www.jped.com.br
ORIGINAL
ARTICLE
Ultra-processed
food
consumption
in
children
from
a
Basic
Health
Unit
夽
Karen
Sparrenberger
a,∗,
Roberta
Roggia
Friedrich
a,
Mariana
Dihl
Schiffner
b,
Ilaine
Schuch
c,
Mário
Bernardes
Wagner
aaPostGraduatePrograminChildandAdolescentHealth,SchoolofMedicine,UniversidadeFederaldoRioGrandedoSul(UFRGS),
PortoAlegre,RS,Brazil
bDepartamentofNutrition,NutritionalAssessmentLaboratory,SchoolofMedicine,UniversidadeFederaldoRioGrandedoSul
(UFRGS),PortoAlegre,RS,Brazil
cDepartmentofNutrition,SchoolofMedicine,UniversidadeFederaldoRioGrandedoSul(UFRGS),CentreofFoodandNutrition
Studies(CESAN),HospitaldeClinicasdePortoAlegre,PortoAlegre,RS,Brazil
Received18September2014;accepted5January2015
Availableonline16June2015
KEYWORDS Foodintake; Nutritionalstatus; Children;
Fast-food
Abstract
Objectives: Toevaluatethecontributionofultra-processedfood(UPF)onthedietary consump-tionofchildrentreatedataBasicHealthUnitandtheassociatedfactors.
Methodology: Cross-sectional study carried outwithaconveniencesample of204children, aged2---10yearsold,inSouthernBrazil.Children’sfoodintakewasassessedusinga24-hrecall questionnaire.Fooditemswereclassifiedasminimallyprocessed,processedforculinaryuse, andultra-processed.Asemi-structuredquestionnairewasappliedtocollectsocio-demographic and anthropometricvariables. Overweightin children was classified using aZ score >2 for childrenyoungerthan5andZscore>+1forthoseagedbetween5and10years,usingthebody massindexforage.
Results: Overweight frequency was 34% (95% CI: 28---41%). Mean energy consumption was 1672.3kcal/day,with 47%(95%CI:45---49%)comingfromultra-processedfood.In the multi-ple linearregressionmodel,maternal education (r=0.23;p=0.001) andchildage(r=0.40; p<0.001) were factors associated with a greater percentage of UPF in the diet (r=0.42; p<0.001).Additionally,astatisticallysignificanttrendforhigherUPFconsumptionwasobserved whendatawerestratifiedbychildageandmaternaleducationallevel(p<0.001).
Conclusions: ThecontributionofUPFissignificantinchildren’sdietsandageappearstobean importantfactorfortheconsumptionofsuchproducts.
©2015SociedadeBrasileiradePediatria.PublishedbyElsevierEditoraLtda.Allrightsreserved.
夽
Pleasecitethisarticleas:SparrenbergerK,FriedrichRR,SchiffnerMD,SchuchI,Wagner MB.Ultra-processedfoodconsumptionin childrenfromaBasicHealthUnit.JPediatr(RioJ).2015;91:535---42.
∗Correspondingauthor.
E-mail:[email protected](K.Sparrenberger). http://dx.doi.org/10.1016/j.jped.2015.01.007
PALAVRAS-CHAVE Consumode alimentos; Estadonutricional; Crianc¸as;
Fast-foods
Consumodealimentosultraprocessadosentrecrianc¸asdeumaUnidadeBásicade
Saúde
Resumo
Objetivos: Avaliar a contribuic¸ão dos alimentos ultraprocessados noconsumo alimentar de crianc¸aspertencentesàárea deabrangênciade umaUnidadeBásicadeSaúdeeos fatores associados.
Metodologia: Estudotransversalcomamostradeconveniênciade204crianc¸as,entre2a10 anosdeidade,noSuldoBrasil.Oconsumoalimentardascrianc¸asfoiobtidopormeiodo Recor-datórioAlimentarde24horase,posteriormente,osalimentosforamclassificadosemalimentos minimamente processados, processados para culinária e ultraprocessados. Umquestionário semiestruturadofoi aplicadopara acoleta das variáveissociodemográficas e antropométri-cas.Oexcessodepesodascrianc¸asfoidefinidopormeiodoescoreZ>2paracrianc¸asmenores de5anoseescoreZ>+1paraaquelascomidadeentre5e10anossegundooÍndicedeMassa Corporalparaidade.
