Manifesto for a European research network into Problematic Usage of the Internet

www.elsevier.com/locate/euroneuro

Manifesto for a European research network

into Problematic Usage of the Internet

NA Fineberg

, Z Demetrovics

, DJ Stein

, K Ioannidis

,

MN Potenza

, E Grünblatt

, M Brand

, J Billieux

,

L Carmi

, DL King

, JE Grant

, M Yücel

, B Dell’Osso

,

HJ Rumpf

, N Hall

, E Hollander

, A Goudriaan

,

J Menchon

, J Zohar

, J Burkauskas

, G Martinotti

, M Van

Ameringen

, O Corazza

, S Pallanti

, COST Action

Network, SR Chamberlain

aHertfordshirePartnershipUniversityNHSFoundationTrust,RosanneHouse,WelwynGardenCity, HertfordshireAL86HG,UK

bCenterforClinical&HealthResearchServices,SchoolofLifeandMedicalSciences,Universityof Hertfordshire,Hatfield,UK

cSchoolofClinicalMedicine,UniversityofCambridge,Cambridge,UK

dInstituteofPsychology,ELTEEötvösLorándUniversity,Budapest,Hungary

eDepartmentofPsychiatryandMentalHealthattheUniversityofCapeTownandSouthAfricanMRC UnitonRisk&ResilienceinMentalDisorders,CapeTown,SouthAfrica

fDepartmentofPsychiatry,UniversityofCambridge,Cambridge,UK

gCambridge&PeterboroughNHSFoundationTrust,Cambridge,UK

hConnecticutMentalHealthCenterandDepartmentsofPsychiatry,NeuroscienceandChildStudy Center,YaleSchoolofMedicine,NewHaven,USA

iConnecticutCouncilonProblemGambling,Wethersfield,CT,USA

jDepartmentofChildandAdolescentPsychiatryandPsychotherapy,PsychiatricUniversityHospital Zurich,UniversityofZurich,Zurich,Switzerland

kNeuroscienceCenterZurich,UniversityofZurichandETHZurich,Zurich,Switzerland

lZurichCenterforIntegrativeHumanPhysiology,UniversityofZurich,Zurich,Switzerland

mGeneralPsychology:CognitionandCenterforBehavioralAddictionResearch(CeBAR),Departmentof ComputerScienceandAppliedCognitiveScienceFacultyofEngineering,UniversityofDuisburg-Essen, Duisburg,Germany

nAddictiveandCompulsiveBehavioursLab,InstituteforHealthandBehaviour,Universityof Luxembourg,Esch-sur-Alzette,Luxembourg

oSchoolofPsychologicalSciences,TelAvivUniversity,TelAviv,Israel

pSchoolofPsychology,UniversityofAdelaide,Adelaide,Australia

∗Correspondingauthorat:HertfordshirePartnershipUniversityNHSFoundationTrust,RosanneHouse,WelwynGardenCity,Hertfordshire AL86HG,UK.

E-mailaddress:naomi.fi[email protected](N.Fineberg).

https://doi.org/10.1016/j.euroneuro.2018.08.004

0924-977X/© 2018TheAuthors.PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBY-NC-NDlicense.

(http://creativecommons.org/licenses/by-nc-nd/4.0/)

qDepartmentofPsychiatry,UniversityofChicago,Chicago,USA

rMonashInstituteofCognitiveandClinicalNeurosciences,SchoolofPsychologicalSciences,Monash University,Melbourne,Australia

sDepartmentofPathophysiologyandTransplantation,UniversityofMilan,FondazioneIRCCSCa’Granda andCRC“AldoRavelli” forneurotechnologyandexperimentalbraintherapeutics,Milan,Italy

tUniversityofLübeck,DepartmentofPsychiatryandPsychotherapy,CentreforIntegrativePsychiatry, Lübeck,Germany

uDepartmentofPsychiatryandCompulsive,ImpulsiveandAutismSpectrumProgram,AlbertEinstein CollegeofMedicineandMontefioreMedicalCenter,Bronx,NY,USA

vDepartmentofPsychiatry,AcademischMedischCentrum(AMC),UniversityofAmsterdam,Amsterdam, Netherlands

wDepartmentofPsychiatry,BellvitgeUniversity,Hospital-IDIBELL,UniversityofBarcelona,Cibersam, Barcelona,Spain

xSacklerMedicalSchool,TelAvivUniversity,andChaimShebaMedicalCenterTelHashomer,TelAviv, Israel

yLaboratoryofBehavioralMedicine,NeuroscienceInstitute,LithuanianUniversityofHealthSciences, Palanga,Lithuania

zDepartmentofNeuroscience,Imaging,ClinicalScience,UniversityG.d’AnnunzioofChieti-Pescara, Chieti,Italy

aaDepartmentofPsychiatryandBehaviouralNeurosciences,McMasterUniversity,Ontario,Canada

abAlbertEinsteinCollegeofMedicine,NewYork,USA

acUniversityofFlorence,Italy

adErwinL.HahnInstituteforMagneticResonanceImaging,Essen,Germany

aeAddictionDivision,DepartmentofMentalHealthandPsychiatry,UniversityHospitalsofGeneva, Switzerland

afCentreforExcessiveGambling,LausanneUniversityHospitals(CHUV),Lausanne,Switzerland

agDepartmentofPsychiatry,UniversityofMilan,FondazioneIRCCSCa’Granda,OspedaleMaggiore Policlinico,Milan,Italy

ahDepartmentofPsychiatryandBehavioralSciences,StanfordUniversity,Stanford,CA,USA

aiCRC“AldoRavelli” forNeurotechnologyandExperimentalBrainTherapeutics,UniversityofMilan, Milan,Italy

ajAmsterdamUMC,UniversityofAmsterdam,DepartmentofPsychiatry,AmsterdamInstitutefor AddictionResearch,Meibergdreef9,Amsterdam,Netherlands

akArkin,Amsterdam,TheNetherlands

Received 23May2018;receivedinrevisedform18July2018;accepted7August2018

KEYWORDS Problematicinternet use;

Compulsive;

Behaviouraladdiction;

Videogaming;

Pornography

Abstract

TheInternet isnowall-pervasiveacross muchofthe globe.While ithaspositive uses(e.g.

promptaccesstoinformation,rapidnewsdissemination),manyindividualsdevelopProblem- aticUseoftheInternet(PUI),anumbrellatermincorporatingarangeofrepetitiveimpairing behaviours.The Internet can act asaconduit for, andmay contribute to,functionallyim- pairingbehaviours includingexcessiveand compulsivevideo gaming,compulsive sexualbe- haviour,buying,gambling,streamingorsocialnetworksuse.ThereisgrowingpublicandNa- tionalhealthauthorityconcernaboutthehealthandsocietalcostsofPUIacrossthelifespan.

GamingDisorderisbeingconsideredforinclusionasamentaldisorderindiagnosticclassifica- tionsystems,andwaslistedintheICD-11versionreleasedforconsiderationbyMemberStates (http://www.who.int/classifications/icd/revision/timeline/en/).Moreresearchisneededinto disorderdefinitions,validationofclinicaltools,prevalence,clinicalparameters,brain-based biology,socio-health-economicimpact,andempiricallyvalidatedinterventionandpolicyap- proaches.Potentialculturaldifferencesinthemagnitudesandnaturesoftypesandpatterns ofPUIneedtobebetterunderstood,toinformoptimalhealthpolicyandservicedevelopment.

