FINAL REPORT - Understanding Eating Behaviors and ... · styles. Conclusions: GMFR’s definitions...
Transcript of FINAL REPORT - Understanding Eating Behaviors and ... · styles. Conclusions: GMFR’s definitions...
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStylesChristineAlbertelli;DanaNotte,MS,RD,CD;AmyNickerson,MS,RDN;Dr.PeterCallas,Ph.D.
UniversityofVermont,BurlingtonVTandGreenMountainatFoxRuninLudlow,VTMasterofScienceinDieteticsEvidenced-BasedResearchProjectFinalPaper
May2017
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 2
ABSTRACT
Background:GreenMountainatFoxRun(GMFR)isanon-diet,mindfulness-basedwellness
retreatforwomenwhostrugglewitheatingandweight.Partoftheprogramistohelp
participantsdeterminetheir“eatingstyle(s)”,whichinformsappropriatebehavioralstrategies
thatguidethemtowardsamorehealth-supportive,“skillful”eatingstyle.Currently,thereare
nostandardeatingstyledefinitionsnoristhereavalidatedtoolfortheiridentification.The
objectiveofthestudywastoevaluateeatingbehaviorsandattitudesofparticipantstomore
accuratelydefineeatingstylesandprovidenextstepsforcreatingavalidatedassessmenttool.
Methods:AliteraturereviewoneatingtypesandpatternsandGMFR’seatingstyles(skillful,
chaotic,dietmentality,emotional,disconnected/habitual,anddazedandconfused)wereused
todevelopaninitialsurveythatwasemailedto185incomingparticipantsenrolledinGMFR’s
program.Qualitativeandquantitativeanalyseswereusedtoobservetrendsamongeating
stylesandevaluatetheinternalconsistencyandreliabilityofthesurvey,respectively.
Results:Theskillfuleatingstylewasnegativelycorrelatedwiththeotherfivestyles.Acceptable
internalconsistencywasobservedinallGMFR’seatingstylesexcept“chaotic”(Cronbach’s
alpha>0.6).Theexploratoryfactoranalysisrevealedsixdistinctfactors;threeofthose
identifiedfactorscorrespondedwiththedietmentality,emotionalanddazedandconfused
styles.
Conclusions:GMFR’sdefinitionsofdietmentality,emotional,anddazedandconfusedstyles
areclearlydefinedanddemonstratedreliability.Theremainingthreestylesrequireadditional
exploration.Further,largersamplesizesareneededtovalidatetheinitialeatingstylessurvey.
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 3
INRODUCTIONANDLITERATUREREVIEW
Thepersistentincreasein“obesity”ratesovermultipledecadesiswelldocumentedand
acknowledgedbymedicalprofessionalsandcommunitymembersacrosstheglobe.Research
hasdemonstratedthatalargerbodysizeisassociatedwithmanyotherchronicdiseases,such
ashighbloodpressure,diabetes,cardiovasculardisease,sleepapneaandcancer.1Eating
patternsandothermodifiableriskfactorscontributetotheprevalenceofthesechronic
diseasesandthusresearcharoundeatingbehaviorshasgrownimmensely.Additionally,
westernizedsocietiesarerespondingwithanintensefocusonweightlosstocombatthe
growing“epidemic,”buttheresearchdoesnotentirelydefendthetraditionalapproachto
weightloss.2Datasupporttheassociationthatdietaryrestraintmayresultinfutureweightgain
anddisorderedeatingbehaviors.2–5Thecomplexityofeatingbehaviorsandassociated
outcomeshascontributedtothegrowinginterestinthisareaofresearch.
Foryears,nutritionandpsychologyresearchershavebeenanalyzingeatingbehaviors,
beliefsandattitudestobetterunderstandcharacteristicsofdifferenttypesofeatingstylesand
dietarypatterns.“Eatingstyles”aregroupsofeatingbehaviors,attitudesandbeliefsthattend
tooccurtogetherintrendsorcategories.Practitionershavecreatedtheirownlistsand
definitionsofeatingstylesandorreferenceexistingonesinpractice.Onepurposeof
identifyingeatingstylesistobetterunderstandpotentialcausesofproblematiceatingbehavior
andinformbehavioralinterventions.Manypractitionershavealsodevelopedquestionnairesto
measuretheseeatingbehaviorsandattitudesinindividualsandpopulationsinanattemptto
providethelinkbetweeneatingbehaviorsandoutcomes(refertoTable1foralistofauthors
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 4
andfacilities,definitionsandquestionnaires).Moreover,todaythereareavarietyofeating
theoriesandstylesrecognizedbyresearchersandpractitionersacrossmanydisciplines.
Intheearlyyearsofthisresearch,Strienandcolleaguesdevelopedanassessmenttool
knownastheDutchEatingBehaviorQuestionnaire(DEBQ).6Thistoolwasdesignedtoenhance
understandingofeatingpatternsamongindividualswithalargerbodysize.Usingthreespecific
psychology-basedtheoriesofeatingbehaviorsasthefoundation,threescaleswerecreated:
emotional,externalandrestrainedeating.Emotionaleatingorthepsychosomatictheorywas
describedaseatinginresponsetoemotions,suchasanger,fearoranxiety,lossofappetite,
whensomeindividualsconsumedlargeamountsoffoodintheformofabinge.Theexternal
eatingphenomenoniseatinginresponsetoexternalstimuliregardlessofhungerlevels,suchas
presenceorevensmelloffood.Therestrainedeatingtheoryhassimilarqualitiestothe
traditionalweightlossapproachandischaracterizedbydietingandrestriction,whichatintense
levelscanleadtoexternalandemotionaleating.Althoughdated,thatstudyoutlinesthree
scalesoreatingtheoriesthatremaininmanylistsofeatingstylestodayandpresentstheidea
thatcharacteristicsofeatingstylesmayoverlapsuchasrestrainedeatingtheory.
Manyotherquestionnaireshavebeendevelopedusingcomparableconstructsand
theoriestolinkeatingbehaviorswithassociatedoutcomes,primarilyrelationshipsbetween
eatingbehaviorsandbodysize,andtoidentifyconstructsthatcapturespecificeatingstyles.7
Multiplestudiessupportedthecausalrelationshipbetweendietaryrestraintandovereating
and,thus,itbecameanimportantconstructandbehaviortoexplore.8Williamsonand
colleaguesconductedastudytotestthevalidityofmultiplequestionnairesthatmeasurethe
dietaryrestraintconstruct,definedas“theintentiontorestrictfoodintakeinordertocontrol
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 5
bodyweight”.Theirresearchrevealedthatthreeofthefoureatingstylequestionnaires[the
DEBQ,theEatingInventory(EI)/Three-FactorEatingQuestionnaireandtheCurrentDieting
Questionnaire(CDQ)],werevalidindicatorsofcurrentdietingandweightsuppression.
However,itwasconcludedthatthefourquestionnairesdidnotmeasurethesametheoretical
construct.8Astheresearchgrew,randomizedcontrolledtrials,factoranalysesandothertests
ofvalidityrevealedthatquestionnaires,whichwereintendedtomeasureseeminglysimilar
eatingstyleconstructs,werenotinterchangeableanddidnotexhibitconstructvalidity.The
fundamentalconclusionfromthesestudiesisthatmeasuringeatingstyleconstructsisan
incrediblycomplexprocess.Nonetheless,thesetoolsandquestionnairesarethelinkbetween
understandingtherelationshipbetweeneatingbehaviorandhealthoutcomes.An
understandingofeatingstylesandvalidquestionnairesareneededtobeginprovidingthe
frameworkforstandardizedapproachesatalllevels,frompolicytointerpersonalinterventions.
ThePresentStudy
Asnotedpreviously,theresponsetothe“obesityepidemic”inwesternsocietieshas
beenaweightfocuseddietculturethatwasusedasaframeworkforweightlossinterventions.
Thetraditionalweightlossapproachischaracterizedbyalteredeatingbehaviors,intheformof
dieting,caloriecountingandrestriction,andincreasedexercisetoachieveanidealbodyweight
andshape.2,9However,researchdemonstratesthatthetraditionalapproachhasbeen
unsuccessfulinmaintainingthegoalofweightloss.Forexample,Pietilaninenetal.foundthat
intentionalweightlossattemptswereindependentlycorrelatedwithfutureweightgain.3
Furthermore,datastronglysuggestthatdietaryrestrictionandovervaluationofweightand
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 6
bodyshapepredictsbingeeating.4Thisisevidenceshowingthatthetraditionalapproachto
weightlossmayleadtodetrimentaloutcomes.
