! " # ! "
$% ! &!'( )*
!*! ) +,& *!* "!( ,) ""!(! -..- Dissertation for the Degree of Doctor of Philosophy (Faculty of Medicine) in Child and Adolescent Psychiatry presented at Uppsala University in 2002.
ABSTRACT
Ramklint, M. 2002. Influence of Child and Adolescent Psychopathology on Adult Personality Disorder. Acta Universitatis Upsaliensis. Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1163. 54 pp. Uppsala. ISBN 91-554-5338-4.
Individuals afflicted with childhood and adolescent mental disorders have an increased risk for poor outcome in adulthood. The progression of psychopathology from childhood to adult life may be influenced by a multitude of interacting variables, both biological and psychosocial. There is limited information on the relationships between child psychopathology and adult personality and personality disorders. The main aim of this thesis was therefore to gain better knowledge concerning adult personality outcome in patients with early onset of mental disorders. Former child psychiatric patients as compared to controls had a significantly higher prevalence of all DSM-IV personality disorders (38.0 vs. 10.9 percent, p<0.001) and also a considerably higher personality disorder co-morbidity. They also had more psychosocial and environmental problems. This was exaggerated in those diagnosed with a personality disorder. Major depression, disruptive disorders and substance use disorders at a young age were strong predictors for adult personality disorder. Patients with an early onset major depression had more personality disorders and more deviant personality traits than those with a late onset. Forensic psychiatric male patients diagnosed with a previous conduct disorder as compared to those without had more cluster B personality disorders, and more repeated violent criminality and mixed abuse. They also exhibited more deviant personality traits and higher psychopathy scores. The instrument "Child and Adolescent Psychiatric Screening Inventory-Retrospect" had acceptable sensitivity and specificity for assessment of child psychiatric disorders. Subscales demonstrated good internal reliability (Crohnbach´s alpha = 0.76-0.93). The results suggest that adult personality disturbances are prevalent in individuals affected with mental problems at young ages. A better understanding of the transition of psychopathology from childhood to adulthood and a better identification of those at risk will be of help in attempts to prevent permanent impact on the adult personality.
Key words: child psychiatry, personality, major depressive disorder, disruptive disorder, conduct disorder, retrospective analysis, self-assessment
Mia Ramklint, Department of Neuroscience, Psychiatry and Child and Adolescent Psychiatry, University Hospital, SE-751 85 Uppsala, Sweden
Mia Ramklint, 2002 ISSN 0282-7476 ISBN 91-554-5338-4 Printed in Sweden by Uppsala University, Tryck&Medier, Uppsala, 2002
2 The childhood shows the man, as morning shows the day
Paradise Regained, 1671 John Milton (1608–1674)
3 LIST OF PAPERS
This thesis is based on the following original papers, which will be referred to in the text by their Roman numerals:
I Ramklint M, von Knorring A-L, von Knorring L, Ekselius L. Personality disorders in former child psychiatric patients. Submitted.
II Ramklint M, von Knorring A-L, von Knorring L, Ekselius L. Child and adolescent psychiatric disorders predicting personality disorders: A follow-up study. Nordic Journal of Psychiatry, 2002. In press.
II Ramklint M, Ekselius L. Personality disorders and personality traits in early onset versus late onset major depression. Journal of Affective Disorders, 2002. In press.
IV Ramklint M, Stålenheim EG, von Knorring A-L, von Knorring L. Conduct disorder and personality in a forensic psychiatric population. The European Journal of Psychiatry, 2001:15:245-254.
V Ramklint M, von Knorring A-L, von Knorring L, Ekselius L. Child and Adolescent Psychiatric Screening Inventory-Retrospect (CAPSI-R): a questionnaire for adults concerning child and adolescent mental disorders. European Psychiatry, 2002:17:61-68.
Reprints were made with the permission of the publishers.
