In . J. En i on. Res. Public Heal h 2012, 9, 1355-1367; doi:10.3390/ije ph9041355
In e na ional Jou nal o
En i onmen al Resea ch and
Public Heal h
ISSN 1660-4601
www.mdpi.com/jou nal/ije ph
A icle
The Impac o an In e en ion Taugh by T ained Teache s on
Childhood O e weigh
Ra aela Rosá io 1,2,*, B uno Oli ei a 3, Ana A aújo 2, Osca Lopes 4, Pa ícia Pad ão 3,
And é Mo ei a 5, Ví o Teixei a 3, Rena a Ba os 3, Bea iz Pe ei a 2 and Ped o Mo ei a 3,6
1 School o Nu sing, Uni e si y o Minho, La go do Paço, 4704-553 B aga, Po ugal
2 Resea ch Cen e in Child S udies, School o Educa ion, Uni e si y o Minho, Campus de Gual a ,
4710-057 B aga, Po ugal; E-Mails: anam_a [email protected] (A.A.); bea [email p o ec ed]minho.p (B.P.)
3 Facul y o Nu i ion and Food Sciences, Uni e si y o Po o, Rua Robe o F ias, 4200-465 Po o,
Po ugal; E-Mails: bmpmo@ cna.up.p (B.O.); pa iciapad ao@ cna.up.p (P.P.);
hugo eixei a@ cna.up.p (V.T.); ena aba os@ cna.up.p (R.B.); ped omo ei a@ cna.up.p (P.M.)
4 Spo s’Medical Cen e , Tempo Li e, Alameda Cidade de Lisboa, C eixomil, 4835-037 Guima ães,
Po ugal; E-Mail: olopes@ empoli e.p
5 Facul y o Medicine, Uni e si y o Po o, Al. P o . He nâni Mon ei o, 4200-319 Po o, Po ugal;
E-Mail: and emo ei [email protected]
6 Resea ch Cen e in Physical Ac i i y, Heal h and Leisu e, Uni e siy o Po o, Rua D . Plácido
Cos a, 91, 4200-450 Po o, Po ugal
* Au ho o whom co espondence should be add essed; E-Mail: osa [email protected] ;
Tel.: +351-253-601-326; Fax: +351-253-601319.
Recei ed: 14 Oc obe 2011; in e ised o m: 14 Decembe 2011 / Accep ed: 4 Janua y 2012 /
Published: 16 Ap il 2012
Abs ac : The pu pose o his s udy was o assess he e ec s o a six-mon hs’ nu i ion
p og am, deli e ed and augh by class oom eache s wi h in-se ice nu i ion aining, on
he p e en ion o o e weigh and obesi y among child en in g ades 1 o 4. In his
andomized ial, ou hund ed and six y ou child en om se en elemen a y schools we e
alloca ed o a nu i ion educa ional p og am deli e ed by hei own eache s. In e ened
eache s had 12 sessions o h ee hou s each wi h he esea che s h oughou six mon hs,
acco ding o he opics nu i ion and heal hy ea ing, he impo ance o d inking wa e and
heal hy cooking ac i i ies. A e each session, eache s we e encou aged o de elop
ac i i ies in class ocused on he lea ned opics. Sociodemog aphic, an h opome ic,
OPEN ACCESS
In . J. En i on. Res. Public Heal h 2012, 9
1356
die a y, and physical ac i i y assessmen s we e pe o med a baseline and a he end o he
in e en ion. In he in e en ion g oup he inc ease in Body Mass Index (BMI) z-sco e was
signi ican ly lowe han in he con ol g oup (p = 0.009); ewe p opo ion o child en
became o e weigh in he in e ened g oup compa ed wi h he con ol (5.6% s. 18.4%;
p = 0.037). Ou s udy p o ides u he suppo o dec ease he o e weigh epidemic,
in ol ing class oom eache s in a aining p og am and making hem dedica ed
in e en ionis s.
