Download Self-Reported Depressive Symptoms and Antisocial

MINNA RITAKALLIO
Self-Reported Depressive Symptoms and
Antisocial Behaviour in Middle Adolescence
ACADEMIC DISSERTATION
To be presented, with the permission of
the Faculty of Medicine of the University of Tampere,
for public discussion in the Auditorium of
Tampere School of Public Health, Medisiinarinkatu 3,
Tampere, on May 23rd, 2008, at 12 o’clock.
U N I V E R S I T Y O F TA M P E R E
ACADEMIC DISSERTATION
University of Tampere, School of Public Health
Tampere University Hospital, Department of Adolescent Psychiatry
Doctoral Programs in Public Health (DPPH)
Finland
Supervised by
Adjunct Professor Riittakerttu Kaltiala-Heino
University of Tampere
Distribution
Bookshop TAJU
P.O. Box 617
33014 University of Tampere
Finland
Reviewed by
Adjunct Professor Kirsi Suominen
University of Helsinki
Professor Pirkko Räsänen
University of Oulu
Tel. +358 3 3551 6055
Fax +358 3 3551 7685
[email protected]
www.uta.fi/taju
http://granum.uta.fi
Cover design by
Juha Siro
Acta Universitatis Tamperensis 1304
ISBN 978-951-44-7276-3 (print)
ISSN 1455-1616
Tampereen Yliopistopaino Oy – Juvenes Print
Tampere 2008
Acta Electronica Universitatis Tamperensis 711
ISBN 978-951-44-7277-0 (pdf )
ISSN 1456-954X
http://acta.uta.fi
3
Contens
LIST OF ORIGINAL PUBLICATIONS ................................................................... 7
ABBREVIATIONS .................................................................................................. 8
ABSTRACT ............................................................................................................10
TIIVISTELMÄ ........................................................................................................12
1. INTRODUCTION ...............................................................................................14
2. REVIEW OF THE LITERATURE.......................................................................15
2.1 Adolescent development ....................................................................................15
2.2 Depression in adolescence..................................................................................16
2.2.1 Definition of depression.............................................................................16
2.2.2 Features of adolescent depression ..............................................................19
2.2.3 Identifying depression ...............................................................................20
2.2.4 Epidemiology of adolescent depression .....................................................21
2.2.4.1 Methodological issues in epidemiological studies
on adolescent depression ............................................................................21
2.2.4.2 Prevalence and incidence of adolescent depression .........................23
2.2.5 Etiology of adolescent depression..............................................................25
2.2.6 Course and consequences of adolescent depression....................................28
2.2.7 Comorbidity in adolescent depression........................................................30
2.3 Antisocial behaviour in adolescence ...................................................................31
2.3.1 Definition of antisocial behaviour ..............................................................31
2.3.2 Features of adolescent antisocial behaviour................................................36
2.3.3 Identifying antisocial behaviour.................................................................39
2.3.4 Epidemiology of adolescent antisocial behaviour.......................................41
2.3.4.1 Methodological issues in epidemiological studies on adolescent
antisocial behaviour ...................................................................................41
4
2.3.4.2 Prevalence and incidence of adolescent antisocial behaviour ..........42
2.3.5 Development, course and consequences of adolescent
antisocial behaviour............................................................................................45
2.3.5.1 Etiology of adolescent antisocial behaviour....................................45
2.3.5.2 Models on the development of adolescent antisocial behaviour ......48
2.3.5.3 Course and consequences of adolescent antisocial behaviour .........51
2.3.6 Comorbidity in adolescent antisocial behaviour .........................................53
2.4 Comorbidity between depression and antisocial behaviour in adolescence..........54
2.4.1 Features of comorbid depression and antisocial behaviour
in adolescence ....................................................................................................54
2.4.2 Epidemiology of comorbid depression and antisocial behaviour
in adolescence ....................................................................................................55
2.4.3 Models on the development of comorbid depression and antisocial
behaviour in adolescence....................................................................................60
2.4.4 Course and consequences of comorbid depression and antisocial
behaviour in adolescence....................................................................................64
2.5 Summary of the literature reviewed ....................................................................65
3. AIMS OF THE STUDY ......................................................................................67
4. MATERIALS AND METHODS..........................................................................68
4.1 The School Health Promotion Study (I and II)....................................................68
4.1.1 Sample and procedures (I and II) ...............................................................68
4.1.2 Measures (I and II) ...................................................................................69
4.1.3 Data analyses (I and II)..............................................................................72
4.2 The Adolescent Mental Health Cohort Study (III and IV)...................................72
4.2.1 Baseline of the AMHC Study (III) .............................................................73
4.2.1.1 Sample and procedure (III).............................................................73
4.2.1.2 Measures (III) ................................................................................73
4.2.1.3 Data analyses (III)..........................................................................75
4.2.2 Two-year follow-up of the AMHC Study (IV) ...........................................76
4.2.2.1 Sample and procedure (IV) ............................................................76
4.2.2.2 Measures (IV) ...............................................................................77
5
4.2.2.3 Data analyses (IV)..........................................................................77
5. RESULTS............................................................................................................78
5.1 Association between self-reported depressive symptoms and delinquent
behaviour (I) ............................................................................................................78
5.2 Patterns of delinquent behaviour among depressed and non-depressed
repeatedly delinquent adolescents (II) ......................................................................79
5.2.1 Frequency..................................................................................................79
5.2.2 Versatility..................................................................................................79
5.2.3 Specialization ............................................................................................80
5.3 Interrelations between self-reported depressive symptoms, antisocial behaviour
and perceived social support (III) .............................................................................80
5.3.1 Association between self-reported depressive symptoms and perceived
social support .....................................................................................................80
5.3.2 Association between antisocial behaviour and perceived social support .....81
5.3.3 Role of perceived social support in the comorbidity between self-reported
depressive symptoms and antisocial behaviour ...................................................81
5.4 Homotypic continuity, concurrent comorbidity and longitudinal associations
between self-reported depressive symptoms and antisocial behaviour (IV)...............82
5.4.1 Homotypic continuity ................................................................................82
5.4.2 Concurrent comorbidity.............................................................................82
5.4.3 Longitudinal associations...........................................................................83
5.4.3.1 Failure model.................................................................................83
5.4.3.2 Acting out model ...........................................................................83
6. DISCUSSION......................................................................................................84
6.1 Associations between perceived social support and self-reported depressive symptoms
and antisocial behaviour (III) ...................................................................................84
6.2 Homotypic continuity of self-reported depressive symptoms and antisocial behaviour
in middle adolescence (IV).......................................................................................84
6.3 Concurrent comorbidity between self-reported depressive symptoms and antisocial
behaviour in middle adolescence (I, II, III, IV).........................................................85
6
6.4 Patterns of delinquent behaviour among depressed and non-depressed adolescents
in middle adolescence (I, II).....................................................................................87
6.5 Longitudinal associations between self-reported depressive symptoms and
antisocial behaviour in middle adolescence (IV).......................................................89
6.6 Limitations of the study......................................................................................90
6.7 Strengths of the study.........................................................................................92
7. CONCLUSIONS .................................................................................................93
8. CLINICAL IMPLICATIONS...............................................................................93
9. RESEARCH IMPLICATIONS ............................................................................95
10. ACKNOWLEDGEMENTS................................................................................96
11. REFERENCES ..................................................................................................98
12. ORIGINAL PUBLICATIONS .........................................................................134
7
LIST OF ORIGINAL PUBLICATIONS
This dissertation is based on the following original publications, which are referred to in the text by
the Roman numerals I-IV:
I Ritakallio M, Kaltiala-Heino R, Kivivuori J and Rimpelä M (2005): Brief report: Delinquent
behaviour and depression in middle adolescence: a Finnish community sample. J Adolesc 28:155–
159.
II Ritakallio M, Kaltiala-Heino R, Kivivuori J, Luukkaala T and Rimpelä M (2006): Delinquency
and the profile of offences among depressed and non-depressed adolescents. Crim Behav Ment
Health 16:100–110.
III Ritakallio M, Luukkaala T, Marttunen M, Pelkonen M and Kaltiala-Heino R. Comorbidity
between depression and antisocial behaviour in middle adolescence: the role of perceived social
support. (Accepted in Nord J Psychiatry)
IV Ritakallio M, Koivisto A, von der Pahlen B, Pelkonen M, Marttunen M, Kaltiala-Heino R (in
press): Continuity, comorbidity and longitudinal associations between depression and antisocial
behaviour in middle adolescence: A 2-year prospective follow-up study. J Adolesc.
The publications are reprinted with the kind permission of Elsevier Ltd (I and IV), John Wiley &
Sons Ltd (II) and Taylor & Francis (III).
8
ABBREVIATIONS
AACAP
American Academy of Child and Adolescent Psychiatry
ADHD
Attention Deficit Hyperactivity Disorder
AL
Adolescent limited group
AMHC Study
The Adolescent Mental Health Cohort Study
APA
American Psychiatric Association
APD
Antisocial Personality Disorder
BDI
Beck Depression Inventory
CAPA
The Child and Adolescent Psychiatric Assessment
CBCL
Child Behavior Checklist
CD
Conduct Disorder
CDI
Children Depression Inventory
CES-D
Center for Epidemiological Studies-Depression Scale
CI
Confidence interval
DISC
Diagnostic Interview Schedule for Children
DOF
Parent Daily Report and Direct Observation Form
DSM
Diagnostic and Statistical Manual of Mental Disorders
DSM-III
Diagnostic and Statistical Manual of Mental Disorders, third edition
DSM-III-R
Diagnostic and Statistical Manual of Mental Disorders, third edition,
revised
DSM-IV
Diagnostic and Statistical Manual of Mental Disorders, fourth edition
FSRD
The Finnish Self-Reported Delinquency Study
HRSD
Hamilton Rating Scale for Depression
ICD-10
International Classification of Diseases, tenth edition
IQ
Intelligence Quotient
ISRD-I
The International Self-Report Delinquency Study I
K-SADS
The Schedule for Affective Disorders and Schizophrenia for SchoolAged Children
LCP
Life-course persistent group
MDD
Major Depression Disorder
MDE
Major Depressive Episode
9
ODD
Oppositional Defiant Disorder
OR
Odds Ratio
PSSS-R
The Perceived Social Support Scale-Revised
RADS
Reynolds Adolescent Depression Scale
RBDI
The Finnish modification of the 13-item Beck Depression Inventory
SDQ
The Strengths and Difficulties Questionnaire
SRD
Self-Report Delinquency Scale
SES
Socioeconomic Status
SHPS
The School Health Promotion Study
STAKES
Sosiaali- ja terveysalan tutkimus- ja kehittämiskeskus (National
Research and Development Centre for Welfare and Health)
TRF
Teacher Report Form
YSR
Youth Self-Report
10
ABSTRACT
The general aim of this study was to analyse associations between self-reported depressive
symptoms and antisocial behaviour both concurrently and longitudinally in a community sample
during middle adolescence. In more detail, the study aimed to investigate the continuity of selfreported depressive symptoms and antisocial behaviour, associations between self-reported
depressive symptoms, antisocial behaviour and perceived social support, the patterns of
delinquency in relation to self-reported depressive symptoms and longitudinal associations between
self-reported depressive symptoms and antisocial behaviour.
The present study is based on two Finnish community studies among adolescents: the School
Health Promotion Study (SHPS) 2002 and the Adolescent Mental Health Cohort Study (AMHC
Study). SHPS is a nationwide cross-sectional survey carried out biannually in the same
municipalities. The AMHC Study is a prospective cohort study conducted in the Finnish cities of
Tampere and Vantaa, and consists of assessments at baseline (9th grade) and at 2-year follow-up.
The subjects were 14 to16-year-old adolescents in SHPS [N 50, 569 (50.1% boys)] and 15 to 17year-old adolescents in the AMHC Study [N 3, 278 (50.9% boys) at baseline, N 2, 070 (43.6%
boys) at follow-up]. The data was collected through school surveys based on self-report scales
previously widely used in depression and antisocial behaviour research.
According to the cross-sectional study perceived social support was inversely associated both with
self-reported depressive symptoms and antisocial behaviour. The results emphasized specially the
role of family support in buffering adolescents against depressive symptoms and antisocial
behaviour. Concurrent comorbidity between self-reported depressive symptoms and antisocial
behaviour in cross-sectional study design was high in both sexes, as adolescents displaying
antisocial behaviour had a up to seven-fold greater risk of being depressed. Although perceived
social support modified comorbidity between self-reported depressive symptoms and antisocial
behaviour, antisocial behaviour still remained independently associated with depressive
symptoms.The risk for self-reported depressive symptoms varied according to various forms of
delinquent acts and increased according to the increased frequency of delinquency. Among females
the strongest association was found in frequent vandalism and among males in frequent violence.
Further, delinquent behaviour of depressed adolescents was more versatile than that of nondepressed adolescents, who had mainly specialized in one type of offence.
11
According to the follow-up study both self-reported depressive symptoms and antisocial behaviour
showed considerable continuity during the 2-year period. There were no significant gender
differences either in the continuity of self-reported depressive symptoms or in the continuity of
antisocial behaviour. Self-reported depressive symptoms predicted subsequent antisocial behaviour
in a 2-year period among females, but among males the history of depressive symptoms protected
against subsequent antisocial behaviour. In either sex antisocial behaviour did not longitudinally
predict subsequent self-reported depressive symptoms when prior depressive symptoms were
controlled for.
In conclusion, the results of the present study suggest that self-reported depressive symptoms and
antisocial behaviour have a fairly stable course and remarkable comorbidity in middle adolescence.
Self-reported depressive symptoms differentiate patterns of delinquent behaviour. Special attention
should be paid to those adolescents who repeatedly behave antisocially because they are at
considerable risk for depressive symptoms. Other potential warning signs for depressive symptoms
are versatile delinquent behaviour as well as specialization on violence and vandalism. Depressed
females are at risk of subsequent antisocial behaviour.
12
TIIVISTELMÄ
Tutkimuksen tavoitteena oli analysoida itseilmoitettujen masennusoireiden ja epäsosiaalisen
käyttäytymisen välisiä yhteyksiä keskinuoruudessa poikkileikkaus- ja seurantatutkimusasetelmassa.
Tutkimuksessa selvitettiin itseilmoitettujen masennusoireiden ja epäsosiaalisen käyttäytymisen
jatkuvuutta; itseilmoitettujen masennusoireiden, epäsosiaalisen käyttäytymisen ja koetun sosiaalisen
tuen välisiä keskinäisiä yhteyksiä; rikollisen käyttäytymisen muotoja masentuneilla ja eimasentuneilla nuorilla sekä itseilmoitettujen masennusoireiden ja epäsosiaalisen käyttäytymisen
välisiä pitkittäisyhteyksiä.
Tutkimus perustuu kahteen suomalaiseen väestötutkimukseen: Kouluterveyskysely 2002 ja Nuorten
mielenterveys –kohorttitutkimus. Kouluterveyskysely on joka toinen vuosi samoissa kunnissa
tehtävä valtakunnallinen poikkileikkaustutkimus. Nuorten mielenterveys –kohorttitutkimus on
Tampereella ja Vantaalla suoritettava prospektiivinen kohorttitutkimus, joka sisältää mittauksen 9.
luokalla (T1) ja 2-vuotisseurantamittauksen. Tämän tutkimuksen tutkimusaineiston muodostivat
14 - 16 –vuotiaat nuoret Kouluterveyskyselyssä (N 50 569, joista 50.1% poikia) ja 15–17 –
vuotiaat nuoret Nuorten mielenterveys –kohorttitutkimuksessa (N 3 278, joista 50.9% poikia T1mittauksessa; N 2 070, joista 43.6% poikia seurantamittauksessa). Aineisto kerättiin
koulukyselyinä, jotka perustuivat monissa aikaisemmissa tutkimuksissa käytettyihin masennuksen
ja epäsosiaalisen käyttäytymisen itsearviointimittareihin.
Poikkileikkaustutkimuksen mukaan nuorten vähäisenä koettu sosiaalinen tuki oli yhteydessä sekä
itseilmoitettujen masennusoireiden että epäsosiaalisen käyttäytymisen esiintyvyyteen. Tulokset
korostavat perheeltä saatavan tuen merkitystä nuoruusiän masennusoireiden ja epäsosiaalisuuden
ennaltaehkäisyssä. Itseilmoitettujen masennusoireiden ja epäsosiaalisen käyttäytymisen
samanaikaissairastavuus oli merkittävää molemmilla sukupuolilla. Poikkileikkausasetelmassa
epäsosiaalisesti käyttäytyvät nuoret olivat jopa 7-kertaisessa riskissä kärsiä masennusoireista eiepäsosiaalisesti käyttäytyviin nuoriin verrattuna. Vaikka koettu sosiaalinen tuki modifioi
itseilmoitettujen masennusoireiden ja epäsosiaalisen käyttäytymisen samanaikaissairastavuutta,
säilyi havaittu yhteys häiriöiden välillä. Masennusoireiden riski vaihteli rikollisen käyttäytymisen
muotojen mukaan sekä kasvoi rikosmäärien lisääntyessä. Tytöillä masennusoireiden riski oli suurin
toistuvasti vandalismiin syyllistyneillä, pojilla suurin riski oli toistuvasti väkivaltatekoja tehneillä.
13
Masennusoireista kärsivien nuorten rikollinen käyttäytyminen oli monimuotoisempaa kuin eimasentuneilla nuorilla, jotka pääsääntöisesti erikoistuivat yhteen rikostyyppiin.
Kahden vuoden seurannassa sekä itseilmoitetut masennusoireet että epäsosiaalinen käyttäytyminen
olivat pysyviä. Häiriöiden jatkuvuudessa ei ollut tilastollisesti merkitseviä sukupuolieroja.
Masennusoireet ennustivat epäsosiaalista käyttäytymistä seurannassa tytöillä. Pojilla aikaisemmat
masennusoireet vähensivät riskiä tulevaan epäsosiaaliseen käyttäytymiseen. Seurannassa
epäsosiaalinen käyttäytyminen ei ennustanut tulevaa masennusoireilua kummallakaan sukupuolella,
kun aikaisemmat masennusoireet oli kontrolloitu.
Tulokset osoittavat, että sekä masennusoireet että epäsosiaalinen käyttäytyminen ovat pysyviä
mielenterveyshäiriöitä keskinuoruudessa ja niiden samanaikaissairastavuus on huomattavan yleistä.
Itseilmoitetut masennusoireet vaikuttavat rikollisen käyttäytymisen muotoihin. Erityistä huomiota
tulisi jatkossa kiinnittää niihin nuoriin, joiden epäsosiaalinen käyttäytyminen on toistuvaa, sillä
heidän masennusriskinsä on erityisen suuri. Muita mahdollisia nuoruusiän masennusoireiden
varomerkkejä ovat toistuva, monimuotoinen rikollinen käyttäytyminen sekä vandalismiin ja
väkivaltaan erikoistuminen. Masennusoireista kärsivät tytöt ovat alttiita tulevalle epäsosiaaliselle
käyttäytymiselle.
14
1. INTRODUCTION
Depression and antisocial behaviour are among the most common mental health disorders in
adolescents. It has been estimated that in general population one third of adolescents have
experienced depression by the age of 19 (Lewinsohn et al. 1998). Similarly, although a certain
degree of norm-violating or delinquent behaviour can be seen as a part of normal adolescent
development, recurring and severe antisocial behaviour is not normative (Steinberg and Morris
2001, Stouthamer-Loeber and Loeber 2002). Estimates of severe antisocial behaviour have ranged
from 5% to 10% in general child and adolescent population (Angold and Costello 2001). Antisocial
behaviour in children/adolescents has increased remarkably in all developed countries in the past
decades (Smith 1995, Costello et al. 2006). Similar conclusions have been reported for increased
rates of adolescent depression (Fombonne 1998, Kessler and Walters 1998) or at least the
identification of adolescent depression and preparedness in treatment seeking has improved (Parker
and Roy 2001).
Depression and antisocial behaviour are accompanied by a range of psychosocial problems which
may disturb adolescent development and psychosocial adjustment (Birmaher et al. 1996, Lewinsohn
et al.1998, Rutter et al. 1998, McMahon and Frick 2005). In addition, depression and antisocial
behaviour are among of the most common reasons for adolescents being referred for psychiatric
evaluation and treatment, yet many of these adolescents are still under-treated (Ritakallio et al.
2003, Sihvola et al. 2007). Comorbidity is very common in both disorders (Angold et al. 1999,
Weller and Weller 2000). Costs and burden to society in adolescence and adulthood are substantial
and affect multiple groups (Scott et al. 2001, Lynch and Clarke 2006). Thus, depression and
antisocial behaviour can be considered a major public health problem.
Reportedly, depression and antisocial behaviour disorders have a fairly stable course in adolescence
(Rutter et al. 2006). However, findings of possible sex differences in the continuity of these
disorders have been mixed. In addition, most of the existing studies on antisocial behaviour have
been conducted only among males. Although, comorbidity between depression and antisocial
behaviour is well established (e.g. Angold et al. 1999), detailed information on the interrelations
between these disorders is still needed. Specially, there are only few studies on the longitudinal
associations between depression and antisocial behaviour, especially those conducted in general
populations. Research on comorbidity between depression and antisocial behaviour and on
15
developmental associations is needed because it has implications for classification systems, etiology
and treatment (Keiley et al. 2003).
2. REVIEW OF THE LITERATURE
2.1 Adolescent development
Adolescence is a challenging developmental stage which begins at puberty around age 11 for
females and 13 for males. Adolescence ends around 22 years when a person has become physically,
psychologically and sexually mature, has achieved emotional and financial autonomy and
independence from parents, and an adult personal identity has developed. In this period a person’s
biological, cognitive, psychological and social characteristics are changing rapidly as they become
more adult-like. The essential challenge in adolescent development is to adapt to these changes
(Aalberg and Siimes 1999, p. 55, Gutgesell and Payne 2004, Christie and Viner 2005). In contrast
to popular belief, viewing adolescence as a extremely troublesome period, most recent studies have
come to the conclusion that only about 20 - 25% of adolescents display extreme psychological and
social disturbances as the majority of adolescents navigate through adolescence without major
disturbances (Arnett 1999, Steinberg and Morris 2001, Gutgesell and Payne 2004).
Adolescence can be divided into three periods: early (around 11 to 14 years), middle (around 15 to
17 years) and late adolescence (around 18 to 21 years). Each period is characterised by a certain
developmental task leading from one period to another, although individual variations and
differences between sexes in the progression through these periods may be considerable. In
addition, occasional regression (i.e. child-like characteristics and behaviours) are very common, and
in fact, they are deemed essential to successful adolescent development (Aalberg and Siimens 1999,
p. 55, Gutgesell and Payne 2004). The key developmental tasks of adolescent development are 1)
the achievement of biological and sexual maturation, 2) the development of personal identity, 3) the
development of intimate sexual relationships, 4) the establishment of independence and autonomy
(Christie and Viner 2005).
16
Early adolescence entails rapid biological changes and reassessment of body image. Difficulties in
impulse control, behavioural changes (e.g. fatigue, irritability and easy embarrassment), and rapid
changes in mood, self-concept and interests as well as increased conflicts with parents are common
at this stage. Early moral concepts, early sexual orientation and emotional separation from parents
begin. The significance of and attachment to peers gradually become more important as closeness
with parents and the amount of time spent with them diminishes. In middle adolescence biological
changes are usually complete for females, whereas males are still maturing during this period.
Abstract thinking and an understanding of law and morality proceed, likewise the development of
personal and sexual identity. Intimate and romantic relationships come into the picture. In late
adolescence the capability for complex abstract thought is acquised, and identity is differentiated
and organised. Equal intimate relationships are gradually achieved. This period is characterised by
further development of emotional, social and financial independence from parents as decisions
about professional and educational goals, leaving home and starting a family become topical. The
parent-adolescent relationship becomes more egalitarian and less contentious. Although the
significance of the family as a source of social support decreases during adolescent development,
the social support from parents still seems to have the most efficient effect in buffering adolescents
against internalizing and externalizing problems. Friendships have become to be intimate,
supportive and close (Garnefski and Diekstra 1996, Aalberg and Siimes 1999, p. 55, Marttunen and
Rantanen 1999, Garnefski 2000, Bru et al. 2001, Steinberg and Morris 2001, Gutgesell and Payne
2004, Christie and Viner 2005). During adolescent development a person’s ability to consider the
long-term consequences of his/her actions, and his/her awareness of the responsibility and of the
effects of his/hers actions on other people increases (Rutter et al. 1998, p. 28).
2.2 Depression in adolescence
2.2.1 Definition of depression
Depression is often used in casual language to refer to lowered mood or depressive feelings as a
reaction to loss or adverse life-events. These temporary “ups and downs”are common to
everybody. The most important distinction between depressive feelings and depression as a disorder
is that depression is characterized by core persistent and pervasive depressed mood and anhedonia
17
associated with significant psychosocial functional impairment (Zalsman et al. 2006). However, in
the earlier literature, the term depression has been used rather inconsistently as some authors (e.g.
Oldehinkel et al. 1999) use depression to refer to solely clinically diagnosed major depression
disorder (MDD) based on diagnostic criteria, whereas in other studies (e.g. Glied and Pine 2002)
self-reported depressive symptoms based on high rating scale scores are also referred to as
depression. The major difference between these perspectives is that disorders based on diagnostic
criteria “are based on specific, operationalized criteria with a certain number of clearly described
symptoms required, occurring with sufficient and specified frequency, intensity, or duration, or a
combination of these”(King et al. 2005, p. 3063). Thus, diagnostic classifications (e.g. DSM-IV,
APA 1994 and ICD-10, STAKES1997) represent a categorical approach which implies that
disorder is either present and absent. This approach works best when there are clear boundaries
between diagnostic classes and when classes are mutually exclusive (APA 2000, p. xxxi, Broberg et
al. 2001, Hankin 2006).
Self-reported depression is based on subjects’own perception and evaluation of depressive
symptoms based on a selected symptom checklist. Depression is seen as a dimensional continuum
in which depressive symptoms vary from non-existent to severe, and there is no sharp boundary
between non-depressed and depressed adolescents. Adolescents scoring above a specified cut-off
point are considered to be in need of clinical attention, and these symptoms are likely to reach
clinical levels over time (Gotlib et al. 1995, Hankin 2006). In fact, several studies (e.g. Kessler and
Walters 1998, Oldehinkel et al. 1999, González-Tejera et al. 2005, Sihvola et al. 2007) have shown
that a substantial proportion of adolescents with symptoms of depression accompanied by
psychosocial functional impairment fail to meet the diagnostic criteria for MDD. These adolescents
nevertheless manifest significant impairment and rates of comorbidity and they use a lot of mental
health services (Gotlib et al. 1995, Lewinsohn et al. 1998, González-Tejera et al. 2005, Rutter et al.
2006). Therefore it is important also to study depression from a dimensional perspective.
According to the DSM-IV (APA 1994, p. 339) MDD is characterized by one or more major
depressive episodes (MDE) without a history of manic, hypomanic or mixed episodes. The
diagnosis of MDD requires a two-week period of either depressed mood or loss of interest/pleasure
and, in addition at least 4 of the following symptoms: significant weight changes or appetite
disturbance, insomnia/hypersomnia, psychomotor agitation or retardation, fatigue or loss of energy,
feelings of worthlessness or guilt, impaired concentration or thinking ability and thoughts of death
or suicidality (Table 1). The diagnostic criteria for MDD are presented only according to the DSM
18
diagnostic classification because it is used more in epidemiological studies than the ICD
classification.
Table 1. DSM-IV criteria for Major Depressive Episode (APA 1994, 327).
A. Five (or more) of the following symptoms have been present during the same 2-week period and represent a change
from previous functioning: at least one of the symptoms is either (1) depressed mood or (2) loss of interest or pleasure.
Note: Do not include symptoms that are clearly due to a general medical condition, or mood-incongruent delusions or
hallucinations.
(1) depressed mood most of the day, nearly every day, as indicated by either subjective report (e.g. feels sad or empty)
or observation made by others (e.g. appears tearful). Note: In children and adolescents, can be irritable mood.
(2) markedly diminished interest or pleasure in all, or almost all, activities most of the day, nearly every day (as
indicated by either subjective account or observation made by others)
(3) significant weight loss when not dieting or weight gain (e.g., a change of more than 5% of body weight in a month),
or decrease or increase in appetite nearly every day. Note: In children, consider failure to make expected weight gains.
(4) insomnia or hypersomnia nearly every day
(5) psychomotor agitation or retardation nearly every day (observable by others, not merely subjective feelings of
restlessness or being slowed down)
(6) fatigue or loss of energy nearly every day
(7) feelings of worthlessness or excessive or inappropriate guilt (which may be delusional) nearly every day (not merely
self-reproach or guilt about being sick)
(8) diminished ability to think or concentrate, or indecisiveness, nearly every day (either by subjective account or as
observed by others)
(9) recurrent thoughts of death (not just fear of dying), recurrent suicidal ideation without a specific plan, or a suicide
attempt or a specific plan for committing suicide
B. The symptoms do not meet criteria for a Mixed Episode.
C. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of
functioning.
D. The symptoms are not due to the direct physiological effects of a substance (e.g., a drug of abuse, a medication) or a
general medical condition (e.g., hypothyroidism)
E. The symptoms are not better accounted for the Bereavement, i.e., after the loss of a loved one, the symptoms persist
for longer than 2 months or are characterized by remarked functional impairment, morbid preoccupation with
worthlessness, suicidal ideation, psychotic symptoms, or psychomotor retardation.
19
2.2.2 Features of adolescent depression
The clinical features of depression are basically very similar among children, adolescents and
adults, although the symptom profile varies slightly according to age (Marttunen and Pelkonen
1998, Masi et al. 1998, APA 2000, p. 353, Kessler et al. 2001). Changes in the symptom profile of
depression seem to be timed around puberty (Hankin 2006). According to DSM-IV there are two
exceptions between symptom profiles with age: among children and adolescents mood may be
irritable instead of depressed and failure to make expected weight gains may be considered along
with weight loss (APA 1994, p. 327). The most typical symptoms of adolescent depression are
depressed mood, loss of interest or pleasure, concentration or thinking difficulties and sleep
disturbances (Roberts et al. 1995, Lewinsohn et al. 1998, Marttunen and Pelkonen 1998, Haarasilta
et al. 2001, Sihvola et al. 2007). Thoughts of death/suicide have been found to be among the least
frequent symptoms among depressed adolescents, at least in community samples (Roberts et al.
1995, Lewinsohn et al. 1998, Haarasilta et al. 2001).
Differences in the symptom profiles of depression between clinical and non-clinical adolescent
samples have been little studied (Roberts et al. 1995). The existing data suggests that in general,
symptoms of depression are fairly similar in both samples (Roberts et al. 1995, Marttunen and
Pelkonen 1998) but there is some evidence that e.g. weight/appetite disturbances and sleep
abnormalities tend to be more prevalent in community samples (Lewinsohn et al. 1998). Suicidal
tendencies have been found to be less prevalent among depressed adolescents in community
samples than among clinically referred counterparts (Roberts et al. 1995, Lewinsohn et al. 1998).
Predictably, referred adolescents are likely to be overall more severe depressed than depressed
adolescents in the community (Hankin and Abramson 1999). Also, co-occurrence with other mental
disorders (comorbidity) is consistently more frequent in clinical than community samples (e.g.
Lewinsohn et al. 1995a, Hankin and Abramson 1999), but this is likely be a consequence of
overpresentation of comorbid adolescents in clinical samples than a true difference in the symptom
profiles given that comorbid adolescents are more likely to receive treatment than non-comorbid
adolescents (Angold et al. 1999).
Some gender differences in symptom profile and severity of adolescent depression have been
reported by some authors, but the results are inconsistent (Roberts et al. 1995, Hankin et al. 1998,
Lewinsohn et al. 2003). Weight/appetite disturbance, feelings of worthlessness/guilt, self-blame,
20
negative body image, concentration problems, fatigue, suicidality and health worries may be more
frequent among depressed females than males (Lewinsohn et al. 1998, Liss et al. 2001, Yorbik et al.
2004, Bennett et al. 2005). Males may express more disturbances in school work and anhedonia
than females (Lewinsohn et al. 2003, Bennett et al. 2005).
2.2.3 Identifying depression
Depression can be measured through diagnostic interviews and rating scales in which adolescent,
parents and teachers can be used as informants (AACAP 1998). In fact, many instruments do have
parallel versions for different informants (e.g. Shaffer et al. 1996, Achenbach 2000). Multifaceted
information is useful to capture a different profile of symptoms which may be observed in different
settings (home vs. school) (AACAP 1998). On the other hand, adults may be unaware of an
adolescent’s depression (Cantwell et al. 1997) resulting in a low agreement between informants
(Sanford et al. 1995, Verhulst et al. 1997, Sourander et al. 1999). In general, adolescents are
considered to be valid informants of their own depression (Cantwell et al. 1997, Lewinsohn et al.
1998, Hankin and Abramson 1999).
Diagnostic interviews can be divided by the level of structure or standardization into unstructured,
semi-structured and highly structured interviews; of these semi-structured and highly structured
diagnostic interviews are mostly used in research settings (Reitman et al. 1998, Roberts et al. 1998,
McClellan and Werry 2000). As semi-structured interviews (or interviewer-based interviews)
provide only broad guidelines for items to be elicited and allow interviewers to use their own probe
questions, they therefore rely on the interviewer’s clinical expertise (Reitman et al. 1998, McClellan
and Werry 2000). The limitations are time-consuming and cost (King et al. 2005). K-SADS (The
Schedule for Affective Disorders and Schizophrenia for School-Aged Children; Kaufman et al.
1997) and CAPA (The Child and Adolescent Psychiatric Assessment; Angold et al. 1995) are
examples of semi-structured interviews. The advantage of highly structured diagnostic interviews
(or respondent-based interviews) is superior diagnostic reliability as they minimize the role of the
interviewer using predetermined standardized questions in a rigid sequence presented precisely the
same way to all participants (McClellan and Werry 2000, King et al. 2005). Due to a specified
protocol, highly structured interviews can be administered by trained lay interviewers (Shaffer et al.
1996, Reitman et al. 1998). DISC (Diagnostic Interview Schedule for Children; Shaffer et al. 1996)
21
is an example of highly structured interviews. The limitation of interviews is that they do not
usually contain normative data as most rating scales do (Reitman et al. 1998). Further, diagnostic
interviews do not provide exact guidelines on how to determine the clinically significant
impairment required in diagnosis (Kessler et al. 2001).
Rating scales can be completed by informants as questionnaires or they can be presented as
interview questions (AACAP 1998). Using rating scales enables comparisons between an
adolescent’s level of depression and a reference group of the same age and gender (e.g. Achenbach
1991).The major limitation of the rating scales is that they cannot be used as diagnostic instruments
as they rely on person’s own perception and evaluation of depressive symptoms. In fact, rating
scales have been criticized for poor specificity (Kessler et al. 2001). However, some recent studies
have reported self-reported depressive symptoms to be approximately equivalent to clinically
diagnosed disorders (e.g. Boyle et al. 1997, Morgan and Cauce 1999, Lasa et al. 2000). For
example, Morgan and Cauce (1999) reported that Youth Self-Report (YSR; Achenbach 1991)
predicted a diagnosis of depression determined by the DISC with a specificity of 90%. In addition,
self-reported depressive symptoms have been reported to be persistent and to be associated with
significant psychosocial impairment both in adolescence and in adulthood (Ferdinand et al. 1995,
Gotlib et al. 1995, Glied and Pine 2002, Twenge and Nolen-Hoeksema 2002). In fact, there is
evidence that adolescents who scored high on a depression questionnaire did not differ significantly
on psychosocial dysfunction compared to adolescents meeting the diagnostic criteria for depression
(Gotlib et al. 1995). The most used rating scales in measuring adolescent depression are the
Children Depression Inventory (CDI), the Beck Depression Inventory (BDI), the Hamilton Rating
Scale for Depression (HRSD), the Reynolds Adolescent Depression Scale (RADS) and the Center
for Epidemiological Studies-Depression Scale (CES-D) (Myers and Winters 2002).
2.2.4 Epidemiology of adolescent depression
2.2.4.1 Methodological issues in epidemiological studies on adolescent depression
The prevalence of a disorder is defined as the proportion of the population with a disorder in a given
timeframe (e.g. current, 12 months or lifetime). Incidence is defined as the rate at which new cases
22
arise in a given period of time (Lewinsohn et al. 1998, Costello et al. 2006). Earlier epidemiological
studies have reported wide variation in the prevalence and incidence rates of adolescent depression.
Several factors contribute to the wide variability in rates including sample characteristics,
assessment instruments, diagnostic criteria and number of informants used, and whether an
additional criterion for functional impairment has been required (Roberts et al. 1998, Hankin and
Abramson 1999, Oldehinkel et al. 1999).
In community studies on adolescent depression sample size has ranged widely from a few hundred
to several thousand (Fergusson et al. 1993, Lewinsohn et al. 1993, Verhulst et al. 1997, Wittchen et
al. 1998, Haarasilta et al. 2001, Kaltiala-Heino et al. 2001, Millikan et al. 2002, Rushton et al. 2002,
Walker et al. 2005). In general, sample sizes are larger in rating scales based studies than in
interview based studies. Considering sample size is important, because due to very large sample
size minor differences without any clinical importance may reach a statistically significant level, but
on the other hand, in smaller samples “true”significant differences may not be discovered
statistically (Läärä and Lammi 1989, p. 224). Many studies (especially interview based studies)
have used random sampling (e.g. Wittchen et al. 1998, Boyd et al. 2000, Romano et al. 2001),
which increases the representativeness of the samples. Birth cohort studies following unselected
samples of children from birth to adulthood (e.g. Fergusson et al. 1993, Newman et al. 1996) are
unique in their representativeness. The age and gender distributions of the sample are significant
factors for generalizability (Roberts et al. 1998) given that the occurrence of depression changes
dramatically with adolescent development (Marttunen and Pelkonen 1998, Twenge and NolenHoeksema 2002, Angold and Costello 2006), and therefore a large age range determines a more
comprehensive picture of the development of adolescent depression than does some specific age.
On the other hand, estimates from prepubertal and postpubertal samples are not comparable. The
sex ratio of the sample is pivotal because depression is far more prevalent among adolescent
females than males (e.g. Newman et al. 1996, Oldehinkel et al. 1999). Regarding school-based
samples, estimates of depression may be underestimated since school non-attenders are omitted
from the estimates.
In general, self-report rating scales tend to yield higher estimates of depression than diagnostic
interviews (Kessler et al. 2001) although different timeframes hamper comparisons greatly.
Estimates of depression vary widely according to different informants, but in general, at least
among older adolescents, the prevalence rate is higher according to adolescent than parent report
(Verhulst et al. 1997, Romano et al. 2001). As there is uncertainty as to how best to combine
23
information from multiple informants, it is recommended to report rates separately by informant
(Roberts et al. 1998, Kessler et al. 2001).
Lastly, incompatibility of results presentation across studies hampers comparability of studies
(Roberts et al. 1998). Studies have used different timeframes of prevalence varying between the
present to lifetime. Unfortunately, the results are sometimes presented inadequately as the
timeframe for example is not clearly presented (e.g. Schraedley et al. 1999) or a number of
respondents is missing (e.g. Kessler and Walters 1998). Therefore, making a critical review of the
findings is rather difficult.
2.2.4.2 Prevalence and incidence of adolescent depression
Depression is one of the most common mental disorders among adolescents. In fact, it has been
estimated that in general population one third of adolescents have experienced depression by the
age 19 which corresponds to the estimates from adult populations (Lewinsohn et al. 1998). In
diagnostic interview based studies in general population the current prevalence of adolescent
depression has ranged from 0.5% to 3.4% (Fergusson et al. 1993, Lewinsohn et al. 1993), 3-month
prevalence has ranged from 0.5% to 4.4% (Simonoff et al. 1997, Costello et al. 2003, Gau et al.
2005), 6-month prevalence has ranged from 0.4% to 7.7% (Verhulst et al. 1997, Romano et al.
2001) and 12-month prevalence has ranged from 2.2% to 21.5% (Fergusson et al. 1993, Kessler and
Walters 1998, Wittchen et al. 1998, Oldenhinkel et al. 1999). The lifetime prevalence of adolescent
depression has varied from 5.4% to 24.8% in diagnostic interview based studies (Lewinsohn et al.
1993, Kessler and Walters 1998, Wittchen et al. 1998, Oldenhinkel et al. 1999). According to rating
scales in general population, estimates of prevalence of adolescent depression have ranged from
2.8% to 32.4% (Connelly et al. 1993, Ivarsson and Gillberg 1997, Wichstrom 1999, Boyd et al.
2000, Glied and Pine 2002, Millikan et al. 2002, Rushton et al. 2002, Walker et al. 2005, Maharajh
et al. 2006, Chabrol et al. in press).
Recent Finnish diagnostic interview based studies conducted in general population have yielded
corresponding rates of adolescent depression as estimates have ranged from 0.9% to 7.8 %
24
depending on timeframe (Aalto-Setälä et al. 2001, Haarasilta et al. 2001, Ristkari et al. 2006,
Sihvola et al. 2007). Similarly, in recent Finnish rating scale based studies estimates of adolescent
depression have ranged from 6% to 14% (Torikka et al. 2001, Kaltiala-Heino et al. 2003, Pelkonen
et al. 2003, Fröjd et al. 2006a, Luopa et al. 2006, Fröjd et al. 2007).
Depression is fairly rare in childhood, as reported prevalence rates have varied from 0.4% to 2.5%
(Birmaher et al. 1996). The prevalence of depression increases in both sexes after puberty
(Birmaher et al. 1996, Marttunen and Pelkonen 1998, Hankin 2006). For example, in a Dunedin
birth cohort 12-month prevalence of depression increased significantly over a 10-year period: 1.1%
at age 11, 2.1% at age 13, 2.8% at age 15, 16.8% at age 18 and 17.9% at age 21 (Hankin et al.
1998). It is not yet clear why the prevalence of depression increases during adolescent development
as diverse biological (e.g. changing levels of estrogen and testosterone), psychosocial (e.g. changes
in social roles) and cognitive factors (e.g. ruminative style) have been reported to be possible
explanations to the rise (AACAP 1998, Costello et al. 2002).
The incidence of adolescent depression increases around ages 13 and 14, and it is estimated to be at
its highest between the ages of 15 and 18 (Hankin et al. 1998, Oldehinkel et al. 1999). 12 to 20month incidence has been reported to range from 3.3% to 5.7% among 11 to 18-year-olds
(Lewinsohn et al. 1993, Garrison et al. 1997, Oldehinkel et al. 1999). The rate of new cases of
depression decreases in adulthood, indicating that the majority of adult depression can be explained
by a background of adolescent depression (Kim-Cohen et al. 2003, Rutter et al. 2006). For example,
Newman et al. (1996) reported that 70.2% of those who were depressed at age 21 had been
previously diagnosed at ages 11, 13, 15 or 18.
In childhood, rates of depression are approximately the same for both genders (AACAP 1998,
Angold and Costello 2006), in spite of some evidence that in childhood females’rates may be lower
than males’(Angold et al. 1998, Cyranowski et al. 2000, Twenge and Nolen-Hoeksema 2002).
After puberty the gender ratio changes dramatically, as numerous studies (e.g. Fergusson et al.
1993, Newman et al. 1996, Oldehinkel et al. 1999, Schraedley et al. 1999, Wichstrom 1999, Fröjd et
al. 2006a) have consistently shown that adolescent females are about twice as likely to be depressed
than males. The gender difference emerges first time in early adolescence around the ages of 12 and
15 (Twenge and Nolen-Hoeksema 2002, Angold and Costello 2006), but the greatest increase in
25
gender difference is estimated to emerge between ages 15 and 18 (Hankin et al. 1998). It seems that
female predominance in depression is explained explicitly by different incidence rates. However,
the underlying mechanisms for the predominance of depressed females are so far unclear, as it
seems that multiple processes may contribute to the development of this gender difference (Hankin
et al. 1998, Hankin and Abramson 1999, Angold and Costello 2006). Gender differences have been
explained by females’tendency to carry more risk factors for depression and to face more social
and biological challenges in adolescent development than males (Nolen-Hoeksema and Girgus
1994, Hankin and Abramson 1999). Pubertal status accompanied by biological, hormonal and
psychosocial factors has also been reported to explain the difference (Angold et al. 1998,
Cyranowski et al. 2000).
Some researchers (e.g. Birmeher et al. 1996a, Fombonne 1998, Kessler and Walters 1998) but not
all (e.g. Twenge and Nolen-Hoeksema 2002, Costello et al. 2006) have reported that the prevalence
of adolescent depression has increased, and the age at onset of depression has decreased in recent
birth cohorts. However, as Parker and Roy (2001) stated, this phenomenon can be at least partly
explained by destigmatization, improved identification of depression or increased willingness to
seek treatment. In Finland there seems to be no evidence of a vast increase in rates of depressive
symptoms among adolescents (Luopa et al. 2006), but there may have been some increase among
children in past decades (Sourander et al. 2004).
2.2.5 Etiology of adolescent depression
Depression is a multietiological disorder in which many genetic, biological/somatic, psychological,
psychosocial and sociodemographic factors usually interact in combination at the onset of
depression. Previous and/or comorbid psychopathology also increases the risk for depression
(Birmaher et al. 1996, AACAP 1998, Masi et al. 1998, Lagges and Dunn 2003). However, many
factors identified with onset of depression are in fact correlates or associations (Costello et al. 2002)
as they have been discovered by cross-sectional designs. Evidence suggests that accumulation of
risk factors has a cumulative effect on the risk of depression (Lewinsohn et al. 1998). In addition,
some of these factors are associated with both onset and maintenance of depression (Lewinsohn et
al. 1998, Parker and Roy 2001), and many of the etiological factors seem also to be non-specific to
26
depression as they also increase the risk of other psychiatric disorders (Birmaher et al. 1996,
Kessler et al. 2001).
