The incidence, psychiatric co-morbidity and pharmacological treatment of severe mental disorders in children and adolescents

(1)

Original article

The incidence, psychiatric co-morbidity and pharmacological treatment of severe mental disorders in children and adolescents

Ragnar Nesvåg

^a,

*, Jørgen G. Bramness

^b

, Marte Handal

^c

, Ingeborg Hartz

^c,d

, Vidar Hjellvik

^c

, Svetlana Skurtveit

^c,e

aNydalenDPS,DepartmentofMentalHealthandAddiction,OsloUniversityHospital,P.O.Box4950,Nydalen,N-0424Oslo,Norway

bNorwegianNationalAdvisoryUnitonConcurrentSubstanceAbuseandMentalHealthDisorders,InnlandetHospitalTrust,P.O.Box104,N- 2381Brumunddal,Norway

cNorwegianInstituteofPublicHealth,P.O.Box4404Nydalen,N-0403Oslo,Norway

dDepartmentofResearch,InnlandetHospitalTrust,P.O.Box104,N-2381Brumunddal,Norway

eNorwegianCenterforAddictionResearch,UniversityofOslo,P.O.Box1171Blindern,N-0318Oslo,Norway

ARTICLE INFO

Articlehistory:

Received22September2017

Receivedinrevisedform15December2017 Accepted17December2017

Availableonline3February2018

Keywords:

Affectivedisorders Schizophreniaandpsychosis Psychopharmacology Epidemiology Adolescentpsychiatry

ABSTRACT

Background:Antipsychoticdruguseamongchildrenandadolescentsisincreasing,andthereisgrowing concernaboutoff-labeluseandadverseeffects.Thepresentstudyaimstoinvestigatetheincidence, psychiatric co-morbidityand pharmacological treatmentof severe mental disorder in Norwegian childrenandadolescents.

Methods:WeobtaineddataonmentaldisordersfromtheNorwegianPatientRegistryon0–18yearolds whoduring2009–2011werediagnosedforthefirsttimewithschizophrenia-likedisorder(International ClassificationofDiseases,10threvisioncodesF20-F29),bipolardisorder(F30-F31),orseveredepressive episodewithpsychoticsymptoms(F32.3orF33.3).Dataonfilledprescriptionsforpsychotropicdrugs wereobtainedfromtheNorwegianPrescriptionDatabase.

Results:Atotalof884childrenandadolescents(25.1per100000personyears)wereﬁrsttimediagnosed with schizophrenia-like disorder (12.6 per 100 000 person years), bipolar disorder (9.2 per 100 000personyears),orseveredepressiveepisodewithpsychoticsymptoms(3.3per100000personyears) during2009–2011.Themostcommonco-morbidmentaldisordersweredepressive(38.1%)andanxiety disorders (31.2%). Antipsychotic drugs were prescribed to 62.4% of the patients, 72.0% of the schizophrenia-likedisorderpatients,51.7%ofthebipolardisorderpatients,and55.4%ofthepatients withpsychoticdepression.Themostcommonlyprescribeddrugswerequetiapine(29.5%),aripiprazole (19.6%),olanzapine(17.3%),andrisperidone(16.6%).

Conclusions:Whenaseverementaldisorderwasdiagnosedinchildrenandadolescents,thepatientwas usuallyalsoprescribedantipsychoticmedication.Cliniciansmustbeawareofthehighprevalenceof depressiveandanxietydisordersamongearlypsychosispatients.

1.Introduction

Theonsetofschizophreniaorbipolardisorderusuallyoccursin earlyadultlife,butanumberofpatientshavetheirﬁrstepisodeof psychosis, mania or depression in adolescence or even in childhood[1,2]. Amongall adultswithschizophrenia, about1%

experiencedonsetofdiseasepriorto13yearsofageand12–33%

hadanonset priorto18 years[3]. Aregistry-basednationwide studyfromspecialisthealthcareinDenmarkdemonstratedthat

0.3%ofthepopulationwerediagnosedwithaschizophrenia-like disorderby20yearsofage[4].Schizophreniawithonsetpriorto18 yearsofagehasapoorerpsychosocialoutcomethanadult-onset schizophrenia [5], with a particularly poor outcome for onset before13yearsofage[6].

Bipolardisorderhastraditionallybeenregardedasanadult- onsetdisorder,butrecentnationwidestudiesfromUKandUShave demonstratedthatbipolardisorderisoftendiagnosedinpaediatric populationsaswell[7,8].Inamulti-centerstudyfromUS,Germany and the Netherlands, the proportion of adult bipolar disorder patientswithonsetpriortoage19was61%intheUSand30%in Europe,indicating substantial geographical differencein prevalence of paediatric bipolar disorder [9]. In the US National

*Correspondingauthor.Presentaddress:TheNorwegianMedicalAssociation,P.

O.Box1152,Sentrum,N-0107Oslo,Norway.

E-mailaddress:ragnar.nesvag@legeforeningen.no(R.Nesvåg).

http://dx.doi.org/10.1016/j.eurpsy.2017.12.009

ContentslistsavailableatScienceDirect

European Psychiatry

j o u r n a l h o m e p a g e : h t tp : / / w w w . e u r o p s y - j o u r n al . c o m

(2)

ComorbiditySurveyofnationallyrepresentativeadolescents,2.9%

ofthe13–18yearoldsmetdiagnosticcriteriaforbipolardisorder typeIorII[8].ThisishigherthanEuropeanﬁgures,butarecent studyfromGermanydemonstratedan18%increaseintherateof discharge diagnosis for bipolar disorder among 0–19year olds between2000and 2013[10]. In theabovementionedregistry- basedstudyfromspecialisthealth carein Denmark,0.1%of the populationwerediagnosed withbipolardisorder bytheageof 20years[4].