Keywo ds: BMI z-sco e; child en; obesi y; o e weigh ; ained eache s; heal h p omo ion
1. In oduc ion
The p e alence o obesi y con inues o inc ease [1] and is a g owing conce n in Po ugal and a ound
he Wo ld. O e weigh and obesi y ha e se ious implica ions in child’s heal h as hey ha e been
associa ed wi h as hma, ype II diabe es, ca dio ascula diso de s, hepa ic and esicula diseases,
s igma and o he psychological p oblems, sleep apnea, a h i is, s oke, ce ain ype o cance s and
quali y o li e [2]. As he p ocess o ea men is complex, discussion abou how o p e en obesi y is
cu en ly high on public heal h agenda and e ec i e heal h p omo ion emains a key s a egy [3].
Al hough he pic u e o how o in e ene is a om comple e, guidance de eloped by e ec i e
p og ams ha e led o an inc ease in obesi y p e en ion esea ch, bu so a only yielded “bes p ac ice”
ecommenda ions. In addi ion, mos s udies ha e been de eloped in he Uni ed S a es, aising
ques ions abou hei applicabili y in Eu opean coun ies gi en he di e en school sys em and school
nu i ion condi ions, as well as di e ences in ea ing habi s o obesi y a es [3,4]. As a esul , he e is a
need o de elop and implemen e ec i e in e en ion p og ams and policies in Eu ope aiming o
imp o e child en’s li es yles.
In e en ions in childhood a e conside ed o ha e la ge e ec s han in olde ages because child en
a e mo e sensi i e o ou side in luences and may conside in e en ions pa icula ly a ac i e [5].
Mo eo e , ea ing and physical ac i i y beha io s lea ned du ing childhood may pe sis h oughou
adul hood [5,6]. The po en ial se ing o he in e en ions is also an impo an aspec o o e weigh and
obesi y p e en ion p og ams. The use o he communi y s uc u es, such as schools, educes he
ba ie s o implemen a ion and ocus in a na u al se ing o child en [5].
Un il now, he ole o eache s in he deli e y ea u es o he in e en ions and i s impac on child’s
body composi ion is unclea [7,8]. In addi ion, ou s udies examine he e ec s o educa ional
p og ams on he an h opome y o child en [9-12] and ew mee he eal knowledge o eache s and
child en’s needs. Theo e ically, eache s a e no able o de o e as much ime and ene gy o p o iding
in e en ions as dedica ed in e en ionis s because eache s ha e class oom esponsibili ies ha ake
p ecedence [8]. To he bes o ou knowledge, his is he i s s udy ha inse s he in e en ion p og am
o p e en childhood o e weigh in he p og ession o eaching ca ee .
The pu pose o he p esen esea ch was o assess he impac o a six mon hs nu i ion p og am,
deli e ed and augh by class oom eache s wi h in-se ice nu i ion aining, on he p e en ion o
o e weigh and obesi y among child en in g ades 1 o 4.
In . J. En i on. Res. Public Heal h 2012, 9
1357
2. Expe imen al Sec ion
2.1. Pa icipan s
Du ing 2007/2008 se en ou o eigh y public elemen a y schools we e andomly selec ed and
in i ed o pa icipa e in his s udy. The numbe o schools in ol ed was acco ding o cons ain s o
pe sonnel o he assessmen and implemen a ion o he p og am. Schools we e he uni o
andomiza ion and h ee we e assigned in o in e en ion, and ou in o con ol g oup (Figu e 1). The
dis ance be ween schools was o 5 Km o less and all o hem we e conside ed u ban. Da a was
collec ed p io o he p og am and immedia ely a e . P io o pa icipa ion on da a collec ion, pa en s
p o ided in o med consen , acco ding wi h he e hical s anda ds laid down in he Decla a ion o
Helsinki, and child en p o ided o al assen . The s udy was app o ed by he schools whe e i was
ca ied ou and by he Po uguese Da a P o ec ion Au ho i y (CNPD-Comissão Nacional de P o ecção
de Dados, p ocess numbe 7613/2008). In addi ion he p o ocol o his s udy was egis e ed in he
clinical ials egis y clinical ials.go , NCT01397123.
