.9� 9.1 21.9� 7.1 NA NAPhysical self-soothing, s 1.5� 2.3 7.2� 1.6a 4.3� 2.7 NA NA

AffectPositive affect during free play, s 54.4� 14.1 31.8� 21.6 54.7� 13.5 NA NAUnengaged affect during free play, s 205.5� 21.6 219.5� 32.9 183.6� 24.8 NA NAPositive affect during stranger approach, % 28.1 7.4 27.1 0.138 (0.023–0.818) 0.459 (0.174–1.215)Unengaged affect during stranger approach, % 50 66.7 44.3 0.284 (0.085–0.953) 0.671 (0.248–1.818)Positive affect during delay of gratification, s 11.0� 3.4 4.7� 5.2a 9.7� 3.2 NA NAUnengaged affect during delay of gratification, s 43.1� 5.2 51.2� 7.9 43.7� 5.0 NA NAPositive affect during novel toy session, s 36.9� 8.3 27.9� 17.5 31.7� 14.5 NA NAUnengaged affect during novel toy session, s 65.4� 8.2 73.1� 12.4 71.4� 9.3 NA NA

Values aremean� SE for continuous variables, with adjustment for background factors. Adjustedmeans are derived frommultiple regression of covariance controlled for age, gender, andmaternal education. Percentages are shown for categorical variables. ORs and 95% confidence intervals are from logistic regression. NA indicates not applicable; CI, confidence interval.a Significantly different from nonanemic group (P� .05).

ARTICLES

PEDIATRICS Volume 127, Number 4, April 2011 e931 at N/A on February 16, 2012pediatrics.aappublications.orgDownloaded from

including postural, coordination, lan-guage, and sociability quotients. Thestudy results indicated that IDA at 10months did not correlate with develop-ment quotient, whereas hemoglobinconcentration at 2 years did; it waspositively associated with overall de-velopmental, postural, coordination,and social quotients in children aged 2years.16 A longitudinal study in CostaRica included a structured mother-child interaction task for childrenaged 5 years. Children with chronic IDin infancy displayed lower levels ofphysical activity, positive affect, ver-balization, and reciprocal interactioncomparedwith childrenwith good ironstatus, despite the correction of theirIDA in infancy.20 In early adolescence,mothers and teachers rated the chil-dren who had chronic ID in infancy ashaving more internalizing behaviorproblems (such as anxiety/depressionand social problems).10 These findingspoint to the potential “functional isola-tion” significance of early behavior al-terations in children with chronic IDAin infancy. Social-emotional altera-tions in children with ID have been in-terpreted in light of the role of iron inthe function of the dopamine sys-tem.30,31 Rats with ID have alterations indopamine metabolism and exhibitedbehaviors; they show more anxious-like behaviors, have reduced explora-tion in new environments, and demon-strate a slower rate of habituation.32,33

Results of studies of the affectivechanges in infants with IDA (wariness,hesitance, absence of positive affect)seem to make sense in this context.9

Neonatal dopamine terminal damageleads to a lifelong hyperreactivity to

novel objects and experiences in anunfamiliar environment31 and seemsto have long-term effects on context-dependent attention and affective re-sponses.9 These findings partially ac-count for the altered affect andbehavior of the chronic-IDA group inthe stranger-approach session anddelay-of-gratification task.

We cannot eliminate the possibilitythat the altered behavior of thechronic-IDA group in this study wascaused by some relevant but unmea-sured factor(s), although the groupswere generally comparable in back-ground characteristics, such as ma-ternal depression and other environ-mental disadvantages, which couldaccount for both poor infant-feedingpractices (leading to IDA) and lack ofstimulation (leading to altered behav-ior and development).

The study was limited with respect tomeasurement of iron status. IDA wasestimated according to hemoglobinconcentration. Hemoglobin concentra-tion generally overestimates the prev-alence of IDA, because it does not ac-count for anemia attributable to othernutritional deficiencies (eg, vitamin Adeficiency), infections, hemoglobinop-athies, or ethnic differences in normalhemoglobin distribution.34 However, itseems that ID was a major cause ofanemia in this population; hemoglobinconcentrations showed an overall in-crease of 10.5 g/L after the supplemen-tation intervention in infancy.35,36 Of the27 children in the chronic-IDA group inthe follow-up sample whose anemiawas not corrected by the time of the24-month assessment, 16 did not re-ceive supplementation (control group)

and 11 received supplementation. Re-garding possible reasons for persis-tent anemia in these 11 children, itshould be noted that the food supple-ment provided a low dose of iron (6mgper day). This amount of iron wasmuch lower than the World Health Or-ganization recommended dose for IDAtreatment (2 mg/kg per day).24 It ispossible that a higher amount of ironwould have corrected IDA in the studychildren by the age of 24 months.

