prevalence of and risk factors for childhood overweight and obesity

7/30/2019 Prevalence of and Risk Factors for Childhood Overweight and Obesity

http://slidepdf.com/reader/full/prevalence-of-and-risk-factors-for-childhood-overweight-and-obesity 1/8

Research

Recherche

Prevalence of and risk factors for childhood

overweight and obesity

Paul |, Veug elers, Ang ela L. Fitzge rald

Abstract

Background: Increases in childhood overweight and obesity have

become an important public health problem in industrialized

nations. Preventive public health action is required, but more

research of risk factors is required before evidence-based ini-

tiatives can be developed and targeted effectively. We investi-

gated the association between childho od overweight and obe-

sity and risk factors relating to dietary habits, actitivities,

[parents and schoo ls.

Methods: In 2003 we surveyed grade 5 students and their parents

and school principals in Nova Scotia. We measured height

and weight and assessed dietary habits (using Harvard's

Youth/Adolescent Food Frequency Questionnaire), physical

and sedentary artivities, and parental and school-based risk

factors. We estimated neighbourhood income by averaging,

per school, the postal-code level means of household income

of residential addresses of children attending that school. We

used multilevel logistic regression to evaluate the significance

of these risk factors for overw eight and obesity.

Results: On the basis of measurements taken of 4298 grade 5 stu-

dents, we estimated the provincial prevalence of overweight to

be 32.9% and of obesity to be 9.9%. Children who bought

lunch at school were at increased risk of ovepA'eight (fully ad-

justed odds ratio [OR] 1.39, 95% confidence interval [CH

1.16-1.67), whereas those who ate supper together with their

family 3 or more times a week were at decreased risk (OR

0.68, 95% CI 0.52-0.88). Physical education classes 2 or more

times a week at school were associated with a decreased risk of

overweight (OR 0 .61 , 95% CI 0.43-0.87) and obesity (OR

0.54, 95% CI 0.33-0.88), Children in high-income neighbour-

hoods were half as likely to be obese as their peers living in

low-income neighbourhoods (OR 0.50, 95% CI 0.36-0.70).

Interpretation: Parents and schools provide important opportuni-

ties for public health initiatives for reducing childhood over-

weight and obesity. Children and schools in low-income

neighbourhoods should receive priority in public health initia-

tives to reduce future socioeconomic inequalities in health.

CM/^y2005;173(6):607-13

I ncreases in childhood overweight and obesity have be-

come a major public healtJi problem in industrialized

nations.'- In Canada, rates of overweight and obesity

among children have more than doubled in past decades,

with the most recent estimates indicating that about 30%of children are overweight or obese.'"* The numerous psy-

chosocial, physical and economic consequences of over-

weight and obesity are well-known. Childhood overweight

affects self-esteem and has negative consequences on cog-

nitive and social development."' Conditions such as type 2

diabetes mellitus, hypertension and bypercholesterolemia,

which were previously seen primarily in adults, are hecoin-

ing more common among children as the prevalence ofobesity increases.' Because childhood overweight often per-

sists into adulthood, a rising numher of adults will be at in-

creased risk of these conditions as well as of cardiovascular

disease, osteo arth ritis and certa in tj'pes of cancer."" As a

whole, the obesity epidemic constitutes a substantial de-

crease in quality of life and life expectancy and accounts for

billions of dollars in health care spending.'""

Insufficient physical activity and poor nutrition are

widely acknowledged as the primary mechanisms underly-

ing the rise in excess body weight.'' Recent studies have de-

scribed geographic and socioeconomic gradients and iden-

tified aspects of children's lifestyle, including physical and

sedentary activities, as risk factors for overweight and obe-

sity. "̂ "̂ M os t re cently , Sw inbu rn and colleagu es identified

risk factors related to parents, family and school as poten-

tially significant and as requiring more evidence so that evi-

dence-based policy and program decision-making for the

prevention of excess weight can be developed.'' The need

for further research is underlined by the fact that a number

of schools and jurisdictions in Canada are already imple-

menting policies and programs without evidence of their

effectiveness.

In 2003, we conducted a comprehensive survey of risk

fectors for childhood overweight and obesity with grade 5

students and their parents and school principals in NovaScotia, a province with particular weight and health con-

cerns among both children and adults."-'- ' Here, we evalu-

ate the prevalence of overweight and obesity and examine

the contribution of dietary habits, activities and sociodemo-

graphic and school-based fac tors to weight problems

among children.

