Content area
The aim of this paper was to systematically review the published evidence on the relationship between the type of childcare and risk of childhood overweight or obesity. The databases PubMed, MEDLINE, Cochrane Library and EMBASE were searched using combinations of the various search terms to identify eligible observational studies published between 2000 and May 2016 in English. Fifteen publications from 7 countries matched the inclusion criteria. The most commonly reported childcare arrangements were centre-based (e.g. crèche) and informal care (e.g. relatives, neighbours, friends). Informal care was most frequently associated with an increased risk of childhood overweight and obesity. Associations were also found for other lifestyle variables such as low maternal education, high birth-weight, maternal employment, ethnicity, maternal overweight/obesity and father's Body Mass Index (BMI).
Conclusion: The relationship between childcare and childhood overweight/obesity is multi-faceted with many aspects linked to childhood adiposity, in particular the age of initiation to care, type of care (i.e. informal care) and shorter breastfeeding duration were related with infant adiposity.
[Table omitted.]
Eur J Pediatr (2016) 175:12771294 DOI 10.1007/s00431-016-2768-9
http://crossmark.crossref.org/dialog/?doi=10.1007/s00431-016-2768-9&domain=pdf
Web End = http://crossmark.crossref.org/dialog/?doi=10.1007/s00431-016-2768-9&domain=pdf
Web End = REVIEW
The association between childcare and risk of childhood overweight and obesity in children aged 5 years and under: a systematic review
Goiuri Alberdi1 & Aoife E. McNamara1 & Karen L. Lindsay1 & Helena A. Scully1 &
Mary H. Horan1 & Eileen R. Gibney2 & Fionnuala M. McAuliffe1
http://orcid.org/0000-0002-8861-137X
Web End = Received: 10 December 2015 /Revised: 15 August 2016 /Accepted: 19 August 2016 /Published online: 8 September 2016 # Springer-Verlag Berlin Heidelberg 2016
Abstract The aim of this paper was to systematically review the published evidence on the relationship between the type of childcare and risk of childhood overweight or obesity. The databases PubMed, MEDLINE, Cochrane Library and EMBASE were searched using combinations of the various search terms to identify eligible observational studies published between 2000 and May 2016 in English. Fifteen publications from 7 countries matched the inclusion criteria. The most commonly reported childcare arrangements were centre-based (e.g. crche) and informal care (e.g. relatives, neigh-bours, friends). Informal care was most frequently associated
with an increased risk of childhood overweight and obesity. Associations were also found for other lifestyle variables such as low maternal education, high birth-weight, maternal employment, ethnicity, maternal overweight/obesity and fathers Body Mass Index (BMI).
Conclusion: The relationship between childcare and childhood overweight/obesity is multi-faceted with many aspects linked to childhood adiposity, in particular the age of initiation to care, type of care (i.e. informal care) and shorter breastfeeding duration were related with infant adiposity.
What is known:
Lifestyle factors during early years affect health outcomes in adulthood, particularly in children with low birth weight.
Pre-school stage influences childrens body composition and growth. What is new:
This is the first systematic review of observational studies examining the association between childcare and childhood overweight and obesity in preschool children.
Informal care is linked to early introduction to solid foods, less physical activity and obesity.
Revisions received: 4 July 2016; 16 August 2016
Communicated by Mario Bianchetti
* Fionnuala M. McAuliffe [email protected]
Goiuri Alberdi [email protected]
Aoife E. McNamara [email protected]
Karen L. Lindsay [email protected]
Helena A. Scully [email protected]
Mary H. Horan [email protected]
Eileen R. Gibney [email protected]
1 UCD Obstetrics and Gynaecology, School of Medicine, University College Dublin, National Maternity Hospital, Dublin 2, Ireland
2 UCD Institute of Food and Health, University College Dublin, Belfield, Dublin 4, Ireland
Keywords Childcare . Childhood overweight . Childhood obesity . Adiposity . Review
AbbreviationsBMI Body Mass IndexOECD Organisation for Economic Co-operation and
DevelopmentSES Socioeconomic Status
1278 Eur J Pediatr (2016) 175:12771294
Introduction
The events that occur in early life are important in shaping how our bodies react to the situations we encounter in life. Langley-Evans et al. found that the quality of diet in early years has been linked with obesity, heart disease and type two diabetes mellitus [1]. Excess weight gain during infancy has been found to increase the risk of obesity during later childhood, which has a tendency to persist into adulthood, giving rise to various chronic diseases [2]. An Australian systematic review also showed that childhood obesity profoundly disturbs the psychological health affecting the quality of life from early age [3]. A high body weight and adiposity gain in the first two years of life has been shown to increase the risk of co-morbidities such as hypertension, sleep apnoea, hyperlipidaemia, asthma and type two diabetes mellitus in childhood as well as in adulthood [4]. A large weight gain is particularly harmful to children with a low birth weight as it is directly associated with a higher body mass index (BMI) and central obesity in childhood, as well as being associated with the same co-morbidities as mentioned above [2].
The environment we are exposed to as young children, and especially the adult-child relationships including childcare, has a role in shaping a childs cognitive, emotional, physical and social behaviour [5]. Women worldwide pursue full-time careers; therefore, the traditional role of the mother representing the primary caregiver is changing. As children spend less time in their mothers care, alternative carers begin to have a strong influence on childrens habits, especially their diet and physical activities [6, 7]. There are many different kinds of childcare as outlined in Fig. 1. Childcare at infant/preschool stage in a childs life may have the potential to influence body composition and growth. For example, a care setting may dictate hours spent engaging in physical activity [8] or may promote sedentary activities that predispose to the development of obesity. In a nationwide survey conducted in Ireland, involving 500 children aged one to four years old took part, parents cited Bother people minding their
children^ as one of the greatest barriers to providing a healthy diet for their child [9].
According to Pearce et al., approximately 50 % of three to six year olds and 25 % of infants under three years old within the Organisation for Economic Co-operation and Development (OECD) countries were in early childcare. The same authors stated that almost 25 % of preschool children in the UK were obese [10]. With the childhood obesity epidemic on the increase it is important to understand the influence childcare settings can have on childrens dietary and physical activity behaviours. Childcare settings have significant potential to promote healthy eating and exercise among children through education and policies. This may help combat the growing prevalence of childhood obesity and obesity-related comorbidities later in life. Hence, this systematic review aims to examine the literature which reports on the relationship between childcare and childhood overweight and obesity, which may aid the implementation of regulations and policies for childcare settings, in order to give children a positive and healthy attitude towards nutrition and physical activity from an early stage in life.
Materials and methods
This is a systematic review of observational studies conducted in accordance with the Preferred Reporting Items for Systematic review and Meta-analysis (PRISMA) guidelines [11].
Study selection
Observational studies (cohort, case-control and cross-sectional) were included regardless of publication status, sample size or follow-up duration. The main exposure variable was childcare and the primary outcome was prevalence or risk of childhood overweight and/or obesity. Thus, studies were included if they reported on both of these variables. Longitudinal studies were included provided they reported on childhood overweight or obesity on at least one time-point at 5 years of age or under. Studies that did not report on attendance in childcare services or any form of non-parental care were excluded.