Resultados: Afrequênciadeexcessodepesofoide34%(IC95%:28%a41%).Oconsumomédiode energiafoide1.672,3kcal/dia,sendo47%(IC95%:45%a49%)provenientedosultraprocessados. Nomodeloderegressãolinearmúltipla,aescolaridadematerna(r=0,23;p=0,001)eaidadeda crianc¸a(r=0,40;p<0,001)foramassociadosàmaiorcontribuic¸ãopercentualdos ultraproces-sadosnaalimentac¸ão(R=0,42;p<0,001).Adicionalmentefoiobservadaumatendêncialinear significativaparamaiorconsumodeultraprocessadosquandoosdadosforamestratificadospela idadedacrianc¸aeníveldeescolaridadematerna(p<0,001).
Conclusões: Acontribuic¸ãodosultraprocessadoséexpressivanaalimentac¸ãoinfantileaidade dacrianc¸amostrou-secomofatorassociadomaisimportanteparaoconsumodestesprodutos. ©2015SociedadeBrasileiradePediatria.PublicadoporElsevierEditoraLtda.Todososdireitos reservados.
Introduction
The prevalence of obesity andnon-communicable chronic diseases (NCDs) associated with diet has grown at a fast pace,andratesinthepediatricpopulationareremarkable.1
According to the National Survey on Demographics and Health of 2006, a national overweight prevalence of 6.6% was recorded in children aged up to 5 years of age.2 However, the results of the Household
Bud-get Survey (HBS) showed that the overweight prevalence ranged from 25% to 40% in children aged between 5 and 9years.3
Scientific evidenceindicates that theincrease in over-weight rates and NCDs is due, among other factors, to the inversion of dietary patterns.4 This inversion is
char-acterizedbythe substitution oftraditionalfood byhighly processedandready-to-consumefoodsandbeverages.5
Ingeneral,ultra-processedfoods(UPFs)havehighenergy density,excessivetotalandsaturatedfat,higher concentra-tionsofsugarand/orsodium,andlowfibercontent.5---7Also,
duetotheircompositionandprocessing,theyare character-isticallyhyper-palatable,lessperishable,andarereadyfor consumption. Thus, theyhave a large commercial advan-tagewhencomparedtofreshorminimallyprocessedfood, inadditiontobeinglowercost.5
HBSdata indicatethat thediet ofBrazilian children is deficientinfruitsandvegetables.Italsoshowsan overcon-sumptionofcookies,coldcuts,beverageswithaddedsugar, sandwiches,andsnacks.8
Amongthefactors thatareassociatedwiththequality offoodinchildren’sdiet,parentalincomeandeducational levelareespeciallysignificant.Researchsuggeststhat high-qualitydiet is directly associated withhighereducational levelsandincome.9,10
Thereisevidencelinkingtheoccurrenceofoverweightin childhoodandearlydevelopmentofdiabetesmellitus, car-diovasculardisease,dyslipidemia,andhypertensioninadult life.11Thus,childhoodisacrucialperiodfortheprevention
of NCDs by encouraging and adoptinghealthy habits that tendtopersistduringadultlife.12Parentshavegreat
influ-enceonthe developmentofthese habitsby thechild, so theymustprovidepositiveexampleswithregardtohealthy eatingassociatedwithphysicalexercise.13
Therefore, the aim of this study was to evaluate the contribution of ultra-processed food in the dietary con-sumptionofchildrentreatedataBasicHealthUnitandits associatedfactors.
Methods
A descriptive, cross-sectionalstudy wasperformed witha conveniencesampleofchildrenaged2---10years,whohad previously scheduled appointmentsand weretreated at a BasicHealthUnit(BHU)inthecityofPortoAlegre,stateof RioGrandedoSul,Brazil.
FamilyHealthStrategyinaBasicHealthUnitofPortoAlegre, Brazil’’.Thesampleincluded204children,whichprovided a statistical power of 90% for this study,to test a differ-enceofmeanswithaneffectmagnitude(E/S)≥0.5standard deviationfor˛=0.05.Regardingthecategoricaldata,this
samplesizeprovidedapowerof80% inthecomparisonof proportionswithdifferences≥20%vs.40%for˛=0.05.