4 CONTENTS
ABSTRACT 2 LIST OF PAPERS 4 CONTENTS 5 ABBREVIATIONS 6 INTRODUCTION 7 Psychopathology in childhood and adult outcome 7 Temperament, personality traits and personality 7 Mental disorders and diagnostic classification 9 Models of the relationship between clinical syndromes and 12 personality disorders Epidemiology 13 Developmental pathways 13 Assessments 16 AIMS OF THE PRESENT THESIS 18 METHODOLOGY 18 Subjects and Methods Papers I, II and V 19 Subjects and Methods Paper III 23 Subjects and Methods Paper IV 25 Ethics 26 RESULTS 26 Personality disorders in former child psychiatric patients (paper I) 26 Predictive value of child and adolescent psychiatric disorders for 27 adult personality disorder (paper II). Personality traits and personality disorders in early onset vs. late 28 onset major depression (paper III) Personality traits and personality disorders in early onset vs. late 29 onset of antisocial behaviour (paper IV) The Child and Adolescent Psychiatric Screening Inventory- 30 Retrospect (Paper V) DISCUSSION 32 Methodological considerations 32 Adult personality disorders 34 Personality traits 35 Major depression and personality characteristics 36 Substance use disorders and personality characteristics 38 Disruptive disorders and personality characteristics 38 The CAPSI-R 40 CONCLUSIONS 41 ACKNOWLEDGEMENTS 42 REFERENCES 44
5 ABBREVIATIONS
ADHD Attention deficit hyperactivity disorder
APA American Psychiatric Association
CAPSI-R The Child and Adolescent Psychiatric Screening Inventory- Retrospect
CI Confidence interval
DIP-Q The DSM-IV and ICD-10 Personality Questionnaire
DSM Diagnostic and Statistical Manual of Mental Disorders
DSM-III-R Diagnostic and Statistical Manual of Mental Disorders, 3rd edition revised
DSM-IV Diagnostic and Statistical Manual of Mental Disorders, 4th edition.
GAF Global Assessment of Functioning
ICD-10 International Classification of Diseases and Related Health Problems, 10th revision
KSP Karolinska Scales of Personality
MADRS Montgomery Åsberg Depression Rating Scale
OR Odds ratio
PCL-R The Psychopathy Checklist-Revised
SCID I Structured Clinical Interview for DSM for Clinical Syndromes
SCID II Structured Clinical Interview for DSM for Personality Disorders
SD Standard deviation
WHO World Health Organization
6 INTRODUCTION Psychopathology in childhood and adult outcome Individuals afflicted with childhood and adolescent mental disorders have an increased risk for poor outcome in adulthood. Thus, there are significantly higher rates of delinquency, disability and mortality in former adolescent psychiatric patients followed-up decades after hospitalisation as compared to the general population (38, 58, 70, 86-89, 93, 95, 138, 160, 177).
Several studies of child psychiatric symptomatology using prospective designs have been conducted (2, 28, 55, 56, 112, 122). Results generally indicate moderate stability of problem behaviour from childhood to adolescence (106). However, specific styles of behaviour in early childhood have been associated with specific deviant personality traits in young adulthood (28, 76). Furthermore, anxious/depressed, disruptive and aggressive syndromes in young adults have been strongly predicted by corresponding adolescent syndromes (2, 122).
Knowledge about the continuity and discontinuity of psychopathology from childhood to adulthood is indispensable for answering questions concerning the origins of adult psychopathology and thus providing possibilities for preventive actions. The course of pathology with respect to both clinical disorders (Axis I) and personality disorders (Axis II) across the child-to-adult transition is, however, still largely unclear (53, 136). Furthermore, there is a considerable lack of understanding about the causal relationship between childhood and adolescent mental disorders and adult clinical disorders and personality disorders. This work consequently focuses upon the relationship between child and adolescent mental disorders and adult personality disorders.
Temperament, personality traits and personality Temperament refers to the raw biological materials from which personality emerges. It reflects the "constitutional soil, the biochemistry, endocrinology, and neurological structure that underlie the way we respond to stimulation in particular areas" (59). The
7 functional thinking that applies to the internal regulatory roles of temperament derives from psychophysiology, the Pavlovian tradition, and the works of Eysenck (49), Gray (61), Sjöbring (147) and Schalling (141). It is abundantly clear that individual children differ markedly from one another in a host of different ways, and that these differences can be measured with reasonable reliability (23, 24, 42). There is consensus that activity level and emotionality are considered to be temperamental dimensions. In addition, some theorists regard reactivity and sociability as temperamental (59). In the first weeks of life, temperamental dimensions constitute nearly the entire personality of the newborn, which makes the infant a prototypical or model system for the study of temperament. As development proceeds, the expression of temperament becomes more influenced by experience and context. It has been well demonstrated that temperamental differences matter in terms of their implications for later development and the risk of psychiatric morbidity (158, 159).