Genetic factors
Several studies on adult samples have shown that depression is a familial and heritable disorder.
Although less studied among adolescents, heritability has been reported to result in a two- to
fourfold risk for depression among first-degree relatives. Genetic factors seem to account for at
least 50% of the variance of depression (Todd and Botteron 2002, Lagges and Dunn 2003, Zalsman
et al. 2006). However, it seems that genes interact with other predisposing factors such as
maladaptive parenting or weak parent-child bonding, which are common in families with parental
depression (Kessler et al. 2001, Parker and Roy 2001, Zalsman et al. 2006, Isometsä 2007).
Moreover, adolescents at high genetic risk appear to be more sensitive to the other etiological
factors (psychosocial stressors) than those at low genetic risk (Hankin 2006).
Biological/somatic factors
Biological factors of depression have mostly been studied among depressed adults. Given that many
of the neurobiological factors associated with the pathophysiology of adult depression are not fully
developed in adolescence, it is at least partly uncertain whether the results from adult samples can
be generalized to adolescent depression (Kaufman et al. 2001, Garber 2006, Hankin 2006). The
crucial biological factors of depression are reviewed below.
Neurochemical changes, that is dysregulation in the serotonergic, noradrenergic and dopaminergic
systems seem to play an important role in the etiology of depression. These neurotransmitters are
closely interrelated. Dysregulation in the serotonergic system has been studied most. Serotonin is a
regulator e.g. of mood, sleep and appetite. Several studies have reported that people with depression
have reduced serotonergic function, and lowered concentration of a metabolite of serotonin (5HIAA) has been found in the cerebrospinal fluid. Reduced serotonergic function is associated
especially with impulsive and aggressive suicidality. The serotogenic system seems to be at least
partly genetically regulated (Mervaala and Räsänen 2004, Isometsä 2007).
There also is evidence of functional and anatomic brain alterations in depression as e.g. metabolism
of prefrontal cortex is decreased among depressed individuals. Many hormonal abnormalities are
27
associated with depression; of these one of the most important is dysregulation in the hypothalamic
pituitary-adrenocortical (HPA) axis. In depression the function of the HPA axis is hyperactivated
caused by hypersecretion of corticotrophin releasing factor (CRF). As a result of dysregulation in
the HPA axis, high cortisol concentrations are reported among depressed individuals. Similarly,
thyroid dysfunction has been reported to be associated with depression as both hypothyroidism and
hyperthyroidism can result in depressive symptoms (Isometsä 2007).
Poor self-perceived health and chronic physical illnesses (e.g. epilepsy, migraine, asthma and
allergy) have also been reported to raise vulnerability to depression (Lewinsohn et al. 1998, Lagges
and Dunn 2003, Haarasilta et al. 2005, Karlsson et al. 2005).
Psychological factors
Among reported psychological factors with increased risk for depression are negative beliefs and
attributional styles, low self-esteem, self-consciousness, high self-criticism, hopelessness, emotional
reliance, pessimistic style and poor coping skills. These features are often referred to as a
depressogenic cognitive style (Birmaher et al. 1996, AACAP 1998, Lewinsohn et al. 1998, Masi et
al. 1998, Steinhausen and Winkler Metzke 2000, Parker and Roy 2001, Pelkonen et al. 2003,
Hankin 2006, Zalsman et al. 2006).
Psychosocial factors
Poor social skills and problems with social interaction and various school problems as psychosocial
stressors may expose adolescents to depression (Birmaher et al. 1996, AACAP 1998, Lewinsohn et
al. 1998, Masi et al. 1998, Pelkonen et al. 2003). Lack of social support from friends and peer
rejection increase vulnerability to depression (Steinhausen and Winkler Metzke 2000, KaltialaHeino et al. 2001, Pelkonen et al. 2003). Stressful life-events (e.g. parental divorce, loss events or
bullying) and overall stress seem also to increase vulnerability to depression. In fact, it has been
estimated that almost all depressed adolescents have experienced at least one significant negative
life event in the month before to the onset of depression (Lewinsohn et al. 1998, Kaltiala-Heino et
al. 1999a, Schraedley et al. 1999, Patton et al. 2003, Hankin 2006). Parental psychopathology,
especially depression, substance abuse and criminality have been consistently associated with
increased risk for adolescent depression (Birmaher et al. 1996, AACAP 1998, Beardslee et al. 1998,
Kessler et al. 2001, Zalsman et al. 2006). Frequent interfamily conflicts, broken family, problems
28
with family dynamics and communication, lack of family cohesion, decreased social support from
family, neglect and abuse including parental rejection are also associated with increased likelihood
of adolescent depression (Birmaher et al. 1996, Sheeber et al. 1997, AACAP 1998, Lewinsohn et al.
1998, Schraedley et al. 1999, Steinhausen and Winkler Metzke 2000, Kaltiala-Heino et al. 2001,
Parker and Roy 2001, Zalsman et al. 2006).
Psychopathology
Externalizing (e.g. bullying, aggression, delinquency) and internalizing symptoms (e.g. shyness,
feeling anxious, behavioural inhibition) are associated with increased risk for adolescent depression
(Lewinsohn et al. 1998, Kaltiala-Heino et al. 2000, Parker and Roy 2001, Pelkonen et al. 2003). A
previous episode of depression undoubtedly increases the risk for subsequent depression
(Lewinsohn et al. 1998, Birmaher et al. 2002, Rutter et al. 2006) likewise a history of suicide
attempt (Lewisohn et al. 1998), anxiety disorders (AACAP 1998, Lewinsohn et al. 1998) and
problematic substance use (Gilvarry 2000, Rey et al. 2002, Fröjd et al. 2006a) has been reported to
be associated with depression.
Sociodemographic factors
Some sociodemographic factors (e.g. age, gender, low SES, single-parent household, low parental
education or unemployment) have been found to be associated with increased risk for depression
(Birmaher et al. 1996, Schraedley et al. 1999, Kaltiala-Heino et al. 2001, Rushton et al. 2002).
2.2.6 Course and consequences of adolescent depression
The mean duration of a depressive episode in adolescence varies according to referral status, being
approximately 4 - 9 months in clinical samples, compared to 3 –7 months in community samples
(Birmaher et al. 2002). According to community studies, between 75 and 90% of depressed
adolescents recovered from depression in a follow-up of 6 –24 months (Birmaher et al. 2002), but a
small minority of depressed adolescents (5-10%) will suffer from chronic, severe depression for
several years (Birmaher et al. 1996, Kessler et al. 2001, Birmaher et al. 2002). Longer duration
seems to be associated with early age of onset, severity of depression, suicidality, poor psychosocial
29
functioning, school problems, comorbidity, somatic symptoms and exposure to negative life events,
interfamily conflicts and parental psychopathology (AACAP 1998, Birmaher et al. 2002, Rushton et
al. 2002, Zalsman et al. 2006). There seems to be no difference in duration of depression according
gender (Birmaher et al. 2002, Angold and Costello 2006).
Both clinical (e.g. Sanford et al. 1995, Weissman et al. 1999, Fombonne et al. 2001a) and
community studies (e.g. Hankin et al. 1998, Beyers and Loeber 2003, Costello et al. 2003, KimCohen et al. 2003) have consistently shown adolescent depression to be highly recurrent both in
later adolescence and in adulthood (Rutter et al. 2006). In later adolescence, approximately 40 –
50% of depressed adolescents experience a relapse during remission (AACAP 1998), and 20 –70%
of them will have at least one recurrent episode in 1 - 6 years (AACAP 1998, Birmaher et al. 2002,
Zalsman et al. 2006). For instance, Costello et al. (2003) found that adolescents with a history of
depression were four times more likely than those with no previous depression to have depression
in later adolescence. In adulthood about 40 –70% of depressed adolescents will experience
recurrent depression resulting a two to sevenfold risk of being depressed compared to nondepressed adolescents (Rutter et al. 2006). Risk of recurrence in adolescent depression is associated
with early age at onset, greater number of previous episodes, severity of previous depression,
psychosis, comorbidity, suicidal attempts, negative attributional style, exposure to stress, familial
depression, low SES and lack of compliance with treatment (AACAP 1998, Masi et al. 1998,
Birmaher et al. 2002, Karlsson et al. 2006a). Evidence regarding gender differences in recurrence
rate is inconsistent, as some studies (Ferdinand et al.1995, Hankin et al. 1998, Kovacs et al. 2003)
found no gender differences in recurrence of depression, but others (Oldehinkel et al. 1999, Costello
et al. 2003, Lewinsohn et al. 2003) found females to be more likely to have recurrences than males.
The stability or continuity described above represents homotypic continuity in which manifestations
of the disorder in question change relatively little over time so that diagnosis remains the same at
different assessments (Angold et al. 1999, Costello et al. 2003).
Given that adolescent depression occurs during an important developmental period it is not
surprising that adolescent depression is associated with many adverse psychosocial consequences in
multiple domains. Adolescent depression, especially recurrent depression, is associated with poor
psychosocial functioning (e.g. poor coping skills and academic performance, problems in social
relationships, negative attributional styles and lower level of self-esteem) and disturbances (e.g.
substance misuse, suicidality), comorbid disorders (e.g. antisocial behaviour and anxiety disorders),
increased physical problems, smoking, obesity and early pregnancy (Sanford et al. 1995, Birmaher
30
et al. 1996, Newman et al. 1996, Lewinsohn et al. 1998, Giaconia et al. 2001, Birmaher et al. 2002,
Pelkonen and Marttunen 2003, Karlsson et al. 2006a, Zalsman et al. 2006, Sihvola et al. 2007). In
adulthood, previously depressed adolescents have shown increased risk for suicidality, lower global
functioning, academic and work difficulties, impairment in parenthood, lower SES, lower
educational level and increased psychiatric and medical hospitalization (Rao et al. 1995, Weissman
et al. 1999, Birmaher et al. 2002). Besides recurrent depression, adolescent depression is also
associated with other adult psychiatric disorders such as anxiety disorders, bipolar disorders and
mania (Rao et al. 1995, Kim-Cohen et al. 2003).
2.2.7 Comorbidity in adolescent depression
Comorbidity is defined as the co-occurence two (or more) different disorders in a defined period of
time (e.g. concurrent, 6 months or lifetime comorbidity). Comorbidity can be divided into
homotypic comorbidity (comorbidity within one diagnostic grouping) and heterotypic comorbidity
(comorbidity between disorders from different diagnostic groupings) (Angold et al. 1999). Studying
comorbidity is indicated, as in general comorbid disorders may lead to more severe symptoms,
increased psychosocial impairment and increased suicidality as well as complicated treatment than
single disorders (Newman et al. 1996, Pelkonen et al. 1997, Lewinsohn et al. 1998, Loeber et al.
2000a, Karlsson et al. 2006b).
Comorbidity is so common among depressed adolescents that it can be considered the rule rather
than the exception (Weller and Weller 2000, Kessler et al. 2001, Parker and Roy 2001).
Approximately 40 –70% of depressed adolescents have at least one comorbid disorder, and 20 –
50% of them have at least two comorbid disorders. Among the most common comorbid disorders
are anxiety disorders (15 –80%), antisocial behaviour (10 –80%), ADHD (10 –60%) and
substance abuse (20 –30 %) (Birmaher et al. 1996, Masi et al. 1998, Angold et al. 1999). Depressed
females are more likely than males to have at least one comorbid anxiety disorder (Kessler et al.
2001), whereas comorbidity between antisocial behaviour and substance abuse is more common
among depressed males (Kessler et al. 2001, Kovacs et al. 2003, Karlsson et al. 2006b). In general,
the majority of the studies have found depression to be temporally secondary to other disorders.
However, comorbid substance abuse may be an exception, as depression has been found to precede
31
substance abuse by several years (Birmaher et al. 1996, Parker and Roy 2001). This finding may be
due to the longer time to reach diagnostic level in substance abuse/dependence than to manifest
depression (Angold et al. 1999).
Comorbidity between depression and antisocial behaviour will be discussed detailed in Chapter 2.4.
2.3 Antisocial behaviour in adolescence
2.3.1 Definition of antisocial behaviour
Antisocial behaviour is a form of behaviour which violates social norms and rules and/or the rights
of others (APA 2000, p. 93). Many adolescents commit some antisocial acts in the form of testing
limits, challenging rules and seeking excitement, but most of these acts are non-serious in nature,
will not persist over a long period and do not require psychiatric treatment. This kind of behaviour
may be seen as part of a developmental process in which prosocial behaviours and social rules are
adopted. However, when antisocial behaviour develops into a persistent pattern demonstrating a
maladaptive trajectory, psychiatric evaluation and interventions are indicated (AACAP 1997,
Loeber and Farrington 2000, Steinberg and Morris 2001, Stouthamer-Loeber and Loeber 2002,
Karnik et al. 2006). In the literature on antisocial behaviour the terms delinquency, antisocial
behaviour/conduct problems, conduct disorder and oppositional defiant disorder, disruptive
disorders and externalizing disorders have been used rather inconsistently. Although these terms are
often interrelated and the conditions they refer to occur within same adolescents, they are not
synonymous, as they all have their own meanings and perspectives (AACAP 1997, Hill 2002).
Delinquency
Delinquency is a legal term referring to criminal, officially unlawful behaviour. Thus, the behaviour
in question is viewed from a legal perspective (AACAP 1997, Vermeiren 2003). The term
delinquent (i.e. offender) is usually used to refer adolescents who have been in contact with the
police and have been sentenced/ incarcerated (Waddell et al. 1999, Bassarath 2001), but as Deptula
32
and Cohen (2004) noted, there is no consistent standard to measure delinquency, and therefore
delinquent behaviour is defined rather differently across studies. In addition to this, as the age of
criminal responsibility between countries differs significantly (e.g. in Finland 15 years vs. in United
States 7 years), comparisons across countries are complicated, as crimes are not compiled uniformly
in official statistics (Rutter et al. 1998, p. 26). Delinquent behaviours can be divided into status
offences (e.g. behaviours which are illegal only for children/adolescents such as alcohol
consumption) and criminal offences (e.g. behaviours which are against the law in any age group
(Pajer 1998, McMahon and Frick 2005).
Antisocial behaviour/conduct problems
Antisocial behaviour (conduct problems) can be considered as an umbrella term in respect to other
terms referring to antisocial behaviour, as it refers broadly to all acts that are considered
inappropriate because they contravene the norms and principles of society or harm others/society or
both (e.g. frequent lying, substance use/abuse, truancy, running away, disobedience, stealing, firesetting, violence or vandalism). Some antisocial behaviours may be also delinquent but not all
(AACAP 1997, Bassarath 2001, Hill 2002, Vermeiren 2003). In addition, although antisocial
behaviour often includes features similar to conduct and oppositional defiant disorders, not all
antisocial behaviour is automatically psychopathological nor does it require psychiatric treatment,
as do conduct and oppositional defiant disorders (AACAP 1997).
Conduct disorder
Conduct disorder (CD) is psychiatric, diagnostic term referring to a group of antisocial and
aggressive symptoms in which a repetitive and persistent pattern is a key symptom but in addition
to this the antisocial behaviour must cause clinically significant impairment in social, academic or
occupational functioning to fulfil the diagnostic criteria of conduct disorder (APA 1994, p. 90,
AACAP 1997, Waddell et al. 1999, Earls and Mezzacappa 2002). CD is described in DSM-IV as “a
repetitive and persistent pattern of behaviour in which the basic rights of others or major ageappropriate societal norms or rules are violated”(APA 1994, p.85). According to DSM-IV (APA
1994, p. 90) CD includes serious aggressive and nonaggressive behaviours, deceitfulness or theft
33
and serious rule violations (Table 2). Adolescents with conduct disorder are often also involved in
delinquency, but these terms are not synonymous as the criteria for conduct disorder include some
behaviours which are not against the law. In addition to this, many delinquents do not show the
social impairment and psychosocial dysfunction required for a diagnosis of conduct disorder. In
fact, it has recently been estimated that approximately 30-70% of delinquents do not fulfil the
diagnostic criteria for conduct disorder (APA 1994, p. 90, Rutter et al. 1998, p.1, Bassarath 2001,
Vermeiren 2003).
34
Table 2. DSM-IV criteria for Conduct Disorder (APA 1994, 90).
A. A repetitive and persistent pattern of behavior in which the basic rights of others or major age-appropriate societal
norms or rules are violated, as manifested by the presence of three (or more) of the following criteria in the past 12
months, with at least one criterion present in the past 6 months:
Aggression to people and animals
(1) often bullies, threatens, or intimidates others
(2) often initiates physical fights
(3) has used a weapon that can cause serious physical harm to others (e.g., a bat, brick, broken bottle, knife, gun)
(4) has been physically cruel to people
(5) has been physically cruel to animals
(6) has stolen while confronting a victim (e.g., mugging, purse snatching, extortion, armed robbery)
(7) has forced someone into sexual activity
Destruction of property
(8) has deliberately engaged in fire setting with the intention of causing serious damage
(9) has deliberately destroyed others’property (other than by fire setting)
Deceitfulness or theft
(10) has broken into someone else’s house, building or car
(11) often lies to obtain goods or favors or to avoid obligations (i.e., “cons”others)
(12) has stolen items of nontrivial value without confronting a victim (e.g., shoplifting, but without breaking and
entering; forgery)
Serious violations of the rules
(13) often stays out at night despite parental prohibitions, beginning before age 13 years
(14) has run away from home overnight at least twice while living in parental or parental surrogate home (or once
without returning for a lengthy period)
(15) is often truant from school, beginning before age 13 years
B. The disturbance in behavior causes clinically significant impairment in social, academic, or occupational
functioning.
C. If the individual is age 18 years or older, criteria are not met for Antisocial Personality Disorder.
Specify type based on age at onset:
Childhood-Onset Type: onset of at least one criterion characteristic of Conduct Disorder prior to age 10 years
Adolescent-Onset Type: absence of any criteria characteristic of Conduct Disorder prior to age 10 years
Specific severity:
Mild: few if any conduct problems in excess of those required to make the diagnosis and conduct problems cause only
minor harm to others
Moderate: number of conduct problems and effect on others intermediate between “mild”and “severe”
Severe: many conduct problems in excess of those required to make the diagnosis or conduct problems cause
considerable harm to others
35
Oppositional defiant disorder
Oppositional defiant disorder (ODD) is defined in DSM-IV (APA 1994, p. 91) as a “recurrent
pattern of negativistic, defiant, disobedient, and hostile behaviour toward authority figures that
persists for at least 6 months”and which causes clinically significant impairment in social,
academic or occupational functioning. ODD consists of often recurring behaviours such as
annoying other people, blaming others for own mistakes, refusing to comply with adult’s requests
or losing one’s temper (Table 3). Recurrent disobedient and hostile behaviour is aimed toward
authority figures (e.g. parents or teachers) but adolescents with ODD do not usually violate the
rights of others as do adolescents with CD (McMahon and Frick 2005). Therefore ODD is viewed
as a milder form than CD (APA 1994, p. 93, Earls and Mezzacappa 2002, McMahon and Frick
2005). The diagnostic criteria of CD and ODD are presented only according to the DSM diagnostic
classification, because this is more generally used in epidemiological studies than is the ICD
classification.
Table 3. DSM-IV criteria for Oppositional Defiant Disorder (APA 1994, 93).
A. A pattern of negativistic, hostile, and defiant behavior lasting at least 6 months, during which four (or more) of the
following are present:
(1) often loses temper
(2) often argues with adults
(3) often actively defies or refuses to comply with adults’requests or rules
(4) often deliberately annoys people
(5) often blames others for his or her mistakes or misbehavior
(6) is often touchy or easily annoyed by others
(7) is often angry and resentful
(8) is often spiteful or vindictive
Note: Consider a criterion met only if the behavior occurs more frequently than is typically observed in individuals of
comparable age and developmental level.
B. The disturbance in behavior causes clinically significant impairment in social, academic, or occupational
functioning.
C. The behaviors do not occur exclusively during the course of a Psychotic or Mood Disorder.
D. Criteria are not met for Conduct Disorder, and, if the individual is age 18 years or older, criteria are not met for
Antisocial Personality Disorder.
36
Disruptive and externalizing disorders
CD, ODD and attention deficit hyperactivity disorder (ADHD) constitute together a broader
category named attention-deficit and disruptive behaviour disorders in DSM-IV (APA 1994, p. 78).
In some studies (e.g. Achenbach 1991, Garnefski and Diekstra 1997) the term externalizing
disorders has been used to refer to both antisocial behaviour and substance misuse/dependence.
Externalizing behaviours are seen as a manifestation of the same underlying construct of
multiproblem behaviour sharing a range of common or similar risk factors (Donovan et al. 1988,
Garnefski and Diekstra 1997, Fergusson 1998). At the adult end of the serious antisocial behaviour
spectrum is antisocial personality disorder (APD) (Loeber et al. 2000a, Karnik et al. 2006).
Although APD cannot be diagnosed until age 18 (APA 1994, p. 649), some of the APD symptoms
(e.g. impulsivity and callousness) may already be present among antisocial adolescents. Some
adolescents with severe CD will continue to behave antisocially in adulthood (Loeber et al. 2000a,
Karnik et al. 2006).
2.3.2 Features of adolescent antisocial behaviour
Features of antisocial behaviour are very heterogeneous as the broad spectrum of antisocial
behaviours varies widely between adolescents in form and severity. Behaviours can range from
minor oppositional behaviours to more severe aggressive or non-aggressive forms of antisocial
behaviour (Maughan and Rutter 2001, McMahon and Frick 2005, Nock et al. 2006). Thus, besides
categorical diagnostic criteria (e.g. DSM-IV, ICD-10), antisocial behaviour can be seen as a
dimensional continuum of problem behaviour in which almost all adolescents show to a greater or
lesser degree (Rutter 2001). In the dimensional (or statistical) approach assessment is based on
quantification of features, and antisocial behaviour is seen as a statistical deviation from normal
behaviour. This approach is useful in describing a phenomenon with no clear boundaries and no
explicit core symptom (APA 2000, p. xxxi, Earls and Mezzacappa 2002, Hill 2002) such as
antisocial behaviour is. The categorical approach is necessary for clinical settings dividing mental
disorders into types which are based on criteria with defined features. This approach, however, is
problematic, as the cut-off-point is inevitably always arbitrary, and there is a risk that important
differences of severity or type of dysfunction below and above the cut-off will be lost. Both
37
orientations occur simultaneously in practice (APA 1994, p. xxii, Earls and Mezzacappa 2002, Hill
2002).
Specialization in delinquency refers to the tendency to commit the same type of crime repeatedly,
whereas versatility of delinquency refers to a criminal career characterised by a random mix of
offences (Kyvsgaard 2003, p. 147). Most antisocial adolescents, at least those with a severe
symptom pattern, engage in many forms of antisocial behaviour (Loeber and Coie 2001). Thus
specialization in one pattern of antisocial behaviour is rare (Gottfredson and Hirschi 1990, p. 91,
Loeber and Stouthamer-Loeber 1998, Lahey et al. 1999a). Antisocial behaviour is usually present in
different settings (e.g. at school, at home and in the community) (APA 2000, p. 94). Adolescents
with repeat antisocial behaviour have been described to be as unempathetic, callous, irritable,
reckless and lacking of feelings of guilt, misperceiving the intentions of others, more prone to
respond with aggression than others and having poor frustration tolerance (APA 2000, p. 95).
Antisocial adolescents often have a number of problems in psychosocial functioning (e.g. poor
performance at school, poor social relationships, substance abuse and increased risk for suicide)
which also impairs their prognosis (Loeber et al. 1998, Renaud et al. 1999, Maughan 2001, Burke et
al. 2002, Deptula and Cohen 2004, McMahon and Frick 2005).
The DSM-IV (APA 1994, p. 90) divides severity levels of symptoms of CD into three subtypes:
mild, moderate and severe (Table 2). Subtyping is based both on number of conduct problems and
on the magnitude of harm caused to others (APA 1994, p. 87), however, number of conduct
problems and the “harm to others”are not clearly specified. Besides severity of symptoms, the
distinction between overt and covert behaviour has been used in subtyping antisocial behaviour
(Stanger et al. 1997, Loeber and Stouthamer-Loeber 1998, Loeber et al. 2000a, Angold and Costello
2001). The overt dimension of antisocial behaviour is described as directly confrontational
behaviours such as bullying, fighting and violence. In contrast, the covert dimension of antisocial
behaviour consists of nonconfrontational behaviours which are, however, concealing or sneaky in
nature (e.g. lying, stealing, truancy and vandalism) (Loeber and Stouthamer-Loeber 1998,
McMahon and Frick 2005). These two types of problem behaviour intercorrelate significantly, and
many adolescents are deviant on both of these dimensions (Stanger et al. 1997, Loeber and Coie
2001).
38
Adolescents with antisocial behaviour often initiate aggressive behaviour (Loeber and StouthamerLoeber 1998, APA 2000, p. 94). Aggression can be divided into several subtypes, of which those
most often referred to are reactive and proactive aggression (Loeber et al. 2000a). In the reactive
subtype, aggression is viewed as a defensive reaction to a perceived threat or provocation and is
characterized by impulsiveness, anger and hostility. This subtype of aggression is common among
adolescents with ODD and those with an early onset of antisocial behaviour. Proactive aggression is
generally unprovoked and premeditated and it is used either for some personal gain or to
coerce/dominate others. Proactive aggression is common among adolescents with CD –especially
among those with delinquent behaviour (Dodge and Coie 1987, Lahey et al. 1999a, Loeber et al.
2000a, Karnik et al. 2006).
Adolescent males tend to engage in more aggressive and more serious forms of antisocial
behaviour, but in milder forms sex differences tend to diminish or even disappear (Junger-Tas
1994a, Rutter et al. 1998, p. 256, Tiet et al. 2001, Kivivuori 2002b, Rey et al. 2005). Regarding
offences, the male preponderance is greatest in sex and drug crimes, and crimes involving the use of
force against a victim or property (Rutter et al. 1998, p. 259). Antisocial behaviour of females tends
to be covert in nature while males’antisocial behaviour tends to be more overt. In addition to this,
females’aggression is often indirect and verbal (e.g. gossiping, backbiting, peer rejection) while
males’aggression is more externally directed and physical (e.g. fighting, destructiveness) (Kann
and Hanna 2000, Loeber et al. 2000a, Angold and Costello 2001, Bassarath 2001).
Features and patterns of antisocial behaviour usually change in the transition from childhood to
adolescence (Loeber and Stouthamer-Loeber 1998, Loeber et al. 2000a). Typically, less serious
forms of antisocial behaviour emerge first, although there are some individual differences (Loeber
and Stouthamer-Loeber 1998, APA 2000, p. 97, Loeber and Coie 2001). Oppositional behaviour
peaks in the early to middle childhood, and minor forms of aggression tend to peak in the early
school years (Loeber and Coie 2001). With age many features of ODD decrease and behaviours
typical of CD become more common during adolescent development (Junger-Tas 1994a, Simonoff
et al. 1997, Lahey et al. 2000, Angold and Costello 2001). In addition, it has been reported that
aggressive behaviours tend to decrease (with the exception of serious violence), while status
offences and non-aggressive, covert behaviours tend to increase with age (Stanger et al. 1997,
Loeber and Stouthamer-Loeber 1998, Lahey et al. 1999b, Maughan et al. 2004, Rey et al. 2005).
39
According to the official statistics the peak of crimes occurs around 17 to 18 years although
estimates vary internationally (Rutter et al. 1998, p. 35).
2.3.3 Identifying antisocial behaviour
Since antisocial behaviour is a complex, heterogeneous disorder associated with multiple
psychosocial problems, it is often recommended to assess antisocial behaviour from
multidimensional perspectives. In general, the use of multiple assessments methods, multiple
informants (adolescent, parents, teacher and other significant persons close to the adolescent) and
assessment in multiple settings (home, school and community) is recommended to obtain a
comprehensive view of an adolescent’s behaviour. The use of multiple informants is advisable
because different informants evaluate adolescents’behaviour from different perspectives and in
different reference groups. Moreover, adolescents tend to behave differently in different settings
(AACAP 1997, Reitman et al. 1998, Kann and Hanna 2000, McMahon and Frick 2005). However,
as in the case of depression (Chapter 2.2.4.1), there is no clear consensus on how to combine
information from different informants (Reitman et al. 1998, Roberts et al. 1998). Besides this,
adolescents are seen as important informants because their parents are not always aware of
adolescents’behaviour, the covert behaviour and antisocial behaviour of older adolescents
especially may be kept in the dark (Rutter et al. 1998, p.40, Lahey et al. 2000, Romano et al. 2001).
On the other hand, some antisocial adolescents tend to minimize their behaviour (AACAP 1997),
but especially if information is gathered anonymously, the reliability of assessment has been
deemed good (Kivivuori 2005a).
Antisocial behaviour can be assessed through diagnostic interviews, behaviour rating scales, crime
statistics and observations (Reitman et al. 1998, Rutter et al. 1998, Kann and Hanna 2000,
McMahon and Frick 2005). Diagnostic interviews can be divided by the level of structure or
standardization into unstructured, semi-structured and structured interviews; of these semiunstructured and structured diagnostic interviews are commonly used in research settings.
Diagnostic interviews on antisocial behaviour have similar advantages and are hampered by
limitations similar to those in interviews on depression (see Chapter 2.2.3).
40
Behaviour rating scales can be completed by adolescents, parent(s) and teachers and/or other
significant persons close to the adolescent or they can be formulated into interview questions. They
are regarded as a useful method in screening for a broad range of antisocial behaviours as they
usually consist of a standard list of specified antisocial behaviours reflecting the number of diverse
acts the respondent has committed and/or the frequency of antisocial behaviour in a specified period
(Reitman et al. 1998, Bendixen et al. 2003, Kivivuori 2005a). Rating scales usually assess common
and victimless forms of antisocial behaviour, because it is not meaningful to inquire about
extremely serious acts as a respondent has greater reason to deny them (Rutter et al. 1998, p. 40,
Kivivuori 2005a). The use of rating scales enables comparisons between an adolescent’s level of
antisocial behaviour and a reference group of the same age and gender. On the other hand, rating
scales are usually seen as unequal to diagnostic interviews (Kann and Hanna 2000, McMahon and
Frick 2005, Mojtabai 2006) although there is some evidence that the differences in reliability and
validity are small between these instruments (Boyle et al. 1997). In addition, although there are
known limitations to the reliability of self-reported antisocial behaviour (e.g. dishonesty,
forgetfulness and exaggeration), they are generally considered reliable and valid methods in
adolescence (Junger-Tas 1994b, Kivivuori 2005a). The most used rating scales of antisocial
behaviour include YSR (the Youth Self-Report), CBCL (the Child Behavior Checklist), the Eyberg
Child Behavior Inventory, SRD (the Self-Report Delinquency Scale), SDQ (the Strengths and
Difficulties Questionnaire) (Reitman et al. 1998, Goodman 2001, McMahon and Frick 2005).
Crime statistics can also be used in assessing antisocial behaviour, but their disadvantage is that
only a minority of delinquent adolescents is caught by the police and thus the majority does not
appear in official crime statistics (Moffitt 1993, Rutter et al. 1998, p. 35, Kivivuori 2005a). In a
Finnish study Kivivuori (2002a) found that only 11% of adolescents reporting shoplifting or
violence were caught by the police. Of course, there are differences between types of offence as
lethal crimes etc. serious crimes are most likely to be included in official crime statistics (Kivivuori
2005a). In addition, not all crimes are included in official statistics if committed by adolescents
under the age of criminal responsibility. Moreover, international comparisons and interpretations of
official statistics are difficult because the age of criminal responsibilities and legal categories of
crimes between countries differ significantly (Junger-Tas 1994b, Rutter et al. 1998, p.26).
Behavioural observations can be conducted either in an adolescent’s natural setting (e.g. school,
home, peer group) or in clinics or laboratories. Observations can provide useful information about
an adolescent’s behaviour which is not filtered through the perceptions of an informant, but on the
41
other hand they are extremely time-consuming and expensive. In addition, covert behaviours (e.g.
theft) are quite impossible to identify reliably through observation. Among the most used
observation paradigms are the Parent Daily Report and the Direct Observation Form (DOF)
(Reitman et al. 1998, McMahon and Frick 2005).
2.3.4 Epidemiology of adolescent antisocial behaviour
2.3.4.1 Methodological issues in epidemiological studies on adolescent antisocial behaviour
Epidemiological studies have previously reported widely varying prevalence rates of antisocial
behaviour. Factors similar to those in the case of depression (see Chapter 2.2.4.1) contribute to the
wide variability in rates (Roberts et al. 1998).
In community studies on antisocial behaviour sample size has ranged widely from a few hundreds
to several thousands (Fergusson et al. 1993, Lewinsohn et al. 1993, Wichstrom et al. 1996, Kratzer
and Hodgins 1997, Verhulst et al. 1997, Weist et al. 1998, Ford et al. 2003, Heyerdahl et al. 2004,
Maughan et al. 2004, Little and Garber 2005). In general, sample sizes are larger in studies based on
rating scales than in interview based studies. Many studies (especially interview based studies) have
used random sampling (e.g. Lewinsohn et al. 1993, Romano et al. 2001, Gau et al. 2005, Rey et al.
2005), which increases the representativeness of the samples. As in the case of depression (see
Chapter 2.2.4.1), age distribution is a significant factor for representatives (Roberts et al. 1998)
because the occurrence and features of antisocial behaviour change with adolescent development
(Junger-Tas 1994a, Angold and Costello 2001). Similarly, it is essential to consider the gender
distribution of the sample. Although a substantial number of studies (e.g. Lahey et al. 1999a,
Pedersen et al. 2001, Tiet et al. 2001) have demonstrated that antisocial behaviour is more prevalent
among males than females, it is important to consider whether different symptom presentations
between sexes might as least partly explain the found differences (Angold and Costello 2001).
Given that e.g. DSM-IV based interviews are mostly based on items that are more common among
males (e.g. physical aggression), it is quite possible that the main items of females’antisocial
behaviour are not captured in the interview and are therefore not identified (Kann and Hanna 2000,
Loeber et al. 2000a). In addition, the gender ratio depends on the age group studied, as males tend
42
to predominate in childhood but as children grow up, the gap between males and females
diminishes (AACAP 1997, Maughan et al. 2004). Most studies so far have been conducted among
males (AACAP 1997, Keenan et al. 1999, Hill 2002). Regarding school-based samples, estimates of
antisocial behaviour may be underestimated since school non-attenders are omitted (Junger-Tas
1994b). Finally, there seems to be large cross-national and cultural variation in the prevalence and
the features of antisocial behaviour. For example, it has been reported that in the United States
crimes involving weapons are approximately 15 times more common than in Europe, and lethal
violence occurs in the context of another crime much more often in the United States than in Europe
(Rutter et al. 1998, p. 40, Savolainen et al. 2000). Most of the existing studies have been conducted
in the United States or in the UK.
In earlier studies, the most used data collection methods have been (semi)structured interview
schedules and self-report rating scales. Especially regarding rating scales, there is wide variation
between scales in the items used as they vary from minor to serious forms of antisocial behaviour.
Similarly, the number of items varies notably between scales. As a heterogeneous disorder, there is
no core group of features of antisocial behaviour, and therefore rates of antisocial behaviour depend
entirely upon which symptoms are included in the instrument (Angold and Costello 2001). It is
noteworthy that there have been major changes in DSM criteria for CD and ODD since 1980
(Angold and Costello 2001) which may produce significant changes in estimates (Loeber et al.
2000a). Based on different informants prevalence rates of antisocial behaviour may vary greatly
(e.g. Verhulst et al. 1997, Romano et al. 2001), but in general, at least among older adolescents, the
prevalence rate is higher according to the adolescent than parental report (Romano et al. 2001). As
there is uncertainty regarding how to best combine information from multiple informants, it is
recommended to report rates separately by informant (Roberts et al. 1998).
2.3.4.2 Prevalence and incidence of adolescent antisocial behaviour
Antisocial behaviour is one of the most common mental disorders among children and adolescents.
It has been estimated that in general population 5 to 10% of the children and adolescents (aged 8 to
16 years) have notable behaviour problems in the spectrum of antisocial behaviour (Angold and
Costello 2001). According to interview based studies in community based adolescent samples
43
lifetime prevalence of CD has ranged from 1.7% to 11.2% (Lewinsohn et al. 1993, Nock et al.
2006), 6-month prevalence from 0.6% to 6.0% (Verhulst et al. 1997, Romano et al. 2001), 3-month
prevalence from 1.6% to 9.0% (Simonoff et al. 1997, Costello et al. 2003, Gau et al. 2005) and
current prevalence rate from 0.3% to 3.3% (Lewinsohn et al. 1993, Ford et al. 2003). Dick et al.
(2005) reported that 15% of 14-year-old boys and 8% of the same aged girls had a lifetime
diagnosis of CD in a Finnish sample. Ristkari et al. (2006) reported in a Finnish sample that 2.6% of
18-year-old boys had an APD diagnosis according to their military clinical records.
According to rating scales estimates of antisocial behaviour in community samples have ranged
from 0.8% to 42.5% in 12-month prevalence (Wichstrom et al. 1996, Vermeiren et al. 2002a,
Heyerdahl et al. 2004) and from 5.3 to 18.2% in 6-month prevalence (Weist et al. 1998, Heyerdahl
et al. 2004, Little and Garber 2005, Lien et al. 2006, Mojtabai 2006). According to recent Finnish
studies estimates of antisocial behaviour have ranged between 12.9 and 29.4% among 14- to 16year-old adolescents (Laukkanen et al. 2002, Kaltiala-Heino et al. 2003). Frequent involvement in
bullying at school has been reported by 9% of the Finnish boys and 2% of the girls (Kaltiala-Heino
et al. 2000).
According to an international survey conducted in thirteen western countries 16% to 34% of
adolescents had committed at least one property offence during the past year, 16% to 35% of them
had used violence at least once and 11% to 26% of them had used drugs at least once (Junger-Tas
1994a). Corresponding rates were reported by a Finnish survey conducted among 15 to 16 year-olds
(Kivivuori 2005b). In 2007, 17% of Finnish adolescents aged 14 to 16 years reported that they had
frequently been involved in delinquency during the past year
(http://info.stakes.fi/kouluterveyskysely). Pelkonen et al. (2003) reported that 7.3% of 16-year-old
boys and 2.0% of same-aged girls had broken the law with consequences during the past year.
Reportedly, antisocial behaviour by children/adolescents has increased remarkably in developed
countries in recent decades (Smith 1995, Fombonne 1998, Costello et al. 2006). However, evidence
from recent Finnish epidemiological studies suggests no significant increase in the prevalence of
antisocial behaviour. In fact, results from the School Health Promotion Study 2007 indicate that in
the period 2002 –2007 frequent delinquency decreased from 21% to 17% among 14 to 16 year-old
44
adolescents (http://info.stakes.fi/kouluterveyskysely). Corresponding results have also been reported
by the National Research Institute of Legal Policy (Kivivuori 2005b).
In general, both in childhood and adolescence males are more likely than females to engage in
antisocial behaviour (Junger-Tas 1994a, Rutter et al. 1998, p. 254, Lahey et al. 1999b, Silverthorn
and Frick 1999). More precisely, males are two to five times more likely to merit a diagnosis of CD
than females (Rutter et al. 1998, p. 263, Loeber et al. 2000a), but in other forms of antisocial
behaviour the sex ratio varies. This is mainly because symptom presentation of antisocial behaviour
differs between genders as discussed in Chapter 2.3.2. For instance, there seems to be very little
gender difference in rates of ODD in adolescence (Simonoff et al. 1997, Lahey et al. 2000, West et
al. 2003, Rowe et al. 2006), but regarding official crime statistics the male preponderance is at least
4:1 (Rutter et al. 1998, p. 259). In general, the more serious the offence, the greater is the difference
between genders (Junger-Tas 1994a), but the gender difference seems to narrow in adolescence as
more females engage in antisocial behaviour (Silverthorn and Frick 1999). Interestingly, the overall
gender ratio for antisocial behaviour has decreased notably in the past 40 years (Fombonne 1998,
Rutter et al. 1998, p.254, Keenan et al. 1999). The underlying mechanisms for the predominance of
antisocial males is still unclear, and further study e.g. on differences in the risk factors of antisocial
behaviour between genders is needed (Lahey et al. 1999b, Silverthorn and Frick 1999).
In general, antisocial behaviour increases with age (e.g. Simonoff et al. 1997, Garnefski 2000,
Gomes et al. 2003), but age trends differ for aggressive and non-aggressive behaviour (Lahey et al.
1999b) as discussed in Chapter 2.3.2. Most of the age-related increase is the result of worsening
behaviour in adolescents with early age at onset, and because of the increase in non-aggressive
behaviours (Lahey et al. 1999b). Delinquency peaks in middle adolescence after which it steadily
declines into young adulthood (Loeber et al. 1998, p. 75, Rutter et al. 1998, p.35, Overbeek et al.
2001) in spite of some variations between offences (Junger-Tas 1994a, Kivivuori 2005a).
Vandalism is most prevalent among 14 to 15 year-olds, property offences among 16 to 17 year-olds
and serious violence reaches its peak among 18 to 20 year-olds (Junger-Tas 1994a). Most studies
have found that male predominance in antisocial behaviour increases steadily with age, while in
females rates remain low until early adolescence (Silverthorn and Frick 1999, Maughan et al. 2004).
Given that features and patterns of antisocial behaviour change in the transition from childhood to
adolescence (Loeber and Stouthamer-Loeber 1998, Loeber et al. 2000a, Loeber and Coie 2001), the
45
age-at-onset of different types of antisocial behaviour also varies. Females usually start somewhat
later than males (Junger-Tas 1994a). In a community sample of males in the Pittsburgh Youth
Study minor to moderate antisocial behaviour (e.g. minor vandalism or theft) started at age 5 to 6
and accelerated between ages 10 and 12, whereas serious forms of antisocial behaviour (e.g. car
theft or serious violence) started at aged 6 to 7, and accelerated between ages 11 to 12 (Loeber et al.
1998, p.75). In a community sample of males and females Lahey et al. (1999a) found, however,
notably higher age-at-onset of the first conduct problems as mean age at onset varied between ages
10 and 11 depending on the informant. The age-at-onset for CD has varied widely between 4.5 and
11.6 years (Rowe et al. 2002, Nock et al. 2006).
2.3.5 Development, course and consequences of adolescent antisocial behaviour
2.3.5.1 Etiology of adolescent antisocial behaviour
Antisocial behaviour is a multietiological disorder in which many genetic, biological,
psychological, psychosocial, psychopathological and sociodemographic risk factors overlap and
interact (Rutter et al. 1998, Waddell et al. 1999, Burke et al. 2002, Ritakallio et al. 2003, McMahon
and Frick 2005). As most studies have mainly been based on males, it is doubtful whether these
findings can be applied to the females, given possible differences in patterns of risk factors (Lewin
et al. 1999, Aalsma and Lapsley 2001, Storvoll and Wichstrom 2002) and differences in the clinical
features between sexes (Kann and Hanna 2000, Bassarath 2001). In addition to this, there is some
evidence that different risk factors may be associated with early onset (individual and family
factors) and later onset (psychosocial and environmental factors) of antisocial behaviour (Moffitt
1993, Lahey et al. 1998, McCabe et al. 2001, Moffitt and Caspi 2001, Taylor et al. 2002). Similarly,
there may be differences in the etiology of different types of antisocial behaviour (McMahon and
Frick 2005). Finally, most risk factors/correlates seem to be non-specific to antisocial behaviour
(Angold and Costello 2001, Maughan 2001).
46
Genetic factors
Several studies have reported that antisocial behaviour is a familial disorder as it has been shown to
aggregate in families, paternal antisocial behaviour especially seems to predict offspring’s antisocial
behaviour. Molecular genetic studies on antisocial behaviour are sparse, and the results are only
preliminary. According to twin and adoption studies there may be some evidence of genetic liability
for antisocial behaviour (although the results are not consistent), but because of difficulties in
differentiating between genetic factors and environmental risk factors associated with familial
antisocial behaviour solid conclusions are difficult to draw. In fact, at present it seems that genetic
and other risk factors combined are most predictive of development of antisocial behaviour. Some
evidence exists that genetic factors may be more predictive for antisocial behaviour persisting into
adult life than for behaviour which is restricted to childhood and adolescence. However, more
research on the nature-nurture interplay is needed (Rutter et al. 1998, Simonoff 2001, Burke et al.
2002, Karnik et al. 2006)
Biological factors
Some neural mechanisms seem to be associated with aggressive behaviour, but the findings are
inconsistent and further investigations are needed (Rutter et al. 1998, Burke et al. 2002, McMahon
and Frick 2005). Regarding neurochemicals, although many of them (e.g. noradrenaline, dopamine,
testosterone and cortisol) appear to be involved in aggressive behaviour, the serotonergic system
has been studied most and seems to play a decisive role. As in depression, several studies have
reported that diminished serotonergic function is associated with both concurrent and future
aggression and especially with impulsive aggression. An association has been observed only
between aggression and serotonin, currently there is no evidence of the same association with
nonaggressive antisocial behaviour (Herbert and Martinez 2001, Burke et al. 2002).