Onlyonechildperhousehold(samemotherorguardian, biologically related or not) was included in the study. When morethanonechild inthisage rangeandfromthe same household was treated at the BHU, the caregiver decidedwhowouldparticipateinthestudy.Exclusion crite-ria included thefollowing: physical incapacity toundergo anthropometric measurements, gastrointestinal tract or oropharyngealdisordersthatcausedchangesinthedietary consumption,andchildrenwithautismspectrumdisorders. Theworkteamconsistedofpreviouslytrained nutrition-istsandnutritionstudentsanddatacollectionoccurredfrom September2012toJuly2013.Theanthropometric measure-mentswereobtainedinduplicateusingstandardtechniques accordingtothe WorldHealthOrganization.14 Weight(kg)
wasmeasuredusingadigitalscalewithacapacityof200kg andaccuracyof50g;height(cm)wasmeasuredusinga sta-diometerfixedtothewall.Excessweight(overweightand obesity)wasestablishedforchildrenyoungerthan5years withaZscoreindicator>2,andforthoseaged5---10years, withaZscoreindicator>+1,accordingtotheBMIforage.15
TheanthropometricdatawereanalyzedusingAnthroPlus® software(Anthro®,WHOAnthroPlus,2007,USA).
To assess food intake, two 24-h food-recall question-naires. The first one was carried out through direct interviewswiththechild’smotherorcaregiver.The ques-tionswereaboutthechild’sfoodintakeonthedaybefore regardingthetype,methodofpreparation,brand,measures used,andquantitiesconsumed.Tominimizerecallbiasand improvethequalityofdataonthesizeoftheconsumed por-tions,a photoalbumshowingutensilsand fooditemswas used.16 Thesecond 24-hrecall wasobtained bytelephone
contactafteranintervalof1---8weekswiththesameperson thatansweredthefirstquestionnaire,onadaythatdidnot correspondtothesamedayofthepreviousweek,inorder tosubsequentlyestimatethemeanconsumption.
The conversion of the reportedfood items from home measurementsintogramswasbasedonthestandardization byPinheiro.17Nutrientanalysiswasperformedaccordingto
theBrazilian Table ofFood Composition (TabelaBrasileira deComposic¸ãodos Alimentos--- TACO),18 andalsothrough
labels,forthosefoodsnotlistedinthetable.Subsequently, the fooditems were grouped according tothe definitions proposedby Monteiro etal.5 asunprocessed or minimally
processedfoods (G1), processedfor culinary use (G2), or ultra-processed(G3).
The studied variables were as follows: (1) child: gen-der,age,weight,height,andfoodintake(calories,proteins, lipids,carbohydrates,fiber,sodium,saturatedfat, monoun-saturatedfat,polyunsaturatedfat,trans fat);(2)mother: age and educationallevel and (3) familyunit: per capita
income.
Macronutrients (carbohydrates, lipids, and proteins) were used for the analysis of the characteristics associ-atedwiththefoodcontributionaccordingtothedegreeof processinginthechildren’sdiet.
Maternal educational level was defined according to the number of years of schooling. This variable was dichotomizedas:<11years(uptoincompletehighschool)or
≥11years(completehighschooland/orcollege/university education).The familyper capitaincomewasassessedin reals(R$)andsubsequentlycategorizedasR$<500andR$
≥500.
For the analysis,the children were stratified into two groups:preschoolers (2---6years)andschool-aged children (7---10 years). The study protocol was approved by the ResearchEthicsCommitteeofHospitaldeClínicasdePorto AlegreunderNo.120124.
Quantitative datawereinitiallydescribedasmeanand standarddeviation.Inthepresenceofasymmetry,medians andinterquartilerange(P25;P75)wereused.Thenormality ofdistributionswastestedusingtheShapiro---Wilktest. Cat-egoricaldataweresummarizedusingabsoluteandrelative frequencies.Themean(standarderror)wasusedtodisplay thecentraltendencyofabsolutecontributionvariablesand thepercentageofnutrientintakeaccordingtothedegree offoodprocessing.
Student’s t-test was used to compare the quantita-tive variables and the chi-squared test was used for the comparison of proportions. In cases of asymmetry, the Mann---Whitneytestwasused.
Toassesstheindependentassociationofthestudyfactors that were significant in the univariate analysis, a multi-plelinearregression wasperformed using thepercentage contributionofultra-processedfood(UPF)asthedependent variable.
Additionally,ananalysisstratifiedbythemother’s educa-tionallevelandchild’sagewasperformed.Thelineartrend assessmentofthisstratificationinrelationtoUPF percent-agewasperformedthroughsimplelinearregressionandfor excessweight,throughlineartrendchi-squaredtest.