Figu e 1. Flow o pa icipan s h ough each s age o he p og am.
The low o he subjec s du ing he s udy is p esen ed in Figu e 1. F om he 574 child en who we e
in i ed o pa icipa e, 464 (239 emale, ages anging om 6 o 12 yea s old) ag eed and e u ned he
w i en consen o ms illed by hei pa en s. O hese, 233 (50.2%) we e alloca ed o he in e en ion
g oup, and 231 (49.8%) o he con ol g oup. Follow up assessmen was a ailable o 63.4% o
child en, 143 (61.9%) in he con ol and 151 (64.8%) in he in e en ion g oups.
In . J. En i on. Res. Public Heal h 2012, 9
1358
BMI and majo sociodemog aphic cha ac e is ics did no di e signi ican ly be ween he child en
who pa icipa ed in he baseline and hose no included in he inal assessmen .
A i ion a es did no di e be ween in e en ion and con ol g oup (35.2% and 38.1%,
espec i ely). Majo easons o nonpa icipa ion we e school ans e (94.1%), pa en e usal (4.1%)
and absence om school (1.8%). Child en and ou comes assesso s we e blinded o g oup assignmen .
A o al o 257 pa en s o he child en in ol ed in he s udy p o ided da a a baseline and 203 (79.0%)
a pos -in e en ion.
2.2. O e iew o he P og am
Teache s om in e en ion schools we e in i ed and ag eed o pa icipa e in he p og am conduc ed
be ween Oc obe 2008 and Ma ch 2009; i een eache s we e in ol ed. This p og am was based on
he Heal h P omo ion Model [13] and he social cogni i e heo y [14] and aimed o p omo e heal hie
ac i e li es yles by encou aging child en o be mo e ac i e and ensu e be e ood selec ion.
The Heal h P omo ion Model a gues ha ac ions and he heal h p omo ion beha io a e a co olla y
o pe sonal cha ac e is ics, p io expe iences, pe cei ed bene i s and ba ie s o ac ion as well as
pe cei ed sel -e icacy [13]. This p og am was awa e o hese in luences on child en beha io and
ocused on a posi i e ision o heal h. Likewise he social cogni i e heo y, he p og am enhanced
cogni i e and beha io al skills by enabling child en o make changes in hei own beha io and o
employ new choices e ec i ely [14].
In Po ugal, in-se ice aining is conside ed he basis o he e icacy o he Educa ion sys em.
Teache s ha e he igh and he obliga ion o de elop hei aining h ough an accumula ion o c edi s.
The e o e, he esea ch eam p oposed he acc edi a ion o he aining sessions de eloped wi h he
eache s o he Minis e o Educa ion, Scien i ic-Pedagogic Council o In-se ice T aining (Conselho
Cien í ico Pedagógico da Fo mação Con ínua, Minis é io da Educação). I was app o ed in he o m o
a “ aining wo kshop” wi h 72 h du a ion, dis ibu ed be ween ac i e lea ning s a egies (36 h wi h he
esea che s) and he deli e y o he lea n con en s o he child en (36 h). Thus, he p og am was
implemen ed o e wo e ms wi h eache s aining deli e ed by esea che s and in e en ion deli e ed
by ained eache s o child en.
Teache s o he in e en ion g oup had 12 sessions o h ee hou s each wi h he esea che s du ing
six mon hs, which included he ollowing con en s: session 1, how o p omo e heal h and p e en
disease, li es yle de e minan s o heal h, obesi y—de ini ions and desc ip ions o he p oblem, isk
ac o s and heal h p oblems; session 2, key concep s in ood and nu i ion; sessions 3 and 4, die a y
guidelines ( he Po uguese Food Wheel), heal hy ea ing ad ice o child en, co e ing he i e main
ood g oups, and in e en ions o help child en and hei amilies o consume heal hy oods and plan
well-balanced meals and snacks; session 5, each child en abou he impo ance o wa e , and eaching
s a egies o eplace consump ion o suga -swee ened be e ages wi h wa e ; sessions 6 and 7,
app op ia e physical ac i i y le els and heal hy ea ing beha iou s such inc easing ui and ege able
in ake and dec easing ene gy-dense mic onu ien -poo oods; session 8, eaching s a egies and
lea ning heo y in he class oom; session 9, s a egies o educe sc een exposu e ime; session 10,
global assessmen o he aining p og am; sessions 11 and 12, heal hy cooking and s a egies o ge
child en and hei amilies in ol ed in heal hy cooking.