Methods

The 2003 Children's Lifestyle and School-performance Study

(CLASS) was a survey of grade 5 students in Nova Scotia on

health, nutrition and lifestyle factors. The procedures used in thisstudy are briefly described below and in more detail elsewhere.-"**^'



Veugelers and Fitzgerald

W e measured standing height to the nearest 0.1 cm after students

had removed their shoes and body weight to the nearest 0.1 kg nn

calibrated digital scales. Overweight and ohesity were defined us-

ing the international body mass index cut-off points established

for children and youth.^" Th ese cut-off points are based on hea lth-

related adult definitions of ovenveight (> 25 Icg/nv') and obesity

(> 30 Icg/in') but are adjusted to specific age and sex categories for

children.-'^

TTie CLA 5S sur\'ey included a modified version of Ha rvard's

Youth/Adolescent Food Frequency Questionnaire,-' which gath-

ers information un both dietary intake and habits pertaining to

mealtime behaviours. It also included validated questions on the

frequency of physical activities and the niunber of hours of seden-

tary activities (watching tele\Tsion, working on a computer, play-

ing video games) taken from the National Longitudinal Survey of

Childr en and Youth (NLS CY ). '" Information on sociode mo-

graphic factors was taken from a survey completed hy parents and

included tbe child's sex, place of birtb and rcsidenc\-\ as well as the

parents' marital status, income level and educational attainment.

Tlie children's ages were not included, since the vast majority ofgrade 5 students were either 10 or 11 years old at the time they

completed the C^L^VSS survey.

W e es t imated ne ighbo urho od incom e by averag ing , per

school, the postal-code level means of household income (avail-

able through the 2001 Canada census) of residential addresses

children attending tliat school. Finally, we developed and admi

istered a short .survey for school principals to collect infomiatio

on school characteristics, including sales of soft drinks, presen

of vending machines, type of food services, frequency of physic

education classes and possible financial restraints for recreatio

and gymnasium equipment.

Because oiu- observations of stud ents are nested within those

their schools, we assessed student and school-based factors at di

tinct levels using multilevel statistical methods. Specifically, st

dent and parental factors w^ere considered as first-level covariate

.Ml school characteristics and neigh bourhoo d income were co

sidered a.s contextual factors and treate d as second-level covar

ates. We first applied multilevel logistic regression methods

determine to what extent risk factors were associated with ove

weight o r oliesity. Second, we grouped each risk factor into 1 of

group s — dietary hab its, activities, and socio deniog raphic an

scbool-hased (actors — and then adjusted for all significant ris

factors within each group. These estimates are referred to

theme-adjusted odds ratios. Tliird, we considered all significarisk lectors simultaneously to quantify their independent impo

tance for excess body weight.

Beeause participation rates were slightly lower in residenti

areas with lower estimates of household income, we calculated r

Table 1: Dietary h abit-related risk factors for overweigbt among grade 5 students in NovaScotia

Risk factor

Breakfast

Usually eats

Does not eat

Lunch

Bring from home

Eat at home

Buy at school

Does not eat

Family supper

< once/wk

1-2 times/wk

3 ^ t imes/wk

> 5 tinifs/wk

Supper in front oftelevision

< once/wk

1-2 times/wk

3-4 timea'wk

> 5 times/wk

Eat at fast foodrestaurant

< once/wk

1-2 times/wk

> 3 times/wk

No, of

students

n = 4209

4054

155

n = 4061

2547

647

805

62

n = 4242

507

643

575

2517

n ^ 4256

1876

1369

478

533

n = 4241

2241

1835

165

Weighted

prevalence,' %

96.3

3.7

61.7

17.5

19.3

1.5

12.0

15.3

13,5

59.2

44.2

31.9

11.4

12.5

52.8

43.3

3.9

Weighted odds ratio [957Q confidence interval)"

1 Unadjusted

1

1.50(1.06-2.13)

1

0.94(0.77-1.14)

1.47(1,23-1.76)

1.19(0.69-2.05)

1

0.81 (0.63-1.03)

0.67 (0.52-0.87)

0.69 (0.57-0.84)

1

1.07(0.93-1.23)

1.22 (1.00-1.50)

1.44(1.18-1.74)

1

1.02 (0.90-1.16)

0.86(0.57-1.12)

Theme-adjustedt

-

-

1

0.93(0.77-1.13)

1,43(1,19-1.71)

1,10(0.64-1.89)

1

0,83(0,65-1.06)

0.71 (0.54-0.94)

0,78(0.63-0.97)

1

1,06(0.92-1.22)

1,14(0.93-1.41)

1,26(1.01-1.28)

-

-

-

Fully adjusted^

-

-

1

0.99(0.82-1.19)

1.39(1.16-1.67)

1.08(0.63-1.86)

1

0.81 (0.63-1.04)

0.68 (0.52-0.88)

0.75(0,61-0.91)

-

-

-

-

-

-

-

•Wffjghieil (tif p)ciiirps|K>nse biiis lo tc fl a i provinc ial esiimaie*,.

tThemp-jtijusted otit b ralios are adjusled for lunch, family supper and supjwr in iron! of the (elevision,

^Fully adjitsteci odds ratios ate arljusted (or signrtkani risk faaurs in all -I iheme griiujw, includ ing lunch and family supper In dietary hablls.panicipatior in physic.il iii:riviries in ariivities 'Table 21, [wrenul eriiicaiion and neigi^bourhood income in sociodemographic factors (Table 3),and frequpficy of physicfll edut ation rl.is&es in school-based faciors (Table A) .