The level of statistical significance was considered as
p<0.05 in all tests. Data were double entered using the EpiData® software(Epi Info, Version 6, Statistics Program forPublicHealth,1995,USA),withconsistencycheck. Sta-tisticalanalyseswereperformedinSPSSsoftware(IBMSPSS StatisticsforWindows,Version20.0,2011,USA).
Results
Thefinalsampleconsistedof204children,withalossoffive, duetolackofcompletionofthe24-hrecallquestionnaire.As forthesamplecharacteristics,therewereahighernumber ofschool-aged children.Regardinggender,theproportions weresimilar.Excessweightfrequencyintheassessedsample was34%(95%CI:28---41%;Table1).
Inrelationtoenergyintake(Table2),onaverage, chil-drenconsumed1672.3kcal/day,and47%(95%CI:45---49%) werederivedfromG3.The contributionofG1 foodis sig-nificantduetotheavailabilityofessentialnutrientssuchas protein,fiber,andmonounsaturatedfat.IntheG3,amore significant contribution of lipids, carbohydrates, sodium, andtransfatwasobserved.
Table1 Sampledistributionaccordingtosociodemographicandanthropometriccharacteristics.
Characteristics Total Agegroup p
Preschooler School-aged
Gender,n(%)(n=204)
Female 102(50.0) 66(55.0) 36(43.4) 0.12
Age,years(n=204) 5.9±2.5 4.1±1.4 8.5±1.1
---Weight,kg(n=202) 26.4±10.8 20.2±5.6 35.2±10.4
---Height,cm(n=199) 199.3±16.5 107.9±10.7 134.9±8.4
---Nutritionalstatus,n(%)(n=199)
Normalweight 131(66.0) 81(70.0) 50(60.2) 0.16
Excessweight 68(34.0) 35(30.0) 33(39.8)
Maternalage,years(n=187) 34.8±8.1 33.3±8.3 37.1±7.2 0.001
Maternalschooling,n(%)(n=184)
<11yearsofschooling 66(36.0) 32(29.0) 34(46.0) 0.01
≥11yearsofschooling 118(64.0) 78(71.0) 40(54.0)
Percapitaincome,R$(n=182) 545.6(339.0;757.5) 533.3(349.9;757.50) 570.8(302.7;783.7) 0.51
Resultsareexpressedasmean±SD,frequency(%),andmedian(P25,P75).Student’st-test;chi-squared;p<0.05.
Table2 Absoluteandpercentagecontributiontothedailynutrientintakeaccordingtothedegreeoffoodprocessing.
Total(n=204) G1mean(SE) G2mean(SE) G3mean(SE)
Energy(kcal/d)
Absolute 1672.3(41.4) 761.8(21.3) 96.9(5.8) 813.6(31.0)
Percentage 100 47.0(1.0) 6.0(0.4) 47.0(1.1)
Protein(g/d)
Absolute 68(2.1) 48.1(1.6) 0.4(0.1) 19.6(1.0)
Percentage 100 70.6(1.1) 0.6(0.1) 28.8(1.1)
Lipids(g/d)
Absolute 56.2(1.8) 21.5(0.8) 6.3(0.3) 28.5(1.5)
Percentage 100 40.6(1.3) 12.0(0.6) 47.4(1.4)
Carbohydrate(g/d)
Absolute 206.5(5.8) 80.3(3.1) 10.2(1.2) 115.9(4.5)
Percentage 100 39.7(1.2) 4.9(0.5) 55.3(1.3)
Fibers(g/d)
Absolute 14.6(0.5) 10.2(0.4) 0.1(0.0) 4.3(0.3)
Percentage 100 68.7(1.3) 0.7(0.1) 30.6(1.3)
Sodium(mg/d)
Absolute 2215.7(71.2) 348.3(20.1) 721.7(24.8) 1147.6(58.7)
Percentage 100 17.3(0.8) 34.9(1.1) 47.8(1.4)
Saturatedfat(g/d)
Absolute 20.7(0.7) 9.5(0.4) 1.1(0.1) 10.1(0.5)
Percentage 100 47.4(1.4) 5.6(0.3) 47.0(1.4)
Monounsaturatedfat(g/d)
Absolute 13.8(0.5) 6.9(0.3) 1.5(0.1) 5.4(0.3)
Percentage 100 50.9(1.4) 12.2(0.6) 36.9(1.5)
Polyunsaturatedfat(g/d)
Absolute 9.5(0.3) 2.5(0.1) 3.6(0.2) 3.4(0.2)
Percentage 100 28.2(1.3) 38.1(1.6) 33.7(1.7)
Transfat(g/d)
Absolute 1.4(0.1) 0.3(0.0) 0.0(0.0) 1.0(0.1)
Percentage 100 29.5(1.8) 5.2(0.5) 65.3(1.9)
Table3 Characteristicsassociatedwiththepercentagecontributiontomacronutrientconsumptionaccordingtothedegree offoodprocessing.