Data sources and search strategy
The electronic bibliographic databases PubMed, MEDLINE, Cochrane Library and EMBASE, as well as the search tool Google Scholar, were searched by two researchers (AEM & KLL) in May 2016 for eligible studies. The search was restricted to studies published in the English language and between the years 2000 and 2016. It was deemed feasible to omit
Childcare
Fig. 1 Childcare flow diagram
Eur J Pediatr (2016) 175:12771294 1279
studies published prior to 2000 as we wished to evaluate the association between attendance in childcare and the more recent rise in the childhood obesity epidemic [12]. The search strategy for potentially eligible studies included the following terms and combinations: [Grandparent OR relative OR childcare OR child care OR day-care OR Preschool] AND [Childhood] AND [overweight OR obesity OR adiposity]. The reference lists of included studies were also searched for additional eligible studies.
Data extraction
Citations identified from the electronic database search were organised and duplicates deleted. The titles and abstracts were independently screened by two reviewers (AEM & KLL) to determine eligibility for inclusion. Where the title and abstract were insufficient to determine eligibility, the full texts were retrieved and evaluated. Any discrepancies that arose between reviewers were resolved by consensus and consultation with FMcA. Information extracted from each article included: publication year, study design, country, sample size, child age range, childcare setting (s), socioeconomic status (SES) and parental BMI of participants (if available), criteria used to define child overweight/obesity; primary and secondary study outcomes and confounding factors. Where studies reported on both childhood overweight and obesity status, separate prevalence rates for each category were obtained where possible. Odds ratios (ORs) were recorded to determine risk of overweight/obesity associated with childcare attendance, with preference given to the adjusted ORs where available.
Quality assessment
The methodological quality of included studies was independently assessed by the two reviewers conducting the literature search using Downs and Black criteria [13]. The original 27-item checklist examines the reporting, external validity, bias and confounding of randomised and non-randomised studies in health care. Seven questions which specifically relate to randomised trials (i.e. reporting adverse events as a consequence of an intervention, blinding of subjects to intervention assignment, blinding of researchers to intervention assignment, reliable compliance to the intervention, randomisation to intervention assignment, concealment of randomised assignment until recruitment complete) were excluded from the quality assessment for the current systematic review of observational studies. The final score for each study was divided by the total eligible score and reported as a percentage. Where disagreement arose between the researchers on the quality scores of studies, a re-assessment was performed together to reach consensus. Due to the heterogeneous nature of included studies, meta-analysis were not performed.
Results
Figure 2 describes the search technique of the literature, which resulted in the inclusion of 15 publications [4, 7, 10, 1425]. Table 1 outlines the characteristics of included studies. The studies sourced were conducted in seven countries, where 11 out of the 15 studies had large sample sizes of over 1000 subjects, 12 of the studies compared more than one type of childcare and the ages of the children included varied from birth up to 6 years. Parental BMI was reported in 11 of the studies and 10 of them specified socioeconomic status levels of participants. The results of the review are presented in Table 2.
Types of childcare and childs body weight
The most commonly reported childcare arrangement was informal care, including relative or non-relative care (neighbours, grandparents, friends), with percentages of attendance to this type of care ranging between 6.2 % and 46.6 % [4, 7, 10, 16, 17, 19, 22, 24, 25].
Some studies showed that the hours spent in any type of childcare were associated with increased risk of childhood overweight/obesity [4, 7, 10, 16], while others found no association between hours spent, or even the years spent, in childcare and weight status [18, 20, 25]. Regarding centre-based care, three studies found an association with overweight/obesity status [14, 16, 17] whereas others found it to be protective [15, 18]. Several studies did not find any association between centre-based care and obesity/overweight status [4, 6, 10, 1921, 25].
Compared to parental care, children who attended a centre-based childcare had higher odds of being overweight or obese during childhood. However, the greatest association between overweight/obesity status and type of childcare arrangement involved attendance in informal care, or care by relatives and non-relatives. Concerning childcare by relatives, six studies found an association with overweight/obesity in children, particularly focusing on the impact of grandparental care [6, 10, 15, 16, 22, 23]. Three studies underlined the role of grandmothers for increasing the odds of a child being overweight. Tanskanen et al. [22] reported an odds ratio (OR) of 1.20 (CI:1.031.40) while Watanabe et al. [23] reported an OR of 1.59 (CI: 1.082.35). Moreover, Pearce et al. [10] found that the increased risk of being overweight was only significant in children who were cared for by grandparents [OR: 1.18; CI:1.051.32] when compared with other informal care settings. From a non-relative informal care perspective,
Tanskanen et al. [22] did not find any association, while others found increased odds of childhood overweight/ obesity [4, 7, 10, 23]. McLaren et al. found that non-relative informal care was associated with higher odds of being overweight/obese but only among boys, while
1280 Eur J Pediatr (2016) 175:12771294
Fig. 2 Flow chart of studies selection process
no association was observed in the female cohort [19]. Another two studies reported no associations with any form of childcare and childhood weight status [21, 25].
Confounding factors associated with the effect of childcare in childs body weight
The most commonly reported confounding factor was maternal overweight/obesity status [15, 17, 18, 21], as well as fathers BMI [17, 21]. Maternal education was also a significant predictor of overweight and obesity among children attending child care [15, 17, 18, 21, 24], along with maternal employment [10, 23], and socioeconomically disadvantaged households [7, 24]. High birth-weight was also considered an important confounding factor [7, 15, 21]. Flores and Lin found that having an unmarried mother increased childs weight [15], while Watanabe et al. found no association between single parents and childhood overweight/obesity [23].
Ethnicity, childcare and childs body weight
There is some evidence that ethnicity may mediate the association of childcare and obesity. Flores and Lin report that
being of African American or from a multiracial extraction could increase the risk for overweight and obesity in children [15]. Wolfenden et al. found an increased likelihood of being overweight or obese in indigenous children, however, after adjusting for age, childcare service hours and other demographic covariates, the association disappeared [24]. Zahir et al. showed that in a cohort of Latino ethnic subjects, neither childcare initiation age nor the number of hours spent in childcare per week were associated with childhood over-weight [25] which defends the protective effect argued by other authors [18]. The Latino ethnic group was most frequently reported to have an association between childcare and overweight/obesity [7, 15], although a protective effect in children attending the Head Start programmes in America was observed, which provides early childhood education, health, nutrition and parental involvement services to low-income families [7, 18]. Flores & Lin also reported a protective effect of attendance in center-based childcare, with more than three times lower odds for developing severe obesity in attenders vs non-attenders [15]. These findings concur with those of Maher et al. who reported that obese children were significantly more likely to be in family, friend or neighbour care [7].
Eur J Pediatr (2016) 175:12771294 1281
Definitionof
obesity
CDCWFLgrowth
charts.
95th percentileon
CDCBMI-for-
agegrowth
charts.
IOTFAge-and
Sex-specificcut-
offs
Childrenclassified
asoverweight/
obese(85th
percentile)using
WHOcut-off
points
CDCWFLand
BMI-for-age
growthcharts.