Individualswithalargebodysizehavebecomesubjecttoanegativeweightbiasin
clinicalandcommunitysettings.10Experiencesofweightbiasorinternalizedweightstigma
increaseanindividual’sriskforimpairedoverallpsychologicalhealth(anxiety,depression,low
self-esteem,bodydissatisfaction)andisassociatedwithdisorderedandunhealthyeating
behaviors,suchasincreasedintakeandlowermotivationforexercise.Moreover,therehas
beenariseinratesofeatingdisordersoverthepasttwodecades.11
Researchexplainingthenegativephysiologicalandpsychologicaleffectsofdieting,
restrictionandintentionalweightlosshasopeneddoorsfornewtreatmentapproaches,such
asthenon-dietapproach.Thenon-dietapproachmaybesuperiortotraditionalweightloss
practicesinachievinglongtermbehaviorchangerelatedtodietaryhabitsandphysicalactivity.2
AsdefinedbyCliffordetal.,thenon-dietapproachencompassesseveraloverlapping
perspectivesofhealthpromotion,includingHealthatEverySize(HAES),theSatterEating
CompetenceModel(ecSatter),IntuitiveEatingandMindfulEatingPractices.TheHAES
paradigmpromotesweightinclusivity,adoptionofhealthpromotingbehaviorsforhealth
ratherthanweight,andencouragespeopletoeatforreasonsincludingbutnotlimitedto
hunger,nutritionalpurposesandpleasure.2,12EcSatter,derivedfromthespectrumofeating
attitudesandbehaviors,isgroundedintheprinciplethatusinginternalcues,hunger,appetite
andsatietytoguidefoodchoiceswillresultinanappropriatebodyweight.Eatingcompetence,
asdefinedbyecSatter,incorporatesfourprinciples:attitudes,foodacceptanceskills,internal
regulationskillsandcontextualskills(ie.planning,acquiringandpreparingfood).2,13
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 7
Additionally,theIntuitiveandMindfulEatingPracticeshavemanyfeaturesthatoverlap,yetare
recognizedindividuallybysomepractitioners.IntuitiveEatingisallowingoneselftoeatwhen
hungrywithnorestrictionoffoodchoices,eatingtofuelthebodyforphysicalreasonsand
usinginternalcuestoguidewhenandhowmuchtoeat.2,9,14Comparatively,MindfulEatingis
the“nonjudgmentalawarenessofphysicalandemotionalsensationswhileeating”byusing
one’ssensesandlearningtobeawareofhungerandsatietycuestoguidefoodchoicesand
eatingbehavior.15,16,17
Asaresultoftheongoingfailureoftraditionalweightlossprograms,manypractitioners
andfacilitieshaveadoptedanon-dietapproach.Intheprocess,attemptshavebeenmadeto
definerelevanteatingstyles.Forexample,GreenMountainatFoxRun(GMFR),anon-
traditionalwellnessretreatlocatedinsouthernVermont,hasbeenusingthenon-dietapproach
since1997tohelpwomenwhostrugglewithweight,eatingandbodyimage.18Theirmulti-
prongedprogramconsistsofwellnessassessments;workshopsandeducationalgroups;courses
suchasSelf-care&Compassion,RedefiningHealthyEating,MindfulEating,etc.;fitnessand
recreationalclassesandactivities;culinarydemonstrations;andmuchmore.18Basedonmany
yearsofworkwithwomenwhostrugglewithweightandtheirrelationshipwithfood,the
GMFRpractitionersestablishedalistofeatingstylesanddefinitions(seeAppendixBforGMFR’s
listofeatingstyles).Fromthesedefinitions,a“GuidetoEatingStylesChecklist”wascreated
andisusedinoneoftheprogram’sclasses,“IdentifyingYourEatingStyle”.Theclassaimsto
helpparticipantsdeterminetheireatingstyle.Theidentifiedeatingstyleguidesthestrategies
usedbyaregistereddietitian(RD)tofacilitatebehaviorchangetowardsamorehealth
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 8
supportive“skillful”eatingstyle,whichencouragesahealthpromotingrelationshipwithfood,
bodyandmind.
Whileunderstandingeatingbehaviorsandattitudesisimportantforoverallhealthandwell-
being,thereiscurrentlynoreliablemeasurementtool.Thecurrentsetoftoolsattemptto
providealinkbetweeneatingbehaviorsandoutcomesandguidepeopletowardsanon-diet
approachtohealthandeating.Althoughthesetoolsarevaluable,theabsenceofuniversally
acceptable,validateddefinitionsispreventingtheareaofresearchandpracticefromfurther
development.Manyofthetoolsfocusononespecificeatingstyle,whichhelpstorecognize
certaincharacteristicsinanindividual’seatingpattern,suchasabilitytorecognizehungerand
satietycues.Forexample,theIntuitiveEatingScalecreatedbyTylkaonlyintendstomeasure
IntuitiveEatingbehaviorsandattitudes.SimilarlytheMindfulEatingQuestionnairesaimto
measureafewconstructsorfactors,suchasawareness,externalcuesandemotional
response.14,19Thentherearesometoolsthatincorporateavarietyofeatingstyles,orcycles,
suchasthe“AmIHungry?”EatingCycleAssessment.20Thistoolidentifiesfoureatingstylesor
constructs(Instinctive,Overeating,RestrictiveandtheEat-Repent-Repeatcycle);yetother
practitionershaveidentifiedstylesbeyondthese.
Manyoftheeatingstyledefinitionsandquestionnairesusedinpractice,suchasGMFR’s
EatingStylesChecklist,arenotvalidated.Therearenostandarddefinitionsforeatingstylesnor
isthereavalidatedtoolfortheiridentification.Thepracticeofidentifyingeatingstyleshas
madelargestrides,however,theneedforfurtherresearchisuniversallydocumented.
Clear,validatedstandarddefinitionsofeatingstylesareneededsothatprofessionalscan
helpclientsunderstandbeliefsandattitudesthatcontributetotheireatingbehaviors.
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 9
Professionalsalsoneedtocreatepracticalstrategiestohelpindividualsovercomeproblematic
behaviorsaroundeatingandfacilitatelastingchanges.GMFR’slong-termgoalistocreatea
user-friendlyquestionnairethatisbasedonvalidatedterminology.Thequestionnairewill
encompassalleatingstyles,andbeusedtodevelopstandardizedapproachesandstrategiesto
workwithclients.Further,suchatoolwillhelpotherpractitionersandfacilitiesadoptsimilar
effectiveapproachesandinterventions.
ThefirststeptowardsGMFR’slongtermgoalofavalidatedmeasureistoclearlydefine
eatingstylesusedintheirprogram.Thepurposeofthepresentstudywastoevaluateeating
behaviorsandattitudesofparticipantstomoreaccuratelydefineeatingstylesandprovide
insightintonextstepsforcreatingavalidatedassessmenttool.Theaimsoftheprojectwereto
investigatetrendsinbehaviorsandattitudesrelatedtoeating;andtoidentifyanyeating
characteristicsorbehaviorsthataremissingfromthecurrentlistofGMFR’seatingstyles.
Additionally,thestudysoughttocollectpreliminarydataonthereliabilityandvalidationofa
surveytool.Theinformationobtainedfromthepresentstudywillbeusedtodetermineifthe
behaviorsandattitudesattachedtoeacheatingstylecorrespondedcorrectlytotheeatingstyle
inwhichitwaslisted.Theresultsoftheanalysiswillbeusedtoinformarevisionofcurrent
GMFR’sEatingStylesChecklist,whichisusedasthebasisforaclassthatseekstoincrease
participantawarenessoftheirpersonaleatingstyles,andinformbestpracticeforcoaching
theseindividualstoachievemoreMindfulandIntuitiveEatingstyles.
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 10
METHODS
LiteratureReviewandQuestionnaireDevelopment
Intheformofadescriptivesurveyresearchdesign,thefirststepofthepresentstudy
wasaliteraturereviewtolearnabouttypesandpatternsofeatingstylesdescribedbyother
professionals.GMFR’soriginallistofeatingstyleswascomparedagainstacollectionof
publisheddefinitionsandcorrespondingquestionnairestoidentifythecommonconstructsand
gapsinbehaviorsandattitudes.Thatinformedthedevelopmentoftheeatingstylessurvey,
whichwastestedtoasamplepopulationofGMFR’sparticipantsinthepresentstudy.