Temperamental traits refer to correlated patterns of reactions, while personality traits refer to a conglomerate of habits. Washing one’s hands repeatedly is a habit, while cleanliness is a trait. Cross-situational consistency is more likely to be assumed for traits than for habits (25). Personality traits are distributed according to a normal bell- shaped curve. The behaviour of people in the middle range would be expected to be less predictable than the behaviour of those scoring in the upper or lower quartiles.
Personality is a much broader concept than the collection of temperamental or personality traits. It involves processing of information, the synthesising and integrative functions of the mind, such as motivation and recognition. Personality is more than temperament; it also includes self-concept, attitudes, specific expectations, perceptual dispositions, motivations, values, standards and defence mechanisms. Allport (7), one of the most influential personality theorists, referred to ”dynamic organisation”, a phrase that emphasises both an active mental role and also the importance of developmental changes. Personality is, in a sense, the patterning that each person, as a thinking being, develops as a way of dealing with the traits with which he/she is endowed, the social contexts that he/she encounters, and the
8 experiences he/she has had. The various characteristics comprising our personality do not display equal degrees of consistency and stability (84).
Mental disorders and diagnostic classification A mental disorder is conceptualised as a clinically significant behavioural or psychological syndrome or pattern that occurs in an individual and is associated with present distress or disability or with a significantly increased risk of suffering death, pain, disability, or an important loss of freedom (10). Mental disorders refer both to clinical disorders and personality disorders. Classification constitutes a means of ordering information, of grouping phenomena, and of providing a language by which to communicate with other people (77). The need for a classification of mental disorders has been obvious throughout the history of medicine. However, there has been little agreement among psychiatrists regarding theoretical concepts, although there has been fairly good agreement on broadly defined, simple phenomenological diagnoses (103). Thus it has become more accepted that psychiatric classifications need to be based on patterns of symptomatology rather than on theories (10). Although the classification of mental illnesses had much earlier origins, modern empirically- based schemes started with Kraepelin’s use of outcome differences to differentiate between schizophrenia and manic-depressive disorder (91). The two classification systems most commonly used today are the International Classification of Diseases, tenth version (ICD-10) (167), and the Diagnostic and Statistical Manual of Mental Disorders, fourth version (DSM-IV) (10).
Some disorders, such as attention deficit hyperactivity disorder (ADHD), are already evident in childhood or adolescence, while others, such as panic disorder, may not come to clinical attention until late adolescence or adulthood. The division between child and adult mental disorders according to age of onset is for convenience only, and is not absolute.
9 The DSM system In 1952 the American Psychiatric Association’s Committee on Nomenclature and Statistics published the first edition of DSM (DSM-I). Four editions have been published since then, the latest in 1994. The approach of DSM is atheoretical with regard to aetiology. The definitions of the disorders usually consist of descriptions of clinical features. DSM-IV is a multiaxial system that evaluates the patient along several variables and that contains five axes (10). Axis I and Axis II comprise the entire classification of mental disorders, including 17 major classifications and more than 300 specific disorders. Axis I consists of clinical disorders, and axis II of personality disorders and mental retardation. In many instances a patient may have more than one disorder and may have disorders on both axes. Axis III is for reporting current general medical conditions that are potentially relevant to the understanding or management of the individual’s mental disorder. Axis IV is used to code the psychosocial and environmental problems that significantly contribute to the development or the exacerbation of the current disorder. Finally, Axis V is the Global Assessment of Functioning (GAF) scale, assessing functioning as a composite of three major areas, social functioning, occupational functioning and psychological functioning. The GAF scale is based on a continuum of mental health and mental illness and is a 100-point scale, with 100 representing the highest level of functioning in all areas.
Dimensions or categories DSM-IV is a categorical classification. Categories assume the existence of discrete boundaries both between different disorders and between normality and abnormality (79). The categorical approach is mainly based on clinical tradition and anchors the system within psychiatry and medicine (62). Much indicates that a categorical concept is a considerable simplification of the underlying biology, and that many symptoms and underlying processes operate in a continuous spectrum. Thus, many of the symptoms that are associated with clinical conditions are observed, but in a milder form, in individuals who do not fulfil the necessary criteria for disease. There are, for example, obvious similarities between social phobia and what can be considered
10 "normal" shyness, between conduct disorder and ”normal” disruptive behaviour, and certainly between personality disorders and ”normal” personalities. Thus, there is empirical support for a dimensional approach, especially with respect to personality disorders and in studies on underlying mechanisms (34, 44). Furthermore, in a study of the adolescent consequences of clinical and threshold-level childhood psychiatric disorders, there was a considerable continuity of psychopathology for those who received a childhood diagnosis, not only in those with distinct disorders but also in those with threshold-level disorders with functional impairment (35).