Childhood overweight and obesity

sponse weights to overcome potential nonresponse bias. On the

basis of average household incomes according to postal code data

from the 2001 census for hoth participant and nonpardcipants,

we calculated response rates per decile of household incomes hy

postal code. These response rates were subsequently converted

into response weights. Because we weighted all prevalence and

risk estimates for nonresponse bias, they represent provincial pop-

ulation estimates for grade 5 students. W e considered missing val-

ues as separate covariate categories, but we do not present their

estimated values.

This study, including data collection and parental informed

consent forms, was approved by the health sciences human re-

search ethics board of Dalhousie University.

Results

Of tbe 291 public schools in Nova Scotia witb grade 5

classes, 282 (96.9%) participated in the study. Parental con-

sent was received for 5517 students, whicb provided an av-erage response rate of 51.1% per school. One of the 7

provincial school boards did not allow measurements of

height and weight. Students without height and weight

measurements were excluded from the analyses, leaving a

sample of 4298 children from 242 schools.

The prevalence of overweight among grade 5 students

in Nova Scot ia in 2003 was 32. 9% , with 9.9% bein g

obese. The prevalence of overweight was about the same

for girls (32.9%) and boys (33.0%), whereas the preva-

lence of obes i ty was lower among gir ls (9.0%) than

among boys (10.9%).

Amo ng grade 5 studen ts, 3.7% did not eat breakfast

(Table 1). In a univariate analysis, these students were 1.5

times (or 50%) more likely to be overweight than stu-dents who usually ate breakfast. Missing lunch was also

associated with an increased risk of excess body weight,

hut this risk was not statistically significant. Relative to

those bringing their lunch from home, children buying

luncb at school were 47% more likely to be overweight

(unadjusted odds ratio [OR] 1.47, 95% confidence inter-

val [CI] 1.23-1.76). Increasing frequencies of eating sup-

per together at home (family supper) and decreasing fre-

quencies of eating supper in front of the television were

associated with a decreased risk of overweight. Because

dietary babits are interrelated, we considered these habits

s imultaneously and found tha t lunch pa t te rn and f re -quency of family supper and supper in front of the televi-

sion were determinants of body weight (Table I: tbeme-

adjusted OR s). After adjustin g for all significant risk

factors across all 4 tbeme groups, only buying lunch at

school significantly increased the risk of overweight, and

family suppers both 3 ^ and 5 or mo re times a week sig-

nificantly d ecreased th e risk (Ta ble 1: fully adjusted O Rs ).

Sedentary activity of mo re than 1 ho ur per day was as-

Table 2: Activity-related risk factors for ove rweight among grade 5 students in

Risk factor

Participation in

physical activities

< twice/wk

> 2 and < 4 times/wk

> 4 and < 7 times/wk

> 7 limes/wk

Participation in

sedentary activities^

< 1 hour/d

> 1 and < 3 hr/d

> 3 and < 6 hr/d

> 6 hr/d

Travel to and frmn

school

Walk or bike

Driven < 15 min

Driven > 15 and< 30 min

Driven > 30 min

No. ofstudents

n = 4280

568

990

2289

433

n = 4270

452

1760

1210

848

n ^ 4000

1224

1580

852

344

Weightedprevalence,* "/I

13.5

23.2

53.2

10.1

10.5

41.0

28.5

20.0

31.7

39.0

28.8

8,5

Nova Scotia

Weighted odds ratio [95% confidence interval)*

) Unadjusted

1

0.87(0.71-1,06)

0.85(0.70-1.02)

0.68 (0.52-0.89)

1

1.28(1.02-1.61)

1.32 (1,04-1.70)

1.37(1,06-1.77)

1

1.15(0.96-1.37)

1,20(0,97-1,49)

1,39(1.05-1,85)

Theme-adjustedt

1

0,87(0,71-1,06)

0,86(0.71-1.03)

0.70(0.54-0.91)

1

1.28(1.02-1.61)

1.30(1.02-1.65)

1.35(1.04-1.74}

1

1.15 (0.96-1.37)

1.20(0.97-1,46)

1.38(1.04-1.84)

Fully adjusted^

1

0,93(0,76-1.15)

0,89(0,70-1,08)

0.74 (0,56-0,97)

-

-

-

-

—

-

-

-

"Weighted for nonresponse hias to reflecl provinc ial estimates.

t Theme-ad usted odds ratios are adjusted for p articipaiion in physical activities, participation in sedenrary activities, and travel to and fromschool,

tFull y adjusted odds ratios are adjusted for significant risk factors in all A theme groups, including lunch and family supper in dietary habils

(Table 1), participation in physical attivilies in activities, parental education and neighbourhood income in sociodemographic factors (Table 31.and frequency of physical education classes in school-based factors (Table 4).tSedentary activities includ e w atching television, using a computer and playing video games.




sociated with a significantly increased risk of overweight,as was being driven to school longer than 30 minutes(Tab le 2: theme-adjusted OR.s). Participa ting in physicalactivity more than 7 times a week was associated with adecreased risk of overweight (Table 2: theme-adjusted

ORs). When considered in conjunction with other deter-minants, frequency of physical activity appears to be theonly activity-related factor independently associated withoverweight. However, sedentary activity was indepen-dently associated with overweight when physical activitywas not considered (data not shown), which indicates acorrelation between these 2 covariates.