Thestatementsbelowdescribethedefinitionsandquestionnairesgleanedfromthe
literaturethatwereprimarilyconsideredtodescribeeacheatingstyleconstruct.However,all
definitions,scales,questionnaires,modelsandconstructswerereviewedcomprehensivelyfor
similaritiesanddifferencestoinformthecreationofthesurveydevelopedandtestedinthe
presentstudy(seeAppendixCfortheGMFREatingStylesSurvey).Forexample,todevelop
questionsthatcapturetheskillfuleatingstyleconstruct,theinvestigatorconsideredsimilar
constructsdescribedbyTribole(IntuitiveEaterdefinition);Tylka[theIntuitiveEatingScale
(IES)];May(theInstinctiveandMindfulEatingCycles);andecSattermodel.9,13,14,21Similarly,the
dietmentalityconstructwascomparedtoeachofthedietingpersonalitysubtypesdescribedby
Tribole,theCarefulEaterandProfessionalDieter,andtheRestrictiveandEat-Repent-Repeat
CyclesbyMay.9,21Additionally,atotaloffiveconstructsandonequestionnairewereusedto
formthequestionsfortheemotionaleatingstyleconstructintheEatingStylesSurvey[1.
EmotionalUnconscioussubtypebyTribole;2-4.theBingeEating,OvereatingandEat-Repent-
RepeatCyclesbyMay;5.theEmotionalResponseconstructintheMindfulEatingQuestionnaire
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 11
(MEQ);andthe6.EmotionalEatingScale(EES)].5,9,19,21–23Lastly,GMFR’schaotic,
disconnected/habitualanddazedandconfusedeatingstyleswerewidelyanalyzedagainstallof
Tribole’ssubtypesoftheUnconsciousEaterpersonality(Emotional,Chaotic,Waste-Not,and
Refuse-Not),manyoftheeatingcyclesbyMay,thehabitualeatingdefinitionbyTylka,the
MEQ,andtheMindfulEatingScale(MES).5,9,14,21,22
Afterthatcomprehensiveanalysis,theinvestigatorsagreedthatGMFR’soriginalsix
eatingstyleswereinclusiveandcapturedtheessenceofawidecontinuumofeatingbehaviors
andattitudes.Next,threeobjectivequestionsforeachconstructweredevelopedandreviewed
bytheauthors,including;skillful,dietmentality,chaotic,emotional,disconnectedorhabitual,
anddazedandconfused.Theobjectivequestions,numbers2-19,weredevelopedusinga5-
pointLikertscale(1-5;when1=stronglydisagree,2=disagree,3=neitheragreenordisagree,
4=agreeand5=stronglyagree).Additionally,threesubjectivequestionsweredevelopedto
elicitfeedbackonanyeatingcharacteristics,behaviorsorattitudesthatmaynothavebeen
capturedorrepresentedintheobjectivequestions(questions1,20-21).
ParticipantsandRecruitment
ThenontraditionalweightmanagementprogramatGMFRtargetswomenintheUnited
Statesaged18+whostrugglewithweight,bodyimageandtheirrelationshipwithfood.
Subjectswererecruitedfromthepopulationofnewandreturningparticipantswhoenrolledin
programatGMFRbetweenNovember7,2016andFebruary15,2017(3-months).Participants
inthestudywerewomenaged19-75yearsold(average=50yearsofage).Uponregistration
foraprogramatGMFR,incomingparticipantsreceivedalinktothesurveyandacorresponding
researchinformationsheetviaemail.Theactofclickingonthesurveylinkwastakenasimplied
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 12
consent.Theresearchinformationsheetprovideddetailsaboutthepurpose,proceduresand
potentialrisksversusbenefitsforparticipatingintheresearch,measurestakentoensure
participants’confidentialityandcontactinformationforfollowup(SeeResearchInformation
SheetinAppendixD).Participationinthesurveywasvoluntaryandsubjectswerenotifiedthat
theycouldwithdrawatanytime.ThestudywasapprovedbytheUniversityofVermont’s
InstitutionalReviewBoard(IRB)CommitteeonHumanResearchintheBehavioralandSocial
Sciences(CHRBSS).
TheEatingStyleSurveywasadministeredusingtheelectronicsurveysoftwareprogram,
Salesforce.Overthecourseofthe3-monthdatacollectionperiod,thesurveywasdistributedto
185individualsandcompletedby62women.Uponcompletionofthesurvey,responseswere
sentanonymouslytoadministrativepersonnelatGMFR,compiledintoanelectronicfileand
thenanalyzedbytheinvestigators.
Qualitativeandquantitativeanalyseswereperformedtoachievethestatedgoalsandaims
oftheproject.Thequalitativeanalysisattemptedtoidentifyanygapsinbehaviorsand
attitudesbycompilinggeneralobservationsfromtheresponsestothesubjectivequestions.
StatisticalAnalysis
QuantitativeanalyseswereperformedusingIBMSPSSStatistics,Version24.Descriptive
datawererecordedasmeansandpercentages.Themeanscoreforeachobjectivequestionon
thesurvey(questions2-19)wasrecordedtoinformtheaverageresponsetoeachquestion
amongthesamplepopulation.Forexample,allresponsestoquestionnumbertwowereused
todeterminehowonaveragethesamplepopulationrespondedtothatquestion.Additionally,
thefrequencyofresponsesforeachitemontheLikertscalewerecomputedaspercentagesfor
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 13
eachquestion.Pearson’scorrelationswerealsocomputedtoexaminerelationshipsbetween
theeatingstyleconstructs.
Cronbach’salphawascomputedtoassessthereliabilityandorinternalconsistencyforeach
construct’ssetofquestions(ie.Questions5-7forchaotic).24Morespecifically,ittestswhether
thesetofsurveyquestionsforeachconstructwereconsistentinmeasuringthesameeating
styleconstruct,giventhatareliablemeasurecontainslittletonoamountofmeasurement
error.25Thestrengthoftheconsistencywascomputedasanalphacoefficient,whichranges
from0-1(0=nocorrelation,1=allitemshavehighco-variances).Intheliterature,thereare
variedreportsontheacceptablevaluesofthealphacoefficient,rangingfrom0.60–0.95;the
alphacoefficientscomputedinthepresentstudywereconsideredacceptablevaluesbetween
therangeof0.60–0.90.25,26
Exploratoryfactoranalysiswasusedtoidentifytheunderlyingrelationshipsbetweenthe
variables(characteristics,behaviorsandattitudesofeating)andthusdeterminethemain
constructsofeatingstyles.27Aprinciplecomponentanalysisextractionmethodwasusedto
condensethematrixofcorrelationsandexplainthevariancesofthatmatrix.Additionally,the
correlationmatrixwasrotatedusingtheQuartimaxmethod.Aminimumloadingfactorof0.5
wasusedtodeterminewhichvariableshadstrongcorrelationswiththeunderlyingfactors.
Thepercentageofsurveyresponderswhoidentifywitheacheatingstylewasdetermined.
“Identify”wasdefinedasasumof>12forthequestionsforeachconstruct.Onascaleof3-15,
asumof12oraboveindicateda“high”or“close”identificationwithaneatingstyleconstruct.
Amongtheparticipantswhoidentifiedwithagivenstyle,crosstabulationwasthenusedto
determinethepercentageofsubjectswhoidentifiedwitheachoftheotherstyles.Forexample,
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 14
thismethodwasusedtodeterminethepercentageofemotionaleaterswhoalsoidentified
withthedietmentalityconstruct.Lastly,thesubjectiveresponsestoquestion20were
qualitativelycodedforthemesaboutmajorchallengestofoodselection,preparationand
eating.Trendsbetweenthethemesandidentificationwithanyeatingstyleconstructswere
computedusingcrosstabulation.
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 15
RESULTS
With185surveyssentouttoincomingparticipantsatGMFR,62surveyswerecompleted
andreturnedforaresponserateof33.5%.Studyparticipantswereallfemalesranginginage
from19-75yearswithanaverageageof50years.Descriptivestatisticsfromtheobjective
surveyresponsesmaybefoundinAppendixE,Table2andTable3.