Tothisend,theEUunderHorizon2020haslaunchedanewfour-yearEuropeanCooperationin ScienceandTechnology(COST)ActionProgramme(CA16207),bringingtogetherscientistsand cliniciansfromacrossthefieldsofimpulsive,compulsive,andaddictivedisorders,toadvance networkedinterdisciplinaryresearchintoPUIacrossEuropeandbeyond,ultimatelyseekingto informregulatorypoliciesandclinicalpractice.Thispaperdescribesninecriticalandachiev-

ableresearchprioritiesidentifiedbytheNetwork,neededinordertoadvanceunderstanding ofPUI,withaviewtowardsidentifyingvulnerableindividualsforearlyintervention.Thenet- workshallenablecollaborativeresearchnetworks,sharedmultinationaldatabases,multicen- trestudiesandjointpublications.

© 2018TheAuthors.PublishedbyElsevierB.V.

ThisisanopenaccessarticleundertheCCBY-NC-NDlicense.

(http://creativecommons.org/licenses/by-nc-nd/4.0/)

1. Introduction

The public health and societal costs of Problematic use of the Internet (PUI) (World Health Organization, 2014) are increasingly recognised, constituting a growing con- cernacrossall agegroups, andan emergingchallengefor mental health research (Ioannidis et al., 2018). We use the umbrellaterm PUI toencompass allpotentially prob- lematicInternetrelatedbehaviours,includingthoserelat- ingtogaming,gambling,buying,pornographyviewing,so- cial networking, ‘cyber-bullying,’ ‘cyberchondria’, among others. PUI may have mental and physical health conse- quences.TheInternetisnowanintegralpartofmodernlife (Andersonetal.,2017),butasitsusehasgrown,ithaspro- videdanewenvironmentinwhichawiderangeofproblem- aticbehavioursmayemerge.

AspectrumofInternetusage,fromcontrolledand“adap- tive” to uncontrolled and “maladaptive”, is recognised (Billieuxetal.,2017).Disorderedonlinebehaviours,suchas excessivevideogaming,pornographyviewing,buying,gam- bling,orstreamingandsocialnetworksuse(Ioannidisetal., 2018)havebeenassociatedwithmarkedfunctionalimpair- ment includingloss of productivity (or reduced scholastic achievement),andmentalhealthsequelaeincludingmood and anxiety disorders (Derbyshire etal., 2013; Ho etal., 2014). Clinicalservicesfor PUI areavailable onlyin some geographicaljurisdictions,andthereforemanyaffectedin- dividualsdonothaveaccesstosupportandtreatment(Lam andLam, 2016;Winkleretal., 2013).National health au- thoritiesareexpressingconcern(Byrneetal.,2016;Parlia- ment (UK),2017) andthereisa growinginterest bysome governmentsinimplementingpoliciesaimedatcurbingPUI (Kiraly etal., 2017a). Psychiatryis beginning toacknowl- edgePUIandInternetGamingDisorderislistedasacandi- datemental disorderinneed offurtherstudyintheDiag- nostic and Statistical Manual (DSM-5) (2012), and Gaming Disorder is being actively considered (King et al., 2018) bythe InternationalClassificationof Diseases(ICD)for in- clusion in the section of disorders due to addictive be- havioursintheupcoming11thEdition.IndeedGamingDis- orderhasrecentlybeen includedin theICD-11versionre- leased to Member States (https://icd.who.int/browse11/

l-m/en#/http://id.who.int/icd/entity/1448597234).

Asnoted,PUIenvelopsawiderangeofactivitiesinclud- ingvideogaming,pornographyviewing(andothercompul- sivesexualbehaviours),buying,gambling, web-streaming, social mediauseandother behaviours.Someofthesebe- havioursmayfallintoanexistingmentaldisorderinpsychi- atricnomenclature (e.g.gamblingdisorder),whereasoth- ersarelikelytobeformallyrecognizedinfutureDSM/ICD

revisions, notably Internet Gaming Disorder (Kim et al., 2016b). Differenttypes of PUI often start in childhoodor adolescence(Volpeetal.,2015),butbroadagerangescan be affected (Ioannidis et al., 2018). Age and gender re- late importantly to PUI behaviours, with younger people typicallyhavingproblems withgamingand mediastream- ing, males withgaming, gambling and pornography view- ingandfemaleswithsocial mediaandbuying (Andreassen etal.,2016).TheInternetprovidesready,immediate,and seeminglylimitless pseudo-anonymous(meaningperceived asanonymous,butinrealitynotnecessarily)accesstothese activities. Furthermore, Internet sites, by collecting user data(‘bigdata’),arepotentiallycapableofbuildingartifi- cialintelligence(machinelearning)modelswiththeaimof reinforcingbehaviour, by generating algorithm-based per- sonalizedcontentandsitearchitecture.Whereasthesein- teractivequalities areviewed onthe one hand as having valuefor theuser,theysuggestthat theInternetmaynot simplyactasapassiveconduit,butitscompositebehaviours mayamplifyasenseorrewardandreliefoftension,thereby providingincreasedreinforcement(GrantandChamberlain, 2014).Being‘cutoff’fromtheInternetinsomeindividuals maygenerateasenseofdysphoriaakintowithdrawal.

Importantinroadshavebeenmadeintodefiningdiagnos- ticcriteriaanddevelopingratingscalesforsomeformsof PUI (i.e. Internet Gaming Disorder) (Kiraly et al., 2015), butnetworkedresearchisneededtounderstandbetterthe nature and extent of the broader range of PUI phenom- enaacrossjurisdictionsandcultures.Thebehaviouralphe- notypes of different types of PUI need to be translated into valid and reliable diagnostic criteria and assessment toolsdevised anddevelopedtosensitivelyandspecifically quantifythemagnitudeoftheproblem.Inturn,thiswould permit thesystematic investigationof aetiological factors (genetic,developmental,psychosocial,cultural)andbrain- based(biological and neuropsychological)mechanisms, as aplatformforthedevelopmentofpreventativeandthera- peuticinterventions.Acoordinatedmulti-nationalresearch approachwouldthus beinvaluabletoallowthe resultsto betestedacrosscountriesandcultures,toproducericher datasets witha view toensuringfindings can be general- ized;andalsoinordertostudylocalculture-specificissues ofrelevancetounderstandingPUI.

Inresponsetotheemergingpublichealthimportanceof PUI, the newly created European Problematic Use of the Internet (EU-PUI)Research Network wasformed, tobring togetherexpertresearchersandnationalandinternational scientificinitiativesunderoneEuropean-ledNetwork(Cost Action;www.cost.eu/COST_Actions/ca/CA16207)tolever- age existing funded research into a more coherent pro- gramme(Fig.1). The Network (currently 109 participants

Fig.1 Mapofthe27COSTMembercountriesand10InternationalObservercountries,with1ormoremembersparticipatingin theCOSTActionEU-PUINetwork.

from37countries)connectsamultidisciplinarygroupofge- ographically diverse expert researchers in neurosciences, genetics, and epidemiology,clinicians includingchild psy- chiatrists andpsychologists,expertsinbio-technologyand information-technology industry,health economists,policy makers,andserviceplanners,todevelopfruitfulcollabora- tions toadvancethe understandingandtreatment of PUI.

Additionally,throughaninteractiveDisseminationPlan(in- cluding website,social media,blogs) weare reachingout torelevantstakeholdersatinternationalandnationallevels withanemphasisonencouragingpeoplewiththelivedex- perienceof PUI tobecomeinvolved in theAction. To this end, our first International Congress (Barcelona, October 10th2018) includesalarge public-facingmeeting devoted tothelivedexperienceofPUI,withpeoplewiththedisor- dersandtheirfamiliesspeakingabouttheir experienceof PUIasanintegralpartofthesymposium.Raisingawareness of thelivedexperienceof peoplewithPUIandharnessing their energy via the network is likely to be of particular value fordirectingscienceandhealth policyandpractice.