With respect to sociodemographic factors, childrenwhose parents had attained higher levels of educadon andhad an income over $60 000 were at a decreased risk of

overweight, as were children who resided in urban areaand who resided in neighbourhoods where the income wain the middle or highest one-third (Table 3).

Of all children, 29.3% attended schools where lunchewere provided by a foodservice company. This may in

clude on-site catering or fast food delivery. Children attending such schools were 12% more likely to be overweight than children who attended schools where nfoodservice companies were used, but this difference wanot statisdcaliy significant (Table 4). Also, no substandaor statistically significant differences were observeamong children attending schools where lunches werdistributed by staff and volunteers or as part of a prepailunch program.

Of all children, 15.3% attended schools where so

Table 3: Sociodemographic risk factors for overweight among grade 5 students in Nova Sco t ia

Nn ni

Risk facto r students

Sex n = 4298

Female 2180

Male 2118

Place of birth n = 4002

Nova Scotia 3508

Other 494

Parents' marita l status n = 396J

Married or commonlaw 3307

Separated or divorced 459

Single or wid ow ed 195

Parental education n = 3985

Completed secondaryschoo i or less 1179

Completedcomm unity college 1502

Com pleted university 934

Completed graduateuniversity 370

Parental income, $ n = 3324

< 20 UOO 33 9

20 000-4 0 000 736

40 000-6 0 000 884

> 60 000 1365

Neighbourhood income§

Lowest one-lhird

Midd le one-third -

Highest one-lhird

Residence location n = 4298

Rural 1700

Urban 2598

Weightedprevalence,*

50.5

49.5

87.6

12.4

83.0

U. 7

5.3

30.1

37.6

23.1

9,2

10.8

22.5

26.4

40.3

-

-

-

38.3

61.7

Weighted odds ratio (95% confidence interval)

% Unadjusted

1

1.01 (0.90-1,13)

1

0.94(0.77-1.15)

1

1.21 (1.00-1.47)

M l (0.83-1.49)

1

0.87(0.73-1.03)

0.67 (0.55-0.82)

0.64 (0.49-0.84)

1

0.94(0.71-1.24)

0.83(0.64-1.07)

0.62 (0.48-0,80)

1

0.78 (0.64-0.99)

0.71 (0.57-0.87)

1

0.77(0,65-0.91)

Them e-adjustedt Fully adjusted^

_

-

-

_

t 1

0.91 (0.76-1,08) 0.90 (0.76-1,07)

0.76 (0.61-0.94) 0,73 (0.60-0.89)

0.75 (0.57-O.99) 0.73 (0.56 -0.96)

1

0.97(0.74-1.29)

0.90(0.69-1,18)

0.73 (0.56-0.95)

1 1

0.84 (0.69-1.03) 0.80 (0.66-0.98)

0.82 (0,66-1,00) 0.76 (0.62-0,95)

-

•Weighted for nonresponse bia? ro teitect provincial eilimales.tThemo-ailjuMed ochls ratios are> .idjusletl for parental eduration, |)arpnlal income and neighbourhood income.

IFully adjuKU^ odcf). ralios are adjusied for iignifirant risk lAcloti in .ill 4 theme groups, including lunch and family mppet in dietary habits iTable1), participation in physical activities in activities iTable 2), p,irenial education and neighbourhood income in sociodemogriiphic factors, and

frequency o l physical ed ucation classes in school-based factors (Table 4).^Npifjhb ourhoo d income was estimated by averaging, pei school, thp poslal-code level means of household inco me (available ihrougti the 2001Canada censuiil of residential addresses of children attending ihai schoo l.



Childhood overweight and obesity

drinks were sold. These children drank an average of 4.0

cans of soda pop per week compared with 3.6 cans drunk

by children attending schools without such sales (p < 0.01),

C'.hildren from schools with and without sales of soft drinks

consumed an average of 33.5 and 32.5 g of sucrose per day

respectively (p = O.U), Availability of soft drinks at schoolswas not associated with significantly increased risks of over-

weight (Table 4),

Children attending schools with more frequent physical

education classes were increasingly more likely to have nor-

mal hody weight. Financial restraints on the purchase of

recreation and gymnasium equipment were not statistically

associated with increased weight (Table 4).

When we compared obese with normal-weight chil-

dren , we obtained a mo del th at identified m ale sex as a

significant risk factor for ohesi ty (Tab le 5). As well, the

effects of parents' educational attainment, neighbourhood

income, and frequency of physical activity and physicaleducation classes were more pronounced for obesity than

for overweight.

