Poverty and Early Childhood OutcomesLeslie L Roos PhDa Elizabeth Wall-Wieler PhDb Janelle Boram Lee BAa

abstractBACKGROUND Children born into poverty face many challenges Exposure to poverty comes indifferent forms and children may also transition into or out of poverty In this study weexamine the relationships among various outcomes and different levels of poverty (householdandor neighborhood poverty) at different points during a childrsquos first 5 years

METHODS We used linkable administrative databases following 46 589 children born inManitoba Canada between 2000 and 2009 to age 7 Poverty is defined as those receivingwelfare and those living in low-income neighborhoods Four outcomes are measured in thefirst 5 years (placement in out-of-home care externalizing mental health diagnosis asthmadiagnosis and hospitalization for injury) with school readiness assessed between ages 5and 7

RESULTSChildren born into poverty had greater odds of not being ready for school than childrennot born into poverty (adjusted odds ratio = 154 159 126 for children born in householdand neighborhood poverty household poverty only and neighborhood poverty onlyrespectively all significant at P 05) Similar patterns were seen across outcomes For thoseborn into neighborhood poverty the odds of school readiness were higher only if childrenmoved before age 2

CONCLUSIONS The level of poverty (household or neighborhood) and its duration modify therelationship between early poverty and childhood outcomes Covariate adjustment generallyweakens but does not eliminate these relationships

WHATrsquoS KNOWN ON THIS SUBJECT Past researchrevealed children born into poverty are less ready forschool and have worse health and social outcomes Itis also known that moving out of poverty in childhoodleads to better long-term outcomes

WHAT THIS STUDY ADDS Household poverty isassociated with worse early childhood outcomes thanneighborhood poverty Links between school readinessand moving into or out of neighborhood poverty areonly seen if moving happened before the childrsquossecond birthday

To cite Roos LL Wall-Wieler E Lee JB Poverty and EarlyChildhood Outcomes Pediatrics 2019143(6)e20183426

aDepartment of Community Health Sciences University of Manitoba Manitoba Canada and bDepartment ofPediatrics Stanford University Stanford California

Dr Roos conceptualized and designed the study drafted the initial manuscript and had full accessto all the data in the study Dr Wall-Wieler designed the study conducted the data analyses draftedthe initial manuscript and had full access to all the data in the study Ms Lee drafted the initialmanuscript and all authors reviewed and revised the manuscript and approved the finalmanuscript as submitted and agree to be accountable for all aspects of the work

DOI httpsdoiorg101542peds2018-3426

Accepted for publication Mar 13 2019

Address correspondence to Leslie L Roos PhD Department of Community Health SciencesUniversity of Manitoba 408-727 McDermot Ave Winnipeg MB R3E 3P5 Canada E-mail leslie_rooscpeumanitobaca

PEDIATRICS (ISSN Numbers Print 0031-4005 Online 1098-4275)

Copyright copy 2019 by the American Academy of Pediatrics

FINANCIAL DISCLOSURE The authors have indicated they have no financial relationships relevant tothis article to disclose

PEDIATRICS Volume 143 number 6 June 2019e20183426 ARTICLE by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

The early life course has been seen asof particular importance childrenborn into poverty face significantchallenges12 Family difficulties arelikely to result in poor educationalsocial and health outcomes34

Poverty is associated with variousfactors leading to poor academicachievement including atypicalstructural brain development5

limited language development6 anda greater likelihood of experiencingfood insecurity78 Lack of schoolreadiness predicts later cognitiveproblems and adult psychosocialadjustment9ndash11 Higher householdfamily incomes and neighborhoodsocioeconomic status have beenlinked with greater school readinessin vocabulary communicationnumeracy knowledge andattention1213 Furthermore povertyis associated with placement ofchildren into out-of-home care14

Poverty may also lead to importantproblems including externalizingmental conditions asthma andinjuries resulting in hospitalizationChildhood attention-deficithyperactivity disorder and conductdisorder start early and predictdisruptive behavior in later childhoodand adolescence15ndash17

The relationships among levels ofpoverty timing and duration ofpoverty and various types ofbehavior have been difficult toinvestigate1819 Several Americanstudies have suggested that bothhousehold and neighborhood povertynegatively affect longer-termeducational achievement and socialmobility Poor neighborhoods canhave multigenerational effects onresidentsrsquo cognitive development20

whereas moving to a higher incomeneighborhood appears to generatebeneficial long-term effects21

Population-level information from theCanadian province of Manitoba allowstracking individualsrsquo povertyexperiences from birth to comparechildren raised under different levels

of poverty (household andneighborhood) with thoseexperiencing more advantageouscircumstances By using these datathe following questions wereexamined How are household andneighborhood-level povertyassociated with early childhoodoutcomes (school readinessexternalizing mental health asthmainjuries) differently To what extentis transitioning out of (neighborhoodandor household) povertyassociated with these outcomes Doesthe timing of that move matterIdentifying how level of poverty atbirth and the duration of poverty arelinked to a range of early childhoodoutcomes can provide insight intowho might benefit most from supportand when that support should beprovided

METHODS

Setting and Data

Manitoba is reasonablyrepresentative of Canada as a wholegenerally ranking in the midrange ofa series of indicators of health statusand health care expenditures2223 In2011 the provincial population was12 million with more than half (n =730018) living in Winnipeg Canadarsquoseighth largest metropolitan area24

Located near Canadarsquos geographiccenter Manitoba has a comparativelylarge aboriginal population (14)25

Manitobans score slightly below thenational average on standardizededucational tests administeredinternationally (although Canadiansdo somewhat better thanAmericans)26 School readinesstesting has shown sim30 ofManitobarsquos 5-year-olds to bevulnerable in at least 1 area ofdevelopment this exceeded theCanadian average of 2627

We used administrative data in thePopulation Research Data Repositoryhoused at the Manitoba Centre for

Health Policy A scrambled personalhealth number allowed linking acrossmultiple deidentified data setsinformation on linkage methodsconfidentiality and privacy andvalidity is reviewed elsewhere2829

Data from the population registry arecombined with individual-levelinformation from hospital dischargeabstracts (containing InternationalClassification of Diseases [ICD]diagnosis codes InternationalClassification of Diseases NinthRevision Clinical Modification [ICD-9-CM] codes before April 1 2004 andInternational Classification ofDiseases 10th Revision Canada [ICD-10-CA] codes after April 1 2004)physician visits (ICD-9-CM codes) theEarly Development InstrumentFamilies First screens (filled in duringroutine home visits by Public HealthNurses for most births and includeinformation on the motherrsquos socialcircumstances) children in out-of-home care and families receivingprotection services by Child andFamily Services monthly receipt ofEmployment and Income Assistancedata (basically welfare) and theCanadian Census (neighborhood-levelmedian income)

Ethics Approval

This study was approved by theUniversity of Manitoba HealthResearch Ethics Board (H2016182)and the Health Information PrivacyCommission at Manitoba HealthSeniors and Active Living (20162017-09) Using deidentifiedadministrative data files did notrequire participantsrsquo informedconsent

Poverty at Birth

Household poverty at birth is definedas a mother receiving Employmentand Income Assistance (analogousto welfare) in the month of birthSuch assistance provides help toManitobans having no other wayto support themselves or their

2 ROOS et al by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

families30 Living in a neighborhoodwith median income in the lowestquintile specifies neighborhoodpoverty Neighborhoods have beenranked from 1 (lowest income) to 5(highest income) created separatelyfor rural and urban Manitobaquintiles are based on censusdissemination areas including sim400individuals31

Poverty at birth is described byusing 4 categories (1) bothhousehold and neighborhoodpoverty (receiving welfare andresided in the lowest income quintileneighborhood) (2) just householdpoverty (receiving welfare butresided in neighborhoods havinghigher median incomes quintiles 2through 5) (3) just neighborhoodpoverty (resided in the poorestneighborhoods but did not receivewelfare) and (4) no poverty (livingin neighborhoods having highermedian incomes and not receivingwelfare)

Cohort Formation

Because performance on the EarlyDevelopment Instrument (generallyadministered biannually) is a primaryoutcome cohort selection began withall children enrolled in kindergartenthe year each Early DevelopmentInstrument was administered2005ndash2006 2006ndash2007 2008ndash20092010ndash2011 2012ndash2013 and2014ndash2015 Of these 84 598 students65 895 (779) had completed theInstrument To ensure thecompleteness of early childhoodinformation we excluded children notliving in Manitoba from birth to age 5those missing key variables andthose not completing a Families Firstscreen The final cohort consisted of46 589 children of whom 11 619(249) were born in poverty (Supplemental Fig 1)

Early Childhood Outcomes

We looked at school readiness and 4other outcomes before age 5 placedin out-of-home care externalizing

mental conditions diagnosis asthmadiagnosis and hospitalization forinjury The Early DevelopmentInstrument is used to assess eachof 5 developmental domains whena child has enrolled in kindergartenphysical health and well-beingsocial competence emotionalmaturity language and cognitivedevelopment and communicationskills and general knowledge Thisindex has acceptable interraterreliability and high internalconsistency32 A child is considerednot ready for school if scoring inthe lowest 10th percentile accordingto national norms in 1 or moredevelopmental areas3233

Children in care have been removedfrom their original families becauseauthorities have deemed their familyunable or unfit to look after themproperly Placement in care of Childand Family Services for at least 1 daybefore age 5 indicated ldquoplaced in out-of-home carerdquo Externalizing mentalconditions asthma andhospitalization for injuries are keychildhood conditions defined by usingICD codes (see SupplementalTable 6)15

Trajectories of Poverty

Trajectories into and out of differentlevels of poverty are examined beforeage 5 with changes classified asoccurring before or after age 2 Wedefined 3 levels of poverty householdpoverty neighborhood poverty (butnot household poverty) and nopoverty Four trajectories areexamined (1) born into householdpoverty and transitioned out ofpoverty (2) born into neighborhoodpoverty and moved out of poverty(3) not born into poverty and movedinto neighborhood poverty and (4)not born into poverty andtransitioned into household povertyIndividuals not falling into 1 of the 4trajectories were excludedSupplemental Figure 2 presents thesetrajectories

Covariates

Early childhood outcomes have beenassociated with both maternal andchild characteristics34 We examineda series of maternal characteristicsduring pregnancy whether shechanged residence received servicesfrom Child and Family Services useddrugs or alcohol smoked had a moodor anxiety disorder or receivedinadequate prenatal care A change in6-digit postal code defined change ofresidence The attention of Child andFamily Services during pregnancycould mean several things adolescentmothers may access expectant parentservices during pregnancy for oldermothers services could includeprotection or support intended toresolve family matters includingcounseling guidance educationand emergency shelter services3536

The Families First screen providedinformation on drug andor alcoholuse and on smoking duringpregnancy The parental careutilization index measures adequacyof care by examining childrsquosgestational age trimester of firstprenatal care and total number ofprenatal visits during pregnancy37

Several maternal variables before andat the childrsquos birth were also includededucation (did not graduate fromhigh school graduated from highschool) age at first birth (2020ndash29 30+) and neighborhoodlocation Maternal education from theFamilies First screen is missing forsim14 of mothers Location isspecified as urban (any neighborhoodin Winnipeg or Brandon) or rural(other Manitoba neighborhood) TheFamilies First screen providedinformation on social isolation andlone parent status Characteristics ofthe child at birth include thefollowing birth order (1 2 3+)sex (male or female) and whetherthe child was of low birth weight(2500 g) or preterm (27 weeks)The Supplemental Informationpresent relevant ICD codes

PEDIATRICS Volume 143 number 6 June 2019 3 by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

families30 Living in a neighborhoodwith median income in the lowestquintile specifies neighborhoodpoverty Neighborhoods have beenranked from 1 (lowest income) to 5(highest income) created separatelyfor rural and urban Manitobaquintiles are based on censusdissemination areas including sim400individuals31

Poverty at birth is described byusing 4 categories (1) bothhousehold and neighborhoodpoverty (receiving welfare andresided in the lowest income quintileneighborhood) (2) just householdpoverty (receiving welfare butresided in neighborhoods havinghigher median incomes quintiles 2through 5) (3) just neighborhoodpoverty (resided in the poorestneighborhoods but did not receivewelfare) and (4) no poverty (livingin neighborhoods having highermedian incomes and not receivingwelfare)

Cohort Formation

Because performance on the EarlyDevelopment Instrument (generallyadministered biannually) is a primaryoutcome cohort selection began withall children enrolled in kindergartenthe year each Early DevelopmentInstrument was administered2005ndash2006 2006ndash2007 2008ndash20092010ndash2011 2012ndash2013 and2014ndash2015 Of these 84 598 students65 895 (779) had completed theInstrument To ensure thecompleteness of early childhoodinformation we excluded children notliving in Manitoba from birth to age 5those missing key variables andthose not completing a Families Firstscreen The final cohort consisted of46 589 children of whom 11 619(249) were born in poverty (Supplemental Fig 1)

Early Childhood Outcomes

We looked at school readiness and 4other outcomes before age 5 placedin out-of-home care externalizing

mental conditions diagnosis asthmadiagnosis and hospitalization forinjury The Early DevelopmentInstrument is used to assess eachof 5 developmental domains whena child has enrolled in kindergartenphysical health and well-beingsocial competence emotionalmaturity language and cognitivedevelopment and communicationskills and general knowledge Thisindex has acceptable interraterreliability and high internalconsistency32 A child is considerednot ready for school if scoring inthe lowest 10th percentile accordingto national norms in 1 or moredevelopmental areas3233

Children in care have been removedfrom their original families becauseauthorities have deemed their familyunable or unfit to look after themproperly Placement in care of Childand Family Services for at least 1 daybefore age 5 indicated ldquoplaced in out-of-home carerdquo Externalizing mentalconditions asthma andhospitalization for injuries are keychildhood conditions defined by usingICD codes (see SupplementalTable 6)15

Trajectories of Poverty

Trajectories into and out of differentlevels of poverty are examined beforeage 5 with changes classified asoccurring before or after age 2 Wedefined 3 levels of poverty householdpoverty neighborhood poverty (butnot household poverty) and nopoverty Four trajectories areexamined (1) born into householdpoverty and transitioned out ofpoverty (2) born into neighborhoodpoverty and moved out of poverty(3) not born into poverty and movedinto neighborhood poverty and (4)not born into poverty andtransitioned into household povertyIndividuals not falling into 1 of the 4trajectories were excludedSupplemental Figure 2 presents thesetrajectories

Covariates

Early childhood outcomes have beenassociated with both maternal andchild characteristics34 We examineda series of maternal characteristicsduring pregnancy whether shechanged residence received servicesfrom Child and Family Services useddrugs or alcohol smoked had a moodor anxiety disorder or receivedinadequate prenatal care A change in6-digit postal code defined change ofresidence The attention of Child andFamily Services during pregnancycould mean several things adolescentmothers may access expectant parentservices during pregnancy for oldermothers services could includeprotection or support intended toresolve family matters includingcounseling guidance educationand emergency shelter services3536

The Families First screen providedinformation on drug andor alcoholuse and on smoking duringpregnancy The parental careutilization index measures adequacyof care by examining childrsquosgestational age trimester of firstprenatal care and total number ofprenatal visits during pregnancy37