Confoundingfactors
considered
Childage,genderandrace/
ethnicity,maternalpre-
pregnancyBMI,smokingin
pregnancyandparity;
paternalBMI;household
income.Breastfeeding
duration,sleepdurationand
televisionviewing
consideredpotential
mediators.
birthweight,no.weeks
premature,mobility
impairment,leveland
frequencyofphysical
activity;familyincome;
numberofchildrenyounger
than18yearsold;ageofthe
youngestchild;andmothers
education,employment,and
maritalstatus.
overweight,ethnicity,ageat
firstlivebirth,smoking
statusinpregnancy;child
birthweightz-score;no.of
childrenlivinginthe
household
Mothersage,smokingstatus,
mothersheightandweight,
educationlevelandannual
householdincome,
breastfeedingactivity,child
age,birthweightandbirth
length.
Various:totalof32parental,21
prenatal/pregnancy,32
infant,and46early
childhoodcharacteristics
considered.
Birthweight,maternalsmoking
duringpregnancy,
Riskofobesitychildsage,race,gender,
RiskofoverweightMaternalpre-pregnancy
Primarystudy
Outcome(s)
Weight-for-lengthz-
scoreat1yearand
BMIz-scoreat
3years
BMIz-scoreat
12months
Predictorsofsevere
obesity
Prevalenceof
overweight/obesity
5yearsCentre-basedAttendancein
exposure
variable
childcare
frombirthto
6months
childcarefor
10hours/
weekwithin
theyear
before
Kindergarten
childcarefor
10hours/
week
Primary
Hoursin
Attendancein
Monthsin
childcare
Multiple
prenatal,
infantand
early
childhood
factors
childcare
Childcaresetting(s)
investigated
1and3yearsCentre-based,someone
elseshome,childs
ownhome
3yearsInformal(relative,non-
relativecarei.e.
27,821Denmark12monthsDay-carehome,crche,
andage-integrated
facility
1,649Canada4yearsPreschool
babysitter),Formal
(Centre-based,
nanny,au-pair)
Non-relative,Centre-
based
anthropometric
examination
Kindergarten
entry,
approx.5-6
years
CountryChildageat
1138United
15,691United
States
States
Kingdom
12,354United
6800United
States
Table1Descriptivestatisticsofincludedstudies
PublicationStudyDesignCohort
size
Prospective
cohort
Longitudinal,
nationally
representative
cohort
Prospective
cohort
Longitudinal
cohort
study
Longitudinal,
nationally
representative
cohort
Prospective
cohort
Benjamin
etal.
(2009)
[4]
Maheretal.
(2008)
[7]
Pearceetal.
(2010)
[10]
Benjaminet
al.
(2015)
[14]
Floresand
Lin,
(2013)
[15]
Geoffroy
etal.
1282 Eur J Pediatr (2016) 175:12771294
Definitionof
obesity
IOTFAge-and
Sex-specificcut-
offs.
Overweightdefined
asBMIz-score
85th centile
usingtheFourth
Dutchnational
GrowthStudy
1997growth
referencecharts.
95th percentileon
CDCBMI-for-
agegrowth
charts.
Atriskofobesity
85th percentile
onCDCBMI-
for-agegrowth
charts
Atriskofobesity
85th percentile
onCDCBMI-
for-agegrowth
charts
Confoundingfactors
considered
breastfeedingdurationby
1.5years,maternalBMI,
childsethnicity,familySES,
maternalemployment,
maternaldepressive
symptoms,family
functioningandmaternal
over-protection.
Numberofhoursof
employmentandoverweight
ofmotherandfather;
maternaleducationallevel,
ageandcountryofbirth;
parentalalternativelifestyle;
childsbirthweight;breast-
feedingduration.
maternaleducation,maternal
occupation,amountoftime
thechildspentwatchingTV
orvideosonausualdayand
thequalityofthechilds
homeenvironment.
RiskofobesityGender,race,maternalBMI,
Incomeadequacy,highest
householdeducational
attainment,no.ofsiblings,
no.ofparentsinthe
household,birthweight,
mothersageatbirth,
provinceofresidence,urban/
ruralresidence.
Childgenderandethnicity;
familyincome;no.ofparents
inthehousehold;maternal
depressionorsensitivity;
controllingparenting
characteristics;reportedand
measuredactivityandtime
spentinactiveorsedentary
pursuits;neighbourhood
safety.
Primarystudy
Outcome(s)
BMIz-scoreat1and
2years;changein
BMIz-score
between1and
2years;changefrom
non-overweightto
overweightbetween
1and2years.
ChangeinBMI
percentilefromage
2-3toage6-7years
ChangesinBMI
betweenagesof2
and12years
exposure
variable
arrangement
atages1.5,
2.5,3.5and
4years
Attendancein
formal
childcare
Numberof
weekinany
childcare;
attendancein
preschool
4days/week
vs.<4days/
week
childcarefor
10hours/
week
centre-based
childcare
Primary
hoursper
Attendancein
between6
and
54months
Childcaresetting(s)
investigated
Centre-based,family-
based,relative,
nanny
(outsideofhomeby
non-relative).
relative,inoroutside
ofownhome;
Centre-based.
Centre-basedAttendancein
2,396Netherlands1and2yearsFormalChildcareonly
3-4yearsRelative&non-
anthropometric
examination
CountryChildageat
3,564Canada2-3yearsRelative,Non-relative,
Centre-based
24,36and
54months
United
States
States
Case-control556;age-
matched
normal
(N=296)
vs.obese
(N=260)
960United
PublicationStudyDesignCohort
size
Prospective
cohort
Longitudinal,
nationally
representative
cohort
Longitudinal
prospective
cohort
Table1(continued)
(2013)
[16]
Gubbelset
al.
(2010)
[17]
Koleilatet
al.
(2012)
[18]
McLarenet
al.
(2012)
[19]
OBrienet
al.
(2007)
[20]
Eur J Pediatr (2016) 175:12771294 1283
NoneCDCBMI-for-age
growthcharts.
CDC,CentreforDiseaseControl;BMI,BodyMassIndex;SES,Socio-EconomicStatus;WFL,Weight-for-Length;IOTF,InternationalObesityTaskForce;WHO,WorldHealthOrganisation;NCHS,
NationalCentreforHealthStatistics
Definitionof
obesity
Overweight>90th
andobese>97th
centileaccording
tonationalage-
andsex-specific
BMIgrowth
charts.
IOTFAge-and
Sex-specificcut-
offs
IOTFAge-and
Sex-specificcut-
offs
IOTFAge-and
Sex-specificcut-
offs
Confoundingfactors
considered
AveragedailydurationofTV-
viewingandcomputer
playing,breastfeeding
duration,maternaleducation
status,maternalsmokingin
pregnancy,parentalweight
andheight.