Abetterunderstandingofthenatureofdifferentformsof PUI andtheirsocietalcostandburden wouldbeexpected tospearheadthedevelopmentofsystematicandadaptive treatmentandpreventionprogrammes.Table1summarises thenineachievableresearchprioritiesthatwebelievewill drivethefieldforwardoverthenexttenyears.

1.1. Reliableconsensus-drivenconceptualisations ofdifferentformsofPUI(phenomenologies, comorbidities,andbrain-basedmechanisms)

The term ‘addiction’ is derived from the antiquated le- gal Latin term addicere which roughly means “enslaved by”, indicating aperson who,havingdifficulties inpaying debts, wasassignedtoa creditor.Overthepast 100years

or so, this term has been mainly used in relation to the poorlycontrolleduseofpsychoactivesubstances,although recentlybehavioural addictions have been receivingmore attention.Addictionisnotaunitaryconstruct.Coreaspects ofaddiction,accordingtotheDSM-5,includeimpairedcon- trol (e.g. unsuccessful attempts to reduce intake), crav- ing(e.g.,strongpreoccupationsormotivationaldrivesthat leadtobehaviouralengagement),impairment(e.g.neglect of other areas of life that may lead tooccupational, re- lational and other problems), risky/harmful use (persist- ingintakedespite awarenessofdamagingpsychological or physiologicaleffectsorothernegativeconsequences),and physiological features (e.g., tolerance, withdrawal). Cer- tainpsychiatricsyndromescharacterisedbyharmfulrepeti- tivebehaviourssharesignificantphenomenological,biolog- ical, clinical, comorbidity, genetic, prevention,treatment andotherparallelswithsubstanceaddictions(Petry,2006;

Potenza, 2006), and have thus been argued to represent candidate ‘behavioural addictions’. Accordingly, the DSM- 5introducedthecategory‘SubstanceRelatedandAddictive Disorders’,andincludedgamblingdisorderasabehavioural addiction(Chamberlainetal.,2016;Potenzaetal.,2009;

Wareham and Potenza, 2010). Additionally, gambling dis- order shows differences from obsessive compulsive disor- der(OCD)insomeofthesemultipledomains,informingthe nosologicdecisiontoclassifyitasaSubstanceRelatedand AddictiveDisorderratherthananObsessiveCompulsiveRe- latedDisorder(Potenzaetal.,2009).

‘Internet addiction’ was considered by the DSM-5 Sub- stanceUseDisorderworkgroupandInternetGamingDisor- derwasincludedinSectionIIIofDSM-5asapotentialdisor- derworthyoffurtherstudy(2012).Morerecently,thepre- liminarybeta-draftofthe11threvisionoftheInternational Classificationof Diseases(ICD-11) included‘Gaming Disor- der’ in itssection on “DisordersDue toSubstance Use or AddictiveBehaviours” andcharacteriseditby(1)impaired

Table1 SummaryofkeyresearchprioritiestoadvancetheunderstandingofPUI.

1. Reliableconsensus-drivenconceptualisationofPUI(definingmainphenotypesandspecifiers,relatedcomorbidityand brain-basedmechanisms)

2. Age-andculture-appropriateassessmentinstrumentstoscreen,diagnoseandmeasuretheseverityofdifferentforms ofPUI

3. CharacterisetheimpactsofdifferentformsofPUIonhealthandqualityoflife 4. DefinetheclinicalcoursesofdifferentformsofPUI

5. Reduceobstaclestotimelyrecognitionandinterventions

6. ClarifythepossibleroleofgeneticsandpersonalityfeaturesindifferentformsofPUI 7. ConsidertheimpactofsocialfactorsinthedevelopmentofPUI

8. Generateandvalidateeffectiveinterventions,bothtopreventPUI,andtotreatitsvariousformsonceestablished 9. Identifybiomarkers,includingdigitalmarkers,toimproveearlydetectionandintervention

control,(2)increasingprioritygiventogamingresultingin precedence over other life interests and daily activities, and (3)continued use despitethe occurrence of negative consequences.Theevidenceforthisdecisionoftheexpert groupof theWHOcan befoundinSaundersetal.(2017).

NotallformsofPUImaymeetthethresholdforadiagnos- ablecondition;forexample,iftherewerenoobviousfunc- tional impairmentattributable to theInternet usage, the diagnosticthresholdforadisorderwouldnotbemet,though such behaviours may presage the development of subse- quent pathology (Billieux et al., 2017). Therefore, alter- nativestodiagnosticentitiesalsowarrantconsiderationso that publichealth maybe promoted. Consistentwiththis notion, definitions for ‘hazardousgaming’ and‘hazardous gambling’arecurrentlyproposedforICD-11,similartowhat currently exists for some addictive behaviours in ICD-10 (e.g., hazardous drinking of alcohol) (Reid et al., 1999).

Although – on the basis of the current evidence – DSM-5 and ICD-11 come tothe conclusion that PUI (in theguise of Gaming Disorder)should beconsidered asan addictive behaviour, alternativeframeworks meritscrutinyinfuture studies.

TheclinicalaspectsofsomeInternet-relatedbehaviours appearphenomenologicallymuchlikeaddiction(e.g.gam- bling or viewingpornography), anddemonstrate impaired control(unsuccessful attemptstoreduce orceasethebe- haviour),preoccupation(craving)associatedfunctionalim- pairment (neglect of other areas of life) and persistence despite damaging effects (Billieux et al., 2015; Ioannidis etal.,2016;Kardefelt-Winther,2017).However,itremains less clear whether, apart from gambling disorder, these other formsofPUImeetthephysiological criteriarelating toaddiction(tolerance,withdrawal).OtherPUIbehaviours maysharemoresimilaritieswithOCDandotherobsessive- compulsiverelateddisorders(e.g.,repeatedlycheckinge- mailsorsocialmedia,digitalhoarding)orsocialanxietydis- order(e.g.,excessiveuseof socialmediaasan avoidance offace-to-facesocialcontact)(Ioannidisetal.,2016).

Interestingly,obsessive-compulsivepersonalitytraitsare commoninexcessiveInternetusersandareassociatedwith problematicInternetuse(Chamberlainetal.,2017b),hint- ing that compulsive behaviours contribute to some forms ofPUI.Someformsofonlineshoppingorcybersex,onthe otherhand,maycloselyresembleICD-10orDSM-IVimpulse control or sexual disorders (Volpe et al., 2015). ‘Cyber-

chondria’isdefinedasexcessiveorrepeatedonlinesearch- ingforhealth-relatedinformationand,viewedthroughthe prismofhypochondriasis,mayrepresentamultidimensional Internet-related psychopathology associated with intoler- anceofuncertaintydrivingcompulsivereassuranceseeking (Volpeet al.,2015).Paradoxically, cyberchondriamay in- creaseratherthanresolveanxiety,perhapsduetothedis- coveryofnewtriggersforhealth-relatedworries,andmay reinforcefurthercompulsivechecking.Suchan“amplifica- tionmechanism” mayrelatetothecharacteristicoftheIn- ternetof not having been designed toprovide unambigu- ousreliable healthinformation,therebyincreasing uncer- taintyandworrying(Fergus,2013;FergusandSpada,2017).

Thus, while phenotypic similarities to behavioural addic- tions may pertain tosome forms of PUI, for others, sim- ilaritieswithsocial anxiety,impulse-control disorders and OCDmaybemoreprominent.Toaddressthisissuefurther, itmaybehelpfultostudyintermediatephenotypes,asdis- cussedlater.