Several maternal variables before andat the childrsquos birth were also includededucation (did not graduate fromhigh school graduated from highschool) age at first birth (2020ndash29 30+) and neighborhoodlocation Maternal education from theFamilies First screen is missing forsim14 of mothers Location isspecified as urban (any neighborhoodin Winnipeg or Brandon) or rural(other Manitoba neighborhood) TheFamilies First screen providedinformation on social isolation andlone parent status Characteristics ofthe child at birth include thefollowing birth order (1 2 3+)sex (male or female) and whetherthe child was of low birth weight(2500 g) or preterm (27 weeks)The Supplemental Informationpresent relevant ICD codes

PEDIATRICS Volume 143 number 6 June 2019 3 by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

Statistical Analysis

We first compared characteristics ofchildren born into poverty and thosenot born into poverty using x2 testsNext we examined the odds of eachoutcome for individuals in the 4groups of poverty (born intohousehold and neighborhood povertyborn into household poverty onlyborn into neighborhood poverty onlyand not born in poverty) usingunadjusted and adjusted logisticregression models The adjustedmodels include all maternal and childcovariates before and at the birth ofthe child

Odds of school readiness were firstcompared between children movingout of different levels of poverty andchildren remaining in poverty in theirfirst 5 years We then looked at theodds of school readiness for childrentransitioning into different levels ofpoverty and those for childrenremaining out of poverty Finally therelationships between trajectories ofpoverty and other childhoodoutcomes between birth and age 5were examined Data managementprogramming and analyses wereperformed by using SAS version 94(SAS Institute Inc Cary NC)38

RESULTS

Household and NeighborhoodPoverty at Birth

Of the 46 589 children in the cohort11 619 (249) were born in povertyOf those born in poverty 2951(254) experienced both types ofpoverty 2766 (238) onlyhousehold poverty and 5902 (508)only neighborhood poverty Table 1highlights the association of povertywith events suggesting familydifficulties The combination ofhousehold and neighborhood povertygenerated the strongest relationshipswith residential mobility receivingservices from Child and FamilyServices and having inadequateprenatal care Drug andor alcoholuse smoking low maternal

education and social isolation werehighest among mothers experiencinghousehold poverty (with or withoutneighborhood poverty)

Children born into both householdand neighborhood poverty had thehighest percentage of placement inout-of-home care (Table 2 andSupplemental Table 7) Theldquohousehold and neighborhoodpovertyrdquo and ldquohousehold povertyonlyrdquo categories differed only slightlyfor school readiness and problemswith mental and physical healthChildren experiencing ldquoneighborhoodpoverty onlyrdquo showed frequencies onthe indicated measures between theircounterparts born into householdpoverty and those not born intopoverty Children with low familyincome showed more externalizingbehavior19

Children born in poverty (householdandor neighborhood) were lesslikely to be ready for school thanthose not born poor (Table 2) Twolevels of poverty (household andneighborhood) led to the highestrates of placement in out-of-homecare (242) rates for householdpoverty (174) were considerablygreater than those for neighborhoodpoverty (31) Household povertywas associated with higher odds ofexternalizing mental conditions andasthma but neighborhood povertywas not Finally children born intoboth household and neighborhoodpoverty were more likely to behospitalized for an injury (21) thantheir more affluent counterparts(06) (Table 2)

In Table 3 we summarize the odds ofeach outcome for individuals in the 4groups of poverty (born intohousehold and neighborhood povertyborn into household poverty onlyborn into neighborhood poverty onlyand not born in poverty) usingunadjusted and adjusted logisticregression models Supplemental Tables 8 through 10 present details onschool readiness

Trajectories Into and Out of PovertyBefore Age 5

The relationship betweentransitioning into and out of differentlevels of poverty before age 5 andschool readiness builds ona simplified cohort This cohortincluded only children who either didnot change poverty level or changedpoverty once before age 5 (n =42 170) Most of those born intohousehold poverty remained in suchpoverty to age 5 (847) Only 452of children born in neighborhoodpoverty remained there to this ageover 900 of children not born inpoverty stayed out of poverty to age5 Supplemental Tables 11 and 12provide detailed information

For children born in householdpoverty transitions out of povertywere associated with a lowerprobability of not being readyfor school (adjusted odds ratio[aOR] = 058 for transitions beforeage 2 aOR = 075 for those afterage 2) (Table 4) School readinesswas significantly higher only ifchildren left neighborhood povertybefore age 2 (aOR = 073) Amongchildren born outside of povertymoving into neighborhood povertybefore age 2 was associated witha greater likelihood of not being readyfor school (aOR = 130) moving intohousehold poverty either before orafter age 2 was associated with notbeing ready (aORs = 171 and 168respectively) Overall householdpoverty was linked with worseoutcomes changes before age 2reveal the largest differencesSupplemental Tables 13 and 14provide additional statistics

Table 5 presents relationshipsbetween type of poverty andoutcomes before age 5 Placement inout-of-home care revealed dramaticdifferences (from 239 to 02across these categories) Placementrates were highest among childrenremaining in household poverty andthose living in such poverty before

4 ROOS et al by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

age 2 Asthma diagnosesranged markedly (from 305 to181) Externalizing mental healthconditions varied substantially butrather irregularly with injuryhospitalization rates low among allgroups

DISCUSSION

We have considered household andneighborhood poverty from several

perspectives Examining thelength of exposure among bothchildren transitioning out ofpoverty and those transitioning intopoverty goes beyond most studiesChildren growing up in poverty mustdeal with a home environment lesssupportive of school readiness andoverall health than those leavingpoverty early in life Moreovermothers transitioning out of povertywhen the child was 2 showed fewer

risk factors than those leavingwhen the child was older Thenumber of years of childhoodpoverty and the lack of schoolreadiness are clearly linked Theassociation of different types ofpoverty with various outcomes(school readiness placement in out-of-home care and indicators ofphysical and mental health) hasemphasized the importance ofhousehold poverty

TABLE 1 Children Born Into Different Levels of Poverty (n = 46 589)

Covariates Group 1A Householdand NeighborhoodPoverty at Birth

(n = 2951)

Group 1B HouseholdPoverty Only at Birth

(n = 2766)

Group 1CNeighborhoodPoverty Only(n = 5902)

Group 2 Not Born inPoverty

(n = 34 970)

n () n () n () n ()

Maternal covariates during pregnancyMoved 940 (319)a 700 (253)a 1155 (196)a 5605 (160)Received services from CFS 2350 (796)a 1962 (709)a 1364 (231)a 4457 (128)Drug or alcohol use 736 (249)a 693 (251)a 633 (107)a 2932 (84)Smoking 1513 (513)a 1403 (507)a 1143 (194)a 4192 (120)Mood or anxiety disorder 469 (159)a 517 (187)a 715 (121) 4046 (116)Inadequate prenatal care 921 (312)a 705 (255)a 989 (168)a 4525 (129)

Maternal covariates before and at birth of childLess than grade 12 education 1603 (543)a 1369 (495)a 1071 (182)a 3240 (93)Social isolation 671 (227)a 623 (225)a 483 (82)a 1374 (39)Lone parent 1566 (531)a 1515 (548)a 606 (103)a 1511 (43)Age of mother at first birth20 1890 (641)a 1696 (613)a 1361 (231)a 4072 (116)20ndash29 995 (337)a 1004 (363)a 3666 (621) 21 932 (627)$30 66 (22)a 66 (24)a 875 (148)a 8966 (256)

Urban neighborhood at birth of child 2601 (881)a 1642 (594) 3918 (664)a 20 659 (591)Child covariates at birthBirth order1 814 (276)a 906 (328)a 2669 (452)a 14 746 (422)2 841 (285)a 753 (272)a 1862 (316)a 12 909 (369)3+ 1296 (439)a 1107 (400)a 1371 (232)a 7315 (209)

Male 1475 (500) 1394 (504) 3033 (514) 17 714 (507)Low birth wt (2500 g) 164 (56)a 166 (60)a 308 (52)a 1597 (46)Preterm (37 wk) 244 (83)a 237 (86)a 445 (75)a 2326 (67)

CFS Child and Family Servicesa Significantly (at P 05) different from Group 2

TABLE 2 Early Childhood Outcomes by Different Levels of Poverty at Birth (n = 46 589)

Outcomes Group 1A Householdand NeighborhoodPoverty at Birth

(n = 2951)

Group 1B HouseholdPoverty Only at Birth

(n = 2766)

Group 1CNeighborhoodPoverty Only(n = 5902)

Group 2 Not BornInto Poverty(n = 34 970)

n () n () n () n ()

Not ready for school (EDI) on 1 or more domain 1499 (508)a 1348 (487)a 1754 (297)a 7740 (221)Before age 5Placed in out-of-home care 713 (242)a 482 (174)a 181 (31)a 269 (08)Diagnosed with externalizing mental condition 233 (79)a 204 (74)a 237 (40)a 1222 (35)Diagnosed with asthma 934 (317)a 799 (289)a 1239 (210)a 6797 (194)Hospitalized for an injury 62 (21)a 44 (16)a 56 (10)a 220 (06)

EDI early development instrumenta Significantly (at P 05) different from Group 2

PEDIATRICS Volume 143 number 6 June 2019 5 by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

TABLE3Odds

RatiosforSchool

ReadinessforChildrenBorn

Into

Povertyby

Trajectories

ofType

ofPoverty

School

Readiness(EDI)

Trajectory

1HouseholdPovertyto

NoPoverty

Trajectory

2Neighborhood

Povertyto

NoPoverty

Trajectory

3No

Povertyto

Neighborhood

Poverty

Trajectory

4No

Povertyto

Household

Poverty

Unadjusted

OR(95

CI)

aORa

(95

CI)

Unadjusted

OR(95

CI)

aORa

(95

CI)

Unadjusted

OR(95

CI)

aORa

(95

CI)

Unadjusted

OR(95

CI)

aORa

(95

CI)

Notreadyon

physical

well-being

domain

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

043

(032ndash058)d

051

(037ndash069)d

059

(048ndash072)d

077

(062ndash095)d

160

(131ndash194)d

140

(114ndash172)d

371

(301ndash458)d

172

(136ndash218)d

Afterage2e

062

(048ndash081)d

069

(053ndash090)d

073

(059ndash091)d

089

(070ndash111)

107

(085ndash133)

091

(072ndash114)

389

(310ndash488)d

206

(161ndash263)d

Notreadyon

social

competencedomain

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

063

(047ndash084)d

071

(053ndash096)d

057

(046ndash070)d

070

(056ndash087)d

138

(112ndash171)d

126

(102ndash157)d

303

(243ndash379)d

157

(123ndash202)d

Afterage2e

086

(066ndash111)

092

(070ndash121)

079

(064ndash098)d

092

(073ndash115)

107

(086ndash135)

094

(075ndash119)

234

(179ndash305)d

132

(099ndash175)

Notreadyon

communicationandgeneral

know

ledgedomain

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

048

(035ndash066)d

053

(038ndash074)d

058

(047ndash071)d

073

(059ndash090)d

161

(132ndash198)d

140

(113ndash172)d

257

(203ndash327)d

154

(118ndash201)d

Afterage2e

053

(039ndash072)d

056

(041ndash076)d

077

(062ndash096)d

089

(071ndash111)

111

(089ndash140)

096

(076ndash121)

216

(163ndash284)d

139

(104ndash188)d

Notreadyon

emotionalmaturity

domain

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

065

(048ndash088)d

073

(053ndash101)

062

(050ndash076)d

074

(060ndash093)d

108

(087ndash134)

097

(078ndash120)

240

(191ndash301)d

137

(107ndash177)d

Afterage2e

115

(089ndash149)

123

(094ndash161)

076

(061ndash095)d

086

(068ndash109)

108

(088ndash134)

100

(080ndash125)

184

(140ndash242)d

114

(085ndash152)

Notready

onlanguage

andcognitive

developm

ent

domain

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

048

(036ndash064)d

054

(040ndash073)d

059

(048ndash072)d

076

(061ndash094)d

143

(115ndash178)d

124

(099ndash155)

357

(287ndash445)d

184

(144ndash235)d

Afterage2e

061

(047ndash080)d

066

(050ndash086)d

072

(058ndash090)d

086

(068ndash108)

123

(099ndash154)

107

(095ndash134)

321

(251ndash412)d

181

(139ndash237)d

Placed

inout-of-hom

ecare0ndash5

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

018

(012ndash029)d

035

(022ndash059)d

012

(006ndash024)d

030

(014ndash065)d

336

(133ndash849)d

152

(057ndash403)

7501(4970ndash11321)d

914

(567ndash1473)

d

Afterage2e

044

(032ndash061)d

067

(047ndash094)d

020

(010ndash039)d

039

(018ndash083)d

254

(091ndash706)

139

(049ndash398)

4419(2699ndash7236)

d637

(373ndash1086)

d

Externalizingmentalcondition

diagnosis0ndash5

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

067

(042ndash109)

078

(048ndash128)

117

(085ndash160)

135

(096ndash189)

092

(064ndash133)

095

(066ndash139)

155

(103ndash232)d

121

(072ndash174)

6 ROOS et al by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

Adjustment for other factors(themselves linked to poverty)weakens the relationships betweenpoverty and the various outcomesHousehold poverty remaineda statistically significant predictorafter controlling for numerouscovariates (although odds ratios weretypically reduced) After adjustmentsneighborhood poverty was generallyless predictive of childhood mentalhealth asthma and hospitalizationfor injuries

Our information provides a baselinefor additional research For exampleeducational performance (grade 3)and our childhood measures can helpassess changes by age 10 Moregenerally having relevant outcomesat several developmental stages aidsin judging the significance of early lifeconditions and events Not onlypoverty but such ldquoshocksrdquo as prenataland early life maternal stress mightwell affect adolescent and adultoutcomes39 Large databases canoperationalize such characteristics associal isolation death of a closerelative domestic violence andmaternal drug andor alcohol useOur findings also supportBronfenbrennerrsquos40 ecologicalsystems theory postulating mother-child interactions to be strongpredictors of child developmentconstructive mother-child bonds aremore difficult for mothers havingdrug or alcohol problemsOvercoming such factors among thedisadvantaged may be more difficultthan among the more affluent andinterventions to increase humancapital might prove more valuableamong the poor Information onseveral such programs (nurse visitingmaternal income supplements) arebeing incorporated into Manitobaanalyses34

This articlersquos limitations include thoseassociated with observational studiesReceipt of welfare was used to definehousehold poverty because access toinformation on household income islacking Our definition mayTA

BLE3

Continued

School

Readiness(EDI)

Trajectory

1HouseholdPovertyto

NoPoverty

Trajectory

2Neighborhood

Povertyto

NoPoverty

Trajectory

3No

Povertyto

Neighborhood

Poverty

Trajectory

4No

Povertyto

Household

Poverty

Unadjusted

OR(95

CI)

aORa

(95

CI)

Unadjusted

OR(95

CI)

aORa

(95

CI)

Unadjusted

OR(95

CI)

aORa

(95

CI)

Unadjusted

OR(95

CI)

aORa

(95

CI)

Afterage2e

134

(093ndash192)