Birthweight;no.ofsiblingsin
household;ethnicity;
maternalsocioeconomic
status,smokingduring
pregnancy,pre-pregnancy
weightstatusandcountryof
origin
Childage,sexandbirthweight;
no.ofsiblings;parental
weightstatus;maternal
employmentstatus;typeof
familyhousehold
Childsex,ageandIndigenous
status;householdincome;
maternaleducation;hours/
weekspentinchild-care
services
Primarystudy
Outcome(s)
Riskofoverweightat
age6yearsand
changeinBMI
between4and
6years
Prevalenceof
overweightor
obesity
Prevalenceandoddsof
overweightor
obesity
Prevalenceof
overweightand
obesity
Prevalenceof
overweight/obesity
Centre-basedAttendancein
childcare)for
minimum
exposure
variable
kindergarten
(German
formal
4hours/day
Attendancein
childcare
between
9monthsand
3years
employment
status;family
household
Maternal
environmen-
talfactors;
childrens
lifestyle
factors
Primary
Attendancein
formal
childcare
Hoursspentin
childcare
Childcaresetting(s)
investigated
3yearsInformal(grandparent,
otherrelative,non-
relative)orformal
4yearsoldInformal(non-centre-
based),Formal
grandparentalcare
outsideofthisformal
childcareisbeing
investigated
(preschoolorcentre-
based)
CountryChildageat
Cross-sectional1,765Japan3-6yearsChildrenrecruitedfrom
Nurseryschoolsand
Kindergartensbut
informal
Cross-sectional764Australia2-5yearsFormal(preschoolor
longday-care
centres)
anthropometric
examination
(retrospec-
tivedata
availableat
4yearsalso)
2,140Germany6years
Kingdom
States
9,000United
201United
PublicationStudyDesignCohort
size
Prospective
cohort
Longitudinal
prospective
cohort
Longitudinal
prospective
cohort
Table1(continued)
Rappetal.
(2005)
[21]
Tanskanen,
2013
[22]
Watanabe
etal.
(2011)
[23]
Wolfenden
etal.
(2011)
[24]
Zahiretal.
(2013)
[25]
1284 Eur J Pediatr (2016) 175:12771294
Maternal
Education
childcare
vs.in
childcare:
high
school
graduate
11.7
vs.6.8%;
some
college
23.8vs.
20.3%;
college
graduate
36.7vs.
36.8%;
graduate
degree27.9
vs.
36.2%
22.6%3rd
level
degree
NotReportedNotin
employment;21.3%
inpart-time
employment;32.3%
not-employed
Maternal
Employment
Notreported46.3%infull-time
childcare
vs.in
childcare:
Mean(SD)
pre-
pregnancy
BMI24.7
(5.4)vs.
25.1
(5.6)
EthnicityMaternal
BMI
Notin
childcare
vs.in
childcare:
Black14.1
vs.15.3%;
White66.5
vs.66.6%;
Hispanic
3.9
vs.5.2%;
Other15.5
vs.
12.9%
Notin
NotReportedNotinchildcare
vs.in
childcare:
breastfeeding
durationat
1year,
mean(SD),
Feeding
Practises
6.7(4.7)
vs.5.8(4.4)
months
Mean7.38lbsNotreportedWhite59.3%;
Latino
18.8%;
Black
14.7%
Birth
Weight
Child
gender
childca
revs.in
childca
re:49.5
vs.
49.6%
female
Notin
49.4%
female
Associationof
childcareand
Ow/Ob.
Onmultivariable
adjusted
regression,
each10h
incrementper
week
ofcarein
someone
elseshomewas
positively
associatedwith
WFLat1year
(co-
efficient0.11;
95%
CI0.010.20;
p=0.02)and
BMI
at3years(0.12;
0.020.21;
p=0.02).No
associations
with
centre-based
careor
careinchilds
ownhome.
Significantlymore
obesechildren
amongthose
receiving
informal
care(20.2vs.
16.5%,
p<0.01).
Onmultivariable
adjusted
regression,
greateroddsof
obesity
associated
withinformal
care
inownhome
(OR:
1.33,CI:1.12
1.57,
p<0.01)and
Head
Startcare
attendance(OR:
1.31,CI:1.06
Timespentin
Childcare
Mean(SD):12.4
(9.0)inany
childcare,
11.3
(8.5)in
centre
basedcare,
11.7
(8.3)
someone
elseshome,
9.6
(9.4)childs
ownhome.
Allchildren
classifiedas
in
childcare
were
inattendance
for10h/
week
Childcare
Attendance
57%ofinfants
attendedany
childcare.Of
these,17%in
centre
based;27%
someoneelses
home;21%
childsown
home.
33%notin
regular
childcare;
16.9%
informalcare,at
least
occasionally
inchildsown
home(relative
or
non-relative);
8.4%informal
careoutside
own
home;9.2%
HeadStart
program;
32.7%
formalcentre-
basedcare.
Table2Resultsofthereview
overweightor
obesity/z-
scores
Notinchildcare
vs.in
childcare:
Mean(SD)
WFLz-score
at1year0.27
(1.01)vs.0.39
(1.05).Mean
(SD)BMIz
scoreat
3years0.32
(0.97)vs.0.51
(1.03).
11.6%obeseAt
thestartof
Kindergarten
11.6%were
obese
PublicationPrevalenceof
Benjaminetal.
(2009)[4]
Maheretal.
(2008)[7]
Eur J Pediatr (2016) 175:12771294 1285
Maternal
Education
55.6%2nd
level;27%
3rdlevel
Notreported
Notreported15.3%none;
Employment
Notreported,though,
childrenassisting
childcarecamefrom
familieswithhigher
householdincomes
Maternal
28.6%pre-
pregnancy
overweight/
obese
Mean(SD)
pre-
pregnancy
BMI23.5
(4.0)and
23.4
(4.3)
EthnicityMaternal
BMI
NotreportedInchildcarevs.
notin
childcare:
88.8%White;
2.6%
Black;
4.2%
Pakistani;
1.9%
Indian;
2.5%
Other
Feeding
30.4%never
breastfed;
36.7%
breastfed<
4 months;
32.8
breastfed
4months
carewereless
oftenbreastfed
for22weeks
(24.4%vs.
35.1
Practises
Childreninchild
%;
p< 0.0001)
Weight
6 %low
(<2.5kg),
92.3%
normal
(2.5
4.5kg),
1.8%
high
(>4.5kg)
Mean(SD):In
childcare
3597.3
(544.6);
Not
in
childcare
3593.0
(590.3)
Birth
gender
reported
children
inchild
carewere
female
(48.4%
vs.
49.8%;
p= 0.
001)
Child
Not
Fewer
Associationof
childcareand
Ow/Ob.
1.60,
p< 0.05).
Pearceetal.
Onadjusted
Poisson
regression
analysis,
informal
childcare
associatedwith
increasedriskof
overweight/
obesityat
3years(RR
1.15,
CI:1.041.27,
p <0.05).
Significantrisk
existsby
grandparents
(RR
1.18,CI:1.05
1.32)butnotby
otherinformal
carers(RR1.15,
CI:0.971.37).
No
associationwith
formalchildcare.
Onmultivariable
adjusted
regression,any
childcareuse
was
associatedwith
0.03units
higherBMIz-
scoreat
12monthsfor
each
additional
30days
ofcare(95%
CI:
0.01,0.05).Each
childcaretype
were
associatedwitha
higherBMIz-
scoreat
12months:Care
in
aday-carehome
(0.03units;
95%
Timespentin
Childcare
Allchildren
classifiedas
in
childcare
were
inattendance
for10h/
week
Mean(SD):2.6
(2.5)months
in
aday-care
home,0.4
(1.2)
inacrche
and
1.5(1.4)inan
age-
integrated
facility
some
spenttimein
morethanone
typeofchild
careoverthe
courseofthe
12months.
Childcare
Attendance
Parent55.7%;
Informal
22.5%
(ofwhich,
Grandparent
care
17.6%;non-
grandparent
care
4.9%);formal
21.9%
63.7%children
attended
childcare
atsomepoint
duringtheirfirst
12monthsof
life
PublicationPrevalenceof
overweightor
obesity/z-
scores
overweight,
5.1%obese
Overweight,
85-94th
percentile:In
childcare
2397(13.5);
notin
childcare
1334(13.2).