Datasuggest thatdifferent formsof PUI areassociated withpsychiatriccomorbiditiesincludingaffectivedisorders (includingbipolar),anxietydisorders,substance-usedisor- ders(SUDs),sleepdisorders,attention-deficit/hyperactivity disorder (ADHD) and impulse-control disorders (Carter et al.,2016;Chamberlainetal.,2017a;Hoetal.,2014;Liuet al.,2011).However,comorbiditiesarealsoseenwithOCD, obsessive-compulsive personality traits (e.g. rigidity, per- fectionism,reluctance to delegate), andautism-spectrum disorder(ASD),implyingtheexistence ofrelationshipsbe- tween PUI and disorders of compulsivity (Chamberlain et al.,2017a;Finkenaueretal.,2012;Ioannidisetal.,2016).

InitialdataalsosuggestthatPUImaybeassociatedwithin- creasedratesofsuicidalityand/orself-harm(Chengetal., 2018;Durkeeetal.,2016;Fischeretal.,2012;Liuetal., 2017;Marchantetal.,2017).

Neuroscience research of other disorders of repetitive urge-driven behaviours, such asobsessive-compulsive and related disorders and SUDs, supports a complex neuro- psychological modelindicating impaired ‘top-down’corti- calbehaviouralcontrolunderpinnedbyabnormalactivation in specific components of brain circuitry encoding affec- tive,cognitiveandmotorcontrolofinstrumentalbehaviour (Fineberg etal., 2018). Implicated neural regions include corticalregions(suchasorbitofrontal, inferior-frontaland anteriorcingulateregions),andthestriatalventralregions

(involved in reward-driven behaviours) and dorsal regions (proposedtobemoreinvolvediningrainedhabitsbutalso relevant to craving in addictions) (Chamberlain and Men- zies,2009;ClarkandLimbrick-Oldfield,2013).Recentmod- elsexplainingthedevelopmentandmaintenanceofdiverse types ofPUIreflect theaforementionedmain mechanisms (e.g.impaired corticaltop-downcontrolandincreasedaf- fectivereactionslinkedtoventralstriatumactivity)(Brand etal.,2016b;DongandPotenza,2014).Althoughfewneu- roimaging investigationshave been conducted toexamine differenttypesofPUI,theproposeddiagnosticcriteriafor InternetGamingDisorderhavefacilitatedresearchintothis population.Inasystematicreviewoftheneuroimaginglit- erature,atleasttwostudiesfoundreducedgreymatterin the anteriorcingulate,insula,supplementary motor area, and inferior temporal gyrus, in individuals with Internet GamingDisordercomparedtonon-affectedcontrolindivid- uals(Weinsteinetal.,2017).Thesestructuralfindingswere confirmedinasubsequentmeta-analysis,whichalsoidenti- fiedabnormalactivationpatterns(fMRI)intheseandother regions (including in the caudate) (Yaoetal., 2017). In a recentmeta-analysisofcognitivefindingsinPUI,significant deficits wereidentifiedincase versus controlparticipants in inhibitory control, decision-making, and working mem- ory(preprint:https://osf.io/5kb7h),therebysuggestingthe likely involvement ofneural regions involved in top-down control.Aninitialcase-controlstudyreportedthatPUIwith- out comorbiditieswasassociatedwithimpaired qualityof life,butthatsignificantcognitivedeficitswerenotapparent onthedomainsexamined;whereasimpairedcognitionwas foundincomorbidcasesforinhibitorycontrolanddecision- making(Chamberlainetal.,2018).Thus,thecognitivepro- fileofPUIisinfluencedbycomorbidities.

Better understanding the similarities and differences amongthegenetic,neurobiologicalandsocio-demographic determinantsofthedifferentformsofPUIandthecharac- teristicsoftheirtreatment-responseswouldhelpclarifythe nosologicalrelationshipsbetweenarangeofmentaldisor- derscharacterisedprimarilybyimpulsive,compulsive,ha- bitualandaddictivebehaviours.Asavitalfirststep,stud- ies are needed to fully describe the range of behaviours anddisordersthatcomprisePUI,acrossgenders,agegroups and cultures, from the perspective of the contentof the repetitivebehaviours,maladaptiveconsequencesandother features,soastoreachconsensusonthediagnosticthresh- olds and criteria that may be used to define the disor- ders and other levels of behaviours warranting consider- ation frompersonal and public health perspectives. Once operational diagnostic criteria are established, standard- iseddatabasescouldbebuilttofacilitatereliableresearch.

However, understanding the determinants of disease and basinganewclassificationonthesedisordersmaynotnec- essarilyimproveclinicaloutcomesforaffectedindividuals;

field-testing of the proposeddiagnostic criteria across an integratednetworkofresearchcentresisalsorequired,as is currently beingconductedfor Gaming Disorder insome jurisdictions,tofacilitatesubsequentrigorousintervention trials, including early intervention in vulnerable individu- alswhohaveyettodevelopfullpathology.Takentogether, such improvements would beexpected tohave apositive impact upon the accuracy of prognostication and lead to betterhealthservicesinformatics,planninganddelivery.

2. Age- and culture-appropriate assessment

instruments to screen, diagnose and measure

the severity of the different forms of PUI

Validatedinstrumentsareneededinordertoidentifymen- taldisordersgermanetoPUI(screeninganddiagnosis),and to quantify their severity. Several such instruments have been developed in this field, as reviewed in detail else- where(Kiralyetal.,2015).Somehavearguedthatasingle instrumentcapturingallthecurrentformsofPUIwouldbe toobulkyandimpracticablebeusedclinically.However,the conversemayalsobetrue,i.e.aseriesofseparateinstru- mentsforeachformofPUImaynotgaintractionoutsidethe researchsetting,exceptforconditionsthatsoonmaybein- cludedinnosologicsystems(e.g.GamingDisorder).Several scaleshavebeendevelopedtomeasurebroadtypesofPUI (Kiralyetal.,2015).Forexample,Young’s InternetAddic- tionDiagnosticQuestionnaire(IADQ)andInternetAddiction Test(IAT)werederivedinpartfromoriginaldiagnosticcri- teriaforgamblingdisorder(Young,2009,1998).Morerecent developments,examiningabroaderrangeoffeatures(not focusingonlyon‘addiction’)includetheProblematicInter- netUseQuestionnaire(PIUQ)(Demetrovicsetal.,2016;Ki- raly et al., 2015) which has a replicable factor structure andgoodtest-retestproperties,alsobeingavailableinshort form.ThePIUQhasafactorstructurerelatingtoobsession, neglect, and loss of control. Instruments have also been developed to assess specific types of problem behaviour (Kiralyetal.,2015),suchastheProblematicOnlineGaming Questionnaire(POGQ),includingashortform,whichagain has a replicable factor structure and sound psychometric properties(Demetrovicsetal.,2012).ThereisalsotheIn- ternet Gaming Disorder 20 (IGD-20), which has good psy- chometricpropertiesand asix-factor structure(Ponteset al.,2014).Anothershortinstrument,theTen-ItemInternet GamingDisorderTest(IGDT-10),assessingIGDasproposedin theDSM-5showedgoodpsychometricpropertiesandmea- surement invariance acrosssevenlanguages (Kiralyetal., 2017b).