In . J. En i on. Res. Public Heal h 2012, 9
1359
A e each session, eache s deli e ed he lea n con en s and de eloped c ea i e and engaging
class oom ac i i ies abou he add essed opic. All he ques ions ha a ose du ing he implemen a ion
o class oom ac i i ies we e add essed and esol ed sho ly wi h he esea che s. Teache s we e
allowed o de elop and e ine he ac i i ies and he lea ning s a egies ha we e p oposed by he
esea che s. A he end o his pe iod, he eache s deli e ed a c i ical epo o ac i i ies ocused on he
in e en ion, e alua ed by he esea ch eam. The au onomy o eache s in he aining p ocess was
p ese ed and he esea che s had a se ious in es men on p o essionalism, p agma ism and policy [15].
Recognizing ha i was no possible o comple e a s uc u ed assessmen o all he in o ma ion and
ma e ials deli e ed in all o he classes, esea che s also pe o med an in o mal assessmen (based on
ace- o- ace in e iews by asking ques ions and cla i ying un esol ed issues) on each eache .
The implemen a ion o he p og am occu ed as planned. All he child en o he in e en ion schools
had con ac wi h ained eache s. Teache s augh he componen s o he p og am as p esc ibed and
he esea che s we e always a ailable o answe any ques ion. In addi ion, eache s epo ed hey we e
en husias ic abou he aining, and had a o al a endance in he sessions wi h he esea che s.
2.3. Assessmen s
In each school, p e iously ained pe sons pe o med an h opome ic e alua ion, using s anda dized
p ocedu es [16]. An h opome ic measu emen s we e pe o med in child en wi h ligh indoo clo hing
and ba e oo ed. Weigh was measu ed in an elec onic scale, wi h an e o o ±100 g (Seca, Model 703,
Ge many), and heigh was measu ed using a s adiome e , wi h he head in he F ank o plane. BMI
was compu ed as mass, kg/heigh m2. The p e alence o unde weigh , no mal weigh , o e weigh and
obesi y was calcula ed acco ding o he In e na ional Obesi y Task Fo ce (IOTF) c i e ia [17].
A z-sco e ( he numbe o s anda d de ia ions (sd) om he e e ence popula ion) was calcula ed o
each child using he LMS me hod and he calcula ion was de e mined using he LMS g ow h add-in
o excel [18]. To es ima e he magni ude o BMI changes du ing he s udy, a new a iable was
compu ed om he di e ence be ween he pos -in e en ion and baseline BMI z-sco e.
Die a y in ake was ga he ed by one day 24-h die a y ecall ob ained by nu i ionis s and/o ained
in e iewe s be o e and a e he p og am. Child en did no ha e p io no i ica ion o when he ecalls
would occu in o de o p e en po en ially biasing epo s and he weekend days we e a oided. Du ing
he 24-h die a y ecall, each child was asked o ecall all ood and be e ages consumed du ing he pas
24 h. Po ion sizes o oods and be e ages consumed we e also es ima ed, using ood models and
pho os, and o he p ops (cups, glasses, ood w appe s o con aine s) as an aid in de e mining se ing
sizes. Daily ou ines we e used as p omp s (waking up, going o bed, ime be ween classes, and be o e
o a e school) o enhance ecall. Ene gy and nu i ional in ake we e es ima ed using an adap ed
Po uguese e sion o he nu i ional analysis so wa e Food P ocesso Plus (ESHA Resea ch Inc.,
Salem, OR, USA).