The range of prevalence estimates for paediatric psychosis illustratestheneedforinvestigationsofdiagnosticpracticeina varietyofsettings.Theincidenceratesfordiagnosedschizophre- nia,bipolarordepressivedisorderwithpsychoticfeaturesamong childrenandadolescentsinNorwayarepresentlyunknown.

Ifdepressiveoranxietydisorders arediagnosedprior tothe ﬁrst-episode of psychosis, they may serve as early signs and prognosticmarkersoftheemergingseverementaldisorder[11].

For about 20% of patients with adult-onset psychosis clinical depressive symptoms are present even 10 years after onset of psychosis[12].Depressive oranxietydisorders mayalsobeco- morbid disorders, present in about half of adult ﬁrst-episode psychosis patients[13,14]. Good prevalence data onco-morbid mental disorders in adolescent-onset psychosis are, however, lacking.

Earlyadequate treatmentofﬁrst-episodepsychosisinyoung adultsmayimprovelong-termoutcomeandmostofteninvolves theuseofantipsychoticdrugs[15].Theuseofantipsychoticdrugs in childrenand adolescents is,however, controversial due toa scarcityofclinicalevidencefromrandomizedcontrolledstudies [16]. Also, a recent meta-analysis has demonstrated that antipsychotics had only modest symptomatic effect in younger patients[17].ADanishrandomizedcontrolledtrialofquetiapine vs aripiprazole among 12–17year old ﬁrst-episode psychosis patientsdemonstrated that both drugs had moderateeffect on alleviating positive symptoms, but most patients experienced unwantedadverseeffectslikesedation,weightgainortremor[18].

Thereisasubstantialdiscontinuationrateduetoweightgainand otheradverse sideeffects [19] andthere isgenerallya growing concernaboutlong-termconsequencesofantipsychoticdrugson somaticandmentalhealth[20,21].Inarecentstudyweshowed thatthemajorityofchildrenandadolescentsusingantipsychotic

drugswerediagnosedwithnon-psychoticdisorderslikehyperkinetic, anxiety ordepressive disorder[22]. The extent towhich younger patientsdiagnosed with psychotic disorders is treated withantipsychoticdrugsinNorwayisunknown.

Data fromnational healthregistrieshavetheadvantagethat theycoverallindividualswhohavebeenincontactwiththehealth caresystem,butthereisconcernregardingvalidityofdiagnostic information and completenessof data [23]. For severe mental disorders like schizophrenia and bipolar disorder, the national clinicalguidelinesrecommendassessmentandstartoftreatment in specialized mental health care [24]. Thus, the prevalence estimatesofseverementaldisordersbasedonnationalhealthcare registrydatawillberepresentativeofthegeneralpopulation.

1.1.Aimsofthestudy

Based on linked individual-level data from national health registriesweaimedtoestimate1)incidenceratesofseveremental disorders, here deﬁned as schizophrenia-like disorder, bipolar disorder,orseveredepressiveepisodewithpsychoticsymptoms, among 0–18year old children and adolescents diagnosed in specialisthealthcareinNorwayduring2009–2011,2)investigate theprevalenceofco-morbidmentaldisorders,and3)exploreuse of antipsychotics and other psychotropic drugs in this patient group.

2.Materialandmethods

ThepresentstudyisbasedonlinkeddatafromtheNorwegian PatientRegistry(NPR)andtheNorwegianPrescriptionDatabase (NorPD). Individual-level registry data were linked using the unique11-digitpersonalidentitynumberassignedtoallindividu- alslivinginNorway.

2.1.Datasources

2.1.1.TheNorwegianPatientRegistry(NPR)

TheNPRisanadministrativedatabaseofrecordsreportedby thespecialisthealthcare,i.e.allhospitalsandoutpatientclinics owned or ﬁnanced by the government, including most private practitioners in child and adolescent psychiatry.Thus, theNPR

Table1

Classiﬁcationofpsychotropicdrugs.

Group Type ATCcode

Antipsychoticdrugs Any N05AexceptN05AN01(lithium)

Firstgenerationantipsychotics Perphenazine N05AB03

Haloperidol N05AD01

Zuclopentixole N05AF05

Secondgenerationantipsychotics Ziprasidone N05AE04

Olanzapine N05AH03

Quetiapine N05AH04

Risperidone N05AX08

Aripiprazole N05AX12

Paliperidone N05AX13

Clozapine N05AH02

Auxiliaryantipsychoticdrugs Chlorpromazine N05AA01

Levomepromazine N05AA02

Lithium n.a. N05AN01

Antidepressant Any N06A

Antiepilepticdrugsusedfortreatmentofbipolardisorder Carbamazepine N03AF01

Valproicacid N03AG01

Lamotrigine N03AX09

Topiramat N03AX11

Anxiolyticdrugs Any N05B

Psychostimulant Any N06BA

Abbreviations:ATC:AnatomicTherapeuticClassiﬁcation;n.a.:notapplicable.

(3)

includes information on patientsthat have been referred by a generalpractitioner (GP)becauseof a needof specialist health care.AllNorwegiancitizensareenrolled inthepatientlistof a dedicatedGPwhorepresentsthelowestlevelofpublichealthcare.

Ifapatientrequirestreatmentatahigherlevel,theGPmayreferto aspecialisthealthcarefacility,i.e.ahospital,outpatientclinicor private practitioner. All referrals and registered contacts with specialist health care are included in the NPR. In Norway, government-funded mental health clinics for children and adolescents are available throughout the country, serving the entirepopulation.Mentalhealthcareforchildrenandadolescents isfree of charge. TheNPR contains nationwideindividual-level specialisthealthcaredatafrom2008andonwards,anddiagnoses are recorded according to the International Classiﬁcation of Diseases,10threvision(ICD-10)[25].