Neuroanatomical studies have revealed that decreased metabolism of the frontal lobe and frontal
lobe damage are associated with aggression (Burke et al. 2002). Some prenatal problems (e.g.
maternal smoking) and birth complications especially combined with other factors (e.g. maternal
rejection) seem to increase the risk of antisocial behaviour (Raine et al. 1994, Fergusson et al. 1998,
Hill 2001, Burke et al. 2002).
47
Psychological factors
Children/adolescents with antisocial behaviour have been described to be impulsive, have difficult
temperament and be keen to seek excitement and risky behaviours. They are often described to have
negative emotionality, to use an aggressive style of coping, lack a sense of responsibility and to
often lack self-control. Deficits in social skills and social-information processing (hostile
attributional biases) are strongly associated with risk of antisocial behaviour (Moffitt 1993,
Tremblay et al. 1994, Kingston and Prior 1995, AACAP 1997, Rutter et al. 1998, KeltikangasJärvinen 2001, Dodge and Pettit 2003, Karnik et al. 2006). Several neuropsychological defects are
also associated with antisocial behaviour, such as impaired verbal ability and executive dysfunction
(e.g. defects in abstract reasoning, concept formation, formulating goals, programming behaviour
and sustaining concentration) (Lynam and Henry 2001). Antisocial adolescents often hold positive
attitudes to aggression and consider own aggressive behaviour justified (Rutter et al. 1998).
Psychosocial factors
School problems, poor scholastic achievement and learning difficulties are psychosocial stressors
which are associated with increased antisocial behaviour (Loeber et al. 1998, p.109, Lewin et al.
1999, Loeber et al. 2001, Grigorenko 2006, Karnik et al. 2006). Early puberty (Graber et al. 1997,
Kaltiala-Heino et al. 2003) and negative life events (Bru et al. 2001) are also associated with
antisocial behaviour. Exposure to community violence, either as a victim or as a witness of violence
has been found to increase the risk of antisocial behaviour (Song et al. 1998, Molnar et al. 2005).
Involvement with antisocial peers strongly predicts and reinforces adolescents’own antisocial
behaviour (Brendgen et al. 2000, McCabe et al. 2001, Deptula and Cohen 2004, McMahon and
Frick 2005). Peer rejection and lack of social support from peers are associated with antisocial
behaviour (Lewin et al. 1999, Bru et al. 2001). Antisocial adolescents usually have as many friends
as others, but they report having fewer close and supportive friends, and these relationships tend to
be unstable and conflictual (Deptula and Cohen 2004).
Antisocially behaving children and adolescents often live in families in which internal conflicts and
poor family functioning, domestic violence, inconsistent parenting or punitive discipline,
insufficient monitoring or distant child-parent relationship and lack of social support are typical
(Wasserman et al. 1996, Rutter et al. 1998, Rey et al. 2000, Bru et al. 2001, Loeber et al. 2001,
48
Loukas et al. 2001, Burke et al. 2002). Familiar criminality and psychopathology, especially
paternal antisocial personality disorder, is associated with increased risk of antisocial behaviour
(Loeber et al. 1995, Stranger et al. 1999, Farrington et al. 2001, Loeber et al. 2001). In addition,
antisocial adolescents often grow up in divorced or single-parent families (Koskinen et al. 2001),
but the risk effect of divorce seems to be mediated through impaired quality of parenting by the
custodial parent and through non-involvement of the non-residential parent rather than by divorce
per se (Simons et al. 1999).
Psychopathology
A number of studies have reported that early aggression, oppositional behaviour and hyperactivity
problems are significant factors for subsequent antisocial behaviour. Shyness and social isolation,
especially in combination with aggression, predicts later antisocial behaviour (Moffitt 1993,
AACAP 1997, Rutter et al. 1998, Lahey et al. 1999b, Sourander et al. 2006). Substance abuse and
dependence are strongly associated with antisocial behaviour (Rutter et al. 1998, Bassarath 2001).
In addition, traumas leading to posttraumatic stress disorder are common among adolescents with
antisocial behaviour (Karnik et al. 2006).
Sociodemographic factors
Reportedly, low SES, low parental education, poor housing, poor and disordered schools and
disadvantaged neighbourhoods have also been associated with increased risk of antisocial behaviour
(Lahey et al. 1999b, Loeber et al. 2001, Dodge and Pettit 2003, Hurtig et al. 2005, Sourander et al.
2006).
2.3.5.2 Models on the development of adolescent antisocial behaviour
Criminological studies have revealed that a small minority of adolescents (approximately 3-6%,
mostly males) are responsible for the majority of the known crimes in this age group (Rutter et
al.1998). Although some of these severely antisocially behaving adolescents tend to continue
49
behaving antisocially in adulthood, there is, however, convincing evidence that the majority of
antisocial adolescents desists in antisocial behaviour over time (Pulkkinen 1988, Moffit 1993,
Smith 1995, Loeber and Hay 1997). A large number of studies and theories have tried to determine
whether persistent, severely antisocial adolescents represent a specific type of antisocial behaviours,
and whether these adolescents differ from others regarding risk factors and developmental
trajectories (Rutter et al. 1998, Loeber and Farrington 2000, Aalsma and Lapsley 2001, Hill 2002,
McMahon and Frick 2005). The theories most often referred to are Moffitt’s model on
developmental taxonomy (e.g. 1993, 1996, 2001, 2002) and Loeber’s and co-workers’model of
three developmental pathways in antisocial behaviour (e.g. 1997, 1998, 2000, 2002).
Moffitt’s development taxonomy model
The key feature of the original Moffitt’s development taxonomy model (1993) is a division of
antisocial adolescents into two separate groups according to the onset of antisocial behaviour: those
with early onset (life-course persistent group, LCP) and those with adolescent onset of antisocial
behaviour (adolescent limited group, AL). Among LCP adolescents antisocial behaviour often starts
at pre-school age with oppositional and defiant behaviour, after which symptoms increase steadily
over time and finally escalate in severe antisocial behaviour during adolescence (Moffitt 1993,
Moffitt et al. 1996). LCP is associated with high rates of neuropsychological deficits (e.g. ADHD,
low IQ) which are seen as the primary basis of antisocial behaviour in interaction with other early
difficulties (e.g. difficult temperament, impulsivity and peer difficulties) and adverse family
environments (e.g. poor parenting) (Moffitt 1993, Moffitt et al. 1996). In addition, comorbid mental
health problems are more prevalent and severe in the LCP group than in the AL group (Vermeiren
2003). LCP is more common among males than females as the sex ratio has been estimated to be as
high as 10:1 (Lahey et al. 1998, Moffitt and Caspi 2001). Some of these adolescents in the LCP
group continue to behave antisocially in adulthood and are later diagnosed as having antisocial
personality disorder (Hill 2002, Karnik et al. 2006).
In contrast to the stable continuity in the LCP group, discontinuity is a key feature of antisocial
behaviour among adolescents in the AL group. In this group antisocial behaviour begins during
adolescence and is usually restricted to this period. Associating with antisocial peers, seeking adult
privileges and experience of a temporal maturity gap are seen as the primary basis of antisocial
50
behaviour in the AL group. A maturity gap is described as the long gap between biological and
social maturity (i.e. adult status). Adolescents in the AL group engage in antisocial behaviour to
gain adult privileges and imitate peers in LCP groups who appear to have achieved autonomy from
parents. As AL adolescents do not usually have a long history of neuropsychological deficits or
antisocial behaviour, they have better psychosocial capacities, low rates of overt aggression and
therefore better adult outcomes than LCP adolescents. Most AL adolescents desists antisocial
behaviour in young adulthood, when prosocial behaviour becomes more rewarding and autonomy is
reached otherwise (Moffitt 1993, Moffitt et al. 1996, Moffitt and Caspi 2001). There are no
remarkable sex differences in the AL group as the sex ratio is close at 1:1 (Lahey et al. 1998,
Moffitt and Caspi 2001, Storvoll and Wichstrom 2003).
There is substantial evidence that early emerging versus later emerging antisocial behaviour differ
both quantitatively and qualitatively (Lahey et al. 1998, Lahey et al. 1999a, McCabe et al. 2001).
Antisocial behaviour in those with early onset is described as more frequent, stable, severe, versatile
and aggressive than in those with later onset of antisocial behaviour. In addition, age at onset is
inversely associated with psychosocial impairment, mental health service use and rates of ODD/ CD
(Moffitt 1993, Moffitt et al. 1996, Lahey et al. 1999a). Moffitt et al. (1996) reported that most
antisocially behaving adolescents are in the AL group (24%) as LCP adolescents constitute only a
small subgroup (7%) of all antisocially behaving adolescents.
Loeber’s model of three developmental pathways
The model by Loeber and co-workers (e.g. 1997, 1998, 2000, 2002) of three developmental
pathways in antisocial behaviour was developed in the longitudinal sample of the Pittsburgh Youth
Study, and has been replicated in several other studies (e.g. Loeber et al. 1997, Stouthamer-Loeber
and Loeber 2002). However, as the model was developed among males, it is unclear whether these
pathways can be generalized to females (Storvoll and Wichstrom 2003). The model hypothesized
that different manifestations of antisocial behaviours at different severity levels are manifestations
of the same underlying deviancy, and that males differ in the extent which they escalate over time
into serious antisocial behaviour (Loeber and Farrington 2000, Stouthamer-Loeber and Loeber
2002). This model consists of three pathways for different behaviour patterns (an overt pathway, a
covert pathway and an authority conflict pathway) and orderly sequencing steps which describe
both the increasing severity of behaviour and the decreasing number of adolescent involved.
51
Adolescents may progress along one or multiple pathways towards serious antisocial behaviour
(Loeber et al. 1997, Stouthamer-Loeber and Loeber 2002). “An overt pathway” starts with minor
aggression (e.g. bullying or annoying others) which gradually proceeds to physical fighting and
finally escalates with more serious forms of violence (e.g. rape or attack). “A covert pathway” stars
before age 15 years with minor covert behaviours at the first stage (e.g. lying or shoplifting),
followed by property damage at the second stage (e.g. vandalism or fire setting) and finally results
in moderate to serious delinquency at the third stage (e.g. burglary or fraud). “An authority conflict
pathway” starts before age 12 years first as stubborn behaviour and defiance, and finally ending
with authority avoidance of e.g. parents and teachers (e.g. truancy, running away) (Loeber and
Farrington 2000). Only a minority of males will develop progressively to the most serious level
within pathways; these males are likely to be those with early onset and persistent and severe
manifestation of antisocial behaviour. Most disturbed males manifest antisocially in multiple
pathways. Desistance of antisocial behaviour is inversely associated with serious antisocial acts,
hard drug use, gang membership and positive attitude to antisocial behaviour. Being at work/school
attendance and low physical punishment is also associated with desistance (Loeber and Farrington
2000, Stouthamer-Loeber and Loeber 2002, Stouthamer-Loeber et al. 2004).
The main difference between Moffitt’s and Loeber’s models is whether they focus on heterogeneity,
by defining the age at onset of antisocial behaviour, or identifying specific patterns or features of
antisocial behaviours (Hill 2002). Nevertheless, it is evident that there are multiple developmental
pathways to antisocial behaviour, all characterised by different causal processes and manifestations
of antisocial behaviour.
2.3.5.3 Course and consequences of adolescent antisocial behaviour
Numerous studies have established that antisocial behaviour has considerable continuity both in
adolescence (Fergusson et al. 1996, Loeber et al. 2000b, Cote et al. 2001, Beyers and Loeber 2003,
Kim et al. 2003, Marmorstein and Iacono 2003, Wiesner 2003) and in adulthood (Ferdinand et al.
1995, Satterfield and Schell 1997, Pulkkinen et al. 2000, Simonoff et al. 2004). Costello et al.
(2003) found that adolescents with a history of antisocial behaviour were ten times more likely than
those with no history of antisocial behaviour to behave antisocially in the future. In a longitudinal
52
study by Satterfield and Schell (1997), 35 % of males with comorbid hyperactivity and childhood
antisocial behaviour engaged in multiple delinquent acts during adolescence, and 70% of them
continued offending in adulthood. Those adolescents especially with an early onset of antisocial
behaviour and those with severe, aggressive antisocial symptoms accompanied by a range of
psychosocial problems are most likely to continue to behave antisocially in later adolescence and in
adulthood (Moffitt 1993, Rutter et al. 1998, p. 98, Loeber et al. 2000a, Cote et al. 2001, Maughan
and Rutter 2001). This is likely because, due to years of involvement in severe antisocial behaviour,
adolescents miss opportunities to adapt and practise prosocial behaviours and are likely to make
irrevocable future decisions (Moffitt 1993, Loeber and Farrington 2000, Taylor et al. 2002,
Fergusson et al. 2005). However, adolescents with later onset of antisocial behaviour may not be
totally without problems in adulthood as they have been reported to have more mental disorders and
more self-reported offending than those without any history of antisocial behaviour (Moffitt et al.
2002).
Antisocial behaviour can lead to multiple problems and consequences throughout the life span
(McMahon and Frick 2005). In general, the greater the versatility of antisocial behaviour, the worse
the outcome (Earls and Mezzacappa 2002, McMahon and Frick 2005). During adolescence, prior
antisocial behaviour is associated with impairment in adjustment, social and scholastic problems,
unemployment, substance abuse and peer delinquency (Achenbach et al. 1998, Fergusson and
Horwood 1998, Loeber et al. 1998, Rutter et al. 1998, Loeber and Farrington 2000, Loeber et al.
2000b, Helstelä and Sourander 2001, Tramontina et al. 2001, Taylor et al. 2002). Antisocially
behaving adolescents also have increased risk of psychosomatic symptoms, suicidal behaviour,
affective disorders (Kivivuori 2000, Pliszka et al. 2000, Ruchkin et al. 2003, Vermeiren 2003).
Antisocial behaviour is one of the most common reasons for which adolescents are referred to
psychiatric evaluation and treatment (Loeber and Stouthamer-Loeber 1998, Hill 2002, Weisbrot and
Ettinger 2002, Karnik et al. 2006). Many of them drift into being clients of child welfare (LehtoSalo et al. 2002, Marttunen et al. 2004, Kitinoja 2005).
In adulthood, prior antisocial behaviour is associated with increased rates of psychiatric disorders,
substance abuse and premature mortality (Kratzer and Hodgins 1997, Pajer 1998, Rutter et al. 1998,
Laub and Vaillant 2000, Hofstra et al. 2002, Kim-Cohen et al. 2003). In a longitudinal study by
Kim-Cohen et al. (2003) for all adult mental disorders at age 26 years, as many as 25% to 60 % of
cases had a history of childhood antisocial behaviour. As adults, antisocial adolescents often
53
manifest dysfunctional relationships, use of violence, marital problems, poor educational
achievement and unemployment and high rates of service utilization (Pajer 1998, Rutter et al. 1998,
Loeber and Farrington 2000, Moffitt et al. 2002, Fergusson et al. 2005). Adult criminality is often
predicted by antisocial behaviour in childhood and adolescence (Kratzer and Hodgins 1997, Pajer
1998, Pulkkinen et al. 2000, Moffitt et al. 2002, Fergusson et al. 2005).
Although the results are somehow inconsistent (Ferdinand et al. 1995, Costello et al. 2003, Storvoll
and Wichstrom 2003), it has been reported that the continuity of severe antisocial behaviour is
higher among females than males (Loeber et al. 2000a). It has also been reported that antisocial
females may be at greater risk of multiple problems and consequences than males (Loeber and
Keenan 1994, Loeber and Loeber-Stouthamer 1998, Tiet et al. 2001). This notion is usually referred
to as the gender paradox, that is the gender with lower prevalence of disorder is actually at higher
risk of poor outcome (Tiet et al. 2001).
2.3.6 Comorbidity in adolescent antisocial behaviour
Comorbidity among antisocially behaving adolescents seems to be the rule rather than the
exception. Comorbidity seems to be more common among antisocial females than males (Keenan et
al. 1999, Kann and Hanna 2000). The most common comorbid disorders with antisocial behaviour
are substance abuse or dependence (11 - 50 %), mood and anxiety disorders (5 - 45 % each) and
ADHD (43-93%) (Jensen et al. 1997, Angold et al. 1999, Costello et al. 1999, Renaud et al. 1999).
Antisocial adolescents also have elevated rates of suicidal behaviour (5 - 46%) even without
concurrent depression (Apter et al. 1995, AACAP 1997, Angold et al. 1999, Loeber et al. 2000a,
King et al. 2001, Pelkonen and Marttunen 2003). The majority of studies have found most
comorbid disorders to be secondary to antisocial behaviour with the possible exception of ADHD,
which begins in early childhood (AACAP 1997, Costello et al. 1999, Loeber et al. 2000a, Angold
and Costello 2001, McMahon and Frick 2005).
Comorbidity between depression and antisocial behaviour will be discussed in detail in Chapter 2.4.
54
2.4 Comorbidity between depression and antisocial behaviour in adolescence
2.4.1 Features of comorbid depression and antisocial behaviour in adolescence
Given that the core symptoms of depression and antisocial behaviour differ notably, the strong cooccurrence of these disorders may be rather surprising (Wolff and Ollendick 2006). On the other
hand, irritability is a frequent symptom both among depressed and antisocially behaving
adolescents (Herkov and Myers 1996, Weisbrot and Ettinger 2002). Thus, numerous studies,
especially in clinical (e.g. Biederman et al. 1995, Grilo et al. 1996, Herkov and Myers 1996,
Karlsson et al. 2006b) and delinquent samples (e.g. Duclos et al. 1998, Ulzen and Hamilton 1998,
Teplin et al. 2002) have reported high comorbidity between adolescent depression and antisocial
behaviour. However, whether the symptom profiles of combination of depression and antisocial
behaviour differ from pure depression and/or from pure antisocial behaviour has been studied less.
Reportedly, comorbid depression with antisocial behaviour may be a distinct disorder from both
pure depression and pure antisocial behaviour based on a different etiology (Birmaher et al. 1996,
Parker and Roy 2001, Simic and Fombonne 2001). Adolescents with a combination of depression
and antisocial behaviour have been reported to witness more family violence, experience more poor
parenting, hostility and physical abuse than those with pure depression (Meller and Borchardt 1996,
Simic and Fombonne 2001), and to have more often a history of physical abuse and events resulting
in loss of self-esteem than those exhibiting pure antisocial behaviour (Simic and Fombonne 2001).
A combination of depression and antisocial behaviour has been reported to be associated with
earlier onset of depression, longer duration of disorders, increased rates of irritability and police
contacts, increased suicidality, lower social competence and decreased rates of anxiety (Capaldi
1991, Harrington et al. 1991, Capaldi 1992, Lewinsohn et al. 1994, Lewinsohn et al. 1995a, Meller
and Borchardt 1996, Renouf et al. 1997, Simic and Fombonne 2001). In clinical samples
adolescents with comorbid depression and antisocial behaviour have been reported to evince more
atypical symptoms of depression such as high energy, inflated sense of self, optimism, lack of social
anxiety, less insomnia, less guilt, fewer ruminations and lower overall severity of depressive
symptoms than adolescents with pure depression (Herkov and Myers 1996, Simic and Fombonne
2001).
55
Further, antisocially behaving adolescents with comorbid depression have been reported to be less
likely to steal, to be less destructive, to be less overtly aggressive and to have overall fewer
symptoms of antisocial behaviour than those adolescents with pure antisocial behaviour (Simic and
Fombonne 2001). Although not consistently reported (Haapasalo and Hämäläinen 1996, Arseneault
et al. 2000), there is also some evidence that depression is more frequent among aggressive than
among non-aggressive delinquents (Rey et al. 2005). In addition, some evidence exists that the
timing of comorbidity affects the profile of antisocial behaviour, as Loeber et al. (1994) found that
early adolescent males with comorbidity tend to show overt and covert behaviour whereas in later
adolescence comorbid adolescents had more conflicts with authority. In light of these results, some
authors (e.g. Simic and Fombonne 2001) have proposed that comorbidity between depression and
conduct disorder also merits separate diagnostic criteria in DSM-IV, as has been done in ICD-10.
However, not all studies have found differences between features of pure disorders and comorbid
disorders (Ezpleta et al. 2005) or in the profile of delinquency according to depression (Elonheimo
et al. 2007).
2.4.2 Epidemiology of comorbid depression and antisocial behaviour in adolescence
Earlier studies have reported widely varying rates of comorbidity between adolescent depression
and antisocial behaviour. Most of these studies have been conducted in sentenced/incarcerated
samples (e.g. Timmons-Mitchell et al. 1997, Duclos et al. 1998, Ulzen and Hamilton1998, Pliszka
et al. 2000, Vermeiren et al. 2000) or in clinical populations (e.g. Mitchell et al. 1988, Grilo et al.
1996, Greene et al. 2002). In clinical samples the rates of comorbidity between depression and
antisocial behaviour have ranged between 15% and 72% (Kovacs et al. 1988, Harrington et al.
1991, Grilo et al. 1996, Herkov and Myers 1996, Greene et al. 2002). In a Finnish outpatient sample
25% of the depressed boys and 7% of the depressed girls were diagnosed to have comorbid
disruptive disorder (Karlsson et al. 2006b). According to the review by Vermeiren (2003)
prevalence rates of depression have ranged between 11% and 33% in instutionalized delinquent
samples, while up to 50% of delinquent adolescents have been reported to have less severe
depressive symptoms. Sailas et al. (2005) reported that 3.5% [odds ratio (OR) 4.3 and 5.8] of the
young prisoners in Finland had had at least one inpatient treatment episode for depression in the
past year. Both of these research frames have some limitations in their generalizability as clinical
samples are hampered by the fact that comorbid adolescents are much more likely to receive
56
treatment than adolescents with only one disorder (Angold et al. 1999, Rowe et al. 2002).
Regarding delinquent samples, the fact that rates of depression may be biased due to the
consequences of incarceration cannot be overlooked. Some studies have found that being caught
accompanied by the feelings of shame and guilt as well as separation from family causes depression
or at least exacerbates it, as others have suggested that depression is associated with the severity and
frequency of antisocial behaviour (Kashani et al. 1980, Vermeiren et al. 2002a, Gomes et al. 2003).
Studies on comorbidity between depression and antisocial behaviour conducted in community
samples give a more extensive picture of antisocial behaviour than studies conducted in delinquent
samples based on official crime statistics (Kaltiala-Heino et al. 2006). In a review of 16
epidemiological studies Angold et al. (1999) found a median OR of 6.6 (95% CI 4.4-11.0) for
comorbidity between depression and conduct disorder among children and adolescents. In pure
adolescent general populations, OR for the comorbidity between depression and antisocial
behaviour ranged from 1.7 to 7.0, and rates of comorbidity have ranged from 6.2% to 45% (Table
4). In Finland, estimates of comorbidity have ranged from 6.2% to 37.2% depending on the
definition of antisocial behaviour used (Kaltiala-Heino et al. 2000, Kaltiala-Heino et al. 2003,
Pelkonen et al. 2003, Sihvola et al. 2007). Studies focusing especially on the association between
delinquent behaviour and depression in a general adolescent population are rare (Vermeiren et al.
2002a).
57
Table 4. Comorbidity between depression and antisocial behaviour in a community based child-adolescent samples.
Reference
N;
gender
Study
population
Age
(years)
Assessment
method; source
of information
Definition of
depression
Definition of antisocial
behaviour
Timeframe of
comorbidity
Rates of comorbidity (%);
OR (CI)
Capaldi
1991
203;
m
At-risk school
sample
Grade 6
Interviews and
questionnaires;
adolescent,
parents,
teachers,
observers
Depressive
symptoms
Conduct problems
Not reported
45%
Lewinsohn
et al. 1991
and Rohde et
al. 1991
1 710;
m and f
Random
school sample
14 - 18
K-SADS;
adolescent
Depressive
disorders
(DSM-III-R)
ODD + CD (DSM-IIIR)
Concurrent
8.0%; 5.3 (1.8-15.7)
Lifetime
12.1%; 2.1 (1.4-3.1)
Fergusson et
al. 1993
965 - 986;
m and f
Birth cohort
15
DISC, SRED,
RBPC;
adolescent,
parent
Mood
disorders
(DSM-III-R)
ODD + CD
Not reported
3.4 (1.9-6.3)
Garnefski
and Diekstra
1997
11 516;
m and f
Subsample
from random
school sample
12 - 18
Monitoring-theFuture
Questionnaire;
adolescent
Emotional
problems
Aggressive/criminal
behaviours
Not reported
35.4%
Loeber et al.
1998
506;
m
Stratified
random
school sample
14
SRD, CBCL,
TRF, YSR,
DISC-P, MFQ,
crime record;
adolescent,
parent, teacher
Depressed
mood
Delinquency
Not reported
1.7 (CI not reported)
MillerJohnson et
al. 1998
340;
m and f
Stratified
random
school sample
Grade 6
CAS;
adolescent
Depressive
symptoms
Conduct symptoms
Not reported
11.2%
58
Reference
N;
gender
Study
population
Age
(years)
Assessment
method; source
of information
Definition of
depression
Definition of antisocial
behaviour
Timeframe of
comorbidity
Rates of comorbidity (%);
OR (CI)
KaltialaHeino et al.
2000
17 643;
m and f
School
sample
14 - 16
Questionnaire;
adolescent
Depressive
symptoms
Frequent bullying at
school
Not reported
m: 13.0%
f: 25.5%
m and f: 4.3 (3.4-5.7)
Flannery et
al. 2001
Costello et
al. 2003
484;
m and f
1 420;
m and f
School
sample
Random
sample
14 - 19
9 - 16
TSC-C, selfreported rating
scale for
violence not
specified;
adolescent
Depression
Dangerous violence
CAPA;
adolescent,
parent
Depressive
disorders
(DSM-IV)
CD (DSM-IV)
Not reported
m: 1.3 (0.7-2.3)
f: 6.4 (2.1-19.0)
Concurrent
m: 0.7 (0.2-2.4)
f: 10.6 (2.0-54.7)
Gomes et al.
2003
2 001;
m and f
Random
school sample
12 - 18
Self-reported
scales*;
adolescent
Emotional
problems
Delinquency
12 months
16.7%
KaltialaHeino et al.
2003
36 549;
m and f
School
sample
14 –16
Questionnaire;
adolescent
Depressive
symptoms
Externalizing problems:
drunkenness, other
substance use, bullying,
truancy
Not reported
m: 6.2%
West et al.
2003
1 860;
m and f
Cohort,
school sample
15
Voice-DISC;
adolescent
Depressive
disorders
(DSM-IV)
ODD + CD (DSM-IV)
Concurrent
5.6 (2.9-10.8)
Maughan et
al. 2004
10 438;
m and f
Systematic
sampling
5 –15
DAWBA;
adolescent,
parent, teacher
Depressive
disorders
(DSM-IV)
CD (DSM-IV)
Not reported
m: 13.2 (5.5-31.7)
SSAGA;
adolescent,
parents
MDD (DSMIV)
ODD
Sihvola et al.
2007
1 854;
m and f
Twins
14
f: 9.7%
f: 3.6 (0.7-17.8)
CD
Lifetime
11.6%; 7.0 (1.3-38.2)
37.2%; 4.8 (2.2-10.6)
(continued)
59
Table 4. (continued)
CAPA = The Child and Adolescent Psychiatric Assessment,
CAS = Child Assessment Schedule,
CBCL = the Child Behavior Checklist,
CD = Conduct Disorder
CI = confidence interval,
DAWBA = the Development and Well-Being Assessment,
DISC = Diagnostic Interview Schedule for Children,
DISC-P = Diagnostic Interview Schedule for Children (parent-version),
DSM-III-R = Diagnostic and Statistical Manual of Mental Disorders, third
edition revised,
DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, fourth
edition,
f = females,
K-SADS = The Schedule for Affective Disorders and Schizophrenia for
School-Aged Children,
m = males,
MDD = Major Depressive Disorder,
MFQ = the Recent Mood and Feelings Questionnaire,
ODD = Oppositional Defiant Disorder,
OR = odds ratio,
RBPC = Revised Behavior Problems Checklist,
SRD = Self-Reported Delinquency Scale,
SRED = the Self-Report Early Delinquency,
SSAGA = the Semi-Structured Assessment for the Genetics of Alcoholism,
TRF = the Teacher Report Form,
TSC-C = Trauma Symptom Checklist for Children,
Voice-DISC = a computer administered version of DISC,
YSR = the Youth Self-Report,
* = emotional problems measured by modified version of scale by Boyle and Offord
(1994), delinquency scale not specified
60
Although not consistently reported (e.g. Kovacs et al. 1988, Mitchell et al. 1988, Lewinsohn et al.
1995a, Costello et al. 1996, Domalanta et al. 2003, Maughan et al. 2004), depression may be more
frequent among antisocially behaving females than males (e.g. Herkov and Myers 1996, Overbeek
et al. 2001, McCabe et al. 2002, Costello et al. 2003, Vermeiren 2003). For instance, Flannery et al.
(2001) found antisocially behaving females to be almost three times more likely to be depressed
than antisocial males. Rates of comorbid depression may vary between subtypes of antisocial
behaviour. For instance, Rowe et al. (2006) found that depression was independently associated
with oppositionality and delinquency but not with physical aggression whereas, Burke et al. (2005)
found that ODD predicted independently subsequent depression whereas CD did not. On the other
hand, some there is evidence that comorbid depression is more prevalent in those with CD than with
ODD (Wolff and Ollendick 2006). Chiles et al. (1980) and Haapasalo and Hämäläinen (1996) found
no differences in delinquency patterns in relation to depression, but as contrary, Vermeiren et al.
(2002a) found a linear relationship between severity of depression and severity of delinquency.
Interaction between age and comorbidity between depression and antisocial behaviour has been less
studied and the results are not consistent. Depression may be more frequent among LCP adolescents
than AL adolescents (Vermeiren 2003) as well younger adolescents seem to have stronger cooccurrence than older adolescents and young adults (Loeber and Keenan 1994, Overbeek et al.
2001, Karlsson et al. 2006b). However, Loeber et al. (1994) and Wu et al. (1999) found comorbidity
between depression and antisocial behaviour to increase from childhood to adolescence and to reach
its peak in middle adolescence. Kovacs et al. (1988) reported that older age-at-onset of depression
was associated with presence of comorbid CD. Herkov and Myers (1996) found no difference in
relation to age between pure depressed adolescents and depressed adolescents with antisocial
behaviour.
2.4.3 Models on the development of comorbid depression and antisocial behaviour in
adolescence
Although numerous studies have demonstrated that depression and antisocial behaviour co-occur
far more often than would be expected by chance (Loeber and Keenan 1994) and that this cooccurrence is not a methodological artefact (Angold et al. 1999, Wolff and Ollendick 2006), the
underlying mechanism of comorbidity and the longitudinal associations between these disorders are
61
still unclear (Fergusson et al. 1996, Measelle et al. 2006, Rutter et al. 2006, Wolff and Ollendick
2006). In fact, there are surprisingly few studies on this topic in adolescence, and the existing
results are inconsistent (Capaldi 1992, Loeber et al. 2000a). Further, different constructions in
defying antisocial behaviour complicate conclusions. Research on comorbidity between antisocial
behaviour and depression, and on developmental associations is, however, important because it has
implications for classification systems, etiology and treatment (Keiley et al. 2003).
In studying developmental associations between depression and antisocial behaviour the concept of
heterotypic continuity is pivotal. Heterotypic continuity refers to a continuous process in which one
disorder generates manifestations of different forms over time or in which one disorder exposes
adolescents to different disorders at different ages (Angold et al. 1999, Costello et al. 2003). There
are basically at least three possible developmental models to explain the heterotypic continuity
between adolescent depression and antisocial behaviour: acting out, failure and mutual influence
models (Fergusson et al. 1996, Overbeek et al. 2001, Wolff and Ollendick 2006). Earlier studies
have found some evidence to support all these models. The models differ in whether they assume
associations to be causal (one disorder creates an increased risk for the other) or non-causal
(association is based on non-specific risk factors) or disorders are reciprocally associated with each
other (Caron and Rutter 1991, Fergusson et al. 1996, Overbeek et al. 2001, Wiesner 2003).
Acting out model
The acting out model is based on psychoanalytic theory proposing that depressed adolescents act
out internalizing problems by masking them with antisocial behaviour (Capaldi 1992, Overbeek et
al. 2001, Wolff and Ollendick 2006).The concept of “masked depression”refers to depressive
features which are not traditionally associated with depression (e.g. stealing or lying) and they are
thought to mask traditional depressive symptoms (Toolan 1962, Glaser 1968). Thus, depression
precedes antisocial behaviour, and the cross-lagged associations are unidirectional and causal
(Capaldi 1992, Overbeek et al. 2001). Depressed adolescents may develop antisocial behaviour as
irritability and hopelessness (which are often associated with depression) become more severe,
which, in turn, may contribute to conflicts with others and increased likelihood that adolescents
engage in antisocial and risky behaviours (Measelle et al. 2006, Wolff and Ollendick 2006).
Similarly, the negative affect associated with depression may be annoying or irritating to others,
resulting in conflictual relationships with others, peer rejection and affiliation with deviant peers,
which increases a risk to a person’s own antisocial behaviour (Oland and Shaw 2005). On the other
62
hand, it has been reported that acting out depression is likely to be associated especially with a
family history of depression (Karnik et al. 2006).
The acting out model was first introduced by Kovacs et al. (1988), who found in a clinical sample
that 56% of the children (aged 8 to 13 years) developed depressive disorders prior to antisocial
behaviour compared to 25% of those who developed antisocial behaviour prior to depression. It was
concluded that antisocial behaviour developed mostly as a complication of the depression. Loeber et
al. (1994) found that depression in early adolescent males was associated with escalation to more
serious delinquency and a variety of delinquent behaviour but by contrast, Leadbeater et al. (1999)
and Vermeiren et al. (2002b) found depression to be protective against recidivism.
Failure model
The failure model assumes that there is a causal, unidirectional association between antisocial
behaviour and depression. Antisocial behaviour emerges first and the ensuing experiences of social
failure contribute to the development or to the vulnerability to depression (Overbeek et al. 2001).
Antisocial adolescents usually face many problems in social relations (e.g. peer rejection, lack of
social support, conflict with parents and teachers), have low social competence and poor coping
skills and encounter problems in academic attainment which all in turn may lead to experiences of
failure. Such failures often affect social development and contribute to or confirm associations with
deviant peers and negative attitudes, which in turn contribute to depression (Capaldi 1992, Capadi
and Stoolmiller 1999, Kiesner 2002). In addition, it has also been reported that when depression is
secondary to antisocial behaviour, antisocially behaving adolescents are likely to have a family
history of antisocial behaviour (Karnik et al. 2006).
The failure model was first introduced by Patterson and Capaldi (1990) and later supported by a
longitudinal study by Capaldi (1992), reporting that antisocial behaviour among 12-year-old boys
predicted increased depression in a 2-year span but not vice versa. Similar results from community
samples have later been reported e.g. by Rohde et al. (1991), Feehan et al. (1993), Kiesner (2002)
and Fergusson et al. (2003). Nock et al. (2006) found that CD preceded depression in 72% of cases
compared to 19 % of those who developed depression prior to CD. In a Finnish sample, Pelkonen et
al. (2003) found that among boys problems with the law at age 16 predicted subsequent depression
at age 22 (OR 3.0).
63
Mutual influence model
According to the mutual influence model comorbidity between depression and antisocial behaviour
results from non-specific (shared or overlapping) risk factors and life pathways which lead to both
disorders, and disorders are also reciprocally reinforcing/effecting on each other over time. The
onset of one disorder increases vulnerability to the other and vice versa. Therefore, cross-lagged and
bidirectional associations between disorders are possible (Overbeek et al. 2001, Wolff and
Ollendick 2006). However, the interaction between depression and antisocial behaviour may not be
symmetrical in that each disorder affects the other to the same degree (Loeber and Keenan 1994).
Wolff and Ollendick (2006) proposed antisocial behaviour to have a stronger effect on depression
than vice versa, but Beyers and Loeber (2003) reported that depression had a more robust effect on
antisocial behaviour than vice versa. Many biological, psychological and psychosocial risk factors
may contribute to the comorbidity (Fergusson et al. 1996, Weisbrot and Ettinger 2002, Wolff and
Ollendick 2006). E.g. O’Connor et al. (1998) found that 45% of the observed covariation of
comorbidity between depression and antisocial behaviour was accounted for by common genetic
liability. Poor family functioning and parenting that is low in warmth are other examples of shared
risk factors for depression and antisocial behaviour (Lahey et al. 2002). The mutual influence model
has been supported by some researchers (e.g. Beyers and Loeber 2003, Wiesner 2003, Wolff and
Ollendick 2006). In a community sample of adolescent females Measelle et al. (2006) found that
initial levels of antisocial behaviour predicted increases in depression, whereas initial level of
depression predicted slower decreases in antisocial behaviour. In a clinical sample of preadolescent
males Lahey et al. (2002) found that depression increased with age and this change was
accompanied by concurrent increases in antisocial behaviour, whereas antisocial behaviour
predicted subsequent depression.
Not all studies have found evidence for heterotypic continuity between depression and antisocial
behaviour (Overbeek et al. 2001, Costello et al. 2003, Marmorstein and Iacono 2003, Roza et al.
2003, Ingoldsby et al. 2006). Several reasons might explain the differing results, such as disparity of
research methods (e.g. characterics of the sample, definitions of antisocial behaviour) or that
longitudinal associations may be age-related or that some unmeasured factor (e.g. family conflict)
may moderate the association (Overbeek et al. 2001, Ingoldsby et al. 2006).
Studies on gender differences among the longitudinal associations between depression and
antisocial behaviour in adolescence are rare. Wiesner (2003) found limited evidence that antisocial
64
behaviour among males preceded depression but not vice versa. Among females a circular process
was found as high level of antisocial behaviour preceded high level of depression, which in turn
preceded de-escalation of antisocial behaviour, which finally resulted in lower level of depression.
Rowe et al. (2002) found that ODD predicted subsequent depression among females, but not among
males. In the study by Hofstra et al. (2002) antisocial behaviour in childhood and adolescence
predicted only males’mood disorders in adulthood. Overbeek et al. (2001) found no gender
differences, suggesting that there are no cross-lagged associations between depression and
antisocial behaviour as homotypic continuity of the disorders best explains the future course of
depression and antisocial behaviour.
2.4.4 Course and consequences of comorbid depression and antisocial behaviour in
adolescence
Results on the predictive power of antisocial behaviour for the recurrence of depression have been
inconsistent, as some studies have reported less frequent (Harrington et al. 1991) and others similar
rates of recurrence of depression (Fombonne et al. 2001a). Kovacs et al. (1988) reported that
comorbid CD did not affect symptom presentation of depression or the duration of depression.
Sanford et al. (1995) found a trend that persistent depression might be more frequent in those with
comorbid antisocial behaviour. Similarly, the results regarding the effect of depression on the
course of antisocial behaviour are inconsistent. For instance, Harrington et al. (1991) and Capaldi
(1992) found that depression had little effect on the course of conduct disorder or on rates of adult
crimes. By contrast, Sourander et al. (2007) reported that children with comorbid conduct and
internalizing problems at age 8 had the poorest outcomes and highest risk of subsequent psychiatric
disorders, criminal offences and overall self-reported problems in late adolescence. On the other
hand, there is some evidence that depression protects adolescents against subsequent antisocial
behaviour (Vermeiren et al. 2002b).
Suspension from school, deficits in academic skills, earlier age of sexual intercourse, self-injurious
behaviour, suicidality, global dysfunctioning, substance abuse and treatment utilization are more
frequent among depressed adolescents with antisocial behaviour than among either of these
disorders alone (Capaldi 1991, Harrington et al. 1991, Capaldi 1992, Lewinsohn et al. 1995a,
Miller-Johnson et al. 1998, Wu et al. 1999, Marmorstein and Iacono 2001 and 2003, Simic and
65
Fombonne 2001, Ingoldsby et al. 2006). When adult, adolescents with both depression and
antisocial behaviour have been reported to have higher rates of occupational problems and
unemployment, low income and educational level, pervasive social dysfunction and problems in
everyday coping, substance abuse, antisocial personality disorder, suicidal behaviours and
criminality than depressed adolescents without antisocial behaviour (Harrington et al. 1991,
Fombonne et al. 2001a and 2001b). However, not all studies have found the combination of
depression and antisocial behaviour to predict the poorer outcome. For instance, Capaldi and
Stoolmiller (1999) found that although both depression and antisocial behaviour were associated
with adverse outcomes in multiple domains in young adulthood, the interaction of these disorders
did not predict greater risk for any single domain over than either disorder alone.
2.5 Summary of the literature reviewed
In the light of current evidence, depression and antisocial behaviour are two of the most common
forms of juvenile psychopathology. Both the incidence of depression and the incidence of
delinquency reach their peak in middle adolescence. Depression is twice as common among
adolescent females as males, and this gender difference seems to be greatest between ages 15 and
18. In general, more males engage in antisocial behaviour, although the sex ratio varies in relation
to the subtypes of antisocial behaviour. Both depression and antisocial behaviour have a
multietiological background and in fact, they are reported to be risk factors for each other. Social
support is inversely associated with both depression and antisocial behaviour. Depression and
antisocial behaviour are accompanied by a variety of psychosocial problems which may have
significant negative effects on adolescent development and psychosocial adjustment. Comorbidity
is very common in both disorders and is likely to increase psychosocial impairment and complicate
treatment. Thus, depression and antisocial behaviour can be considered as major public health
problems.
Reportedly, depression and antisocial behaviour disorders have considerable continuity both in later
adolescence and in adulthood. However, possible gender differences in the continuity of these
disorders need to be further studied. Notably, although studies of antisocial behaviour based on
female samples have increased in recent years, there is still relatively little psychiatric research on
antisocial females.
66
While the magnitude and importance of the comorbidity between adolescent depression and
antisocial behaviour has already been acknowledged, further detailed knowledge about the
interrelations between these disorders is still needed. Major issues not sufficiently studied so far are
whether depression affects the patterns of antisocial behaviour, and possible gender differences in
these patterns. Similarly, although a number of studies have reported inverse associations both
between social support and depression/antisocial behaviour, apparently no study has so far
investigated whether social support mediates the comorbidity between depression and antisocial
behaviour or if comorbidity is independent of social support. In addition, comorbidity between
depression and antisocial behaviour has mostly been investigated among referred and
institutionalized adolescents, which limits the generalizability of the findings. Thus, research
conducted in community samples is indicated. Further, most of the studies have been conducted in
the United States or in the UK, in continental Europe and especially in the Nordic countries only
few studies have focused on this topic in general community samples. Because cultural and social
differences may influence the association between disorders, it is essential to assess comorbidity in
different countries and in different cultures.
Surprisingly few studies on the longitudinal associations between depression and antisocial
behaviour have been conducted in adolescent community samples, and the findings are inconsistent.
Thus, the underlying mechanism of comorbidity between these disorders is still unclear. Many
studies have found that antisocial behaviour precedes depression, but because of the wide disparity
of research methods (e.g. age distributions of the samples, sex ratio, length of follow-up period,
definitions of antisocial behaviour) strict conclusions are difficult to draw. In addition, most of the
studies so far have investigated longitudinal associations between these disorders using only one
developmental model as a theoretical framework. It would be interesting to simultaneously test
different developmental models to explain the longitudinal association between depression and
antisocial behaviour.
Studies on gender differences on the longitudinal associations between depression and antisocial
behaviour in adolescence are rare. Given that the rates of depression and antisocial behaviour and
the course of these disorders differ among males and females, it could be assumed that longitudinal
associations between these disorders also differ between genders. Knowledge about developmental
associations is essential for the development of both effective prevention and interventions for
depression and antisocial behaviour.
67
3. AIMS OF THE STUDY
The general aim of this study was to analyse associations between self-reported depressive
symptoms and antisocial behaviour both concurrently and longitudinally in a community sample
during middle adolescence. Unless otherwise stated, the term antisocial behaviour is used as an
umbrella term to refer to diverse delinquent behaviours and symptoms of CD and ODD which
violate societal norms and/or the rights of others and/or are against the law.
The specific aims of the study were:
I
To investigate the association between self-reported depressive symptoms and
delinquent behaviour.
II
To investigate the patterns of delinquent behaviour (frequency, versatility and
specialization) among depressed and non-depressed repeatedly delinquent adolescents.
III
To investigate interrelations between self-reported depressive symptoms, antisocial
behaviour and perceived social support.
IV
To investigate homotypic continuity, concurrent comorbidity and longitudinal
associations (the failure model and the acting out model) between self-reported
depressive symptoms and antisocial behaviour during a two-year follow-up in middle
adolescence.
68
4. MATERIALS AND METHODS
The present study is based on two Finnish community studies among adolescents: the School
Health Promotion Study and the Adolescent Mental Health Cohort Study.
4.1 The School Health Promotion Study (I and II)
The School Health Promotion Study (SHPS) is a nationwide cross-sectional school survey carried
out biannually in the same municipalities since 1996 among Finnish adolescents in about 90% of all
municipalities in Finland. SHPS examines adolescents’health-related behaviour, health, school as a
work environment and living conditions. SHPS is co-ordinated by STAKES, the University of
Jyväskylä and the University of Tampere. The study has been approved by the Ethical Committee
of Tampere University Hospital. The survey is conducted during a regular school lesson in
comprehensive schools (grades 8th and 9th) and in high schools (grades 1st and 2nd) (Luopa et al.
2003, http://info.stakes.fi/kouluterveyskysely).