Obese,95th
percentile:In
childcare
1859(10.5);
notin
childcare998
(9.9)
Table2(continued)
18%
(2010)[10]
Benjaminetal.
(2015)[14]
1286 Eur J Pediatr (2016) 175:12771294
NotReportedNotreported
Maternal
Education
NotReported
62.1%highly
educated
(completed
2nd
levelorat
Employment
15.7%inany
employmentupto
timechildaged
4years
Mean(SD)hours/week:
18.6(12.0)
Maternal
Notseverely
obesevs.
severely
obese:
Maternal
pre-
pregnancy
severe
obesity
(BMI
40kg/m2 )
2.6vs.
9.3%
EthnicityMaternal
BMI
(p<0.01)
Mean(SD):
23.64(4.6)
whenchild
aged
1.5years
overweight
(BMI
25kg/m2 )
31.6%
Notseverely
obesevs.
severely
obese
(p<0.01):
White55.2
vs.35.5%;
Latino23.9
vs.43.1%;
African-
American
14.0vs.
13.8%
91.2%white,
8.8%
others
mothers
bornin
Netherlands
Practises
Notseverely
obesevs.
severely
obese:Mean
(SD)agefirst
fedformula
3.7(0.1)
vs.2.5(0.2)
months
(p<0.01).
Mean
(SD)age
when
4.5(0.04)vs.
4.1
(0.1)months
(p=0.04).
Notreported96.5%
Feeding
solids
introduced
Mean(SD)
duration
of
breastfeeding:
3.79(4.58)
months
obese:
Mean(SD)
3.30
(1.10)vs.
3.48(3.37)
Kg
(p< 0.01)
Weight
Notseverely
obesevs.
severely
normal
(2.54.0
Kg);
3.5%low
(<2.5
Kg);
11.4%
high
(>4.0Kg)
3526(505)
g
Birth
85.2%
Mean(SD)
gender
severely
obese
vs.
severely
obese:
50.7vs.
58.8%
male
(p= 0.
02)
reported
Child
female
Not
Not
48.5%
Associationof
childcareand
Ow/Ob.
NotreportedProvisionofnon-
relative
childcare
>10h/weekPrevalenceof
CI:0.01,0.05);
crche
(0.05units;
95%CI:0.01,
0.10);andage
integrated
facility
(0.08units;
95%
CI:0.04,0.12).
FloresandLin
(2013)[15]
wassignificantly
higheramong
non
severelyobese
children
(P<0.04).On
multivariable
regression
analyses,ever
attendingcentre-
basedcarewas
protective
against
severeOb(OR:
0.3;95%CI:
0.1
0.9).
overweight/
obesity
atage4didnot
differbyprimary
childcare
arrangement
between1.5and
4years
(p= 0.326),or
at
eachagepointof
examination
(1.5,
2.5,3.5,4years
of
age).
Onadjusted
multivariate
logistic
regression
analysis,
significant
Timespentin
Childcare
Mean(SD)
hours/
weekat
1year20.1
(7.4);at
Childcare
Attendance
Non-relative
childcarein
childsown
home:
15.8%among
not
severelyobese
children;3.7%
amongseverely
obese
Mainchildcare
arrangement
between1.5
and
4years:29.9%
Centre
based,46.6%
Relative,4.3%
Non-
relative,19.2%
Parental
26.2%attended
childcareat
each
stageoffollow
up
(7months,
overweightor
obesity/z-
scores
5.7%(N= 400)
severely
obese
(BMI>99th
percentile);
94.3%
(N=6400)
PublicationPrevalenceof
notseverely
obese.
Overweight/
obeseat
4yearsold
12.9%
Overweightat
1year
15.2%,at
2years
14.4%
Table2(continued)
Geoffroyetal.
(2013)[16]
Gubbelsetal.
(2010)[17]
Eur J Pediatr (2016) 175:12771294 1287
NotreportedNotreportedNotreportedNotreportedFamiliesof
Maternal
27.52%inemployment83.63%high
school
education
orless
Education
leastsome
3rdlevel)
55.3%
girlsand
57.2%
boyshad
complete
3rdlevel
education
Maternal
Employment
Mean(SD)
30.99(6.86)
EthnicityMaternal
BMI
Hispanic
Feeding
Practises
NotreportedNotreported96.03%
birthweight
Birth
Weight
92.1%girls
and
94.9%
boys
normal
Child
gender
49.82%
female
49.4%
female
Associationof
childcareand
Ow/Ob.
association
between
overweight
status
at1yearwith
childcare
attendanceat
7months(OR:
1.32,CI:1.04
1.69,p< 0.05)
but
notat1year
(OR:
1.23,CI:0.97
1.57).No
association
between
childcare
attendanceat
any
stageand
overweightat
2years.
Onadjusted
multivariate
logistic
regression
analysis,lower
oddsof
overweight
associatedwith
attendancein
preschool4or
moredays/week.
Noassociation
withhoursspent
in
otherformof
childcare.
Directassociation
between
childcare
attendanceand
weightstatusat
age
23yearsnot
reported.In
adjusted
ordinaryleast
squares
regression,non
relativecareof
boyspositively
Timespentin
Childcare
2years19.5
(8.0)
Mean(SD)
hours/weekin
non-
preschool
childcare6.07
(13.94).31%
attend
preschool4or
more
days/week
Allchildren
classifiedas
in
childcare
were
inattendance
for
10h/week
Childcare
Attendance
2years&
3years).
Attendanceat
1year33.2%;
at
2years38.6%
20%totalsample
attended
childcare
(preschoolor
othernon-
formal
care)
25.5%girlsand
28.8%boysin
non-relative
care;
13.7%girls
and
13.4%boysin
relativecare;
11.9%girls
and
9.0%boysin
centre-based
care;
57.0%girls
overweightor
obesity/z-
scores
46.8%obese
(cases),
53.2%
normal
weight
(controls)
45.8%girlsand
47.4%boys
atriskof
overweight/
obesityat
2/3yearsof
age
Table2(continued)
PublicationPrevalenceof
Koleilatetal.
(2012)[18]
McLarenetal.
(2012)[19]
1288 Eur J Pediatr (2016) 175:12771294
NotreportedNotreportedNotreported
NotreportedMothersof
NotreportedNotreported
Maternal
Education
German
children:
32.6%
low,
36.0%
medium,
47.7%
high.
Mothersof
non-
German
children:
52.8%
low,
24.7%
medium,
22.4%
high
Maternal
Employment
EthnicityMaternal
BMI
Mean(SD),
mothersof
German
children:
22.7
(3.7);of
non-
German
children:
24.3
(4.2)
Overweight
pre-
pregnancy
29.7%
NotreportedNotreportedNotreportedThosewho
werenever
Owwere
lesslikely
to
benon-
white(OR:
0.88).
69%German,
31%Other
Practises
Germanchildren:
<
2months
exclusive
breastfeeding
27%,
26months
57%,
>6months
33%.
Non-German
children:
<2months
exclusive
breastfeeding
33%,
26months
47%,
>6months
19%
Feeding
Mean3.4KgNotreported87.6%White;
2.2%
Mixed;
2.6%
Weight
3.99kg):
Girls
87%,boys
80%.