139

(096ndash201)

067

(044ndash103)

074

(048ndash115)

087

(060ndash125)

086

(059ndash125)

141

(089ndash224)

110

(068ndash178)

Asthmadiagnosis0ndash5

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

076

(059ndash098)d

091

(070ndash118)

081

(069ndash094)d

084

(071ndash099)d

099

(084ndash117)

102

(087ndash120)

130

(105ndash160)d

108

(087ndash136)

Afterage2e

101

(080ndash128)

107

(084ndash136)

082

(069ndash098)d

084

(070ndash101)

094

(080ndash111)

095

(081ndash112)

128

(101ndash161)d

113

(089ndash144)

Injury

hospitalization

0ndash5

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

041

(013ndash131)

069

(021ndash224)

052

(026ndash105)

054

(026ndash111)

037

(009ndash149)

033

(008134)

183

(075ndash448)

106

(040ndash276)

Afterage2e

053

(019ndash146)

071

(026ndash197)

035

(014ndash090)d

035

(014ndash093)d

193

(105ndash357)d

163

(088ndash304)

132

(042ndash416)

083

(025ndash269)

CFSChild

andFamily

ServicesC

IconfidenceintervalEDIearlydevelopm

entinstrumentORoddsratio

aAdjusted

formaternalcovariatesduring

pregnancy(m

ovedreceivedservices

from

CFSmoodandanxietydisordersdrug

andor

alcoholusesmokingprenatalcare

use)m

aternalcovariatesat

theindexdate(educationsocialisolationlone

parentagen

eighborhoodlocation)and

child

covariates

atbirth(birth

ordersexbirthwtgestationalage)

bNo

change

inpovertylevelandor

type

before

age5

cChange

inpovertylevelandor

type

betweenbirthandage2

dP

05

eChange

inpovertylevelandor

type

betweenage2andage5

PEDIATRICS Volume 143 number 6 June 2019 7 by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

underestimate such poverty Canadahas no official poverty line howeverhouseholds having incomes less thanhalf of the median household income(for a family of a given size) areconsidered low income41 ManitobaEmployment and Income Assistance(and other benefits) generally payless than this amount Manitobarsquosaverage total income of householdswas below the Canadian nationalaverage in 2015 (Manitoba $85373Canada $92764)42 The CanadianIncome Survey data suggested that

Manitobarsquos median employmentincome was consistently lower thanthe Canadian national median from2012 to 201543 In 2015 Manitobarsquosprevalence of low income based onthe low-income cutoff after tax was99 which was slightly higher thanthe national prevalence (92)42

Additionally although children areclustered within neighborhoods wedid not account for this in ouranalysis (ie use a multilevel model)This was done because in many caseschildren lived in 1 neighborhood

between birth and age 5 with theamount of time spent in eachneighborhood varying by childchildren would often belong inseveral clusters over time

Because the poverty variable couldnot be manipulated in experimentalor quasi-experimental fashioncausality cannot be ascertainedSelective movement out of poverty byless challenged families (or intopoverty by their more challengedcounterparts) might be responsible

TABLE 4 Odds Ratios for Not Being Ready For School Trajectories by Type of Poverty

Not Ready on 1 or More Domain (EDI) Unadjusted OR (95 CI) aORa (95 CI)

Trajectory 1 born in household poverty (n = 4610)Stayed in household poverty to age 5 (n = 3904) Reference ReferenceTransitioned out of poverty before age 2 (n = 347) 049 (039ndash061)b 058 (046ndash075)b

Transitioned out of poverty after age 2 (n = 359) 067 (054ndash083)b 075 (060ndash094)b

Trajectory 2 born in neighborhood poverty (n = 5070)Stayed in neighborhood poverty to age 5 (n = 2290) Reference ReferenceMoved out of poverty before age 2 (n = 1656) 059 (051ndash068)b 073 (063ndash086)b

Moved out of poverty after age 2 (n = 1124) 080 (068ndash094)b 094 (080ndash111)Trajectory 3 not born in poverty (n = 31 581)Stayed out of poverty to age 5 (n = 29 561) Reference ReferenceMoved into neighborhood poverty before age 2 (n = 981) 145 (125ndash167)b 130 (112ndash151)b

Moved into neighborhood poverty after age 2 (n = 1039) 117 (101ndash136)b 102 (088ndash119)Trajectory 4 not born in poverty (n = 30 470)Stayed out of poverty to age 5 (n = 29 561) Reference ReferenceTransitioned into household poverty before age 2 (n = 497) 327 (274ndash391)b 171 (140ndash209)b

Transitioned into household poverty after age 2 (n = 412) 289 (237ndash352)b 168 (135ndash208)b

CFS Child and Family Services CI confidence interval EDI early development instrument OR odds ratioa Adjusted for maternal covariates during pregnancy (moved received services from CFS mood and anxiety disorders drug andor alcohol use smoking prenatal care use) maternalcovariates at the index date (education social isolation lone parent age neighborhood location) and child covariates at birth (birth order sex birth wt gestational age) These variablesare used for adjustment in all tablesb P 05

TABLE 5 Proportion of Population With Childhood Outcomes Before Age 5 by Trajectories and Type of Poverty

Placed in Out-of-Home Care

Externalizing MentalCondition Diagnosis

AsthmaDiagnosis

InjuryHospitalization

Trajectory 1 born in household poverty (n = 4610)Stayed in household poverty to age 5 (n = 3904) 239 79 305 21Transitioned out of poverty before age 2 (n = 347) 55 55 251 09Transitioned out of poverty after age 2 (n = 359) 123 103 306 11

Trajectory 2 born in neighborhood poverty (n = 5070)Stayed in neighborhood poverty to age 5 (n = 2290) 40 38 223 13Moved out of poverty before age 2 (n = 1656) 05 44 188 07Moved out of poverty after age 2 (n = 1124) 08 26 193 04

Trajectory 3 not born in poverty (n = 31 581)Stayed out of poverty to age 5 (n = 29 561) 02 33 190 06Moved into neighborhood poverty before age 2 (n = 981) 05 31 189 02Moved into neighborhood poverty after age 2 (n = 1039) 04 29 181 11

Trajectory 4 not born in poverty (n = 30 470)Stayed out of poverty to age 5 (n = 29 561) 02 33 190 06Transitioned into household poverty before age 2 (n = 497) 103 50 233 10Transitioned into household poverty after age 2 (n = 412) 63 46 231 07

8 ROOS et al by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

for the findings Moreover the rangeof our income measures is limitedTransition out of poverty (whetherhousehold or neighborhood) isunlikely to be associated withsubstantial income changesMovement out of the lowest incomequintile (defining neighborhoodpoverty) is primarily to quintile 2(437) and quintile 3 (234)neighborhoods

Finally the absolute andrelative importance of povertyin Canada compared with in theUnited States is intriguingNeighborhood variation and therole of neighborhood characteristicsappear greater in the UnitedStates44 Canadarsquos safety net is moreextensive than those in the UnitedStates and the United Kingdom4546

Seven-country comparisons havenoted steep American socioeconomic

status gradients along severaldimensions of well-being (includingcognitive and socioemotional)47

More detailed study of thesegradients might draw on ouranalyses

CONCLUSIONS

Children growing up in povertymust deal with more riskfactors for poor outcomes thanthose never experiencing povertyor experiencing poverty fora short time Children born intohousehold poverty have muchworse early outcomes than thoseborn into neighborhood povertySupport for children whose familiesare receiving welfare to transitionout of poverty when the childis young could yield the greatestbenefit

ACKNOWLEDGMENTS

In recognition of the high proportionof the children in care being of FirstNations descent the larger programof research using Child and FamilyServices data has been presented toseveral general and First Nationsaudiences

ABBREVIATIONS

aOR adjusted odds ratioICD International Classification of

DiseasesICD-9-CM International

Classification ofDiseases Ninth RevisionClinical Modification

ICD-10-CA InternationalClassification ofDiseases 10th RevisionCanada

FUNDING This work was supported by a Graduate Enhancement of Tri-Council Stipend from the University of Manitoba Data used in this study are from the

Population Health Research Data Repository housed at the Manitoba Centre for Health Policy University of Manitoba and were derived from data provided by

Manitoba Health Seniors and Active Living Manitoba Families and Healthy Child Manitoba under project 20132014-04 The results and conclusions are those of the

authors and no official endorsement by the Manitoba Centre for Health Policy Manitoba Health Seniors and Active Living or other data providers is intended or

should be inferred

POTENTIAL CONFLICT OF INTEREST The authors have indicated they have no potential conflicts of interest to disclose

COMPANION PAPER A companion to this article can be found online at wwwpediatricsorgcgidoi101542peds2019-0195

REFERENCES

1 Power C Kuh D Morton S Fromdevelopmental origins of adult diseaseto life course research on adult diseaseand aging insights from birth cohortstudies Annu Rev Public Health 2013347ndash28

2 Wood D Effect of child and familypoverty on child health in the UnitedStates Pediatrics 2003112(3 pt 2)707ndash711

3 Aber JL Bennett NG Conley DC Li J Theeffects of poverty on child health anddevelopment Annu Rev Public Health199718463ndash483

4 Brooks-Gunn J Duncan GJ The effectsof poverty on children Future Child19977(2)55ndash71

5 Hair NL Hanson JL Wolfe BL Pollak SDAssociation of child poverty braindevelopment and academicachievement [published correctionappears in JAMA Pediatr 2015169(9)878] JAMA Pediatr 2015169(9)822ndash829

6 Hart B Risley TR MeaningfulDifferences in the Everyday Experienceof Young American Children BaltimoreMD Paul H Brookes Publishing 1995

7 Cook JT Frank DA Levenson SM et alChild food insecurity increases risksposed by household food insecurity toyoung childrenrsquos health J Nutr 2006136(4)1073ndash1076

8 Casey PH Szeto KL Robbins JM et alChild health-related quality of life and

household food security Arch PediatrAdolesc Med 2005159(1)51ndash56

9 Romano E Babchishin L Pagani LSKohen D School readiness and laterachievement replication and extensionusing a nationwide Canadian surveyDev Psychol 201046(5)995ndash1007

10 Campbell FA Wasik BH Pungello E et alYoung adult outcomes of theAbecedarian and CARE early childhoodeducational interventions Early ChildRes Q 200823(4)452ndash466

11 Fothergill KE Ensminger ME Green KMCrum RM Robertson J Juon HS Theimpact of early school behavior andeducational achievement on adult druguse disorders a prospective study

PEDIATRICS Volume 143 number 6 June 2019 9 by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

Drug Alcohol Depend 200892(1ndash3)191ndash199

12 Bradley RH Corwyn RF Socioeconomicstatus and child development Annu RevPsychol 200253371ndash399

13 Browne DT Wade M Prime H JenkinsJM School readiness amongst urbanCanadian families risk profiles andfamily mediation J Educ Psychol 2018110(1)133ndash146

14 Putnam-Hornstein E Needell BPredictors of child protective servicecontact between birth and age five anexamination of Californiarsquos 2002 birthcohort Child Youth Serv Rev 201133(11)2400ndash2407

15 Currie J Stabile M Manivong P Roos LLChild health and young adult outcomesJ Hum Resour 201045(3)517ndash548

16 Duncan GJ Dowsett CJ Claessens Aet al School readiness and laterachievement Dev Psychol 200743(6)1428ndash1446

17 Shaw DS Shelleby EC Early-startingconduct problems intersection ofconduct problems and poverty AnnuRev Clin Psychol 201410503ndash528

18 Duncan GJ Magnuson K Votruba-DrzalE Moving beyond correlations inassessing the consequences of povertyAnnu Rev Psychol 201768413ndash434

19 McEwen CA McEwen BS Socialstructure adversity toxic stress andintergenerational poverty an earlychildhood model Annu Rev Sociol 201743(1)445ndash472

20 Sharkey P Elwert F The legacy ofdisadvantage multigenerationalneighborhood effects on cognitiveability AJS 2011116(6)1934ndash1981

21 Chetty R Hendren N Katz LF The effectsof exposure to better neighborhoods onchildren new evidence from the movingto opportunity experiment Am EconRev 2016106(4)855ndash902

22 Oreopoulos P Stabile M Walld R RoosLL Short- medium- and long-termconsequences of poor infant health ananalysis using siblings and twinsJ Hum Resour 200843(1)88ndash138

23 OrsquoGrady K Deussing M Scerbina T FungK Muhe N Measuring Up CanadianResults of the OECD PISA Study TorontoCanada Council of Ministers ofEducation Canada 2016

24 Statistics Canada Focus on GeographySeries 2011 Ottawa Canada StatisticsCanada 2014 Available at httpswww12statcangccacensus-recensement2011as-safogs-spgIndex-engcfm Accessed September 252018

25 Statistics Canada Aboriginal PeoplesFact Sheet for Manitoba OttawaCanada Statistics Canada 2016Available at httpswww150statcangccan1pub89-656-x89-656-x2016008-enghtm Accessed September 25 2018

26 OrsquoGrady K Fung K Servage L Khan GReport on the Pan-CanadianAssessment of Reading Mathematicsand Science Toronto Canada Councilof Ministers of Education Canada 2018

27 Canadian Institute for HealthInformation Children Vulnerable inAreas of Early Development ADeterminant of Child Health OttawaCanada Canadian Institute for HealthInformation 2014 Available at httpssecurecihicafree_productsChildren_Vulnerable_in_Areas_of_Early_Development_ENpdf Accessed September 252018

28 Roos LL Gupta S Soodeen RA JebamaniL Data quality in an information-richenvironment Canada as an exampleCan J Aging 200524(suppl 1)153ndash170

29 Roos LL Nicol JP A research registryuses development and accuracy J ClinEpidemiol 199952(1)39ndash47

30 Government of Manitoba Employmentand Income Assistance (EIA) Availableat httpswwwgovmbcafseiaAccessed September 25 2018

31 Manitoba Centre for Health PolicyConcept Income Quintile RankingProcedure Winnipeg Canada ManitobaCentre for Health Policy 2002 httpmchp-appservcpeumanitobacaviewConceptphpconceptID=1164Accessed September 25 2018

32 Janus M Brinkman S Duku E et al TheEarly Development Instrument APopulation-Based Measure forCommunities A Handbook onDevelopment Properties and UseHamilton Canada Offord Centre forChild Studies 2007

33 Snow KL Measuring school readinessconceptual and practical

considerations Early Educ Dev 200617(1)7ndash41

34 Brownell M Nickel NC Chartier M et alAn unconditional prenatal incomesupplement reduces populationinequities in birth outcomes Health Aff(Millwood) 201837(3)447ndash455

35 Government of Manitoba Familyservices and housing Child ProtectionServices Available at httpswwwgovmbcafschildfamindexhtmlAccessed March 27 2019

36 Manitoba Centre for Health PolicyTerm apprehended at birth taken intoCFS care at birth Available at httpmchp-appservcpeumanitobacaviewDefinitionphpdefinitionID=104764Accessed January 10 2019

37 Alexander GR Kotelchuck M Assessingthe role and effectiveness of prenatalcare history challenges and directionsfor future research Public Health Rep2001116(4)306ndash316