StatisticalAnalysis
TheCronbach’salphatestrevealedthatallconstructsdemonstratedinternalconsistency,or
reliability,exceptforthe“chaotic”eatingstyle(acceptablevalues0.60-0.90;seeAppendixE,
Table3).Theskillfulconstruct’salphacoefficientwas0.625,chaotic0.456,dietmentality0.668,
emotional0.791,disconnected/habitual0.609anddazedandconfused0.678.Acceptable
valuesindicatedthatthesurveyquestionsforeachconstructwereconsistentinmeasuringthe
sameeatingstyleconstruct.ThesecondpartoftheCronbach’salphatestwastodeterminethe
alphacoefficientsoftheconstructs(indicativeofinternalconsistency)whenaquestionor
“item”wasdeletedfromthesurvey.Therewerethreenoteworthyresultsfromthisportionof
thetest.Fortheemotionalconstruct,thealphacoefficientincreasedto0.868(by0.77)if
question13wasdeleted,andforthedisconnected/habitualeatingstyleconstruct,thealpha
coefficientincreasedto0.691(by0.082)ifquestionfourteenwasdeleted.Similarly,when
questionseventeenwasdeleted,thedazedandconfusedalphacoefficientincreasedto0.758
(by0.08).
Theexploratoryfactoranalysisrevealedsixfactors,orstrongunderlyingrelationships
betweenvariables(eatingstyleconstructs),indicatedbyaninitialeigenvalueof1.0orgreater.
A0.5loadingfactorwasusedtodetermineastrongcorrelationbetweenavariableandan
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 16
underlyingfactor.Values0.5andabovearehighlightedinyelloworlightblueinTable4in
AppendixEontherotationcomponentmatrix.Threeoftheidentifiedfactorscorresponded
withthedietmentality(Q8-10),emotionalanddazedandconfused(Q17-19)eatingstyles.
Questions8-10correlatedstronglywiththefourthfactorandisrepresentativeofthediet
mentalityconstruct.Questions11-13correlatedstronglywithfirstfactorontherotation
componentmatrix,whichsuggeststhatthefirstunderlyingfactoristheemotionaleatingstyle.
Additionally,questions17-19correlatedstronglywiththethirdfactor,whichisrepresentative
ofthedazedandconfusedconstruct.Theremainingfactors(2,5and6)didnotcorrespond
withtheotherthreeeatingstyles:skillful,chaoticanddisconnected/habitual.Thesecondfactor
ontherotationcomponentmatrixwasstronglycorrelatedwithquestionstwo,six,fifteenand
sixteen.Factorfivecorrelatedwithtwoofthethreequestionsrelatedtotheskillfulconstruct(2
and3),buthasastrongcorrelationwithquestionsevenfromthechaoticstyle(andnot
questionfourfromtheskillfulconstruct).Thesixthfactorwasstronglycorrelatedwithfour
questions:#5,#7,#14and#17.Lastly,thereweretwoquestionsthatcorrelatedwithmorethan
onefactor,questionsseven(correlatedwithfactorfiveandsix)andseventeen(correlatedwith
factorthreeandsix).
TherelationshipsbetweeneatingstyleconstructsevaluatedbyPearson’scorrelationsare
foundinAppendixE,Table5.Theskillfuleatingstylehasanegativecorrelationwiththeother
fiveeatingstyles,chaotic,dietmentality,emotional,disconnected/habitualanddazedand
confused.Allconstructsweresignificantlycorrelatedwithskillfuleatingstyleata0.01level
exceptfordazedandconfused.Thissuggeststhatasskillfulbehaviorsincrease,theother
variablesorbehaviorsoftheothereatingstylesdecrease.Additionally,thedazedandconfused
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 17
eatingstylewasnotsignificantlycorrelatedfourofthefiveothereatingstyles.Dazedand
confusedwassignificantlycorrelatedwiththechaoticeatingstyleatthe0.01levelanddiet
mentalityatthe0.05level,butwasnotsignificantlycorrelatedwiththeskillful,emotionaland
disconnected/habitualstyles.Thus,ifsomeonebecomeslessconfusedbyfood,nutritionand
dietinformation,theywillnotnecessarilybecomemoreskillfuleaters.
ParticipantIdentificationwithEatingStyles
Thepercentofsubjectswhoidentifiedwitheacheatingstylewascomputedtodescribe
thesamplepopulation;31%ofsubjectsidentifiedwiththeemotionaleatingstyle,21%with
disconnected/habitual,18%withdietmentality,18%withchaotic,8%withdazedand
confused,and4%withskillful(seeAppendixE,Figure1).
Crosstabulationwasusedtodeterminetrendsamongidentificationwithmultipleeating
styles.Whenparticipantsidentifiedwithtwoeatingstyles,themostcommontrendwas
identificationwiththeemotionalanddisconnected/habitualstyles(90%ofemotionaleaters
alsoidentifiedwithdisconnected/habitual).Thesecondandthirdmostcommontrendswere
identificationwithdazedandconfusedandemotional(72.7%ofthedazedandconfusedeaters
alsoidentifiedwithemotional);anddazedandconfusedandchaoticstyles(72.7%ofthedazed
andconfusedeatersalsoidentifiedwithchaotic).
Additionaltrendswereobservedamongtheeatingstyles.Forchaotic,53.3%ofthe
disconnected/habitualeaters;52.0%ofdietmentalityeaters;and40.9%ofemotionaleaters
alsoidentifiedwithchaotic.Trendsamongthedietmentalityareasfollows:54.5%ofdazedand
confused,50%ofdisconnected/habitual,and45.5%oftheemotionaleatersalsoidentifiedwith
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 18
dietmentality.Lastly,54.5%ofthedazedandconfusedeatersalsoidentifiedwiththe
disconnectedeatingstyle.
MajorChallengestoFoodSelection,PreparationandEating
Whenanalyzingthesubjectiveresponsestoquestion20,challengestofoodselection,
preparationandoreatingdescribedbythesubjectswerecodedintotrends.Of49responsesto
question20,thetopthreechallengeswere“mealplanning”(n=15;30.6%),“time”(n=9;18.4%),
“cravings”(n=8;10.8%)andother(n=17;35%).Mealplanningchallengesincludeddifficulties
shoppingandpreparingfood,andfeelingoverwhelmedbytheprocessofmealplanning.For
example,onesubjectreportedthat“havingtopreparefoodformultiplepeople(husband,2
kids,me)makesitdifficulttoeatwhatIwantorwhenIwant”(mealplanningcategory).The
timechallengeswererelatedtobusyschedulesandreportsofnotenoughtimeinthedayto
planandprepareadequatefood.Experiencesmentionedinthecravingscategoryincluded
feelingsofintensecravings,lossofcontroloverfoodchoices,orhighlypalatableortrigger
foods.Forexample,“whenIaminthemiddleofacraving,allmyreasoningquicklyfliesoutthe
window.Allofit.Andthere'snoretrievingit.BeforeIknowit,I'meatingravenously.”
The“other”categoryincludedavarietyofchallengesdescribedlessfrequentlyby
subjects.Forexample,afewsubjectsdescribedchallengestofoodselection,preparationor
eatingthatincludedhealthoramedicalcondition,suchasmenopause.Othersdescribedwhat
maybecategorizedas“habitualeating”,where“Ihavetheknowledgeofwhatishealthytofix
butIhavegottenintoahabitoffixingthesametypesofmealsformyhusbandthatIthinkhe
expectsnow.IhavecookedhealthyforhiminthepastwhenI'vebeenonahealthiereating
kickandIdon'trememberhimcomplainingbutsomehowI'vegottenintoarutoflookingand
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 19
shoppingformore"comfort"mealsthattendtobemorefattening.Igetstuckonbadeating
habits.”Otherchallengesincludedpreoccupationwithfood,foodpreferences(ie.pickyeater),
cost,needforconveniencefood,accesstohealthyfoods,navigatingrestaurants,externalcues
toeating(presenceoffoodaroundthehouse)andlackofcookingskills.
Theresearchattemptedtoidentifytrendsbetweenchallengesandeatingstylesusing
crosstabulationmethods.Thedazedandconfusedeatingstylewashighlycorrelatedwiththe
“mealplanning”challenges(62.5%ofdazedandconfusedeatermentionedmealplanning).