CONCLUSIONS AND IMPLICATIONS

The results of this study showed thatpreschool-aged children who hadchronic IDA in infancy showed less pos-itive affect, less frustration tolerance,more passive behavior, and morephysical self-soothing during a delay-of-gratification situation. Childrenwhose anemia was corrected beforeage 24 months were comparable insocial-emotional behavior to childrenwho were nonanemic throughout in-fancy and early childhood. These re-sults point to the potential benefits ofpreventing ID in infancy and treating itbefore it becomes chronic or severe.

ACKNOWLEDGMENTSThe 4-year follow-up survey and thiswork were supported by InternationalLife Sciences Institute and Ellison Med-ical Foundation-International NutritionFoundation.

We are grateful to the families for theircommitment and their continued par-ticipation. We especially appreciatethe efforts of students in Beijing Uni-versity in coding the children’sbehavior.

REFERENCES

1. Leung AK, Chan KW. Iron deficiency anemia.Adv Pediatr. 2001;48:385–408

2. Zhu Y, Liao Q. Prevalence of iron deficiencyin children aged 7 months to 7 years inChina. Zhonghua Er ke Za Zhi. 2004;42(12):886–891

3. Lozoff B. Iron deficiency and child development.Food Nutr Bull. 2007;28(4 suppl):S560–S571

4. Lozoff B, Brittenham GM, Wolf AW, et al. Irondeficiency anemia and iron therapy effectson infant developmental test performance.Pediatrics. 1987;79(6):981–995

5. Lozoff B, Wolf AW, Jimenez E. Iron-deficiencyanemia and infant development: effects ofextended oral iron therapy. J Pediatr. 1996;129(3):382–389

6. Hokama T, Gushi Ken M, Nosoko N. Irondeficiency anaemia and child develop-

e932 CHANG et al at N/A on February 16, 2012pediatrics.aappublications.orgDownloaded from

ment. Asia Pac J Public Health. 2005;17(1):19 –21

7. Walter T, Kovalskys J, Stekel A. Effect of mildiron deficiency on infant mental develop-ment scores. J Pediatr . 1983;102(4):519–522

8. Lozoff B, Brittenham GM, Viteri FE, Wolf AW,Urrutia JJ. The effects of short-term oraliron therapy on developmental deficits iniron-deficient anemic infants. J Pediatr.1982;100(3):351–357

9. Lozoff B, Beard J, Connor J, Barbara F, Geor-gieff M, Schallert T. Long-lasting neural andbehavioral effects of iron deficiency in in-fancy. Nutr Rev. 2006;64(5 pt 2):S34–S43;discussion S72–S91

10. Lozoff B, Jimenez E, Hagen J, Mollen E, WolfAW. Poorer behavioral and developmentaloutcome more than 10 years after treat-ment for iron deficiency in infancy. Pediat-rics. 2000;105(4). Available at: www.pediatrics.org/cgi/content/full/105/4/e51

11. Stoltzfus RJ, Mullany L, Black RE. Iron defi-ciency anemia. In: Ezzati M, Lopez AD, Rodg-ers A, Murray C, eds. Comparative Quantifi-cation of Health Risks: Global and RegionalBurden of Disease Attributable to SelectedMajor Risk Factors. Geneva, Switzerland:World Health Organization; 2004

12. Levitsky DA. Malnutrition and hunger tolearn. In: Levitsky DA, eds. Malnutrition, En-vironment, and Behavior. Ithaca, NY: CornellUniversity; 1979:161–179

13. Brown JL, Pollitt E. Malnutrition, povertyand intellectual development. Sci Am. 1996;274(2):38–43

14. Grantham-McGregor S, Ani C. A review ofstudies on the effect of iron deficiency oncognitive development in children. J Nutr.2001;131(2S-2):649–668

15. De Andraca I, Walter T, Castillo M. Iron defi-ciency anemia and its effects upon psycho-logical development at preschool age: a lon-gitudinal study. In: Nestle FoundationNutrition Annual Report (1990). Vevey,Switzerland: Nestec Ltd; 1991:53–62

16. Dommergues JP, Archambeaud MP, Ducot

B, et al. Iron deficiency and psychomotordevelopment tests: longitudinal study be-tween 10 months and 4 years of age [inFrench]. Arch Fr Pediatr. 1989;46(7):487–490

17. Palti H, Pevsner B, Adler B. Does anemia ininfancy affect achievement on developmen-tal and intelligence tests? Hum Biol. 1983;55(1):183–194

18. Lozoff B, Jimenez E, Wolf AW. Long-term de-velopmental outcome of infants with irondeficiency. N Engl J Med. 1991;325(10):687–694

19. Tamura T, Goldenberg RL, Hou J, et al. Cordserum ferritin concentrations and mentaland psychomotor development of childrenat five years of age. J Pediatr. 2002;140(2):165–170

20. Corapci F, Radan AE, Lozoff B. Iron deficiencyin infancy and mother-child interaction at 5years. J Dev Behav Pediatr. 2006;27(5):371–378

21. Chen C, Wang Y, Chang S. Effect of in-homefortification of complementary feeding onintellectual development of Chinese chil-dren. Biomed Environ Sci. 2010;23(2):83–91