38 SAS Institute Inc Statistical AnalysisSoftware (SAS) version 94 2013

39 Almond D Currie J Duque V Childhoodcircumstances and adult outcomes ActII Journal of Economic Literature 201856(4)1360ndash1446

40 Bronfenbrenner U The Ecology ofHuman Development Experiments byNature and Design Cambridge MAHarvard University Press 1979

41 Government of Canada A backgrounderon poverty in Canada 2016 Available athttpwww12edscgccasgpe-pmpsservletsgpp-pmps-publang=engampcurjsp=p5bd2t13ls-engjspampcuractn=dwnldamppid=54405ampdid=4879Accessed January 10 2019

42 Statistics Canada Census Profile 2016Census Ottawa Canada StatisticsCanada 2017 Available at httpswww12statcangccacensus-recensement2016dp-pdprofdetailspagecfmLang=EampGeo1=PRampCode1=46ampGeo2=PRampCode2=01ampData=CountampSearchText=ManitobaampSearchType=BeginsampSearchPR=01ampB1=IncomeampTABID=1 Accessed March 27 2019

43 Statistics Canada Canadian IncomeSurvey 2015 Ottawa Canada StatisticsCanada 2017 Available at httpswww150statcangccan1endaily-quotidien170526dq170526a-engpdfst=kFoEj4AM Accessed March 27 2019

10 ROOS et al by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

TABLE3Odds

RatiosforSchool

ReadinessforChildrenBorn

Into

Povertyby

Trajectories

ofType

ofPoverty

School

Readiness(EDI)

Trajectory

1HouseholdPovertyto

NoPoverty

Trajectory

2Neighborhood

Povertyto

NoPoverty

Trajectory

3No

Povertyto

Neighborhood

Poverty

Trajectory

4No

Povertyto

Household

Poverty

Unadjusted

OR(95

CI)

aORa

(95

CI)

Unadjusted

OR(95

CI)

aORa

(95

CI)

Unadjusted

OR(95

CI)

aORa

(95

CI)

Unadjusted

OR(95

CI)

aORa

(95

CI)

Notreadyon

physical

well-being

domain

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

043

(032ndash058)d

051

(037ndash069)d

059

(048ndash072)d

077

(062ndash095)d

160

(131ndash194)d

140

(114ndash172)d

371

(301ndash458)d

172

(136ndash218)d

Afterage2e

062

(048ndash081)d

069

(053ndash090)d

073

(059ndash091)d

089

(070ndash111)

107

(085ndash133)

091

(072ndash114)

389

(310ndash488)d

206

(161ndash263)d

Notreadyon

social

competencedomain

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

063

(047ndash084)d

071

(053ndash096)d

057

(046ndash070)d

070

(056ndash087)d

138

(112ndash171)d

126

(102ndash157)d

303

(243ndash379)d

157

(123ndash202)d

Afterage2e

086

(066ndash111)

092

(070ndash121)

079

(064ndash098)d

092

(073ndash115)

107

(086ndash135)

094

(075ndash119)

234

(179ndash305)d

132

(099ndash175)

Notreadyon

communicationandgeneral

know

ledgedomain

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

048

(035ndash066)d

053

(038ndash074)d

058

(047ndash071)d

073

(059ndash090)d

161

(132ndash198)d

140

(113ndash172)d

257

(203ndash327)d

154

(118ndash201)d

Afterage2e

053

(039ndash072)d

056

(041ndash076)d

077

(062ndash096)d

089

(071ndash111)

111

(089ndash140)

096

(076ndash121)

216

(163ndash284)d

139

(104ndash188)d

Notreadyon

emotionalmaturity

domain

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

065

(048ndash088)d

073

(053ndash101)

062

(050ndash076)d

074

(060ndash093)d

108

(087ndash134)

097

(078ndash120)

240

(191ndash301)d

137

(107ndash177)d

Afterage2e

115

(089ndash149)

123

(094ndash161)

076

(061ndash095)d

086

(068ndash109)

108

(088ndash134)

100

(080ndash125)

184

(140ndash242)d

114

(085ndash152)

Notready

onlanguage

andcognitive

developm

ent

domain

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

048

(036ndash064)d

054

(040ndash073)d

059

(048ndash072)d

076

(061ndash094)d

143

(115ndash178)d

124

(099ndash155)

357

(287ndash445)d

184

(144ndash235)d

Afterage2e

061

(047ndash080)d

066

(050ndash086)d

072

(058ndash090)d

086

(068ndash108)

123

(099ndash154)

107

(095ndash134)

321

(251ndash412)d

181

(139ndash237)d

Placed

inout-of-hom

ecare0ndash5

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

018

(012ndash029)d

035

(022ndash059)d

012

(006ndash024)d

030

(014ndash065)d

336

(133ndash849)d

152

(057ndash403)

7501(4970ndash11321)d

914

(567ndash1473)

d

Afterage2e

044

(032ndash061)d

067

(047ndash094)d

020

(010ndash039)d

039

(018ndash083)d

254

(091ndash706)

139

(049ndash398)

4419(2699ndash7236)

d637

(373ndash1086)

d

Externalizingmentalcondition

diagnosis0ndash5

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

067

(042ndash109)

078

(048ndash128)

117

(085ndash160)

135

(096ndash189)

092

(064ndash133)

095

(066ndash139)

155

(103ndash232)d

121

(072ndash174)

6 ROOS et al by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

Adjustment for other factors(themselves linked to poverty)weakens the relationships betweenpoverty and the various outcomesHousehold poverty remaineda statistically significant predictorafter controlling for numerouscovariates (although odds ratios weretypically reduced) After adjustmentsneighborhood poverty was generallyless predictive of childhood mentalhealth asthma and hospitalizationfor injuries

Our information provides a baselinefor additional research For exampleeducational performance (grade 3)and our childhood measures can helpassess changes by age 10 Moregenerally having relevant outcomesat several developmental stages aidsin judging the significance of early lifeconditions and events Not onlypoverty but such ldquoshocksrdquo as prenataland early life maternal stress mightwell affect adolescent and adultoutcomes39 Large databases canoperationalize such characteristics associal isolation death of a closerelative domestic violence andmaternal drug andor alcohol useOur findings also supportBronfenbrennerrsquos40 ecologicalsystems theory postulating mother-child interactions to be strongpredictors of child developmentconstructive mother-child bonds aremore difficult for mothers havingdrug or alcohol problemsOvercoming such factors among thedisadvantaged may be more difficultthan among the more affluent andinterventions to increase humancapital might prove more valuableamong the poor Information onseveral such programs (nurse visitingmaternal income supplements) arebeing incorporated into Manitobaanalyses34

This articlersquos limitations include thoseassociated with observational studiesReceipt of welfare was used to definehousehold poverty because access toinformation on household income islacking Our definition mayTA

BLE3

Continued

School

Readiness(EDI)

Trajectory

1HouseholdPovertyto

NoPoverty

Trajectory

2Neighborhood

Povertyto

NoPoverty

Trajectory

3No

Povertyto

Neighborhood

Poverty

Trajectory

4No

Povertyto

Household

Poverty

Unadjusted

OR(95

CI)

aORa

(95

CI)

Unadjusted

OR(95

CI)

aORa

(95

CI)

Unadjusted

OR(95

CI)

aORa

(95

CI)

Unadjusted

OR(95

CI)

aORa

(95

CI)

Afterage2e

134

(093ndash192)

139

(096ndash201)

067

(044ndash103)

074

(048ndash115)

087

(060ndash125)

086

(059ndash125)

141

(089ndash224)

110

(068ndash178)

Asthmadiagnosis0ndash5

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

076

(059ndash098)d

091

(070ndash118)

081

(069ndash094)d

084

(071ndash099)d

099

(084ndash117)

102

(087ndash120)

130

(105ndash160)d

108

(087ndash136)

Afterage2e

101

(080ndash128)

107

(084ndash136)

082

(069ndash098)d

084

(070ndash101)

094

(080ndash111)

095

(081ndash112)

128

(101ndash161)d

113

(089ndash144)

Injury

hospitalization

0ndash5

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

041

(013ndash131)

069

(021ndash224)

052

(026ndash105)

054

(026ndash111)

037

(009ndash149)

033

(008134)

183

(075ndash448)

106

(040ndash276)

Afterage2e

053

(019ndash146)

071

(026ndash197)

035

(014ndash090)d

035

(014ndash093)d

193

(105ndash357)d

163

(088ndash304)

132

(042ndash416)

083

(025ndash269)

CFSChild

andFamily

ServicesC

IconfidenceintervalEDIearlydevelopm

entinstrumentORoddsratio

aAdjusted

formaternalcovariatesduring

pregnancy(m

ovedreceivedservices

from

CFSmoodandanxietydisordersdrug

andor

alcoholusesmokingprenatalcare

use)m

aternalcovariatesat

theindexdate(educationsocialisolationlone

parentagen

eighborhoodlocation)and

child

covariates

atbirth(birth

ordersexbirthwtgestationalage)

bNo

change

inpovertylevelandor

type

before

age5

cChange

inpovertylevelandor

type

betweenbirthandage2

dP

05

eChange

inpovertylevelandor

type

betweenage2andage5

PEDIATRICS Volume 143 number 6 June 2019 7 by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

underestimate such poverty Canadahas no official poverty line howeverhouseholds having incomes less thanhalf of the median household income(for a family of a given size) areconsidered low income41 ManitobaEmployment and Income Assistance(and other benefits) generally payless than this amount Manitobarsquosaverage total income of householdswas below the Canadian nationalaverage in 2015 (Manitoba $85373Canada $92764)42 The CanadianIncome Survey data suggested that

Manitobarsquos median employmentincome was consistently lower thanthe Canadian national median from2012 to 201543 In 2015 Manitobarsquosprevalence of low income based onthe low-income cutoff after tax was99 which was slightly higher thanthe national prevalence (92)42

Additionally although children areclustered within neighborhoods wedid not account for this in ouranalysis (ie use a multilevel model)This was done because in many caseschildren lived in 1 neighborhood

between birth and age 5 with theamount of time spent in eachneighborhood varying by childchildren would often belong inseveral clusters over time

Because the poverty variable couldnot be manipulated in experimentalor quasi-experimental fashioncausality cannot be ascertainedSelective movement out of poverty byless challenged families (or intopoverty by their more challengedcounterparts) might be responsible

TABLE 4 Odds Ratios for Not Being Ready For School Trajectories by Type of Poverty

Not Ready on 1 or More Domain (EDI) Unadjusted OR (95 CI) aORa (95 CI)

Trajectory 1 born in household poverty (n = 4610)Stayed in household poverty to age 5 (n = 3904) Reference ReferenceTransitioned out of poverty before age 2 (n = 347) 049 (039ndash061)b 058 (046ndash075)b

Transitioned out of poverty after age 2 (n = 359) 067 (054ndash083)b 075 (060ndash094)b

Trajectory 2 born in neighborhood poverty (n = 5070)Stayed in neighborhood poverty to age 5 (n = 2290) Reference ReferenceMoved out of poverty before age 2 (n = 1656) 059 (051ndash068)b 073 (063ndash086)b

Moved out of poverty after age 2 (n = 1124) 080 (068ndash094)b 094 (080ndash111)Trajectory 3 not born in poverty (n = 31 581)Stayed out of poverty to age 5 (n = 29 561) Reference ReferenceMoved into neighborhood poverty before age 2 (n = 981) 145 (125ndash167)b 130 (112ndash151)b

Moved into neighborhood poverty after age 2 (n = 1039) 117 (101ndash136)b 102 (088ndash119)Trajectory 4 not born in poverty (n = 30 470)Stayed out of poverty to age 5 (n = 29 561) Reference ReferenceTransitioned into household poverty before age 2 (n = 497) 327 (274ndash391)b 171 (140ndash209)b

Transitioned into household poverty after age 2 (n = 412) 289 (237ndash352)b 168 (135ndash208)b

CFS Child and Family Services CI confidence interval EDI early development instrument OR odds ratioa Adjusted for maternal covariates during pregnancy (moved received services from CFS mood and anxiety disorders drug andor alcohol use smoking prenatal care use) maternalcovariates at the index date (education social isolation lone parent age neighborhood location) and child covariates at birth (birth order sex birth wt gestational age) These variablesare used for adjustment in all tablesb P 05

TABLE 5 Proportion of Population With Childhood Outcomes Before Age 5 by Trajectories and Type of Poverty

Placed in Out-of-Home Care

Externalizing MentalCondition Diagnosis

AsthmaDiagnosis

InjuryHospitalization

Trajectory 1 born in household poverty (n = 4610)Stayed in household poverty to age 5 (n = 3904) 239 79 305 21Transitioned out of poverty before age 2 (n = 347) 55 55 251 09Transitioned out of poverty after age 2 (n = 359) 123 103 306 11

Trajectory 2 born in neighborhood poverty (n = 5070)Stayed in neighborhood poverty to age 5 (n = 2290) 40 38 223 13Moved out of poverty before age 2 (n = 1656) 05 44 188 07Moved out of poverty after age 2 (n = 1124) 08 26 193 04

Trajectory 3 not born in poverty (n = 31 581)Stayed out of poverty to age 5 (n = 29 561) 02 33 190 06Moved into neighborhood poverty before age 2 (n = 981) 05 31 189 02Moved into neighborhood poverty after age 2 (n = 1039) 04 29 181 11

Trajectory 4 not born in poverty (n = 30 470)Stayed out of poverty to age 5 (n = 29 561) 02 33 190 06Transitioned into household poverty before age 2 (n = 497) 103 50 233 10Transitioned into household poverty after age 2 (n = 412) 63 46 231 07

8 ROOS et al by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

for the findings Moreover the rangeof our income measures is limitedTransition out of poverty (whetherhousehold or neighborhood) isunlikely to be associated withsubstantial income changesMovement out of the lowest incomequintile (defining neighborhoodpoverty) is primarily to quintile 2(437) and quintile 3 (234)neighborhoods

Finally the absolute andrelative importance of povertyin Canada compared with in theUnited States is intriguingNeighborhood variation and therole of neighborhood characteristicsappear greater in the UnitedStates44 Canadarsquos safety net is moreextensive than those in the UnitedStates and the United Kingdom4546

Seven-country comparisons havenoted steep American socioeconomic

status gradients along severaldimensions of well-being (includingcognitive and socioemotional)47

More detailed study of thesegradients might draw on ouranalyses

CONCLUSIONS

Children growing up in povertymust deal with more riskfactors for poor outcomes thanthose never experiencing povertyor experiencing poverty fora short time Children born intohousehold poverty have muchworse early outcomes than thoseborn into neighborhood povertySupport for children whose familiesare receiving welfare to transitionout of poverty when the childis young could yield the greatestbenefit

ACKNOWLEDGMENTS

In recognition of the high proportionof the children in care being of FirstNations descent the larger programof research using Child and FamilyServices data has been presented toseveral general and First Nationsaudiences

ABBREVIATIONS

aOR adjusted odds ratioICD International Classification of

DiseasesICD-9-CM International

Classification ofDiseases Ninth RevisionClinical Modification

ICD-10-CA InternationalClassification ofDiseases 10th RevisionCanada

FUNDING This work was supported by a Graduate Enhancement of Tri-Council Stipend from the University of Manitoba Data used in this study are from the