In o de o assess he le el o physical ac i i y o child en, pa en s we e asked i e ques ions wi h
ou answe choices (4-poin scale) anging om 1 o 4 abou child en’s ac i i y [19]: (a) ou side he
school does you child ake pa in o ganized spo ? (b) ou side school does you child ake pa in non-
o ganized spo ? (c) ou side school, how many imes a week does you child ake pa in spo o
physical ac i i y o a leas 20 min? (d) ou side school hou s, how many hou s a week does you child

In . J. En i on. Res. Public Heal h 2012, 9
1360
usually ake pa in physical ac i i y so much ha ge s ou o b ea h o swea ? (e) does you child ake
pa in compe i i e spo ?. This ques ionnai e was de eloped by Telama e al. [20] and i s appliance o
Po uguese popula ion has been desc ibed p e iously [19,21]. O e all a maximum o 20 poin s can be
eached. The Physical Ac i i y Index was ob ained, spli ing he sample in o ou ac i i y classes:
seden a y g oup [5]; low ac i i y g oup [6–10]; mode a ely ac i e g oup [11–15]; and igo ously ac i e
g oup [16–20], on he basis o hei epo ed physical ac i i y [19,22].
Social, demog aphic and amily cha ac e is ics we e assessed by ques ionnai e. The su ey sen o
pa en s con ained ques ions abou gende and age o child en, educa ion o he pa en s ( eco ded in
i e ca ego ies: 0, 1–4, 5–9, 10–12, and mo e han 12 yea s o o mal educa ion). This in o ma ion was
u he g ouped o analysis in o h ee ca ego ies: up o 9 yea s, 10–12 yea s, and mo e han 12 yea s
o educa ion.
2.4. S a is ical Analyses
Mean and sd we e used o desc ibe con inuous a iables. S uden ’s - es s, Mann-Whi ney U,
K uskall Wallis and Chi-squa e es s we e used o compa e se e al a iables g ouped by in e en ion
and con ol school g oups and sex. These es s we e also conduc ed o assu e compa abili y o
an h opome ic measu emen s be ween g oups a baseline. A 0.05 le el o signi icance was conside ed.
Fo he pu pose o sample size es ima ion, he p ima y ou come was a change in BMI-z sco e be ween
pos -in e en ion and baseline. Acco ding o p e ious s udies [23], a sd o BMI z-sco e nea 0.5 and a
d opou a e nea 30% was es ima ed. F om hese assump ions a powe analyses was conduc ed using
sample size o each g oup o 143, ob aining an es ima ed powe o 0.84 o - es .
The e ec o he p og am was e alua ed based on changes in an h opome y be ween baseline and
pos in e en ion, compa ing in e en ion o con ol schools. The es s examining hese di e ences
we e de eloped using Gene alized Linea Models and ook in o accoun he nes ed na u e o he da a
(child en we e nes ed wi hin schools). Hence, he BMI z-sco e a ia ion was used as dependen
a iable and he adjus men was made o gende , age, baseline o al ene gy in ake, baseline BMI
z-sco e and pa en s’ educa ion in o de o maximize p ecision.
Incidence o o e weigh and obesi y a e in e en ion (pe cen age o child en who we e no
ini ially o e weigh o obese bu who became o e weigh o obese), p e alence (pe cen age o child en
who we e o e weigh o obese) and emission (pe cen age o child en who we e o e weigh o obese
a baseline bu we e no o e weigh o obese a e in e en ion) we e analyzed. Analyses o o e weigh
and obesi y we e conduc ed sepa a ely. The p e alence, incidence and emission assessmen was
conduc ed by bina y ou comes.
The da a analysis was pe o med using SPSS®, Ve sion 18.0 (SPSS Inc; Chicago, IL, USA).