4.1.1 Sample and procedures (I and II)
The subjects of the present study were adolescents from 8th and 9th grades of the comprehensive
schools who participated in SHPS in 2002. In 2002 SHPS was conducted in three of the six
administrative divisions in Finland (Lapland, East Finland and South Finland) and covered 150
municipalities. These divisions include urban and rural municipalities of different sizes, so the
sample is representative of Finnish adolescents from diverse living environments. The survey was
completed anonymously during a regular school lesson under the supervision of a teacher.
Adolescents absent on the survey day were not contacted. 416 comprehensive schools took part in
the survey. These schools had a total of 66, 039 pupils, of whom 53, 524 adolescents (81%)
responded to the survey (Luopa et al. 2003).
69
In Study I, 2, 955 of the responses (5.5% of the respondents) were excluded from the study because
of the incomplete responses on the depressive symptoms scale. Thus, the final sample was 50, 569
adolescents (50.1% males). Subjects’mean age was 15.3 years (std. 0.6, range 14-16 years). Of the
subjects 65.4% lived in urban municipalities, 76.5% were living with both parents and 74.5% had
lived in the same area for over 9 years. Stable employment of their parents (not unemployed during
the past years) was reported by 72.5% of the subjects and 29.8% of the subjects had parents (one or
both) who had completed an academic degree.
Study II focused on a subsample of the SHPS in 2002. Adolescents in this subsample (n=3, 825)
had committed at least one type of offence at least five times during the past 12 months. Of these
adolescents, 146 (3.8% of the subsample) were excluded because of incomplete responses on the
depressive symptoms scale. Thus, the final sample in Study II was 3, 679 (69.3% males), and these
adolescents are called “repeatedly delinquent adolescents”. Subjects’mean age was 15.3 years (std.
0.6, range 14-16 years). Of these subjects, 66.2% lived in urban municipalities, 67.4% were living
with both parents and 67.7% had lived in the same area for over 9 years. Stable employment during
the past year of at least one parent was reported by 63.6% of the subjects and 31.6% of the subjects
had parents (one or both) who had completed an academic degree.
4.1.2 Measures (I and II)
Self-reported depressive symptoms were assessed by a Finnish modification (RBDI; Raitasalo and
Notkola 1987, Raitasalo 1995, Kaltiala-Heino et al. 1999b) of the 13-item Beck Depression
Inventory (BDI; Beck and Beck 1972, Beck et al. 1974). The 13-item BDI is a shortened version of
the original 21-item Beck Depression Inventory (Beck and Beck 1972, Beck et al. 1988, Myers and
Winters 2002). The 13-item BDI was developed for screening purposes, and it has been reported to
represent mainly one cognitively oriented symptom dimension (Beck et al. 1988, Raitasalo 1995).
The diagnostic efficiency of the 13-item BDI is good (Beck et al. 1988, Bennett et al. 1997,
Fountoulakis et al. 2003). In the RBDI an introductory question and one positive choice answer
were added for each item. Thus, RBDI constructs a dimensional continuum in which positive mood
and depressive symptoms are the two end points of the continuum (Raitasalo 1995, Kurki et al.
2000, Konu 2002). The RBDI has good reliability and validity, and it has been widely used among
70
diverse Finnish populations (e.g. Raitasalo and Notkola 1987, Hänninen and Aro 1996, KaltialaHeino et al. 1999b and 2001, Fröjd et al. 2006a).
Each of the 13 items of the RBDI is scored 0 –3 according to the severity of the symptom as
estimated currently (today). The original version of the RBDI includes one item on suicidal
ideation. This item, however, was excluded from the SHPS 2002 because of the decision of the
Finnish Ministry of Education, which did not allow the use of this item for fear that asking about
suicidality might provoke it in adolescents. Results from previous samples have demonstrated that
the original threshold of the RBDI can be used even without the suicidality item (Kaltiala-Heino et
al. 1999b). Thus, in the present study sum scores of the 12 items were dichotomised to none/mild
depressive symptoms (0- 7 scores) and moderate/severe depressive symptoms (8-36 scores) as
recommended by Beck and Beck (1972) and as has been done in earlier studies on Finnish
adolescent population samples (e.g. Kaltiala-Heino et al. 1999a, Fröjd et al. 2006a). Adolescents
scoring in the none/mild range are referred to as “non-depressed adolescents”and adolescents
scoring in the moderate/severe range are referred to as “depressed adolescents”.
In Studies I and II term delinquency as a definition of antisocial behaviour was used because the
behaviours in question are criminal offences. The self-report questions on delinquent behaviour in
SHPS 2002 were adoptd from the Finnish Self-Reported Delinquency Study (FSRD) conducted by
the National Research Institute of Legal Policy of Finland (Kivivuori 2002a). The FSRD in turn is a
modified version of the International Self-Report Delinquency Study I (ISRD-I) instrument which
covers widely different delinquency measures (Junger-Tas 1994b). Originally the ISRD-I was used
for comparing delinquency among adolescents in 13 western countries in the 1990’s (Junger-Tas
1994b), since then it has been widely used in many studies and has been shown in test-retest studies
to have adequate reliability (Zhang et al. 2000).
In SHPS 2002 involvement in delinquent behaviour during the past 12 months was originally
measured by six domains of delinquency: graffiti writing/drawing (question 1), vandalism at school
(question 2), vandalism at other public places than school (question 3), shoplifting (question 4),
physical fighting (question 5) and beating someone up (question 6). Originally the response
categories were “never”, “once”, “two to four times”and “five or more times”. Preliminary
analysis, however, revealed that the two vandalism variables (questions 2 and 3) and the two violent
behaviour variables (questions 5 and 6) correlated (the weighted Kappa was 0.69 and 0.62
respectively). Therefore questions 2 and 3 were combined into a single variable, called vandalism,
71
and questions 5 and 6 were combined into a single variable, called violent behaviour. In Study I,
these four different domains of delinquent behaviour (graffiti writing/drawing, shoplifting,
vandalism and violent behaviour) were used in the main analyses. Response categories in these
analyses were “never”, “once to four times”and “five or more times”in the past year. In Study II,
graffiti writing/drawing was excluded from the final analyses because versatility and specialization
of delinquency was to be studied between different domains of delinquency, and vandalism and
graffiti writing/drawing were considered to be qualitatively similar to each other. Thus, in Study II
three different domains of delinquent behaviour were used in the main analyses: shoplifting,
vandalism and violent behaviour. The original response categories from the SHPS 2002 were used
to capture the versatility and specialization of delinquency.
Association between self-reported depressive symptoms and delinquent behaviour was studied by
comparing the frequencies of delinquency between depressed and non-depressed adolescents (I).
Differences in delinquent activities between depressed and non-depressed repeatedly delinquent
adolescents were assessed by comparing the frequencies of delinquency between the groups (II).
Versatility of delinquency among depressed and non-depressed adolescents was assessed according
to how many different types of delinquent acts respondents had repeatedly committed (II). For this
analysis, delinquency committed at least five or more times in the past year was considered as
repeated. Each type of delinquent acts was dichotomized to “five or more times”vs. “four or less
times”in the past year. Then a sum score of these different delinquent acts was formed to illustrate
the versatility of delinquent behaviour. The resulting categories were: “one delinquent act
repeatedly”, “two different types of delinquent acts repeatedly”and “three different types of
delinquent acts repeatedly”. Specialization in given types of delinquent acts in relation to selfreported depressive symptoms (II) was studied by classifying adolescents as “repeat specialist
shoplifters”if they had stolen at least five times in the past year but had not also committed
repeated acts of vandalism or violence over the same period. “Specialist vandals”and “specialist
violent offenders”were classified by similar procedures. Those adolescents who had repeatedly
committed several offences from all three categories were classified as “versatile delinquents”.
Age, sex, family structure (nuclear family, step-parent family, single-parent family and adolescent
living apart from parents), parental educational level (academic, vocational and basic school),
parental unemployment during the past year (none, one parent and both parents), years of living in
the same municipality (nine years or more, five to nine years, one to four years or less than one
year) and degree of urbanization of place of residence (rural, semi-urban and urban) were
72
investigated as sociodemographic variables (I). These variables were included as covariates because
earlier studies have shown them to be associated both with adolescent depression and delinquency
(Wichstrom et al. 1996, Loeber et al. 1998, Kaltiala-Heino et al. 2001).
4.1.3 Data analyses (I and II)
To study association between self-reported depressive symptoms and delinquent behaviour adjusted
by sociodemographic variables multivariate logistic regression analyses [Odds ratio (OR), 99%
confidence interval (CI)] was used (I). Delinquent behaviours were entered simultaneously as
independent variables and depressive symptoms as the dependent variable. Analyses were
conducted separately for males and females using SPSS for Windows version 11.0.
To study patterns of delinquency among depressed and non-depressed repeatedly delinquent
adolescents Pearson ²-test was used (II). Multinominal logistic regression models (OR, 95% CI)
were also used to examine the association between versatility of delinquency (dependent variable)
and self-reported depressive symptoms (independent variable). All analyses were conducted
separately for males and females. SPSS for Windows version 11.0 and Statxact version 4.0 were
used.
4.2 The Adolescent Mental Health Cohort Study (III and IV)
The Adolescent Mental Health Cohort Study (AMHC Study) is an ongoing prospective cohort study
conducted at baseline in two Finnish cities: Tampere (200, 000 inhabitants) and Vantaa (180, 000
inhabitants). The study examines epidemiology, comorbidity and risk and protective factors of nonpsychotic mental disorders in adolescence starting from 15-year-olds. The study is carried out by
the Tampere School of Public Health; Department of Mental Health, the National Public Health
Institute; Tampere University Hospital and Helsinki University Hospital. The Ethical Committee of
Tampere University Hospital approved the study. The AMHC Study will be conducted in three
waves: baseline assessment, 2-year follow-up and 5-year follow-up (Fröjd et al. 2004, 2006b). The
present study consists of the assessments at baseline (III) and 2-year follow-up (IV).
73
4.2.1 Baseline of the AMHC Study (III)
4.2.1.1 Sample and procedure (III)
The baseline study of the AMHC Study was conducted during the school term 2002 –2003.
Ninth grade pupils of all the Finnish-speaking comprehensive schools (36) in the participating cities
completed a person-identifiable questionnaire during a school lesson supervised by a teacher. The
subjects were identified in the school registers of the participating cities. The parents were informed
in advance by letter, but parental consent to participation was not required since a 15-year-old
subject can decide alone on his/her participation according to the law on medical research. Subjects
received written information on the study and signed an informed consent. For pupils absent from
school on the original survey day, a separate opportunity to participate was offered in the school
within a couple of weeks of the original data collection. For pupils not present on either occasion,
the questionnaires were sent twice by post.
The adolescents in the participating schools totalled 3, 809, of whom 3, 597 responded to the survey
(response rate 94.4%). Six respondents had to be excluded due to obvious facetiousness. Since it is
possible for an adolescent under 15 years to be attending the 9th grade, respondents who were not
yet 15 years old (n=313) were later excluded. Thus, the final sample was 3, 278 (50.9% males). The
mean age was 15.5 (range 15.0 -19.9, sd 0.39). Of the respondents, 71% were living in two-parent
families. Residential stability was high, as 67% of the respondents had been living in the same
municipality for at least 9 years. At least one of the parents had a college or university degree in the
case of 37% of respondents. In 75% of the subjects neither of the parents had been unemployed
during the past year.
4.2.1.2 Measures (III)
Self-reported depressive symptoms were assessed by the RBDI (see Chapter 4.1.2). In Study III, all
13 items of the RBDI were included in the survey. Items were scored similarly as in Studies I and
74
II, resulting a sum score ranging from 0 to 39. Sum scores were dichotomised with similar cut-off
points as in Studies I and II.
In Studies III and IV the term antisocial behaviour was used because some of the behaviours in
question were not illegal although they violate social norms and/or the rights of others, and
resemble CD and ODD. Antisocial behaviour was assessed by the externalizing scale of the Youth
Self-Report (YSR). YSR, Child Behavior Checklist (CBCL) and Teacher Report Form (TRF) are
parallel versions of symptom rating scales developed by Achenbach (1991) to assess a wide range
of the competencies, emotional and behavioural problems of children and adolescents. The YSR is
based on self-reports from adolescents aged 11-18 years. It consists of three broad-band scales of
problems (Internalizing, Externalizing and Total Problems), and it has been widely used in earlier
studies showing a good reliability and validity, and considerable consistency across different
countries (Achenbach 1991, Kvernmo and Heyerdahl 1998, Achenbach 2000, Broberg et al. 2001,
Helstelä and Sourander 2001, Verhulst et al. 2003, Hurtig et al. 2005, Kapi et al. 2007, Rescorla et
al. 2007). For example, Morgan and Cauce (1999) found that high YSR scores on the Externalizing
scale predicted a DISC-C-diagnosis of ODD or CD with reasonable sensitivity (0.74) and
specificity (0.73). The Externalizing scale consists of 11 variables concerning delinquency and 18
variables concerning aggressive behaviour during the past 6 months. Each variable is scored 0 - 2
according to the frequency of antisocial behaviour. Sum scores of the 29 items (range 0 - 58) were
dichotomised using the 90th percentile as a cut-off point into those within normal range versus those
within clinical range. Because antisocial behaviour is usually more common among adolescent
males than females, sex-specific cut-off points were used (Achenbach 1991). Cut-off points of
antisocial behaviour for males and females were 24.0 and 25.0 respectively.
Perceived social support was assessed by the Perceived Social Support Scale-Revised (PSSS-R).
The PSSS-R was presented by Blumenthal et al. (1987), and measures persons’subjective
perceptions of social support and emotional closeness, not the actual number of supportive contacts.
It has been shown to be a valid method in assessing perceived social support among Finnish
adolescents (e.g. Katainen et al. 1999). The PSSS-R contains 12 items on a five-point Likert-type
scale. Three factor-analytically derived sum scores (each ranging 4 to 20) were used addressing
perceived social support from family (4 items), from friends (4 items) and from significant other (4
items). High sum scores indicate high perceived social support. PSSS-R was used as continuous
variable in associations between perceived social support and self-reported depressive symptoms
and antisocial behaviour. In multivariate associations the PSSS-R was used as a categorical variable
75
dichotomised according to the 25th percentile into those with low level of social support versus
those with high level of social support. This was done because the distributions of the perceived
social support scores were skewed.
Sex, family structure (nuclear family, step-parent family, single-parent family or other custodian),
parental education level (academic, vocational or basic), parental unemployment during the past
year (none, one parent or both parents) and years of living in the same municipality (9 years or
more, 5-9 years, 1-4 years or less than 1 year) were investigated as sociodemographic variables.
These variables were included as covariates because other studies have shown them to be associated
with adolescent depression and antisocial behaviour (Wichstrom et al. 1996, Garnefski and Diekstra
1997, Loeber et al. 1998, Kaltiala-Heino et al. 2001, Harland et al. 2002).
4.2.1.3 Data analyses (III)
Associations between perceived social support and self-reported depressive symptoms and between
perceived social support and antisocial behaviour were analysed by Mann-Whitney test, where all
three types of perceived social support (family, friends and significant other) were analysed
separately. Univariate associations between antisocial behaviour and self-reported depressive
symptoms were analysed by Pearson ²-test and by logistic regression (Model 1). Multivariate
associations between self-reported depressive symptoms as a dependent variable and antisocial
behaviour as an independent variable were adjusted by family support, friend support and
significant other support, which were entered separately into the models (Model 2, Model 3 and
Model 4). In Model 5 three perceived social support types were entered simultaneously to
investigate the association of the antisocial behaviour for self-reported depressive symptoms.
Finally covariates were added into the model to analyse the effect of potential confounding factors
(Model 6). All analyses were conducted separately for males and females. Statistical analyses were
performed using SPSS version 11.0.
76
4.2.2 Two-year follow-up of the AMHC Study (IV)
4.2.2.1 Sample and procedure (IV)
The 2-year follow-up of the AMHC Study was conducted during the academic year 2004 –2005.
Eligible participants in Study IV (now after T2) were those adolescents who participated in the
baseline assessment of the AMHC Study (III, now after T1). Multiple approaches were used to
contact the adolescents at T2 (Fröjd et al. 2006b). School based surveys like that at T1 were
organised in upper secondary schools and vocational schools. Adolescents not reached through
schools were contacted by postal survey. Finally, an Internet based survey was offered to those who
had not yet responded via school or post.
A total of 2, 070 adolescents completed surveys at both T1 and T2. The response rate of the sample
at the follow-up was 63.1% (2, 070/3, 278). Of the respondents 43.6% (903) were males. The mean
age at T2 was 17.6 years (range 16.9 to 21.6, std. 0.41). At T2, 62.4% of the adolescents were living
in an intact family. Stable employment of the parents was high at T2, as only 6.2% of parents (one
or both) had been unemployed during the past year. Over 80% of the adolescents were studying
full-time at T2. 66.1% of the adolescents were studying at upper secondary school at T2. Stability
of residence was high, as 93.3% of the adolescents had lived in the same municipality more than
one year.
Non-participants at T2 were more likely to be males (63.4% vs. females 36.6%, p<0.001) and living
more frequently with both parents (35.2%) than those adolescents who participated in both surveys
(27.5%, p<0.001). Non-responders in the second survey were more likely to be depressed (11.7%)
at T1 than those who participated in both surveys (9.1%) (p= 0.020). Drop-outs were also more
likely to have displayed antisocial behaviour at T1 (12.3%) than participants (8.7%) (p<0.001).
77
4.2.2.2 Measures (IV)
Self-reported depressive symptoms and antisocial behaviour were assessed by the RBDI as in Study
III and by the YSR as in Study III (see Chapter 4.2.1.2). In the YSR cut-off points at T1 were for
females 23.0 (males 24.0) and at T2 for females 21.0 (males 22.0).
Sex, family structure (intact vs.non-intact), parental education level (academic vs. non-academic),
parental unemployment during the past year (no vs. one or both) and residential stability (living 10
years or more vs. up to 9 years in the same municipality) at T1 were controlled for as
sociodemographic variables. Previous studies have shown these variables to be associated with both
adolescent depression and antisocial behaviour (Loeber et al. 1998, Kaltiala-Heino et al. 2001,
Fröjd et al. 2006a).
4.2.2.3 Data analyses (IV)
Homotypic continuity of self-reported depressive symptoms and antisocial behaviour were first
analysed by Pearson χ²-test, after which logistic regression analyses were performed to control for
sociodemographic variables. To ascertain if there were significant sex differences in homotypic
continuity, interaction terms for sex and self-reported depressive symptoms/antisocial behaviour
were added into the basic model.
Concurrent comorbidity between self-reported depressive symptoms and antisocial behaviour was
analysed both at T1 and at T2. Analyses were performed similarly at both waves. The proportions
of those reporting depressive symptoms by antisocial behaviour were calculated by Pearson χ²-test.
Similarly, the proportions of those reporting antisocial behaviour by depressive symptoms were
calculated by Pearson χ²-test. Next, odds ratio (OR) with 95% confidence interval (CI) for
comorbidity between antisocial behaviour (dependent variable) and depressive symptoms
(independent variable) was performed by logistic regression analysis. To ascertain if there were
significant sex differences in comorbidity, interaction terms for sex and depressive symptoms
(independent variable) were added into the model.
78
To test whether the failure model explained longitudinal associations between self-reported
depressive symptoms and antisocial behaviour, the proportions of those reporting depressive
symptoms at T2 were calculated according to antisocial behaviour at T1 by Pearson χ²-test. Next,
crude odds ratio (OR, 95% CI) for depressive symptoms at T2 (dependent variable) according to
antisocial behaviour at T1 (independent variable) was calculated using logistic regression analysis
(Model 1). Subsequently, odds ratio for depressive symptoms at T2 by antisocial behaviour at T1
was adjusted for depressive symptoms at T1 (Model 2) and finally also for sociodemograhpic
variables (Model 3).
To test the acting out model in explaining longitudinal associations between self-reported
depressive symptoms and antisocial behaviour, the proportions those reporting antisocial behaviour
at T2 were calculated according to depressive symptoms at T1 by Pearson χ²-test. Next, crude odds
ratio (OR, 95% CI) for antisocial behaviour at T2 (dependent variable) according to depressive
symptoms at T1 (independent variable) was calculated using logistic regression analysis (Model 1).
Subsequently, odds ratio for antisocial behaviour at T2 by depressive symptoms at T1 was adjusted
for antisocial behaviour at T1 (Model 2), and finally also for sociodemographic variables (Model 3).
All analyses were conducted separately for males and females. Statistical analyses were performed
with the SPSS 13.0 software package.
5. RESULTS
5.1 Association between self-reported depressive symptoms and delinquent behaviour (I)
Graffiti writing/drawing, vandalism and violent behaviour were all independently associated with
self-reported depressive symptoms in both sexes after controlling for age and sociodemographic
factors (I: Table 1). Shoplifting was independently associated with self-reported depressive
symptoms only in females after adjustment. Among females risk for self-reported depressive
symptoms varied between 1.3 and 3.1 according to the various delinquent behaviours. The
equivalent risk among males was 1.3–2.5. Among females the strongest associations with selfreported depressive symptoms were detected for frequent vandalism (5 or more times) and among
79
males for frequent violent behaviour (5 or more times). In both sexes self-reported depressive
symptoms increased according to the frequency of delinquent behaviour.
5.2 Patterns of delinquent behaviour among depressed and non-depressed repeatedly
delinquent adolescents (II)
5.2.1 Frequency
Even among repeatedly delinquent adolescents, self-reported depressive symptoms differentiated
frequencies of delinquent behaviour. Depressed males and females had both committed more
repeated (5 or more times in the past year) shoplifting, vandalism and violent behaviour than nondepressed males and females (II: Table 1). The only exception was for the category of shoplifting,
where non-depressed females had more often stolen repeatedly than had depressed females.
5.2.2 Versatility
Analysis of the versatility of repeated delinquency revealed that most of the non-depressed
delinquent males (75.1%) had specialized in one offence type (vs. depressed delinquent males
50.1%, p< 0.001). The difference was in the same direction but less marked for non-depressed and
depressed females (87.4% vs. 77.3% respectively, p<0.001). In other words, the delinquent
behaviour of depressed adolescents was more versatile than that of non-depressed adolescents. It
was over five times more likely that depressed males had repeatedly committed offences from all
three categories of delinquent acts than had non-depressed males (OR 5.49, CI 4.26-7.07).
Depressed females were about three times more likely to have repeated offences from all categories
than non-depressed females (OR 2.78, CI 1.58-4.86) (II: Table 2).
80
5.2.3 Specialization
Non-depressed repeatedly delinquent males specialized mostly in violent offences (36.7%) (II:
Table 3). Depressed repeatedly delinquent males were most frequently classified as versatile
offenders (49.9%), but second most frequently they had also specialized in violent offences
(27.1%). Among repeatedly delinquent females specialization in shoplifting was most typical
among both depressed (41.0%) and non-depressed females (59.4%). Depressed females had more
often specialized in repeat vandalism (19.2%) than non-depressed females (11.6%), and versatile
offending was also more common among depressed females (22.7%) than among non-depressed
females (12.6%). There were no differences in specialization in violent offences between depressed
and non-depressed females (17.1% and 16.3% respectively).
5.3 Interrelations between self-reported depressive symptoms, antisocial behaviour and
perceived social support (III)
5.3.1 Association between self-reported depressive symptoms and perceived social support
Perceived social support from family, friends and significant other were inversely associated with
self-reported depressive symptoms (III: Table 1). Both depressed males and females reported
significantly less social support than did non-depressed males and females. The difference in
relation to social support between depressed and non-depressed adolescents was present in all
sources of social support, but the difference in social support was most evident in both sexes in
decreased social support from family (depressed males 11.3 vs. non-depressed males 16.6, p<0.001;
depressed females 12.4 vs. non-depressed females 16.8, p<0.001).
81
5.3.2 Association between antisocial behaviour and perceived social support
Perceived social support from family, friends and significant other were inversely associated with
antisocial behaviour among males (III: Table 1) as antisocially behaving males reported receiving
less social support from all 3 sources than non-antisocial males (13.5 vs. 16.5, 13.9 vs. 14.8 and
13.1 vs. 14.6 respectively). Among females, only perceived social support from family was
inversely associated with antisocial behaviour, as antisocially behaving females perceived less
support from family than non-antisocial females (13.5 vs. 16.6 respectively).
5.3.3 Role of perceived social support in the comorbidity between self-reported depressive
symptoms and antisocial behaviour
As expected, also in this sample antisocial behaviour was associated with self-reported depressive
symptoms in both sexes. Depressive symptoms were far more common (p<0.001) among antisocial
males (25.3%) and females (35.1%) than among non-antisocial males (4.5%) and females (11.1%).
Antisocial males were seven times (OR 7.2, CI 4.7-11.0) more likely to be depressive than nonantisocial males (III: Table 2). Similarly, antisocial females were four times more likely to be
depressive than non-antisocial females (OR 4.3, CI 3.0-6.2). Adding perceived social support
factors separately into the model modified the comorbidity between self-reported depressive
symptoms and antisocial behaviour differently. When the association between depressive symptoms
and antisocial behaviour was adjusted by social support from family (III: Table 2, Model 2), the
association of the antisocial behaviour for depressive symptoms decreased in both sexes compared
to the unadjusted associations. Social support from friends (III: Table 2, Model 3) and social
support from significant other (III: Table 2, Model 4) had an opposite effect on the association of
the antisocial behaviour for depressive symptoms, as social support adjusted associations were
slightly higher in both sexes than the unadjusted associations. When all social support factors were
entered simultaneously into the model (III: Table 2, Model 5), the associations of the antisocial
behaviour for depressive symptoms were OR 5.3 (CI 3.4-8.5) among males and OR 3.2 (CI 2.1-4.7)
among females. Adding covariates into the model (III: Table 2, Model 6) did not remarkably alter
the associations between depressive symptoms and antisocial behaviour, as the final odds ratios
were 5.5 (CI 3.3-9.1) among males and 2.9 (CI 1.9-4.5) among females. Thus, regardless of
82
perceived social support, antisocial behaviour was independently associated with self-reported
depressive symptoms.
5.4 Homotypic continuity, concurrent comorbidity and longitudinal associations between selfreported depressive symptoms and antisocial behaviour (IV)
5.4.1 Homotypic continuity
Self-reported depressive symptoms and antisocial behaviour showed considerable continuity in both
sexes during the 2-year follow-up (IV: Table 1 and Table 2). Of the males depressed at T1, 27.8%
of them were also depressed at T2. The corresponding proportion of females was 46.3%. Males
with depressive symptoms at T1 had a 7.4-fold risk (CI 3.6-15.2) for depressive symptoms at T2
after adjustment for sociodemographic variables. The corresponding odds ratio for females was 12.5
(CI 7.8-20.0). There were no significant gender differences in continuity of self-reported depressive
symptoms (p=0.108). Of the males and females reporting antisocial behaviour at T1, 43.5% of
males and 42.0% of females still displayed antisocial behaviour at T2. Males with antisocial
behaviour at T1 had a 14.6-fold risk (CI 8.3-25.6) for antisocial behaviour at T2 after adjustment for
sociodemographic variables. The corresponding odds ratio for females was 8.3 (CI 5.4-12.9).There
were no significant gender differences in continuity of antisocial behaviour (p=0.342).
5.4.2 Concurrent comorbidity
In the follow-up sample, concurrent comorbidity between self-reported depressive symptoms and
antisocial behaviour was also high (IV: Table 3). Females with antisocial behaviour at T1 were
three times more likely to be depressive than non-antisocial females (OR 3.5, CI 2.3-5.5). The
corresponding risk at T2 was OR 5.1 (CI 3.3-7.9). Similarly, males with antisocial behaviour at T1
were seven times more likely to be depressive than non-antisocial males (OR 7.3, CI 4.0-13.2). The
corresponding risk at T2 was OR 5.5 (CI 3.1-10.1). There were no significant gender differences in
83
comorbidity (at T2 p=0.836) although comorbidity at T1 almost reached statistical significance
(p=0.055), suggesting that comorbidity may be more common among males.
5.4.3 Longitudinal associations
5.4.3.1 Failure model
Antisocial behaviour at T1 predicted self-reported depressive symptoms at T2 in both sexes when
entered alone into the logistic regression analysis (IV: Table 4, Model 1), but when depressive
symptoms at T1 and sociodemographic variables were controlled for (IV: Table 4, Model 2 and
Model 3), the association between antisocial behaviour at T1 and depressive symptoms at T2 lost
significance in both sexes. In other words, the failure model did not explain the longitudinal
associations between self-reported depressive symptoms and antisocial behaviour during a 2-year
follow-up in either sex.
5.4.3.2 Acting out model
Prior self-reported depressive symptoms predicted subsequent antisocial behaviour differently
among females and males (IV: Table 5). Among females, depressive symptoms at T1 predicted
antisocial behaviour at T2 after controlling for antisocial behaviour at T1 and sociodemographic
variables. Females with depressive symptoms at T1 had a 2.3-fold risk (CI 1.3-3.9) for antisocial
behaviour at T2. Among males depressive symptoms at T1 predicted inversely antisocial behaviour
at T2 after controlling for antisocial behaviour at T1 and for sociodemographic variables. Thus prior
self-reported depressive symptoms protected males from subsequent antisocial behaviour (OR 0.4,
CI 0.1-1.0).
84
6. DISCUSSION
6.1 Associations between perceived social support and self-reported depressive symptoms and
antisocial behaviour (III)
In line with earlier studies (Garnefski and Diekstra1996, Lewinsohn et al. 1998, Helsen et al. 2000,
Kaltiala-Heino et al. 2001), social support was inversely associated with self-reported depressive
symptoms (III). Likewise consistent with earlier findings (Garnefski and Diekstra 1996, Lewinsohn
et al. 1998, Helsen et al. 2000, Kaltiala-Heino et al. 2001), the difference in relation to social
support between depressed and non-depressed adolescents was present in all sources of social
support (family, friends and significant other), but the effect of the social support from family was
most obvious in both sexes (III). Also in accordance with earlier findings (e.g. Garnefski and
Diekstra 1996, Brook et al. 1999, Bru et al. 2001) social support was inversely associated with
antisocial behaviour (III). However, the effect of social support from all sources was systematic
only among males as among females only social support from family was inversely associated with
antisocial behaviour (III). The finding parallels with the study of Bru et al. (2001) reporting that
parental support was negatively associated with antisocial behaviour in both sexes, but lack of
friend support was modestly associated with antisocial behaviour only among males. These findings
emphasise the role of parents and family in a socialisation process and as prosocial role models in
preventing antisocial behaviour. Earlier findings on the psychological closeness of antisocial
adolescents’friendships have been mixed (Deptula and Cohen, 2004). In addition, little attention
has been paid to the role of gender in the friendships of antisocial adolescents as many studies have
focused exclusively on males.
6.2 Homotypic continuity of self-reported depressive symptoms and antisocial behaviour in
middle adolescence (IV)
Consistent with earlier findings, self-reported depressive symptoms (e.g. Sanford et al. 1995,
Beyers and Loeber 2003, Costello et al. 2003) and antisocial behaviour (e.g. Fergusson et al. 1996,
Kim et al. 2003, Wiesner 2003) showed considerable homotypic continuity during the 2-year span
85
in middle adolescence (IV). Reportedly, the risk of continuity in adolescent depression is associated
with high comorbidity (AACAP 1998, Birmaher et al. 2002) which was also common in the present
study (IV). The findings emphasize the importance of prevention, early identification and efficient
treatment of depressive symptoms and antisocial behaviour, given that both of these disorders are
accompanied by range of long-term psychosocial problems and impairment (Rutter et al. 1998,
Birmaher et al. 2002). This is crucial because important and long-reaching occupational and
educational decisions are made in middle adolescence which may be adversely affected by
depressive symptoms and antisocial behaviour.
The results in the present study (IV) concur with those of earlier studies reporting no significant
gender differences either in the continuity of depressive symptoms (e.g. Ferdinand et al. 1995,
Hankin et al. 1998 ) or in the continuity of antisocial behaviour (e.g. Ferdinand et al. 1995, Costello
et al. 2003). However, since it has been reported that the continuity of severe antisocial behaviour
may be higher among females than males (Loeber et al. 2000a), it is at least partly possible that in
the present study (IV) this gender difference was not detected because of the use of the YSR which
may not measure severe antisocial behaviour rather than more common forms of antisocial
behaviour. In addition, because the present findings on continuity are based on 2-year interval
assessments during follow-up (IV), it cannot with certainty be verified whether in the continiuty of
depressive symptoms it is a case of long duration or recurrence of symptoms. However, given that
75 to 90% of depressed adolescents tend to recover from depression in 6 to 24 months (Birmaher et
al. 2002), the latter is likely to be the case. Nevertheless, the findings indicate that once they occur,
depressive symptoms and antisocial behaviour are equally persistent in both sexes.
6.3 Concurrent comorbidity between self-reported depressive symptoms and antisocial
behaviour in middle adolescence (I, II, III, IV)
Despite different rating scales of antisocial behaviour (FSRD and YSR) and different samples,
concurrent comorbidity between self-reported depressive symptoms and antisocial behaviour was
high in both sexes (I, II, III, IV). Estimates of the comorbidity between depressive symptoms and
antisocial behaviour from the present study (I, II, III, IV) fell into the range reported earlier in
studies on community adolescent samples (e.g. Loeber et al. 1998, p. 109, Flannery et al. 2001,
West et al. 2003). Adolescents with antisocial behaviour had up to seven-fold risk of being
86
depressed (IV). The finding is remarkable, given that comorbidity between depressive symptoms
and antisocial behaviour is reportedly a potential risk factor for increased psychosocial impairment
in multiple domains, both in adolescence and in adulthood (Wu et al. 1999, Fombonne et al. 2001a
and 2001b, Marmorstein and Iacono 2001 and 2003). Thus, research on comorbidity has
implications for both the prevention and treatment of both depressive symptoms and antisocial
behaviour.
The comorbidity between depression and antisocial behaviour has previously been reported
especially in clinical and institutionalized samples (e.g. Herkov and Myers 1996, Pliszka et al.
2000, Vermeiren et al. 2000, Greene et al. 2002). However, studies focusing specifically on the
association between delinquent behaviour and depression in a general adolescent population are
rare. The present study (I and II) focused especially on this topic. Studies based on general
adolescent population are needed because of the possible limitations of the generalizability of the
findings conducted among incarcerated or referred adolescents. Delinquently behaving adolescents
in the community may not yet have come to the attention of parents, teachers and mental health
professionals although at least a subset of those may have depressive symptoms requiring
immediate intervention. In addition, most of the earlier studies have been conducted in the United
States or in the UK, in continental Europe and especially in the Nordic countries only few studies
have focused on this topic in general community samples. Because cultural and social differences
may influence the association between disorders, it is essential to assess comorbidity in different
countries and in different cultures. Reportedly, e.g. differences in lethal violence between European
and American adolescents are substantial (Rutter et al. 1998, p. 40, Savolainen et al. 2000), and it is
still uncertain whether such factors interact with comorbidity between depression and delinquent
behaviour. Further, it is unsure whether rates of depression in American samples are elevated given
that even 7-year-old children can be convicted in the American judicial system.
Although a number of studies have reported inverse associations both between social support and
depression/antisocial behaviour (e.g. Garnefski and Diekstra 1996, Peiser and Heaven 1996,
Lewinsohn et al.1998), apparently no study has so far investigated whether social support mediates
the comorbidity between depression and antisocial behaviour or if the comorbidity is independent
of social support. The present study (III) addressed this gap in the research by controlling for the
effect of perceived social support on comorbidity. The results indicated that, although perceived
social support from family, friends and significant others modified comorbidity between selfreported depressive symptoms and antisocial behaviour, antisocial behaviour was still
87
independently associated with depressive symptoms (III). In other words, perceived social support
did not explain the comorbidity identified between depressive symptoms and antisocial behaviour
(III). In fact, the results indicated that antisocial adolescents are even more vulnerable to depressive
symptoms if they simultaneously lack support from friends (III). However, it may also be that
depressed antisocially behaving adolescents push away offers of social support or that they do not
realise that they actually have social support as a part of feeling of loneliness. On the other hand, the
measure of social support used in this study measures explicitly adolescents’subjective perceptions
of social support and emotional closeness, not actual number of supportive contacts. Therefore, the
adolescents’subjective perceptions of decreased support are valuable.
Consistent with some earlier studies (e.g. Kovacs et al. 1988, Mitchell et al. 1988, Domalanta et al.
2003), no significant gender differences in the comorbidity between self-reported depressive
symptoms and antisocial behaviour were found (IV). The finding differs from most earlier studies
(e.g. Flannery et al. 2001, Overbeek et al. 2001, McCabe et al. 2002, Vermeiren 2003) suggesting
that depression is more prevalent among antisocial females than males. Differences in the findings
may be a result of different samples as e.g. Flannery et al. (2001) focused on seriously violent
adolescents whereas, the present study (IV) investigated antisocial behaviour in a general school
sample. Disparity of assessment methods and variations in participants’ages may also contribute to
the difference.
6.4 Patterns of delinquent behaviour among depressed and non-depressed adolescents in
middle adolescence (I, II)
Little is known about how adolescent depression is associated with the various forms of
delinquent/antisocial behaviour. In the present study the risk for self-reported depressive symptoms
varied according to the various forms of delinquent behaviours (I). Among females, the strongest
association was found in frequent vandalism and among males in frequent violence (I). The findings
concur with the study by Rey et al. (2005) reporting depression to be more prevalent among
aggressive than non-aggressive delinquents, but does not corroborate with the study of Simic and
Fombonne (2001) reporting that adolescents with comorbid depression and antisocial behaviour are
less likely to destroy and to be less overtly aggressive than those with pure antisocial behaviour.
Previously, Chiles et al. (1980) and Hämäläinen and Haapasalo (1996) found no differences in
88
patterns of delinquency in relation to depression. However, different definitions of antisocial
behaviour/delinquency complicate the conclusions.
In the present study the risk for self-reported depressive symptoms increased according to the
increased frequency of delinquency (I). It is specially noteworthy that even among repeatedly
delinquent adolescents, self-reported depressive symptoms still differentiated frequencies of
delinquency as repeat delinquency was far more common among depressed adolescents than among
non-depressed adolescents (II). The reason for the exception that repeat shoplifting was more
common among non-depressed girls (II) needs further study. The finding of increased intensity
parallels with the studies by Loeber et al. (1994) and Vermeiren et al. (2002a) reporting depression
to be associated with escalation to more serious antisocial behaviour, but does not concur with the
study by Simic and Fombonne (2001). However, although rates of self-reported depressive
symptoms increased as a function of increased delinquency, it is worth noting that the risk for
depressive symptoms was increased even when there had been one single offence (I).
Earlier studies focusing on the possible differences between specialization and versatility of
delinquency according to depression are scarce. In the current study specialization in one
delinquency type was significantly more common among non-depressed repeatedly delinquent
adolescents than their depressed counterpairs (II). Earlier, specialization into one pattern of
antisocial behaviour has been reported to be rare in adolescence (Gottfredson and Hirschi 1990, p.
91, Loeber and Stouthamer-Loeber 1998, Lahey et al. 1999a). Again, a number of methodological
factors may contribute to this difference. The findings of the current study may at least partially
reflect the relatively broad delinquency categories in the questionnaire used (Kyvsgaard 2003, p.
148). When specializing in one type of delinquency, both depressed and non-depressed delinquent
males specialized mainly in repeat violent offences (II) which concurs with earlier studies reporting
that in general, boys’antisocial behaviour is overt and aggressive by the nature (Kann and Hanna
2000, Loeber et al. 2000a, Bassarath 2001). Among repeatedly delinquent females specialization in
shoplifting was the most typical form among both depressed and non-depressed females (II) which
also concurs with earlier studies reporting that in general, females’antisocial behaviour is likely to
be covert. Specialization in repeat vandalism was more frequent among depressed females than
their non-depressed counterparts (II). Interestingly, self-reported depressive symptoms did not
differentiate specialization in violent offences among females (II). Further study is needed to
explain and replicate these findings.
89
The result indicating that the delinquent behaviour of depressed adolescents was more versatile than
non-depressed adolescents was new (II). The versatile offending by depressed adolescents may be
at least partly explained by anger and irritability, which are often associated with depression
(Herkov and Myers 1996, Weisbrot and Ettinger 2002). These feelings could predispose depressed
adolescents to misbehave in various ways rather than distinguishing between different types of
offences. Further research on the exact mechanisms leading to the finding is needed. Nevertheless,
the finding of versatile offending among depressed adolescents is worrying, given that in general,
the more versatile antisocial behaviour is, the poorer outcome is to be expected (McMahon and
Frick 2005).
6.5 Longitudinal associations between self-reported depressive symptoms and antisocial
behaviour in middle adolescence (IV)
There are surprisingly few studies on longitudinal associations between depression and antisocial
behaviour in adolescence, and the findings are mixed. The present study addressed this gap in the
research by investigating simultaneously two different developmental models in a large community
sample with both sexes (IV). Studying longitudinal associations between depression and antisocial
behaviour yields important implications for the diagnostic classifications, etiology and course for
both disorders, and therefore for prevention and for interventions regarding both disorders
(Overbeek et al. 2001, Hankin 2006).
In contrast to several other studies (e.g. Capaldi 1992, Feehan et al. 1993, Kiesner 2002, Fergusson
et al. 2003, Nock et al. 2006), the failure model did not explain the longitudinal associations
between self-reported depressive symptoms and antisocial behaviour during follow-up in either
sexes (IV). However, most studies have focused on the transition from late childhood to early
adolescence (e.g. Capaldi 1992) or they have followed up participants from childhood to young
adulthood (e.g. Fergusson et al. 2003). The present study covered a 2-year period in middle
adolescence. The relatively short follow-up may account for the non-realisation of failure model in
the present study. On the other hand, developmental associations between depressive symptoms and
antisocial behaviour may depend on the time when antisocial behaviour begins (Lewinsohn et al.
1995b), and failure experiences triggered by antisocial behaviour may be age-related and differ
90
across developmental stages (Wiesner 2003). The present results suggest that the failure model is
not adequate in middle adolescence (IV).
The present study demonstrated that self-reported depressive symptoms predicted subsequent
antisocial behaviour in a 2-year span, but the acting out model held true only among females (IV).
The acting out model has previously been supported by some studies (e.g. Kovacs et al. 1988,
Loeber et al. 1994). In contrast to the finding for females, a history of depressive symptoms seemed
to protect males against subsequent antisocial behaviour (IV), which concurs with the findings of
Leadbeater et al. (1999) and Vermeiren et al. (2002b). Although the precise mechanism underlying
this effect remains to be identified, it has been supposed that subsequent antisocial behaviour may
be reduced due to the apathy and diminished energy associated with depression or that depressive
symptoms and experiences of guilt/shame may indicate the potential for reflecting on one’s actions
and their consequences for others (Vermeiren et al. 2002b).
Gender comparisons in longitudinal associations between depression and antisocial behaviour have
so far been relatively uncommon. Regarding the acting out model (IV), there are at least two
possible explanations. Firstly, the developmental course of antisocial behaviour seems to differ by
sex, as boys generally begin antisocial behaviour earlier than girls do (Silverthorn and Frick 1999).
Therefore, the longitudinal association between depressive symptoms and antisocial behaviour may
emerge earlier among boys and be non-apparent in middle adolescence. Secondly, there is some
evidence that acting out depression is likely to be associated especially with a family history of
depression rather than a family history of antisocial behaviour, as in the case of the failure model
(Karnik et al. 2006), and that females are more vulnerable to parental depression than males
(Crawford et al. 2001). Obviously, because the evidence supporting this model is limited, and
because the present study did not investigate gender differences in risk factors for depressive
symptoms, the proposed mechanism needs to be verified in future research.
6.6 Limitations of the study
A major limitation of the present study is the lack of diagnostic interviews and lack of additional
information (e.g. police records) as both depressive symptoms and antisocial behaviour were
measured by self-report scales. Exclusive reliance on self-reports may have influenced the results.
91
However, the measures used in this study are widely used in depression and antisocial behaviour
research and have demonstrated good psychometric properties and ability to predict DSM diagnosis
(Beck et al. 1974, Achenbach 1991, Junger-Tas 1994b, Bennett et al. 1997, Kaltiala-Heino et al.
1999b, Morgan and Cauce 1999, Helstelä and Sourander 2001, Fountoulakis et al. 2003). In
addition, because hidden criminality is estimated to be considerable, and official statistics cover
only a small fraction of all antisocial behaviour, self-report methods offer more extensive picture of
the whole phenomenon. In addition to this, anonymous responses, as in Studies I and II, are
considered to increase reliability in the assessment of antisocial behaviour (Kivivuori 2005a).
Lastly, self-report methods provide valuable information on adolescents’subjective perceptions of
their own well-being and behaviour.
Another limitation in data collection was that different versions of the RBDI were used in Studies I
and II (12 items) vs. III and IV (13 items). Moreover, as the RBDI measures depressive symptoms
at a given point in time (today), the duration of symptoms or the number of depressive episodes
could not be assessed. It can therefore only be assumed that adolescents reporting depressive
symptoms had been also depressed during the assessment of antisocial behaviour. In the light of
earlier studies this assumption is likely to be valid, as adolescent depression tends to be persistent
(e.g. Rutter et al. 2006). However, as the follow-up period in Study IV was relatively short, more
research on the long-term course of depressive symptoms and antisocial behaviour and on
longitudinal associations between disorders is needed.
It is also important to acknowledge that due to the cross-sectional study design in Studies I, II and
III, no firm conclusions about causality between self-reported depressive symptoms and antisocial
behaviour could be drawn from these samples. However, these findings also indicate a robust
association between depressive symptoms and antisocial behaviour.