n G1a G2b G3a
Agegroup
Preschooler 121 50.9(1.2) 5.5(0.6) 43.7(1.4)
School-aged 83 40.6(1.6) 4.7(0.5) 54.7(1.7)
p-value <0.001 0.60 <0.001
Gender
Male 102 45.7(1.5) 5.1(0.6) 49.2(1.6)
Female 102 47.7(1.4) 5.2(0.5) 47.1(1.5)
p-value 0.33 0.65 0.36
Nutritionalstatus,BMI/age
Normalweight 131 47.0(1.3) 4.8(0.4) 48.2(1.4)
Excessweight 68 45.2(1.7) 5.8(0.8) 49.0(2.0)
p-value 0.41 0.55 0.73
Maternalschooling
<11yearsofschooling 66 49.4(1.8) 5.8(0.8) 44.8(1.9)
≥11yearsofschooling 118 45.5(1.4) 4.7(0.5) 49.8(1.5)
p-value 0.09 0.11 0.04
Percapitaincome,R$
<500 77 46.8(1.6) 5.3(0.6) 48.0(1.7)
≥500 105 46.9(1.5) 4.7(0.5) 48.4(1.6)
p-value 0.95 0.05 0.85
G1,unprocessedorminimallyprocessedfood;G2,processedforculinaryuse;G3,ultra-processedfood;BMI,bodymassindex.Results areexpressedasmean(standarderror).
a Student’st-test. b Mann---Whitneytest.
p<0.05.
increaseinthechild’sage(p<0.001).Incontrast,the pro-portionofG3consumptionshowedadirectassociationwith theincreaseinthechild’sage(p<0.001).Intermsof gen-der,nutritionalstatus,andpercapitaincome,therewere nosignificantdifferencesinthecontributionofthedifferent groups.
When comparingthepercentageof macronutrient con-sumptionbythechildrenandmaternaleducationallevel,it wasobservedthatchildrenwhosemothershad<11yearsof schoolingtendedtoconsumemoreG1food(p=0.09),while thosewhosemothershad≥11yearsofschoolinghadahigher G3contributionintheirdiets(p=0.04).
Based on the multiple linear regression model, with the G3 contribution percentage as the dependent vari-able, the factors maternal educational level (r=0.23; p=0.001)andchild’sage(r=0.40;p<0.001)showeda mod-erate multiplecorrelation andweresignificant factorsfor higher contributionof G3 in the children’sdiets (r=0.42; p<0.001).
Additionally,anincreaseintheconsumptionofUPFwas observedwhendatawerestratifiedbymaternaleducational level and the child’s age. This finding reached statisti-calsignificanceat thelineartrendtest (p<0.001).Excess weightalso increasedwhen comparing extreme groups in this stratification (18% vs. 38%); however, in the linear trendanalysistheincreasereachedborderline significance (p=0.079).
Discussion
Based on the results, a high rate of excess weight was observed among the assessed children, thus making this nutritional problem in this age group noteworthy. The increasingoverweightandobesity ratesin childhoodhave beenreported inthe scientificliterature, andare consid-ered important predictors of obesity anddevelopment of NCDsinadultlife.19,20
Theexcessweightfrequency observedinthisstudy are consistentwiththeresultsobservedinBrazil.Similarvalues werefoundinaschool-basedsurveyconductedinthecity ofItajaí,stateofSantaCatarina,inwhichtheobserved fre-quencywas30%inchildrenaged6---11years.9Theresultsof
astudycarriedoutinareascoveredbyBasicHealthUnits inColombo,state of Parana,indicated lowerfrequencies, showing that 12% of children aged 2---5 years had excess weight.21
In this research, the caloric contribution fromG3 was higherthan that found for the Brazilian population, esti-matedat28%.6AsforthepopulationofCanada,thisvalue
becomes even more significant, representing 61% of the dailyenergy.7However,noneoftheabovementionedstudies
rich in fats and sugars, such as sandwich cookies, baked goods,candy,andsoftdrinks.22
Theresultsofthisstudyarealsosimilartothosefoundin otherstudies.InCanada,onestudycomparedalistofstaple foodconsistingoffooditemsfromG1plusG2andanother thatincludedonlyitemsfromG3.Thelistthatdidnot con-taintheultra-processedfoodshowedhigherproteincontent (19%vs.10%)andfiber(14.8gvs. 6.8g),loweramount of totalfat(33.8%vs.39.3%),freesugars(3.8%vs.18.6%),and sodium(3.1gvs. 3.8g).7 Brazilian studiesindicate similar
findingsregardingthesenutrients.5,23
Another negative aspect of UPF is its high sodium content.Excesssodiumintakeisassociatedwiththe devel-opment of hypertension.24 Blood pressure alterations in
childhoodareassociatedwiththisprobleminadultlife.25
Itshouldalsobenotedthatthesaltusedinfood prepara-tionoratthetableisaG2component.Therefore,addedto theintrinsicsodiumcontentofUPF,itincreasestheestimate ofdailysodiumintakebythechildreninthestudy.