3. Resul s
Table 1 shows he baseline cha ac e is ics o he pa icipan s. Subjec s included 239 gi ls and 225
boys wi h a mean (sd) age o 8.3 (1.2) yea s. The a e age BMI was 17.9 kg/m2 (sd = 2.7, ange 11.9 o
26.9 kg/m2) and BMI z-sco e was 0.8 (1.1). O e all, 23.3% o he child en we e classi ied as
o e weigh and 9.5% as obese. The la ge majo i y o he child en we e classi ied in seden a y
o low ac i i y g oup. Mean ene gy in ake was no signi ican ly di e en be ween in e en ion and
In . J. En i on. Res. Public Heal h 2012, 9
1361
con ol schools ( espec i ely, 2091 (684) kcal e sus 2024 (582) kcal, p = 0.257). No signi ican
di e ences we e ound o weigh , BMI and BMI z-sco e a he baseline assessmen be ween
in e en ion and con ol g oups.
Table 1. Cha ac e is ics o he sample a baseline in in e en ion and con ol schools.
N To al In e en ion Con ol
p
464 231 233
Age (yea s) 8.3 (1.2) 8.3 (1.2) 8.2 (1.2) 0.846
Boys 225 (48.5) 116 (49.8) 109 (47.2)
Gi ls 239 (51.5) 117 (50.2) 122 (52.8) 0.575
Mo he ’s educa ion
Up o 9 yea s 244 (64.0) 116 (58.6) 128 (69.9)
10–12 yea s 88 (23.1) 52 (26.3) 36 (19.7)
>12 yea s 49 (10.6) 30 (15.2) 19 (10.4) 0.021
Fa he ’s educa ion
Up o 9 yea s 254 (69.0) 122 (62.9) 132 (75.9)
10–12 yea s 70 (19.0) 39 (20.1) 31 (17.8)
>12 yea s 44 (12.0) 33 (17.0) 11 (6.3) 0.003
Weigh (kg) 30.9 (7.4) 30.9 (7.2) 30.9 (7.5) 0.901
Heigh (cm) 130.5 (7.8) 129.3 (7.8) 131.3 (7.8) 0.046
BMI (kg/m2) 17.9 (2.7) 18.1 (2.7) 17.7 (2.8) 0.062
Da a p esen ed as mean (sd) o age, an h opome ic a iables, and n (%) o o he a iables.
The e we e signi ican di e ences be ween g oups wi h ega d o mo he and a he educa ion
(p = 0.021 and p = 0.003 espec i ely), and heigh (p = 0.045). The e we e highe le els o pa en s
educa ion in he in e en ion g oup and child en in he con ol g oup we e alle [1.29 (0.06) s. 1.27
(0.06) m]. To accoun o hese di e ences a baseline, hese a iables we e con olled o in
subsequen analyses.
A e in e en ion (Table 2), he BMI z-sco e a ia ion (pos in e en ion—baseline) was highe in
he con ol han in he in e en ion subjec s [ espec i ely, mean (se) 0.34 (0.05) e sus 0.13 (0.04)].
A e adjus ing o gende , age, baseline o al ene gy in ake, baseline BMI z-sco e and pa en s’
educa ion, he BMI z-sco e inc eased 0.176 uni s mo e in he con ol g oup han in he in e en ion
g oup [95% CI = (0.044;0.308), p = 0.009], no ep esen ed in he able.
Table 2. Ou comes a pos -in e en ion.
Measu e
Baseline Pos -in e en ion
p
Con ol In e en ion Con ol In e en ion
n = 233 n = 231 n = 143 n = 151
BMI z-sco e 0.66 (1.12) 0.84 (1.07) 0.92 (1.0) 0.90 (0.97) 0.875
Ene gy (kcal/day) 2024.2 (581.8) 2091 (683.9) 2475.6 (684.9) 2388.0 (1036.5) 0.399
IOTF
Unde weigh 10 (2.1) 7 (1.5) 0 (0.0) 2 (0.7)
No mal 157 (33.8) 138 (29.7) 90 (62.9) 95 (62.9)
O e weigh 41 (8.8) 67 (14.4) 40 (28.0) 44 (29.1)
Obesi y 23 (5.0) 21 (4.5) 13 (9.1) 10 (6.6) 0.610
In . J. En i on. Res. Public Heal h 2012, 9
1362
Table 2. Con .