There nevertheless remains room for the development of astandardisedcontemporary measurefocussing onPUI severitythatissensitivetochange(e.g.treatmenteffects) and that could be broadly applied across all – or almost all clinicallyrelevant forms ofPUI. Examples fromdiffer- entfieldsincludethe differentversionsof theYaleBrown Obsessive CompulsiveScale that wassuccessfully adapted fora broadrangeofdisorders(Pallantietal.,2005).Con- siderationofPUIfromdifferentbutnon-mutuallyexclusive andcomplementaryperspectivesincludingthoserelatedto addiction,impulsivity,andcompulsivity,mayinthefuture mean that more of its features can be captured, under- stoodandaddressed. Otherkeyobjectivesfor researchin thisareaincludetheneedtomorerigorouslyassesstheim- pactofcomorbidities(e.g.OCD,ADHD,impulse-controldis- orders)onfunctionalimpairmentandtreatmentoutcomes, furtherworkonthelongitudinalprofilesofdifferenttypes ofPUI,andtheextenttowhichinstrumentscanbeshownto besensitivetotheeffectsoftreatment.Inparticular,there is a need for the validation of instruments besides facto- rialvalidityusingexternalmeasuresconfirmingconcurrent andpredictivevalidity.Suchexternalmeasuresneedtobe

based on gold standard assessments such as standardized clinicalinterviewsandmeasuresoffunctionalimpairment.

To date,onlyfew clinicalinstrumentsexistandvalidation needstobeconfirmed.Ingeneral,thereisanurgentneed toharmonizeassessmenttoolsinordertoassurecompara- bilityoffindings.Consideringthedigitalformofthedisor- der,digitaltools maybeusedasobjectiveandcontinuous monitorofthepathologicalbehaviour.Oncedevelopedand validated inclinical trials,theseInternet-basedtoolsmay servebothasmonitoringandasinterventionalutensils.

3. Characterise the impacts of different

forms of PUI on health and quality of life

IncreasingexposuretotheInternet,coupledwithpoorper- sonalcontroloveritsuse,suggeststhatthesocietalriskof PUI may begrowing. PUI is presently attracting consider- ablemediaattention.Parentalconcernisunderstandable, sincePUImaystartinchildhoodoradolescenceandpoten- tiallyalterdevelopmentaltrajectoriesorincreaseinsever- ityover time.Forexample,ina nationallyrepresentative sample of students,adolescents whospent more time on screenandengagedinelectroniccommunication(including Internet use) had lower psychological well-being and the converse alsoappeared tobethe case (greaterhappiness withloweruse)(Twengeetal.,2018).Thus,theeconomic burdenofPUImaygrowasdirecthealthcostsaccumulate.

SeveralexistingepidemiologicalstudiesofPUIhavepro- videdusefuldataontheprevalenceandfunctionalimpact of PUI (Dell’Osso et al., 2006; Kuss et al., 2014). Differ- encesinmethodologiescomplicatecomparisonsacrossdif- ferent studies, and have contributedto widevariation in estimates.Forexample,forInternetAddiction,prevalence rates of0.8–26.7% werereported(Kussetal.,2014).Sim- ilarly, for Internet gaming disorder, a systematic review reported prevalence estimates of 0.7–27.5% (Mihara and Higuchi, 2017). Suchvariabilityacross studies mayreflect differences in assessment instruments and thresholds ap- plied acrossstudies;discrepancies inInternetavailability;

social,culturalanddemographicdifferencesinInternetuse behavioursandattitudes; aswellasotherfactors. Onthe otherhand,thepublichealthburdenofPUImaybegreater thanitsprevalencemaysuggestifonealsoconsiderscomor- bidities.

Attheindividuallevel,PUIhasbeenassociatedwithre- lationshipdifficulties,academicandemploymentunderper- formance, financialproblems and other concerns(Billieux et al., 2017). Among college students, PUI was associ- ated with a poorer ‘grade point’ average, less frequent exercise,andhigherdistress(stressscores andlowmood) (Derbyshireetal.,2013).Amonghigh-schoolstudents,PUI occurred in ∼4% of the sample and was significantly as- sociated withsubstance-use disorders,depression,aggres- sionandother concerns(Liuetal.,2011).Thereis thusa clear need for further large-scale community-based stud- iestorobustlyexplorethedirectandindirectcostsofPUI acrossthelifespan.Specifically,weneedtoobtainreliable estimates of incidence, cumulative prevalence across the age-rangeand remissionrates alongwithassociatedclini- calanddemographicdeterminants thatmayalterthetra- jectoriesofspecificformsofPUI.Suchprimarydatawould

also be required to calculate the relative cost and out- come of ‘evidence-based’ interventions, as theydevelop, forindividualswithvariouslevelsofseverityandcomorbidi- ties.Withthesedatawecouldrefinesocio-health-economic models,developmeaningfulpublichealthpolicies,anddi- rectresources towardsthe greatest need alsotaking into consideration thedegree of suffering andlife impairment associatedwiththedifferentformsofPUI.

4. Defining the clinical courses of different

forms of PUI using longitudinal studies

Remarkablylittleprospectiveresearchhasbeenconducted on the courses of different forms of PUI and we remain relativelyignorant of keyfactorsaffecting long-term out- comes.Suchdataareofcrucialimportanceinunderstand- ingaetiology,planningtreatmentandimprovingprognosti- cation.Forexample,forsomeindividualsPUImayrepresent atemporaryphenomenonandspontaneouslyresolve(e.g., in some young people asbrain systems mature), whereas for others PUI may become chronic. Consequently, there isapressingneed forprospectivelongitudinalcommunity- basedanalysisofincidence,comorbidityandremission,us- ing age-defined cohorts including young children, adoles- centsandtheelderly.Studiessuchasthesewouldprovide freshdata onthecriticaltrajectories thatoccur over the lifespan and may informnew aetiological theories linking thesebehavioursanddisorders.Inatwo-yearlongitudinal studyinschoolstudents,depression,anxiety,socialphobia, andlowerschoolperformance,appearedtoariseasacon- sequenceof pathologicalgaming(Gentileetal.,2011).In alargecross-sectionaltwo-sitequestionnairestudy,PUIin olderagewasparticularlystronglyassociatedwithOCDand generalizedanxietydisorder;whereasinyoungindividuals, theassociations werestrongest for ADHDand social anxi- etydisorder(Ioannidisetal.,2018).Thus,furtherlongitu- dinalevaluationofdifferentformsofPUIandtheirclinical associationswouldbevaluableassuchinformationmayin- formtheneedfordifferentinterventionsandpreventative measures.Furthermore,longitudinalstudiesareneededto examineprotectivefactors andwitha viewtofacilitating resilience.

5. Reduce obstacles to timely recognition

and intervention

Lack of awareness, and the shame and secrecy that may surrounddifferent formsof PUI may constitute important obstaclestorecognition,diagnosisandultimatelyinterven- tion.Consequently,althoughtheincreasingnumberofpub- lishedreportsoftreatmentseekingcasesin2016–2017sug- geststhenumbersarerising(Billieuxetal.,2017),itislikely thatonlyaminorityofcasesaredetectedandofferedtreat- ment.Comorbidity(e.g.depression)mayincreasethelikeli- hoodthattheindividualwillpresenttoaclinician,butmay oftenfailtoimprovedetectionofPUIasPUImaycommonly beoverlookedbyhealthcareproviders,duetoalackofes- tablisheddiagnosticcriteria,trainingregardingrecognition andtreatment,awareness(education), reimbursement,or otherfactors.