Amongthechaoticeaters,55%mentionedmealplanningasachallenge.Thetimebarrierwas
highlycorrelatedwiththedisconnected/habitualanddazedandconfusedeatingstyles,
reportedby25%ofthethesetypesofeaters.Lastly,thesubjectswhoidentifiedwiththe
disconnected/habitualeatingstyleweremorelikelytoexperience“cravings”comparedtoany
othereatingstyle(20.8%ofdisconnected/habitualeatermentionedcravings).Similarly,19.4%
oftheemotionaleatersalsomentionedcravings.
GapsinEatingBehaviorsandAttitudes
Generalobservationsweregatheredfromthesubjectiveresponsestoidentifyany
eatingbehaviorsandattitudesnotcapturedbytheinitialsurvey.Multiplesubjectsmentioned
ordescribedanoteworthypreoccupationwithfood.Forexample,“Ijustwanttostopobsessing
aboutitandleadanormallife.”Otherbehaviorsreportedinthesubjectiveresponsesincluded
eatinginsecrecyandeatingoutofboredom.Lastly,multiplesubjectsmentionedeatingata
fasterpacerelatedtoconcernthattherewouldnotbeenoughforasecondserving,“justas
whenIwasakidwithseveralsiblings,whenIamsharingwithmyhusband,Iameatingfastto
getmyshare.”
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 20
DISCUSSION
Thisstudypresentedthedevelopmentofaninitialeatingstylesurveyderivedfrom
yearsofexperienceworkingwithwomenwhostrugglewitheatingandweightandpooled
itemsfromexistingquestionnairesregardingeatingbehaviorsandattitudes.Theresults
providedinsightintoGMFR’sparticipanteatingbehaviorsandattitudes,andabaseline
overviewofthereliability,validityandcomprehensivenessofthesurveytool.Bysurveyingthe
participantsusingquantitativeandqualitativequestions,trendsbetweeneatingstylesandgaps
inbehaviorsandattitudesnotinitiallycapturedbythesurveywereidentified.Tothe
researchers’knowledge,thesurveydevelopedinthisstudyisthefirstassessmenttoolcreated
thatincorporatesacomprehensivelistofeatingbehaviorsandattitudes,andaimstohelp
peopleidentifytheirindividualeatingstyle.Althoughtheidentifiedeatingstyleconstructsand
correspondingsurveyareatanearlystageofdevelopment,theresultsprovideinsightintonext
stepsforcreatingavalidatedeatingstylesassessmenttool.
Theobservednegativerelationshipbetweentheskillfuleatingstyleandtheremainingfive
styleswasexpected.GivenGMFR’saimtoguideindividualstowardsa“skillful”eatingstyle,itis
reassuringtoknowthatwhensomeonebecomesamoreskillfuleater,problematiceating
behaviorsinthedietmentality,emotional,chaotic,anddisconnected/habitualcategoriesare
likelyminimized.Thisisconsistentwithotherliteraturethatconfirmsintuitiveeatingis
negativelyrelatedtoeatingdisordersymptomatology,andmindfulnessisassociatedwith
decreasedemotionalandexternaleating.14,28Thedazedandconfusedstylewasnot
significantlycorrelatedwiththeskillful,dietmentality,emotionalordisconnected/habitual
styles.Thissuggeststhatspecifictargetedinterventionstrategiesmayberequiredtoimprove
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 21
understandingoffoodandnutritioninformation,orincreaseconfidenceinplanningand
preparingmeals;thesecharacteristicswillnotbeimprovedsimplybyincreasingskillful
behaviorsorattitudes.
Theresultsdemonstratedthatthedietmentality,emotionalanddazedandconfusedeating
stylesarethemostreliableandclearlydefinedconstructs.Thereisoverlapbetweenthese
factorsandotherassessmenttoolsthatincludemultipleconstructs.Similarly,the“AmI
Hungry?”assessmenttoolandtheDEBQincorporateconstructsorscalesthathaveparallels
withthedietmentality,suchastherestrictiveeatingcycleandtherestraintscale.6,20The
emotionaleatingstyleoverlapswiththeemotionaleatingscaleontheDEBQ.Theoverlapwith
the“AmIHungry?”cyclesandtheemotionalstyleislessstraightforwardgiventhecomplex
structureofeatingcycles.TheDEBQandthe“AmIHungry?”questionnairesdonotoverlap
withthedazedandconfusedeatingstyle,however,itissimilartothecontextualskills
describedbytheecSattermodel(“skillsandresourcestoprocureadequateamountsof
rewardingfood”).13Theresultsandthiscomparisonstronglysuggestthatthedietmentality,
emotionalanddazedandconfusedstylesaredistinctconstructsandshouldremainas
individualeatingstylesinthesurvey.
Asexpected,additionaleatingstylecharacteristicsemergedthatwerenotincludedonthe
initialsurvey.Eatingoutofboredomwasmentionedafrequentlybyparticipantsinrelationto
thequestion,“onceIstarteating,Ioftenfinditdifficulttostop”,whichwasdevelopedunder
theemotionaleatingconstruct.Emotionaleatingisacomplexconstructsinceitconsidersa
varietyofemotionsandexternalstimulithatinfluenceeating,suchasstress,anger,loneliness,
boredom,etc.Thereisquestionwhethereatinginresponsetoboredomcategorizesasa
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 22
componentofemotionalorexternaleating.23Koballetal.’sresearchsuggeststhatboredomis
animportantandindependentdimension,orsubscale,ofemotionaleating.Therefore,adding
“eatingoutofboredom”totheeatingstyleassessmenttoolmustbecarefullyconsidered.
Lastly,futureresearchinvestigatinghowtobestcategorizeboredomwouldfurtherinformthe
developmentofausefulassessmenttool.
Limitations
Thereareseverallimitationstothisstudythatmayhaveimpactedtheoutcomesandability
todrawconclusions.First,withonly62responses,repeatmeasuresofthestatisticalanalyses
areneeded.Morespecifically,researchersusuallyrecommendasamplesizeof100participants
ormoretomakeconclusionsfromanexploratoryfactoranalysis.27However,inthisstudythe
exploratoryfactoranalysiswasintentionallypreliminaryandprovidedapreviewofthe
underlyingfactors.Additionally,thesamplepopulationlimitsthegeneralizabilityoftheresults
becauseGMFRparticipantsarenotonlyauniquepopulationoffemaleswhostrugglewith
eatingandweight,theyarealsoanaffluentpopulation.Asapilotstudy,thepurposeofthe
researchwastoprovideinsightabouteatingstylesandwaystocategorizeeatingcharacteristics
andbehaviors.Identificationandvalidationofeatingstylesmaybeusefulforother
practitionersworkingwithclientswhoexperiencedisorderedeatingandissueswithbody
image.Therefore,incorporatingotherpopulationswhodonothavethesamefinancial
resourcesisnecessarybeforetheassessmenttoolcouldbevalidatedforusewithother
populationsandpractices.
Thesurveymethodwasadequateforprovidingpreliminaryinformationaboutobservable
trendsamongeatingstylesandinvestigatinggapsinthedefinitions.However,thesamplesize
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 23
andthesurveymethodlimitedtheabilitytodrawconcreteconclusionsabouttrendsamong
eatingstylesandwherenewcharacteristicsshouldbecategorized.Forexample,theresultsare
notclearenoughtomakeconclusionsaboutwhichbarriersshouldbelistedundereacheating
stylebecausethereweresofewresponsesineachcategory.Organizedfocusedgroupsarethe
nextlogicalstepstotrulyevaluatetheeatingbehaviorsandattitudesofparticipants.
Lastly,afewoftheresultswerebasedontheinferencethataparticipant“identified”with
aneatingstyle,indicatedbyasumof12orgreater,onascaleof3-15.Thisrequiresa
considerableamountofevidenceandassessmenttoensurethatthecutoffpointvalidly
capturesareasonablepointofidentifyingwiththateatingstyle,whichwasnotdoneto
determinethisindicator.Insuchearlystagesofresearch,thecutoffpointof12orgreateris
reasonablebecauseitreflectsanindividualwhoprimarilyansweredagreeorstronglyagreeto
thethreequestionsregardingthateatingstyle.However,furtherresearchinvestigatingan
appropriatescaleandcutoffpointfordeterminingidentificationwithaneatingstyleisneeded.