AccordingtoSarnoetal.,24excessofsodiumconsumption
canbemotivatedbytheincreasedconsumptionofprocessed foods.AstudycarriedoutintheUSshowedthat,ofthetotal dailyconsumedsodium, 44%camefrombreads,processed meats, pizzas, soups, sandwiches, cheese, and pasta- or meat-basedmeals.26
In Brazil, the estimates do not differ from those in othercountries.The HBSdataindicatethatUPF consump-tion(pizza,processedmeats,snacks,sandwichcookies,and soda)wasassociatedwithhigherintakeofsodium.8
Vitolo et al.27 suggest that thereis a positive
associa-tionbetweensodiumintakeandabnormalbloodpressurein preschoolers.The researchershighlightthe factthat they didnotassessthesaltaddedtothepreparationsand, there-fore,theybelievethemainsourceofthisnutrientcamefrom UPF.
The resultsregardingfiberconsumptionfromUPF rein-forcewhathasbeenshownintheliterature,i.e.,foodsin thisgroupareextremelypoorsourcesofthisnutrient.5
Furthermore,UPFwerethemainsourceoftransfatfound inthechildren’sdiets.Thistypeoffatiswidelyusedinthe foodindustrytoimprove thephysicaland sensoryaspects oftheproducts.Excesstransfatconsumptionisassociated withincreasedLDL-cholesterol,risk of cardiovascular dis-ease,diabetes,andhypertension.28
Althoughtheresultsdidnotreachstatisticalsignificance atthetestoflineartrendbetweenthevariablesofexcess weightandthechild’sageandmaternaleducationallevel, whichmaybeduetothesmallnumberofchildrenineach group,itcan beobserved(Fig.1)that therewas increas-inggrowthinthefirstthreecategories.Similarly,thedata showninepidemiologicalstudiesindicatethattherewasa directassociationbetweenexcessweightinchildrenandthe maternaleducationallevel.19
Thehigherconsumptionofmacronutrientsderivedfrom G3 by schoolchildren, compared to preschoolers, can be explained by the fact that they have greater autonomy regardingfoodchoices.Therefore,theyaremorelikelyto suffertheinfluence ofthe environmentinwhich theyare inserted,whichoffersunhealthyfoodchoices.29
Inthepresentstudy,unlikewhatwasindicatedinsome references,9,30highermaternaleducationallevelwas
associ-atedwithincreasedcontributionofUPFinchildren’sdiets.
Percentage
Ultra-processed food consumption
p (trend)<0.001 p (trend)=0.079 <11 years 0 10 20 39 45 50 59 18 33 44 38 30 40 50 60 70 Preschooler School-aged <11 years
≥11 years ≥11 years Maternal educational level
Age group Excess weight
Figure1 Percentageofcontributionofultra-processedfoods andchildren’sexcessweightaccordingtomaternaleducational levelandagegroup.Ultra-processedfoodconsumptiontrend: simplelinearregression;excessweighttrend:chi-squaredtest oflineartrend.
However,this association withmaternal educational level showedlowmagnitude(r=0.23)andwasconsiderablylower than that observed between consumption of UPF and the child’sage(r=0.40).
Furthermore, unlike what has been reported by some studies,10 the present study did not find an association
betweenhigherconsumptionofUPFandlowerincome. How-ever,thisfindingmaybeexplainedbythefactthatasample ofusersofaBHUhasrelativelylowincomeintheeconomic context of society. Thus, this relative homogeneity leads to low variability regarding income in the group, making itdifficult(ormisrepresenting)todetectassociationswith this factor and other collinear factors (e.g., educational level).