Because those adolescents who were absent from school on the survey day were not contacted in
Studies I and II, it is possible that the rates of self-reported depressive symptoms and delinquent
behaviour may have been underestimated given that those adolescents may be more depressed and
behave more delinquently than other pupils. A further limitation is that the response rate in the 2year follow-up sample (IV) was modest. Given that study dropouts are more likely to have
psychiatric disorders than those attending the study (Oldenhinkel et al. 1999), and only limited
information about non-participants was available at follow-up, the results may not be generalizable.
Similarly, given that both depression and antisocial behaviour reportedly have considerable
92
continuity in adolescence (e.g. Rutter et al. 2006), the rates of self-reported depressive symptoms
and antisocial behaviour may also be underestimated at follow-up. However, the associations found
between these disorders are not likely to be affected by non-participants. A further limitation in the
follow-up sample is that in studying homotypic continuity of antisocial behaviour no distinction
was made between adolescents with violent antisocial behaviour vs. adolescents with non-violent
antisocial behaviour. Reportedly, this differentation would have been valid as violence and
persistent delinquency are strongly associated (Rutter et al. 1998, p. 105).
Because of the anonymous responses in Studies I and II, no attrition analysis was conducted. In
Study III attrition analysis was not conducted as only very limited information (age) about nonparticipants at baseline assessment was available. In Study IV attrition analysis was conducted.
Finally, the results from Studies III and IV may not be representative of the whole Finnish
adolescent population since none of these adolescents came from rural areas.
6.7 Strengths of the study
Most of the studies investigating comorbidity between depression and antisocial behaviour have
been conducted in the United States or in the UK, and many of them have been conducted among
referred or institutionalized male adolescents, which impairs the generalizability of the results. The
key strength of the current study is that a large number of females and males made it possible to
study gender differences. This is important, because although studies on antisocial behaviour based
on female samples have increased in past years, there is still relatively little psychiatric research on
antisocial females (Maughan and Rutter 2001). Similarly, the use of a community samples enables
the extension of previous findings conducted in more selective populations into general adolescent
population in Nordic countries. Regarding longitudinal associations between self-reported
depressive symptoms and antisocial behaviour, the large longitudinal sample made it possible to
test two models simultaneously. Most of the studies so far have focused on only one developmental
model.
The sample of Studies I and II were geographically representative of Finnish adolescent population
as they consisted different sized urban and rural municipalities representing adolescents from
different living environments. Lastly, the response rate in Studies I, II and III was excellent.
93
7. CONCLUSIONS
Based on the results of this study the following conclusions can be drawn.
1. Social support from family is essential in buffering adolescents against depressive
symptoms and antisocial behaviour.
2. Depressive symptoms and antisocial behaviour have a considerable continuity in middle
adolescence in both sexes. The finding emphasizes the importance of the prevention, early
identification and efficient treatment of disorders.
3. Adolescents with antisocial behaviour are more likely to be depressed than non-antisocial
adolescents. The finding of comorbidity between depressive symptoms and antisocial
behaviour previously found mainly among institutionalized males was confirmed in the
current study on males and females from a general school population.The finding is
important, given that adolescents with co-occurring depressive symptoms and antisocial
behaviour are at increased risk for diverse psychosocial impairments both in later
adolescence and in adulthood.
4. The risk for depressive symptoms varies and increases according to the types and the
frequencies of delinquent behaviour. Among females the strongest association is in frequent
vandalism and among males in frequent violence.
5. Depressive symptoms differentiate patterns of delinquent behaviour. Delinquent behaviour
in depressed adolescents is more versatile than in non-depressed adolescents. The finding is
worrying, given that versatile antisocial behaviour is associated with the poorest outcome.
6. The course of antisocial behaviour is altered by the presence of depressive symptoms.
Depressed females are at risk of subsequent antisocial behaviour. Former depressive
symptoms protect males from subsequent antisocial behaviour.
8. CLINICAL IMPLICATIONS
The findings of this study have many implications for both the prevention and treatment of both
depressive symptoms and antisocial behaviour. Firstly, the early identification of those adolescents
with decreased social support is crucial to prevent depressive symptoms and antisocial behaviour.
94
Because social support from family seems to be especially important in preventing both depressive
symptoms and antisocial behaviour, it is important to identify families at risk and provide them with
adequate support. Secondly, based on the earlier research it is already well acknowledged that both
depression and antisocial behaviour have considerable continuity in adolescence, but besides this,
the present study detected no significant gender differences either in the continuity of depressive
symptoms or in the continuity of antisocial behaviour. Thus, the findings indicate that once they
have emerged, depressive symptoms and antisocial behaviour are at least as persistent in both sexes.
Therefore, targeted interventions should be initiated as soon as possible before the psychosocial
adjustment of later adolescence and adulthood is impaired.
Thirdly, given that concurrent comorbidity between depressive symptoms and antisocial behaviour
was also remarkable in a general school population sample, it is crucial that adolescents presenting
with one of these disorders are carefully assessed also for the presence of the other disorder. Health
care professionals have a special challenge in recognizing these multiproblem adolescents.
Identification of comorbid antisocial behaviour may be difficult if attention is paid solely to
externalizing symptoms. Efficient collaboration between parents, teachers, social workers and
police is pivotal in detecting and in treating antisocially behaving adolescents. Special attention
should be paid to those adolescents who repeatedly behave delinquently because they are at
considerable risk for depressive symptoms.
Fourthly, the findings that depressive symptoms differentiate patterns of delinquent behaviour have
implications for planning crime-prevention programmes, in developing treatment programmes for
repeatedly delinquently behaving adolescents and in detecting depressive symptoms among
delinquent adolescents. The risk for depressive symptoms increases as a function of increased
delinquent behaviour. Other potential warningsigns for depressive symptoms are versatile
delinquent behaviour and specialization in violence among males and vandalism among females.
Finally, the course of antisocial behaviour is altered by the presence of depressive symptoms. The
finding that depressive symptoms precede antisocial behaviour among females indicates that to
prevent antisocial behaviour, interventions should not focus solely on antisocial behaviour, but also
on depressive symptoms.
95
9. RESEARCH IMPLICATIONS
While depressive symptoms and antisocial behaviour were associated in a significant and robust
manner in the current study, it is noteworthy that some of the antisocially behaving adolescents
were not depressed. It seems that some adolescents behave antisocially without becoming
depressed, and behave antisocially for other reasons than acting out depression. In the future,
further research will be needed on how depressed and non-depressed antisocially behaving
adolescents differ from one other in terms of social and psychosocial background factors and
antisocial causation (e.g. shared risk factors).
Little is known about how adolescent depression is associated with the various forms of antisocial
behaviour. This study indicated that repeat delinquency was far more common among depressed
adolescents than among their non-depressed counterparts, but the reason for the exception that
repeat shoplifting was more common among non-depressed girls needs further study. Similarly,
more research is needed on why depressive symptoms did not differentiate specialization in violent
offences among females as it did among males. The finding indicating that delinquent behaviour in
depressed adolescents was more versatile than in non-depressed adolescents was new. Replication
of the findings and further research on the exact mechanisms are needed.
There are only few studies focusing on the developmental associations between depression and
antisocial behaviour in general adolescent population, gender comparisons are scarce, and need
further study. Although studies on antisocial behaviour based on female samples have increased in
past years, there is still relatively little psychiatric research on antisocial females.This study yielded
evidence for the acting out model among females, but more research is needed to identify the
precise mechanism underlying the finding that a history of depressive symptoms protected males
from subsequent antisocial behaviour. Similarly, another important aim for future studies will be to
ascertain why this protective effect of depressive symptoms did not operate among females. Finally,
as the present results of longitudinal associations were based on a period in middle adolescence, it
would be interesting to investigate whether depressive symptoms also predict antisocial behaviour
in adulthood and whether the gender differences detected persist.
96
10. ACKNOWLEDGEMENTS
This study was carried out at the University of Tampere, School of Public Health and Tampere
University Hospital, Department of Adolescent Psychiatry. Financial support was received from the
Competitive Research Funding of the Pirkanmaa Hospital District, from the Yrjö Jahnsson
Foundation and from the Finnish Cultural Foundation.
I owe my sincere gratitude to Adjunct Professor Riittakerttu Kaltiala-Heino for supervising this
study and for teaching me psychiatric research. Ever since my master’s thesis, our cooperation has
been inspiring and equal. Without her patient guidance and encouragement this study would never
have been completed.
I also want to express my warm thanks to the members of the follow-up group, Professor Mauri
Marttunen and Adjunct Professor Mirjami Pelkonen for their valuable support and for co-authoring
articles. Also other co-authors, Research Professor Matti Rimpelä, Director of the Criminological
Unit of the National Research Institute of Legal Policy, Janne Kivivuori and Professor Bettina von
der Pahlen merit my sincere gratitude. It has been pleasure to collaborate.
I wish to express my profound gratitude to the official reviewers of this study, Adjunct Professor
Kirsi Suominen and Professor Pirkko Räsänen, for their careful review and constructive comments
which helped me to improve the final version of the study.
I warmly thank the members of research team of the Social Psychiatry Unit, especially Professor
Matti Joukamaa and Teaching Coordinator Helena Rantanen for providing the facilities of the
institution and for their interest in my research work. I am most grateful to Anna-Maija Koivisto,
Tiina Luukkaala and Heini Huhtala for their help with statistical problems. Virginia Mattila has
patiently revised the language of the study and I owe my best thanks to her.
Members of the research team of the Adolescent Mental Health Cohort Study merit my thanks for
productive and inspiring discussions. The distress of studying is easier when shared!
I am grateful to my superior Maarit Selander, who made leave of absence from work possible and
gave me her encouragement. Former Chief Psychiatrist Päivi Rantanen deserves my thanks for her
97
comments and interest in my research work. I also want to thank all the clinical teams on wards
NPS1 and NPS2 in the Department of Adolescent Psychiatry. I truly appreciate the understanding
shown towards my absence from work.
I am deeply indebted to my parents, Pirkko and Jorma Jäkälä, for always supporting me in many
ways in my academic goals. My deepest thanks and love are dedicated to my beloved husband,
Mikko. He provides me with much else to think about besides research work and he has patiently
tolerated my absent-mindedness when physically present. I will also always remember the
beginning of this study, which was associated with a humorous chat about depressed foxes in the
sauna by Lake Jäkälä. Since then, too, Mikko has provided me with critical comments and
encouragement, and has made writing at home possible. Our sweet children, Lauri and Lotta, have
literally grown up during this research process. Henceforth Mom will be around more.
Tampere, 2008
Minna Ritakallio
98
11. REFERENCES
Aalberg V and Siimens MA (1999): Lapsesta aikuiseksi. Nuoren kypsyminen naiseksi tai mieheksi.
Gummerus Kirjapaino Oy, Jyväskylä.
Aalsma MC and Lapsley DK (2001): A typology of adolescent delinquency: sex differences and
implications for treatment. Crim Behav Ment Health 11:173-191.
Aalto-Setälä T, Marttunen M, Tuulio-Henriksson A, Poikolainen K and Lönnqvist J (2001): Onemonth prevalence of depression and other DSM-IV disorders among young adults. Psychol Med
31:791-801.
Achenbach TM (1991): Manual for the Youth Self-Report and 1991 Profile. VT: University of
Vermont Department of Psychiatry, Burlington.
Achenbach TM (2000): Child Behavior Checklists (CBCL/2-3 and CBCL/4-18), Teacher Report
Form (TRF), and Youth Self Report (YSR). In: Handbook of Psychiatric Measures, pp. 310-314.
Eds. AJ Rush, HA Pincus, MB First and D Blacker, American Psychiatric Association (APA),
Washington DC.
Achenbach TM, Howell CT, McConaughy SH, Stanger C (1998): Six-Year Predictors of Problems
in a National Sample: IV. Young Adult Signs of Disturbance. J Am Acad Child Adolesc Psychiatry
37:718-727.
American Academy Child and Adolescent Psychiatry (AACAP) (1997): Practice Parameters for the
Assessment and Treatment of Children and Adolescents With Conduct Disorder. J Am Acad Child
Adolesc Psychiatry 36:122S-139S.
American Academy Child and Adolescent Psychiatry (AACAP) (1998): Practice Parameters for the
Assessment and Treatment of Children and Adolescents With Depressive Disorders. J Am Acad
Child Adolesc Psychiatry 37:63S-83S.
99
American Psychiatric Association (APA) (1994): Diagnostic and Statistical Manual of Mental
Disorders, 4th edition (DSM-IV). American Psychiatric Association, Washington DC.
American Psychiatric Association (APA) (2000): Diagnostic and Statistical Manual of Mental
Disorders, 4th edition. Text Revision (DSM-IV-TR). American Psychiatric Association, Washington
DC.
Angold A and Costello EJ (2001): The epidemiology of disorders of conduct: nosological issues
and comorbidity. In: Conduct disorders in childhood and adolescence, pp. 126-168. Eds. J Hill and
B Maughan, Cambridge University Press, New York.
Angold A and Costello EJ (2006): Puberty and Depression. Child Adolesc Psychiatric Clin N Am
15:919-937.
Angold A, Prendergast M, Cox A, Harrington R, Simonoff E and Rutter M (1995): The Childhood
and Adolescent Psychiatric Assessment (CAPA). Psychol Med 25:739-754.
Angold A, Costello EJ and Worthman CM (1998): Puberty and depression: the roles of age,
pubertal status and pubertal timing. Psychol Med 28:51-61.
Angold A, Costello EJ and Erkanli A (1999): Comorbidity. J Child Psychol Psychiatry 40:57-87.
Apter A, Gothelf D, Orbach I, Weizman R, Ratzoni G, Har-even D and Tyano S (1995): Correlation
of Suicidal and Violent Behavior in Different Diagnostic Categories in Hospitalized Adolescent
Patients. J Am Acad Child Adolesc Psychiatry 34:912-918.
Arnett JJ (1999): Adolescent Storm and Stress, Reconsidered. Am Psychol 54:317-326.
Arseneault L, Moffitt TE, Caspi A, Taylor PJ, Silva PA (2000): Mental Disorders and Violence in a
Total Birth Cohort: Results From the Dunedin Study. Arch Gen Psychiatry 57:979-986.
Bassarath L (2001): Conduct Disorder: A Biopsychological Review. Can J Psychiatry 46:609-617.
100
Beardslee WR, Versage EM and Gladstone TR (1998): Children of affectively ill parents: a review
of the past 10 years. J Am Acad Child Adolesc Psychiatry 37:1134-1141.
Beck AT and Beck RW (1972): Screening Depressed Patients in Family Practice. A Rapid Technic.
Postgrad Med 52:81-85.
Beck AT, Rial WY and Rickels K (1974): Short form of depression inventory: cross-validation.
Psychol Rep 34:1184-1186.
Beck AT, Steer RA and Garbin MC (1988): Psychometric properties of the Beck Depression
Inventory: twenty-five years of evaluation. Clin Psychol Rev 8:77-100.
Bendixen M, Endresen IM and Olweus D (2003): Variety and frequency scales of antisocial
involvement: Which one is better? Leg Crim Psychology 8:135-150.
Bennett DS, Ambrosini PJ, Bianchi M, Barnett D, Metz C and Rabinovich H (1997): Relationship
of Beck Depression Inventory factors to depression among adolescents. J Affect Disord 45:127-134.
Bennett DS, Ambrosini PJ, Kudes D, Metz C and Rabinovich H (2005): Gender differences in
adolescent depression: Do symptoms differ for boys and girls? J Affect Disord 89:35-44.
Beyers JM and Loeber R (2003): Untangling Developmental Relations Between Depressed Mood
and Delinquency in Male Adolescents. J Abnorm Child Psychol 31:247-266.
Biederman J, Faraone S, Mick E and Lelon E (1995): Psychiatric Comorbidity among Referred
Juveniles with Major Depression: Fact or Artifact. J Am Acad Child Adolesc Psychiatry 34:579590.
Birmaher B, Ryan ND, Williamson DE, Brent DA, Kaufman J, Dahl RE, Perel J and Nelson B
(1996): Childhood and Adolescent Depression: A Review of the Past 10 Years. Part I. J Am Acad
Child Adolesc Psychiatry 35:1427-1439.
Birmaher B, Arbelaez C and Brent D (2002): Course and outcome of child and adolescent major
depressive disorder. Child Adolesc Psychiatric Clin N Am 11:619-637.
101
Blumenthal JA, Burg MM, Barefood J, Williams RB, Haney T and Zimet G (1987): Social support,
type A behavior, and coronary artery disease. Psychosom Med 49:331-340.
Boyd CP, Kostanski M, Gullone E, Ollendick TH and Shek DTL (2000): Prevalence of Anxiety and
Depression in Australian Adolescents: Comparisons With Worldwide Data. J Genet Psychol
161:479-492.
Boyle MH, Offord DR, Racine YA, Szatmari P, Sanford M and Fleming JE (1997): Adequacy of
Interviews vs Checklists for Classifying Childhood Psychiatric Disorder Based on Parent Reports.
Arch Gen Psychiatry 54:793-799.
Brendgen M, Vitaro F and Bukowski WM (2000): Deviant Friends and Early Adolescents’
Emotional and Behavioral Adjustment. J Res Adolesc 10:173-189.
Broberg AG, Ekeroth K, Gustafsson PA, Hansson K, Hägglöf B, Ivarsson T and Larsson B (2001):
Self-reported competencies and problems among Swedish adolescents: a normative study of the
YSR. Eur Child Adolesc Psychiatry 10:186-193.
Brook JS, Brook DW, De La Rosa M, Whiteman M and Montoya I (1999): The Role of Parents in
Protecting Colombian Adolescents From Delinquency and Marijuana Use. Arch Pediatr Adolesc
Med 153:457-464.
Bru E, Murberg TA and Stephens P (2001): Social support, negative life events and pupil
misbehaviour among young Norwegian adolescents. J Adolesc 24:715-727.
Burke JD, Loeber R and Birmaher B (2002): Oppositional Defiant Disorder and Conduct Disorder:
A Review of the Past 10 Years, Part II. J Am Acad Child Adolesc Psychiatry 41:1275-1293.
Burke JD, Loeber R, Lahey BB and Rathouz PJ (2005): Developmental transitions among affective
and behavioral disorders in adolescent boys. J Child Psychol Psychiatry 46:1200-1210.
Cantwell DP, Lewinsohn PM, Rohde P and Seeley JR (1997): Correspondence between adolescent
report and parent report of psychiatric diagnostic data. J Am Acad Child Adolesc Psychiatry
36:610-619.
102
Capaldi DM (1991): Co-occurrence of conduct problems and depressive symptoms in early
adolescent boys: I. Familial factors and general adjustment at Grade 6. Dev Psychopathol 3:277300.
Capaldi DM (1992): Co-occurrence of conduct problems and depressive symptoms in early
adolescent boys: II. A 2-year follow-up at Grade 8. Dev Psychopathol 4:125-144.
Capaldi DM and Stoolmiller M (1999): Co-occurrence of conduct problems and depressive
symptoms in early adolescent boys: III. Prediction to young-adult adjustment. Dev Psychopathol
11:59-84.
Caron C and Rutter M (1991): Comorbidity in Child Psychopathology: Concepts, Issues and
Research Strategies. J Child Psychol Psychiatry 32:1063-1080.
Chabrol H, Rodgers R and Rousseau A (in press): Relations between suicidal ideation and
dimensions of depressive symptoms in high-school students. J Adolesc.
Chiles J, Miller M and Cox G (1980): Depression in an adolescent delinquent population. Arch Gen
Psychiatry 37:1179-1184.
Christie D and Vinter R (2005): Adolescent development. BMJ 330:301-304.
Connelly B, Johnston D, Brown ID, Mackay S and Blackstock EG (1993): The prevalence of
depression in a high school population. Adolescence 28:149-158.
Costello E J, Angold A, Burns BJ, Stangl DK, Tweed D, Erkanli A and Worthman C (1996): The
Great Smoky Mountains Study of Youth: Goals, Design, Methods, and the Prevalence of DSM-IIIR Disorders. Arch Gen Psychiatry 53:1129-1136.
Costello EJ, Erkanli A, Federman E and Angold A (1999): Development of Psychiatric
Comorbidity With Substance Abuse in Adolescents: Effects of Timing and Sex. J Clin Child
Psychol 28:298-311.
103
Costello EJ, Pine DS, Hammen C, March JS, Plotsky PM, Weissman MM, Biederman J, Goldsmith
HH, Kaufman J, Lewinsohn PM, Hellander M, Hoagwood K, Koretz DS, Nelson CA and Leckman
JF (2002): Development and Natural History of Mood Disorders. Biol Psychiatry 52:529-542.
Costello EJ, Mustillo S, Erkanli A, Keeler G and Angold A (2003): Prevalence and Developmnet of
Psychiatric Disorders in Childhood and Adolescence. Arch Gen Psychiatry 60:837-844.
Costello EJ, Foley DL and Angold A (2006): 10-Year Research Update Review: The Epidemiology
of Child and Adolescent Psychiatric Disorders: II. Developmental Epidemiology. J Am Acad Child
Adolesc Psychiatry 45:8-25.
Cote S, Zoccolillo M, Tremblay RE, Nagin D and Vitaro F (2001): Predicting Girls’Conduct
Disorder in Adolescence From Childhood Trajectories of Disruptive Behaviors. J Am Acad Child
Adolesc Psychiatry 40:678-684.
Crawford TN, Cohen P, Midlarsky E and Brook JS (2001): Internalizing Symptoms in Adolescents:
Gender Differences in Vulnerability to Parental Distress and Discord. J Res Adolesc 11:95-118.
Cyranowski JM, Frank E, Young E and Shear MK (2003): Adolescent Onset of the Gender
Difference in Lifetime Rates of Major Depression: A Theoretical Model. Arch Gen Psychiatry
57:21-27.
Deptula DP and Cohen R (2004): Aggressive, rejected, and delinquent children and adolescents: a
comparison of their friendships. Agg Viol Behav 9:75-104.
Dick DM, Viken RJ, Kaprio J, Pulkkinen L and Rose RJ (2005): Understanding the Covariation
Among Childhood Externalizing Symptoms: Genetic and Environmental Influences on Conduct
Disorder, Attention Deficit Hyperactivity Disorder, and Oppositional Defiant Disorder Symptoms. J
Abnorm Child Psychol 33:219-229.
Dodge KA and Coie JD (1987): Social-Information-Processing Factors in Reactive and Proactive
Aggression in Children’s Peer Groups. J Pers Soc Psychol 53:1146-1158.
104
Dodge KA and Pettit GS (2003): A Biopsychosocial Model of the Development of Chronic
Conduct Problems in Adolescence. Dev Psychol 39:349-371.
Domalanta DD, Risser WL, Roberts RE and Risser JM (2003): Prevalence of Depression and Other
Psychiatric Disorders Among Incarcerated Youths. J Am Acad Child Adolesc Psychiatry 42:477484.
Donovan JE, Jessor R and Costa FM (1988): Syndrome of Problem Behavior in Adolescence: A
Replication. J Consult Clin Psychol 56:762-765.
Duclos CW, Beals J, Novins DK, Martin C, Jewett CS and Manson S (1998): Prevalence of
common psychiatric disorders among American Indian adolescent detainees. J Am Acad Child
Adolesc Psychiatry 37:866-873.
Earls F and Mezzacappa E (2002): Conduct and Oppositional Disoders. In: Child and Adolescent
Psychiatry, pp. 419-436. Eds. M Rutter and E Taylor, Blackwell Science, Cornwall.
Elonheimo H, Niemelä S, Parkkola K, Multimäki P, Helenius H, Nuutila A and Sourander A
(2007): Police-registred offences and psychiatric disorders among young males. Soc Psychiatry
Psychiatr Epidemiol 42:477-484.
Ezpeleta L, Granero R and Doménech JM (2005): Differential contextual factors of comorbid
conduct and depressive disorders in Spanish children. Eur Child Adolesc Psychiatry 14:282-291.
Farrington DP, Jolliffe D, Loeber R, Stouthamer-Loeber M and Kalb LM (2001): The concentration
of offenders in families, and family criminality in the prediction of boys’delinquency. J Adolesc
24:579-596.
Feehan M, McGee R and Williams SM (1993): Mental Health Disorders from Age 15 to Age 18
Years. J Am Acad Child Adolesc Psychiatry 32:1118-1126.
Ferdinand RF, Verhulst FC and Wiznitzer M (1995): Continuity and Change of Self-Reported
Problem Behaviors from Adolescence into Young Adulthood. J Am Acad Child Adolesc Psychiatry
34:680-690.
105
Fergusson DM (1998): Stability and change in externalising behaviours. Eur Arch Psychiatry Clin
Neurosci 248:4-13.
Fergusson DM and Horwood LJ (1998): Early Conduct Problems and Later Life Opportunities. J
Child Psychol Psychiatry 39:1097-1108.
Fergusson DM, Horwood LJ and Lynskey MT (1993): Prevalence and Comorbidity of DSM-III-R
Diagnoses in a Birth Cohort of 15 Year Olds. J Am Acad Child Adolesc Psychiatry 32:1127-1134.
Fergusson DM, Lynskey MT and Horwood LJ (1996): Origins of Comorbidity between Conduct
and Affective Disorders. J Am Acad Child Adolesc Psychiatry 35:451-460.
Fergusson DM, Woorward L and Horwood LJ (1998): Maternal Smoking During Pregnancy and
Psychiatric Adjustment in Late Adolescence. Arch Gen Psychiatry 55:721-727.
Fergusson DM, Wanner B, Vitaro F, Horwood LJ and Swain-Campbell N (2003): Deviant Peer
Affiliations and Depression: Confounding or Causation? J Abnormal Child Psychology 31:605-618.
Fergusson DM, Horwood LJ and Ridder EM (2005): Show me the child at seven: the consequences
of conduct problems in childhood for psychosocial functioning in adulthood. J Child Psychol
Psychiatry 46:837-849.
Flannery DJ, Singer MI and Wester K (2001): Violence Exposure, Psychological Trauma, and
Suicide Risk in a Community Sample of Dangerously Violent Adolescents. J Am Acad Child
Adolesc Psychiatry 40:435-442.
Fombonne E (1998): Increased rates of psychosocial disorders in youth. Eur Arch Psychiatry Clin
Neurosci 248:14-21.
Fombonne E, Wostear G, Cooper V, Harrington R and Rutter M (2001a): The Maudsley long-term
follow-up of child and adolescent depression. Br J Psychiatry 179:210-217.
106
Fombonne E, Wostear G, Cooper V, Harrington R and Rutter M (2001b): The Maudsley long-term
follow-up of child and adolescent depression: 2. Suicidality, criminality and social dysfunction in
adulthood. Br J Psychiatry 179:218-223.
Ford T, Goodman R and Meltzer H (2003): The British Child and Adolescent Mental Health Survey
1999: The Prevalence of DSM-IV Disorders. J Am Acd Child Adolesc Psychiatry 42:1203-1211.
Fountoulakis KN, Iacovides A, Kleanthous S, Samolis S, Gougoulias K, St Kaprinis G and Bech P
(2003): The Greek translation of the symptoms rating scale for depression and anxiety: preliminary
results of the validation study. BMC Psychiatry 3:doi:10.1186/1471-244X-3-21.
Fröjd S, Charpentier P, Luukkaala T, Pelkonen M, Ranta K, Ritakallio M, von der Pahlen B,
Marttunen M and Kaltiala-Heino R (2004): 9-luokkalaisten mielenterveys Tampereella ja Vantaalla.
Pirkanmaan sairaanhoitopiirin julkaisuja 7/2004. Tampereen Yliopistopaino Oy, Tampere.
Fröjd S, Marttunen M, Pelkonen M, von der Pahlen B and Kaltiala-Heino R (2006a): Perceived
financial difficulties and maladjustment outcome in adolescence. Eur J Public Health 16:542-548.
Fröjd S, Kaltiala-Heino R and Marttunen M (2006b): 9-luokkalaisten mielenterveys Tampereella ja
Vantaalla, kaksivuotisseuranta. Pirkanmaan sairaanhoitopiirin julkaisuja, 6/2006. Tampereen
Yliopistopaino Oy, Tampere.
Fröjd S, Kaltiala-Heino R and Rimpelä M (2007): The association of parental monitoring and
family structure with diverse maladjustment outcomes in middle adolescent boys and girls. Nord J
Psychiatry 61:296-303.
Garber J (2006): Depression in Children and Adolescents. Linking Risk Research and Prevention.
Am J Prev Med 31:S104–S125.
Garnefski N (2000): Age Differences in Depressive Symptoms, Antisocial Behavior, and Negative
Perceptions of Family, School, and Peers Among Adolescents. J Am Acad Child Adolesc
Psychiatry 39:1175-1181.
107
Garnefski N and Diekstra RFW (1996): Perceived Social Support from Family, School, and Peers:
Relationship with Emotional and Behavioral Problems among Adolescents. J Am Acad Child
Adolesc Psychiatry 35:1657-1664.
Garnefski N and Diekstra RFW (1997): “Comorbidity”of Behavioral, Emotional, and Cognitive
Problems in Adolescence. J Youth Adolesc 26:321-338.
Garrison CZ, Waller JL, Cuffe SP, McKeown RE, Addy CL and Jackson KL (1997): Incidence of
major depressive disorder and dysthymia in young adolescents. J Am Acad Child Adolesc
Psychiatry 36:458-465.
Gau SS, Chong MY, Chen TH and Cheng AT (2005): A 3-Year Panel Study of Mental Disorders
Among Adolescents in Taiwan. Am J Psychiatry 162:1344-1350.
Giaconia RM, Reinherz HZ, Paradis AD, Carmola HA and Stashwick CK (2001): Major
Depression and Drug Disorders in Adolescence: General and Specific Impairments in Early
Adulthood. J Am Acad Child Adolesc Psychiatry 40:1426-1433.
Gilvarry E (2000): Substance Abuse in Young People. J Child Psychol Psychiatr 41:55-80.
Glaser K (1968): Masked depression in children and adolescents. Annu Progr Child Psychiatr Dev
1:345-355.
Glied S and Pine DS (2002): Consequences and Correlates of Adolescent Depression. Arch Pediatr
Adolesc Med 156:1009-1014.
Gomes JT, Bertrand LD, Paetsch JJ and Hornick JP (2003): Self-reported delinquency among
Alberta’s youth: findings from a survey of 2,001 junior and senior high school students.
Adolescence 38:75-91.
González-Tejera G, Canino G, Ramírez R, Chávez L, Shrout P, Bird H, Bravo M, Martínez-Taboas
A, Ribera J and Bauermeister J (2005): Examining minor and major depression in adolescents. J
Child Psychol Psychiatry 46:888-899.
108
Goodman R (2001): Psychometric Properties of the Strengths and Difficulties Questionnaire. J Am
Acad Child Adolesc Psychiatry 40:1337-1345.
Gotlib JH, Lewinsohn PM and Seeley JR (1995): Symptoms Versus a Diagnostic Depression:
Differences in Psychosocial Functioning. J Consult Clin Psychol 63:90-100.
Gottfredson M and Hirschi T (1990): A General Theory of Crime. Stanford University Press,
Stanford.
Graber JA, Lewinsohn PM, Seeley JR and Brooks-Gunn J (1997): Is Psychopathology Associated
With the Timing of Pubertal Development? J Am Acad Child Adolesc Psychiatry 36:1768-1776.
Greene RW, Biederman J, Zerwas S, Monuteaux MC, Goring JC and Faraone SV (2002):
Psychiatric Comorbidity, Family Dysfunction, and Social Impairment in Referred Youth With
Oppositional Defiant Disorder. Am J Psychiatry 159:1214-1224.
Grigorenko EL (2006): Learning Disabilities in Juvenile Offenders. Child Adolesc Psychiatric Clin
N Am 15:353-371.
Grilo CM, Becker DF, Fehon DC, Edell WS and McGlashan TH (1996): Conduct Disorders,
Substance Use Disorders, and Coexisting Conduct and Substance Use Disorders in Adolescent
Inpatients. Am J Psychiatry 153:914-920.
Gutgesell ME and Payne N (2004): Issues of Adolescent Psychological Development in the 21st
Century. Pediatr Rev 25:79-85.
Haapasalo J and Hämälainen T (1996): Childhood family problems and current psychiatric
problems among young violent and property offenders. J Am Acad Child AdolescPsychiatry
35:1394–1401.
Haarasilta L, Marttunen M, Kaprio J and Aro H (2001): The 12-month prevalence and
characteristics of major depressive episode in a representative nationwide sample of adolescents and
young adults. Psychol Med 31:1169-1179.
109
Haarasilta L, Marttunen M, Kaprio J and Aro H (2005): Major depressive episode and physical
health in adolescents and young adults: results from a population-based interview survey. Eur J
Public Health 15:489-493.
Hankin BL (2006): Adolescent depression: Description, causes, and interventions. Epilepsy Behav
8:102-114.
Hankin BL and Abramson LY (1999): Development of gender differences in depression:
description and possible explanations. Ann Med 31:372-379.
Hankin BL, Abramson LY, Moffitt TE, Silva PA, McGee R and Angell KE (1998): Development
of Depression From Preadolescence to Young Adulthood: Emerging Gender Differences in a 10Year Longitudinal Study. J Abnorm Psychol 107:128-140.
Harland P, Reijneveld SA, Brugman E, Verloove-Vanhorick SP and Verhulst FC (2002): Family
factors and life events as risk factors for behavioural and emotional problems in children. Eur Child
Adolesc Psychiatry 11:176-184.
Harrington R, Fudge H, Rutter M, Pickles A and Hill J (1991): Adult Outcomes of Childhood and
Adolescent Depression: II. Links with Antisocial Disorders. J Am Acad Child Adolesc Psychiatry
30:434-439.
Helsen M, Vollebergh W and Meeus W (2000): Social Support from Parents and Friends and
Emotional Problems in Adolescence. J Youth Adolesc 29:319-335.
Helstelä L and Sourander A (2001): Self-reported competence and emotional and behavioral
problems in a sample of Finnish adolescents. Nord J Psychiatry 55:381-385.
Herbert J and Martinez M (2001): Neural mechanisms underlying aggressive behaviour. In:
Conduct disorders in childhood and adolescence, pp. 67-102. Eds. J Hill and B Maughan,
Cambridge University Press, New York.
Herkov MJ and Myers WC (1996): MMPI profiles of depressed adolescents with and without
conduct disorder. J Clin Psychol 52:705-710.
110
Heyerdahl S, Kvernmo S and Wichstrom L (2004): Self-reported behavioural/emotional problems
in Norwegian adolescents from multiethnic areas. Eur Child Adolesc Psychiatry 13:64-72.
Hill J (2001): Biosocial influences on antisocial behaviours in childhood and adolescence. In:
Conduct disorders in childhood and adolescence, pp. 103-125. Eds. J Hill and B Maughan,
Cambridge University Press, New York.
Hill J (2002): Biological, psychological and social processes in the conduct disorders. J Child
Psychol Psychiatry 43:133-164.
Hofstra MB, van deer Ende J and Verhulst FC (2002): Child and Adolescent Problems Predict
DSM-IV Disorders in Adulthood: A 14-Year Follow-up of a Dutch Epidemiological Sample. J Am
Acad Child Adolesc Psychiatry 41:182-189.
http://info.stakes.fi/kouluterveyskysely (16.3.2007 and 9.11.2007)
Hurtig T, Taanila A, Ebeling H, Miettunen J and Moilanen I (2005): Attention and behavioural
problems of Finnish adolescents may be related to the family environment. Eur Child Adolesc
Psychiatry 14:471-478.
Hänninen V and Aro H (1996): Sex differences in coping and depression among young adults. Soc
Sci Med 43:1453-1460.
Ingoldsby EM, Kohl GO, McMahon RJ, Lengua L and The Conduct Problems Prevention Research
Group (2006): Conduct Problems, Depressive Symptomatology and Their Co-Occurring
Presentation in Childhood as Predictors of Adjustment in Early Adolescence. J Abnorm Child
Psychol 34:603-621.
Isometsä E (2007): Masennushäiriöt. Kirjassa Psykiatria, ss. 156-195. Toim. J Lönnqvist, M
Heikkinen, M Henriksson, M Marttunen ja T Partonen, Gummerus Kirjapaino Oy, Jyväskylä.
Ivarsson T and Gillberg C (1997): Depressive symptoms in Swedish adolescents: Normative data
using the Birleson Depression Self-Rating Scale (DSRS). J Affect Disord 42:59-68.
111
Jensen PS, Martin D and Cantwell DP (1997): Comorbidity in ADHD: Implications for Research,
Practice, and DSM-V. J Am Acad Child Adolesc Psychiatry 36:1065-1079.
Junger-Tas J (1994a): Delinquency in thirteen western countries: some preliminary conclusion. In:
Delinquent behavior among young people in the western world: first results of the international selfreport delinquency study, pp. 370-380. Eds. J Junger-Tas, G Terlouw and MW Klein, Kugler
Publications, Amsterdam.
Junger-Tas J (1994b): The international self-report delinquency study: some methodological and
theoretical issues. In: Delinquent behavior among young people in the western world: first results of
the international self-report delinquency study, pp. 1-13. Eds. J Junger-Tas, G Terlouw and MW
Klein, Kugler Publications, Amsterdam.
Kaltiala-Heino R, Rimpelä M, Marttunen M, Rimpelä A and Rantanen P (1999a): Bullying,
depression, and suicidal ideation in Finnish adolescents: school survey. BMJ 319:348-351.
Kaltiala-Heino R, Rimpelä M, Rantanen P and Laippala P (1999b): Finnish modification of the 13item Beck Depression Inventory in screening an adolescent population for depressiveness and
positive mood. Nord J Psychiatry 53:451-457.
Kaltiala-Heino R, Rimpelä M, Rantanen P and Rimpelä A (2000): Bullying at school –and
indicator of adolescents at risk for mental disorders. J Adolesc 23:661–674.
Kaltiala-Heino R, Rimpelä M, Rantanen P and Laippala P (2001): Adolescent depression: the
role of discontinuities in life course and social support. J Affect Disord 64:155-166.
Kaltiala-Heino R, Marttunen M, Rantanen P and Rimpelä M (2003): Early puberty is associated
with mental health problems in middle adolescence. Soc Sci Med 57:1055-1064.
Kaltiala-Heino R, Kaivosoja M and Ritakallio M (2006): Nuorten rikostentekijöiden mielenterveys
ja psykiatrisen hoidon tarve. Suom Lääkäril 61:971–977.
Kann RT and Hanna FJ (2000): Disruptive Behavior Disorder in Children and Adolescents: How
Do Girls Differ From Boys. J Couns Dev 7:267-275.
112
Kapi A, Veltsista A, Sovio U, Järvelin M and Bakoula C (2007): Comparison of self-reported
emotional and behavioural problems in adolescents from Greece and Finland. Acta Paediatr
96:1174-1179.
Karlsson L, Pelkonen M, Aalto-Setälä T ja Marttunen M (2005): Nuorten masennus –vakava
sairaus, jonka hoitoa tutkittu vähän. Suom Lääkäril 60:2879–2883.
Karlsson L, Pelkonen M, Heilä H, Holi M, Kiviruusu O, Tuisku V, Ruuttu T and Marttunen M
(2006a): Differences in the clinical characteristics of adolescent depressive disorders. Depress
Anxiety 0:1-12.
Karlsson L, Pelkonen M, Ruuttu T, Kiviruusu O, Heilä H, Holi M, Kettunen K, Tuisku V, TuulioHenriksson A, Törrönen J and Marttunen M (2006b): Current comorbidity among consecutive
adolescent psychiatric outpatients with DSM-IV mood disorders. Eur Child Adolesc Psychiatry
15:220-231.
Karnik NS, McMullin MA and Steiner H (2006): Disruptive Behaviors: Conduct and Oppositional
Disorders in Adolescence. Adolesc Med 17:97-114.
Kashani JH, Manning GW, McKnew DH, Cytryn L, Simonds JF and Wooderson PC (1980):
Depression Among Incarcerated Delinquents. Psychiatry Res 3:185-191.
Katainen S, Räikkönen K and Keltikangas-Järvinen L (1999): Adolescent Temperament, Perceived
Social Support, and Depressive Tendencies as Predictors of Depressive Tendencies in Young
Adulthood. Eur J Personal 13:183-207.
Kaufman J, Birmaher B, Brent D, Rao U, Flynn C, Moreci P, Williamson D and Ryan N (1997):
Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime
Version (K-SADS-PL): Initial Reliability and Validity Data . J Am Acad Child Adolesc Psychiatry
36:980-988.
Kaufman J, Martin A, King RA and Charney D (2001): Are Child-, Adolescent-, and Adult-Onset
Depression One and the Same Disorder? Biol Psychiatry 49:980-1001.
113
Keenan K, Loeber R and Green S (1999): Conduct Disorder in Girls: A Review of the Literature.
Clin Child Fam Psychol Rev 2:3-19.
Keiley MK, Lofthouse N, Bates JE, Dodge KA and Pettit GS (2003): Differential Risks of
Covarying and Pure Components in Mother and Teacher Reports of Externalizing and Internalizing
Behavior Across Ages 5 to 14. J Abnorm Child Psychol 31:267-283.
Keltikangas-Järvinen L (2001): Aggressive behaviour and social problem-solving strategies: a
review of the findings of a seven-year follow-up from childhood to late adolescence. Crim Behav
Ment Health 11:236-250.
Kessler RC and Walters EE (1998): Epidemiology of DSM-III-R Major Depression and Minor
Depression Among Adolescents and Young Adults in the National Comorbidity Survey. Depress
Anxiety 7:3-14.
Kessler RC, Avenevoli S and Merikangas KR (2001): Mood Disorders in Children and
Adolescents: An Epidemiologic Perspective. Biol Psychiatry 49:1002-1014.
Kiesner J (2002): Depressive Symptoms in Early Adolescence: Their Relations with Classroom
Problem Behavior and Peer Status. J Res Adolesc 12:463-478.
Kim KJ, Conger RD, Elder GH and Lorenz FO (2003): Reciprocal Influences Between Stressful
Life Events and Adolescent Internalizing and Externalizing Problems. Child Dev 74:127-143.
Kim-Cohen J, Caspi A, Moffitt TE, Harrington H, Milne BJ and Poulton R (2003): Prior Juvenile
Diagnoses in Adults With Mental Disorder: Developmental Follow-Back of a ProspectiveLongitudinal Cohort. Arch Gen Psychiatry 60:709-717.
King R, Scwab-Stone M, Flisher AJ, Greenwald S, Kramer RA, Goodman SH, Lahey BB, Shaffer
D and Gould MS (2001): Psychosocial and Risk Behavior Correlates of Youth Suicide Attempts
and Suicidal Identiation. J Am Acad Child Adolesc Psychiatry 40:837-846.
114
King RA, Schwab-Stone ME, Peterson BS and Thies AP (2005): Psychiatric Examination of the
Infant, Child, and Adolescent. In: Kaplan & Sadock’s Comprehensive Textbook of Psychiatry,
volume two, pp. 3044-3075. Eds. BJ Sadock and VA Sadock, Lippincott Williams & Wilkins,
Philadelphia.
Kingston L and Prior M (1995): The Development of Patterns of Stable, Transient, and School-Age
Onset Aggressive Behavior in Young Children. J Am Acad Child Adolesc Psychiatry 34:348-358.
Kitinoja M (2005): Kujan päässä on koulukoti. Tutkimus koulukoteihin sijoitettujen lasten
lastensuojeluasiakkuudesta ja kouluhistoriasta. Stakes, Tutkimuksia150. Gummerus Kirjapaino,
Saarijärvi.
Kivivuori J (2000): Delinquent Behaviour, Psychosomatic Symptoms and the Idea of “Healthy
Delinquency”. J Scand Stud Crim Crime Prev 1:121-139.
Kivivuori J (2002a): Piilorikollisuuden tutkimus. Kirjassa Nuoret rikosten tekijöinä ja uhreina, ss.
1-13. Oikeuspoliittisen tutkimuslaitoksen julkaisuja 188/2002. Toim. J Kivivuori, Hakapaino Oy,
Helsinki.
Kivivuori J (2002b): Nuoret rikosten tekijöinä, uhreina ja kontrollin kohteina 1995-2001. Kirjassa
Nuoret rikosten tekijöinä ja uhreina, ss. 14-63. Oikeuspoliittisen tutkimuslaitoksen julkaisuja
188/2002. Toim. J Kivivuori, Hakapaino Oy, Helsinki.
Kivivuori J (2005a): Piilorikollisuuden tutkimus. Kirjassa Nuorten rikoskäyttäytyminen 1995 –
2004. Teemana sosiaalinen pääoma, ss. 1-10. Oikeuspoliittisen tutkimuslaitoksen julkaisuja
214/2005. Toim. J Kivivuori ja V Salmi, Hakapaino Oy, Helsinki.
Kivivuori J (2005b): Nuoret rikosten tekijöinä, uhreina ja kontrollin kohteina 1995 –2004. Kirjassa
Nuorten rikoskäyttäytyminen 1995 - 2004. Teemana sosiaalinen pääoma, ss. 11-70.
Oikeuspoliittisen tutkimuslaitoksen julkaisuja, 214/2005. Toim. J Kivivuori ja V Salmi, Hakapaino
Oy, Helsinki.
Konu A (2002): Oppilaiden hyvinvointi koulussa. Acta Universitatis Tamperensis 887. University
of Tampere. Tampereen Yliopistopaino Oy Juvenes Print, Tampere.
115
Koskinen O, Sauvola A, Valonen P, Hakko H, Järvelin M and Räsänen P (2001): Increased risk of
violent recidivism among adult males is related to single-parent family during childhood: the
Northern Finland 1966 Birth Cohort Study. J Forens Psychiatry 12:539-548.