The likely long-term, societal costs of early-onset PUI highlight theimportanceofdeliveringinterventionsatthe earliestsignsofillnessandatthegreatestlevelofefficacy andtolerability.Earlyinterventionmaybeparticularlyrel- evantfordisordersassociatedwithconsiderablepsychiatric comorbidity,sincetheearlydetectionandtreatmentofthe primary disorder couldalsoprevent theonset ofthe sec- ondarycomorbidity.Atthesametimeearlyinterventionin PUImightpreventthedevelopmentofsecondarycomorbid disorders.DatafromcompulsivedisorderssuchasOCDin- dicate that longer durationof illness is associated witha poorerclinicaloutcome(Dell’Ossoetal.,2013;Jakubovski etal.,2013).Apotentialforearlydetectionliesinidenti- fyingearlysignsofPUI.MaladaptiveuseoftheInternetin adolescencemaypredictthedevelopmentofPUIandhave severeconsequencesforthelongerterm.Educatingparents around appropriate use of the Internet, throughdevelop- ingpubliceducationprograms,toenablethemtoidentify young people at increasedrisk ofPUI is anapproachthat couldbeadoptedwithoutdelay.Digitalmonitoringwithap- propriate consentcouldbe anotherpromising toolasitis inexpensiveand couldbeusedforscreening andmonitor- ingofPUI.Developingandvalidatinganapplication(‘app’), whichremotely,objectivelymonitorsInternetuseandsends analerttothepatientand‘trust-buddy’(i.e.,familymem- ber or friend) might bean innovative approach to bein- cludedinfuturestudiesandsurveys.

6. Clarify the possible role of genetics and

personality features in different forms of PUI

PUI includes a complex spectrum of behaviours, and ge- netic andenvironmentalfactorsarethereforelikelytoin- dicate multifactorialaetiologies,aswithsubstance addic- tions(Kreeketal.,2005)orgambling(Gyollaietal.,2014).

Indeed,therehasbeensomeinitialresearchintotheheri- tabilityofPUI,inwhichbehaviouraltraitswerealsoconsid- ered.

HeritabilityestimatesofgeneralizedInternetaddictionin onestudywere58%forfemalesand66%formales,whilethe remaining variance wasexplained by unique environmen- tal influences(Lietal.,2014).InaTurkishstudyofyoung twin-pairs, geneticinfluencesuponPUIwere estimatedat 19−86% (Deryakulu and Ursavas, 2014). Elsewhere, in a study in theNetherlands, the heritabilityof PUI wasesti- mated at 48% (Vinket al.,2016).Similarly, a studyinves- tigatingthefrequencyofInternetuseinyoungadulttwins fromBrisbanefoundthat41%ofvarianceinthefrequencyof Internetusewasattributabletogeneticfactors(Longetal., 2016).Ontheotherhand,in784adultGermantwins,gen- eralizedPUIcouldbeexplainedbyshared andnon-shared environmentalfactorswhilegeneticinfluencedidnotplaya significantrole(Hahnetal.,2017).However,inoneperson- alitytrait(‘self-directedness’),ahighestimatedheritabil- ity wasfoundamounting to59%, while theremaining41%

wasaccountedtonon-sharedenvironmentalfactors(Hahn etal.,2017).Asummaryofallcurrentheritabilitystudiesis representedinFig.2.Despitelimitedresearch,maladaptive Internetusehasbeenshowntobemoderatelyheritablein somestudies,perhapsmorestronglysoinmales.However, duetovarieddefinitions,ageattimeofstudyandcultural

differences,thisissueneedstobefurtherinvestigatedina moresystematicmanner.

Preliminary data exist from molecular genetic studies showing possible specific genetic associations with PUI.

Thesestudieshavesuggestedthatvariantsinthedopamine receptor gene (DRD2) (Han et al., 2007), catecholamine- O-methyltransferase gene (COMT) (Han et al., 2007), serotonin transporter gene (5HTTLPR/ SLC6A4) (Lee et al.,2008), nicotineacetylcholinereceptorgene(CHRNA4) (Jeongetal.,2017;Montagetal.,2012)andneurotrophic tyrosine kinase type 3 receptor gene (NTRK3) (Ki et al., 2016a) are associated with PUI. Variants in these genes (DRD2, COMT, 5HTTLPR, CHRNA4) have also been associ- atedwithsubstanceuseandsubstanceaddiction(Breitling etal.,2009;Fengetal.,2004;Hanetal.,2011;LeFollet al.,2009;Suetal.,2015;TammimakiandMannisto,2010;

Yangetal.,2013b),possiblysuggesting thatPUImight be influencedby similargenes asinaddiction. Similarly, sev- eral communalities exist between these candidate genes and OCD as well as ADHD (Brem et al., 2014; Grunblatt et al., 2018), perhaps pointing to some overlapping risk genes andtraits for thesedisorders.However,in orderto have reliable results, replicationsand well powered phe- notypingstudy populationsarerequired. Moreover,future studiesshouldfocusonfindingrelevantgenesbyconduct- ing genome-wide association studies (GWAS) that enable hypothesis-free(data-driven)searchofriskgenes.Similarly, tomorefullyconsiderenvironmentalinfluencesongeneex- pression,epigeneticmeasurementswouldbenecessary.

Personalitytraits(e.g.,relatingtoimpulsivityandrigid- ity) have been associated with PUI (Chamberlain et al., 2017b; Ioannidis et al., 2016). For example, low self- directednessexceededhigh neuroticisminpredicting high Internet addiction test scores (Montag etal., 2010). This trait was highly heritable in an adolescent twin study, though PUI was not (Hahn et al., 2017). Indeed, such personalityfactors(e.g.neuroticism,extraversion,consci- entiousness, openness and impulsivity) were already sug- gested in GWAS studies to be highly heritable with over- lapping genetic componentsinfluencing various brain dis- orders (Anttila et al.,2016; Gray etal., 2018; Lo etal., 2017).Moreover,personalityfeaturesseemtoinfluencedi- versesubstanceusedisorders(Fengetal.,2004;Fewetal., 2014;Kreeketal.,2005;Munafoetal.,2007;Pintoetal., 2008; Su et al., 2015). Therefore, it would be important to undertake GWAS analysis of PUI, including personality featuresasan additionalfacet.In thisregardit wouldbe importanttoinclude notonly measuresof impulsivity but alsorecentlydevelopedtrans-diagnosticmeasuresofcom- pulsivity (Chamberlain andGrant, 2018), aswell as other intermediatephenotypes.

7. Consider the impact of social factors in

the development of PUI

While genetics andpersonality factors mayplay a role in PUI,social factors associatedwithPUI suchasfamilyand cultural factors and website architecture may be more amenabletomodificationandpublichealthpolicyinterven- tion.

Fig.2 PUIheritabilityestimateslinkedtopopulationsstudied.

Specific familial influences may contribute to the in- creasedordecreased likelihoodofanadolescent develop- ingproblemgaming.Inareview of14recentstudies, rel- evant family-related variables included: (a) parent status (e.g.,socioeconomicstatusandmentalhealth),(b)parent–

child relationship (e.g., warmth, conflict, andabuse), (c) parentalinfluenceongaming(e.g.,supervisionofgaming, modeling,andattitudestowardgaming),and(d)familyen- vironment(e.g.,householdcomposition)(Schneideretal., 2017).Mostofthestudieshavefocusedonparent-childre- lationships,reportingthatpoorerqualityrelationshipsmay be associatedwithincreased severityof problemgaming, withsomegender-specificinfluences.However,thereview authors found that much of this research was limited by relianceonadolescent self-reports,without corroboration fromparentsandotherfamilymembers,andsotheinter- generationaleffectsofproblemgamingrequirefurtherre- search.Notwithstanding,thereis evidencetosuggestthat preventionprogramsandinterventionsforadolescentsmay be more effective in some cases if they can address fa- milial influences on problem gaming, with the active co- participationofparents,ratherthanfocussingonindividual- basedmethodstotheexclusionofthefamily(Schneideret al.,2017).