Measu e
Baseline Pos -in e en ion
p
Con ol In e en ion Con ol In e en ion
n = 233 n = 231 n = 143 n = 151
Physical Ac i i y Index
Seden a
y
21
(
15.6
)
23
(
14.0
)
6
(
7.1
)
5
(
5.9
)
Low-ac i i
y
72
(
53.3
)
82
(
50.0
)
48
(
56.5
)
40
(
47.1
)
Mode a el
y
ac i i
y
35
(
25.9
)
49
(
29.9
)
26
(
30.6
)
30
(
35.3
)
Vi
g
o ousl
y
Ac i i
y
7
(
5.2
)
10
(
6.1
)
5
(
5.9
)
10
(
11.8
)
0.133
Da a p esen ed as mean (sd) o BMI z-sco e, ene gy and n (%) o o he a iables.
Signi ican ly ewe child en in he in e en ion schools (5.6%) han in he con ol schools (18.4%)
became o e weigh a e he in e en ion (unadjus ed means; Table 3). A e con olling o
con ounde s, he p edic ed odds o incidence was 75% lowe o he in e en ion g oup (odds
a io [OR]: 0.25; 95% CI: 0.07–0.92; p < 0.05). Fu he mo e, as expec ed, o e weigh child en wi h a
BMI close o no mal weigh s a us we e mo e likely o ha e emission o o e weigh han o e weigh
child en wi h a highe BMI (p = 0.018). Simila ly, no mal weigh child en on he h eshold o being
o e weigh we e mo e likely o su e o o e weigh compa ed wi h no mal weigh child en wi h a
lowe BMI (p = 0.001), no ep esen ed in he able. Mo e in e ened child en had emission o
o e weigh compa ed o con ol g oup (13% and 12.5%, espec i ely) al hough non-signi ican ly.
A e in e en ion he e we e no signi ican di e ences in he p e alence, incidence o emission o
obesi y be ween g oups o schools.
Table 3. P e alence, incidence and emission o o e weigh and obesi y a e in e en ion.
Measu e Sample,
n
Baseline,
n (%)
Pos -in e en ion,
n (%)
Unadjus ed
change (%) a
Adjus ed Odds
(95% CI) b p
O e weigh P e alence
Con ol 143 24 (16.8) 40 (28.0) 11.2 1.00 0.880
In e en ion 151 46 (30.5) 44 (29.1) 1.4 1.10 (0.31; 3.97)
Incidence
Con ol 98 - 18 (18.4) 18.4 1.00 0.037
In e en ion 89 - 5 (5.6) 5.6 0.25 (0.07; 0.92)
Remission
Con ol 24 - 3 (12.5) 12.5 1.00 0.835
In e en ion 46 - 6 (13.0) 13.0 1.24 (0.17; 9.31)
Obesi y P e alence
Con ol 143 14 (9.8) 13 (9.1) 0.7 1.00 0.493
In e en ion 151 9 (6.0) 10 (6.6) 0.6 0.42 (0.04; 4.94)
Incidence
Con ol 24 - 4 (16.7) 16.7 1.00 0.979
In e en ion 46 - 5 (10.9) 10.9 0.98 (0.16; 5.64)
Remission
Con ol 14 - 5 (35.7) 35.7 1.00 0.493
In e en ion 9 - 4 (44.4) 44.4 2.36 (0.20; 27.45)
N = 294 (indi iduals wi h da a a baseline and pos -in e en ion. Sample sizes o p e alence included all 294
pa icipan s, whe eas sample sizes o incidence and emission we e dependen on ini ial weigh s a us (e.g.,
incidence o o e weigh was based only on indi iduals who we e no mal weigh a baseline, whe eas emission
o obesi y was conside ed using only indi iduals who we e obese a baseline). - indica es no da a a ailable;
a Da a a e unadjus ed pe cen ages; b Odds we e adjus ed o gende , age, baseline o al ene gy in ake, pa en s’
educa ion and baseline BMI z-sco e.