Personality disorders Personality pathology comprises stable and consistent traits that persist inflexibly, are exhibited inappropriately, and foster vicious circles that perpetuate and intensify already present difficulties. According to the DSM-IV (10), a personality disorder is "an enduring pattern of inner experience and behaviour that deviates markedly from the expectations of the individual’s culture". This pattern has an onset in adolescence or early adulthood, is stable over time, and leads to distress and impairment. Personality disorders can be organised in three distinct clusters based on characteristics common to each cluster. The cluster organisation was largely the end result of a committee process (108). Cluster A, the odd and eccentric cluster, consists of the paranoid, schizoid and schizotypal personality disorders. Common features of this cluster are constricted emotions, a reduced range of affect, mistrust of others, aloofness and social withdrawal (10, 108). Schizotypal personality disorder also includes psychotic-like symptoms and has been linked genetically and phenomenologically to schizophrenia (83). Cluster B includes antisocial, borderline, histrionic and narcissistic personality disorders. Individuals with a cluster B diagnosis are often described as dramatic, emotional and erratic. Impulsiveness is a core characteristic of at least three of the diagnoses in this cluster (10, 108). Cluster C, the anxious cluster, includes avoidant, dependent, and obsessive-compulsive personality disorders. Individuals with a cluster C personality disorder tend to manifest anxious and fearful behaviour (10, 108).
11 Psychopathy Psychopathy, as defined by Cleckley (30) and subsequently by Hare (67), offers an alternative conceptualisation to the DSM antisocial personality disorder. Hare defined psychopaths as grandiose, egocentric, manipulative and cold-hearted (67). The Psychopathy Checklist-Revised (PCL-R) was explicitly constructed as a basis for a reliable assessment (66). Factor analysis has given a two-factor solution, with one factor that appears to represent a narcissistic personality variant of the psychopathic pattern, including callous-unemotional interpersonal traits, and another factor that appears more directly related to those with an overtly antisocial lifestyle (68). A significant correlation between antisocial personality disorder and psychopathy has been reported in male prisoners (67) and in female methadone patients (133). Psychopathy has been shown to be a robust predictor of violence and violent recidivism (71).
Models of the relationship between clinical syndromes and personality disorders Personality disorders and clinical syndromes are related in several ways (37, 76, 108, 168). Quantitative descriptions of clinical and general populations have revealed substantial co-morbidity between classes of child and adolescent psychiatric disorders (9, 107). Co-morbidity is also common between adult clinical syndromes (85), among personality disorders (45), and between clinical syndromes and personality disorders (33, 113, 149). Different models for these relationships have been suggested (41, 108). According to the vulnerability model, personality disorders dispose the individual to the development of an axis I disorder. When coping strategies are limited or impoverished, the probability of developing an axis I disorder such as anxiety or depression is greatly increased. Personality-disordered patients often evoke the very stressors under which these disorders will be developed. In the complication model, the causality of the vulnerability model is reversed. Axis I conditions create a predisposition to personality change. For example, patients who experience a severe and long-lasting depression may internalise pessimism and hopelessness, creating a personality change at the trait level. In the spectrum model, personality disorders and
12 axis I conditions are seen as developing from the same constitutional soil and therefore as existing on a continuum. This model has been used to describe the nature of relationships between e.g. the cluster A personality disorders and psychotic disorders, most typically schizophrenia. The pathoplasty model holds that personality influences the course of an axis I disorder but does not itself dispose to the development of the disorder. All these models are possible and likely to be true for different individuals. In fact, it is possible that all are applicable to some degree in a single individual (41, 108).