Theroleofculturalfactorsisofparticularinterest,given thehigherestimatedprevalenceofPUIinsomegeographi- callocationsthanothers,yetmoststudieshavenotspecifi- callyexploredtransculturalissues.Inarecentonlinesurvey of2775youngadults(aged18–29years)recruitedacrossdif- ferentEuropeanUniversities coveringawidegeographical range,Northernand Southernregionsreportedthe heavi- est useof mobilephones, whereasperceived dependence waslessprevalentintheEasternregion(Lopez-Fernandez et al., 2017). The proportion of highly dependent mobile phone userswas highest in Belgium, UK, and France. Re-

gressionanalysisidentifiedseveralriskfactorsforincreased self-reported mobile-phone dependence, including higher frequencyofusage,specificInternetapplicationusage(so- cialnetworking, playingvideo games, shoppingand view- ingTVshowsthroughtheInternet,chattingandmessaging), andbeingfemale.Inasmallerquestionnairestudy(Yanget al.,2013a),inwhichJapanesestudentsweremorelikelyto demonstratePUI thantheir Chinesecounterparts,depres- sionandperceivedmaternalcarewereidentifiedasmoder- atingfactorsassociatedwiththiscross-nationaldifference.

EngagementinaddictivebehavioursontheInternetmay bestronglylinkedtoexposuretoreward-dependentcues.

For this reason, specific aspects of website architecture (form, content, reinforcing schedules) are likely to have an important effect on rates of Internet usage. Research in thisarea however is difficult toassess, as much is un- dertakenincommercialsettingsandisnotavailabletothe scientificcommunity.Specific psychoactivesubstancesand forms of gambling differ in their propensities for repeti- tion,duetovariableeffectsonthebrain’srewardcircuitry.

Forgambling, maximal engagementin gambling activities maybefacilitatedbyincorporatingpsychologicallyrelevant features such as variable ratio of reinforcement, losses- disguised-as-wins,andsenseofcontrol;andbyencouraging perceptionof‘nearmisses’(MurchandClark,2016).There- fore,gamblingresearchershavehighlightedtheimportance offocusingontheinteractionbetweenthe‘player’andthe

‘product’fromapublichealthperspective(KornandShaf- fer,1999;MurchandClark,2016).Wesuggestthatasimilar strategyisneededforunderstandingPUI.

Itislikelythatstructuralelementsofwebsitesandappli- cationsservetofavourexcessiveuseinvulnerableindivid- uals,butthattheseelementsmaydifferdependingonthe precisetypeof onlineactivity considered.Forvideo gam- ing,itappearsthatspecificstructuralcharacteristicsofthe

games (e.g.,in-game reward systemcapitalisingoninter- mittent reinforcement, daily updated international rank- ings) may potentiate addictive/excessive usage (Griffiths andNuyens,2017).Forproblematiconlinesexualbehaviour (e.g.cybersex),threestructuralelementshavebeenhigh- lighted asbeingimportant contributorsper theso-termed Triple A Model involving: accessibility, affordability, and anonymity(Cooper,1998;Cooperetal.,1999),thoughmore researchisneededonthistopic(Brandetal.,2016a;Wery and Billieux, 2017). Another similar proposed framework is the ACE Model (anonymity, convenience, and escape) (Young,2008).Forexcessivestreaming(watchingvideosex- cessively), important structural features may include the abilityofgivenprogrammestograbattentionbyactivating abiological‘orientatingresponse’,mediatedthroughtech- niquesincludingtheuseofattention-grabbingnoises,zoom- ing/panning,andpresentationofrewardingstimuli(e.g.,of a sexual or thrilling nature)(Flayelle et al., 2017, 2018).

Collectively, public research into the structural elements thatmaypromotePUIindifferentcontextsislacking.

8. Generate and validate effective

interventions, both to prevent PUI, and to

treat its various forms once established

Early identificationofindividualsat riskofdevelopingdif- ferenttypesofPUI,withaviewtoearlyinterventionstrate- gies, couldinthefuturereducetheburdenofdiseaseand help to prevent untoward functional consequences. This suggestion draws parallels from positive findings in other relatedareas ofmental health,suchasearlyintervention for substance addictions,to mitigatesuffering and public healthburden(Tanner-SmithandRisser,2016).Inasystem- aticreviewoftheavailableliterature,onlyfivedatapapers wereidentifiedthatdescribedandassessedpreventativein- terventionsforPUI(VondrackovaandGabrhelik,2016).The reviewauthorsfoundthattherewashighriskofbiasinthe coreempiricalpapersexamined,andreportedthatthemain targetgroups forinterventionwereyoungpeople,parents of youngpeople,andemployees.Inviewof theheteroge- nousnatureoftheinterventionsused,inconsistentmethod- ologies,andabove-notedhighriskofbias,rigorousearlyin- terventionstudiesareurgentlyneededbeforeanydefinitive conclusionsmaybedrawnastowhetherparticularearlyin- terventionsareclinicallyeffectiveandcosteffective.

Formost forms ofPUI, no‘evidencebased’ treatments exist and clinical research remains at early stages. Sev- eralmeta-analysesofavailabletreatmentshavebeencon- ducted,whichhavehighlightedmethodologicalflawsinthe extantliterature,suchas(butnotlimitedto)smallsamples, lack ofappropriatecontrolconditions(e.g.,lack ofactive controlconditionsasopposedtoawaitinglistcontrol),lim- itedinformationonadherencetotreatment, andinconsis- tent diagnostic definitions andmeasurements of symptom severity(Kingetal.,2017;Zajacetal.,2017).Todate,the strongest evidence appears tobe for cognitive behaviour therapy(CBT)butduetostudylimitationsdefinitivestate- mentsastoitsbenefits awaitfurthertesting (King etal., 2017).ACBTmodel specificallydesigned forPUIhasbeen piloted (Young, 2013). Appropriate controlled pharmaco-

logical treatment trials for PUI are absent. Future phar- macologicaltreatmenttrialsfor PUIcouldbeinformedby findingsfromexistingaddictionsandimpulsive-compulsive disorders. Forexample, glutamate modulatorsand opioid antagonistsshowpromiseincertainbehaviouraladdictions (Grantetal.,2016;Grantetal.,2008;Grantetal.,2014) and serotonin reuptake inhibitors are first-line treatment forOCD(Bandelowetal.,2008;Finebergetal.,2015;Koran andBlairSimpson,2013).Futurework couldalsoconsider evaluatingneuromodulatorytreatmentsthatarepromising inothercontexts,suchastranscranialmagneticstimulation (Carmi et al., 2018). In addition, thereis a need for de- velopingearlyinterventionsandpreventivemeasureswhich should include behavioural interventions as well asstruc- turalpreventionrelatedtogamedesign,protectionofyouth andothermeasuresthatcounteractexcessiveoraddictive use.

9. Identify biomarkers, to improve early

detection and intervention

TreatingvariousformsofPUIfromtheoutsetusingthemost appropriatestrategies(balancingpatientrisksandbenefits, along withresourcecosts) islikelytoproducebetterout- comesandrepresentbettervaluethantreatingmoresevere presentations later in their courses. Translational studies targeting pathogenicmechanisms at geneticandinterme- diatephenotype levels(i.e.,early stagevulnerabilityfac- tors), may clarify crucialneurobiological mechanisms and formthebasis for earlyillness-detection.Specifically, the discovery ofa reliable setof markersof psychopathology, neuropsychologyandneuroimagingmayprovideaninnova- tiveplatformofscreeningtoolswithasufficientlyhighlevel ofspecificityandsensitivitytodetectPUIearlyinitsonset.