Interpretation

T h e results of this study show that 3 2.9% of grade 5 stu-

dents in Nova Scotia are overweight and 9,9% are obese.

This rate of overweight is higher than the 26% found

among participants of the same age in the 1996 NLSCY."

Our obesity estimates are more than twice as high as those

of 10- and 11 -year-old NLS CY participan ts and abo ut

50% higher than those of 10- and 11-year-old NLSCY

participants residing in Nova Scotia." It is important to

note tha t we measured ch i ldren ' s he ight and we ight ,

whereas the NLSCY used parental reports of height and

weight. Because the latter generally results in overestima-

tion of overweight and obesitj',"' the actual difference be-

tween the rates is likely to be even more pronounced,

Swinburn and associates called for the investigation ofrisk factors relating to parents, family and school environ-

me nt. '' W e found that children from families tha t eat to -

gether regularly are less likely to be overweight or obese.

One reason for this is that these children generally eat a

more healthy diet.''"'^' A second reason is that family meals

prevent children from eating in front of the television,

which may lead to "mindless eadng" and higher energy in-

take. '^" There may also be broader benefits of better com-

mumcadon and relationships between children and parents

resulting from the daily interactions during meals.'-

In terms of school environment, the association be-

tween obesity levels and frequency of physical educadonclasses was striking. This finding supports policies that in-

crease the frequency of physical education classes, such as

those recently announced for all elementary schools in Al-

berta. School lunches represent another opportunity to

reduce overweight and obesity. Halting the sales of sofr

drinks in schools has been suggested to reduce the intake

of refined sugars"" and is an advocated strategy to pre-

ven t overweight,"^ W e observed that childre n atten din g

schools that sell soft drinks consumed somewhat more

soft drinks and sugar, but the amounts were likely insuffi-

cient to bring about differences in body weight. Previ-

ously we showed that integrated school programs that in-

clude more physical education, healthy lunches, health

Table 4: School-based

Risk factor

Lunch provided by

Foodservice company

Staff or volunteers

Prepaid lunch program

Available at school

Soft drinks

Vending machine(s)

Frequency of physicaleducation classes

< 1.5 times/wk

a 1.5 and<2 timesAvk

> 2 times/wk

Financial restraints for

Outdoor play equipment

Gymnasium equipment

risk factors

No. of

studentsn = 4209

1269

54 0

658

658

2229

185

2097

1927

2352

1741

for overweight among grade 5 students in Nova Scotia

weigti tedprevalence,*

29.3

12.3

35.6

15,3

59,2

3.9

49,5

46.6

32.8

33.6

Weighted odds ratio (957o confidence interval)*

% Unadjusted Theme-adjustedt Fully adjusted^:

1,12(0.93-1.34)

1,02(0.77-1.35)

1.05(0.88-1.25)

0.99(0.77-1.27)

0.90(0.76-1,07)

1 1 1

0.61 (0.43-0.88) 0.61 (0,43-0.88) 0.76 (0,53 -1.10 )

0.53 (0.36-0.76) 0.53 (0.36-0.76) 0.61 (0.43-0.87)

1.08(0.91-1.29)

1.17(0.98-1.40)

"Weighted for iionrespo nje bias to reflect provin cial estimates,

tTheme-ddjustpd odds ralios are adjusted for (requency of physical edu ration classes,

iFu lly adjusted odds ralios are adjusted for significant risk factors in all 4 theme groups, inclu ding !unch and family supper in dietary habils, (Table 1),parricipation in physical activities in activities (Table 2], parental education and neighbourhood inrnme in sociodemoRraphic faaors (Table 3). andfrequency of physical education classes (school-based faciors).




and nutr i t ion educat ion, t ra in ing of staff, paren ta l in-

volvement, and a halt to the sales of soft drinks were suc-

cessfiil in improving children's diets and reducing over-

weight by 59% and obesity by 7 2 % . "

In our study, as in other studies of the effect of risk fac-

tors related to children's lifestyle, normal-weight childrenwere more physically active and engaged less in sedentary

activities.'''"'' Similarly in keeping with other studies, we

observed a gradient whereby children of socioeconomically

disadvantaged families were more likely to be overweight

or obese."'̂ ••' Studies in the United States and Europe have

shown neighbourhood differentials with respect to over-

weight. '^'" Our study confirms this in a Canadian setting,

with childhood obesity rates that were twice ashigh in low-

Table 5: Risk factors for obesity among grade 5 students

In Nova Scotia

Risk factor

Adjusted odds ralio*

(95% confidence interval

Lunch

Bring from home 1

Eat at home 1.04(0.77-1.40)

Buy at school 1.39(1.09-1.77)

Skips 1.47(0.77-2.83)

Family supper

< once/wk 1

1-2 times/wk 0.88 (0.64-1.22)

3-4 times/wk 0.68 (0.48-0.96)

> 5 times/wk 0.71 (0.53-0.95)Participation in physical activities

£ twice/week 1

> 2 and < 4 times/wk 0.81 (0.58-1.13)