Population Health Research Data Repository housed at the Manitoba Centre for Health Policy University of Manitoba and were derived from data provided by

Manitoba Health Seniors and Active Living Manitoba Families and Healthy Child Manitoba under project 20132014-04 The results and conclusions are those of the

authors and no official endorsement by the Manitoba Centre for Health Policy Manitoba Health Seniors and Active Living or other data providers is intended or

should be inferred

POTENTIAL CONFLICT OF INTEREST The authors have indicated they have no potential conflicts of interest to disclose

COMPANION PAPER A companion to this article can be found online at wwwpediatricsorgcgidoi101542peds2019-0195

REFERENCES

1 Power C Kuh D Morton S Fromdevelopmental origins of adult diseaseto life course research on adult diseaseand aging insights from birth cohortstudies Annu Rev Public Health 2013347ndash28

2 Wood D Effect of child and familypoverty on child health in the UnitedStates Pediatrics 2003112(3 pt 2)707ndash711

3 Aber JL Bennett NG Conley DC Li J Theeffects of poverty on child health anddevelopment Annu Rev Public Health199718463ndash483

4 Brooks-Gunn J Duncan GJ The effectsof poverty on children Future Child19977(2)55ndash71

5 Hair NL Hanson JL Wolfe BL Pollak SDAssociation of child poverty braindevelopment and academicachievement [published correctionappears in JAMA Pediatr 2015169(9)878] JAMA Pediatr 2015169(9)822ndash829

6 Hart B Risley TR MeaningfulDifferences in the Everyday Experienceof Young American Children BaltimoreMD Paul H Brookes Publishing 1995

7 Cook JT Frank DA Levenson SM et alChild food insecurity increases risksposed by household food insecurity toyoung childrenrsquos health J Nutr 2006136(4)1073ndash1076

8 Casey PH Szeto KL Robbins JM et alChild health-related quality of life and

household food security Arch PediatrAdolesc Med 2005159(1)51ndash56

9 Romano E Babchishin L Pagani LSKohen D School readiness and laterachievement replication and extensionusing a nationwide Canadian surveyDev Psychol 201046(5)995ndash1007

10 Campbell FA Wasik BH Pungello E et alYoung adult outcomes of theAbecedarian and CARE early childhoodeducational interventions Early ChildRes Q 200823(4)452ndash466

11 Fothergill KE Ensminger ME Green KMCrum RM Robertson J Juon HS Theimpact of early school behavior andeducational achievement on adult druguse disorders a prospective study

PEDIATRICS Volume 143 number 6 June 2019 9 by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

Drug Alcohol Depend 200892(1ndash3)191ndash199

12 Bradley RH Corwyn RF Socioeconomicstatus and child development Annu RevPsychol 200253371ndash399

13 Browne DT Wade M Prime H JenkinsJM School readiness amongst urbanCanadian families risk profiles andfamily mediation J Educ Psychol 2018110(1)133ndash146

14 Putnam-Hornstein E Needell BPredictors of child protective servicecontact between birth and age five anexamination of Californiarsquos 2002 birthcohort Child Youth Serv Rev 201133(11)2400ndash2407

15 Currie J Stabile M Manivong P Roos LLChild health and young adult outcomesJ Hum Resour 201045(3)517ndash548

16 Duncan GJ Dowsett CJ Claessens Aet al School readiness and laterachievement Dev Psychol 200743(6)1428ndash1446

17 Shaw DS Shelleby EC Early-startingconduct problems intersection ofconduct problems and poverty AnnuRev Clin Psychol 201410503ndash528

18 Duncan GJ Magnuson K Votruba-DrzalE Moving beyond correlations inassessing the consequences of povertyAnnu Rev Psychol 201768413ndash434

19 McEwen CA McEwen BS Socialstructure adversity toxic stress andintergenerational poverty an earlychildhood model Annu Rev Sociol 201743(1)445ndash472

20 Sharkey P Elwert F The legacy ofdisadvantage multigenerationalneighborhood effects on cognitiveability AJS 2011116(6)1934ndash1981

21 Chetty R Hendren N Katz LF The effectsof exposure to better neighborhoods onchildren new evidence from the movingto opportunity experiment Am EconRev 2016106(4)855ndash902

22 Oreopoulos P Stabile M Walld R RoosLL Short- medium- and long-termconsequences of poor infant health ananalysis using siblings and twinsJ Hum Resour 200843(1)88ndash138

23 OrsquoGrady K Deussing M Scerbina T FungK Muhe N Measuring Up CanadianResults of the OECD PISA Study TorontoCanada Council of Ministers ofEducation Canada 2016

24 Statistics Canada Focus on GeographySeries 2011 Ottawa Canada StatisticsCanada 2014 Available at httpswww12statcangccacensus-recensement2011as-safogs-spgIndex-engcfm Accessed September 252018

25 Statistics Canada Aboriginal PeoplesFact Sheet for Manitoba OttawaCanada Statistics Canada 2016Available at httpswww150statcangccan1pub89-656-x89-656-x2016008-enghtm Accessed September 25 2018

26 OrsquoGrady K Fung K Servage L Khan GReport on the Pan-CanadianAssessment of Reading Mathematicsand Science Toronto Canada Councilof Ministers of Education Canada 2018

27 Canadian Institute for HealthInformation Children Vulnerable inAreas of Early Development ADeterminant of Child Health OttawaCanada Canadian Institute for HealthInformation 2014 Available at httpssecurecihicafree_productsChildren_Vulnerable_in_Areas_of_Early_Development_ENpdf Accessed September 252018

28 Roos LL Gupta S Soodeen RA JebamaniL Data quality in an information-richenvironment Canada as an exampleCan J Aging 200524(suppl 1)153ndash170

29 Roos LL Nicol JP A research registryuses development and accuracy J ClinEpidemiol 199952(1)39ndash47

30 Government of Manitoba Employmentand Income Assistance (EIA) Availableat httpswwwgovmbcafseiaAccessed September 25 2018

31 Manitoba Centre for Health PolicyConcept Income Quintile RankingProcedure Winnipeg Canada ManitobaCentre for Health Policy 2002 httpmchp-appservcpeumanitobacaviewConceptphpconceptID=1164Accessed September 25 2018

32 Janus M Brinkman S Duku E et al TheEarly Development Instrument APopulation-Based Measure forCommunities A Handbook onDevelopment Properties and UseHamilton Canada Offord Centre forChild Studies 2007

33 Snow KL Measuring school readinessconceptual and practical

considerations Early Educ Dev 200617(1)7ndash41

34 Brownell M Nickel NC Chartier M et alAn unconditional prenatal incomesupplement reduces populationinequities in birth outcomes Health Aff(Millwood) 201837(3)447ndash455

35 Government of Manitoba Familyservices and housing Child ProtectionServices Available at httpswwwgovmbcafschildfamindexhtmlAccessed March 27 2019

36 Manitoba Centre for Health PolicyTerm apprehended at birth taken intoCFS care at birth Available at httpmchp-appservcpeumanitobacaviewDefinitionphpdefinitionID=104764Accessed January 10 2019

37 Alexander GR Kotelchuck M Assessingthe role and effectiveness of prenatalcare history challenges and directionsfor future research Public Health Rep2001116(4)306ndash316

38 SAS Institute Inc Statistical AnalysisSoftware (SAS) version 94 2013

39 Almond D Currie J Duque V Childhoodcircumstances and adult outcomes ActII Journal of Economic Literature 201856(4)1360ndash1446

40 Bronfenbrenner U The Ecology ofHuman Development Experiments byNature and Design Cambridge MAHarvard University Press 1979

41 Government of Canada A backgrounderon poverty in Canada 2016 Available athttpwww12edscgccasgpe-pmpsservletsgpp-pmps-publang=engampcurjsp=p5bd2t13ls-engjspampcuractn=dwnldamppid=54405ampdid=4879Accessed January 10 2019

42 Statistics Canada Census Profile 2016Census Ottawa Canada StatisticsCanada 2017 Available at httpswww12statcangccacensus-recensement2016dp-pdprofdetailspagecfmLang=EampGeo1=PRampCode1=46ampGeo2=PRampCode2=01ampData=CountampSearchText=ManitobaampSearchType=BeginsampSearchPR=01ampB1=IncomeampTABID=1 Accessed March 27 2019

43 Statistics Canada Canadian IncomeSurvey 2015 Ottawa Canada StatisticsCanada 2017 Available at httpswww150statcangccan1endaily-quotidien170526dq170526a-engpdfst=kFoEj4AM Accessed March 27 2019

10 ROOS et al by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

Adjustment for other factors(themselves linked to poverty)weakens the relationships betweenpoverty and the various outcomesHousehold poverty remaineda statistically significant predictorafter controlling for numerouscovariates (although odds ratios weretypically reduced) After adjustmentsneighborhood poverty was generallyless predictive of childhood mentalhealth asthma and hospitalizationfor injuries

Our information provides a baselinefor additional research For exampleeducational performance (grade 3)and our childhood measures can helpassess changes by age 10 Moregenerally having relevant outcomesat several developmental stages aidsin judging the significance of early lifeconditions and events Not onlypoverty but such ldquoshocksrdquo as prenataland early life maternal stress mightwell affect adolescent and adultoutcomes39 Large databases canoperationalize such characteristics associal isolation death of a closerelative domestic violence andmaternal drug andor alcohol useOur findings also supportBronfenbrennerrsquos40 ecologicalsystems theory postulating mother-child interactions to be strongpredictors of child developmentconstructive mother-child bonds aremore difficult for mothers havingdrug or alcohol problemsOvercoming such factors among thedisadvantaged may be more difficultthan among the more affluent andinterventions to increase humancapital might prove more valuableamong the poor Information onseveral such programs (nurse visitingmaternal income supplements) arebeing incorporated into Manitobaanalyses34

This articlersquos limitations include thoseassociated with observational studiesReceipt of welfare was used to definehousehold poverty because access toinformation on household income islacking Our definition mayTA

BLE3

Continued

School

Readiness(EDI)

Trajectory

1HouseholdPovertyto

NoPoverty

Trajectory

2Neighborhood

Povertyto

NoPoverty

Trajectory

3No

Povertyto

Neighborhood

Poverty

Trajectory

4No

Povertyto

Household

Poverty

Unadjusted

OR(95

CI)

aORa

(95

CI)

Unadjusted

OR(95

CI)

aORa

(95

CI)

Unadjusted

OR(95

CI)

aORa

(95

CI)

Unadjusted

OR(95

CI)

aORa

(95

CI)

Afterage2e

134

(093ndash192)

139

(096ndash201)

067

(044ndash103)

074

(048ndash115)

087

(060ndash125)

086

(059ndash125)

141

(089ndash224)

110

(068ndash178)

Asthmadiagnosis0ndash5

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

076

(059ndash098)d

091

(070ndash118)

081

(069ndash094)d

084

(071ndash099)d

099

(084ndash117)

102

(087ndash120)

130

(105ndash160)d

108

(087ndash136)

Afterage2e

101

(080ndash128)

107

(084ndash136)

082

(069ndash098)d

084

(070ndash101)

094

(080ndash111)

095

(081ndash112)

128

(101ndash161)d

113

(089ndash144)

Injury

hospitalization

0ndash5

Nochange

bReference

Reference

Reference

Reference

Reference

Reference

Reference

Reference

Before

age2c

041

(013ndash131)

069

(021ndash224)

052

(026ndash105)

054

(026ndash111)

037

(009ndash149)

033

(008134)

183

(075ndash448)

106

(040ndash276)

Afterage2e

053

(019ndash146)

071

(026ndash197)

035

(014ndash090)d

035

(014ndash093)d

193

(105ndash357)d

163

(088ndash304)

132

(042ndash416)

083

(025ndash269)

CFSChild

andFamily

ServicesC

IconfidenceintervalEDIearlydevelopm

entinstrumentORoddsratio

aAdjusted

formaternalcovariatesduring

pregnancy(m

ovedreceivedservices

from

CFSmoodandanxietydisordersdrug

andor

alcoholusesmokingprenatalcare

use)m

aternalcovariatesat

theindexdate(educationsocialisolationlone

parentagen

eighborhoodlocation)and

child

covariates

atbirth(birth

ordersexbirthwtgestationalage)

bNo

change

inpovertylevelandor

type

before

age5

cChange

inpovertylevelandor

type

betweenbirthandage2

dP

05

eChange

inpovertylevelandor

type

betweenage2andage5

PEDIATRICS Volume 143 number 6 June 2019 7 by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

underestimate such poverty Canadahas no official poverty line howeverhouseholds having incomes less thanhalf of the median household income(for a family of a given size) areconsidered low income41 ManitobaEmployment and Income Assistance(and other benefits) generally payless than this amount Manitobarsquosaverage total income of householdswas below the Canadian nationalaverage in 2015 (Manitoba $85373Canada $92764)42 The CanadianIncome Survey data suggested that

Manitobarsquos median employmentincome was consistently lower thanthe Canadian national median from2012 to 201543 In 2015 Manitobarsquosprevalence of low income based onthe low-income cutoff after tax was99 which was slightly higher thanthe national prevalence (92)42

Additionally although children areclustered within neighborhoods wedid not account for this in ouranalysis (ie use a multilevel model)This was done because in many caseschildren lived in 1 neighborhood

between birth and age 5 with theamount of time spent in eachneighborhood varying by childchildren would often belong inseveral clusters over time

Because the poverty variable couldnot be manipulated in experimentalor quasi-experimental fashioncausality cannot be ascertainedSelective movement out of poverty byless challenged families (or intopoverty by their more challengedcounterparts) might be responsible

TABLE 4 Odds Ratios for Not Being Ready For School Trajectories by Type of Poverty

Not Ready on 1 or More Domain (EDI) Unadjusted OR (95 CI) aORa (95 CI)

Trajectory 1 born in household poverty (n = 4610)Stayed in household poverty to age 5 (n = 3904) Reference ReferenceTransitioned out of poverty before age 2 (n = 347) 049 (039ndash061)b 058 (046ndash075)b

Transitioned out of poverty after age 2 (n = 359) 067 (054ndash083)b 075 (060ndash094)b

Trajectory 2 born in neighborhood poverty (n = 5070)Stayed in neighborhood poverty to age 5 (n = 2290) Reference ReferenceMoved out of poverty before age 2 (n = 1656) 059 (051ndash068)b 073 (063ndash086)b

Moved out of poverty after age 2 (n = 1124) 080 (068ndash094)b 094 (080ndash111)Trajectory 3 not born in poverty (n = 31 581)Stayed out of poverty to age 5 (n = 29 561) Reference ReferenceMoved into neighborhood poverty before age 2 (n = 981) 145 (125ndash167)b 130 (112ndash151)b

Moved into neighborhood poverty after age 2 (n = 1039) 117 (101ndash136)b 102 (088ndash119)Trajectory 4 not born in poverty (n = 30 470)Stayed out of poverty to age 5 (n = 29 561) Reference ReferenceTransitioned into household poverty before age 2 (n = 497) 327 (274ndash391)b 171 (140ndash209)b