CONCLUSION
Overall,thepresentstudyaccomplishedthedevelopmentofaninitialsurvey,provided
insightonGMFR’sparticipants’eatingbehaviorsandattitudes,abaselineassessmentoftrends
amongeatingstyles,andanevaluationofthesurvey’sreliability.Mostimportantly,asskillful
eatingbehaviorsandattitudesincrease,thenotherproblematiceatingbehaviorsarelikelyto
decrease,exceptfordazedandconfusedcharacteristics.Althoughtheresultsidentified
additionalcharacteristicstoincludeontheassessmenttool,furtherresearchisneededto
gatherinsightonbehaviorsandattitudesthatalignwiththesixeatingstyles.Specifically,the
chaoticeatingstyleappearstobetheweakestconstructandtheemotional,
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 24
disconnected/habitualanddazedandconfusedstylesneedtoberevisitedtoimproveinternal
consistency.ThestudyprovidedGMFRwithastartingplaceonthemostcommonbarrierstheir
populationfaces.Lastly,theresultswillinformGMFR’snextstepstowardscreatingavalidated
eatingstylesassessmenttoolandstandardizedbehavioralstrategiestoguideindividualsto
becomemore“skillful”eaters.
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 25
Citations
1. EffectsofObesity|StanfordHealthCare.https://stanfordhealthcare.org/medical-
conditions/healthy-living/obesity.html.AccessedMarch1,2017.
2. CliffordD,OzierA,BundrosJ,MooreJ,KreiserA,MorrisMN.ImpactofNon-Diet
ApproachesonAttitudes,Behaviors,andHealthOutcomes:ASystematicReview.JNutr
EducBehav.2015;47(2):143-155.e1.doi:10.1016/j.jneb.2014.12.002.
3. PietiläinenKH,SaarniSE,KaprioJ,RissanenA.Doesdietingmakeyoufat?Atwinstudy.
IntJObes.2012;36(3):456-464.doi:10.1038/ijo.2011.160.
4. AndrésA,SaldañaC.Bodydissatisfactionanddietaryrestraintinfluencebingeeating
behavior.NutrRes.2014;34(11):944-950.doi:10.1016/j.nutres.2014.09.003.
5. Hulbert-WilliamsL,NichollsW,JoyJ,Hulbert-WilliamsN.InitialValidationoftheMindful
EatingScale.Mindfulness(NY).2014;5(6):719-729.doi:10.1007/s12671-013-0227-5.
6. StreinTvan,FrijtersJ,BergersG,DefaresP.TheDutchEatingBehaviorQuestionnaire
(DEBQ)forassessmentofrestrained,emotional,andexternaleatingbehavior.IntJEat
Disord.1986;5(2):295-315.
7. BroganA,HeveyD.Eatingstylesinthemorbidlyobese:restrainteating,butnot
emotionalandexternaleating,predictsdietarybehaviour.PsycholHealth.
2013;28(6):714-725.doi:10.1080/08870446.2012.760033.
8. WilliamsonDA,MartinCK,York-CroweE,AntonS,RedmanLM,HanH,RavussinE.
Measurementofdietaryrestraint:Validitytestsoffourquestionnaires.Appetite.
2007;48(2):183-192.doi:10.1016/j.appet.2006.08.066.
9. TriboleE,ReschE.IntuitiveEating.ARevolutionaryProgramThatWorks.3rded.New
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 26
York,NY:St.Martin’sGriffin;2012.
10. PapadopoulosS,BrennanL.Correlatesofweightstigmainadultswithoverweightand
obesity:Asystematicliteraturereview.Obesity.2015;23(9):1743-1760.
doi:10.1002/oby.21187.
11. NationalEatingDisordersAssociation.2016.https://www.nationaleatingdisorders.org/.
AccessedMarch1,2017.
12. HAES–HealthAtEverySize.https://haescommunity.com/.AccessedMarch1,2017.
13. SatterE.EatingCompetence:DefinitionandEvidencefortheSatterEatingCompetence
Model.JNutrEducBehav.2007;39(5SUPPL.).doi:10.1016/j.jneb.2007.01.006.
14. TylkaTL.Developmentandpsychometricevaluationofameasureofintuitiveeating.J
CounsPsychol.2006;53(2):226-240.doi:10.1037/0022-0167.53.2.226.
15. MonroeJT.MindfulEating:PrinciplesandPractice.AmJLifestyleMed.2015;9(3):217-
220.doi:10.1177/1559827615569682.
16. ThePrinciplesofMindfulEating.CentMindfulEat.
http://www.thecenterformindfuleating.org/Resources/Documents/principles_handout_
1_22.pdf.AccessedMarch1,2017.
17. FungTT,LongMW,HungP,CheungLWY.AnExpandedModelforMindfulEatingfor
HealthPromotionandSustainability:IssuesandChallengesforDieteticsPractice.JAcad
NutrDiet.2016;116(7):1081-1086.doi:10.1016/j.jand.2016.03.013.
18. ImagineALifeFreeofWeightLossWorries.GreenMtFoxRun.
https://www.fitwoman.com/?gclid=Cj0KEQiAzZHEBRD0ivi9_pDzgYMBEiQAtvxt-
IoS7Gn3EdKd0nv3R80bQnojV2Da5HfrI1JeS1qfEPoaArMq8P8HAQ.AccessedJanuary14,
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 27
2017.
19. FramsonC,KristalAR,SchenkJM,LittmanAJ,ZeliadtS,BenitezD.Developmentand
ValidationoftheMindfulEatingQuestionnaire.JAmDietAssoc.2009;109(8):1439-1444.
doi:10.1016/j.jada.2009.05.006.
20. MayM.EatingCycleAssessment-AmIHungry?http://amihungry.com/eating-cycle-
assessment/.AccessedAugust2,2016.
21. MayM.EatWhatYouLoveLoveWhatYouEatforBingeEating.First.Austin,Texas:
GreenleafBookGroupPress;2010.
22. ArnowB,KenardyJ,AgrasWS.TheEmotionalEatingScale:Thedevelopmentofa
measuretoassesscopingwithnegativeaffectbyeating.IntJEatDisord.1995;18(1):79-
90.doi:10.1002/1098-108X(199507)18:1<79::AID-EAT2260180109>3.0.CO;2-V.
23. KoballAM,MeersMR,Storfer-IsserA,DomoffSE,Musher-EizenmanDR.Eatingwhen
bored:RevisionoftheEmotionalEatingScalewithafocusonboredom.HealPsychol.
2012;31(4):521-524.doi:10.1037/a0025893.
24. UsingandInterpretingCronbach’sAlpha.UnivVirginiaLibrResDataServ+Sci.
http://data.library.virginia.edu/using-and-interpreting-cronbachs-alpha/.Accessed
March8,2017.
25. TavakolM,DennickR.MakingsenseofCronbach’salpha.IntJMedEduc.2011;2:53-55.
doi:10.5116/ijme.4dfb.8dfd.
26. WhatdoesCronbach’salphamean?InstDigitResEduc.
http://stats.idre.ucla.edu/spss/faq/what-does-cronbachs-alpha-mean/.AccessedMarch
8,2017.
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 28
27. KlineP.AnEasyGuidetoFactorAnalysis.NewYork:Routledge;1994.
28. OuwensMA,SchifferAA,VisserLI,RaeijmaekersNJC,NyklekI.Mindfulnessandeating
behaviourstylesinmorbidlyobesemalesandfemales.Appetite.2015;87:62-67.
doi:10.1016/j.appet.2014.11.030.
29. EatingStylesDefinitions.ReferenceDocument.2016:1-2.
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 29
AppendixA.
Table1.Overviewofeatingstyledefinitionsandassessmenttools.CreatorofDefinitions TypesorCategoriesofEating“Styles”GreenMountainatFoxRun(GMFR)29
EatingStyles Skillful,Chaotic,DietMentality,Emotional,DisconnectedorHabitual,Dazed&ConfusedMeasure:EatingStylesChecklist
MichelleMay,MD21 EatingCycles Instinctive,Overeating,Mindful,Restrictive,Binge,Eat-Repent-RepeatMeasure:“AmIHungry?”