22. Lozoff B, Corapci F, Burden MJ, et al.Preschool-aged children with iron defi-ciency anemia show altered affect and be-havior. J Nutr. 2007;137(3):683–689

23. Chen X, Hastings PD, Rubbin KH, Chen H, CenG, Stewart SL. Child-rearing attitudes andbehavioral inhibition in Chinese and Cana-dian toddlers: a cross-cultural study. DevPsychol. 1998;34(4):677–686

24. World Health Organization. Iron DeficiencyAnemia: Assessment, Prevention, and Con-trol. A Guide for Programme Managers.Geneva, Switzer land: World HealthOrganization; 2001.Available at: www.who.int/nutrition/publications/micronutrients/anaemia_iron_deficiency/WHO_NHD_01.3/en/index.html. Accessed February 2, 2011

25. Mischel W, Shoda Y, Rodriguez ML. Delay ofgratification in children. Science. 1989;244(4907):933–938

26. Rubin KH, Hastings PD, Stewart SL, Hender-

son HA, Chen XY. The consistency and con-comitants of inhibition: Some of the chil-dren, all of the time. Child Dev. 1997;68(3):467–483

27. Caspi A, Henry B, McGee RO, Moffitt TE, SilvaPA. Temperamental origins of child and ad-olescent behavior problems: from age 3 toage 15. Child Dev. 1995;66(1):55–68

28. Lozoff B, Klein NK, Nelson EC, McClish DK,Manuel M, Chacon ME. Behavior of infantswith iron-deficiency anemia. Child Dev.1998;69(1):24–36

29. Angulo-Kinzler RM, Peirano P, Lin E, AlgarinC, Garrido M, Lozoff B. Twenty-four-hour mo-tor activity in human infants with and with-out iron deficiency anemia. Early Hum Dev.2002;70(1–2):85–101

30. Lozoff B, Clark KM, Jing Y, Armony-Sivan R,Angelilli MR, Jacobson SW. Dose-responserelationships between iron deficiency withor without anemia and infant social-emotional behavior. J Pediatr. 2008;152(5):696–702

31. Beard J, Felt B, Schallert T, Burhans M, Con-nor J, Georgieff M. Moderate iron deficiencyin infancy: biology and behavior in youngrats. Behav Brain Res. 2006;170(2):224–232

32. Beard J, Chen Q, Connor J, Jones B. Alteredmonoamine metabolism in caudate-putamen of iron-deficient rat. PharmacolBiochem Behav. 1994;3(48):621–624

33. Beard JL, Connor JR. Iron status and neuralfunctioning. Annu Rev Nutr. 2003(23):41–58

34. Zimmermann MB, Hurrell RF. Nutritionaliron deficiency. Lancet. 2007;370(9586):511–520

35. Wang Y, Chen C, Jia M, Fang J. Effect of com-plementary food supplements on anemia ininfant and young children. Wei Sheng YanJiu. 2004;33(3):334–336

36. Dallman PR, Reeves JD, Driggers DA, Lo EY.Diagnosis of iron deficiency: the limitationsof laboratory tests in predicting responseto iron treatment in 1-year-old infants. J Pe-diatr. 1981;99(3):376–381

ARTICLES

PEDIATRICS Volume 127, Number 4, April 2011 e933 at N/A on February 16, 2012pediatrics.aappublications.orgDownloaded from

DOI: 10.1542/peds.2010-1659; originally published online March 14, 2011; 2011;127;e927Pediatrics

and Chunming ChenSuying Chang, Li Wang, Yuying Wang, Inge D. Brouwer, Frans J. Kok, Betsy Lozoff

Preschool-Aged Chinese ChildrenIron-Deficiency Anemia in Infancy and Social Emotional Development in

  

ServicesUpdated Information &

tmlhttp://pediatrics.aappublications.org/content/127/4/e927.full.hincluding high resolution figures, can be found at:

References

tml#ref-list-1http://pediatrics.aappublications.org/content/127/4/e927.full.hat:This article cites 30 articles, 4 of which can be accessed free

Rs)3Peer Reviews (PPost-Publication

  http://pediatrics.aappublications.org/cgi/eletters/127/4/e927

R has been posted to this article: 3One P

Subspecialty Collections

ntal:behaviorhttp://pediatrics.aappublications.org/cgi/collection/developmeDevelopmental/Behaviorthe following collection(s):This article, along with others on similar topics, appears in

Permissions & Licensing

mlhttp://pediatrics.aappublications.org/site/misc/Permissions.xhttables) or in its entirety can be found online at: Information about reproducing this article in parts (figures,

Reprints http://pediatrics.aappublications.org/site/misc/reprints.xhtml

Information about ordering reprints can be found online:

rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Grove Village, Illinois, 60007. Copyright © 2011 by the American Academy of Pediatrics. All and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elkpublication, it has been published continuously since 1948. PEDIATRICS is owned, published, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

at N/A on February 16, 2012pediatrics.aappublications.orgDownloaded from