For the present study we retrieved data reported by the specialisthealthcareduring2008–12forallindividualswhowere 0–18 years of ageatﬁrst registrationin the NPRwithan ICD- 10 diagnosis of either schizophrenia-like disorder (F20-F29), bipolar disorder (F30-F31), or severe depressive episode with psychoticsymptoms(F32.3,F33.3)during2009–2011.

To investigate prevalence of co-morbid mental disorders,all registered contacts with the following ICD-10 diagnoses were noted: Substance use disorders (SUD, F10-F19), depressive disorder(F32-F34),anxietydisorder(F40-F48),mentalretardation (F70-F79),autism-spectrumdisorder(F84),hyperkineticdisorder (F90),andconductdisorder(F91).

2.1.2.TheNorwegianPrescriptionDatabase(NorPD)

DataonpsychotropicdrugusewereobtainedfromtheNorPD, whichcoversalldispensedprescriptionsatpharmaciesinNorway.

SinceJanuary2004allNorwegianpharmacieshavebeenobligedto senddataelectronicallytotheNorwegianInstituteofPublicHealth onall prescribeddrugs (irrespectiveof reimbursementor not) dispensedtoindividualsinambulatorycare.Drugsadministeredto patientswhileinhospitalarenotreportedtotheNorPD.Thedrugs intheNorPDareclassiﬁedaccordingtotheAnatomicalTherapeu- ticChemical(ATC)classiﬁcationsystem[26].

Forthepresentstudyweincludedthepatients’uniqueidentity number(encrypted),sex,age,dateofdispensingandgenericdrug information (ATC code). Information about all prescriptions of psychotropicdrugsin theperiod2008–2012 wasobtained. The ATC codes included in each psychotropic drug category are presentedinTable1.

2.2.Analyticalapproach

Incidenceratesofeachofthespeciﬁedseverementaldisorders (schizophrenia-likedisorder,bipolardisorder,andseveredepres- sive episode withpsychotic symptoms) werecalculated asthe numberof boys and girls per100 000 personyears who were

registered with the disorder between January 1, 2009 and December31,2011,followingaminimumwashoutperiodofone year,i.e.noregistrationwithanyofthethreetypesofdisorders fromthestartof2008.Numberofpersonyears(3521950)was calculatedasthesumofboysandgirlsaged0–18yearslivingin NorwayasperJanuary1,2009(1168235),2010(1174347)and 2011(1179368)accordingtodataobtainedfromStatisticsNorway.

A washout period of one year may have resulted in some prevalentcasesbeingmisidentiﬁedasincidentones(cf.Section4.1 Strengthsandlimitations).To enablecomparisonwithprevious incidencestudies,one-yearincidencerateswerecalculatedforthe year2011,allowingforawashoutperiodofatleastthreeyears.The incidencerateswerecalculatedasnumberofboysandgirlsper100 000 person years who werediagnosed withschizophrenia-like disorder, bipolar disorder, or severe depressive episode with psychotic symptomsin2011, butnot in2008–2010.Numberof personyearswascalculatedasthesumofboysandgirlsaged0– 18yearslivinginNorwayasperJanuary1,2011(605201boysand 574167girls)accordingtodataobtainedfromStatisticsNorway.

Theprevalenceofco-morbidmentaldisorderswascalculated as the proportion of patients who were registered with SUD, depressiveillness,anxietydisorders,mentalretardation,autism- spectrum disorder, hyperkinetic disorder or conduct disorder betweenoneyearpriortoandoneyearaftertheincidentsevere mentaldisorderdiagnosisintheNPR.Insearchofearlydiagnostic markers of psychotic disorder, theproportion of patientsdiag- nosedwithSUD,depressiveillness,anxietydisorderorhyperki- neticdisorderwithinoneyearpriortotheincidentseveremental disorderdiagnosiswascalculatedseparately.

Theproportionofpatientsusingantipsychoticmedicationand othertypesofpsychotropicdrugsbetweenoneyearpriortoand oneyearaftertheincidentseverementaldisorderdiagnosiswas determinedfromregistrationsofdispensedprescriptionsin the NorPD. Having at least one prescription dispensed during this periodwasdefinedasuseofthespecificdrug.Theproportionof patientsfillingmorethanoneantipsychoticmedicationprescrip- tionwascalculatedasaproxyforregularuse.

AllanalyseswereperformedusingSPSS22.0forWindows.

2.3.Ethicalconsiderations

ThestudyhasbeenapprovedbytheRegionalCommitteefor Medical Research Ethics (2010/131–REK Sør-Øst) and by the NorwegianDataProtectionAuthority.

3.Results

3.1.Incidenceofseverementaldisordersinchildrenandadolescents During 2009–2011 a total of 884 children and adolescents 18yearsoryounger(incidencerate25.1per100000personyears)

Table2

Incidenceofseverementaldisordersinchildrenandadolescents.DatafromtheNorwegianPatientRegistryonall0–18yearoldsregisteredwithaﬁrstdiagnosisof schizophrenia-likedisorder,bipolardisorder,ordepressiveepisodewithpsychoticsymptomsduring2009–2011.

SCZ(n=443) BIP(n=323) DEPw/psych(n=118) Total(n=884)

Incidence(per100000personyears) 12.6 9.2 3.3 25.1

Boys(n,%) 239 54.0 119 36.8 31 26.3 389 44.0

Ageat1stdiagnosis(mean,SD) 16.1 2.5 16.2 2.4 16.1 1.7 16.1 2.3

Age0–11at1stdiagnosis(n,%) 18 4.1 13 4.0 <5 n.a. 33 3.7

Age12–15at1stdiagnosis(n,%) 106 23.9 73 22.6 32 27.1 211 23.9

Age16–18at1stdiagnosis(n,%) 319 72.0 237 73.4 84 71.2 640 72.4

Abbreviations:SCZ:schizophrenia-likedisorder(InternationalClassiﬁcationofDiseases,10threvisioncodeF20-F29),BIP:bipolardisorder(F30-F31),DEPw/psych:severe depressiveepisodewithpsychoticsymptoms(F32.3orF33.3),n.a.:notapplicable.