Kovacs M, Paulauskas S, Gatsonis C and Richards C (1988): Depressive disorders in childhood III.
A longitudinal study of comorbidity with and risk for conduct disorder. J Affect Disord 15:205-217.
Kovacs M, Obrosky DS and Sherrill J (2003): Developmental changes in the phenomenology of
depression in girls compared to boys from childhood onward. J Affect Disord 74:33-48.
Kratzer L and Hodgins S (1997): Adult Outcomes of Child Conduct Problems: A Cohort Study. J
Abnorm Child Psychol 25:65-81.
Kurki T, Hiilesmaa V, Raitasalo R, Mattila H and Ylikorkala O (2000): Depression and Anxiety in
Early Pregnancy and Risk for Preeclampsia. Obstet Gynecol 95:487-490.
Kvernmo S and Heyerdahl S (1998): Influence of Ethnic Factors on Behavior Problems in
Indigenous Sami and Majority Norwegian Adolescents. J Am Acad Child Adolesc Psychiatry
37:743-751.
Kyvsgaard B (2003): The Criminal Career. The Danish Longitudinal Study. Cambridge University
Press, Cambridge.
Lagges AM and Dunn DW (2003): Depression in children and adolescents. Neurol Clin N Am
21:953-960.
Lahey BB, Loeber R, Quay HC, Applegate B, Shaffer D, Waldman I, Hart EL, McBurnett K, Frick
PJ, Jensen PS, Dulcan MK, Canino G and Bird HR (1998): Validity of DSM-IV Subtypes of
Conduct Disorder Based on Age of Onset. J Am Acad Child Adolesc Psychiatry 37:435-442
Lahey BB, Goodman SH, Waldman ID, Bird H, Canino G, Jensen P, Regier D, Leaf PJ, Gordon R
and Applegate B (1999a): Relation of Age of Onset to the Type and Severity of Child and
Adolescent Conduct Problems. J Abnorm Child Psychol 27:247-260.
116
Lahey BB, Waldman ID and McBurnett K (1999b): Annotation: The Development of Antisocial
Behavior: An Integrative Causal Model. J Child Psychol Psychiatry 40:669-682.
Lahey BB, Goodman SH, Canino G, Bird H, Schwab-Stone M, Waldman ID, Rathouz PJ, Miller
TL, Dennis KD (2000): Age and Gender Differences in Oppositional Behvior and Conduct
Problems: A Cross-Sectional Household Study of Middle Childhood and Adolescence. J Abnorm
Psychol 109:488-503.
Lahey BB, Loeber R, Burke J, Rathouz PJ and McBurnett K (2002): Waxing and Waning in
Concert: Dynamic Comorbidity of Conduct Disorder With Other Disruptive and Emotional
Problems Over 7 Years Among Clinic-Referred Boys. J Abnorm Psychol 111:556-567.
Lasa L, Ayuso-Mateos JL, Vàzquez-Barquero JL, Díez-Manrique FJ and Dowrick CF (2000): The
use of the Beck Depression Inventory to screen for depression in the general population: a
preliminary analysis. J Affect Disord 57:261-265.
Laub J and Vaillant G (2000): Delinquency and mortality: a 50 year follow-up study of 1000
delinquent and non-deliquent boys. Am J Psychiatry 157:96-102.
Laukkanen E, Shemeikka S, Notkola I, Koivumaa-Honkanen H and Nissinen A (2002):
Externalizing and internalizing problems at school as signs of health-damaging behaviour and
incipient marginalization. Health Prom Intern 17:139-146.
Leadbeater BJ, Kuperminc GP, Hertzog C and Blatt SJ (1999): A Multivariate Model of Gender
Differences in Adolescents’Internalizing and Externalizing Problems. Dev Psychol 35:1268-1282.
Lehto-Salo P, Kuuri A, Marttunen M, Mahlanen A, Toivonen V, Toivola P, Närhi V, Ahonen T ja
Koponen H (2002): POLKU-tutkimus. Tutkimus nuorista kolmessa koulukodissa. Psykiatrinen ja
neurologinen näkökulma. Stakes, aiheita 16/2002. Stakes, Helsinki.
Lewin LM, Davis B and Hops H (1999): Childhood Social Predictors of Adolescent Antisocial
Behavior: Gender Differences in Predictive Accuracy and Efficacy. J Abnorm Child Psychol
27:277-292.
117
Lewinsohn PM, Rohde P, Seeley JR and Hops H (1991): Comorbidity of Unipolar Depression: I.
Major Depression With Dysthymia. J Abnorm Psychol 100: 205-213.
Lewinsohn PM, Hops H, Roberts RE, Seeley JR and Andrews JA (1993): Adolescent
Psychopathology: I. Prevalence and Incidence of Depression and Other DSM-III-R Disorders in
High School Students. J Abnorm Psychol 102:133-144.
Lewinsohn PM, Clarke GN, Seeley JR and Rohde P (1994): Major Depression in Community
Adolescents: Age at Onset, Episode Duration, and Time to Recurrence. J Am Acad Child Adolesc
Psychiatry 33:809-818.
Lewinsohn PM, Rohde P and Seeley JR (1995a): Adolescent Psychopathology: III. The Clinical
Consequences of Comorbidity. J Am Acad Child Adolesc Psychiatry 34:510-519.
Lewinsohn PM, Gotlib JH and Seeley JR (1995b): Adolescent Psychopathology: IV. Specificity of
Psychosocial Risk Factors for Depression and Substance Abuse in Older Adolescents. J Am Acad
Child Adolesc Psychiatry 34:1221-1229.
Lewinsohn PM, Rohde P and Seeley JR (1998): Major depressive disorder in older adolescents:
prevalence, risk factors, and clinical implications. Clin Psychol Rev 18:765-794.
Lewinsohn PM, Pettit JW, Joiner TE and Seeley JR (2003): The Symptomatic Expression of Major
Depressive Disorder in Adolescents and Young Adults. J Abnorm Psychol 112:244-252.
Lien L, Lien N, Heyerdahl S, Thoresen M and Bjertness E (2006): Consumption of Soft Drinks and
Hyperactivity, Mental Distress, and Conduct Problems Among Adolescents in Oslo, Norway. Am J
Public Health 96:1815-1820.
Liss H, Phares V and Liljequist L (2001): Symptom Endorsement Differences on the Children’s
Depression Inventory With Children and Adolescents on an Inpatient Unit. J Pers Assess 76:396411.
118
Little SA and Garber J (2005): The Role of Social Stressors and Interpersonal Orientation in
Explaining the Longitudinal Relation Between Externalizing and Depressive Symptoms. J Abnorm
Psychol 114:432-443.
Loeber R and Coie J (2001): Continuities and discontinuities of development, with particular
emphasis on emotional and cognitive components of disruptive behaviour. In: Conduct disorders in
childhood and adolescence, pp. 379-407. Eds. J Hill and B Maughan, Cambridge University Press,
New York.
Loeber R and Farrington DP (2000): Young children who commit crime: Epidemiology,
developmental origins, risk factors, early interventions, and policy implications. Dev Psychopathol
12:737-762.
Loeber R and Hay D (1997): Key Issues in the Development of Aggression and Violence from
Childhood to Early Adulthood. Annu Rev Psychol 48:371-410.
Loeber R and Keenan K (1994): Interaction between conduct disorder and its comorbid conditions:
effects of age and gender. Clin Psychol Rev 14:497-523.
Loeber R and Stouthamer-Loeber M (1998): Development of Juvenile Aggression and Violence.
Some Common Misconceptions and Controversials. Am Psychol 53:242-259.
Loeber R, Russo MF, Stouthamer-Loeber M and Lahey BB (1994): Internalizing Problems and
Their Relations to the Development of Distruptive Behaviors in Adolescence. J Res Adolesc 4:615637.
Loeber R, Green SM, Keenan K and Lahey BB (1995): Which Boys Will Fare Worse? Early
Predictors of the Onset of Conduct Disorder in a Six-Year Longitudinal Study. J Am Acad Child
Adolesc Psychiatry 34:499-509.
Loeber R, Keenan K and Zhang Q (1997): Boys’Experimentation and Persistence in
Developmental Pathways Toward Serious Delinquency. J Child Fam Stud 6:321-357.
119
Loeber R, Farrington DP, Stouthamer-Loeber M and Van Kammen WB (1998): Antisocial
Behaviour and Mental Health Problems. Explanatory Factors in Childhood and Adolescence.
Lawrence Erlbaum Associates, Mahwah.
Loeber R, Burke JD, Lahey BB, Winters A and Zera M (2000a): Oppositional Defiant and Conduct
Disorder: A Review of the Past 10 Years, Part I. J Am Acad Child Adolesc Psychiatry 39:14681484.
Loeber R, Green SM, Lahey BB and Kalb L (2000b): Physical Fighting in Childhood as a Risk
Factor for Later Mental Health Problems. J Am Acad Child Adolesc Psychiatry 39:421-428.
Loeber R, Farrington DP, Stouthamer-Loeber M, Moffitt TE, Caspi A and Lynam D (2001): Male
Mental Health Problems, Psychopathy, Personality Traits: Key Findings from the First 14 Years of
the Pittsburgh Youth Study. Clin Child Fam Psychol Rev 4:273-297.
Luokas A, Fitzgerald HE, Zucker RA and von Eye A (2001): Parental Alcoholism and CoOccurring Antisocial Behavior: Prospective Relationships to Externalizing Behavior Problems in
their Young Sons. J Abnorm Child Psychol 29:91-106.
Luopa P, Rimpelä M ja Jokela J (2003): Nuorten hyvinvointi Etelä-Suomen, Itä-Suomen ja Lapin
lääneissä. Kouluterveyskysely 2000 ja 2002. Stakes aiheita 7/2003. Stakes, Helsinki.
Luopa P, Pietikäinen M ja Jokela J (2006): Nuorten elinolot, koulutyö, terveys ja
terveystottumukset 1996 - 2005. Kouluterveyskysely 2005. Stakesin työpapereita 25/2006.
Valopaino Oy, Helsinki.
Lynch FL and Clarke GN (2006): Estimating the Economic Burden of Depression in Children and
Adolescents. Am J Prev Med 31:S143-S151.
Lynam DR and Henry B (2001): The role of neuropsychological deficts in conduct disorders.
In: Conduct disorders in childhood and adolescence, pp. 235-263. Eds. J Hill and B Maughan,
Cambridge University Press, New York.
120
Läärä E and Lammi S (1989): Tilastotieteen perusteet lääketiedettä ja lähialoja varten. Kuopion
yliopiston painatuskeskus, Kuopio.
Maharajh HD, Ali A and Konings M (2006): Adolescent depression in Trinidad and Tobago. Eur
Child Adolesc Psychiatry 15:30-37.
Marmorstein NR and Iacono WG (2001): An Investigation of Female Adolescents Twins With Both
Major Depression and Conduct Disorder. J Am Acad Child Adolesc Psychiatry 40:299-306.
Marmorstein NR and Iacono WG (2003): Major Depression and Conduct Disorder in a Twin
Sample: Gender, Functioning, and Risk for Future Psychopathology. J Am Acad Child Adolesc
Psychiatry 42:225-233.
Marttunen M and Pelkonen M (1998): Phenomenology of adolescent depression. Psychiatr Fenn
29:29-39.
Marttunen M and Rantanen P (1999): Nuorisopsykiatria. Kirjassa Psykiatria, ss. 497–527. Toim. J
Lönnqvist, M Heikkinen, M Henriksson, M Marttunen and T Partonen, Gummerus Kirjapaino Oy,
Jyväskylä.
Marttunen M, Kataja H, Henttonen A, Hokkanen T, Tuominen T ja Ebeling H (2004): Hyötyykö
käytöshäiriöinen nuori nuorisopsykiatrisesta osastohoidosta? Duodecim 120:43-49.
Masi G, Favilla L and Mucci M (1998): Depressive disorder in children and adolescents. Eur J
Paediatr Neurol 2:287-295.
Maughan B (2001): Conduct disorder in context. In: Conduct disorders in childhood and
adolescence, pp. 169-201. Eds. J Hill and B Maughan, Cambridge University Press, New York.
Maughan B and Rutter M (2001): Antisocial children grown up. In: Conduct disorders in childhood
and adolescence, pp. 507-552. Eds. J Hill and B Maughan, Cambridge University Press, New York.
121
Maughan B, Rowe R, Messer J, Goodman R and Meltzer H (2004): Conduct Disorder and
Oppostional Defiant Disorder in a national sample: developmental epidemiology. J Child Psychol
Psychiatry 45:609-621.
McCabe KM, Hough R, Wood PA and Yeh M (2001): Childhood and Adolescent Onset Conduct
Disorder: A Test of the Developmental Taxonomy. J Abnorm Child Psychol 29:305-316.
McCabe KM, Lansing AE, Garland A and Hough R (2002): Gender Differences in
Psychopathology, Functional Impairment, and Familial Risk Factors Among Adjudicated
Delinquents. J Am Acad Child Adolesc Psychiatry 41:860-867.
McClellan JM and Werry JS (2000): Introduction. Special Section: Research Psychiatric Diagnostic
Interviews for Children and Adolescents. J Am Acad Child Adolesc Psychiatry 39:19-27.
McMahon R and Frick P (2005): Evidence-based assessment of conduct problems in children and
adolescents. J Clin Child Adolesc Psychol 34:477-505.
Measelle JR, Stice E and Hogansen JM (2006): Developmental Trajectories of Co-occurring
Depressive, Eating, Antisocial, and Substance Abuse Problems in Female Adolescents. J Abnorm
Psychol 115:524-538.
Meller WH and Borchardt CM (1996): Comorbidity of major depression and conduct disorder. J
Affect Disord 39:123-126.
Mervaala E ja Räsänen E (2004): Aivot ja mieli. Kirjassa Lasten- ja nuorisopsykiatria, ss. 26-33.
Toim. I Moilanen, E Räsänen, T Tamminen, F Almqvist, J Piha ja K Kumpulainen, Gummerus
Kirjapaino Oy, Jyväskylä.
Miller-Johnson S, Lochman JE, Coie JD, Terry R and Hyman C (1998): Comorbidity of Conduct
and Depressive Problems at Sixth Grade: Substance Use Outcomes Across Adolescence. J Abnorm
Child Psychol 26:221-232.
122
Millikan E, Wamboldt MZ and Bihun JT (2002): Perceptions of the Family, Personality
Characteristics, and Adolescent Internalizing Symptoms. J Am Acad Child Adolesc Psychiatry
41:1486-1494.
Mitchell J, McCauley E, Burke P.M and Moss SJ (1988): Phenomenology of Depression in
Children and Adolescents. J Am Acad Child Adolesc Psychiatry 27:12-20.
Moffitt TE (1993): Adolescence-Limited and Life-Course-Persistent Antisocial Behaviour: A
Developmental Taxonomy. Psychol Rev 100:674-701.
Moffitt TE and Caspi A (2001): Childhood predictors differentiate life-course persistent and
adolescence-limited antisocial pathways among males and females. Dev Psychopathol 13:355-375.
Moffitt TE, Caspi A, Dickson N, Silva P and Stanton W (1996): Childhood-onset versus
adolescent-onset antisocial conduct problems in males: Natural history from ages 3 to 18 years. Dev
Psychopathol 8:399-424.
Moffitt TE, Caspi A, Harrington H and Milne BJ (2002): Males on the life-course-persistent and
adolescence-limited antisocial pathways: Follow-up at age 26 years. Dev Psychopathol 14:179-207.
Mojtabai R (2006): Serious Emotional and Behavioral Problems and Mental Health Contacts in
American and British Children and Adolescents. J Am Acad Child Adolesc Psychiatry 45:12151223.
Molnar BE, Browne A, Cerda M and Buka SL (2005): Violent Behaviour by Girls Reporting
Violent Victimization: a Prospective Study. Arch Pediatr Adolesc Med 159:731-739.
Morgan CJ and Cauce AM (1999): Predicting DSM-III-R Disorders From the Youth Self-Report:
Analysis of Data From a Field Study. J Am Acad Child Adolesc Psychiatry 38:1237-1245.
Myers K and Winters NC (2002): Ten-Year Review of Rating Scales. II: Scales for Internalizing
Disorders. J Am Acad Child Adolesc Psychiatry 41:634-659.
123
Newman DL, Moffitt TE, Silva PA, Caspi A and Magdol L (1996): Psychiatric Disorders in a Birth
Cohort of Young Adults: Prevalence, Comorbidity, Clinical Significance, and New Case Incidence
From Ages 11 to 21. J Consult Clin Psychol 64:552-562.
Nock MK, Kazdin AE, Hiripi E and Kessler RC (2006): Prevalence, subtypes, and correlates of
DSM-IV conduct disorder in the National Comorbidity Survey Replication. Psychol Med 36:699710.
Nolen-Hoeksema S and Girgus JS (1994): The Emergence of Gender Differences in Depression
During Adolescence. Psychol Bull 115:424-443.
O’Connor TG, McGuire S, Reiss D, Hetherington EM and Plomin R (1998): Co-occurrence of
Depressive Symptoms and Antisocial Behavior in Adolescence: A Common Genetic Liability. J
Abnorm Psychol 107:27-37.
Oland AA and Shaw DS (2005): Pure Versus Co-occurring Externalizing and Internalizing
Symptoms in Children: The Potential Role of Socio-Developmental Milestones. Clin Child Fam
Psychol Rev 8:247-270.
Oldehinkel AJ, Wittcehn H-U and Schuster P (1999): Prevalence, 20-month incidence and outcome
of unipolar depressive disorders in a community sample of adolescents. Psychol Med 29:655-668.
Overbeek G, Vollebergh W, Meeus W, Engels R and Luijpers E (2001): Course, Co-occurrence,
and Longitudinal Associations of Emotional Disturbance and Delinquency From Adolescence to
Young Adulthood: A Six-Year Three-Wave Study. J Youth Adolesc 30:401-426.
Pajer K (1998): What Happens to “Bad”Girls? A Review of the Adult Outcomes of Antisocial
Adolescent Females. Am J Psychiatry 155:862-870.
Parker G and Roy K (2001): Adolescent depression: a review. Aust N Z J Psychiatry 35:572-580.
124
Patterson GR and Capaldi DM (1990): A mediational model for boys’depressed mood.
In: Risk and protective factors in the development of psychopathology, pp. 141-163. Eds. J Rolf,
AS Masten, D Cicchetti, KH Nuechterlein and S Weintraub, Cambridge University Press, New
York.
Patton GC, Coffey C, Posterino M, Carlin JB and Bowes G (2003): Life events and early onset
depression: cause or consequence? Psychol Med 33:1203-1210.
Pedersen W, Mastekaasa A and Whichstrom L (2001): Conduct problems and early cannabis
initiation: a longitudinal study of gender differences. Addiction 96:415-31.
Peiser NC and Heaven PCL (1996): Family influences on self-reported delinquency among high
school students. J Adolesc19:557-568.
Pelkonen M and Marttunen M (2003): Child and Adolescent Suicide. Epidemiology, Risk Factors,
and Approaches to Prevention. Pediatr Drugs 5:243-265.
Pelkonen M, Marttunen M, Laippala P, Lönnqvist J and Aro H (1997): Psychosocial functioning in
adolescent psychiatric patients: a prospective study on changes in psychosocial functioning among
severely and moderately impaired adolescent out-patients. Acta Psychiatr Scand 96:217-224.
Pelkonen M, Marttunen M and Aro H (2003): Risk for depression: a 6-year follow-up of Finnish
adolescents. J Affect Disord 77:41-51.
Pliszka SR, Sherman JO, Barrow MV and Irick S (2000): Affective Disorder in Juvenile Offenders:
A Preliminary Study. Am J Psychiatry 157:130-132.
Pulkkinen L (1988): Delinquent development: theoretical and empirical considerations. In: Studies
of psychosocial risk: the power of longitudinal data, pp: 184-199. Ed. M Rutter, Cambridge
University Press, Cambridge.
Pulkkinen L, Virtanen T, Af Klinterberg B and Magnusson D (2000): Child behaviour and adult
personality: comparisons between criminality in Finland and Sweden. Crim Behav Ment Health
10:155-169.
125
Raine A, Brennan P and Mednick SA (1994): Birth Complications Combined With Early Maternal
Rejection at Age 1 Year Predispose to Violent Crime at Age 18 Years. Arch Gen Psychiatry
51:984-988.
Raitasalo R (1995): Elämänhallinta sosiaalipolitiikan tavoitteena. Sosiaali- ja terveysturvan
tutkimuksia I. Kansaneläkelaitos, Helsinki.
Raitasalo R and Notkola V (1987): Viljelijäväestön mielenterveyden ongelmien seulonta ja
työterveyshuolto. Sosiaalilääk Aikak 24:232-241.
Rao U, Ryan ND, Birmaher B, Dahl RE, Williamson DE, Kaufman J, Rao R and Nelson B (1995):
Unipolar Depression in Adolescents: Clinical Outcome in Adulthood. J Am Acad Child Adolesc
Psychiatry 38:846-851.
Reitman D, Hummel R, Franz DZ and Gross AM (1998): A review of methods and instruments for
assessing externalizing disorders: theoretical and practical considerations in rendering a diagnosis.
Clin Psychol Rev 18:555-584.
Renaud J, Brent DA, Birmaher B, Chiappetta L and Bridge J (1999): Suicide in Adolescence With
Disruptive Disorders. J Am Acad Child Adolesc Psychiatry 38:846-851.
Renouf AG, Kovacs M and Mukerji P (1997): Relationship of Depressive, Conduct, and Comorbid
Disorders and Social Functioning in Childhood. J Am Acad Child Adolesc Psychiatry 36:9981004.
Rescorla L, Almqvist F, Bird H, Dobrean A, Erol N, Hannesdottir H, Lambert MC, Minaei A,
Novik TS, Roussos A, Simsek Z, Weintraub S, Wolanczyk R, Achenbach TM, Ivanova MY,
Dumenci L, Bilenberg N, Broberg A, Dupfer M, Forns M, Kanbayashi Y, Leung P, Mulatu MS. Oh
K, Sawyer M, Steinhausen H, Metzke CW, Zilber N and Verhulst F (2007): Epidemiological
Comparisons of Problems and Positive Qualities Reported by Adolescents in 24 Countries. J
Consult Clin Psychol 75:351-358.
Rey JM, Walter G, Plapp JM and Denshire E (2000): Family environment in attention deficit
hyperactivity, oppositional defiant and conduct disorders. Aus N Z J Psychiatry 34:453-457.
126
Rey JM, Sawyer MG, Raphael B, Patton G and Lynskey M (2002): Mental health of teenagers who
use cannabis: Results of an Australian survey. Br J Psychiatry 180:216-221.
Rey JM, Sawyer MG and Prior MR (2005): Similarities and differences between aggressive and
delinquent children and adolescents in a national sample. Aust N Z J Psychiatry 39:366-372.
Ristkari T, Sourander A, Ronning J and Helenius J (2006): Self-reported psychopathology, adaptive
functioning and sense of coherence, and psychiatric diagnosis among young men. Soc Psychiatry
Psychiatr Epidemiol 41:523-531.
Ritakallio M, Kaltiala-Heino R, Pelkonen M ja Marttunen M (2003): Miten ehkäistä nuorten
käytöshäiriöitä? Duodecim 119:1752-1760.
Roberts RE, Lewinsohn PM and Seeley JR (1995): Symptoms of DSM-III-R Major Depression in
Adolescence: Evidence from an Epidemiological Survey. J Am Acad Child Adolesc Psychiatry
34:1608-1617.
Roberts RE, Attkinsson CC and Rosenblatt A (1998): Prevalence of Psychopathology Among
Children and Adolescents. Am J Psychiatry 155:715-725.
Rohde P, Lewinsohn PM and Seeley JR (1991): Comorbidity of Unipolar Depression: II.
Comorbidity With Other Mental Disorders in Adolescents and Adults. J Abnorm Psychol 100:214222.
Romano E, Tremblay RE, Vitaro F, Zoccolillo M and Pagani L (2001): Prevalence of Psychiatric
Diagnoses and the Role of Perceived Impairment: Findings from a Adolescent Community Sample.
J Child Psychol Psychiatry 42:451-461.
Rowe R, Maughan B, Pickles A, Costello EJ and Angold A (2002): The relationship between DSMIV oppositional defiant disorder and conduct disorder: findings from the Great Smoky Mountains
Study. J Child Psychol Psychiatry 43:365-373.
Rowe R, Maughan B and Eley TC (2006): Links Between Antisocial Behavior and Depressed
Mood: The Role of Life Events and Attributional Style. J Abnorm Child Psychol 34:293-302.
127
Roza SJ, Hofstra MB, van der Ende J and Verhulst FC (2003): Stable Prediction of Mood and
Anxiety Disorders Based on Behavioral and Emotional Problems in Childhood: A 14-Year FollowUp During Childhood, Adolescence, and Young Adulthood. Am J Psychiatry 160:2116-2121.
Ruchkin VV, Schwab-Stone M, Koposov RA, Vermeiren R and King RA (2003): Suicidal ideations
and attempts in juvenile delinquents. J Child Psychol Psychiatry 44:1058-1066.
Rushton JL, Forcier M and Schectman RM (2002): Epidemiology of Depressive Symptoms in the
National Longitudinal Study of Adolescent Health. J Am Acad Child Adolesc Psychiatry 41:199205.
Rutter M (2001): Conduct disorder: future directions. An afterword. In: Conduct disorders in
childhood and adolescence, pp. 553-572. Eds. J Hill and B Maughan, Cambridge University Press,
New York.
Rutter M, Giller H and Hagell A (1998): Antisocial Behavior by Young People. Cambridge
University Press, New York.
Rutter M, Kim-Cohen J and Maughan B (2006): Continuities and discontinuities in
psychopathology between childhood and adult life. J Child Psychol Psychiatry 47:276-295.
Sailas ES, Feodoroff B, Virkkunen M and Wahlbeck K (2005): Mental disorders in prison
populations aged 15-21: national register study of two cohorts in Finalnd. BMJ 330:1364-1365.
Sanford M, Szatmari P, Spinner M, Munroe-Blum H, Jamieson E, Walsh C and Jones D (1995):
Predicting the One-Year Course of Adolescent Major Depression. J Am Acad Child Adolesc
Psychiatry 34:1618-1628.
Satterfield JH and Schell A (1997): A Prospective Study of Hyperactive Boys With Conduct
Problems and Normal Boys: Adolescent and Adult Criminality. J Am Acad Child Adolesc
Psychiatry 36:1726-1735.
Savolainen J, Messner SF and Kivivuori J (2000): Crime is a Part of the Problem: Contexts of
Lethal Violence in Finland and the USA. J Scand Stud Crim Crime Prev 1:41-55.
128
Schraedley PK, Gotlib IH and Hayward C (1999): Gender Differences in Correlates of Depressive
Symptoms in Adolescents. J Adolesc Health 25:98-108.
Scott S, Knapp M, Henderon J and Maughan B (2001): Financial cost of social exclusion: follow up
study of antisocial children into adulthood. BMJ 323:191-194.
Shaffer D, Fisher P, Dulcan M, Davies M, Piacentini J, Schwab-Stone M.E, Lahey BB, Bourdon K,
Jensen PS, Bird H, Canino G and Regier DA (1996): The NIMH Diagnostic Interview Schedule for
Children Version 2.3 (DISC-2.3): Description, Acceptability, Prevalence Rates, and Performance in
the MECA Study. J Am Acad Child Adolesc Psychiatry 35:865-877.
Sheeber L, Hops H, Alpert A, Davis B and Andrews J (1997): Family Support and Conflict:
Prospective Relations to Adolescent Depression. J Abnorm Child Psychol 25:333-344.
Sihvola E, Keski-Rahonen A, Dick DM, Pulkkinen L, Rose RJ, Marttunen M and Kaprio J (2007):
Minor depression in adolescence: Phenomenology and clinical correlates. J Affect Disord 97:211218.
Silverthorn P and Frick PJ (1999): Developmental pathways to antisocial behavior: The delayedonset pathway in girls. Dev Psychopathol 11:101-126.
Simic M and Fombonne E (2001): Depressive conduct disorder: symptom patterns and correlates in
referred children and adolescents. J Affect Disord 62:175-185.
Simonoff E (2001): Genetic influences on conduct disorder. In: Conduct disorders in childhood and
adolescence, pp. 202-234. Eds. J Hill and B Maughan, Cambridge University Press, New York.
Simonoff E, Pickles A, Meyer JM, Silberg JL, Maes HH, Loeber R, Rutter M, Hewitt JK and Eaves
LJ (1997): The Virginia Twin Study of Adolescent Behavioral Development: Influences of Age,
Sex, and Impairment on Rates of Disorder. Arch Gen Psychiatry 54:801-808.
Simonoff E, Elander J, Holmshaw J, Pickles A, Murray R and Rutter M (2004): Predictors of
antisocial personality. Continuities from childhood to adult life. Br J Psychiatry, 184:118-127.
129
Simons RL, Lin K, Gordon LC, Conger RD and Lorenz FO (1999): Explaining the Higher
Incidence of Adjustment Problems Among Children of Divorce Compared with Those in TwoParent Families. J Marriage Fam 61:1020-1033.
Smith DJ (1995): Youth Crime and Conduct Disorders: Trends, Patterns, and Causal Explanations.
In: Psychosocial Disorders in Young People. Time Trends and Their Causes, pp. 389-489. Eds. M
Rutter and DJ Smith, Wiley & Sons Ltd, West Sussex.
Song L, Singer MI and Anglin TM (1998): Violence Exposure and Emotional Trauma as
Contributors to Adolescents’Violent Behaviors. Arch Pediatr Adolesc Med 152:531-536.
Sourander A, Helstelä L and Helenius H (1999): Parent-adolescent agreement on emotional and
behavioral problems. Soc Psychiatry Psychiatric Epidemiol 34:657-663.
Sourander A, Santalahti P, Haavisto A, Piha J, Ikäheimo K and Helenius H (2004): Have There
Been Changes in Children’s Psychiatric Symptoms and Mental Health Service Use? A 10-Year
Comparison From Finland. J Am Acad Child Adolesc Psychiatry 43:1134-1145.
Sourander A, Elonheimo H, Niemelä S, Nuutila A, Helenius H, Sillanmäki L, Piha J, Tamminen T,
Kumpulainen K, Moilanen I and Almqvist F (2006): Childhood Predictors of Male Criminality: A
Prospective Population-Based Follow-up Study From Age 8 to Late Adolescence. J Am Acad Child
Adolesc Psychiatry 45:578-586.
Sourander A, Jensen P, Davies M, Niemelä S, Elonheimo H, Ristkari T, Helenius H, Sillanmäki L,
Piha J, Kumpulainen K, Tamminen T, Moilanen I and Almqvist F (2007): Who Is at Greatest Risk
of Adverse Long-Term Outcomes? The Finnish From a Boy to a Man Study. J Am Acad Child
Adolesc Psychiatry 46:1148-1161.
STAKES (1997): Psykiatrian luokituskäsikirja. Tautiluokitus ICD-10. Psykiatriaan liittyvät
diagnoosit. Stakes, Ohjeita ja luokituksia 1997:4. Kirjapaino Oy West Point, Rauma.
Stanger C, Achenbach TM and Verhulst FC (1997): Accelerated longitudinal comparisons of
aggressive versus delinquent syndromes. Dev Psychopathol 9:43-58.
130
Steinberg L and Morris AS (2001): Adolescent Development. Annu Rev Psychol 52:83-110.
Steinhausen H and Winkler Metzke C (2000): Adolescent Self-Rated Depressive Symptoms in a
Swiss Epidemiological Study. J Youth Adolesc 29:427-440.
Storvoll EE and Wichstrom L (2002): Do the risk factors associated with conduct problems in
adolescents vary according to gender? J Adolesc 25:183–202.
Storvoll EE and Wichstrom L (2003): Gender differences in changes in and stability of conduct
problems from early adolescence to early adulthood. J Adolesc 26:413-429.
Stouthamer-Loeber M and Loeber R (2002): Lost opportunities for intervention: undetected
markers for the development of serious juvenile delinquency. Crim Behav Ment Health 12:69-82.
Stouthamer-Loeber M, Wei E, Loeber R and Masten AS (2004): Desistance from persistent serious
delinquency in the transition to adulthood. Dev Psychopathol 16:897-918.
Stranger C, Higgins ST, Bickel WK, Elk R, Grabowski J, Schmitz J, Amass L, Kirby KC and
Seracini AM (1999): Behavioral and Emotional Problems Among Children of Cocaine- and OpiateDependent Parents. J Am Acad Child Adolesc Psychiatry 38:421-428.
Taylor J, Malone S, Iacono WG and McGue M (2002): Development of Substance Dependence in
Two Delinquent Subgroups and Nondelinquents From a Male Twin Sample. J Am Acad Child
Adolesc Psychiatry 41:386-393.
Teplin LA, Abram KM, McClelland GM, Duclan MK and Mericle AA (2002): Psychiatric
Disorders in Youth in Juvenile Detention. Arch Gen Psychiatry 59:1133-1143.
Tiet QQ, Wasserman GA, Loeber R, McReynolds LS and Miller LS (2001): Developmental and
Sex Differences in Types of Conduct Problems. J Child Fam Stud 10:181-197.
Timmons-Mitchell J, Brown C, Schulz SC, Webster SE, Underwood LA and Semple WE (1997):
Comparing the Mental Health Needs of Female and Male Incarcerated Juvenile Delinquent. Behav
Sci Law 15:195-202.
131
Todd RD and Botteron KN (2002): Etiology and genetics of early-onset mood disorders. Child
Adolesc Psychiatric Clin N Am 11:499-518.
Toolan JM (1962): Depression in children and adolescents. Am J Orthopsychiatry 34:404- 415.
Torikka A, Kaltiala-Heino R, Rimpelä A, Rimpelä M and Rantanen P (2001): Depression, drinking,
and substance use among 14 to 16-year-old Finnish adolescents. Nord J Psychiatry 55:351-357.
Tramontina S, Martins S, Michalowski MB, Ketzer CR, Eizirik M, Biedeman J and Rohde LA
(2001): School Dropout and Conduct Disorder in Brazilian Elementary School Students. Can J
Psychiatry 46:941-947.
Tremblay RE, Phil RO, Vitaro F and Dobkin PL (1994): Predicting Early Onset of Male Antisocial
Behavior From Preschool Behavior. Arch Gen Psychiatry 51:732-739.
Twenge JM and Nolen-Hoeksema S (2002): Age, Gender, Race, Socioeconomic Status, and Birth
Cohort Differences on the Children’s Depression Inventory: A Meta-Analysis. J Abnorm Psychol
111:578-588.
Ulzen T P and Hamilton H (1998): The nature and characteristics of psychiatric comorbidity in
incarcerated adolescents. Can J Psychiatry 43:57–63.
Verhulst FC, van der Ende JM, Ferdinand RF and Kasius MC (1997): The Prevalence of DSM-IIIR Diagnoses in a National Sample of Dutch Adolescents. Arch Gen Psychiatry 54:329-336.
Verhulst FC, Achenbach TM, van der Ende J, Erol N, Lambert MC, Leung PWL, Silva MA, Zilber
N and Zubrik SR (2003): Comparisons of Problems Reported by Youths from Seven Countries. Am
J Psychiatry 160:1479-1485.
Vermeiren R (2003): Psychopathology and delinquency in adolescents: a descriptive and
developmental perspective. Clin Psychol Rev 23:277-318.
Vermeiren R, De Clippele A and Deboutte D (2000): A descriptive survey of Flemish delinquent
adolescents. J Adolesc 23:277–285.
132
Vermeiren R, Deboutte D, Ruchkin V and Schwab-Stone M (2002a): Antisocial behaviour and
mental health: findings from three communities. Eur Child Adolesc Psychiatry 11:168-175.
Vermeiren R, Schwab-Stone M, Ruchkin V, De Clippele A and Deboutte D (2002b): Predicting
Recidivism in Delinquent Adolescents From Psychological and Psychiatric Assessment. Compr
Psychiatry 43:142-149.
Waddell C, Lipman E and Offord D (1999): Conduct Disorder: Practice Parameters for Assessment,
Treatment, and Prevention. Can J Psychiatry Suppl 2, 44:35S-40S.
Walker L, Merry S, Watson PD, Robinson E, Crengle S and Shaaf D (2005): The Reynolds
Adolescent Depression Scale in New Zealand adolescents. Aust N Z J Psychiatry 39:136-140.
Wasserman GA, Miller LS, Pinner E and Jaramillo B (1996): Parenting Predictors of Early Conduct
Problems in Urban, High-Risk Boys. J Am Acad Child Adolesc Psychiatry 35:1227-1236.
Weisbrot DM and Ettinger AB (2002): Aggression and violence in mood disorders. Child Adolesc
Psychiatric Clin N Am 11:649-671.
Weissman MM, Wolk S, Golstein RB, Moreau D, Adams P, Greenwald S, Klier C.M, Ryan ND,
Dahl RE and Wickramaratne P (1999): Depressed Adolescents Grown Up. JAMA 281:1707-1713.
Weist MD, Paskewitz DA, Jackoson CY and Jones D (1998): Self-Reported Delinquent Behavior
and Psychosocial Functioning in Inner-City Teenagers: A Brief Report. Child Psychiatry Hum Dev
28:241-248.
Weller EB and Weller RA (2000): Depression in adolescents: growing pains or true morbidity? J
Affect Disord 61:S9-S13.
West P, Sweeting H, Der G, Barton J and Lucas C (2003): Voice-DISC Identified DSM-IV
Disorders Among 15-Year-Olds in the west of Scotland. J Am Acad Child Adolesc Psychiatry
42:941-949.
133
Wichstrom L (1999): The Emergence of Gender Differences in Depressed Mood During
Adolescence: The Role of Intensified Gender Socialization. Dev Psychol 35:232-245.
Wichstrom L, Skogen K and Oia T (1996): Increased Rate of Conduct Problems in Urban Areas:
What is the Mechanism? J Am Acad Child Adolesc Psychiatry 35:471-479.
Wiesner M (2003): A Longitudinal Latent Variable Analysis of Reciprocal Relations Between
Depressive Symptoms and Delinquency During Adolescence. J Abnorm Psychol 112:633-645.
Wittchen H-U, Nelson CB and Lachner G (1998): Prevalence of mental disorders and psychosocial
impairments in adolescents and young adults. Psychol Med 28:109-126.
Wolff JC and Ollendick TH (2006): The Comorbidity of Conduct Problems and Depression in
Childhood and Adolescence. Clin Child Fam Psychol Rev 9:201-20.
Wu P, Hoven CW, Bird HR, Moore RE, Cohen P, Alegria M, Dulcan MK, Goodman SH, McCue
Horwitz S, Lichtman JH, Narrow WE, Rae DS, Regier DA and Roper MT (1999): Depressive and
Disruptive Disorders and Mental Health Service Utilization in Children and Adolescents. J Am
Acad Child Adolesc Psychiatry 38:1081-1090.
Yorbik O, Birmaher B, Axelson D, Williamson DE and Ryan ND (2004): Clinical Characteristics
of Depressive Symptoms in Children and Adolescents With Major Depressive Disorder. J Clin
Psychiatry 65:1654-1659.
Zalsman G, Brent DA and Weersing VR (2006): Depressive Disorders in Childhood and
Adolescence: An Overview. Epidemiology, Clinical Manifestation and Risk Factors. Child Adolesc
Psychiatric Clin N Am 15:827-841.
Zhang S, Benson T and Deng X (2000): A test-retest reliability assessment of the international selfreport delinquency instrument. J Crim Justice 28:283-295.
134
12. ORIGINAL PUBLICATIONS
ARTICLE IN PRESS
Journal of
Adolescence
Journal of Adolescence 28 (2005) 155–159
www.elsevier.com/locate/jado
Brief report: Delinquent behaviour and depression in middle
adolescence: a Finnish community sample
Minna Ritakallioa,, Riittakerttu Kaltiala-Heinob, Janne Kivivuoric,
Matti Rimpeläd
a
Tampere School of Public Health, University of Tampere and Tampere University Hospital, Adolescent ward,
Tampere, Finland
b
Psychiatric Treatment and Research Unit for Adolescent Intensive Care (EVA), Tampere University Hospital,
Tampere, Finland
c
National Research Institute of Legal policy, Criminological Unit, Helsinki, Finland
d
National Research and Development Centre for Welfare and Health (STAKES), Helsinki, Finland
Abstract
A large number (N 50 569) of 14–16 year old Finnish adolescents taking part in the School Health
Promotion Study were surveyed for delinquent behaviour in relation to depression. The results indicate a
robust association between delinquency and depression. Among girls risk for depression varied between 1.3
and 3.1 according to various antisocial behaviours (the equivalent risk among boys was 1.3–2.5).
Depression increased according to the frequency of delinquent behaviour. The results emphasize the
importance of comprehensive clinical assessment and treatment of delinquent adolescents.
r 2004 The Association for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights
reserved.
Introduction
The association between delinquent behaviour and depression is well established among
sentenced or incarcerated adolescents. These studies have shown a high prevalence of major
Corresponding author.
E-mail address: minna.ritakallio@uta.fi (M. Ritakallio).
0140-1971/$30.00 r 2004 The Association for Professionals in Services for Adolescents. Published by Elsevier Ltd. All
rights reserved.
doi:10.1016/j.adolescence.2004.07.002
ARTICLE IN PRESS
156
M. Ritakallio et al. / Journal of Adolescence 28 (2005) 155–159
depression (10–30%). Depression is more prevalent among delinquent girls than boys. (Duclos,
Beals, Novins, Martin, Jewett, & Manson, 1998; Ulzen & Hamilton, 1998; Teplin, Abram,
McClelland, Duclan, & Mericle, 2002) Only a minority of adolescents behaving delinquently are
convicted or incarcerated (Rutter, Giller, & Hagell, 1998). In a Finnish study 10% of those
adolescents who reported that they had stolen or beaten up someone in the past year, were caught
by the police (Kivivuori, 2002). Thus as the group of incarcerated adolescents represents only part
of all delinquent adolescents, there is a need for studies conducted in general population.
Few studies have investigated the relationship between delinquent behaviour and depression in
a general adolescent population. These studies have suggested increased risk of depression among
delinquent adolescents (Loeber, Stouthamer-Loeber, & White, 1999; Vermeiren, 2002). Most of
these studies have been conducted in the United States, in Europe only a few studies have focused
on the relationship between delinquency and depression in community samples. Because cultural
and social differences may influence the association between delinquency and depression, it is
essential to assess this relationship in different countries.
The purpose of current study was to investigate the relationship between self-reported
delinquent behaviour and depression in a large community-based middle-adolescence sample.
Methods
The subjects of the study were adolescents aged 14–16 years who participated in 2002 in the
School Health Promotion Study. The survey was conducted in three of the six administrative
divisions in Finland. These areas include urban and rural municipalities of different sizes; thus the
data of the current study is a very representative sample of Finnish adolescents. The survey was
completed anonymously during a school lesson (8th and 9th grades) under the supervision of
teachers. 53 524 adolescents (81% of the pupils) responded to the survey. 2 955 of the responses
(5.5% of the respondents) were excluded from the study because of the incomplete responses on
the depression scale. Thus the final sample for the present study was 50 569 adolescents (50.1%
boys). Respondents’ mean age was 15.3 years (std 0.6).
Depression was assessed by a Finnish modification of the short version of Beck Depression
Inventory (R-BDI)(Beck & Beck, 1972; Raitasalo, 1995). R-BDI is a self-reported instrument
measuring the intensiveness of depressive symptoms. Originally R-BDI contains 13 items, but in
the current study one item concerning suicidal ideation was excluded because the Finnish Ministry
of Education decided not to allow the use of this item for fear that asking about suicidality might
provoke it in adolescents. Each of remaining twelve item was scored 0–3 according to the
intensiveness of the symptom. Sum scores of depression were dichotomised to none/mild
depression (0–7) and moderate/severe depression (8–36) (Beck & Beck, 1972).
Involvement in delinquent behaviour was measured by six different forms of delinquent
behaviour by asking regarding each the question ‘‘During the past twelve months have you
written/drawn graffiti on walls, buses, bus shelters, windows or other similar places (similarly:
deliberately damaged or destroyed the property of the school or the school building/deliberately
damaged or destroyed the property of other public places than school/stolen something from
shops or from stalls/taken part in a fight/beaten someone up)? Response categories were ‘‘never’’,
‘‘once to four times’’ and ‘‘five times or more’’. The preliminary analysis revealed that the two
ARTICLE IN PRESS
M. Ritakallio et al. / Journal of Adolescence 28 (2005) 155–159
157
vandalism variables and similarly the two violent behaviour variables greatly overlap. Therefore
the two questions about vandalism were combined into one variable (vandalism), and the two
questions about violent behaviour were combined into one (violent behaviour).
Age, sex, family structure, parental educational level, parental unemployment during the past
year, years of living in the same municipality and degree of urbanization of place of residence were
investigated as the sociodemographic variables.
Data was analysed by logistic regression in which delinquent behaviours were entered
simultaneously as independent variables and depression as dependent variable. Sociodemographic
variables and age were controlled. Analyses were conducted separately for boys and girls using
SPSS for Windows version 11.0.