> 4 and 5 7 times/wk 0.77 (0.58-1.03)

> 7 tImes/wk 0.65 (0.44-0.97)

Sex

Girls 1

Boys 1.20(1.01-1.44)

Parental education

Completed secondary school or less 1

Completed community college 0.79 (0.61-1.01)

Completed university 0.58 (0.43-0.79)Completed graduate university 0.55 (0.36-0.83)

Neighhourhood incomef

Lowest one-third 1

Middle one-third 0.82 (0.61-1.11)

Highest one-third 0.50 (0.36-0.70)

Frequency of physical education

classes at school

< 1.5/wk 1

> 1.5 and < 2/wk 0.91 (0.55-1.52)

a 2/wk 0.54 (0.33-0.88)

•Odds ratios areadjusted for all of the factors listed In the lable and are weighted for a

nonresponse bias lo reflecl provinei. il eslima te.

tNeighbourhood income was estimated by averaging perstHoot, Ihe posial-codetewl meani ofhousehold income (available through the2001 Canada census} of

resldenliai addresses ofchildren attending that school.

i ncome ne ighbourhoods as in h igh- income ne ighbour

hoods. Because of these dramatic differences, children and

schools in low-income neighbourhoods should receive pri

ority in public health initiatives to reduce future socioeco

nomic inequalities in health.

Strengths ofour study include the population-based anmiiltidisciplinary approach. Other strengths include mea

surements of participants' height and weight, the adjust

men t for nonresponse bias and the consideration of variou

potential confounders. Although most of the questions and

questionnaires we administered were validated, response

remain subjective andprone to error. An additional limita

tion is the cross-secdona! design, which introduces uncer

tainties regarding the sequence of cause and effect of th

observed associations. Longitudinal studies overcome thi

limitation and will strengthen the evidence regarding risk

factors for childhood overweight and obesity.

In summar) ' , we have shown that school- and tamilj 'based factors are associated with an increased risk of child

hood ovenveight and obesity, and therefore they presen

oppor tuni t ies for target ing tb is problem. Recommenda

tions for action include more physical education classes and

promotion of healthy lunches at schools and of family sup

pers at home. Preventive public bealth actions should b

targeted first toward low-income neighbourhoods.

TTiis article has been peer reviewed.

From the Dep ar o n en t of Public Heakb Sciences, University of .Mberta, Edmonton, Alta. (Veugelers) and the Dep ar tmen t of Commtini ty i teal th and Epicfemioogy, Dalhousie University. HalifaJi, NS (Veugelers, Fitzgerald)

Competing interests: None declared.

Contributors: Paul Veugelers, as the principal investigator of the CLASS projecconceived and designed the study. Heand Angela Fit7,gerald both contributed sub

stantially to the analysis and interpretation ofdata, revised tbe manuscript criticallfor important intellectual content and read and appro\'ed the final version.

Acknowledgements: We thank all of the grade 5 students, parents and schools fo

their participation in tbe CLASS project. We thank all of the research assistanand public health staff who assisted in the data collection, Jason L iang for tb

data management of the project, and Cvnthia Colapin to for her feedback on th

manuscript.

This research wasfunded by a Canadian Populatian Health [nitiative opera

ing grant and tbrough a Canadian Institutes of Health Research New InvestigatoAward to Dr. Veugeiers.

References1. libbeiing CB, Pawlak DB, Ludwig DS. Childbood obesity: Public-health cr

sis, commo n sense cure. Lancet 2OO2;36O:473-82,

2. Andersen RE. 1 he spread of tbe childbood obesity epidemic [commeiuaty

S.

7.

8.

3. Tre mb lay M S, VVillms JD . Secular trends in the body mass index of Canadichildren. CAt-J7200O;163:l429-33.

4. Tremblay MS, Katzmarzyk PT, Willins J D . Temporal trends inoverweight an

obesity in Canada. \9S\-\W6.1ntJOhesRelatMetahDisord2m2-.26:5i^-^l.Tr emb iay MS, lnman JW, Willm s JD , Relationships between physical actiity, self esteem, and academic achievements in ten- and eleven-year-old ch

dren. Pediatr ExerSalOOOM-} 12-2}.

Hesketh K, Wak e M, Wate r s E. Body mass index and parent- repor ted se

esteem in elementar) ' school children: evidence for a causal relationship. InObesReiaiMetab Disord20m;2S:\2ii-7.

Must A, Strauss RS. Risks and consequences of childbood and adolescent obsity. ImJ Obes Relat Mmb Disord iyy'>;23(Suppl 2):S2- II .

Dietz \VH. Overweight in childhood and adolescence. N En^l J Med 200350:855-7,



9. Ma nson JE , Ba.ssuk SS. Obe sity in tbe U nited States. A fresh look at its high

roll JAMA 2QQ}i2S9:22'>-ili.