(4)

wereregisteredwithafirst-episodeofseverementaldisorderin specialisthealthcareinNorway,ofwhom12.6,9.2,and3.3per100 000personyearswerediagnosedwithschizophrenia-likedisorder,bipolardisorderorseveredepressiveepisodewithpsychotic symptoms,respectively(Table2).Overallmoregirls(n=495)than boys (n=389) were diagnosed with a first-episode of severe mentaldisorder,duetoahighernumberofgirlsbeingdiagnosed withbipolardisorderorseveredepressiveepisodewithpsychotic symptoms.However,moreboysthangirlswerediagnosedwith schizophrenia-like disorder.Mean age at first registered severe mentaldisorderwas16years.Only3.7%ofthe884patientshad theirfirstregisteredpsychoticdisorderdiagnosispriortotheageof 12,while23.9%werefirstdiagnosedbetween12and15yearsof age,and72.4%werefirstdiagnosedbetween16and18yearsofage.

In2011,113boysand160girlswerediagnosedwithasevere mentaldisorderwithnoprior registrationsintheperiod2008– 2010,yieldinganincidencerateof18.6per100000personyears forboysand27.9per100000personyearsforgirls.Corresponding rateswereforschizophrenia-likedisorder12.4forboysand11.0for girls,for bipolardisorder4.6forboysand 11.5forgirls,andfor severedepressiveepisodewithpsychoticsymptoms1.7forboys and5.6forgirls.

3.2.Co-morbidmentaldisorders

The most prevalent co-morbid mental disorders diagnosed betweenoneyearpriortoandoneyearaftertheincidentsevere mental disorder diagnosis were depressive disorder (38.1%), anxiety disorder (31.2%) and hyperkinetic disorder (16.6%) (Table 3). Analysis of early diagnostic markers revealed that within one year prior to the incident severe mental disorder diagnoses, 13.3% of schizophrenia patients had a depressive disorderdiagnosis,14.8%hadananxietydisorderdiagnosis,and 10.2%hada hyperkineticdisorderdiagnosis.The corresponding percentagesforbipolardisorderpatientswere21.1%(depressive disorder), 13.9% (anxiety disorder), and 14.6% (hyperkinetic disorder), and for patients with psychotic depression the percentages were 28.0% (depressive disorder), 16.9% (anxiety disorder),and2.5%(hyperkineticdisorder).Inthetotalgroupof patients, 10.1% were diagnosed with a SUD, and 5.7% were

diagnosed withSUD duringtheyearprior totheﬁrstdiagnosis ofaseverementaldisorder.

3.3.Useofpsychotropicdrugs

Amongthe884childrenandadolescentsfirstdiagnosedwitha severe mental disorder during 2009–2011, 552 (62.4%) were prescribedanantipsychoticdrugbetweenoneyearpriortoand oneyearaftertheincidentdiagnosisofseveremental disorder, predominantlyasecondgenerationantipsychotic(Table4).Among the patients using antipsychotic medication, 486 (88.0%) filled morethanoneprescriptionduringthetwoyearsofobservation, indicatingregularuse.Theproportionofpatientsfillingmorethan one prescription was higher among those diagnosed with schizophrenia-likedisorder(90.9%),thanamongthosediagnosed withbipolardisorder(85.6%) orseveredepressiveepisodewith psychotic symptoms (80.3%). The most commonly prescribed antipsychoticdrugwasquetiapine(29.5%),followedbyaripipra- zole(19.6%),olanzapine(17.3%)andrisperidone(16.6%).Ingeneral, patientsdiagnosed withschizophrenia-like disorderweremore likelytobeprescribedanantipsychoticdrug(72.0%)comparedto patients diagnosed with bipolar disorder (51.7%) or severe depressiveepisodewithpsychoticsymptoms(55.4%).

Anantiepilepticdrugwasprescribedto26.4%ofthepatients, predominantly to patients diagnosed with a bipolar disorder.

Antidepressantswereprescribedto35.1%,andanxiolyticdrugsto 13.2%. Lithiumwas prescribed to8.0% of patients withbipolar disorder, and psychostimulants were prescribed to 19.2% of patientswith schizophrenia-likedisorder and 20.1% ofpatients withbipolardisorder.

4.Discussion

Inthepresentstudywereportthat25per100000childrenand adolescents in Norway werediagnosed with a ﬁrst episode of severementaldisordereachyearduring2009–2011.Halfofthe patients were diagnosed with schizophrenia-like disorder, 37%

withbipolardisorder,and13%withpsychoticdepression.About one third of the patients were diagnosed with co-morbid depressiveoranxietydisorderaroundthetimeoftheirincident

Table3

Co-morbidmentaldisordersinchildrenandadolescentswithseverementaldisorders.DatafromtheNorwegianPatientRegistryonall0–18yearoldboysandgirlsregistered withafirst-episodeofschizophrenia-likedisorder,bipolardisorder,ordepressivedisorderwithpsychoticsymptomsduring2009–2011.Co-morbiddiagnosesaredefinedas anyregistereddiagnosiswithsubstanceusedisorder(ICD-10codesF10-F19),depressivedisorder(F32-F34),anxietydisorders(F40-F48),hyperkineticdisorder(F90),mental retardation(F70-F79),autism-spectrumdisorder(F84),orconductdisorder(F91)betweenoneyearpriortoandoneyearafterthefirstregistereddiagnosisofaseveremental disorder.