High
weight
(4 kg):
Girls8%,
boys
13%].
German
ethnicity
[Normal
weight
(2.5
Non-
German
ethnicity,
[Normal
weight:
girls90%,
boys
81%.
High
weight:
girls
6.6%,
boys
12.6%].
Birth
gender
German
ethnicity;
47.1%
female
among
non-
47.6%
female
among
German
ethnicity
reported
Child
Not
Associationof
childcareand
Ow/Ob.
associatedwitha
0.1unitincrease
in
BMIpercentile
changebyage
6
7years(CI:
0.02
0.18,p<0.05).
Noassociation
detectedamong
girls.
OBrienetal.
attendedcentre
basedcarethan
everOw(OR:
0.70)
Noassociation
betweentypeof
Kindergarten
childcare
arrangementand
riskof
overweight/
obesityatage
6yearsamong
Germanornon-
German
children,
orwithchange
in
BMIbetween
ages
4and6years.
associatedwith
NotreportedNotreportedNeverOwchildren
lesslikelyto
have
NotreportedPrimarycareby
maternal
grandmother
Timespentin
Childcare
Morningonly,
4 h/
day;morning
andafternoon
withno
lunch,
6 h/day;
morningto
early
afternoon,
5.5h/
day;full-time
care,upto
10h/
day
Childcare
Attendance
and
56.0%boys
not
inchildcare
22%attended
Kindergartenin
morningonly;
13%attended
morningand
afternoonwith
luncheatenat
home;44%
attended
morning
throughtoearly
afternoonwith
snacksonly
provided;15%
attendedfull-
timewithlunch
provided
62.2%parental
care;
12.8%
maternal
overweightor
obesity/z-
scores
Overweightat
24months
15%;
36months
18%;
54months
25%
Atage4years:
German
ethnicity,
Overweight:
Girls7.5%,
boys5.8%;
Obese:girls
1.4%,boys
1.5%.Non-
German
ethnicity,
Overweight:
girls13.1%,
boys16.0%;
Obese:girls
4.4%,boys
6.6%
overweight
Table2(continued)
PublicationPrevalenceof
23.6%
(2007)[20]
Rappetal.
(2005)[21]
Tanskanen
(2013)[22]
Eur J Pediatr (2016) 175:12771294 1289
Maternal
Education
Notreported
employment
NotreportedNotreported36%with3rd
level
education
Maternal
Employment
75.5%in
Overweight/
obesityin
motheronly
5.4%;in
bothparents
69.2%
EthnicityMaternal
BMI
Indian;
4.7%
Pakistanior
Bangladesh
i;1.6%
Black;
1.3%
Other
Japanese)
Indigenous
Australian;
89%non-
indigenous
NotreportedNotreported
(assumedall
Feeding
Practises
NotreportedNotreported11%
Girls
3.06(0.41)
Kg
Birth
Mean(SD)
Boys3.17
(0.42)Kg;
Weight
Child
gender
48.4%
female
female
50%
NotreportedAfteradjustingfor
maternal
employment
status
andother
Associationof
childcareand
Ow/Ob.
overweight/
obesity
atage3years
(OR:
1.20,CI:1.03
1.40,
p=0.019).
Noassociations
withothercare-
givers.
confounders,
livingin3
generation
households(i.e.
witha
grandparent)
wassignificantly
associatedwith
child
overweight/
obesity(OR:
1.59,
CI:1.082.35)
Asallrecruited
from
formalchildcare
centres,no
associations
with
individual
childcaretypes
couldbe
examined.
Adjustingfor
hoursspentin
childcare,
maternal
educationlevel
(2ndlevel/
diploma)was
only
significant
predictorof
child
overweight/
obesity
(OR:2.06,
CI:1.163.66)
Timespentin
Childcare
Mean(SD)17.4
(9.6)hours/
week
Childcare
Attendance
grandmother;
5.3%paternal
grandmother;
2.9%other
relative;1%
non-relative
informal;
15.7%formal.
Allchildren
attending
formal
childcare
facilities
(e.g.
nursery).52%
livingin
householdwith
grandparent(s);
48%living
withparent(s)
only
Allchildren
attending
formalcentre-
basedcare
PublicationPrevalenceof
overweightor
obesity/z-
scores
Overweight/
obesityin
Boys8.4%;
Girls9.9%
Overweight
amongtotal
sample
12.7%;boys
11.8%;girls
13.6%.
Obeseamong
total4.0%;
boys3.8%;
girls4.2%
Table2(continued)
Watanabeetal.
(2011)[23]
Wolfendenetal.
(2011)[24]
1290 Eur J Pediatr (2016) 175:12771294
BMI,BodyMassIndex;WFL,Weight-for-Length;mo,monthsold;Ow,Overweight;Ob,Obesity;OR,OddsRatio;95%CI,ConfidenceInterval;P,Probability;RR,RiskRatio;WHO,WorldHealth
Organisation;IOTF,InternationalObesityTaskForce;
Maternal
Education
76%high
school
diplomaor
less
Maternal
Employment
71.6%
unem
ployed
Mean(SD)
25.8
(5.5)Kg/m2
EthnicityMaternal
BMI
Mexican;
38.8%
Central
American
Feeding
Practises
NotreportedNotreported61.2%
Birth
Weight
Child
gender
reported
Not
Associationof
childcareand
Ow/Ob.
Noassociation
between
childcare
attendanceatage
4yearsand
weightstatus:
47.7%of
overweight/
obese
inchildcarevs.
52.3%normal
weight;p= 0.98
forOwand0.56
forOb
Nodifference
betweenweight
statusgroups
and
dailyhoursin
childcare
(p= 0.68)or
ageat
commencing
childcare
attendance
(p= 0.82).
Timespentin
Childcare
Mean(SD)6.4
(2.1)hours/
day
or31.2(11.3)
hours/week
Childcare
Attendance
71.1%in
childcare
(98.5%of
which
wascentre-
based
care)
overweightor
obesity/z-
scores
PublicationPrevalenceof
overweight;
24.9%obese.
47.3%
Table2(continued)
Zahiretal.
(2013)[25]
Eur J Pediatr (2016) 175:12771294 1291
Breastfeeding, childcare and childs body weight
Children attending childcare were breastfed for shorter durations [4]. A protective effect of breastfeeding against overweight/obesity has been reported [4, 21], while shorter durations of breastfeeding increased the risk [15, 17]. On the contrary, other studies found breastfeeding did not mediate an association [10, 17]. Flores and Lin found that early introduction to solids, normally occurring with non-parental childcare, was associated with overweight and severe obesity [15].
Quality index
The mean scores obtained from the assessment of the methodological quality of each of the studies included in this review are shown in Table 3. Two of the studies scored less than 60 %, nine studies scored between 60 and 80 % and the remaining four had the highest scores of more than 80 %.