TheInternetalsoopensnewavenuesfordifferenttypesof biomarkersforPUI,e.g.,digitalfingerprinting/phenotyp- ing.Withtheaidofappropriatealgorithms,carefulcontin- uousmonitoringofbehavioursofindividualsviadigitaltools withinformedconsentmayhelpidentifyaccuratebiomark- ersofPUI.

Extending ‘biomarkers’researchtoyoung individuals at theoreticalrisk ofdeveloping PUI (suchastheunaffected relatives of peoplewithPUI) or prodromal phasesof PUI, may permit early interventions that might alter the tra- jectory of the disorder toward a better long-term out- come.Forexample,ifvariousformsofPUIcouldbereliably predicted from recognised traits and symptomatologies, such as tendencies toward disordered compulsive (urge- driven) behaviours, this couldpotentially have important implicationsforearlyidentificationof“at risk” individuals and timely intervention beforeproblems take hold. How- ever, the promise of identifying “at-risk” individuals and effectively preventing the development of such disorders hasremainedchallenging,ascontinuedyouthengagement in addictive behaviours and disorders over time suggest (Potenza, 2013). Nonetheless,the Internet offers perhaps some unique opportunities for public policy interventions founduseful forother riskbehaviors(e.g.,taxationtore- duce tobaccouse) (Gearhardt et al.,2011).Given differ- ences across jurisdictions in Internet regulation, having a

research network focused on PUI will bevaluable in pro- motinghealthworldwide.

10. Conclusions

PUIrepresentsanumbrellatermforanemergingrangeof costly and burdensome behaviours that can occur across a large age range. Despite dedicated research and some breakthroughs in the scientific understanding of relevant neurobiologicalandpsychosocialfactors,theriskfactorsfor PUIneedfutureresearch.Thecurrentlevelofevidencehas tobeincreasedtopromotetheaccuratediagnosis,predic- tion ofprognosis, and developmentof effective interven- tional approaches. Action is needed todevelop strategies toidentifythosewhomightbeat increasedriskof PUIat the earlieststages, in ordertofacilitate prompt andreli- ablediagnosisanddevisepreventativeortherapeuticinter- ventionsatlocalandsocietallevels.InlinewithEUhealth priorities (Europe2020, European Pactfor MentalHealth, EPMH),theseactionswouldbeexpectedtooptimisehealth outcomes,improvequalityoflifeandproductivityandre- duceoveralldiseaseburden.TheEU-PUINetworkbringsto- gether individuals withthe knowledge and skills toset in motionaprogrammeofresearchaimedataddressingthese needs through networked research. Priorities include the followingareasfor differentformsofPUI:reachingacon- sensusonconceptualisation,describinganddefiningthedi- agnosticcriteria, developing andvalidating reliable tools, andquantifyingtheclinicalandhealtheconomicimpact.It willalsobeimportant,fordifferentformsofPUI,toiden- tifytheunderpinningbrain-basedmechanismsincludingthe contributionofcompulsive,impulsiveandhabitmechanisms aspotentialtargetsforinterventionandbiomarkers(includ- ing digital biomarkers) to advance early diagnosis, moni- toring,managementand publichealthpolicy. Throughes- tablishingsharedmultinationaldatabasesandcollaborative research networks, it is anticipated that progress will be made.Amultinationaltrans-culturalapproachwouldensure thegeneralisabilityoftheresultsacrossEuropeandbeyond.

Funding

This publication is based upon work from COST Action CA16207 “EuropeanNetwork for ProblematicUsage ofthe Internet”, supported by COST (European Cooperation in Scienceand Technology:www.cost.eu). The publicationis alsobased onworksupported by theEuropean Collegeof Neuropsychopharmacology Networks andthe International College of Obsessive Compulsive Spectrum Disorders. Dr Chamberlain’s involvement in this project was supported by a Clinical Fellowship from the Wellcome Trust (UK;

110049/Z/15/Z).ZsoltDemetrovics acknowledgesthesup- portoftheHungarianNationalResearch,Developmentand InnovationOffice(Grantnumbers:K111938,KKP126835).Dr.

Ioannidis involvement was supported by Health Education EastofEnglandHigherTrainingSpecialInterestsessions.Dr.

Potenza’sinvolvementwassupportedbytheNationalCen- terforResponsibleGamingthroughaCenterofExcellence grant, the Connecticut Council on Problem Gambling and theConnecticutDepartmentofMentalHealthandAddiction

Services.Dr.King’sinvolvementwassupportedbyaDiscov- ery Early Career Researcher Award (DECRA) DE170101198 fundedbytheAustralianResearchCouncil(ARC).Thefund- ing agencies did not have input into the content of the manuscriptandthe viewsdescribed inthe manuscriptre- flectthoseoftheauthorsandnotnecessarilythoseofthe funding agencies. Dr. Yucel was supported by a National Health and Medical Research Council of Australia Fellow- ship(#APP1117188)and the David WinstonTurnerEndow- mentFund.

Conflict of interest

The authorsreport noconflict of interest withrespectto thecontentofthismanuscript.

Disclosures

In the past year, NF has received research support from EuropeanCollegeofNeuropsychopharmacology(ECNP),Na- tionalInstituteforHealthResearch,WellcomeFoundation.

Dr Fineberg has received honoraria for lecturesat scien- tificmeetingsfrom Abbott,Wiley, Sun Pharmaceutical In- dustries, and British Association for Psychopharmacology (BAP).DrFineberghasreceivedfinancialsupporttoattend scientific meetings from the International Society for Af- fective Disorders, International Forum of Mood and Anxi- etyDisorders,ECNP,BAPandRoyalCollegeofPsychiatrists.

DrFineberghasreceivedfinancialroyaltiesforpublications fromOxfordUniversityPressandpaymentforeditorialdu- ties from Taylor and Francis. Dr Chamberlain receives an editorialfeefromElsevierinhiscapacityasassociateedi- toratNeuroscience andBiobehavioralReviews(NBBR). Dr Chamberlain consults for Cambridge Cognition and Shire.

Dr.Hollanderhasreceivedresearchgrantsinthepastyear fromDepartmentofDefense (DOD),OrphanProductsDivi- sionof Foodand DrugAdministration (FDA),Roche, Cure- mark,Takeda, Avanir, andNeurocrine, Dr.Pallanti hasre- ceivedresearchgrantinthepastandcurrentyearfromNa- tionalInstituteofMentalHealth(NIMH)foraR21onTheta BurstStimulationinGamblingDisorder.Dr.Dell’Ossohasre- ceivedresearchgrantsinthepastyearfromAngelini,Lund- beckandFBHealth.Dr.Potenzahasconsultedforandad- visedRivermendHealth,Opiant/LakelightTherapeuticsand JazzPharmaceuticals; received researchsupport(to Yale) fromtheMoheganSunCasinoandthe NationalCenterfor ResponsibleGaming;consultedfororadvisedlegalandgam- blingentitiesonissuesrelatedtoimpulsecontrolandaddic- tive behaviours;provided clinical care related toimpulse controlandaddictivebehaviours;performedgrantreviews;

editedjournals/journal sections; given academic lectures in grand rounds, CME events and other clinical/scientific venues;andgeneratedbooksor chaptersforpublishersof mentalhealth texts.In thepast 3years,Dr.Stein hasre- ceivedresearchgrants and/or consultancyhonorariafrom Biocodex, Lundbeck, Servier, and Sun. Joseph Zohar has received grant/research support from Lundbeck, Servier, BrainswayandPfizer,hasserved asaconsultantor onad- visoryboardsforServier,Pfizer,Abbott,Lilly,Actelion,As- traZenecaandRoche,andhasservedonspeakers’bureaus