Results
Of the boys 7.5% (1899/25 352) and of the girls 15.7% (3960/25 217) reported symptoms
indicating to moderate or severe depression (po0:001). All the studied delinquent behaviours
were independently associated with depression in both sexes after controlling for age and
sociodemographic factors (Table 1). Among girls risk for depression varied between 1.3 and 3.1
according to various antisocial behaviours (the equivalent risk among boys was 1.3–2.5). The
strongest associations with depression were detected for frequent vandalism and violent
behaviour. Depression increased according to the frequency of delinquent behaviour in both
sexes. In addition to delinquent behaviour, depression was associated in both sexes with living
apart from parents, low parental education level, unemployment of both parents and living in the
area for less than 1 year.
Discussion
The present study confirmed on the general population level the association between delinquent
behaviour and depression previously found in incarcerated adolescents. Depression was the more
common the more frequently the adolescent had been involved in the delinquent acts studied, but
it is worth noting that the risk of depression was increased even when there had been one single
offence.
The association between antisocial behaviour and depression is a complex phenomenon and
there are conflicting opinions on the causal relationship between these features. Delinquent
behaviour can be considered a result of depression so that delinquency is one form of acting out,
or depression can be considered a consequence of criminal behaviour. It has been suggested that
depression among incarcerated adolescents may be associated with the incarceration and not with
delinquency per se (Vermeiren, De Clippele, & Deboutte, 2000), but the present results cannot be
explained by this.
Some limitations of the present study must be conceded. Depression was evaluated with a selfreported questionnaire (R-BDI). Although the R-BDI is not a diagnostic instrument, its quality as
a screening method in assessing depressive symptoms has been proved good. (Bennett, Ambrosini,
Bianchi, Barnett, Metz, & Rabinovich, 1997). Adolescents who score moderately or severely
ARTICLE IN PRESS
158
M. Ritakallio et al. / Journal of Adolescence 28 (2005) 155–159
Table 1
Risk of depression [Odds ratio (OR), 99% confidence interval (CI)] according to delinquent behaviour among 14- to 16year-old Finnish adolescents by sex when age and sociodemographic factors are controlled for
Girls
Boys
OR
CI 99%
OR
CI 99%
Graffiti writing/drawing
No
Once to four times
Five or more times
1.0
1.5
1.5
1.3–1.7
1.1–2.1
1.0
1.3
1.8
1.1–1.6
1.3–2.3
Shoplifting
No
Once to four times
Five or more times
1.0
1.3
1.4
1.1–1.5
1.1–1.7
1.0
1.2
1.2
1.0–1.5
1.0–1.7
Vandalism
No
Once to four times
Five or more times
1.0
1.6
3.1
1.4–1.9
2.2–4.5
1.0
1.5
2.3
1.3–1.8
1.7–3.1
Violent behaviour
No
Once to four times
Five or more times
1.0
1.8
1.9
1.5–2.1
1.4–2.8
1.0
1.3
2.5
1.1–1.6
2.0–3.2
Family structure
Nuclear family
Step-parent family
Single-parent family
Living apart from parents
1.0
1.3
1.2
1.7
1.1–1.5
1.1–1.4
1.0–2.6
1.0
1.4
1.3
2.4
1.1–1.7
1.1–1.6
1.4–3.7
Parental education
Academic
Vocational
Basic school
1.0
1.1
1.4
1.0–1.2
1.1–1.7
1.0
1.0
1.5
0.8–1.1
1.2–1.9
Parental unemployment
None
One parent
Both parents
1.0
1.3
1.8
1.2–1.5
1.4–2.2
1.0
1.3
2.1
1.1–1.5
1.5–2.8
Years living in the same area
9 years or more
5–9 years
1–4 years
Under 1 year
1.0
1.0
1.1
1.5
0.9–1.2
1.0–1.3
1.2–2.0
1.0
1.0
1.1
1.8
0.8–1.2
0.8–1.3
1.3–2.5
Degree of urbanization
Rural
Semi-urban
Urban
1.0
0.9
1.0
0.8–1.1
0.8–1.1
1.0
0.9
1.0
0.7–1.1
0.8–1.2
ARTICLE IN PRESS
M. Ritakallio et al. / Journal of Adolescence 28 (2005) 155–159
159
depressed can be considered to be in need of clinical attention. The evaluation of delinquent
behaviour was also based on self-reported information. Although there are known limitations in
the reliability of self-reported delinquency, such as dishonesty, self-report methods are generally
considered valid methods in assessing delinquency. Because the hidden criminality is estimated to
be substantial and official reports cover only a small fraction of all delinquency, self-report
methods offer sufficient reliability for assessing delinquent behaviour (Rutter et al., 1998;
Kivivuori, 2002).
The present study offers implications for both the prevention and treatment of depression and
delinquent behaviour. Health care professionals have a special challenge in recognizing depression
among delinquent adolescents. Special attention should be paid to those young people who
repeatedly commit offences because they are at considerable risk of depression. Comprehensive
clinical assessment and treatment of delinquent adolescents is needed. On the other hand,
delinquent behaviour should be screened among depressed adolescents, to prevent marginalization.
References
Beck, A. T., & Beck, R. W. (1972). Screening depressed patients in family practice. A rapid technic. Postgraduate
Medicine, 52, 81–85.
Bennett, D. S., Ambrosini, P. J., Bianchi, M., Barnett, D., Metz, C., & Rabinovich, H. (1997). Relationship of beck
depression inventory factors to depression among adolescents. Journal of Affective Disorders, 45, 127–134.
Duclos, C. W., Beals, J., Novins, D. K., Martin, C., Jewett, C. S., & Manson, S. (1998). Prevalence of common
psychiatric disorders among American Indian adolescent detainees. Journal of the American Academy of Child &
Adolescent Psychiatry, 37, 866–873.
Kivivuori, J. (2002). Nuoret rikosten tekijöinä ja uhreina. Oikeuspoliittinen tutkimuslaitos. Julkaisuja 188. Helsinki:
Hakapaino Oy.
Loeber, R., Stouthamer-Loeber, M., & White, H. R. (1999). Developmental aspects of delinquency and internalizing
problems and their association with persistent juvenile substance use between ages 7 and 18. Journal of Clinical Child
Psychology, 28, 322–332.
Raitasalo, R. (1995). Elämänhallinta sosiaalipolitiikan tavoitteena. Sosiaali- ja terveysturvan tutkimuksia I. Helsinki:
Kansaneläkelaitos.
Rutter, M., Giller, H., & Hagell, A. (1998). Antisocial behavior by young people. New York: Cambridge University
Press.
Teplin, L. A., Abram, K. M., McClelland, G. M., Duclan, M. K., & Mericle, A. A. (2002). Psychiatric disorders in
youth in juvenile detention. Archives of General Psychiatry, 59, 1133–1143.
Ulzen, T. P., & Hamilton, H. (1998). The nature and characteristics of psychiatric comorbidity in incarcerated
adolescents. Canadian Journal of Psychiatry, 43, 57–63.
Vermeiren, R. (2002). Delinquents disordered? Psychopathology and neuropsychological deficits in delinquent adolescents.
Antwerp: Univerisity of Antwerp.
Vermeiren, R., De Clippele, A., & Deboutte, D. (2000). A descriptive survey of Flemish delinquent adolescents. Journal
of Adolescence, 23, 277–285.
1
Comorbidity between depression and antisocial behaviour in middle adolescence:
the role of perceived social support
Authors: Minna Ritakallio, Tiina Luukkaala, Mauri Marttunen, Mirjami Pelkonen and
Riittakerttu Kaltiala-Heino
2
Comorbidity between depression and antisocial behaviour in middle adolescence:
the role of perceived social support
Abstract
A sample of 3, 278 students (aged 15 to 16 years) were surveyed to investigate the
interrelations between depression (RBDI; The Finnish modification of the 13-item Beck
Depression Inventory), antisocial behaviour (YSR; Youth Self-Report) and perceived
social support from family, friends and significant other (PSSS-R; The Perceived Social
Support Scale-Revised). The study aimed especially to control for the effect of
perceived social support in the comorbidity between depression and antisocial
behaviour. The results indicate firstly that decreased perceived social support was
associated with both depression and antisocial behaviour in both sexes. Similarly,
antisocial behaviour was associated with depression in both sexes. Perceived social
support modified comorbidity between depression and antisocial behaviour, but
antisocial behaviour still remained independently associated with depression. Thus,
perceived social support did not explain the found comorbidity between depression and
antisocial behaviour.
Key words: comorbidity, depression, antisocial behaviour, perceived social
support, adolescence
3
Comorbidity between depression and antisocial behaviour in middle adolescence:
the role of perceived social support
Authors:
Minna Ritakallio, M.Sc. (1,2)
(1) University of Tampere, Tampere School of Public Health, Tampere, Finland
(2) Tampere University Hospital, Adolescent ward, Tampere, Finland
Tiina Luukkaala, M.Sc. (1,2)
(1) Tampere University Hospital, Research Unit, Tampere, Finland
(2) University of Tampere, Tampere School of Public Health, Tampere, Finland
Mauri Marttunen, M.D., Ph.D. (1,2,3)
(1) Department of Adolescent Psychiatry, University of Kuopio, Kuopio, Finland
(2) National Public Health Institute, Department of Mental Health and Alcohol
Research, Helsinki, Finland
(3) Peijas Hospital, Department of Adolescent Psychiatry, Helsinki University Hospital,
Vantaa, Finland
Mirjami Pelkonen, Ph.D. (1,2,3)
(1) National Public Health Institute, Department of Mental Health and Alcohol
Research, Helsinki, Finland (Senior Researcher)
(2) University of Tampere, Tampere School of Public Health, Tampere, Finland (Senior
Lecturer in Social Psychiatry)
4
(3) Department of Adolescent Psychiatry, Peijas Hospital, Helsinki University Hospital,
Vantaa, Finland (Consulting Psychologist)
Riittakerttu Kaltiala-Heino, MD, DrMedSci (1,2)
(1) Chief Psychiatrist, Director of Department of Adolescent Psychiatry, Tampere
University Hospital, Tampere, Finland
(2) Tampere School of Public Health, University of Tampere, Tampere, Finland
Correspondence:
Minna Ritakallio
Tampere School of Public Health
33104 University of Tampere
Finland
E-Mail: [email protected]
5
Introduction
Several studies have revealed that depression and antisocial behaviour associate in
adolescence. Among sentenced or incarcerated adolescents the prevalence of major
depression has ranged from 10% to 30 % (1-3). Similarly, in community samples rates
of comorbidity between depression and antisocial behaviour have ranged from 8% to
45% (4-7). In a recent community study the odds ratio (OR) for depression varied from
1.3 to 3.1 according to various forms of antisocial acts (8). Although antisocial
behaviour itself is more common among boys than girls (9), depression is more
prevalent among antisocial girls than antisocial boys (2-3). The risk for depression
seems to increase in both sexes according to the frequency and severity of antisocial
behaviour (10,8). However, the underlying mechanism of comorbidity between
adolescent depression and antisocial behaviour are yet unclear. There are basically at
least three possible models to consider. Antisocial behaviour can be considered to be a
result of depression so that antisocial behaviour is one form of acting out or depression
can be considered as a consequence of antisocial behaviour accompanied with social
failure experiences. Thirdly, comorbidity between depression and antisocial behaviour
may result from common or shared risk factors (11-15).
Lack of social support is associated with adolescent depression. Although the
significance of the family as a support source decreases during the adolescent
development, and also other forms of social support (e.g. social support from friends
and teachers) are inversely related to depression, the social support from parents seems
to have the most efficient effect in buffering adolescents against depression (16-20).
6
Some of the previous studies have found that girls are more vulnerable to decreased
social support than boys in relation to depression (21,18), but others have found no
gender differences in the association between emotional problems and decreased social
support (22,20). In addition to depression, social support is associated also with
antisocial behaviour. Previous studies have shown that antisocially behaving
adolescents receive less social support and have more negative family relations than
non-antisocial counterparts (23-26). As in the case of depression, it seems that
particularly social support from parents buffers adolescents most efficiently against
antisocial behaviour. According to the review of Deptula & Cohen (27) there are
inconsistent results concerning psychological closeness of delinquents’friendships.
Some studies have found that antisocial adolescents receive less social support from
their friends than non-antisocial adolescents, but others have not found any differences
between these groups in the closeness of friend relations (27).
In sum, given that low social support is associated both with adolescent depression and
antisocial behaviour, more detailed research on the role of social support in the
comorbidity between depression and antisocial behaviour is indicated. Apparently, no
previous study has investigated whether social support mediates the found comorbidity
between depression and antisocial behaviour or is the comorbidity independent of social
support. Therefore, the purpose of the current study was, firstly to investigate whether
perceived social support from family, friends and significant other person are similarly
associated with depression and with antisocial behaviour in middle adolescence.
Secondly, the study aimed to control for the effect of perceived social support in the
comorbidity between depression and antisocial behaviour in order to reveal whether the
7
comorbidity between disorders is independent of perceived social support. Finally, the
study investigated whether the interrelations between depression, antisocial behaviour
and perceived social support occur similarly in both sexes.
Materials and Methods
Study sample and procedures
The present study is a part of an ongoing prospective follow-up study Mental Health in
Adolescence. Ninth grade students (aged 15-16 years) of all the Finnish-speaking
secondary schools in two Finnish cities, Tampere (200 000 inhabitants) and Vantaa (180
000 inhabitants) completed a person-identifiable questionnaire during a school lesson
supervised by a teacher. The subjects were identified in the school registers of the
participating cities. The parents of the subjects were informed in advance by a letter, but
parental consent to participation was not required since the Finnish legislation on
medical research allows a 15-year-old subject to decide alone. The students received
written information on the study and signed written informed consent forms before
completing the questionnaires. The students returned the informed consent forms and
the completed questionnaires in separate closed envelopes. For students absent from the
school on the original survey day, a separate opportunity to participate was offered in
the school within a couple of weeks of the original data collection. For students not
present on either occasion, the questionnaires were sent by post. If there was no
response after a second reminder, we concluded that the student was not willing to
participate.
8
Pupils in the ninth grades of the participating schools totalled 3,809, of whom 3,597
responded (response rate 94.4%). Six respondents had to be excluded due to obvious
facetiousness. Since it is possible for a person under 15 years to be attending the ninth
grade, we later excluded respondents who were not yet 15 years old (n=313). Thus, the
final sample consisted of 1609 girls and 1669 boys. The mean age was 15.5 (sd 0.39).
Of the respondents 71% were living in two-parent families, 15% in single-parent
families, 12% in stepparent families and 1% had some other legal guardian. There were
no statistically significant differences between cities in the family structure variables.
Residential stability was high: 84% of the respondents had been living in the same
municipality for at least five years and 67% had been living there at least nine years.
64% of the respondents had not moved even within the municipality for at least five
years. Highest education of the father was primary/comprehensive in 18%,
primary/comprehensive with vocational education in 39%, high school/high school with
vocational education in 17% and college/university degree in 26% of the respondents.
Highest education of the mother was primary/comprehensive in 15%,
primary/comprehensive with vocational education in 31%, high school/high school with
vocational education in 29% and college/university degree in 25% of the respondents.
At least one of the parents had a college or university degree in 37% of the respondents.
There were 242 (8%) families where both of the parents had completed primary or
comprehensive education only. Low education of the parents was significantly more
common in Vantaa. In 75% of the cases neither of the parents had been unemployed
during the past year. Parental unemployment was significantly more common in
Tampere (28%) than in Vantaa (23%). These sociodemographic variables resemble
9
those conducted previously from a large geographically representative sample of
Finnish adolescents (28-29).
Attrition analysis was not conducted as only very limited information (age) about nonparticipants was available. Therefore it is possible that the rates of depression and
antisocial behaviour may have been underestimated given that non-responders may be
more depressed and behave more antisocially than other pupils.
Measures
A Finnish modification (RBDI, 30) of the 13-item Beck Depression Inventory (BDI, 3132) was used to assess depression. The 13-item BDI is a widely used self-report scale
measuring the severity of depressive symptoms (emotional, cognitive and physical), and
its reliability and validity are good (33). For example, Fountoulakis et al. (34) reported
that the 13-item BDI with cut-off point 8 predicted a SCAN-diagnosis of depression
with good sensitivity (0.93) and specificity (0.88). Further, the Finnish modification
(RBDI) has been shown to be an appropriate method for measuring depression in
Finnish adolescents (35-37). Each of the 13 items (e.g. “I am so sad or unhappy that I
can’t stand it”and “I can’t make any decisions at all anymore”) were scored 0 –3
according to the severity of the symptom. Sum scores (range 0-39) were dichotomised
to none/mild depression (0- 7 scores) and moderate/severe depression (8-39 scores) as
recommended by Beck & Beck (31) and as has been done in the previous analyses of
Finnish adolescent samples (e.g. 20,8).
10
Antisocial behaviour was assessed by the externalizing scale of the Youth Self-Report
(YSR, 38). YSR is based on self-reports from adolescents aged 11-18 years, and it is
widely used in previous studies showing high reliability and validity (38-40). For
example, Morgan & Cauce (41) found that high YSR scores on Externalizing scale
predicted a DISC-C-diagnosis of oppositional defiant or conduct disorder with
respectable rates of sensitivity (0.74) and specificity (0.73). In addition, recent
epidemiological comparisons of the YSR in 24 countries revealed considerable
consistency in adolescents’self-reported problems (42).
The Externalizing scale consists of 11 variables concerning delinquency (e.g. “I lie or
cheat”, “I run away from home”, “I steal at home”) and 18 variables concerning
aggressive behaviour (e.g. “I am mean to others”, “I destroy things belonging to others”,
“I physically attack people”) during the past 6 months. Each variable was scored 0- 2
according to the frequency of antisocial behaviour. Sum scores of the 29 items (range 0
- 58) were dichotomised using the 90th percentile as cut-off into those within normal
range versus those within clinical range. Because antisocial behaviour is more common
among adolescent boys than girls, sex-specific cut-off-points (boys 24.0 and girls 25.0)
were used (38).
Perceived social support was assessed by The Perceived Social Support Scale-Revised
(PSSS-R). PSSS-R is presented by Blumenthal et al. (43), and measures persons’
subjective perceptions of social support and emotional closeness, not actual number of
supportive contacts. It has been shown to be a useful method in assessing perceived
social support in Finnish adolescents (44). PSSS-R contains 12 items on a five-point
11
Likert-type scale. Three factor-analytically derived sum scores (each ranging 4 to 20
scores) were used addressing perceived social support from family (4 items, e.g. “I get
emotional help and support I need from my family”), from friends (4 items, e.g. “My
friends really try to help me”) and from significant other (4 items, e.g. “There is a
special person with whom I can share may joys and sorrows”). High sum scores indicate
high perceived social support. PSSS-R was used as continuous variable in associations
between perceived social support and depression and antisocial behaviour. In
multivariable associations PSSS-R was used as categorical variable dichotomised
according to 25th percentile into those with low level of perceived social support versus
those with high level of perceived social support. This was done because the
distributions of the perceived social support-scores were skew.
Sex, family structure (nuclear family, step-parent family, single-parent family or other
custodian), parental education level (academic, vocational or basic level), parental
unemployment during the past year (none, one parents or both parents) and years of
living in the same municipality (9 years or more, 5-9 years, 1-4 years or less than 1
year) were investigated as sociodemographic variables. Parental educational level was
classified as academic if at least one of the parents had an academic education, as
vocational if at least one of the parents had some vocational training and as basic level if
neither of the parents had not completed any vocational education. These variables were
included as covariates because previous studies have shown them to associate with
depression (6,20,45).
12
Statistical analysis
All analyses were conducted separately for boys and girls. Association between
perceived social support and depression and association between perceived social
support and antisocial behaviour were analysed by Mann-Whitney test, where all 3
types, of perceived social support (family, friends and significant other) were analyzed
separately. Univariate associations between antisocial behaviour and depression was
analysed by Pearson chi-square test and by logistic regression (Model 1). Multivariate
associations between depression as a dependent variable and antisocial behaviour as
independent variable were adjusted by family support, friend support and significant
other support, which were entered separately into the models (Model 2, Model 3 and
Model 4). In the model 5 three perceived social support types were entered
simultaneously to investigate the association of the antisocial behaviour for depression.
Finally family structure, parental education, parental unemployment and stability of
living were added as covariates into the model to analyse the effect of potential
confounding factors (Model 6). Statistical analyses were performed using SPSS for
Windows, version 11.0. (SPSS Inc., Chigaco, IL, USA).
Results
Decreased perceived social support was associated with depression (Table 1). Both
depressed boys and girls reported significantly less social support than non-depressed
boys and girls. The difference in relation to perceived social support between depressed
and non-depressed adolescents was present in all sources of social support (family,
13
friends and significant other), but the difference in the perceived social support was
most evident in both sexes in the decreased social support from family.
All three types of perceived social support were inversely associated with antisocial
behaviour among boys (Table 1). Antisocially behaving boys reported receiving less
social support from family, friends and other significant person than non-antisocial
boys. Among girls, only decreased perceived social support from family was associated
with antisocial behaviour.
Insert Table 1 about here.
Antisocial behaviour was associated with depression in both sexes. Depression was far
more common (p<0.001) among antisocial boys (25.3%) and girls (35.1%) than among
non-antisocial boys (4.5%) and girls (11.1%). Antisocial boys were seven times more
likely to be depressive than non-antisocial boys. Respectively, antisocial girls were four
times more likely to be depressive than non-antisocial girls (Table 2, Model 1).
Adding perceived social support factors separately into the model modified differently
the association between depression and antisocial behaviour. When the association
between depression and antisocial behaviour was adjusted by perceived social support
from family (Model 2), the association of the antisocial behaviour for depression
decreased in both sexes compared to the unadjusted associations. Perceived social
support from friends (Model 3) and perceived social support from significant other
(Model 4) had an opposite effects on the association of the antisocial behaviour for
14
depression as social support adjusted associations were slightly higher in both sexes
than the unadjusted associations. When all perceived social support factors were entered
simultaneously into the model (Model 5), the associations of the antisocial behaviour
for depression were OR 5.3 [95% CI 3.4-8.5] among boys and OR 3.2 [95% CI 2.1-4.7]
among girls. Adding covariates (family structure, parental education level, parental
unemployment during the past year and years of living in the same municipality) into
the model (Model 6) did not remarkable alter associations between depression and
antisocial behaviour (Table 2).
Insert Table 2 about here.
Discussion
Consistent with previous studies (e.g. 17) decreased perceived social support was
associated with adolescent depression. Also in accord with previous studies (e.g. 16),
social support from family seems to have the most efficient buffering effect in relation
to depression. The finding of no gender differences in the associations between
decreased social support and depression is in contrast to some previous studies (e.g. 44),
but in accordance with others (e.g. 22). Perceived social support from family, friends
and significant other were inversely associated with antisocial behaviour among boys.
Among girls, only decreased perceived social support from family was associated with
antisocial behaviour. The finding is similar to the study of Bru et al. (26) reporting that
parental support was negatively associated with antisocial behaviour in both sexes, but
lack of friend support was modestly associated with antisocial behaviour only among
15
boys. These results emphasise the role of parents and family in a socialisation process
and as prosocial role models in preventing antisocial behaviour. Previous findings on
psychological closeness of delinquents’friendships have been mixed (27). In addition,
little attention has been paid to the role of gender in the friendships of antisocial
adolescents as many studies have focused exclusively on boys.
The main purpose of the present study was to control the effect of perceived social
support in the comorbidity between depression and antisocial behaviour. Although
associations between low social support and depression/antisocial behaviour are well
established, apparently no previous study has focused on the interrelations of these
phenomenons. The current study indicated that although perceived social support from
family, friends and significant others modified the comorbidity between depression and
antisocial behaviour, antisocial behaviour still remained independently associated with
depression. In other words, decreased perceived social support did not explain the found
comorbidity between depression and antisocial behaviour.
Different factors of perceived social support (family, friends and significant other)
modified differently the comorbidity between depression and antisocial behaviour, but
in all cases, antisocial behaviour still remained independently associated with
depression. When taking account the perceived social support from family, the risk
effect of antisocial behaviour for depression reduced in both sexes as decreased social
support from family explained part of the depression. In contrast, adjusting by perceived
social support from friends slightly strengthened the risk effect of antisocial behaviour
for depression in both sexes. It seems that antisocial adolescents are even more
16
vulnerable to depression if they simultaneously lack support from friends. However, it
may also be that depressed antisocially behaving adolescents push away offers of social
support or that they do not realise that they actually have social support as a part of
feeling of loneliness. On the other hand, the measure of social support used in this study
measures explicitly adolescents’subjective perceptions of social support and emotional
closeness, not actual number of supportive contacts. Therefore, the adolescents’
subjective perceptions of decreased support are valuable. After controlling by all
sources of perceived social support and the sociodemographic variables, depression was
five times more common among antisocial boys and almost three times more common
among antisocial girls than among non-antisocial counterparts. The finding is in
accordance with previous studies suggesting increased risk of depression among
antisocial adolescents (e.g. 8,10,46).
A major limitation of the current study is the lack of diagnostic interviews and lack of
additional information (e.g. parents, teachers) as both depression and antisocial
behaviour were measured by self-report scales (RBDI and YSR). Exclusive reliance on
self-reports may have influenced the results. However, although RBDI is not a
diagnostic instrument, it has been shown to be efficient and valid instrument in
distinguishing between depressed and non-depressed adolescents (33, 36). Adolescents
scoring moderate to severe on depressive continuum can be considered to be in need of
clinical attention, and these depressive symptoms are likely to reach clinical levels over
time (47). Similarly, although there are known limitations in the reliability of selfreported antisocial behaviour, such as dishonesty, forgetfulness and exaggeration, selfreport methods are generally considered reliable and valid methods in assessing
17
antisocial behaviour (48). In addition to this, the instrument used has shown good
reliability and validity in assessing antisocial behaviour (38-39). Finally, self-report
methods give valuable information on adolescents’subjective perceptions of their own
well-being and behaviour.
Findings of the current study indicate a robust association between depression and
antisocial behaviour. However, given that depressed adolescents may be more likely to
accuse themselves of wrongdoing because of feelings of guilt and self-accusations
related with depression, rates of antisocial behaviour may be overestimated. The study
sample was a large school-based sample of Finnish adolescents, the response rate was
high (94.4%), and thus representative sample of medium-sized Finnish cities. However,
as none of adolescents came from rural areas, the results may not be representative of
the whole Finnish population of adolescents. Although antisocial behaviour was
independently associated with depression, due to the cross-sectional study design no
firm conclusion of a causal relationship can be drawn. Nevertheless, results indicate that
antisocial behaviour and depression associate independently of decreased perceived
social support, and therefore health care professionals and other professionals working
with adolescents have a special challenge in recognizing these multiproblem
adolescents.
The current study offers implications for prevention and treatment of both depression
and antisocial behaviour. The early identification of those adolescents with decreased
perceived social support is central in preventing depression and antisocial behaviour.
The results emphasize the role of the family as source of perceived social support, and
18
therefore it is important to identify families at risk and provide adequate support for
them. Although perceived social support modified the comorbidity between depression
and antisocial behaviour, antisocial behaviour still remained independently associated
with depression and therefore comprehensive clinical assessment and treatment of
antisocial adolescents should be considered.
Acknowledgments: The study was supported by Yrjö Jahnsson Foundation, Pirkanmaa
Hospital District Research Fund and The Finnish Cultural Foundation.
19
References
1. Duclos CW, Beals J, Novins DK, Martin C, Jewett CS, Manson S. Prevalence of
common psychiatric disorders among American Indian adolescent detainees. Journal of
the American Academy of Child & Adolescent Psychiatry 1998; 37: 866-873.
2. Ulzen TP, Hamilton H. The Nature and Characteristics of Psychiatric Comorbidity in
Incarcerated Adolescents. Canadian Journal of Psychiatry 1998;43:57-63.
3. Teplin LA, Abram KM, McClelland GM, Duclan MK, Mericle AA. Psychiatric
disorders in youth in juvenile detention. Archives of General Psychiatry 2002;59:11331143.
4. Capaldi DM. Co-occurrence of conduct problems and depressive symptoms in early
adolescent boys: I. Familial factors and general adjustment at Grade 6. Development
and Psychopathology 1991;3:277-300.
5. Lewinsohn PM, Rohde P, Seeley JR, Hops H. Comorbidity of Unipolar Depression:
I. Major Depression With Dysthymia. Journal of Abnormal Psychology 1991;100:205213.
6. Garnefski N, Diekstra RFW. Adolescents from one parent, stepparent and intact
families: emotional problems and suicide attempts. Journal of Adolescence 1997;20:
201-208.
20
7. Gomes JT, Bertrand LD, Paetsch JJ, Hornick JP. Self-reported delinquency among
Alberta’s youth: findings from a survey of 2,001 junior and senior high school students.
Adolescence 2003;38:75-91.
8. Ritakallio M, Kaltiala-Heino R, Kivivuori J, Rimpelä M. Brief report: delinquent
behaviour and depression in middle adolescence: a Finnish community sample. Journal
of Adolescence 2005;28:155-159.
9. Rutter M, Giller H, Hagell A. Antisocial Behavior by Young People. New York:
Cambridge University Press; 1998.
10. Vermeiren R, Deboutte D, Ruchkin V, Schwab-Stone M. Antisocial behaviour and
mental health. Findings from three communities. European Child & Adolescent
Psychiatry 2002;1:168-175.
11. Capaldi DM. Co-occurrence of conduct problems and depressive symptoms in early
adolescent boys: II. A 2-year follow-up at Grade 8. Development and Psychopathology
1992;4:125-144.
12. Fergusson DM, Lynskey MT, Horwood LJ. Origins of Comorbidity between
Conduct and Affective Disorders. Journal of the American Academy of Child and
Adolescent Psychiatry 1996;35:451-460.
21
13. Overbeek G, Vollebergh W, Meeus W, Engels R, Luijpers E. Course, CoOccurrence, and Longitudinal Associations of Emotional Disturbance and Delinquency
From Adolescence to Young Adulthood: A Six-Year Three-Wave Study. Journal of
Youth and Adolescence 2001;30:401-426.
14. Measelle JR, Stice E, Hogansen JM. Developmental Trajectories of Co-Occurring
Depressive, Eating, Antisocial, and Substance Abuse Problems in Adolescents Females.
Journal of Abnormal Psychology 2006;115:524-538.
15. Wolff JC, Ollendick TH. The Comorbidity of Conduct Problems and Depression in
Childhood and Adolescence. Clinical Child and Family Psychology Review 2006;9:
201-20.
16. Garnefski N, Diekstra RFW. Perceived Social Support from Family, School, and
Peers: Relationship with Emotional and Behavioral Problems among Adolescents.
Journal of the American Academy of Child & Adolescent Psychiatry 1996;35:16571664.
17. Lewinsohn PM, Rohde P, Seeley JR. Major depressive disorder in older
adolescents: prevalence, risk factors, and clinical implications. Clinical Psychology
Review 1998;18:765-794.
22
18. Garnefski N. Age Differences in Depressive Symptoms, Antisocial Behavior, and
Negative Perceptions of Family, School, and Peers Among Adolescents. Journal of the
American Academy of Child & Adolescent Psychiatry 2000;39:1175-1181.
19. Helsen M, Vollebergh W, Meeus W. Social Support from Parents and Friends and
Emotional Problems in Adolescence. Journal of Youth and Adolescence 2000;29:319335.
20. Kaltiala-Heino R, Rimpelä M, Rantanen P, Laippala P. Adolescent depression: the
role of discontinuities in life course and social support. Journal of Affective Disorders
2001;64:155-166.
21. Schraedley PK, Gotlib IH, Hayward C. Gender Differences in Correlates of
Depressive Symptoms in Adolescents. Journal of Adolescent Health 1999;25:98-108.
22. Sheeber L, Hops H, Alpert A, Davis B, Anders J. Family Support and Conflict:
Prospective Relations to Adolescent Depression. Journal of Abnormal Child
Psychology 1997;25:333-344.
23. Peiser NC, Heaven PCL. Family influences on self-reported delinquency among
high school students. Journal of Adolescence 1996;19:557-568.
23
24. Weist MD, Paskewitz DA, Jackoson CY, Jones D. Self-Reported Delinquent
Behavior and Psychosocial Functioning in Inner-City Teenagers: A Brief Report. Child
Psychiatry and Human Development 1998;28:241-248.
25. Brook JS, Brook DW, De La Rosa M, Whiteman M, Montoya I. The Role of
Parents in Protecting Colombian Adolescents From Delinquency and Marijuana Use.
Archives of Pediatrics & Adolescent Medicine 1999;153:457-464.
26. Bru E, Murberg TA, Stephens P. Social support, negative life events and pupil
misbehaviour among young Norwegian adolescents. Journal of Adolescence 2001;24:
715-727.
27. Deptula DP, Cohen R. Aggressive, rejected, and delinquent children and
adolescents: a comparison of their friendships. Aggression and Violent Behavior 2004;
9:75-104.
28. Luopa P, Pietikäinen M and Jokela J. Nuorten elinolot, koulutyö, terveys ja
terveystottumukset 1996 - 2005. Kouluterveyskysely 2005. Stakesin työpapereita
25/2006. Valopaino Oy, Helsinki.
29. Ritakallio M, Kaltiala-Heino R, Kivivuori J, Luukkaala T and Rimpelä M.
Delinquency and the profile of offences among depressed and non-depressed
adolescents. Criminal Behaviour and Mental Health 2006; 16:100–110.
24
30. Raitasalo R. Elämänhallinta sosiaalipolitiikan tavoitteena. Sosiaali- ja terveysturvan
tutkimuksia I. Helsinki: Kansaneläkelaitos; 1995.
31. Beck AT, Beck RW. Screening depressed patients in family practice. A rapid
technic. Postgraduate Medicine 1972;52:81-85.
32. Beck AT, Rial WY, Rickels K. Short form of depression inventory: cross-validation.
Psychological Reports 1974; 34:1184-1186.
33. Bennett DS, Ambrosini PJ, Bianchi M, Barnett D, Metz C, Rabinovich H.
Relationship of Beck Depression Inventory factors to depression among adolescents.
Journal of Affective Disorders 1997;45:127-134.
34. Fountoulakis KN, Iacovides A, Kleanthous S, Samolis S, Gougoulias K, St
Kaprinis, et al. The greek translation of the symptoms rating scale for depression and
anxiety: preliminary results of the validation study. BMC Psychiatry 2003;3:21,
doi:10.1186/1471-244X-3-21.
35. Hänninen V, Aro H. Sex differences in coping and depression among young adults.
Social Science of Medicine 1996;43:1453-1460.
36. Kaltiala-Heino R, Rimpelä M, Rantanen P, Laippala P. Finnish modification of the
13-item Beck Depression Inventory in screening an adolescent population for
depressiveness and positive mood. Nordic Journal of Psychiatry 1999;53:451-457.
25
37. Fröjd S, Marttunen M, Pelkonen M, von der Pahlen B, Kaltiala-Heino R. Perceived
financial difficulties and maladjustment outcomes in adolescence. The European Journal
of Public Health 2006;16:542-8.
38. Achenbach TM. Manual for the Youth Self-Report and 1991 Profile. Burlington:
University of Vermont Department of Psychiatry; 1991.
39. Helstelä L, Sourander A. Self-reported competence and emotional and behavioral
problems in a sample of Finnish adolescents. Nordic Journal of Psychiatry 2001;55:381385.
40. Verhulst FC, Achenbach TM, van der Ende J, Erol N, Lambert MC, Leung PWL, et
al. Comparisons of Problems Reported by Youths from Seven Countries. Am J
Psychiatry 2003;160:1479-1485.
41. Morgan CJ, Cauce AM. Predicting DSM-III-R Disorders From the Youth SelfReport: Analysis of Data From a Field Study. Journal of the American Academy of
Child and Adolescent Psychiatry 1999;38:1237-1245.
42. Rescorla L, Almqvist F, Bird H, Dobrean A, Erol N, Hannesdottir H, et al.
Epidemiological Comparisons of Problems and Positive Qualities Reported by
Adolescents in 24 Countries. Journal of Consulting and Clinical Psychology 2007; 75:
351-358.
26
43. Blumenthal JA, Burg MM, Barefood J, Williams RB, Haney T, Zimet G. Social
support, type A behavior, and coronary artery disease. Psychosomatic Medicine 1987;
49:331-340.
44. Katainen S, Räikkönen K, Keltikangas-Järvinen L. Adolescent Temperament,
Perceived Social Support, and Depressive Tendencies as Predictors of Depressive
Tendencies in Young Adulthood. European Journal of Personality 1999;13:183-207.
45. Harland P, Reijneveld SA, Brugman E, Verloove-Vanhorick SP, Verhulst FC.
Family factors and life events as risk factors for behavioural and emotional problems in
children. European Child & Adolescent Psychiatry 2002;11:176-184.
46. Loeber R, Stouthamer-Loeber M, White HR. Developmental Aspects of
Delinquency and Internalizing Problems and Their Association With Persistent Juvenile
Substance Use Between Ages 7 and 18. Journal of Clinical Child Psychology 1999;28:
322-332.
47. Gotlib JH, Lewinsohn PM, Seeley JR. Symptoms Versus a Diagnostic Depression:
Differences in Psychosocial Functioning. Journal of Consulting and Clinical
Psychology 1995;63:90-100.
48. Junger-Tas J, Terlouw G, Klein M, editors. Delinquent behaviour among young
people in the western world: first results of the international self-report delinquency
study. Amsterdam: RDC, Ministry of Justice; 1994.
27
Table 1.
Perceived social support from family, friends and significant other according to depression and antisocial behaviour are shown by mean (standard
deviation) separately for boys and girls. Differences between depression or antisocial behaviour groups are tested by Mann-Whitney test.
Perceived social support
From family
N
Mean (Sd)
From friends
Mean (Sd)
From significant other
Mean (Sd)
Boys
Depression
No
Yes
1506
106
16.6 (3.7)
11.3 (5.2)
p<0.001
14.9 (4.3)
11.6 (5.5)
p<0.001
14.7 (4.6)
11.2 (5.8)
p<0.001
Antisocial
behaviour
No
Yes
1447
167
16.5 (3.8)
13.5 (5.2)
p<0.001
14.8 (4.4)
13.9 (4.9)
p=0.028
14.6 (4.7)
13.1 (5.5)
p=0.001
Depression
No
Yes
1365
215
16.8 (3.4)
12.4 (4.8)
p<0.001
17.8 (3.0)
15.3 (4.8)
p<0.001
18.1 (2.9)
15.9 (4.9)
p<0.001
Antisocial
behaviour
No
Yes
1407
175
16.6 (3.7)
13.5 (4.8)
p<0.001
17.5 (3.4)
17.4 (3.5)
p=0.650
17.8 (3.3)
17.6 (3.5)
p=0.391
Girls
Note: Due to the missing cases on the RBDI, YSR and PSSS-R, number of cases in the analyses differs from total sample.
28
Table 2. Risk for depression by antisocial behaviour in 15 year old boys and girls as unadjusted (Model
1), adjusted by the support from family (Model 2), by the support from friends (Model 3) and
by support from significant other (Model 4). In the model 5 antisocial behaviour is adjusted by
all the effect of perceived social support (family, friends and significant other) together. To the
latest model (Model 6) was added also family structure, parental education level, parental
unemployment during the past year and years of living in the same municipality. Logistic
regression models were used, with results given as the odds ratios (OR) and 95 % confidence
intervals (CI). Reference group of depression is non-depressive.
Depression
Boys
N
Model 1
Model 2
Model 3
Model 4
Model 5
OR [95% CI]
Girls
N
OR [95% CI]
Antisocial behaviour
No
Yes
1472 1.0
170 7.2 [4.7-11.0]
1425 1.0
174 4.3 [3.0-6.2]
Antisocial behaviour
No
Yes
1441 1.0
165 5.3 [3.3-8.4]
1405 1.0
173 2.8 [1.9-4.2]
Support from family
15-20 points
Under 15 points
1191 1.0
415 7.8 [4.9-12.3]
1173 1.0
405 6.1 [4.4-8.3]
Antisocial behaviour
No
Yes
1441 1.0
165 7.7 [4.9-11.9]
1405 1.0
173 4.7 [3.2-6.8]
Support from friends
15-20 points
Under 15 points
961 1.0
645 2.7 [1.8-4.2]
1339 1.0
240 4.2 [3.0-5.9]
Antisocial behaviour
No
Yes
1441 1.0
165 7.3 [4.7-11.3]
1405 1.0
173 4.4 [3.0-6.3]
Support from s.o*
15-20 points
Under 15 points
909 1.0
697 2.8 [1.8-4.3]
1381 1.0
197 3.5 [2.5-5.1]
Antisocial behaviour
No
Yes
1441 1.0
165 5.3 [3.4-8.5]
1405 1.0
173 3.2 [2.1-4.7]
29
Model 6
Support from family
15-20 points
Under 15 points
1191 1.0
415 7.7 [4.4-13.3]
1173 1.0
405 4.9 [3.5-6.8]
Support from friends
15-20 points
Under 15 points
961 1.0
645 1.1 [0.6-2.0]
1338 1.0
240 2.6 [1.7-4.0]
Support from s.o.*
15-20 points
Under 15 points
909 1.0
697 0.9 [0.5-1.7]
1381 1.0
197 1.1 [0.7-1.8]
Antisocial behaviour
No
Yes
1307 1.0
145 5.5 [3.3-9.1]
1279 1.0
154 2.9 [1.9-4.5]
Support from family
15-20 points
Under 15 points
1086 1.0
366 6.9 [3.8-12.5]
1078 1.0
355 5.4 [3.7-7.9]
Support from friends
15-20 points
Under 15 points
879 1.0
573 1.2 [0.6-2.2]
1213 1.0
220 3.0 [1.9-4.8]
Support from s.o.*
15-20 points
Under 15 points
829 1.0
623 1.0 [0.5-1.9]
1259 1.0
174 0.7 [0.5-1.5]
Family structure
Nuclear family
Step-parent
Single parent
Other
Parental education level
Either academic
Either vocational
Both only basic level
Parental unemployment
Neither
Either mother or father
Both of parents
Living in same municipality
Over 9 years
5-9 years
1-4 years
Under 1 year
1052
180
209
11
1.0
1.2 [0.6-2.2]
1.5 [0.8-2.8]
2.5 [0.5-12.2]
1023
178
221
11
1.0
1.2 [0.7-2.0]
2.0 [1.2-3.1]
2.3 [0.5-10.1]
526 1.0
794 0.8 [0.5-1.3]
132 0.6 [0.3-1.5]
519 1.0
813 0.7 [0.5-1.0]
101 0.8 [0.4-1.5]
1121 1.0
293 1.3 [0.7-2.2]
38 2.5 [0.9-7.2]
1056 1.0
340 1.6 [1.1-2.4]
37 2.3 [1.0-5.4]
970
241
190
51
1.0
1.1 [0.6-2.1]
1.2 [0.6-2.4]
2.0 [0.8-4.9]
983
231
162
57
1.0
1.1 [0.7-1.8]
1.0 [0.6-1.7]
3.3 [1.6-6.6]
*s.o.= significant other
Note: Due to the missing cases on the RBDI, YSR and PSSS-R, number of cases in the analyses differs from total sample.
ARTICLE IN PRESS
Journal of
Adolescence
Journal of Adolescence ] (]]]]) ]]]–]]]
www.elsevier.com/locate/jado
Continuity, comorbidity and longitudinal associations between
depression and antisocial behaviour in middle adolescence:
A 2-year prospective follow-up study
Minna Ritakallioa,b,, Anna-Maija Koivistoa,c, Bettina von der Pahlend,e,
Mirjami Pelkonena,e,f, Mauri Marttunene,f,g, Riittakerttu Kaltiala-Heinoa,h
a
University of Tampere, Tampere School of Public Health, Tampere, Finland
b
Tampere University Hospital, Adolescent ward, Tampere, Finland
c
Tampere University Hospital, Research Unit, Tampere, Finland
d
Department of Psychology, University of Åbo Akademi, Åbo, Finland
e
National Public Health Institute, Department of Mental Health and Alcohol Research, Helsinki, Finland
f
Department of Adolescent Psychiatry, Peijas Hospital, Helsinki University Hospital, Vantaa, Finland
g
Department of Adolescent Psychiatry, University of Kuopio, and Kuopio University Hospital, Kuopio, Finland
h
Department of Adolescent Psychiatry, Tampere University Hospital, Tampere, Finland
Abstract
The study investigated continuity, comorbidity and longitudinal associations between depression Beck
depression inventory (RBDI) and antisocial behaviour Youth self-report (YSR) in middle adolescence.
Data were used from a community sample of 2070 adolescents who participated in a 2-year prospective
follow-up study. The results indicate that both depression and antisocial behaviour had considerable
continuity, and concurrent comorbidity between these disorders was strong. In contrast to several previous
studies, antisocial behaviour did not predict subsequent depression, but conversely, depression predicted
subsequent antisocial behaviour among girls. Among boys history of depression seemed to protect from
subsequent antisocial behaviour. Gender differences in longitudinal associations are discussed.
r 2007 The Association for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights
reserved.
Corresponding author. University of Tampere, Tampere School of Public Health, Tampere, Finland.
E-mail address: minna.ritakallio@uta.fi (M. Ritakallio).