Discussion
This review found evidence for an association between childcare and childhood overweight and obesity. The childcare environment may promote excessive weight gain in early childhood, because childcare providers may be less likely than parents to encourage healthful nutrition and physical activity behaviours [4]. Informal care (relative and non-relative) seems to be the most risky environment regarding obesogenic childcare. Many informal carers may lack childcare qualifications and are less exposed to child health advice [26]. It has been hypothesised that the increased risk in
obesity associated with relative care may be largely attributed to the care of grandparents, both full and part-time [22, 23]. Grandparents tend to mind children in their own home which is less, if at all, regulated by public authority [16]. They tend to Bspoil^ young children more than parents, giving them more palatable and calorific foods [6, 16, 26, 27], leading to inappropriate feeding practises [4]. Grandparents often provide group care for several grandchildren, being responsible for feeding a number of age groups with different feeding requirements. This may lead to encouraging infants to be advanced onto solids too early in order to make mealtimes easier for grandparents [4, 14]. Kim et al. found that infant care by a relative, compared to parental care, was associated with early introduction of solids before 4 months of age and increased infant weight gain at 9 months of age [2]. Group care may also lead to less individual interaction between the carer and the child, as well as increased periods of restricted movement, i.e. baby walker. This means less physical activity for the child, leading to increased risk of obesity [4, 28]. Physical inactivity is particularly common in the care of grandparents as they may not be physically capable of keeping up with children and may promote more sedentary activities such as TV viewing [7, 16]. Long screen times, along with short sleep durations, have been associated with obesity in preschool children [28]. On the contrary, Benjamin et al. stated that a nanny caring for a child in the childs own home was more likely to adhere to parental feeding and physical activity preferences than if care was provided in someone elses home [4].
Centre-based care is a regulated form of childcare provided by qualified staff and therefore there is potential to encourage healthy eating and physical activity, such as described in the Head Start programmes. A potential explanation for the protective relationship between centre-based childcare and risk for obesity is the decreased exposure to obesogenic risk factors found at home and the increase in healthy behaviours for those at risk for severe obesity [15]. Differences in the quality or irregularities in the implementation regulation of childcare centres in regard to nutrition and physical activities might explain variability in results [6, 16]. Nevertheless, Maher et al. stated that infrequent childcare (<10 h/week) was unlikely to significantly influence the childs development [7].
On the other hand, studies showing an association between childcare attendance and overweight and obesity occurrence have been found to occur in sample populations where mothers had a high education, hence the protective effect of formal childcare might be lost when mothers have a high education background [17]. The children of managerial/ professional mothers were more likely to be obese than those whose mothers either did not work or work at an intermediate level, possibly due to longer hours spent in care [5].
Parental weight status, particularly maternal weight status has a strong effect on childhood overweight and obesity,
Table 3 Overall mean scores of quality assessment
Publication Mean Total score/21 Mean % Score
Benjamin et al. (2009) [4] 17.5 83.3 Maher et al. (2008) [7] 15.5 73.3 Pearce et al. (2010) [10] 15.5 73.3 Benjamin et al. (2015) [14] 16 76.2 Flores & Lin (2013) [15] 13.5 64.3 Geoffroy et al. (2013) [16] 19 90.5 Gubbels et al. (2010) [17] 16 76.2 Koleilat et al. (2011) [18] 14.5 69.0 McLaren et al. (2012) [19] 18.5 88.1 OBrien et al. (2007) [20] 14.5 69.0 Rapp et al. (2005) [21] 12.5 59.5 Tanskanen AO (2013) [22] 16 76.2 Watanabe et al. (2011) [23] 16 76.2 Wolfenden et al. (2010) [24] 19 9.05 Zahir et al. (2012) [25] 10.5 50.0
1292 Eur J Pediatr (2016) 175:12771294
highlighting the possible effect of genetics and environmental factors on childhood weight status. The fact that parental overweight/obesity was the most commonly reported confounder of childhood obesity is unsurprising, as if the parents are leading unhealthy lifestyles, so will too their children, following nutritionally deficient diets with little or no promotion of physical activity [15, 17, 18, 21].
The higher obesity risk among the Latino ethnic group may be due to the fact that these studies involved low-income populations unable to afford high quality childcare and may have been fed cheap, calorific ready-meals instead of home-prepared nutritious meals [2, 7, 15]. However, the slight protective effect of family, friend and neighbour and centre-based care on obesity among the Latino children in these studies may be attributed to the likelihood of traditional Latino food being prepared in these households [7]. Publication bias may also explain why Latinos appeared to be the ethnic group most at risk, as five out of the 15 included studies investigated the associations in this subgroup specifically. The remaining studies either did not have this subgroup present or did not break their sample down into ethnic groups. Zahir et al. [25] and Koleilat et al. [18] had the highest proportion of Hispanic subjects with 100 % and 96 % respectively, Flores and Lin [15] had almost a quarter and Maher et al. [7] and Benjamin et al. [4] had 19 % and 5 %, respectively.
The length of time spent breastfeeding is directly associated with the introduction of solids to the babys diet [2]. Kim et al. observed that children who entered full-time childcare younger than three months were less likely to have been breastfed and 1.73 times more likely to have been introduced to solid foods too early [2]. The association between childcare attendance and shorter breastfeeding durations [4, 15, 17] could be due to the separation of mother and child during care periods, automatically reducing available time for breastfeeding.
Possible limitations of this review include heterogeneity due to a wide range of countries and cultures under study, making it difficult to synthesise data. In line with that, the exclusion of non-English articles could limit the number of studies on childcare, as informal childcare is more prevalent in developing countries. The same can be said for the use of several different diagnostic criteria for overweight and obesity. Additionally, some of the studies had a large time lag between the time when subjects would have been the appropriate age for this review and the time of publishing, likely attributed to the long follow-up periods involved in these longitudinal studies. This could mean that the obesity rates at the time studies were conducted are different to present day figures as well as differences in society. Many of the studies collected data retrospectively or relied solely on parental-report, which leaves the studies prone to recall bias and inaccuracies of data. Furthermore, a few studies did not collect data on diet or physical activity both in and out of childcare, indicating that
the overweight or obesity at this age could be due to other effector variables such as feeding practises and physical environment at home. Furthermore, some papers did not report on other potential effector variables such as parental BMI. In some of the studies there were certain subgroups that were under-represented, with very small numbers of ethnic minorities or small or none low SES numbers. Many of the studies did not report on all types of childcare and only examined main childcare arrangement. Finally, due to the cross-sectional nature of two of the studies included in this review, causality of a relationship between childcare and overweight and obesity could not be determined [23, 24]. It should be noted that the methods used for classification of overweight or obesity varied. Most studies used global standards but others used more locally/nationally specific guidelines to determine weight status, this meant that the prevalence of over-weight and obesity in each study could not be directly compared.
Despite these limitations, this review also has several strengths. To the best of our knowledge, this is the first systematic review of observational studies examining the association between childcare and childhood over-weight and obesity in preschool children. Secondly, the broad database search returned a wide selection of published studies with a variety of data on different aspects of childcare. Thirdly, the overall score [13] provided for study quality of reporting, internal validity (bias and confounding factors) and also external validity (generalisability) shows that the studies included in the present review are methodologically strong. Fourthly, publications from a number of countries were returned and many of the studies reviewed were nationally-representative [29]. Finally, the majority of the studies were longitudinal, and thus the follow-up interview or measurements allowed a more comprehensive analysis of the reported associations between childcare and obesity.