Transitioned into household poverty after age 2 (n = 412) 289 (237ndash352)b 168 (135ndash208)b

CFS Child and Family Services CI confidence interval EDI early development instrument OR odds ratioa Adjusted for maternal covariates during pregnancy (moved received services from CFS mood and anxiety disorders drug andor alcohol use smoking prenatal care use) maternalcovariates at the index date (education social isolation lone parent age neighborhood location) and child covariates at birth (birth order sex birth wt gestational age) These variablesare used for adjustment in all tablesb P 05

TABLE 5 Proportion of Population With Childhood Outcomes Before Age 5 by Trajectories and Type of Poverty

Placed in Out-of-Home Care

Externalizing MentalCondition Diagnosis

AsthmaDiagnosis

InjuryHospitalization

Trajectory 1 born in household poverty (n = 4610)Stayed in household poverty to age 5 (n = 3904) 239 79 305 21Transitioned out of poverty before age 2 (n = 347) 55 55 251 09Transitioned out of poverty after age 2 (n = 359) 123 103 306 11

Trajectory 2 born in neighborhood poverty (n = 5070)Stayed in neighborhood poverty to age 5 (n = 2290) 40 38 223 13Moved out of poverty before age 2 (n = 1656) 05 44 188 07Moved out of poverty after age 2 (n = 1124) 08 26 193 04

Trajectory 3 not born in poverty (n = 31 581)Stayed out of poverty to age 5 (n = 29 561) 02 33 190 06Moved into neighborhood poverty before age 2 (n = 981) 05 31 189 02Moved into neighborhood poverty after age 2 (n = 1039) 04 29 181 11

Trajectory 4 not born in poverty (n = 30 470)Stayed out of poverty to age 5 (n = 29 561) 02 33 190 06Transitioned into household poverty before age 2 (n = 497) 103 50 233 10Transitioned into household poverty after age 2 (n = 412) 63 46 231 07

8 ROOS et al by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

for the findings Moreover the rangeof our income measures is limitedTransition out of poverty (whetherhousehold or neighborhood) isunlikely to be associated withsubstantial income changesMovement out of the lowest incomequintile (defining neighborhoodpoverty) is primarily to quintile 2(437) and quintile 3 (234)neighborhoods

Finally the absolute andrelative importance of povertyin Canada compared with in theUnited States is intriguingNeighborhood variation and therole of neighborhood characteristicsappear greater in the UnitedStates44 Canadarsquos safety net is moreextensive than those in the UnitedStates and the United Kingdom4546

Seven-country comparisons havenoted steep American socioeconomic

status gradients along severaldimensions of well-being (includingcognitive and socioemotional)47

More detailed study of thesegradients might draw on ouranalyses

CONCLUSIONS

Children growing up in povertymust deal with more riskfactors for poor outcomes thanthose never experiencing povertyor experiencing poverty fora short time Children born intohousehold poverty have muchworse early outcomes than thoseborn into neighborhood povertySupport for children whose familiesare receiving welfare to transitionout of poverty when the childis young could yield the greatestbenefit

ACKNOWLEDGMENTS

In recognition of the high proportionof the children in care being of FirstNations descent the larger programof research using Child and FamilyServices data has been presented toseveral general and First Nationsaudiences

ABBREVIATIONS

aOR adjusted odds ratioICD International Classification of

DiseasesICD-9-CM International

Classification ofDiseases Ninth RevisionClinical Modification

ICD-10-CA InternationalClassification ofDiseases 10th RevisionCanada

FUNDING This work was supported by a Graduate Enhancement of Tri-Council Stipend from the University of Manitoba Data used in this study are from the

Population Health Research Data Repository housed at the Manitoba Centre for Health Policy University of Manitoba and were derived from data provided by

Manitoba Health Seniors and Active Living Manitoba Families and Healthy Child Manitoba under project 20132014-04 The results and conclusions are those of the

authors and no official endorsement by the Manitoba Centre for Health Policy Manitoba Health Seniors and Active Living or other data providers is intended or

should be inferred

POTENTIAL CONFLICT OF INTEREST The authors have indicated they have no potential conflicts of interest to disclose

COMPANION PAPER A companion to this article can be found online at wwwpediatricsorgcgidoi101542peds2019-0195

REFERENCES

1 Power C Kuh D Morton S Fromdevelopmental origins of adult diseaseto life course research on adult diseaseand aging insights from birth cohortstudies Annu Rev Public Health 2013347ndash28

2 Wood D Effect of child and familypoverty on child health in the UnitedStates Pediatrics 2003112(3 pt 2)707ndash711

3 Aber JL Bennett NG Conley DC Li J Theeffects of poverty on child health anddevelopment Annu Rev Public Health199718463ndash483

4 Brooks-Gunn J Duncan GJ The effectsof poverty on children Future Child19977(2)55ndash71

5 Hair NL Hanson JL Wolfe BL Pollak SDAssociation of child poverty braindevelopment and academicachievement [published correctionappears in JAMA Pediatr 2015169(9)878] JAMA Pediatr 2015169(9)822ndash829

6 Hart B Risley TR MeaningfulDifferences in the Everyday Experienceof Young American Children BaltimoreMD Paul H Brookes Publishing 1995

7 Cook JT Frank DA Levenson SM et alChild food insecurity increases risksposed by household food insecurity toyoung childrenrsquos health J Nutr 2006136(4)1073ndash1076

8 Casey PH Szeto KL Robbins JM et alChild health-related quality of life and

household food security Arch PediatrAdolesc Med 2005159(1)51ndash56

9 Romano E Babchishin L Pagani LSKohen D School readiness and laterachievement replication and extensionusing a nationwide Canadian surveyDev Psychol 201046(5)995ndash1007

10 Campbell FA Wasik BH Pungello E et alYoung adult outcomes of theAbecedarian and CARE early childhoodeducational interventions Early ChildRes Q 200823(4)452ndash466

11 Fothergill KE Ensminger ME Green KMCrum RM Robertson J Juon HS Theimpact of early school behavior andeducational achievement on adult druguse disorders a prospective study

PEDIATRICS Volume 143 number 6 June 2019 9 by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

Drug Alcohol Depend 200892(1ndash3)191ndash199

12 Bradley RH Corwyn RF Socioeconomicstatus and child development Annu RevPsychol 200253371ndash399

13 Browne DT Wade M Prime H JenkinsJM School readiness amongst urbanCanadian families risk profiles andfamily mediation J Educ Psychol 2018110(1)133ndash146

14 Putnam-Hornstein E Needell BPredictors of child protective servicecontact between birth and age five anexamination of Californiarsquos 2002 birthcohort Child Youth Serv Rev 201133(11)2400ndash2407

15 Currie J Stabile M Manivong P Roos LLChild health and young adult outcomesJ Hum Resour 201045(3)517ndash548

16 Duncan GJ Dowsett CJ Claessens Aet al School readiness and laterachievement Dev Psychol 200743(6)1428ndash1446

17 Shaw DS Shelleby EC Early-startingconduct problems intersection ofconduct problems and poverty AnnuRev Clin Psychol 201410503ndash528

18 Duncan GJ Magnuson K Votruba-DrzalE Moving beyond correlations inassessing the consequences of povertyAnnu Rev Psychol 201768413ndash434

19 McEwen CA McEwen BS Socialstructure adversity toxic stress andintergenerational poverty an earlychildhood model Annu Rev Sociol 201743(1)445ndash472

20 Sharkey P Elwert F The legacy ofdisadvantage multigenerationalneighborhood effects on cognitiveability AJS 2011116(6)1934ndash1981

21 Chetty R Hendren N Katz LF The effectsof exposure to better neighborhoods onchildren new evidence from the movingto opportunity experiment Am EconRev 2016106(4)855ndash902

22 Oreopoulos P Stabile M Walld R RoosLL Short- medium- and long-termconsequences of poor infant health ananalysis using siblings and twinsJ Hum Resour 200843(1)88ndash138

23 OrsquoGrady K Deussing M Scerbina T FungK Muhe N Measuring Up CanadianResults of the OECD PISA Study TorontoCanada Council of Ministers ofEducation Canada 2016

24 Statistics Canada Focus on GeographySeries 2011 Ottawa Canada StatisticsCanada 2014 Available at httpswww12statcangccacensus-recensement2011as-safogs-spgIndex-engcfm Accessed September 252018

25 Statistics Canada Aboriginal PeoplesFact Sheet for Manitoba OttawaCanada Statistics Canada 2016Available at httpswww150statcangccan1pub89-656-x89-656-x2016008-enghtm Accessed September 25 2018

26 OrsquoGrady K Fung K Servage L Khan GReport on the Pan-CanadianAssessment of Reading Mathematicsand Science Toronto Canada Councilof Ministers of Education Canada 2018

27 Canadian Institute for HealthInformation Children Vulnerable inAreas of Early Development ADeterminant of Child Health OttawaCanada Canadian Institute for HealthInformation 2014 Available at httpssecurecihicafree_productsChildren_Vulnerable_in_Areas_of_Early_Development_ENpdf Accessed September 252018

28 Roos LL Gupta S Soodeen RA JebamaniL Data quality in an information-richenvironment Canada as an exampleCan J Aging 200524(suppl 1)153ndash170

29 Roos LL Nicol JP A research registryuses development and accuracy J ClinEpidemiol 199952(1)39ndash47

30 Government of Manitoba Employmentand Income Assistance (EIA) Availableat httpswwwgovmbcafseiaAccessed September 25 2018

31 Manitoba Centre for Health PolicyConcept Income Quintile RankingProcedure Winnipeg Canada ManitobaCentre for Health Policy 2002 httpmchp-appservcpeumanitobacaviewConceptphpconceptID=1164Accessed September 25 2018

32 Janus M Brinkman S Duku E et al TheEarly Development Instrument APopulation-Based Measure forCommunities A Handbook onDevelopment Properties and UseHamilton Canada Offord Centre forChild Studies 2007

33 Snow KL Measuring school readinessconceptual and practical

considerations Early Educ Dev 200617(1)7ndash41

34 Brownell M Nickel NC Chartier M et alAn unconditional prenatal incomesupplement reduces populationinequities in birth outcomes Health Aff(Millwood) 201837(3)447ndash455

35 Government of Manitoba Familyservices and housing Child ProtectionServices Available at httpswwwgovmbcafschildfamindexhtmlAccessed March 27 2019

36 Manitoba Centre for Health PolicyTerm apprehended at birth taken intoCFS care at birth Available at httpmchp-appservcpeumanitobacaviewDefinitionphpdefinitionID=104764Accessed January 10 2019

37 Alexander GR Kotelchuck M Assessingthe role and effectiveness of prenatalcare history challenges and directionsfor future research Public Health Rep2001116(4)306ndash316

38 SAS Institute Inc Statistical AnalysisSoftware (SAS) version 94 2013

39 Almond D Currie J Duque V Childhoodcircumstances and adult outcomes ActII Journal of Economic Literature 201856(4)1360ndash1446

40 Bronfenbrenner U The Ecology ofHuman Development Experiments byNature and Design Cambridge MAHarvard University Press 1979

41 Government of Canada A backgrounderon poverty in Canada 2016 Available athttpwww12edscgccasgpe-pmpsservletsgpp-pmps-publang=engampcurjsp=p5bd2t13ls-engjspampcuractn=dwnldamppid=54405ampdid=4879Accessed January 10 2019

42 Statistics Canada Census Profile 2016Census Ottawa Canada StatisticsCanada 2017 Available at httpswww12statcangccacensus-recensement2016dp-pdprofdetailspagecfmLang=EampGeo1=PRampCode1=46ampGeo2=PRampCode2=01ampData=CountampSearchText=ManitobaampSearchType=BeginsampSearchPR=01ampB1=IncomeampTABID=1 Accessed March 27 2019

43 Statistics Canada Canadian IncomeSurvey 2015 Ottawa Canada StatisticsCanada 2017 Available at httpswww150statcangccan1endaily-quotidien170526dq170526a-engpdfst=kFoEj4AM Accessed March 27 2019

10 ROOS et al by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

underestimate such poverty Canadahas no official poverty line howeverhouseholds having incomes less thanhalf of the median household income(for a family of a given size) areconsidered low income41 ManitobaEmployment and Income Assistance(and other benefits) generally payless than this amount Manitobarsquosaverage total income of householdswas below the Canadian nationalaverage in 2015 (Manitoba $85373Canada $92764)42 The CanadianIncome Survey data suggested that

Manitobarsquos median employmentincome was consistently lower thanthe Canadian national median from2012 to 201543 In 2015 Manitobarsquosprevalence of low income based onthe low-income cutoff after tax was99 which was slightly higher thanthe national prevalence (92)42

Additionally although children areclustered within neighborhoods wedid not account for this in ouranalysis (ie use a multilevel model)This was done because in many caseschildren lived in 1 neighborhood

between birth and age 5 with theamount of time spent in eachneighborhood varying by childchildren would often belong inseveral clusters over time

Because the poverty variable couldnot be manipulated in experimentalor quasi-experimental fashioncausality cannot be ascertainedSelective movement out of poverty byless challenged families (or intopoverty by their more challengedcounterparts) might be responsible

TABLE 4 Odds Ratios for Not Being Ready For School Trajectories by Type of Poverty

Not Ready on 1 or More Domain (EDI) Unadjusted OR (95 CI) aORa (95 CI)

Trajectory 1 born in household poverty (n = 4610)Stayed in household poverty to age 5 (n = 3904) Reference ReferenceTransitioned out of poverty before age 2 (n = 347) 049 (039ndash061)b 058 (046ndash075)b

Transitioned out of poverty after age 2 (n = 359) 067 (054ndash083)b 075 (060ndash094)b

Trajectory 2 born in neighborhood poverty (n = 5070)Stayed in neighborhood poverty to age 5 (n = 2290) Reference ReferenceMoved out of poverty before age 2 (n = 1656) 059 (051ndash068)b 073 (063ndash086)b

Moved out of poverty after age 2 (n = 1124) 080 (068ndash094)b 094 (080ndash111)Trajectory 3 not born in poverty (n = 31 581)Stayed out of poverty to age 5 (n = 29 561) Reference ReferenceMoved into neighborhood poverty before age 2 (n = 981) 145 (125ndash167)b 130 (112ndash151)b

Moved into neighborhood poverty after age 2 (n = 1039) 117 (101ndash136)b 102 (088ndash119)Trajectory 4 not born in poverty (n = 30 470)Stayed out of poverty to age 5 (n = 29 561) Reference ReferenceTransitioned into household poverty before age 2 (n = 497) 327 (274ndash391)b 171 (140ndash209)b

Transitioned into household poverty after age 2 (n = 412) 289 (237ndash352)b 168 (135ndash208)b

CFS Child and Family Services CI confidence interval EDI early development instrument OR odds ratioa Adjusted for maternal covariates during pregnancy (moved received services from CFS mood and anxiety disorders drug andor alcohol use smoking prenatal care use) maternalcovariates at the index date (education social isolation lone parent age neighborhood location) and child covariates at birth (birth order sex birth wt gestational age) These variablesare used for adjustment in all tablesb P 05

TABLE 5 Proportion of Population With Childhood Outcomes Before Age 5 by Trajectories and Type of Poverty