EvelynTribole,MS,RD9
EatingPersonalities&Styles
IntuitiveEaterCarefulEater,ProfessionalDieter,UnconsciousEater(Chaotic,Refuse-Not,Waste-Not,EmotionalMeasure:None
EllynSatter,MS,RD,LCSW,BCD13
EatingCompetenceModel
EatingAttitudes,FoodAcceptanceSkills,InternalRegulationSkillsandContextualSkillsMeasure:EatingCompetenceQuestionnaire
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 30
AppendixB.GreenMountainatFoxRunEatingStylesDefinitions29SkillfulIngeneral,askillfuleaterlistenstoandtrustsherbody,totellherwhat,when,andhowmuchtoeat.Fooddecisionsarenotdictatedbyexternalcontrolsbutdrivenbyindividualneedsandpreferencesinagivenmoment.Sheusesfoodtofuelandnourishherbodyanddoesnotfeelguiltyaboutherchoices,evenifsheischoosingtoeataparticularfoodpurelyforpleasure.Sheenjoysherfood,savoringeachbite,andselectsfromawidevarietyoffoods.ChaoticDietlacksroutineandstructure.Eatingtimesmaybesporadicandinconsistent.Individualisoftenlessattunedtohungercues,notrespondinguntiltheyfeelveryhungry,whichleadstoconsumptionoflessnutritionallybalancedandmoreconvenience/highly-processedfoods.Italsooftenleadstoeatingbeyondthepointofcomfortablesatisfactionbeforerecognizingfeelingsoffullness.Apersonwhoeatschaoticallyoftenreportslackoftimeasabarriertohealth-supportiveeating.Foodselectionandpreparationtakesabackseattootherperceivedhigherpriorities.Asaresult,theindividualofteneatswhilemultitaskingand/orinarushedmanner.Dietmayalsolackvarietyduetotimeconstraintsandlowprioritizationoffoodselection.DietMentalityCharacterizedbyrigidrules-focusedeatingbehaviors.Individualsoften,butdonotalways,havealonghistoryoffollowingspecificdietprograms(WeightWatchers,Atkins,SouthBeach,Nutrisystem,etc.).Ifnotfollowingaspecificdiet,theyhaveinternalizedthedietmessagesthataresopervasiveinoursociety–viewingfoodasgoodandbad;viewingexerciseasawaytoearn,orpaypenancefor,certainfooddecisions;hyper-focusedoncontrollingportionsizesandcalorieintake;andverystrictaboutwhat,when,andhowmuchtheywilleat.Theyfeelshameandguiltovereatingcertainfoodsanddon’ttrustthemselvestomakedecisionsaboutwhat,when,andhowmuchtoeat,oftenrelyingonexternalcontrols.EmotionalFoodisbeingusedasamajorcopingmechanismand,insomecases,istheonlycopingmechanismformanagingchallengingemotions.Thedecisiontoeat,aswellasthetypeandamountoffood,isinfluencedbymoodandemotionsratherthanphysicalhungerandsatietycuesandnutrientneeds.Individualsmightfeeloutofcontrolandasiftheyareunabletostopeatingoncetheybegin.
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 31
Disconnected/HabitualThisisbestdescribedaseatingonautopilot.Individualsarelessawareduringthefoodselectionoreatingprocess,notrespondingtophysiologicalneedsbutratherenvironmentalcuesandtriggers.Theytendtochoosethesamefoodsoverandoveragain,becausetheyarenotattunedtotheirbodies’needsorevenpersonalpreferences.Theymayevenbegintoeatcertainfoodswithoutgivingmuchthoughttowhetherornottheyevenwanttobeeating.Theyareoftendistractedwhileeatingandarelessabletorecognizecuesoffullness;asaresulttheygenerallyrelyonexternalfactorstodictatetheendpointintheeatingoccasion(e.g.,whenthebagisempty,theplateisclean).Dazed&ConfusedIndividualsfeelcompletelyoverwhelmedbytheamountofnutritioninformationtheyreadandhearandarecompletelyconfusedabouthowtoeathealthfully.Additionally,theymayfeeluncomfortableinthekitchenhavinglimitedcookingskillsandareunawareofhowtocreategood-tasting,health-supportivemeals.Theymaynotknowwheretobeginwhenitcomestomealplanning,shopping,andpreparingmeals,andalthoughtheymayhaveadesiretopreparemoremealsthemselves,theyfeeloverwhelmedbytheprocess.Alternatively,theymayhavelittledesiretopreparetheirownmealsandareunsurehowtogoaboutprocuringhealth-supportivemealspreparedoutsideofthehome.
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 32
AppendixC.Eating Styles Survey
Please answer the following questions to the best of your ability. You may skip any question you prefer to not answer and still remain in the study.
1. Whatwouldyouliketochangeaboutyoureatingbehaviorsand/orthoughtsaboutfood? For questions 2 – 19, please select the response that best describes your behaviors or attitudes. When prompted, please provide more details about your experiences. 1=strongly disagree | 2=disagree | 3=neither agree nor disagree | 4=agree | 5=strongly agree
2. Iusuallytrustmybodytotellmewhat,whenandhowmuchtoeat.1 2 3 4 5
3. Usually,IsavormyfoodwhenIeat.1 2 3 4 5
4. Iamusuallyflexiblewithfoodchoices(forexample,myfoodchoicesarenotdictatedbyasetof
rules;Ifeelcomfortableeatingavarietyofdifferentfoodsbasedonthecircumstancesormyfoodpreferencesinthemoment,etc.).1 2 3 4 5
5. IamoftenveryhungrywhenIstarteatingandthenoftenuncomfortablyfullwhenIstop.
1 2 3 4 5
6. Myeatingroutineusuallylacksstructure,andissporadicandunpredictable.1 2 3 4 5
7. Iusuallyeatveryquickly.1 2 3 4 5Ifyourated3,4,or5,pleasedescribethefactorsthataffectthespeedatwhichyoueat.
8. Icategorizemostfoodsasbeing“good”or“bad”basedontheircaloriecontentand/ornutritionalvalue.1 2 3 4 5
9. Ioftenfeelguiltyorshamefulaftereatingcertainfoods.
1 2 3 4 5
10. Iusuallyfeeltheneedtocontrolmyfoodintakebycountingcalories,controllingportionsizes,and/orfollowingstrictdietplans.1 2 3 4 5
11. IofteneatwhenI’mfeelingunpleasantoruncomfortableemotions.
1 2 3 4 5
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 33
12. Foodoftenprovidesmecomfortanddistractionfromwhatisgoingoninmylife.
1 2 3 4 5
13. OnceIstarteating,Ioftenfinditdifficulttostop.1 2 3 4 5Ifyourated3,4,or5,pleasedescribewhenthisismostlikelytohappenandwhatmaytriggerthis.
14. Iofteneatwhiledistractedorengagedinotheractivities.1 2 3 4 5
15. IofteneatoutofhabitwithoutgivingmuchthoughttowhatorhowmuchIameating.1 2 3 4 5
16. IusuallyrelyonexternalcuestodecidewhenIwillstopeating(forexample,thepackageis
empty,myplateisclean,thefoodisgone,myTVshowisover,etc.).1 2 3 4 5
17. Ioftenfeeloverwhelmedbytheamountoffood,nutrition,anddietinformationIhear.
1 2 3 4 5
18. Ioftendon’tknowwhatfoodtoeatthatwillsupportmyhealth.1 2 3 4 5
19. Ioftenlackconfidenceinmyabilitytoplanandpreparemealsthatarenourishingandenjoyable.1 2 3 4 5
20. Pleasedescribeanyothermajorchallengeswhenitcomestofoodselection,preparation,
and/oreatingthatwerenotcapturedinthissurvey?
21. Isthereanythingelseyouwouldliketoshareaboutyoureatingbehaviorsorexperiences?
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 34
AppendixD.Research Information Sheet
Title of Study: Understanding Eating Behaviors and Attitudes to Define Eating Styles
Principal Investigator (PI): Christine Albertelli
Faculty Sponsor: Amy Nickerson, MS, RDN, CD
Funder: University of Vermont, Department of Nutrition and Food Sciences & Green Mountain at Fox Run
Introduction You are being invited to take part in this research study because you are an incoming participant at Green Mountain at Fox Run (GMFR). This study is being conducted by Christine Albertelli, a candidate in the Master of Science in Dietetics Program at the University of Vermont.
Purpose This study is being conducted to gather information about and discern patterns within individual eating behaviors and attitudes to better define different types of eating styles.