Epidemiology A systematic review (129) of epidemiological studies done over the past four decades has found prevalence estimates of psychopathology during childhood and adolescence ranging from one to nearly 51 percent (mean =15.8). The high variability in prevalence rates obtained in previous studies reflects differences in sampling, diagnostic methods and case definitions. The cumulative incidence of psychiatric disorders (meeting DSM-III criteria) in a total population under 25 years of age in Sweden was 13.9 percent for boys and 14.2 percent for girls. These figures comprise those applying for professional help who are therefore registered at a hospital or at a clinic (170). The prevalence of personality disorders in the adult general population is estimated to be between six and 14 percent depending on the diagnostic criteria and methods of assessments used (47, 139, 161, 165). Research results indicate that the prevalence of personality disorders varies between different age groups, with a slight decrease in older groups (102). Among psychiatric patients the prevalence is much higher, with frequencies between 30 and 80 percent (8, 19, 78, 92).
Developmental pathways Biological and psychosocial factors influence development. The biological substrate (genetic and non-genetic), the interactive shaping of the environment, cognitive and social skills, self-esteem and self-efficacy, habits, cognitive sets and coping styles are all possible mediating factors. The timing as well as the nature of experiences is likely to influence their impact (136). There are both risk and protective factors and
13 interactions between them (134, 135, 173). Mental disorders are much more common in adults who have been treated for such disorders in childhood or adolescence (69, 131, 177). In a 22 to 25-year follow-up study of former child psychiatric in-patients, Thomsen (160) found that approximately one-third of the sample had at least one readmission after the age of 18 years. Childhood emotional and behaviour problems are associated with a greater risk of developing adolescent personality disorder (14). Furthermore, childhood mental disorders increase the risk of personality disorders in young adulthood (1, 127, 128, 160). Finally, both adolescent clinical disorders and adolescent personality disorders are independent risk factors for young adult personality disorders (81). An increased number of adolescent mental disorders has also been shown to be associated with an elevated personality disorder dimensional score in young adults (97).
Many individuals fulfilling criteria for one Axis I and/or Axis II diagnosis also fulfil criteria for one or more independent diagnosis(es). Such co-morbidity (see above) has been extensively studied, as it offers a possibility of understanding underlying developmental processes (9). Although there is little evidence that any one disorder directly causes any other disorder, the presence of shared risk factors is likely to be an important component in the causation of co-morbidity. It is also likely that some co- morbid patterns, such as depression with dysthymia and oppositional defiant disorder with conduct disorder, represent developmental sequences of unitary underlying developmental psychopathological processes. Furthermore, certain patterns of co- morbidity may represent separate subtypes of their component diagnoses that might have different outcomes (9). Moreover, co-morbidity is of considerable clinical importance, as it is associated with a higher level of impairment and service utilisation (17).
Disruptive disorders and personality disorders Using a sociological approach, Robins (130) showed a convincing link between childhood conduct problems and adult sociopathy (later named antisocial personality disorder). This link has been confirmed in several studies (73, 100). However, among
14 children defined as aggressive, antisocial or delinquent on one occasion, only 30 to 50 percent develop adult criminality (130, 153). Children with conduct problems constitute a very heterogeneous group (73, 94). They develop various patterns of offending. It seems that the chronic offenders represent a distinct subgroup, with an early onset of symptoms and a pattern where minor antisocial behaviours are followed by more serious ones (51, 99).
In both clinical and epidemiological populations there is an overlap between the diagnoses of conduct disorder and ADHD in 30 to 50 percent of cases (157). Children with both ADHD and conduct disorder are considered to be at the greatest risk for developing antisocial personality in adulthood, particularly psychopathy (99). Consistent with findings in adults, Frick (57) found two separable psychological dimensions in clinically referred children with conduct disorder. The first involved a callous-unemotional interpersonal style. The second dimension involved poor impulse control and antisocial behaviour. The presence of both callous-unemotional traits and significant conduct problems characterises a subgroup of antisocial children who show a very severe pattern of antisocial behaviour that corresponds closely to the adult conceptualisation of psychopathy (29). Finally, there is an association between childhood disruptive behaviours and personality disorders in all clusters, particularly between conduct disorder and adult antisocial personality disorder (81, 128, 131, 176).