Placed in Out-of-Home Care

Externalizing MentalCondition Diagnosis

AsthmaDiagnosis

InjuryHospitalization

Trajectory 1 born in household poverty (n = 4610)Stayed in household poverty to age 5 (n = 3904) 239 79 305 21Transitioned out of poverty before age 2 (n = 347) 55 55 251 09Transitioned out of poverty after age 2 (n = 359) 123 103 306 11

Trajectory 2 born in neighborhood poverty (n = 5070)Stayed in neighborhood poverty to age 5 (n = 2290) 40 38 223 13Moved out of poverty before age 2 (n = 1656) 05 44 188 07Moved out of poverty after age 2 (n = 1124) 08 26 193 04

Trajectory 3 not born in poverty (n = 31 581)Stayed out of poverty to age 5 (n = 29 561) 02 33 190 06Moved into neighborhood poverty before age 2 (n = 981) 05 31 189 02Moved into neighborhood poverty after age 2 (n = 1039) 04 29 181 11

Trajectory 4 not born in poverty (n = 30 470)Stayed out of poverty to age 5 (n = 29 561) 02 33 190 06Transitioned into household poverty before age 2 (n = 497) 103 50 233 10Transitioned into household poverty after age 2 (n = 412) 63 46 231 07

8 ROOS et al by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

for the findings Moreover the rangeof our income measures is limitedTransition out of poverty (whetherhousehold or neighborhood) isunlikely to be associated withsubstantial income changesMovement out of the lowest incomequintile (defining neighborhoodpoverty) is primarily to quintile 2(437) and quintile 3 (234)neighborhoods

Finally the absolute andrelative importance of povertyin Canada compared with in theUnited States is intriguingNeighborhood variation and therole of neighborhood characteristicsappear greater in the UnitedStates44 Canadarsquos safety net is moreextensive than those in the UnitedStates and the United Kingdom4546

Seven-country comparisons havenoted steep American socioeconomic

status gradients along severaldimensions of well-being (includingcognitive and socioemotional)47

More detailed study of thesegradients might draw on ouranalyses

CONCLUSIONS

Children growing up in povertymust deal with more riskfactors for poor outcomes thanthose never experiencing povertyor experiencing poverty fora short time Children born intohousehold poverty have muchworse early outcomes than thoseborn into neighborhood povertySupport for children whose familiesare receiving welfare to transitionout of poverty when the childis young could yield the greatestbenefit

ACKNOWLEDGMENTS

In recognition of the high proportionof the children in care being of FirstNations descent the larger programof research using Child and FamilyServices data has been presented toseveral general and First Nationsaudiences

ABBREVIATIONS

aOR adjusted odds ratioICD International Classification of

DiseasesICD-9-CM International

Classification ofDiseases Ninth RevisionClinical Modification

ICD-10-CA InternationalClassification ofDiseases 10th RevisionCanada

FUNDING This work was supported by a Graduate Enhancement of Tri-Council Stipend from the University of Manitoba Data used in this study are from the

Population Health Research Data Repository housed at the Manitoba Centre for Health Policy University of Manitoba and were derived from data provided by

Manitoba Health Seniors and Active Living Manitoba Families and Healthy Child Manitoba under project 20132014-04 The results and conclusions are those of the

authors and no official endorsement by the Manitoba Centre for Health Policy Manitoba Health Seniors and Active Living or other data providers is intended or

should be inferred

POTENTIAL CONFLICT OF INTEREST The authors have indicated they have no potential conflicts of interest to disclose

COMPANION PAPER A companion to this article can be found online at wwwpediatricsorgcgidoi101542peds2019-0195

REFERENCES

1 Power C Kuh D Morton S Fromdevelopmental origins of adult diseaseto life course research on adult diseaseand aging insights from birth cohortstudies Annu Rev Public Health 2013347ndash28

2 Wood D Effect of child and familypoverty on child health in the UnitedStates Pediatrics 2003112(3 pt 2)707ndash711

3 Aber JL Bennett NG Conley DC Li J Theeffects of poverty on child health anddevelopment Annu Rev Public Health199718463ndash483

4 Brooks-Gunn J Duncan GJ The effectsof poverty on children Future Child19977(2)55ndash71

5 Hair NL Hanson JL Wolfe BL Pollak SDAssociation of child poverty braindevelopment and academicachievement [published correctionappears in JAMA Pediatr 2015169(9)878] JAMA Pediatr 2015169(9)822ndash829

6 Hart B Risley TR MeaningfulDifferences in the Everyday Experienceof Young American Children BaltimoreMD Paul H Brookes Publishing 1995

7 Cook JT Frank DA Levenson SM et alChild food insecurity increases risksposed by household food insecurity toyoung childrenrsquos health J Nutr 2006136(4)1073ndash1076

8 Casey PH Szeto KL Robbins JM et alChild health-related quality of life and

household food security Arch PediatrAdolesc Med 2005159(1)51ndash56

9 Romano E Babchishin L Pagani LSKohen D School readiness and laterachievement replication and extensionusing a nationwide Canadian surveyDev Psychol 201046(5)995ndash1007

10 Campbell FA Wasik BH Pungello E et alYoung adult outcomes of theAbecedarian and CARE early childhoodeducational interventions Early ChildRes Q 200823(4)452ndash466

11 Fothergill KE Ensminger ME Green KMCrum RM Robertson J Juon HS Theimpact of early school behavior andeducational achievement on adult druguse disorders a prospective study

PEDIATRICS Volume 143 number 6 June 2019 9 by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

Drug Alcohol Depend 200892(1ndash3)191ndash199

12 Bradley RH Corwyn RF Socioeconomicstatus and child development Annu RevPsychol 200253371ndash399

13 Browne DT Wade M Prime H JenkinsJM School readiness amongst urbanCanadian families risk profiles andfamily mediation J Educ Psychol 2018110(1)133ndash146

14 Putnam-Hornstein E Needell BPredictors of child protective servicecontact between birth and age five anexamination of Californiarsquos 2002 birthcohort Child Youth Serv Rev 201133(11)2400ndash2407

15 Currie J Stabile M Manivong P Roos LLChild health and young adult outcomesJ Hum Resour 201045(3)517ndash548

16 Duncan GJ Dowsett CJ Claessens Aet al School readiness and laterachievement Dev Psychol 200743(6)1428ndash1446

17 Shaw DS Shelleby EC Early-startingconduct problems intersection ofconduct problems and poverty AnnuRev Clin Psychol 201410503ndash528

18 Duncan GJ Magnuson K Votruba-DrzalE Moving beyond correlations inassessing the consequences of povertyAnnu Rev Psychol 201768413ndash434

19 McEwen CA McEwen BS Socialstructure adversity toxic stress andintergenerational poverty an earlychildhood model Annu Rev Sociol 201743(1)445ndash472

20 Sharkey P Elwert F The legacy ofdisadvantage multigenerationalneighborhood effects on cognitiveability AJS 2011116(6)1934ndash1981

21 Chetty R Hendren N Katz LF The effectsof exposure to better neighborhoods onchildren new evidence from the movingto opportunity experiment Am EconRev 2016106(4)855ndash902

22 Oreopoulos P Stabile M Walld R RoosLL Short- medium- and long-termconsequences of poor infant health ananalysis using siblings and twinsJ Hum Resour 200843(1)88ndash138

23 OrsquoGrady K Deussing M Scerbina T FungK Muhe N Measuring Up CanadianResults of the OECD PISA Study TorontoCanada Council of Ministers ofEducation Canada 2016

24 Statistics Canada Focus on GeographySeries 2011 Ottawa Canada StatisticsCanada 2014 Available at httpswww12statcangccacensus-recensement2011as-safogs-spgIndex-engcfm Accessed September 252018

25 Statistics Canada Aboriginal PeoplesFact Sheet for Manitoba OttawaCanada Statistics Canada 2016Available at httpswww150statcangccan1pub89-656-x89-656-x2016008-enghtm Accessed September 25 2018

26 OrsquoGrady K Fung K Servage L Khan GReport on the Pan-CanadianAssessment of Reading Mathematicsand Science Toronto Canada Councilof Ministers of Education Canada 2018

27 Canadian Institute for HealthInformation Children Vulnerable inAreas of Early Development ADeterminant of Child Health OttawaCanada Canadian Institute for HealthInformation 2014 Available at httpssecurecihicafree_productsChildren_Vulnerable_in_Areas_of_Early_Development_ENpdf Accessed September 252018

28 Roos LL Gupta S Soodeen RA JebamaniL Data quality in an information-richenvironment Canada as an exampleCan J Aging 200524(suppl 1)153ndash170

29 Roos LL Nicol JP A research registryuses development and accuracy J ClinEpidemiol 199952(1)39ndash47

30 Government of Manitoba Employmentand Income Assistance (EIA) Availableat httpswwwgovmbcafseiaAccessed September 25 2018

31 Manitoba Centre for Health PolicyConcept Income Quintile RankingProcedure Winnipeg Canada ManitobaCentre for Health Policy 2002 httpmchp-appservcpeumanitobacaviewConceptphpconceptID=1164Accessed September 25 2018

32 Janus M Brinkman S Duku E et al TheEarly Development Instrument APopulation-Based Measure forCommunities A Handbook onDevelopment Properties and UseHamilton Canada Offord Centre forChild Studies 2007

33 Snow KL Measuring school readinessconceptual and practical

considerations Early Educ Dev 200617(1)7ndash41

34 Brownell M Nickel NC Chartier M et alAn unconditional prenatal incomesupplement reduces populationinequities in birth outcomes Health Aff(Millwood) 201837(3)447ndash455

35 Government of Manitoba Familyservices and housing Child ProtectionServices Available at httpswwwgovmbcafschildfamindexhtmlAccessed March 27 2019

36 Manitoba Centre for Health PolicyTerm apprehended at birth taken intoCFS care at birth Available at httpmchp-appservcpeumanitobacaviewDefinitionphpdefinitionID=104764Accessed January 10 2019

37 Alexander GR Kotelchuck M Assessingthe role and effectiveness of prenatalcare history challenges and directionsfor future research Public Health Rep2001116(4)306ndash316

38 SAS Institute Inc Statistical AnalysisSoftware (SAS) version 94 2013

39 Almond D Currie J Duque V Childhoodcircumstances and adult outcomes ActII Journal of Economic Literature 201856(4)1360ndash1446

40 Bronfenbrenner U The Ecology ofHuman Development Experiments byNature and Design Cambridge MAHarvard University Press 1979

41 Government of Canada A backgrounderon poverty in Canada 2016 Available athttpwww12edscgccasgpe-pmpsservletsgpp-pmps-publang=engampcurjsp=p5bd2t13ls-engjspampcuractn=dwnldamppid=54405ampdid=4879Accessed January 10 2019

42 Statistics Canada Census Profile 2016Census Ottawa Canada StatisticsCanada 2017 Available at httpswww12statcangccacensus-recensement2016dp-pdprofdetailspagecfmLang=EampGeo1=PRampCode1=46ampGeo2=PRampCode2=01ampData=CountampSearchText=ManitobaampSearchType=BeginsampSearchPR=01ampB1=IncomeampTABID=1 Accessed March 27 2019

43 Statistics Canada Canadian IncomeSurvey 2015 Ottawa Canada StatisticsCanada 2017 Available at httpswww150statcangccan1endaily-quotidien170526dq170526a-engpdfst=kFoEj4AM Accessed March 27 2019

10 ROOS et al by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

for the findings Moreover the rangeof our income measures is limitedTransition out of poverty (whetherhousehold or neighborhood) isunlikely to be associated withsubstantial income changesMovement out of the lowest incomequintile (defining neighborhoodpoverty) is primarily to quintile 2(437) and quintile 3 (234)neighborhoods

Finally the absolute andrelative importance of povertyin Canada compared with in theUnited States is intriguingNeighborhood variation and therole of neighborhood characteristicsappear greater in the UnitedStates44 Canadarsquos safety net is moreextensive than those in the UnitedStates and the United Kingdom4546

Seven-country comparisons havenoted steep American socioeconomic

status gradients along severaldimensions of well-being (includingcognitive and socioemotional)47

More detailed study of thesegradients might draw on ouranalyses

CONCLUSIONS

Children growing up in povertymust deal with more riskfactors for poor outcomes thanthose never experiencing povertyor experiencing poverty fora short time Children born intohousehold poverty have muchworse early outcomes than thoseborn into neighborhood povertySupport for children whose familiesare receiving welfare to transitionout of poverty when the childis young could yield the greatestbenefit

ACKNOWLEDGMENTS

In recognition of the high proportionof the children in care being of FirstNations descent the larger programof research using Child and FamilyServices data has been presented toseveral general and First Nationsaudiences

ABBREVIATIONS

aOR adjusted odds ratioICD International Classification of

DiseasesICD-9-CM International

Classification ofDiseases Ninth RevisionClinical Modification

ICD-10-CA InternationalClassification ofDiseases 10th RevisionCanada

FUNDING This work was supported by a Graduate Enhancement of Tri-Council Stipend from the University of Manitoba Data used in this study are from the

Population Health Research Data Repository housed at the Manitoba Centre for Health Policy University of Manitoba and were derived from data provided by

Manitoba Health Seniors and Active Living Manitoba Families and Healthy Child Manitoba under project 20132014-04 The results and conclusions are those of the

authors and no official endorsement by the Manitoba Centre for Health Policy Manitoba Health Seniors and Active Living or other data providers is intended or

should be inferred

POTENTIAL CONFLICT OF INTEREST The authors have indicated they have no potential conflicts of interest to disclose

COMPANION PAPER A companion to this article can be found online at wwwpediatricsorgcgidoi101542peds2019-0195

REFERENCES

1 Power C Kuh D Morton S Fromdevelopmental origins of adult diseaseto life course research on adult diseaseand aging insights from birth cohortstudies Annu Rev Public Health 2013347ndash28

2 Wood D Effect of child and familypoverty on child health in the UnitedStates Pediatrics 2003112(3 pt 2)707ndash711

3 Aber JL Bennett NG Conley DC Li J Theeffects of poverty on child health anddevelopment Annu Rev Public Health199718463ndash483

4 Brooks-Gunn J Duncan GJ The effectsof poverty on children Future Child19977(2)55ndash71

5 Hair NL Hanson JL Wolfe BL Pollak SDAssociation of child poverty braindevelopment and academicachievement [published correctionappears in JAMA Pediatr 2015169(9)878] JAMA Pediatr 2015169(9)822ndash829

6 Hart B Risley TR MeaningfulDifferences in the Everyday Experienceof Young American Children BaltimoreMD Paul H Brookes Publishing 1995

7 Cook JT Frank DA Levenson SM et alChild food insecurity increases risksposed by household food insecurity toyoung childrenrsquos health J Nutr 2006136(4)1073ndash1076

8 Casey PH Szeto KL Robbins JM et alChild health-related quality of life and

household food security Arch PediatrAdolesc Med 2005159(1)51ndash56

9 Romano E Babchishin L Pagani LSKohen D School readiness and laterachievement replication and extensionusing a nationwide Canadian surveyDev Psychol 201046(5)995ndash1007