Study Procedures If you take part in the study, you will be asked to complete an anonymous, online survey. The survey will consist of several open-ended and Likert-scale based questions regarding your current eating behaviors and attitudes. You may skip any question you prefer not to answer and still remain in the study. The survey must be completed in one sitting and will take between 15-20 minutes to complete, depending on the detail of your responses.
Benefits As a participant in this research study, there may not be any direct benefit for you, other than gaining insight into your current eating behaviors and attitudes prior to your arrival at Green Mountain at Fox Run; however, information from this study may benefit other people now or in the future by helping to better inform programming and curriculum development at Green Mountain at Fox Run. This study may also contribute to the field at large by providing information that will help practitioners assess eating styles in clients and thus, better inform intervention approaches.
Risks We will take measures to fully protect the information we collect from you during this study. We will not collect any information that will identify you to protect your confidentiality and avoid any potential risk for an accidental breach of confidentiality.
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 35
Costs There will be no costs to you for participation in this research study.
Compensation You will not be paid for taking part in this study.
Confidentiality All information collected about you during the course of this study will be stored without any identifiers (anonymous). No one will be able to match you to your answers.
Survey responses will be kept in a secure electronic data file, on password protected computers, managed by Green Mountain at Fox Run. Anonymous survey responses will be compiled and provided to the PI via an electronic data file to be analyzed. Voluntary Participation/Withdrawal Taking part in this study is voluntary. You are free to not answer any questions or withdraw at any time. You may choose not to take part in this study, or if you decide to take part, you can change your mind later and withdraw from the study.
Questions If you have any questions about this study now or in the future, you may contact me Christine Albertelli at the following phone number (314) 610-5597 or Dana Notte, Nutrition Lead at Green Mountain at Fox Run, at (802) 228-8885 ext. 255. If you have questions or concerns about your rights as a research participant, you may contact the Director of the Research Protections Office at (802) 656-5040.
Participation Your participation is voluntary, and you may refuse to participate without penalty or discrimination at any time. Please print this information sheet for your records before continuing.
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 36
AppendixE.Results.
DescriptiveStatistics
Table2.Meanresponsestoobjectivequestions. Mean Std.Deviation NQ2 2.14 .973 59Q3 2.69 1.021 59Q4 3.56 1.303 59Q1 3.29 1.260 59Q2 3.41 1.315 59Q3 3.86 1.196 59Q8 3.78 1.233 59Q9 4.08 1.134 59Q10 2.92 1.343 59Q11 4.39 .983 59Q12 4.27 1.048 59Q13 3.78 1.340 59Q14 3.76 1.208 59Q15 3.61 1.017 59Q16 3.97 1.144 59Q17 3.42 1.499 59Q18 2.14 1.306 59Q19 2.80 1.448 59
Table3.FrequenciesofresponsesontheLikertscaleforeachquestion.
1 2 3 4 5Q2 Count 19 22 15 6 0
RowValidN% 30.6% 35.5% 24.2% 9.7% 0.0%Q3 Count 6 20 27 4 5
RowValidN% 9.7% 32.3% 43.5% 6.5% 8.1%Q4 Count 6 7 11 19 19
RowValidN% 9.7% 11.3% 17.7% 30.6% 30.6%Q5 Count 7 10 14 21 10
RowValidN% 11.3% 16.1% 22.6% 33.9% 16.1%Q6 Count 7 8 12 19 16
RowValidN% 11.3% 12.9% 19.4% 30.6% 25.8%Q7 Count 2 9 10 17 24
RowValidN% 3.2% 14.5% 16.1% 27.4% 38.7%Q8 Count 5 5 13 15 22
RowValidN% 8.3% 8.3% 21.7% 25.0% 36.7%Q9 Count 3 3 8 18 28
RowValidN% 5.0% 5.0% 13.3% 30.0% 46.7%Q10 Count 11 12 17 9 10
RowValidN% 18.6% 20.3% 28.8% 15.3% 16.9%Q11 Count 2 2 3 16 37
RowValidN% 3.3% 3.3% 5.0% 26.7% 61.7%Q12 Count 0 6 8 9 37
RowValidN% 0.0% 10.0% 13.3% 15.0% 61.7%Q13 Count 5 7 10 13 25
RowValidN% 8.3% 11.7% 16.7% 21.7% 41.7%Q14 Count 4 6 12 18 20
RowValidN% 6.7% 10.0% 20.0% 30.0% 33.3%Q15 Count 2 5 19 21 13
RowValidN% 3.3% 8.3% 31.7% 35.0% 21.7%Q16 Count 4 3 6 24 23
RowValidN% 6.7% 5.0% 10.0% 40.0% 38.3%Q17 Count 9 12 8 9 22
RowValidN% 15.0% 20.0% 13.3% 15.0% 36.7%Q18 Count 28 10 9 10 3
RowValidN% 46.7% 16.7% 15.0% 16.7% 5.0%Q19 Count 16 13 10 11 10
RowValidN% 26.7% 21.7% 16.7% 18.3% 16.7%
KEY:Percentages 30-40% >41-50% >51-60%
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 37
Table3.Internalconsistencyofeatingstyleconstructs.Construct Cronbach’salpha(a) Questions Cronbach’sAlphaifitemdeletedSkillful 0.625* Q2 0.595
Q3 0.386Q4 0.594
Chaotic 0.456 Q5 0.250
Q6 0.457Q7 0.362
DietMentality 0.668* Q8 0.519
Q9 0.626Q10 0.566
Emotional 0.791* Q11 0.620Q12 0.681Q13 0.868**
Disconnected/Habitual
0.609* Q14 0.691**
Q15 0.433
Q16 0.390Dazed&Confused
0.678* Q17 0.758**Q18 0.455Q19 0.522
*Acceptablevaluesindicativeofinternalconsistency/reliability=0.60-0.90.**Internalconsistencyincreaseswhenthequestionisdeletedfromtheconstruct. Table 4. Identification of underlying factors among eating behaviors and attitudes (factor analysis: rotated component matrix). Component (Factors)
1 2 3 4 5 6 Q2 -.126 -.629 -.244 -.168 .393 -.244 Q3 -.307 -.175 -.128 .101 .777 -.034 Q4 .051 .128 .081 -.441 .751 -.045 Q5 .130 .269 .435 .375 .217 .502 Q6 -.014 .772 .051 .000 .114 .104 Q7 .065 .223 .014 .169 -.545 .523 Q8 .316 .181 .084 .659 -.029 .032 Q9 .309 .238 .178 .565 -.191 -.231 Q10 -.003 .153 .140 .801 -.106 .318 Q11 .815 .254 .048 .294 -.082 .155 Q12 .910 -.004 .010 .036 -.015 .190 Q13 .616 .455 .016 .151 -.207 -.119 Q14 .185 .272 -.081 .069 -.220 .724 Q15 .219 .722 -.164 .202 -.032 .058 Q16 .413 .542 .166 .312 -.187 .067 Q17 .153 -.049 .552 .034 .070 .513 Q18 -.003 .045 .911 .095 .014 -.051 Q19 .014 .030 .811 .118 -.130 -.006 Extraction Method: Principal Component Analysis. Rotation Method: Quartimax. KEY: Factor loading: >0.5 (yellow or blue highlights; see written report for further explanation).
UnderstandingEatingBehaviorsandAttitudestoDefineEatingStyles|May2017 38
Table5.RelationshipsbetweeneatingstylesevaluatedbyPearson’sCorrelations. Chaotic DietMentality Emotional Disconnected/
HabitualDazed&Confused
Skillful -.369** -.459** -.366** -.437** -.188
Chaotic .485** .369** .588** .337**
DietMentality .455** .468** .305*
Emotional .594** .122Disconnected/Habitual
.140
*Correlationissignificantatthe0.05level(2-tailed).**Correlationissignificantatthe0.01level(2-tailed).
[CATEGORY NAME]
[PERCENTAGE]
(n=6) [CATEGORY
NAME] [PERCENTAG
E] (n=26)
[CATEGORYNAME]
[PERCENTAGE](n=25)
[CATEGORY NAME]
[PERCENTAGE] (n=44)
[CATEGORYNAME]
[PERCENTAGE](n=30)
[CATEGORYNAME]
[PERCENTAGE](n=11)
FIGURE1.PERCENTOFSUBJECTSWHO"IDENTIFIED"WITHEACHEATINGSTYLE