n % n % n % n %

Substanceusedisorder 41 9.3 37 11.5 11 9.3 89 10.1

Substanceusedisorder1yearpriortoincidentseverementaldisorderdiagnosis 23 5.2 22 6.8 5 4.2 50 5.7 Substanceusedisorder1yearafterincidentseverementaldisorderdiagnosis 34 7.7 23 7.1 8 6.8 65 7.4

Depressivedisorder 123 27.8 96 29.7 118 100 337 38.1

Depressivedisorder1yearpriortoincidentseverementaldisorderdiagnosis 59 13.3 68 21.1 33 28.0 160 18.1 Depressivedisorder1yearafterincidentseverementaldisorderdiagnosis 104 23.5 68 21.1 111 94.1 283 32.0

Anxietydisorder 141 31.8 81 25.1 54 45.8 276 31.2

Anxietydisorder1yearpriortoincidentseverementaldisorderdiagnosis 79 14.8 45 13.9 20 16.9 144 16.3 Anxietydisorder1yearafterincidentseverementaldisorderdiagnosis 117 26.4 61 18.9 52 44.1 230 26.0

Hyperkineticdisorder 63 14.2 75 23.2 9 7.6 147 16.6

Hyperkineticdisorder1yearpriortoincidentseverementaldisorderdiagnosis 45 10.2 47 14.6 <5 n.a. 95 10.7 Hyperkineticdisorder1yearafterincidentseverementaldisorderdiagnosis 55 12.4 70 21.7 8 6.8 133 15.0

Mentalretardation 36 8.1 12 3.7 <5 n.a. 50 5.7

Autismspectrumdisorder 44 9.9 13 4.0 <5 n.a. 61 6.9

Conductdisorder 8 1.8 20 6.2 <5 n.a. 29 3.3

(5)

severementaldisorderdiagnosis.Morethan60%ofthepatients ﬁlled prescriptions for antipsychoticmedication, mostlysecond generationantipsychotics,and 88%of patientsonantipsychotic medication ﬁlled more than one prescription which indicates regularuse.

Thepresentincidenceratesfromspecialisthealthcaresettings aresubstantiallylowerthanwhatwas foundintheUS National Comorbidity Survey of mental disorders on a representative community sample of 13–18year olds, finding that 2.9% met diagnosticcriteriaforbipolardisordertypeIorII[8].However,our findingsaremoreinlinewithaDanishregistry-basedstudyfrom specialisthealthcare,whichdemonstrateda12-yearcumulative incidenceof0.3%forschizophrenia-likedisorderandabout0.1%for bipolardisorderby20yearsofage[4].Thecurrentfindingsarealso similar to incidence rates of bipolar disorder in children and adolescents found in health care settings in the UK [7] and Germany [10]. In a registry-based study from Denmark, the incidence of schizophrenia-like disorder (F20-F29) among 5– 18yearoldsincreasedsignificantlyduringtheperiod2000–2012 [27],demonstratinghigherincidenceratesofschizophreniawhen outpatient contacts were included in the Danish Psychiatric CentralRegister[28].Inthepresentstudy,theincidenceratefor schizophrenia-likedisorder(F20-F29)in2011was11–12per100 000 person years. In comparison, the mean age-standardized incidencerate for the more narrowdiagnosis of schizophrenia (F20)among0–18yearoldswas5.2per100000personyearsin theDanishstudy[28].

Prevalenceestimatesofseverementaldisordersin European childrenandadolescentsarescarce,sincethepaediatricpopula- tionhas beenexcluded in previous interview-based surveys of mentaldisorders inthegeneral populationin Norway[29],the Netherlands[30],andintheEuropeanStudyoftheEpidemiology ofMentalDisorders[31],leavingthepresentﬁguresvaluable.

Co-morbid diagnoses of depressive, anxiety or hyperkinetic disorderswereregisteredin38.1%,31.2%,and16.6%,respectively, ofpatientswithseverementaldisorder.Anxietydisorders were mostcommoninpatientsﬁrstdiagnosedwithschizophrenia-like disorder (31.8%) or severe depressive episode with psychotic symptoms(45.8%),whilehyperkineticdisorderwasmostcommon among patients ﬁrst diagnosed with bipolar disorder (23.2%).

Depressiveillnessandanxietydisordersarecommonlyreportedin clinicalstudiesof adultﬁrst-episodepsychosis patients[13,14], and the results of the present study further emphasize that symptomsofdepressiveillnessoranxietydisordermayserveas earlysigns,albeitnon-speciﬁc,ofseverementaldisordersalsoin

childrenandadolescents.DatafromtheAvonLongitudinalStudy of Parents and Children (ALSPAC) birth cohort have shown a considerableoverlapinearlyriskfactorsforanxiety,depression andpsychotic-likesymptomsinadolescents,indicatingacontin- uousunderlying commonfactor of mental distress [32]. In the presentstudy,13–21%ofthepatientshadeitherananxietyora depressivediagnosisduringoneyearpriortohavingfirst-episode schizophrenia-likeorbipolardisorder.Further,thefactthat23%of patientswithfirst-episodebipolardisorderwerealsodiagnosed with hyperkinetic disorder concurs with findings from clinical studieslinkingirritabilityorhyperactivitytobipolardisorderin youngage[33].

Amongchildrenandadolescentsdiagnosedwithseveremental disorders,62.4%wereprescribedantipsychoticmedication,most commonlyasecond-generationantipsychotic.Theproportionon antipsychoticdrugs washigher inpatientsfirstdiagnosed with schizophrenia-like disorder (72.0%) compared to patients first diagnosedwithbipolardisorder(51.7%)orpsychoticdepression (55.4%).Wehavepreviouslyshownthatchildrenandadolescents whouseantipsychoticmedicationarecommonlydiagnosedwith non-psychoticdisorderslikehyperkinetic,anxietyordepressive disorder[22].Thefindingsofthepresentstudydemonstratethat childrenand adolescentsdiagnosed withschizophrenia, bipolar disorder or psychotic depression are usually also prescribed antipsychotic medication, and the great majority of patients treatedwithantipsychoticmedicationuseitregularly.