In . J. En i on. Res. Public Heal h 2012, 9
1363
4. Discussion
Ou s udy shows ha a nu i ion p og am, deli e ed and augh by in-se ice eache s ained in
nu i ion, is e ec i e in lowe ing he inc ease in he incidence o o e weigh among schoolchild en.
In e ened child en ha e a signi ican ly lowe inc ease in BMI and a lowe p opo ion o child en
become o e weigh . This is pa icula ly impo an as i p o ides u he e idence o he WHO
ecommenda ions o schools o include bo h die a y and physical ac i i y componen s in he
cu iculum augh by ained eache s [24].
Ou s udy has some weaknesses ha should be men ioned. We did no explo e di e ences be ween
schools selec ed and no selec ed gi en human and ma e ials cons ain s, al hough he schools we e
om he same geog aphical a ea and o he bes o ou knowledge, no da a is a ailable epo ing
sociodemog aphic and income signi ican di e ences be ween schools om his pa icula a ea.
In addi ion, o he bes o ou knowledge he e we e no o he “an i-obesi y” p og ams going on in
in e en ion and con ol schools.
I is possible ha ou sample size was no la ge enough o de ec o he signi ican di e ences han
hose epo ed. Fu he mo e, we ailed o ob ain iden ically equi alen g oups a e andomiza ion,
namely in pa en s educa ion le el and child en’s heigh , mainly because we andomized by school and
no by subjec s, aiming o a oid c oss con amina ion be ween in e en ion and con ol g oups.
Ne e heless, hese di e ences we e aken in accoun in all o he s a is ical models. Also, physical
ac i i y le els we e ob ained upon sel - epo ed da a making ecall bias and o e es ima ion possible.
Howe e , he ques ionnai e was alida ed o Po uguese adolescen s [19] and we ha e no eason o
assume hese biases would a ec g oups di e en ly. Fu u e s udies would be imp o ed by using
accele ome y o be e classi y physical ac i i y.
On he o he hand, he p esen s udy has impo an s eng hs ha also should be acknowledged.
Fi s , o he bes o ou knowledge his is he i s s udy ha included he p og am in he p og ession o
eaching ca ee . This p obably allowed eache s o inc ease hei mo i a ion in he deli e y o he
in e en ion. I would be desi able ha o he simila p og ams could be ecognized on he ca ee
p og ession o eache s in Po ugal and o he coun ies in o de o engage and ewa d eache s o hei
e o s. Second, he long e m eache s’ in-se ice aining and he subsequen ne wo k de eloped
be ween hemsel es, esea che s and child en. We know ha in Po ugal s uden s o educa ion cou ses
do no ha e heal h p omo ion in hei aining cu icula, nei he conside i s changes owa ds an
inc ease in heal h con en impo an [25]. In addi ion, eache s a e desc ibed as a g oup ha is
no able o de o e much ime and ene gy o p o ide obesi y p e en ion in e en ions as dedica ed
in e en ionis s [8]. Being awa e o his need and ha long- e m p og ams a e mo e e ec i e han hose
o sho du a ion [8] we p omo ed a six mon hs du a ion aining p og am expec ing eache s o become
also nu i ion educa o s. We belie e his pe iod allowed eache s o ecognize jus how impo an
heal hy ea ing habi s and physical ac i i y a e. Thi d, ou app oach was o s anda dize ecommenda ions
o eache s, allowing hem enough lexibili y o c ea e in e ac i e in e en ions and pedagogic
ins umen s o be used wi h child en. The eache s we e assessed a e hei in e en ion o ma e ial
and lessons de eloped, and i was ound ha he in o ma ion inco po a ed in hei lessons and/o he
quali y o eaching ma e ials did no a y be ween hem. This is in line wi h he de elopmen o
“schola ship o eaching and lea ning” [15] and is con a y o p e ious school-based in e en ions ha