U ), Fontaine K R, Redden D T. Wa ng C, Westfall AO , Allison DB , Years of life

los tdue to obes i ty .7 / lAM20O 3;289:187-93,

1 1 . Kaczmarzyk P T, Janssen I. Tb e ec onomic costs associated with physical inac-tivity and ohesity in Canada: An update. Can J Appl Physio! 2004;29\90-\\$.

\2 . Nicklas T, Johnson R. Position of the American Dietetic .Association: dietaryguidant-e for healthy children ages 2 to 11 years. J Am Diet Assoc 20O^;\(y\:fi60-77.

1 3 . Katzmarzyk PT, Ardern CI , Overweight and obes iry mor tabty t rends in

Canada, 1985-2000. C a n J Public Health 2OO4;95:16-20.

1 4 . Power C, G raham H, Due P , Hallqvi.st J , Joung 1, Kuh D, et al. Th e contribu-

tion of childhood and adult socioeconomic position to aduk obesity and smok-ing behaviour: an international comparison, hi J Epidemiol 2()0S;i'i\'ii5-44.

15 . Lamerz A, Kiiepper-Nybelen J, Wehle C, Bruning N, Trost-Brinkhiies G, Bren-

ner H, et al. Social class, parental education, and obesity pre^'alence in a study- ofsix-year-old children in Germany. Int] Obes Relat Metab Disord 20O5;29;373-80.

1 6 . Trem blay M S, Willms JD . Is the Canadian cbildbood obesity epidemic re-

lated to physical inactivity? Ini J Obes Relat Metab Disord 2OOi\27:\]OO-5.

1 7 . Hancox RJ, Milne BJ, Poulton R. Association hetween child and adolescenttelevision viewing and adult health: a longitudinal birtb cohort study. Lancet2004:17;364:2S7-62.

1 8. Graf C, Koch B, Dordel S, Schindler-Marlow S, Icks A, Schuller A, et al.

Physical activity, leisure babits and obesity in first-grade cbildren. EurJ Car-dtovasc Prei' Rebabil 2004;l 1:284-90.

1 9 . Willms JD , Early Child hood Obesity: A call for early surveillance and pre -

ventive measures. CAW72004;171:243-4,2 0 . Cann ing P M, C ourage M L. Frizzell, Prevalence of overweight and obesity in a

provincial population of Canadian prescbool cbildren. CM472l)tM;171:24O-2.

2 1 . Swinburn B, Gill T , Kumanyika S. Obesity preven tion: a proposed frame-

work for translating evidence into action, Obes Rev 2QQS•,6:21-^ii.2 2 . Gaudet te LA, Altmayer CA, Wysocki M, Gao RN, Cancer incidence and

mortality across Canada, Health R e p 1998;10;51-66.2 3 . Willms JD ,Tre mh iay MS, Katzmarzyk PT , Geographic and demographic varia-

tion in tbe prevalence of overweight Canadian children. O/'crR«2003;l 1:668-73.

Z 4 . Veugelers PJ, Fitzgerald AL. Effectiveness of scbool programs in tbe preven-tion of childhood obesity, Am J Public Health 2OO5;95:432-5,

2 5 . Veugelers PJ, Fitzgerald A L , Johnston E, Dietary intake and risk factors for poor

diet quality among children in Nova Scoria. Can J Public Health 2OO5;96;212-6.

2 6 . Cole ' 0 , Bellizzi MC, Flegal KM , Diec/, W H , Establishing a standard defini-

rion for child overweight and obesity worldwide: international survey. BM J2OO0;320:1240-3,

2 7 . Rockett HR , Wolf AM, Colditz GA . Deve lopmen t and reproducibility of a

food frequency q uestio nnaire co assess diets of older child ren and adol escent s.

J / i m D(>(/lr,«.f 1995;95:336-4O.2 8 . Stat is t ics Canada, Nat ional longi tudinal survey of cbi ldren and youth

(NL SCY ). Avai lable: ww w,s tatcan .ca/cgi-b in / imd b/p2SV.pl?Fun ct ion=get

Surve ys SDDS=4450&lang=en&db".IMDB&dhg=f&adm=8&dis=2 (accessed2OO5Jul26).

2 9 . Burton P, Lethhridge L, Osberg L, Phipps S. Measuring obesity in young

children. Ca n Public Policy 2004;30:349-64,

3 0 . Gillman MW, Rifas -Shiman SL, Frazier L, Rocket t HRH, Camargo CA,Field AE, Family dinner and diet quality among older cbildren and adoles-

cents. Arch Fam Med 2 OOO;9 i2 3 5 ^ .

3 1 . Neu mark -Sztain er D , Hannan PJ, Story M, Croll J , Perry C. Family meal

patterns: Associations witb sociodemograpbic characteristics and improveddietary in take am ong adolescents , J -^w D/e i^w of 2003; 103:317-22.

3 2 . Robinson T N . Television viewing and childhood obesity, Pediatr Clin North^m2 OOl ;4 8 : I Oi7 - 2 5 .