Tothebestofourknowledge,therearenopublishedstudies fromNorwaylinkinginformationondiagnosesofseveremental disorders in children and adolescents to information on their psychotropic drug use, but a previous registry-based study indicatedanincreaseinuseofmoodstabilizersamong0–17year oldsingeneralbetween2004and2007inNorway[34].Themost commonly used drugs among the patientsin our study were quetiapine (29.5%),aripiprazole (19.6%),olanzapine(17.3%),and risperidone(16.6%).Thesefourtypesofantipsychoticdrugsalso representthemostcommonlyprescribedantipsychoticdrugsto Norwegianchildrenandadolescentsinthegeneralpopulationas describedinourpreviousstudy[22].InNorway,onlyrisperidone, aripiprazoleandziprasidoneareapprovedforuseinchildrenand adolescents,anduseofquetiapineandolanzapinemusttherefore beregardedasoff-label.Thesesubstancesare,however,approved forpaediatricuseinUSA,whichmayexplainuseelsewhere.During the later years there has been an increasing off-label use of especiallyquetiapineinchildrenandadolescentsinNorway[35].

Oneexplanationtothismaybethatweight-gainseemstobea Table4

Psychotropicdruguseamongchildrenandadolescentsdiagnosedwithseverementaldisorder.DatafromtheNorwegianPrescriptionDatabaseonalldispensedprescriptions betweenoneyearpriortoandoneyearafteranincidentdiagnosisofschizophrenia-likedisorder,bipolardisorder,orseveredepressiveepisodewithpsychoticsymptoms among0–18yearoldsasregisteredintheNorwegianPatientRegistryfortheperiod2009–2011.

Drugtype n % N % n % n %

Anyantipsychotic 319 72.0 167 51.7 66 55.4 552 62.4

Secondgenerationantipsychotic 307 69.3 158 48.9 59 50.0 524 59.3

Risperidone 99 22.3 28 6.3 20 16.9 147 16.6

Aripiprazole 115 26.0 47 14.6 11 9.3 173 19.6

Quetiapine 136 30.7 87 26.9 38 32.2 261 29.5

Olanzapine 100 22.6 41 12.7 12 10.2 153 17.3

Antiepileptic 37 8.4 187 57.9 9 7.6 233 26.4

Lithium <5 n.a. 26 8.0 0 0 29 3.3

Antidepressant 122 27.5 117 36.2 71 60.2 310 35.1

Anxiolytic 62 14.0 48 14.9 7 5.9 117 13.2

Psychostimulant 57 19.2 65 20.1 7 5.9 129 14.6

(6)

lesser problem with quetiapine, compared to for example olanzapineamongchildrenandadolescentsingeneral.However, a recently publishedrandomized controlled trialdemonstrated thatadolescentsusingquetiapinefor12weeksgained3kgmore weightthanadolescentsonaripiprazole[18].Thusthechoiceof antipsychotic drugs for children and adolescents strikes a fine balancebetween efficacy and adverse effects profile. Clinicians whotreatyoungpatientswithantipsychoticdrugsshouldcarefully monitorbodyweight,bloodlipidsandsubjectivesignsofadverse effects.

4.1.Strengthsandlimitations

Themajorstrengthofthepresentstudywastheuseofdata frommandatoryhealthregistriescoveringtheentirecountryand allavailablespecialisthealthcarefacilities.Thisensuredacloseto complete coverage of patients in treatment for severe mental disorderswithnoselectionbias. However,the resultsmust be interpretedwiththefollowinglimitationsinmind.First,thedata onlyincludepatientsinspecialisthealthserviceandnotpatients treatedby their GP only. Mental health care for children and adolescentsisfreeofchargeandreadilyavailablethroughoutthe country,presumably lowering the thresholdfor seeking treatment. National clinical guidelines stress that patients with suspectedpsychoticillnessshouldbeevaluatedbyapsychiatrist orpsychologistwithintwoweeks[24].GPsarethereforestrongly encouraged to refer children and adolescents with suspected psychosistomentalhealthcareforevaluationbyaspecialistas soonaspossible.Second,individual-leveldatainNPRhavebeen availablesince2008,andtheshortobservationperiodlimitsthe opportunitytoobtaintrueincidentcases.Inthecurrentstudy, patientswereconsideredincidentcasesat thedateofthefirst registereddiagnosisofaseverementaldisorderafter31Decem- ber2008iftherewerenopreviousregistrationssincebeginning of2008.However,lessthan1%ofpatientsdiagnosedwithF20- F31,F32.3orF33.3in2011,butnotin2010,werediagnosedwith anyoftherelevantdiagnosesduring2008–2009,whichindicates thatmore than 99% of our assumed incident cases were true incidentcases. Nevertheless,wecannot excludethepossibility thatsomepatientsregisteredintheNPRforthefirsttimeinthe period 2009–2011 could have been diagnosed with a severe mentaldisorderpriorto2008,whenindividualdatafromNPR were not available. Sensitivity analyses revealed that the prevalence of co-morbid disorders and psychotropic drug use was similar for patients registered for the first time in 2011 comparedto thetotalgroup registered for the first time during 2009–2011 (data not shown). Third, diagnostic data registeredintheNPRarereportedbythetreatingpsychiatristor psychologist and not subjectedtoexternalquality assessment.