0140-1971/$30.00 r 2007 The Association for Professionals in Services for Adolescents. Published by Elsevier Ltd. All
rights reserved.
doi:10.1016/j.adolescence.2007.06.006
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006
ARTICLE IN PRESS
M. Ritakallio et al. / Journal of Adolescence ] (]]]]) ]]]–]]]
2
Introduction
Previous studies have established that both depression and antisocial behaviour in adolescence
have considerable continuity (Fergusson, Lynskey, & Hoorwood, 1996; Kim et al., 2003; Wiesner,
2003). In a longitudinal study by Beyers and Loeber (2003) 6–35% of the variance of current
depression was explained by prior depression, the corresponding variance predicting current
delinquency was 25–34%. Similarly, Costello, Mustillo, Erkanli, Keeler, and Angold (2003) found
that adolescents with a history of depression were four times more likely than those with no
previous depression to have depression in the future. The corresponding odds ratio (OR) for
antisocial behaviour among those with a history of antisocial behaviour was about 10. Girls
showed more continuity of depression than boys, but no gender differences were found in the
continuity of antisocial behaviour. In contrast, Ferdinand, Verhulst, and Wiznitzer (1995) found
that continuity of both depression and antisocial behaviour did not differ by sex. According to
Angold et al. (1999) and Costello et al. (2003) the stability described above represents homotypic
continuity in which manifestations of the disorder in question change relatively little over time so
that diagnosis remains the same at different assessments.
High rates of concurrent comorbidity of adolescent depression and antisocial behaviour have
been well established in both clinical (Biederman, Faraone, Mick, & Lelon, 1995; Goodyer,
Herbert, Secher, & Pearson, 1997; Grilo et al., 1996) and population-based samples (Beyers &
Loeber, 2003; Fergusson & Woodward, 2002; Loeber, Farrington, Stouthamer-Loeber, & Van
Kammen, 1998; Ritakallio, Kaltiala-Heino, Kivivuori, & Rimpelä, 2005; Vermeiren, Deboutte,
Ruchkin, & Schwab-Stone, 2002). In a meta-analysis of community studies Angold et al. (1999)
found that after controlling for other comorbidities, conduct disorder was about seven times more
common in depressed than in non-depressed adolescents. Previous research on gender differences
in comorbidity, however, is inconsistent as some studies have found that depression is more
prevalent among antisocial girls than boys (Costello et al., 2003; Flannery et al., 2001) while
others have found the opposite (Maughan, Rowe, Messer, Goodman, & Meltzer, 2004).
In studying developmental associations between depression and antisocial behaviour the
concept of heterotypic continuity is pivotal. Heterotypic continuity refers to a continuous process
in which one disorder generates manifestations of different forms over time or in which one
disorder exposes adolescents to different disorders at different ages (Angold et al., 1999; Costello
et al., 2003). There are at least three possible developmental models (acting out, failure and
mutual influence model) to explain the heterotypic continuity between adolescent depression
and antisocial behaviour. The models differ in whether they assume associations to be causal
(one disorder creates an increased risk for the other) or non-causal (association based on nonspecific risk factors) or disorders are reciprocally associated with each other (Caron &
Rutter, 1991; Fergusson et al., 1996; Overbeek, Vollebergh, Meeus, Engels, & Luijpers, 2001;
Wiesner, 2003).
The acting out model proposes that depressed adolescents act out internalising problems and
depression is masked out by antisocial behaviour. This model assumes that depression precedes
antisocial behaviour and the cross-lagged associations are unidirectional and causal (Capaldi,
1992; Overbeek et al., 2001). The failure model assumes that there is a causal, unidirectional
association between antisocial behaviour and subsequent depression. Antisocial adolescents face
many problems in social relations (e.g. rejection, lack of support, conflict with parents and
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006
ARTICLE IN PRESS
M. Ritakallio et al. / Journal of Adolescence ] (]]]]) ]]]–]]]
3
teachers), have low social competence and coping skills and problems in academic attainment
which in turn lead to failure experiences which increases vulnerability for depression (Capaldi,
1992; Kiesner, 2002). According to the mutual influence model both depression and antisocial
behaviour result from non-specific (shared or overlapping) risk factors, and disorders are also
reciprocally reinforcing each other over time. This model assumes that the onset of one disorder
increases vulnerability to the other and vice versa. Therefore, cross-lagged and bidirectional
associations between disorders are possible (Overbeek et al., 2001). As the current study
investigated prospective associations between depression and antisocial behaviour we focused on
acting out and failure models.
There are surprisingly few studies of the longitudinal associations between depression and
antisocial behaviour in adolescence, and the results appear inconsistent ( Loeber et al., 2000). It
should, however, be noted that due to differences in methodology (e.g. age range of the sample,
sex ratio of the sample and period of follow-up), previous studies are difficult to compare. Most
previous studies have focused on the transition from (late) childhood to early adolescence (e.g.
Kiesner, 2002) or they have followed children through adolescence into adulthood (e.g. Roza,
Hofstra, van der Ende, & Verhulst, 2003).
Many studies conducted in community samples propose that antisocial behaviour precedes the
onset of depression and that the effect is unidirectional (Capaldi, 1992; Feehan, McGee, &
Williams, 1993; Fergusson et al., 2003; Kiesner, 2002; Rohde, Lewinsohn, & Seeley, 1991).
Different results from these have been reported by Loeber, Russo, Stouthamer-Loeber, and Lahey
(1994) who found that depression in early adolescent boys was associated with escalation to more
serious delinquency and a variety of delinquent behaviour, but Vermeiren, Schwab-Stone,
Ruchkin, De Clippele, and Deboutte (2002) found depression to be protective against recidivism.
Harrington, Fudge, Rutter, Pickles, and Hill (1991) found that rates of adult depression were
lower in those who as adolescents had comorbid antisocial behaviour and depression than in
those who had had pure antisocial behaviour or pure depression in adolescence. Furthermore,
some studies found no heterotypic continuity between these disorders (Costello et al., 2003;
Lewinsohn et al., 1994; Overbeek et al., 2001; Roza et al., 2003) while others suggest that
depression and antisocial behaviour affect each other in a bidirectional way (Beyers and Loeber,
2003).
Studies on gender differences among the longitudinal associations between depression and
antisocial behaviour in adolescence are relatively rare. The study by Wiesner (2003) found limited
evidence that antisocial behaviour among boys preceded depression but not vice versa. Among
girls a circular process was found as high level of antisocial behaviour preceded high level of
depression, which in turn, preceded de-escalation of antisocial behaviour, which finally resulted in
lower level of depression. In the study by Hofstra et al. (2002) antisocial behaviour in childhood
and adolescence predicted only boys’ mood disorders in adulthood. Overbeek et al. (2001) found
no gender differences, suggesting that there are no cross-lagged associations between depression
and antisocial behaviour as homotypic continuity of disorders best explain the future course of
depression and antisocial behaviour.
In sum, the findings of previous studies on comorbidity and longitudinal associations between
adolescent depression and antisocial behaviour have been inconsistent, especially with regard to
gender differences. Although there are a few studies which includes both boys and girls, gender
comparisons are still relatively uncommon. Given that the rates of both adolescent depression and
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006
ARTICLE IN PRESS
M. Ritakallio et al. / Journal of Adolescence ] (]]]]) ]]]–]]]
4
antisocial behaviour and the course of these disorders differ among boys and girls (Maughan
et al., 2004; Overbeek et al., 2001), it could be assumed that longitudinal associations between
these disorders also differ between genders. Most previous studies have investigated longitudinal
associations between adolescent depression and antisocial behaviour using only one developmental model as a theoretical framework. It would be interesting to simultaneously test different
developmental models to explain the longitudinal association between depression and antisocial
behaviour. Ascertaining the developmental associations between depression and antisocial
behaviour might provide important information regarding the etiology and course of these
disorders for use in prevention and treatment.
The purpose of the present study was to investigate continuity, comorbidity and longitudinal
associations between depression and antisocial behaviour using data from a 2-year longitudinal
study with adolescent boys and girls. In detail, the study investigated (1) how common homotypic
continuity of depression and antisocial behaviour is during a 2-year follow-up in middle
adolescence, (2) whether there is concurrent comorbidity between depression and antisocial
behaviour at age 15 and at age 17, (3) which developmental model—the failure model or the
acting out model would best explain the longitudinal associations between depression and
antisocial behaviour during a 2-year follow-up, and (4) whether there are any significant sex
differences in any of these themes.
Methods
Study samples and procedures
The present study is a part of an ongoing prospective cohort study, Mental Health in
Adolescence, conducted in two Finnish cities; Tampere (200 000 inhabitants) and Vantaa (180 000
inhabitants). The present samples consist of surveys at two waves (T1 and T2) during a 2-year
follow-up conducted during the academic year 2002–2003 (T1) and during the academic year
2004–2005 (T2).
At T1 data were collected by a school survey completed by the ninth graders in all the Finnishspeaking secondary schools in the two cities. The students were identified from school registers. A
person-identifiable survey was completed during a school lesson under the supervision of a
teacher. For students absent from school on the original survey day, another opportunity to
participate was offered in the school within a couple of weeks. For students not present on either
occasion the questionnaires were sent by post twice, after which it was concluded that the student
was not willing to participate. The parents of the students were informed in advance by letter, but
parental consent to participation was not required since the Finnish legislation on medical
research allows a 15-year-old subject to decide alone. The procedure at T1 is described in detail in
Ritakallio, Luukkaala, Marttunen, Pelkonen, and Kaltiala-Heino (submitted for publication).
At T1 the ninth graders of the participating schools totalled 3809, of whom 3597 responded to
the survey (response rate 94.4% at T1). Six respondents were excluded due to obvious
facetiousness, and 313 respondents were excluded as they were completed by adolescents under 15
years old. Thus, the initial sample at T1 was 3278 students of whom 1609 were girls (49.1%). The
mean age at T1 was 15.5 (sd ¼ 0.4).
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006
ARTICLE IN PRESS
M. Ritakallio et al. / Journal of Adolescence ] (]]]]) ]]]–]]]
5
Eligible participants at T2 data were students who had participated at T1. Multiple approaches
were used to contact the adolescents at follow-up. School-based surveys like that at T1 were
organised in upper secondary schools and vocational schools. Adolescents not reached through
schools were contacted by postal survey. Finally, the same survey was offered by Internet to those
who had not yet responded via their schools or by post.
The subjects of the present study were the 2070 adolescents who completed the survey both at
T1 and at T2. The response rate of the final sample was 63.1% (2070/3278). Of the respondents
56.4% (1167) were girls. The mean age at T2 was 17.6 (sd 0.4). At T2 62.4% of the adolescents
were living in intact families (boys 65.0% vs. girls 60.5%, po0.001). Permanent employment of
the parents was high at T2 as only 6.2% of parents (one or both) had been unemployed during the
past year (no gender differences). Over 80% of the adolescents were full-time students at T2 (girls
89.6% vs. boys 82.0, po0.001). 66.1% of the adolescents were studying at upper secondary school
at T2 (girls 72.3% vs. boys 58.1%, po0.001). Stability of residence was high as 93.3% of the
adolescents had lived in the same municipality for more than 1 year (girls 93.6% vs. boys 92.8,
p ¼ 0.443).
Measures
A Finnish modification (RBDI) (Raitasalo, 1995) of the 13-item Beck depression inventory
(BDI) was used to assess depression (Beck & Beck, 1972; Beck, Rial, & Rickels, 1974). The brief
BDI is a widely used self-report scale measuring the severity of depressive symptoms (emotional,
cognitive and physical), and its reliability and validity are good (Bennett et al., 1997). The Finnish
modification of the brief BDI has been shown to be an appropriate method for measuring
depression in Finnish adolescents (Kaltiala-Heino, Rimpela, Rantanen, & Laippala, 1999). Each
item is scored 0–3 according to the severity of the symptom. Sum scores of the 13 items (range
0–39 scores) were dichotomised to none/mild depression (0–7 scores) and moderate/severe
depression (8–39 scores) as recommended by Beck and Beck (1972) and as has been done in the
previous analyses of Finnish adolescent population samples (e.g. Kaltiala-Heino et al., 2001).
Fountoulakis et al. (2003) reported that 13-item BDI with cut-off point 8 predicted a SCANdiagnosis of depression with good sensitivity (0.93) and specificity (0.88).
Antisocial behaviour was assessed by the externalising scale of the Youth self-report (YSR).
The YSR was developed by Achenbach (1991) for the assessment of psychopathology in
adolescents aged 11–18 years. It is widely used and has shown high reliability and validity
(Achenbach, 1991; Helstelä & Sourander, 2001). There is also evidence that YSR has good ability
to predict DSM diagnosis. For example, Morgan and Cauce (1999) found that high YSR scores
on externalising scale predicted a DISC-C-diagnosis of oppositional defiant or conduct disorder
with respectable rates of sensitivity (0.74) and specificity (0.73). The externalising scale consists of
11 variables concerning delinquency and 18 variables concerning aggressive behaviour. Each
variable was scored 0–2 according to the frequency of antisocial behaviour. Sum scores of the 29
items (range 0–58) were dichotomised using the 90th percentile as cut-off into those within normal
range vs. those within clinical range. Because antisocial behaviour is more common among
adolescent boys than girls, sex-specific cut-off-points were used (Achenbach, 1991.) At T1 the cutoff-point was 23.0 for girls and 24.0 for boys, and at T2 21.0 for girls and 22.0 for boys.
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006
ARTICLE IN PRESS
M. Ritakallio et al. / Journal of Adolescence ] (]]]]) ]]]–]]]
6
Sex, family structure (intact vs. non-intact), parental education level (academic vs. nonacademic), parental unemployment during the past year (no vs. none or both) and residential
stability (living 10 or more vs. up to 9 years in the same municipality) at T1 were controlled for as
sociodemographic variables. Parental education was classified according to the higher educated
parent. Previous studies have shown these variables to be associated with both depression and
antisocial behaviour (Fröjd, Marttunen, Pelkonen, von der Pahlen, & Kaltiala-Heino,2006;
Kaltiala-Heino et al., 2001; Loeber et al., 1998).
Drop out
Non-participants at T2 were more likely to be boys (63.4% vs. girls 36.6%, po0.001) living
more frequently with both parents (35.2%) than those adolescents who participated in both
surveys (27.5%, po0.001). Adolescents who dropped out of the second survey were more likely to
be depressed (11.7%) at baseline than those who participated in both surveys (9.1%) (p ¼ 0.020).
Dropouts were also more likely to have displayed antisocial behaviour at baseline (12.3%) than
participants (8.7%) (po0.001).
Statistical analyses
All analyses were conducted separately for boys and girls. Homotypic continuity of
depression was first studied by calculating proportions of those reporting depression at T2
according to depression at T1. The significance of differences in proportions was analysed by
w2-test. Logistic regression analysis was performed to control for confounding by sociodemographic variables (family structure, parental education level, parental unemployment during
the past year and residential stability). Depression at T2 was entered as a dependent variable,
and in the first step depression at T1 was entered alone as an independent variable (basic
model). Finally, all sociodemographic variables were added as independent variables. To ascertain
if there were significant sex differences in homotypic continuity of depression, interaction
terms for sex and depression at T1 were added into the basic model. Homotypic continuity
of antisocial behaviour was analysed by similar procedures as homotypic continuity of
depression.
Concurrent comorbidity between depression and antisocial behaviour was analysed both at
T1 and at T2. Analyses were performed similarly at both waves. The proportions of those
reporting depression by antisocial behaviour were calculated and the significance of differences
in proportions was analysed by w2-test. Similarly, the proportions of those reporting antisocial behaviour by depression were calculated and the significance of differences was analysed
by w2-test. Next, OR with 95% confidence interval (CI) for comorbidity between antisocial
behaviour (dependent variable) and depression (independent variable) was calculated by
logistic regression analysis. To ascertain, if there were significant sex differences in comorbidity, interaction terms for sex and depression (independent variable) were added into the
model.
To test whether the failure model would explain longitudinal associations between depression
and antisocial behaviour, the proportions of those reporting depression at T2 were calculated
according to antisocial behaviour at T1. The significance of differences between groups was
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006
ARTICLE IN PRESS
M. Ritakallio et al. / Journal of Adolescence ] (]]]]) ]]]–]]]
7
analysed using w2-test. Next, crude OR (OR, 95% CI) for depression at T2 (dependent variable)
according to antisocial behaviour at T1 (independent variable) was calculated using logistic
regression analysis (Model 1). Subsequently, OR for depression at T2 by antisocial behaviour at
T1 was adjusted for depression at T1 (Model 2) and finally also for sociodemograhpic variables
(family structure, parental education level, parental unemployment during the past year and
residential stability) (Model 3).
To test the acting out model in explaining longitudinal associations between depression and
antisocial behaviour, the proportions those reporting antisocial behaviour at T2 were calculated
according to depression at T1. The significance of differences between groups was studied by
w2-test. Next, crude OR (OR, 95% CI) for antisocial behaviour at T2 (dependent variable)
according to depression at T1 (independent variable) was calculated using logistic regression
analysis (Model 1). Subsequently, OR for antisocial behaviour at T2 by depression at T1 was
adjusted for antisocial behaviour at T1 (Model 2), and finally also for sociodemographic variables
(family structure, parental education level, parental unemployment during the past year and
residential stability) (Model 3).
Statistical analyses were performed with the SPSS 13.0 software package.
Results
Homotypic continuity
Both depression and antisocial behaviour showed considerable continuity during the 2-year
follow-up in both sexes (Tables 1 and 2). Of the boys depressed at T1, 27.8% of them were also
depressed at T2. Of the girls depressed at T1, 46.3% of them were also depressed at T2. Boys with
depression at T1 had a 7.4-fold risk for depression at T2 after adjustment for sociodemographic
variables. The corresponding OR for girls was 12.5. There were no significant gender differences
in continuity of depression according to the interaction analysis (p ¼ 0.108). Of the boys and girls
reporting antisocial behaviour at T1, 43.5% of boys and 42.0% of girls still displayed antisocial
behaviour at T2. Boys with antisocial behaviour at T1 had a 14.6-fold risk for antisocial
behaviour at T2 after adjustment for sociodemographic variables. The corresponding OR for girls
was 8.3.There were no significant gender differences in continuity of antisocial behaviour
according to the interaction analysis (p ¼ 0.342).
Concurrent comorbidity at T1 and T2
Concurrent comorbidity between depression and antisocial behaviour was high both at T1 and
at T2 (Table 3). Girls with antisocial behaviour at T1 were three times more likely to be depressed
than non-antisocial girls (OR 3.5, CI 2.3–5.5). Corresponding risk at T2 was OR 5.1 (CI 3.3–7.9).
Similarly, boys with antisocial behaviour at T1 were seven times more likely to be depressed than
non-antisocial boys (OR 7.3, CI 4.0–13.2). Corresponding risk at T2 was OR 5.5 (CI 3.1–10.1).
There were no significant gender differences in comorbidity (at T2 p ¼ 0.836) although
comorbidity at T1 almost reached statistical significance (p ¼ 0.055), suggesting that comorbidity
may be more common among boys.
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006
ARTICLE IN PRESS
M. Ritakallio et al. / Journal of Adolescence ] (]]]]) ]]]–]]]
8
Table 1
Rates of [% (n)] and risk for [odds ratio (OR), 95% confidence interval (CI)] depression at 2-year follow-up (T2)
according to depression at 15-year-old (T1) adolescents
% (n)
Boys at T1 (n)
Depression
No (836)
Yes (54)
4.9 (41)
27.8 (15)
Model 1a
Model 2b
OR
95% CI
OR
95% CI
Ref.
7.5
3.8–14.6
Ref.
7.4
3.6–15.2
Family structure
Intact
Non-intact
#
#
Ref.
0.8
0.4–1.6
Parental education level
Academic
Non-academic
#
#
Ref.
0.9
0.5–1.6
Parental unemployment
No
One or both
#
#
Ref.
1.4
0.7–2.6
Residential stability
Over 10 years
Less than 9 years
#
#
Ref.
1.7
0.9–3.1
Ref.
12.5
7.8–20.0
Girls at T1 (n)
Depression
No (1025)
Yes (134)
5.7 (58)
46.3 (62)
Ref.
14.4
9.3–22.1
Family structure
Intact
Non-intact
#
#
Ref.
1.8
1.1–3.0
Parental education level
Academic
Non-academic
#
#
Ref.
0.9
0.5–1.4
Parental unemployment
No
One or both
#
#
Ref.
0.9
0.6–1.6
Residential stability
Over 10 years
Less than 9 years
#
#
Ref.
1.1
0.7–1.8
Ref. ¼ reference group in logistic regression analysis and # ¼ variable not in the model.
Differences between non-depressed and depressed adolescents are significant (po0.001) in w2-test.
a
In Model 1, crude OR are presented.
b
In Model 2, OR are adjusted for sociodemographic variables (family structure, parental education level, parental
unemployment during the past year and residential stability).
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006
ARTICLE IN PRESS
M. Ritakallio et al. / Journal of Adolescence ] (]]]]) ]]]–]]]
9
Table 2
Rates of [%(n)] and risk for [odds ratio (OR), 95% confidence interval (CI)] antisocial behaviour at 2-year follow-up
(T2) according to antisocial behaviour at 15-year-old (T1) adolescents
% (n)
Model 1a
Model 2b
OR
95% CI
OR
95% CI
Ref.
11.4
6.9–18.9
Ref.
14.6
8.3–25.6
#
#
Ref.
1.0
0.5–1.8
Parental education level
Academic
Non-academic
#
#
Ref.
0.5
0.3–0.9
Parental unemployment
No
One or both
#
#
Ref.
1.5
0.8–2.6
Residential stability
Over 10 years
Less than 9 years
#
#
Ref.
1.5
0.9–2.6
Ref.
8.3
5.4–12.9
Boys at T1 (n)
Antisocial
No (794)
Yes (92)
Family structure
Intact
Non-intact
Girls at T1 (n)
Antisocial
No (1028)
Yes (131)
6.3 (50)
43.5 (40)
8.0 (82)
42.0 (55)
Ref.
8.3
5.5–12.6
Family structure
Intact
Non-intact
#
#
Ref.
1.2
0.7–1.9
Parental education level
Academic
Non-academic
#
#
Ref.
0.8
0.5–1.2
Parental unemployment
No
One or both
#
#
Ref.
1.1
0.7–1.8
Residential stability
Over 10 years
Less than 9 years
#
#
Ref.
1.0
0.6–1.6
Ref. ¼ reference group in logistic regression analysis and # ¼ variable not in the model.
Differences between non-antisocial and antisocial adolescents are significant (po0.001) in w2-test.
a
In Model 1, crude OR are presented.
b
In Model 2, OR are adjusted for sociodemographic variables (family structure, parental education level, parental
unemployment during the past year and residential stability).
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006
ARTICLE IN PRESS
M. Ritakallio et al. / Journal of Adolescence ] (]]]]) ]]]–]]]
10
Table 3
Concurrent comorbidity of depression and antisocial behaviour: prevalence [%(n)] of depression by antisocial
behaviour and vice versa among 15 years old (T1) and at 2-year follow-up (T2)
Girls (N ¼ 1167) [%(n)]
Boys (N ¼ 903) [%(n)]
T1
Rate of depression by antisocial behaviour (n)
No (1028a, 795b)
Yes (128a, 94b)
9.6 (99)
27.3 (35)
4.0 (32)
23.4 (22)
Rate of antisocial behaviour by depression (n)
No (1022a, 835b)
Yes (134a, 54b)
9.1 (93)
26.1 (35)
8.6 (72)
40.7 (22)
T2
Rate of depression by antisocial behaviour (n)
No (1026a, 802b)
Yes (137a, 88b)
7.7 (79)
29.9 (41)
4.7 (38)
21.6 (19)
Rate of antisocial behaviour by depression (n)
No (1043a, 833b)
Yes (120a, 57b)
9.2 (96)
34.2 (41)
8.3 (69)
33.3 (19)
a
Girls.
Boys.
b
Longitudinal associations between adolescent depression and antisocial behaviour
Failure model
Of those boys with antisocial behaviour at T1, 12.1% (11/91) were depressed at T2 (vs. 5.8%
(46/796) of non-antisocial boys at T1) (p ¼ 0.020). Correspondingly, of those girls with antisocial
behaviour at T1, 16.0% (21/131) were depressed at T2 (vs. 9.6% (99/1029) of non-antisocial girls
at T1) (p ¼ 0.023). Antisocial behaviour at T1 predicted depression at T2 in both sexes when
entered alone in logistic regression analysis (Table 4: Model 1). Adolescents with antisocial
behaviour at T1 were approximately two times more likely to be depressed at T2 than those
adolescents with no antisocial behaviour at T1. When depression at T1 and sociodemographic
variables were controlled for (Table 4: Models 2 and 3), the association between antisocial
behaviour at T1 and depression at T2 lost significance in both sexes.
Acting out model
Of boys depressed at T1, 14.8% (8/54) displayed antisocial behaviour at T2 (vs. 9.8% (82/834)
of non-depressed boys at T1, p ¼ 0.240). Similarly, of girls depressed at T1, 26.9% (36/134)
displayed antisocial behaviour at T2 (vs. 9.8% (100/1024) of non-depressed girls at T1)
(po0.001).
In logistic regression analysis depression at T1 predicted antisocial behaviour at T2 differently
among girls and boys (Table 5). Among girls depression at T1 predicted antisocial behaviour at
T2 after controlling for antisocial behaviour at T1 and sociodemographic variables. Girls with
depression at T1 had a 2.3-fold risk for antisocial behaviour at T2. Among boys depression at T1
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006
ARTICLE IN PRESS
M. Ritakallio et al. / Journal of Adolescence ] (]]]]) ]]]–]]]
11
Table 4
Risk for [odds ratio (OR), 95% confidence interval (CI)] depression at 2-year follow-up (T2) according to antisocial
behaviour among 15-year-old (T1) adolescents
Girls
Boys
Model 1a
Model 2b
Model 3c
Model 1a
Model 2b
Model 3c
OR (95% CI)
OR (95% CI)
OR (95% CI)
OR (95% CI)
OR (95% CI)
OR (95% CI)
Ref.
1.0 (0.5–1.7)
Ref.
1.0 (0.6–2.0)
Ref.
2.2 (1.1–4.5)
Ref.
1.3 (0.6–2.8)
Ref.
1.1 (0.4–2.6)
Antisocial behaviour at T1
No
Ref.
Yes
1.8 (1.1–3.0)
Depression at T1
No
Yes
#
#
Ref.
14.4
(9.2–22.4)
Ref.
12.4
(7.6–20.1)
#
#
Ref.
6.8 (3.3–14.0)
Ref.
7.3 (3.4–15.7)
Family structure
Intact
Non-intact
#
#
#
#
Ref.
1.8 (1.1–3.0)
#
#
#
#
Ref.
0.8 (0.4–1.6)
Parental education level
Academic
Non-academic
#
#
#
#
Ref.
0.9 (0.5–1.4)
#
#
#
#
Ref.
0.9 (0.5–1.5)
Parental unemployment
No
One or both
#
#
#
#
Ref.
0.9 (0.6–1.6)
#
#
#
#
Ref.
1.4 (0.7–2.6)
Residential stability
Over 10 years
Less than 9 years
#
#
#
#
Ref.
1.1 (0.7–1.8)
#
#
#
#
Ref.
1.7 (0.9–3.1)
Ref. ¼ reference group in logistic regression analysis and # ¼ variable not in the model.
a
In Model 1, crude OR are presented.
b
In Model 2, OR are adjusted for depression at T1.
c
In Model 3, OR are adjusted for depression at T1 and for sociodemographic variables (family structure, parental
education level, parental unemployment during the past year and residential stability).
predicted inversely antisocial behaviour at T2 after controlling for antisocial behaviour at T1 and
for sociodemographic variables. Thus depression at T1 protected boys from antisocial behaviour
at T2 (OR 0.4, p ¼ 0.05).
Discussion
Regarding homotypic continuity of depression and antisocial behaviour, the current study
supports the findings from previous studies (e.g. Costello et al., 2003) indicating that these
disorders have a fairly stable course in adolescence. The finding that continuity of either
depression or antisocial behaviour did not differ by gender is consistent with the results of
Ferdinand et al. (1995), but different from the results of Costello et al. (2003) proposing girls to
show more continuity of depression than boys. The differences may be due to different time
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006
ARTICLE IN PRESS
M. Ritakallio et al. / Journal of Adolescence ] (]]]]) ]]]–]]]
12
Table 5
Risk for [odds ratio (OR), 95% confidence interval (CI)] antisocial behaviour at 2-year follow-up (T2) according to
depression among 15-year-old (T1) adolescents
Girls
Boys
Model 1a
Model 2b
Model3c
Model 1a
Model 2b
Model 3c
OR (95% CI)
OR (95% CI)
OR (95% CI)
OR (95% CI)
OR (95% CI)
OR (95% CI)
Ref.
3.4 (2.2–5.2)
Ref.
2.4 (1.5–3.9)
Ref.
2.3 (1.3–3.9)
Ref.
1.6 (0.7–3.5)
Ref.
0.5 (0.2–1.3)
Ref.
0.4 (0.1–1.0)
Antisocial behaviour at T1
No
#
Yes
#
Ref.
7.4 (4.8–11.3)
Ref.
7.6 (4.8–11.9)
#
#
Ref.
13.0 (7.6–22.1)
Ref.
18.0 (9.9–32.9)
Family structure
Intact
Non-intact
#
#
#
#
Ref.
1.1 (0.7–1.8)
#
#
#
#
Ref.
1.1 (0.6–2.0)
Parental education level
Academic
#
Non-academic
#
#
#
Ref.
0.8 (0.5–1.3)
#
#
#
#
Ref.
0.5 (0.3–0.9)
Parental unemployment
No
#
One or both
#
#
#
Ref.
1.0 (0.6–1.6)
#
#
#
#
Ref.
1.5 (0.8–2.7)
Residential stability
Over 10 years
Less than 9 years
#
#
Ref.
1.0 (0.6–1.6)
#
#
#
#
Ref.
1.5 (0.9–2.7)
Depression at T1
No
Yes
#
#
Ref. ¼ reference group in logistic regression analysis and # ¼ variable not in the model.
a
In Model 1, crude OR are presented.
b
In Model 2, OR are adjusted for antisocial behaviour at T1.
c
In Model 3, OR are adjusted for antisocial behaviour at T1 and for sociodemographic variables (family structure,
parental education level, parental unemployment during the past year and residential stability).
intervals or different case ascertainment used. Consistent with previous studies (e.g. Fergusson
and Woodward, 2002), the concurrent comorbidity between depression and antisocial behaviour
was strong during the 2-year follow-up. There were no significant gender differences in
comorbidity, although gender differences at T1 almost reached significance, suggesting stronger
comorbidity among boys. This finding is consistent with the results of Domalanta, Risser,
Roberts, and Risser (2003), but differs from the results of Flannery et al. (2001) proposing
depression to be more prevalent among antisocial girls than boys. Flannery et al. (2001), however,
focused on extremely violent adolescents, whereas the present study investigated antisocial
behaviour in a general population samples.
The main purpose of this study was to investigate longitudinal associations between depression
and antisocial behaviour. In univariate analyses subsequent depression was more prevalent
among adolescents of both sexes with a history of antisocial behaviour, but after controlling for
previous depression this association lost significance. The finding that the failure model did not
explain the developmental associations between antisocial behaviour and depression seems
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006
ARTICLE IN PRESS
M. Ritakallio et al. / Journal of Adolescence ] (]]]]) ]]]–]]]
13
inconsistent with several previous studies (e.g. Kiesner, 2002). However, most previous studies
have focused on the transition from late childhood to early adolescence (e.g. Capaldi, 1992) or
they have had a considerably longer follow-up period from childhood to young adulthood (e.g.
Fergusson et al., 2003). The present study covered the period from middle to late adolescence. It
has been suggested that developmental associations between depression and antisocial behaviour
may depend on the time when antisocial behaviour initiates (Lewinsohn, Gotlib, & Seeley, 1995),
and that failure experiences triggered by antisocial behaviour may be age-related and differ across
developmental stages (Wiesner, 2003). Our results suggest that the failure model is not adequate in
middle adolescence.
The present study demonstrated that depression predicted subsequent antisocial behaviour, but
the acting out model held true only among girls after controlling for previous antisocial
behaviour. Girls with a history of depression were twice more likely than non-depressed girls to
display subsequent antisocial behaviour. In contrast to the finding for girls, history of depression
seemed to protect boys from subsequent antisocial behaviour. The protective influence on
depression for subsequent antisocial behaviour is consistent with the study of Vermeiren, SchwabStone et al. (2002) who found depression to be protective against recidivism. It has been suggested
that subsequent antisocial behaviour may be reduced due to apathy and diminished energy
associated with depression or that depression and experiences of guilt/shame may indicate the
potential for reflecting on one’s actions and their consequences for others (Vermeiren, SchwabStone et al., 2002). Further research is needed to ascertain why in the present study these
mechanisms did not apply to girls. The developmental course of antisocial behaviour may differ
by sex, as boys generally initiate antisocial behaviour earlier than girls do (e.g. Overbeek et al.,
2001). Therefore, the longitudinal association between depression and antisocial behaviour may
occur earlier among boys and to be non-apparent in middle adolescence.
The results of the present study need to be viewed in the light of certain limitations. Depression
and antisocial behaviour were measured by self-report scales. However, although the RBDI is not
a diagnostic instrument, it has been shown to be efficient and valid in distinguishing between
depressed and non-depressed adolescents (Beck et al., 1974; Bennett et al., 1997; Kaltiala-Heino
et al., 1999). Further, although there are known limitations in the reliability of self-reported
antisocial behaviour, such as dishonesty, forgetfulness and exaggeration, self-report methods are
generally considered reliable and valid methods in assessing antisocial behaviour (Junger-Tas,
Terlouw, & Klein, 1994). In addition to this, the instrument used (YSR) has shown good
psychometric properties and ability to predict DSM oppositional defiant and conduct disorder
(Achenbach, 1991; Helstelä & Sourander, 2001; Morgan & Cauce, 1999).
The key strengths of the study are a large community-based cohort of adolescents with a fair
response rate. However, because the present analyses are based on 63% of the initial sample, and
because only limited information about non-participants at T2 was available, the results may not
be generalisable. Further, attrition analysis showed that non-participants at T2 were more likely
to be depressed and behave antisocially at baseline than participants. On the other hand, given
that both depression and antisocial behaviour have considerable continuity in adolescence (e.g.
Costello et al., 2003), the rates of depression and antisocial behaviour may be underestimated at
T2, but the found associations between these disorders are not likely to be affected by nonparticipants. The results may not be representative of the Finnish population of adolescents since
none of these adolescents came from rural areas. Nevertheless, the large longitudinal samples
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006
ARTICLE IN PRESS
14
M. Ritakallio et al. / Journal of Adolescence ] (]]]]) ]]]–]]]
made it possible to test two models simultaneously for developmental associations between
depression and antisocial behaviour. Most of the previous studies have focused on only one
developmental model. The current data also allowed us to study gender differences. Most of the
previous studies have not included gender comparisons.
Conclusion
Our findings indicated that antisocial behaviour and depression had considerable continuity in
middle adolescence. Concurrent comorbidity between depression and antisocial behaviour was
also remarkable. Among girls depression predicted subsequent antisocial behaviour in a 2-year
span, but among boys depression seemed to protect them from subsequent antisocial behaviour.
The present study offers implications for the prevention and treatment of adolescent depression
and antisocial behaviour. Prevention should be initiated as soon as possible before the
psychosocial adjustment of adulthood is impaired. In clinical settings, antisocial behaviour
among adolescent girls should be considered as a possible marker of severe depression.
References
Achenbach, T. M. (1991). Manual for the youth self-report and 1991 profile. Burlington, VT: University of Vermont
Department of Psychiatry.
Angold, A., Costello, E. J., & Erkanli, A. (1999). Comorbidity. Journal of Child Psychology and Psychiatry, 40,
57–87.
Beck, A. T., & Beck, R. W. (1972). Screening depressed patients in family practice. A rapid technic. Postgraduate
Medicine, 52, 81–85.
Beck, A. T., Rial, W. Y., & Rickels, K. (1974). Short form of depression inventory: Cross-validation. Psychological
Reports, 34, 1184–1186.
Bennett, D. S., Ambrosini, P. J., Bianchi, M., Barnett, D., Metz, C., & Rabinovich, H. (1997). Relationship of beck
depression inventory factors to depression among adolescents. Journal of Affective Disorders, 45, 127–134.
Beyers, J. M., & Loeber, R. (2003). Untangling developmental relations between depressed mood and delinquency in
male adolescents. Journal of Abnormal Child Psychology, 31, 247–266.
Biederman, J., Faraone, S., Mick, E., & Lelon, E. (1995). Psychiatric comorbidity among referred juveniles with major
depression: Fact or artifact. Journal of the American Academy of Child and Adolescent Psychiatry, 34, 579–590.
Capaldi, D. M. (1992). Co-occurrence of conduct problems and depressive symptoms in early adolescent boys: II.
A 2-year follow-up at grade 8. Development and Psychopathology, 4, 125–144.
Caron, C., & Rutter, M. (1991). Comorbidity in child psychopathology: Concepts, issues and research strategies.
Journal of Child Psychology and Psychiatry, 32, 1063–1080.
Costello, E. J., Mustillo, S., Erkanli, A., Keeler, G., & Angold, A. (2003). Prevalence and development of psychiatric
disorders in childhood and adolescence. Archives of General Psychiatry, 60, 837–844.
Domalanta, D. D., Risser, W. L., Roberts, R. E., & Risser, J. M. (2003). Prevalence of depression and other psychiatric
disorders among incarcerated youths. Journal of the American Academy of Child and Adolescent Psychiatry, 42,
477–484.
Feehan, M., McGee, R., & Williams, S. M. (1993). Mental health disorders from age 15 to age 18 years. Journal of the
American Academy of Child and Adolescent Psychiatry, 32, 1118–1126.
Ferdinand, R. F., Verhulst, F. C., & Wiznitzer, M. (1995). Continuity and change of self-reported problem behaviors
from adolescence into young adulthood. Journal of the American Academy of Child and Adolescent Psychiatry, 34,
680–690.
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006
ARTICLE IN PRESS
M. Ritakallio et al. / Journal of Adolescence ] (]]]]) ]]]–]]]
15
Fergusson, D. M., Lynskey, M. T., & Hoorwood, L. J. (1996). Origins of comorbidity between conduct and affective
disorders. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 451–460.
Fergusson, D. M., Wanner, B., Vitaro, F., Horwood, J., & Swain-Campbell, N. (2003). Deviant peer affiliations and
depression: Confounding or causation? Journal of Abnormal Child Psychology, 31, 605–618.
Fergusson, D. M., & Woodward, L. J. (2002). Mental health, educational, and social role outcomes of adolescents with
depression. Archives of General Psychiatry, 59, 225–231.
Flannery, D., Singer, M., & Wester, K. (2001). Violence exposure, psychological trauma, and suicide risk in community
sample of dangerously violent adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 40,
435–442.
Fountoulakis, K. N., Iacovides, A., Kleanthous, S., Samolis, S., Gougoulias, K., St Kaprinis, G., et al. (2003). The
greek translation of the symptoms rating scale for depression and anxiety: Preliminary results of the validation
study. BMC Psychiatry, 3, 21.
Fröjd, S., Marttunen, M., Pelkonen, M., von der Pahlen, B., & Kaltiala-Heino, R. (2006). Perceived financial difficulties
and maladjustment outcomes in adolescence. The European Journal of Public Health.
Goodyer, I. M., Herbert, J., Secher, S., & Pearson, J. (1997). Short-term outcome of major depression: I comorbidity
and severity at presentation as predictors of persistent disorder. Journal of the American Academy of Child and
Adolescent Psychiatry, 36, 179–187.
Grilo, C. M., Becker, D. F., Fehon, D. C., Edell, W. S., & McGlashan, T. H. (1996). Conduct disorders, substance use
disorders, and coexisting conduct and substance use disorders in adolescent inpatiens. American Journal of
Psychiatry, 153, 914–920.
Harrington, R., Fudge, H., Rutter, M., Pickles, A., & Hill, J. (1991). Adult outcomes of childhood and adolescent
depression: II. Links with antisocial disorders. Journal of the American Academy of Child and Adolescent Psychiatry,
30, 434–439.
Helstelä, L., & Sourander, A. (2001). Self-reported competence and emotional and behavioral problems in a sample of
Finnish adolescents. Nordic Journal of Psychiatry, 55, 381–385.
Hofstra, M. B., van deer Ende, J., & Verhulst, F. C. (2002). Child and adolescent problems predict DSM-IV disorders
in adulthood: A 14-year follow-up of a Dutch epidemiological sample. Journal of the American Academy of Child and
Adolescent Psychiatry, 41, 182–189.
Junger-Tas, J., Terlouw, G., & Klein, M.(Eds.). (1994). Delinquent behaviour among young people in the western world:
first results of the international self-report delinquency study. Amsterdam: RDC, Ministry of Justice.
Kaltiala-Heino, R., Rimpela, M., Rantanen, P., & Laippala, P. (1999). Finnish modification of the 13-item beck
depression inventory in screening an adolescent population for depressiveness and positive mood. Nordic Journal of
Psychiatry, 53, 451–457.
Kaltiala-Heino, R., Rimpela, M., Rantanen, P., & Laippala, P. (2001). Adolescent depression: The role of
discontinuities in life course and social support. Journal of Affective Disorders, 64, 155–166.
Kiesner, J. (2002). Depressive symptoms in early adolescence: Their relations with classroom problem behavior and
peer status. Journal of Research on Adolescence, 12, 463–478.
Kim, K. J., Conger, R. D., Elder, G. H., & Lorenz, F. O. (2003). Reciprocal influences between stressful life events and
adolescent internalizing and externalizing problems. Child Development, 74, 127–143.
Lewinsohn, P. M., Gotlib, I. H., & Seeley, J. R. (1995). Adolescent Psychopathology: IV. Specificity of psychosocial
risk factors for depression and substance abuse in older adolescents. Journal of the American Academy of Child and
Adolescent Psychiatry, 34, 1221–1229.
Lewinsohn, P. M., Roberts, R. E., Seeley, J. R., Rohde, P., Gotlib, I. H., & Hyman, H. (1994). Adolescent
psychopathology: II Psychosocial risk factors for depression. Journal of Abnormal Psychology, 103, 302–315.
Loeber, R., Burke, J. D., Lahey, B. B., Winters, A., & Zera, M. (2000). Oppositional defiant and conduct disorder:
A review of the past 10 years, Part I. Journal of the American Academy of Child and Adolescent Psychiatry, 39,
1468–1484.
Loeber, R., Farrington, D.P., Stouthamer-Loeber, M., & Van Kammen, W.B. (1998). Antisocial behavior and mental
health problems. Explanatory factors in childhood and adolescence. New Jersey.
Loeber, R., Russo, M. F., Stouthamer-Loeber, M., & Lahey, B. B. (1994). Internalizing problems and their relations to
the development of disruptive behaviors in adolescence. Journal of Research on Adolescence, 4, 615–637.
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006
ARTICLE IN PRESS
16
M. Ritakallio et al. / Journal of Adolescence ] (]]]]) ]]]–]]]
Maughan, B., Rowe, R., Messer, J., Goodman, R., & Meltzer, H. (2004). Conduct disorder and oppositional defiant
disorder in a national sample: Developmental epidemiology. Journal of Child Psychology and Psychiatry, 45,
609–621.
Morgan, C. J., & Cauce, A. M. (1999). Predicting DSM-III-R disorders from the youth self-report: Analysis of data
from a field study. Journal of the American Academy of Child and Adolescent Psychiatry, 38, 1237–1245.
Overbeek, G., Vollebergh, W., Meeus, W., Engels, R., & Luijpers, E. (2001). Course, co-occurrence, and longitudinal
associations of emotional disturbance and delinquency from adolescence to young adulthood: A six-year three-wave
study. Journal of Youth and Adolescence, 30, 401–426.
Raitasalo, R. (1995). Elämänhallinta sosiaalipolitiikan tavoitteena. Sosiaali- ja terveysturvan tutkimuksia I. Helsinki:
Kansaneläkelaitos.
Ritakallio, M., Kaltiala-Heino, R., Kivivuori, J., & Rimpelä, M. (2005). Brief report: Delinquent behaviour and
depression in middle adolescence: A Finnish community sample. Journal of Adolescence, 28, 155–159.
Ritakallio, M., Luukkaala, T., Marttunen, M., Pelkonen, M., & Kaltiala-Heino, R. Depression and delinquency in
middle adolescence: the role of perceived social support. American Journal of Orthopsychiatry, [submitted for
publication].
Rohde, P., Lewinsohn, P. M., & Seeley, J. R. (1991). Comorbidity of unipolar depression: II. Comorbidity with other
mental disorders in adolescents and adults. Journal of Abnormal Psychology, 100, 214–222.
Roza, S. J., Hofstra, M. B., van der Ende, J., & Verhulst, F. C. (2003). Stable prediction of mood and anxiety disorders
based on behavioral and emotional problems in childhood: A 14-year follow-up during childhood, adolescence, and
young adulthood. American Journal of Psychiatry, 160, 2116–2121.
Vermeiren, R., Deboutte, D., Ruchkin, V., & Schwab-Stone, M. (2002). Antisocial behaviour and mental health.
Findings from three communities. European Child & Adolescent Psychiatry, 1, 168–175.
Vermeiren, R., Schwab-Stone, M., Ruchkin, V., De Clippele, A., & Deboutte, D. (2002). Predicting recidivism in
delinquent adolescents from psychological and psychiatric assessment. Comprehensive Psychiatry, 43, 142–149.
Wiesner, M. (2003). A Longitudinal latent variable analysis of reciprocal relations between depressive symptoms and
delinquency during adolescence. Journal of Abnormal Psychology, 112, 633–645.
Please cite this article as: Ritakallio, M., et al. Continuity, comorbidity and longitudinal associations between depression and
antisocial behaviour in middle adolescence: A 2-year.... Journal of Adolescence (2007), doi:10.1016/j.adolescence.2007.06.006