Accordingtothe study byVitolo etal.,27 when
assess-inglow-incomechildreninacityinthemetropolitanregion of Porto Alegre, it was observed that UPF showed great contribution in the children’s diet. The most consumed food items were breads (78.8%), sugary drinks (75.6%), sweet snacks (63.2%), cookies (52.5%), cold cuts (42.9%), potato chips (17.7%), and instant noodles (11.0%). How-ever, it is noteworthy that most types of bread (e.g., Frenchbread)haveironandfolicacidintheircomposition, whichareimportantnutrientsinthechildren’sdiets.This typeoffood,in appropriateamounts,ispartof ahealthy diet.
However, there are indications that the increase in consumptionofultra-processedfoodaffectsboththe pop-ulationwiththelowestincomeandthosewiththehighest income.Also,itissuggestedthatthereductioninG1food consumption is more significant among those with higher income.
habits.Furthermore,theremaybearecallbias,considering thattheintervieweehadtoreportthefoodconsumptionof thepreviousday.
Itisconcluded,baseduponthisstudy,thatthe contribu-tion of UPF is significant in the assessed children’sdiets, demonstrating a poor quality regarding the presence of protective foods and nutrients, as well as health risk. Also,therewasahigherfrequencyofUPFconsumptionby schoolchildrenandchildrenofmotherswithhigher educa-tionallevels.Thehighfrequencyofexcessweightfoundin thestudypopulationisalsonoteworthy.
Therefore,theauthorsreinforcetheneedfor foodand nutritioneducationalstrategiesaimedatchildrenand par-ents, considering that childhood is an important period fortheencouragementanddevelopmentofhealthyeating habits.Itisalsoemphasizedthatfurtherstudiesshouldbe carriedout toassess theimpact ofUPF on thequalityof children’sdietandnutritionalstatus.
Conflicts
of
interest
Theauthorsdeclarenoconflictsofinterest.
Acknowledgments
Karen Sparrenberger received a CAPES Master’s degree scholarship.Theauthorsalsowouldliketothankthe Post-Graduate Program in Child and Adolescent Health of the SchoolofMedicine ofUniversidadeFederal doRioGrande doSul.
References
1.VernarelliJA,MitchellDC,HartmanTJ,RollsBJ.Dietaryenergy densityis associatedwithbody weightstatus and vegetable intakeinU.S.children.JNutr.2011;141:2204---10.
2.Brasil.MinistériodaSaúde.SecretáriadeCiência,Tecnologia eInsumosEstratégicos.DepartamentodeCiênciaeTecnologia. PesquisaNacionaldeDemografiaeSaúdedaCrianc¸aedaMulher (PNDS-2006).Brasília:MinistériodaSaúde;2009.
3.Brasil. Ministério do Planejamento, Orc¸amento e Gestão. PesquisadeOrc¸amentosFamiliares2008---2009:Antropometria eestado nutricionalde crianc¸as, adolescentes e adultos no Brasil.RiodeJaneiro:InstitutoBrasileirodeGeográfiae Estatís-tica;2010.
4.Enes CC, Slater B. Obesidade na adolescência e seus prin-cipais fatores determinantes. Rev Bras Epidemiol. 2010;13: 163---71.
5.MonteiroCA,LevyRB,ClaroRM,CastroIR,CannonG.Anew classificationoffoodsbasedontheextentandpurposeoftheir processing.CadSaudePublica.2010;26:2039---49.
6.Martins AP, Levy RB, Claro RM, Moubarac JC, Monteiro CA. Participac¸ãocrescentedeprodutosultraprocessadosnadieta brasileira(1987---2009).RevSaudePublica.2013;47:656---65. 7.MoubaracJC,MartinsAP,ClaroRM,LevyRB,CannonG,Monteiro
CA.Consumption of ultra-processed foods and likely impact onhumanhealth.EvidencefromCanada.PublicHealthNutr. 2013;16:2240---8.
8.Brasil. Ministério do Planejamento, Orc¸amento e Gestão. PesquisadeOrc¸amentosFamiliares2008---2009:Análisedo con-sumo alimentar pessoal no Brasil. Rio de Janeiro: Instituto BrasileirodeGeográfiaeEstatística;2010.