However, a recent study demonstrated a high degree of agreement between the clinical diagnoses as reported to the NPRanddiagnosesbasedonstructureddiagnosticinterviewina clinical research setting for adult patients withsevere mental disorder[36].Fourth,alimitationwiththeNorPDisthatitonly includes data on ﬁlled prescriptions with no information on secondarycompliancetotreatment.

5.Conclusion

During 2009–2011, theincidenceof severemental disorders among children and adolescents in Norway was 25 per 100 000personyears.Abouthalfofthepatientswerediagnosedwith schizophrenia-like disorder, and a third with bipolar disorder.

Morethan30%ofthepatientshadco-morbiddepressiveillnessor anxietydisorder,and10%hadSUD.Mostofthediagnosedpatients usedantipsychoticmedication,andabout90%ofthepatientson

antipsychoticmedicationﬁlledmorethanoneprescriptionwhich indicatesregularuse.Secondgenerationantipsychoticswerebyfar themostcommonlyprescribedtypes,andstudiesofthelong-term efﬁcacy and safety of antipsychotic drug treatment in young patientswithpsychoticdisordersareneeded.

Conﬂictofintereststatement

The authors declare that they have no conﬂict of interest.

AuthorSShadfullaccesstothedataandperformedallstatistical analyses.DatafromtheNPRandtheNorPDhavebeenusedinthis publication.Theinterpretationandreportingofthesedataarethe sole responsibility of theauthors, and noendorsement by the registryownersisintendednorshouldbeinferred.

Funding

Thisresearchdidnotreceiveanyspeciﬁcgrantfromfunding agenciesinthepublic,commercial,ornot-for-proﬁtsectors.

Authorcontribution

RNconceptualizedanddesignedthestudy,andwrotethefirst draft of the manuscript. JGB, MH, and IH conceptualized and designed the study. VH prepared the data for analysis. SS conceptualizedanddesignedthestudyandcarriedouttheinitial analyses. All authors participated in interpretation of the data, critical review of themanuscript for scientific content, and all authorshaveapprovedofthefinalversionofthemanuscript.

References

[1]HafnerH,NowotnyB.Epidemiologyofearly-onsetschizophrenia.EurArch PsychiatryClinNeurosci1995;245:80–92.

[2]RemschmidtH,TheisenFM.Schizophreniaandrelateddisordersinchildren andadolescents.JNeuralTransm2005;(Suppl.69):121–41.

[3]Schimmelmann BG, Schmidt SJ, Carbon M, Correll CU. Treatment of adolescentswithearly-onsetschizophreniaspectrumdisorders:insearch ofarational,evidence-informedapproach.CurrOpinPsychiatry2013;26:219–

30.

[4]PedersenCB,MorsO,BertelsenA,WaltoftBL,AgerboE,McGrathJJ,etal.A comprehensivenationwidestudyoftheincidencerateandlifetimeriskfor treatedmentaldisorders.JAMAPsychiatry2014;71:573–81.

[5]ClemmensenL,VernalDL,SteinhausenHC.Asystematicreviewofthelong- termoutcomeofearlyonsetschizophrenia.BMCPsychiatry2012;12:150.

[6]Rapoport JL, Gogtay N. Childhood onset schizophrenia: support for a progressiveneurodevelopmentaldisorder.IntJDevNeurosci2011;29:251–8.

[7]SharmaA,NeelyJ,CamilleriN,JamesA,GrunzeH,LeCouteurA.Incidence, characteristicsandcourseofnarrowphenotypepaediatricbipolarIdisorderin theBritishIsles.ActaPsychiatrScand2016;134:522–32.

[8]MerikangasKR,HeJP,Burstein M,SwansonSA,AvenevoliS,CuiL,etal.

LifetimeprevalenceofmentaldisordersinU.S.adolescents:resultsfromthe NationalComorbiditySurveyReplication-AdolescentSupplement(NCS-A).J AmAcadChildAdolescPsychiatry2010;49:980–9.

[9]PostRM,LuckenbaughDA,LeverichGS,AltshulerLL,FryeMA,SuppesT,etal.

Incidenceofchildhood-onsetbipolarillnessintheUSAandEurope.BrJ Psychiatry2008;192:150–1.

[10]RaoP,MooreJK,StewartR,RunionsK,BearN,WongJW,etal.Bipolardisorder inchildrenandadolescents:diagnosticinpatientratesfrom2000to2013in Germany.IntJBipolarDisord2016;4:23.

[11]IyerSN,BoekestynL,Cassidy CM,KingS,JooberR, MallaAK. Signsand symptomsinthepre-psychoticphase:descriptionandimplicationsfor diagnostictrajectories.PsycholMed2008;38:1147–56.

[12]SönmezN,RøssbergJI,EvensenJ,BarderHE,HaahrU,TenVeldenHegelstadW, etal.Depressivesymptomsinﬁrst-episodepsychosis:a10-yearfollow-up study.EarlyIntervPsychiatry2016;10:227–33.

[13]RommKL,MelleI,ThoresenC,AndreassenOA,RossbergJI.Severesocial anxietyinearlypsychosisisassociatedwithpoorpremorbidfunctioning, depression,andreducedqualityoflife.ComprPsychiatry2012;53:434–40.

[14]Sonmez N, Romm KL, AndreasssenOA, MelleI, Rossberg JI.Depressive symptomsinﬁrstepisodepsychosis:aone-yearfollow-upstudy.BMC Psychiatry2013;13:106.

[15]FriisS,MelleI,JohannessenJO,RøssbergJI,BarderHE,EvensenJH,etal.Early predictorsoften-yearcourseinﬁrst-episodepsychosis.PsychiatrServ 2016;67:438–43.

(7)

[16]PathakP,WestD,MartinBC,HelmME,HendersonC.Evidence-baseduseof second-generationantipsychoticsinastateMedicaidpediatricpopulation, 2001–2005.PsychiatrServ2010;61:123–9.