3 3 . Caroli M, Argentieri L, Cardone M, Masi A. Role of television in childhood

obesity prevention, htj Obes Relat Metab Disord 2004;28(Suppl 3):S104-8.

3 4 . National Center for Addiction and Substance Abuse at Columbia University,

The importance of amily dinners. New York: l"be Center ; 2003, Available:w\ww.casacoiumhia.org/Absolutenm/articlefiles/Famiiy_Dinners_9_03_03.pdf(accessed 2005 Mar 21),

3 5 . Swinbum BA, Caterson 1, SeldellJC, James W P, D iet, nutrition and tbe preven-tion of excess weight gain and obesity. Public Health Nutr2OM;7(]\y.]2i-^.

3 6 . Fried EJ, Nestle M. Tbe growing polidcal movement against soft drinks inschools. 7^A M 2OO2;288:2181.

3 7 . EUaway A, Anderson A, Macintyre S. Does area of residence affect body size

and shape? IntJ Obes Relat Metab Disord 1997;21:304-8,

3 8 . Van Len the FJ, Mackenbach JP, Neighbou rhood d eprivation and overweight:tbe GLOBE s tudy. Iftj Obes Relat Metab Disord 2OO2;26:234-4O.

, 39 , Robert SA, Reither EN. A mulrilevel analysis of race, community disadvantage,and body mass index among adults in theUS. Sof Sir(Mei/2OO4;59:242I-34.

Correspondence to: Dr. Paul). Veugelers, Departm ent of Public

Health Sciences, Faculty of Medicine and Dentistry, University of

Alherta, 13-106D Clinical Sciences Building, Edmonton ABT6G 2G3; fax 780 492-0364; paul.veugelers©ualberta.ca

Now w ith VaD datain the Product Monograph^

ke O ^ K stdi

• Over 7 years of excellent efficacy supported

by more than 2 bil l ion pat ient days worldwide"" '

• Excellent safety and tolera bility profiles^

• Convenient once-a-day dosing^

• A wealth of clinical studies

i s i n d i c a t e d f o r H i e s y m p t o m a t i c t r e a t m e n t o l p a t i e n t s w i t n m i l O - t o - m o d e r a t e

d e m e n t ia o f t h e A l z ii e im e r 's t y p e a n d d o e s n o t c i ia n g e th e u n d e t t y i n g c o u r s e o t t t i e d i s e a s e .

t A i i c e p r is n o t i n d i c a t e d for t t i e t r e a t m e n t

o f p a r e n t s w i t i i v a s c u la r d e m e n t i a . M h o u g h f fi e s e

d a t a a r e [ y o m i s i n g , fiey a r e c o n s i d e r e d h y p o t h e s i s -

g e n e r a t i n g , a n d c o n f i r m a t o i y t r ia l s a re r e q u i r e d .

W i th a p p r o p r i a t e d o s e e s c a l a t io n , 5 m g / d ,

1 0 m g / d a n d p l a c e b o w e r e s t i o w n to h a v e

c o m p a r a b l e a d v e r s e e v e n t s , t h e m o s t c o m m o n

{ o c c u r r i n g in at l e a s t 5 % of p a t i e n t s ) b e i n g

d i a n h e a , n a u s e a , i n s o m n i a , f a t i g u e , v o m i t i n g ,

m u s c l e c r a m p s a n d a n o r e x i a . T h e s e a re u s u a l l y

m i ld a n d t r a n s i e n t , re s o i v i n g w i t h c o n t i n u e d t r e a t - ,

m e n t w i t i o u t n e e d or d o s e m o d i f i c a t i o n .

A R I C E P T '

IS T H E * 1

R E C O M M E N D E D

A D

T H E R A P Y

I N C A N A D A . '

1 I n a b 2 - m X i m i l t o l e , o au C fe M,l o re c B ir e ^ i c m ' 5 m g / d i a ? a t te y s f n = M 2 ) J o lt o ii ie j b y 1 0 m f t' rt , as | » [ Jn i ra n ' s j u i ^

§ IT a 5 4 w

In a 2 i - m k , r a r K to r iz e d , d o u M e - M m l , p l a c e ii o a r t r o l le d s i x t y i ) < A i l c « i l ' i i 2 0 7 m o d e t a le W p a f c f f c P ^ 1 0 - 1 7] .

p a l i 9 i l 5 re c E T O d e t h e r W i c a i l " 5 m g / d oi ie t d 2 B d a is a W 1 0 m ^ J r a e a t l a , 8S p a c W c i a ii 's ) u d p n e n (

M O 2 ) , o ( [p la c eb o ( n = lO 5 1

• T M H s a i C o L W , T o ln o , Ja p a n , P t l r e t C a n a c t a I n c , k c e n s e e

O 2 0 0 5 , P I C H C a n a d a i n c , K W O a n d, Q u e b e c H 9 J 2 m

Once-a-day

NEUROSCIENCES

•ilHri:',& 10 mgMMn.1

Experience defines us.For prescribing iniormation see page 679

prevalence of and risk factors for childhood overweight and obesity

Documents