9.Momm N, Höfelmann DA. Qualidade da dieta e fatores associados em crianc¸asmatriculadas em umaescola munici-pal de Itajaí, Santa Catarina.Cad SaudeColetiva. 2014;22: 32---9.
10.ZarnowieckiDM,DollmanJ,ParlettaN.Associationsbetween predictors of children’s dietary intake and socioeconomic position: a systematic review of the literature. Obes Rev. 2014;15:375---91.
11.Reilly JJ,Kelly J.Long-term impact ofoverweightand obe-sityinchildhoodandadolescenceonmorbidityandpremature mortalityinadulthood:systematicreview.IntJObes.2011;35: 891---8.
12.PetersJ,DollmanJ,PetkovJ,ParlettaN.Associationsbetween parentingstylesandnutritionknowledgeand2---5-year-old chil-dren’sfruit,vegetableandnon-corefoodconsumption.Public HealthNutr.2013;16:1979---87.
13.FriedrichRR,SchuchI,WagnerMB.Efeitodeintervenc¸õessobre oíndice demassacorporalem escolares.RevSaúdePública. 2012;46:551---60.
14. World Health Organization (WHO). Physical status: the use and interpretation of anthropometry: report of a WHO Expert Committee. Geneva: World Health Organization; 1995.
15.World Health Organization (WHO). Department of Nutri-tion for Health and, Development Child Growth Stan-dards: length/height-for-age, weight-for age, weight-for-length, weight-for-height and body mass index-for-age. Methodsanddevelopment.Geneva:WorldHealthOrganization; 2006.
16.ZabotoCB.Registrofotográficoparainquéritosdietéticos: uten-sílioseporc¸ões.Campinas:NEPA-UNICAMP;1996.
17.PinheiroAB.Tabelaparaavaliac¸ãodoconsumoalimentarem medidascaseiras.4thed.SãoPaulo:Atheneu;2004.
18.Nepa-Unicamp.TabelaBrasileiradeComposic¸ãodosAlimentos (TACO).4thed.Campinas:NEPA-UNICAMP;2011.
19.Silveira JA, Colugnati FA, Cocetti M, Taddei JA. Secular trendsand factors associatedwithoverweightamong Brazil-ianpreschoolchildren:PNSN-1989,PNDS-1996,and2006/07.J Pediatr(RioJ).2014;90:258---66.
20.ParkMH,FalconerC,VinerRM,KinraS.Theimpactofchildhood obesityonmorbidityandmortalityinadulthood:asystematic review.ObesRev.2012;13:985---1000.
21.Monteiro F, Schmidt ST, Costa IB, Almeida CC, Matuda NS. BolsaFamília:inseguranc¸aalimentarenutricionaldecrianc¸as menores de cinco anos. Cienc Saude Coletiva. 2014;19: 1347---58.
22.NobreLN,LamounierJA,FranceschiniSC.Preschoolchildren dietary patterns and associated factors. J Pediatr (Rio J). 2012;88:129---36.
23.Monteiro CA, Levy RB, Claro RM, de Castro IR, Cannon G. Increasing consumption of ultra-processed foods and likely impactonhumanhealth:evidencefromBrazil.PublicHealth Nutr.2011;14:5---13.
24.SarnoF,ClaroRM,LevyRB,BandoniDH,MonteiroCA.Estimativa deconsumodesódiopelapopulac¸ãobrasileira,2008---2009.Rev SaudePublica.2013;47:571---8.
25.PintoSL,SilvaRdC,PrioreSE,AssisAM,PintoEdeJ. Prevalên-ciade pré-hipertensãoede hipertensãoarterial eavaliac¸ão de fatores associados em crianc¸as e adolescentes de esco-las públicas de Salvador, Bahia, Brasil. Cad Saude Publica. 2011;27:1065---75.
26.CentersforDiseaseControlandPrevention(CDC).Vitalsigns: foodcategoriescontributingthemosttosodiumconsumption --- United States, 2007---2008. MMWR Morb Mortal Wkly Rep. 2012;61:92---8.
28.SantosRD,GagliardiAC,XavierHT,MagnoniCD,CassaniR, Lot-tenbergAM,et al.I Diretriz sobreo consumodegorduras e saúdecardiovascular.ArqBrasCardiol.2013;100:1---40. 29.Rossi A,Moreira EA,Rauen MS.Determinantesdo
comporta-mento alimentar: uma revisão com enfoquena família.Rev Nutric¸ão.2008;21:739---48.