[17]PagsbergAK,TarpS,GlintborgD,StenstrømAD,Fink-JensenA,CorrellCU,etal.

Acuteantipsychotictreatmentofchildrenandadolescentswith

schizophrenia-spectrumdisorders:asystematicreviewandnetworkmeta- analysis.JAmAcadChildAdolescPsychiatry2017;56:191–202.

[18]PagsbergAK,JeppesenP,KlauberDG,JensenKG,RudåD,Stentebjerg-Olesen M,etal.Quetiapineextendedreleaseversusaripiprazoleinchildrenand adolescentswithﬁrst-episodepsychosis:themulticentre,double-blind, randomisedtolerabilityandefﬁcacyofantipsychotics(TEA)trial.Lancet Psychiatry2017;4:605–18.

[19]StaffordMR,Mayo-WilsonE,LoucasCE,JamesA,HollisC,BirchwoodM,etal.

Efﬁcacyandsafetyofpharmacologicalandpsychologicalinterventionsforthe treatmentofpsychosisandschizophreniainchildren,adolescentsandyoung adults:asystematicreviewandmeta-analysis.PLoSOne2015;10:e0117166.

[20]BenAmorL.Antipsychoticsinpediatricandadolescentpatients:areviewof comparativesafetydata.AffectDisord2012;(Suppl.138):S22–30.

[21]KumraS,OberstarJV,SikichL,FindlingRL,McClellanJM,VinogradovS,etal.

Efﬁcacyandtolerabilityofsecond-generationantipsychoticsinchildrenand adolescentswithschizophrenia.SchizophrBull2008;34:60–71.

[22]NesvågR,HartzI,BramnessJG,HjellvikV,HandalM,SkurtveitS.Mental disorderdiagnosesamongchildrenandadolescentswhouseantipsychotic drugs.EurNeuropsychopharmacol2016;26:1412–8.

[23]Munk-JorgensenP,OkkelsN,GolbergD,RuggeriM,ThornicroftG.Fiftyyears' developmentandfutureperspectivesofpsychiatricregisterresearch.Acta PsychiatrScand2014;130:87–98.

[24]The Norwegian Directorate of Health. Nasjonal faglig retningslinje for utredning,behandlingogoppfølgingavpersonermedpsykoselidelser.

Helsedirektoratet. IS-1957.(inNorwegian).

[25]World Health Organizatioin. The ICD-10 Classiﬁcation of mental and behaviouraldisorders.Clinicaldescriptionsanddiagnosticguidelines.

Geneva,Switzerland:WorldHealthOrganization;1992.

[26]ATCClassiﬁcationIndexwithDDDsOslo:WHOCollaboratingCentreforDrug StatisticsMethodology.[Availablefrom:http://www.whocc.no/].

[27]KuhlJOG,LaursenTM,ThorupA,NordentoftM.Theincidenceofschizophrenia andschizophreniaspectrumdisordersinDenmarkintheperiod2000–2012.A register-basedstudy.SchizophrRes2016;176:533–9.

[28]OkkelsN,VernalDL, Jensen SO,McGrathJJ,NielsenRE. Changesin the diagnosedincidenceofearlyonsetschizophreniaoverfourdecades.Acta PsychiatrScand2013;127:62–8.

[29]KringlenE,TorgersenS,CramerV.ANorwegianpsychiatricepidemiological study.AmJPsychiatry2001;158:1091–8.

[30]TenHaveM,VolleberghW,BijlR,NolenWA.Bipolardisorderinthegeneral populationinTheNetherlands(prevalence,consequencesandcare utilisation):resultsfromTheNetherlandsMentalHealthSurveyandIncidence Study(NEMESIS).JAffectDisord2002;68:203–13.

[31]AlonsoJ,AngermeyerMC,BernertS,BruffaertsR,BrughaTS,BrysonH,etal.

ESEMeD/MHEDEA2000investigators,EuropeanStudyoftheEpidemiologyof MentalDisorders(ESEMeD)project:prevalenceofmentaldisordersinEurope:

resultsfromtheEuropeanStudyoftheEpidemiologyofMentalDisorders (ESEMeD)project.ActaPsychiatrScand2004;420:21–7Suppl.

[32]NiarchouM,ZammitS,LewisG.TheAvonLongitudinalStudyofParentsand Children(ALSPAC)birthcohortasaresourceforstudyingpsychopathologyin childhoodandadolescence:asummaryofﬁndingsfordepressionand psychosis.SocPsychiatryPsychiatrEpidemiol2015;50:1017–27.

[33]RylesF,MeyerTD,Adan-ManesJ,MacmillanI,ScottJ.Asystematicreviewof thefrequencyandseverityofmanicsymptomsreportedinstudiesthat comparephenomenologyacrosschildren,adolescentsandadultswithbipolar disorders.IntJBipolarDisord2017;5:4.

[34]BramnessJG,GrøholtB,EngelandA,FuruK.Theuseoflithium,valproateor lamotrigineforpsychiatricconditionsinchildrenandadolescentsinNorway 2004–2007–aprescriptiondatabasestudy.JAffectDisord2009;117:208–11.

[35]GjerdenP,BramnessJG,TveteIF,SlordalL.Theantipsychoticagentquetiapine isincreasinglynotusedassuch:dispensedprescriptionsinNorway2004–

2015.EurJClinPharmacol2017;73(9):1173–9.

[36]NesvågR,JönssonEG,BakkenIJ,KnudsenGP,BjellaTD,Reichborn-Kjennerud T,etal.Thequalityofseverementaldisorderdiagnosesinanationalhealth registryascomparedtoresearchdiagnosesbasedonstructuredinterview.

BMCPsychiatry2017;17:93.