Conclusions
This review highlights the existence of an important relationship between childcare and the childs body composition. All the studies including relative care in their research concluded that this type of care is linked to a higher BMI. BInformal^ care, most frequently provided by grandparents, is linked to early introduction to solid foods and less physical activity. This systematic review has shown that age of initiation to care, type of care, hours spent in care, mothers education and occupation and parental overweight/obesity and breastfeeding are all positively associated with weight gain and adiposity among children. Further research is
Eur J Pediatr (2016) 175:12771294 1293
required to examine some of these relationships in more detail, which may provide an opportunity to improve regulations and policies in the field of childcare (i.e. family-focused behavioural strategies among relatives caring for children), aiming to tackle the growing burden of childhood obesity, as lifestyle habits are established in early life.
Acknowledgments This study was supported by the Health Research Board Ireland, the Health Research Centre for diet, and Health research and the European Unions Seventh Framework Programme (FP7/2007-2013), project Early Nutrition under grant agreement no. 289346.
Author contributions AEM conducted the literature search, completed data collection and interpretation and wrote the manuscript. GA and KLL assisted with the literature search, data interpretation and writing of the manuscript. HAS and MHH assisted with writing of the paper. ERG and FMM conceived the idea for this project and approved the final draft.
Compliance with ethical standards
Conflicts of interest Goiuri Alberdi declares that she has no conflict of interest. Aoife E. McNamara declares that she has no conflict of interest. Karen L. Lindsay declares that she has no conflict of interest. Helena Scully declares that she has no conflict of interest. Mary H. Horan declares that she has no conflict of interest. Eileen R. Gibney declares that she has no conflict of interest. Fionnuala M. McAuliffe declares that she has no conflict of interest. Goiuri Alberdi, Karen L. Lindsay and Mary H. Horan were funded by European Unions Seventh Framework Programme (FP7/20072013).
Ethical approval This article does not contain any studies with human participants or animals performed by any of the authors.
References
1. S. C. Langley-Evans, Nutrition in early life and the programming of adult disease: a review, J Hum Nutr Diet, vol. 28 Suppl 1, pp. 114 2015.
2. Kim J, Peterson KE (2008) Association of infant child care with infant feeding practices and weight gain among US infants. Arch Pediatr Adolesc Med 162:627633
3. Sanders RH, Han A, Baker JS, Cobley S (2015) Childhood obesity and its physical and psychological co-morbidities: a systematic review of Australian children and adolescents. Eur J Pediatr 174:715 746
4. Benjamin SE, Rifas-Shiman SL, Taveras EM, Haines J, Finkelstein J, Kleinman K, et al. (2009) Early child care and adiposity at ages 1 and 3 years. Pediatrics 124:555562
5. Growing-Up-Ireland ST (2011) Key findings: infant cohort (at 9 months) no. 3 childcare and parenting support , ed. Dublin, Ireland, Office of the Minister for Children and Youth Affairs
6. Lin SL, Leung GM, Hui LL, Lam TH, Schooling CM (2011) Is informal child care associated with childhood obesity? Evidence from Hong Kongs children of 1997 birth cohort. Int J Epidemiol 40:12381246
7. Maher EJ, Li G, Carter L, Johnson DB (2008) Preschool child care participation and obesity at the start of kindergarten. Pediatrics 122: 322330
8. Takahashi E, Yoshida K, Sugimori H, Miyakawa M, Izuno T, Yamagami T, et al. (1999) Influence factors on the development of obesity in 3-year-old children based on the Toyama study. Prev Med 28:293296
9. Walton J (2012) National Pre-School Nutrition Survey. Summary report on: food and nutrient intakes, physical measurements and barriers to healthy eating, ed. Irish University Nutrition Alliance, Dublin
10. Pearce A, Li L, Abbas J, Ferguson B, Graham H, Law C, et al. (2010) Is childcare associated with the risk of overweight and obesity in the early years? Findings from the UK millennium cohort study. Int J Obes 34:11601168
11. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gtzsche PC, Ioannidis JP, et al. (2009) The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med 6:e1000100
12. de Onis M, Blssner M, Borghi E (2010) Global prevalence and trends of overweight and obesity among preschool children. Am J Clin Nutr 92:12571264
13. Downs SH, Black N (1998) The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health 52:377384
14. Benjamin Neelon SE, Schou Andersen C, Schmidt Morgen C, Kamper-Jrgensen M, Oken E, Gillman MW, et al. (2015) Early child care and obesity at 12 months of age in the Danish National Birth Cohort. Int J Obes 39:3338
15. Flores G, Lin H (2013) Factors predicting severe childhood obesity in kindergarteners. Int J Obes 37:3139
16. M. C. Geoffroy, C. Power, E. Touchette, L. Dubois, M. Boivin, J. R. Sguin, et al., Childcare and overweight or obesity over 10 years of follow-up, J Pediatr, vol. 162, pp. 753758.e (1), 2013.
17. Gubbels JS, Kremers SP, Stafleu A, Dagnelie PC, de Vries NK, van Buuren S, et al. (2010) Child-care use and the association with body mass index and overweight in children from 7 months to 2 years of age. Int J Obes 34:14801486
18. Koleilat M, Harrison GG, Whaley S, McGregor S, Jenks E, Afifi A (2012) Preschool enrollment is associated with lower odds of childhood obesity among WIC participants in LA County. Matern Child Health J 16:706712
19. McLaren L, Zarrabi M, Dutton DJ, Auld MC, Emery JC (2012) Child care: implications for overweight / obesity in Canadian children? Chronic Dis Inj Can 33:111
20. OBrien M, Nader PR, Houts RM, Bradley R, Friedman SL, Belsky J, et al. (2007) The ecology of childhood overweight: a 12-year longitudinal analysis. Int J Obes 31:14691478
21. Rapp K, Schick KH, Bode H, Weiland SK (2005) Type of kindergarten and other potential determinants of overweight in pre-school children. Public Health Nutr 8:642649
22. Tanskanen AO (2013) The association between grandmaternal investment and early years overweight in the UK. Evol Psychol 11: 417425
23. Watanabe E, Lee JS, Kawakubo K (2011) Associations of maternal employment and three-generation families with pre-school childrens overweight and obesity in Japan. Int J Obes 35:945952
24. Wolfenden L, Hardy LL, Wiggers J, Milat AJ, Bell C, Sutherland R (2011) Prevalence and socio-demographic associations of over-weight and obesity among children attending child-care services in rural and regional Australia. Nutr Diet 68:1520
25. Zahir N, Heyman MB, Wojcicki JM (2013) No association between childcare and obesity at age 4 in low-income Latino children. Pediatr Obes 8:e24e28
1294 Eur J Pediatr (2016) 175:12771294
26. Kuczmarski RJ, Ogden CL, Guo SS, Grummer-Strawn LM, Flegal KM, Mei Z, et al. (2002) 2000 CDC growth charts for the United States: methods and development. Vital Health Stat 11:1190
27. Garza C, de Onis M (2004) Rationale for developing a new international growth reference. Food Nutr Bull 25:S514
28. Sijtsma A, Koller M, Sauer PJ, Corpeleijn E (2015) Television, sleep, outdoor play and BMI in young children: the GECKO Drenthe cohort. Eur J Pediatr 174:631639
29. CASP-NHS, Critical Appraisal Skills Programme, ed. Oxford: Public Health Resource Unit, NHS, 2003.
Springer-Verlag Berlin Heidelberg 2016