10 Campbell FA Wasik BH Pungello E et alYoung adult outcomes of theAbecedarian and CARE early childhoodeducational interventions Early ChildRes Q 200823(4)452ndash466

11 Fothergill KE Ensminger ME Green KMCrum RM Robertson J Juon HS Theimpact of early school behavior andeducational achievement on adult druguse disorders a prospective study

PEDIATRICS Volume 143 number 6 June 2019 9 by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

Drug Alcohol Depend 200892(1ndash3)191ndash199

12 Bradley RH Corwyn RF Socioeconomicstatus and child development Annu RevPsychol 200253371ndash399

13 Browne DT Wade M Prime H JenkinsJM School readiness amongst urbanCanadian families risk profiles andfamily mediation J Educ Psychol 2018110(1)133ndash146

14 Putnam-Hornstein E Needell BPredictors of child protective servicecontact between birth and age five anexamination of Californiarsquos 2002 birthcohort Child Youth Serv Rev 201133(11)2400ndash2407

15 Currie J Stabile M Manivong P Roos LLChild health and young adult outcomesJ Hum Resour 201045(3)517ndash548

16 Duncan GJ Dowsett CJ Claessens Aet al School readiness and laterachievement Dev Psychol 200743(6)1428ndash1446

17 Shaw DS Shelleby EC Early-startingconduct problems intersection ofconduct problems and poverty AnnuRev Clin Psychol 201410503ndash528

18 Duncan GJ Magnuson K Votruba-DrzalE Moving beyond correlations inassessing the consequences of povertyAnnu Rev Psychol 201768413ndash434

19 McEwen CA McEwen BS Socialstructure adversity toxic stress andintergenerational poverty an earlychildhood model Annu Rev Sociol 201743(1)445ndash472

20 Sharkey P Elwert F The legacy ofdisadvantage multigenerationalneighborhood effects on cognitiveability AJS 2011116(6)1934ndash1981

21 Chetty R Hendren N Katz LF The effectsof exposure to better neighborhoods onchildren new evidence from the movingto opportunity experiment Am EconRev 2016106(4)855ndash902

22 Oreopoulos P Stabile M Walld R RoosLL Short- medium- and long-termconsequences of poor infant health ananalysis using siblings and twinsJ Hum Resour 200843(1)88ndash138

23 OrsquoGrady K Deussing M Scerbina T FungK Muhe N Measuring Up CanadianResults of the OECD PISA Study TorontoCanada Council of Ministers ofEducation Canada 2016

24 Statistics Canada Focus on GeographySeries 2011 Ottawa Canada StatisticsCanada 2014 Available at httpswww12statcangccacensus-recensement2011as-safogs-spgIndex-engcfm Accessed September 252018

25 Statistics Canada Aboriginal PeoplesFact Sheet for Manitoba OttawaCanada Statistics Canada 2016Available at httpswww150statcangccan1pub89-656-x89-656-x2016008-enghtm Accessed September 25 2018

26 OrsquoGrady K Fung K Servage L Khan GReport on the Pan-CanadianAssessment of Reading Mathematicsand Science Toronto Canada Councilof Ministers of Education Canada 2018

27 Canadian Institute for HealthInformation Children Vulnerable inAreas of Early Development ADeterminant of Child Health OttawaCanada Canadian Institute for HealthInformation 2014 Available at httpssecurecihicafree_productsChildren_Vulnerable_in_Areas_of_Early_Development_ENpdf Accessed September 252018

28 Roos LL Gupta S Soodeen RA JebamaniL Data quality in an information-richenvironment Canada as an exampleCan J Aging 200524(suppl 1)153ndash170

29 Roos LL Nicol JP A research registryuses development and accuracy J ClinEpidemiol 199952(1)39ndash47

30 Government of Manitoba Employmentand Income Assistance (EIA) Availableat httpswwwgovmbcafseiaAccessed September 25 2018

31 Manitoba Centre for Health PolicyConcept Income Quintile RankingProcedure Winnipeg Canada ManitobaCentre for Health Policy 2002 httpmchp-appservcpeumanitobacaviewConceptphpconceptID=1164Accessed September 25 2018

32 Janus M Brinkman S Duku E et al TheEarly Development Instrument APopulation-Based Measure forCommunities A Handbook onDevelopment Properties and UseHamilton Canada Offord Centre forChild Studies 2007

33 Snow KL Measuring school readinessconceptual and practical

considerations Early Educ Dev 200617(1)7ndash41

34 Brownell M Nickel NC Chartier M et alAn unconditional prenatal incomesupplement reduces populationinequities in birth outcomes Health Aff(Millwood) 201837(3)447ndash455

35 Government of Manitoba Familyservices and housing Child ProtectionServices Available at httpswwwgovmbcafschildfamindexhtmlAccessed March 27 2019

36 Manitoba Centre for Health PolicyTerm apprehended at birth taken intoCFS care at birth Available at httpmchp-appservcpeumanitobacaviewDefinitionphpdefinitionID=104764Accessed January 10 2019

37 Alexander GR Kotelchuck M Assessingthe role and effectiveness of prenatalcare history challenges and directionsfor future research Public Health Rep2001116(4)306ndash316

38 SAS Institute Inc Statistical AnalysisSoftware (SAS) version 94 2013

39 Almond D Currie J Duque V Childhoodcircumstances and adult outcomes ActII Journal of Economic Literature 201856(4)1360ndash1446

40 Bronfenbrenner U The Ecology ofHuman Development Experiments byNature and Design Cambridge MAHarvard University Press 1979

41 Government of Canada A backgrounderon poverty in Canada 2016 Available athttpwww12edscgccasgpe-pmpsservletsgpp-pmps-publang=engampcurjsp=p5bd2t13ls-engjspampcuractn=dwnldamppid=54405ampdid=4879Accessed January 10 2019

42 Statistics Canada Census Profile 2016Census Ottawa Canada StatisticsCanada 2017 Available at httpswww12statcangccacensus-recensement2016dp-pdprofdetailspagecfmLang=EampGeo1=PRampCode1=46ampGeo2=PRampCode2=01ampData=CountampSearchText=ManitobaampSearchType=BeginsampSearchPR=01ampB1=IncomeampTABID=1 Accessed March 27 2019

43 Statistics Canada Canadian IncomeSurvey 2015 Ottawa Canada StatisticsCanada 2017 Available at httpswww150statcangccan1endaily-quotidien170526dq170526a-engpdfst=kFoEj4AM Accessed March 27 2019

10 ROOS et al by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

Drug Alcohol Depend 200892(1ndash3)191ndash199

12 Bradley RH Corwyn RF Socioeconomicstatus and child development Annu RevPsychol 200253371ndash399

13 Browne DT Wade M Prime H JenkinsJM School readiness amongst urbanCanadian families risk profiles andfamily mediation J Educ Psychol 2018110(1)133ndash146

14 Putnam-Hornstein E Needell BPredictors of child protective servicecontact between birth and age five anexamination of Californiarsquos 2002 birthcohort Child Youth Serv Rev 201133(11)2400ndash2407

15 Currie J Stabile M Manivong P Roos LLChild health and young adult outcomesJ Hum Resour 201045(3)517ndash548

16 Duncan GJ Dowsett CJ Claessens Aet al School readiness and laterachievement Dev Psychol 200743(6)1428ndash1446

17 Shaw DS Shelleby EC Early-startingconduct problems intersection ofconduct problems and poverty AnnuRev Clin Psychol 201410503ndash528

18 Duncan GJ Magnuson K Votruba-DrzalE Moving beyond correlations inassessing the consequences of povertyAnnu Rev Psychol 201768413ndash434

19 McEwen CA McEwen BS Socialstructure adversity toxic stress andintergenerational poverty an earlychildhood model Annu Rev Sociol 201743(1)445ndash472

20 Sharkey P Elwert F The legacy ofdisadvantage multigenerationalneighborhood effects on cognitiveability AJS 2011116(6)1934ndash1981

21 Chetty R Hendren N Katz LF The effectsof exposure to better neighborhoods onchildren new evidence from the movingto opportunity experiment Am EconRev 2016106(4)855ndash902

22 Oreopoulos P Stabile M Walld R RoosLL Short- medium- and long-termconsequences of poor infant health ananalysis using siblings and twinsJ Hum Resour 200843(1)88ndash138

23 OrsquoGrady K Deussing M Scerbina T FungK Muhe N Measuring Up CanadianResults of the OECD PISA Study TorontoCanada Council of Ministers ofEducation Canada 2016

24 Statistics Canada Focus on GeographySeries 2011 Ottawa Canada StatisticsCanada 2014 Available at httpswww12statcangccacensus-recensement2011as-safogs-spgIndex-engcfm Accessed September 252018

25 Statistics Canada Aboriginal PeoplesFact Sheet for Manitoba OttawaCanada Statistics Canada 2016Available at httpswww150statcangccan1pub89-656-x89-656-x2016008-enghtm Accessed September 25 2018

26 OrsquoGrady K Fung K Servage L Khan GReport on the Pan-CanadianAssessment of Reading Mathematicsand Science Toronto Canada Councilof Ministers of Education Canada 2018

27 Canadian Institute for HealthInformation Children Vulnerable inAreas of Early Development ADeterminant of Child Health OttawaCanada Canadian Institute for HealthInformation 2014 Available at httpssecurecihicafree_productsChildren_Vulnerable_in_Areas_of_Early_Development_ENpdf Accessed September 252018

28 Roos LL Gupta S Soodeen RA JebamaniL Data quality in an information-richenvironment Canada as an exampleCan J Aging 200524(suppl 1)153ndash170

29 Roos LL Nicol JP A research registryuses development and accuracy J ClinEpidemiol 199952(1)39ndash47

30 Government of Manitoba Employmentand Income Assistance (EIA) Availableat httpswwwgovmbcafseiaAccessed September 25 2018

31 Manitoba Centre for Health PolicyConcept Income Quintile RankingProcedure Winnipeg Canada ManitobaCentre for Health Policy 2002 httpmchp-appservcpeumanitobacaviewConceptphpconceptID=1164Accessed September 25 2018

32 Janus M Brinkman S Duku E et al TheEarly Development Instrument APopulation-Based Measure forCommunities A Handbook onDevelopment Properties and UseHamilton Canada Offord Centre forChild Studies 2007

33 Snow KL Measuring school readinessconceptual and practical

considerations Early Educ Dev 200617(1)7ndash41

34 Brownell M Nickel NC Chartier M et alAn unconditional prenatal incomesupplement reduces populationinequities in birth outcomes Health Aff(Millwood) 201837(3)447ndash455

35 Government of Manitoba Familyservices and housing Child ProtectionServices Available at httpswwwgovmbcafschildfamindexhtmlAccessed March 27 2019

36 Manitoba Centre for Health PolicyTerm apprehended at birth taken intoCFS care at birth Available at httpmchp-appservcpeumanitobacaviewDefinitionphpdefinitionID=104764Accessed January 10 2019

37 Alexander GR Kotelchuck M Assessingthe role and effectiveness of prenatalcare history challenges and directionsfor future research Public Health Rep2001116(4)306ndash316

38 SAS Institute Inc Statistical AnalysisSoftware (SAS) version 94 2013

39 Almond D Currie J Duque V Childhoodcircumstances and adult outcomes ActII Journal of Economic Literature 201856(4)1360ndash1446

40 Bronfenbrenner U The Ecology ofHuman Development Experiments byNature and Design Cambridge MAHarvard University Press 1979

41 Government of Canada A backgrounderon poverty in Canada 2016 Available athttpwww12edscgccasgpe-pmpsservletsgpp-pmps-publang=engampcurjsp=p5bd2t13ls-engjspampcuractn=dwnldamppid=54405ampdid=4879Accessed January 10 2019

42 Statistics Canada Census Profile 2016Census Ottawa Canada StatisticsCanada 2017 Available at httpswww12statcangccacensus-recensement2016dp-pdprofdetailspagecfmLang=EampGeo1=PRampCode1=46ampGeo2=PRampCode2=01ampData=CountampSearchText=ManitobaampSearchType=BeginsampSearchPR=01ampB1=IncomeampTABID=1 Accessed March 27 2019

43 Statistics Canada Canadian IncomeSurvey 2015 Ottawa Canada StatisticsCanada 2017 Available at httpswww150statcangccan1endaily-quotidien170526dq170526a-engpdfst=kFoEj4AM Accessed March 27 2019

10 ROOS et al by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

44 Ross NA Wolfson MC Dunn JRBerthelot J-M Kaplan GA Lynch JWRelation between income inequalityand mortality in Canada and in theUnited States cross sectionalassessment using census data andvital statistics BMJ 2000320(7239)898ndash902

45 UNICEF Child Poverty in Perspective AnOverview of Child Well-Being in RichCountries Florence Italy UNICEF 2007Available at httpswwwunicef-ircorgpublications445-child-poverty-in-perspective-an-overview-of-child-well-being-in-rich-countrieshtml AccessedMarch 27 2019

46 Wolfson M Murphy B Income inequalityin North America does the 49th parallelstill matter Can Econ Obs 200031ndash24

47 Ermisch J Jantti M Smeeding T eds FromParents to Children The IntergenerationalTransmission of Advantage New York NYRussell Sage Foundation 2012

PEDIATRICS Volume 143 number 6 June 2019 11 by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

DOI 101542peds2018-3426 originally published online May 20 2019 2019143Pediatrics

Leslie L Roos Elizabeth Wall-Wieler and Janelle Boram LeePoverty and Early Childhood Outcomes

ServicesUpdated Information amp

httppediatricsaappublicationsorgcontent1436e20183426including high resolution figures can be found at

Referenceshttppediatricsaappublicationsorgcontent1436e20183426BIBLThis article cites 29 articles 3 of which you can access for free at

Subspecialty Collections

ubhttpwwwaappublicationsorgcgicollectionpsychosocial_issues_sPsychosocial Issuesmilestones_subhttpwwwaappublicationsorgcgicollectiongrowthdevelopment_GrowthDevelopment Milestonesal_issues_subhttpwwwaappublicationsorgcgicollectiondevelopmentbehaviorDevelopmentalBehavioral Pediatricsfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpwwwaappublicationsorgsitemiscPermissionsxhtmlin its entirety can be found online at Information about reproducing this article in parts (figures tables) or

ReprintshttpwwwaappublicationsorgsitemiscreprintsxhtmlInformation about ordering reprints can be found online

by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from

DOI 101542peds2018-3426 originally published online May 20 2019 2019143Pediatrics

Leslie L Roos Elizabeth Wall-Wieler and Janelle Boram LeePoverty and Early Childhood Outcomes

httppediatricsaappublicationsorgcontent1436e20183426located on the World Wide Web at

The online version of this article along with updated information and services is

httppediatricsaappublicationsorgcontentsuppl20190517peds2018-3426DCSupplementalData Supplement at

by the American Academy of Pediatrics All rights reserved Print ISSN 1073-0397 the American Academy of Pediatrics 345 Park Avenue Itasca Illinois 60143 Copyright copy 2019has been published continuously since 1948 Pediatrics is owned published and trademarked by Pediatrics is the official journal of the American Academy of Pediatrics A monthly publication it

by guest on September 9 2021wwwaappublicationsorgnewsDownloaded from