Mood disorders and personality disorders Studies have demonstrated that subjects with early onset major depression are more likely than those with late onset major depression to have co-morbid personality disorders (52, 140). Personality disorders from cluster B and cluster C are most commonly associated with major depression (113, 119, 120, 149). The co-morbidity between personality disorders and mood disorders has raised questions about etiological factors (146) and developmental pathways (5). Mood disorders with onset in childhood or adolescence are common, with prevalence rates for major depressive disorder ranging between 0.4 and 2.5 percent in children and between 0.4 and 8.3
15 percent in adolescents (for review see (18)). Early onset major depression is a condition with estimated recurrence rates of 60 to 70 percent (69, 126). Youths with major depressive disorder convert to bipolar illness more frequently than do adults (90), which was taken as an argument indicating that early onset major depression is a particularly serious form of affective illness. On the other hand, a recent genetic aggregation study suggests a separation of the genetic linkage for early onset major depression from that of bipolar disorder (101).
Assessments Diagnostic instruments such as interviews and questionnaires are based on diagnostic criteria. An assessment instrument must be safe, feasible and have good psychometric properties (for review see (32)). It should be reliable, i.e. give consistent results when used by different examiners (inter-rater reliability) or at different times (test-retest reliability). It should be valid, i.e. measure what it purports to measure. The three most central aspects of validity are termed content validity, criterion-related validity and construct validity. Content validity reflects how well a test as such covers the aspects of a person, situation, behaviour, life, etc., it is meant to cover. Criterion-related validity, or empirical validity, is a judgement regarding how adequately a test score can be used to infer information about an individual’s most probable standing with respect to a defined measure of interest. Construct validity refers to a demonstration that an instrument does measure what it intends to measure with respect to certain “constructs”. The test should also be sensitive, i.e. be able to detect all the true cases of the disorder identified by the criterion instrument, and specific, i.e. identify the true non-cases identified by the criterion instrument.
The diagnostic method most often recommended in epidemiological child psychiatric research is the structured interview (75), e.g. the Diagnostic Interview Schedule for Children (144). One reason for this is that there have been arguments against the adequacy of using questionnaires, or checklists, for classifying childhood psychiatric disorders (137). In a recent study, however, Boyle et al. (21) found only small differences in reliability and validity between structured interviews and checklists. A
16 consensus has evolved that multiple measures are required, preferably involving different informants and repeated measures over time, a routine that has led to more robust findings (54, 137).
Retrospective reporting Studies addressing complex issues, like the long-term course of mental disorders in this project, require reliable information on all parameters under study. In order to assure this, some investigators have used a longitudinal approach, where up-to date information has been obtained from the subjects under study at various time points (81, 176). Such studies are cumbersome, require a long time, and are often associated with a very high attrition rate. This approach has been supported by a widespread claim that alternative strategies such as retrospective reporting are unreliable. A number of recent reports, however, suggest that such problems have been exaggerated (22, 104). In an epidemiological longitudinal study (72), retrospective reports thus showed good agreement with earlier measures of factual issues, such as family residence changes, and also for the subject’s reporting of some aspects of his/her own performance and characteristics in childhood. Bernstein et al. (15) found strong support for the reliability and validity of the Childhood Trauma Questionnaire that retrospectively measures child abuse and neglect. However, it is possible that different life events are more easily recalled than problems of behaviour and emotional regulation. In the study by Henry et al. (72), agreement was much poorer on questions dealing with hyperactivity or depressed mood at particular ages. In a study by Fendrich et al. (54), both mothers and children showed good recall of a child's major depression, whereas the recall of both mothers and children regarding any anxiety disorder was less satisfactory. The Wender Utah Rating Scale (WURS) (164) was constructed to retrospectively establish a childhood diagnosis of an ADHD. In a recent study (105), the WURS was found to be sensitive in detecting ADHD, but it miss-classified approximately half of those who did not have this disorder.
17 AIMS OF THE PRESENT THESIS
The progression of psychopathology from childhood to adult life may be influenced by a multitude of interacting variables, both biological and psychosocial. The objective in most studies has been to address the relationship between child psychopathology and clinical disorders in adulthood, while there is limited information available on what characteristics presage adult personality disorders. The principal aim of this thesis was therefore to gain better knowledge concerning adult personality outcome in patients with early onset of mental disorders.
The specific aims were:
• To investigate the long-term outcome with respect to DSM-IV personality disorders in former child psychiatric in-patients as compared to matched controls from the general population. • To examine the predictive value of different child or adolescent mental disorders for adult personality disorders in former child psychiatric patients. • To determine whether early onset major depression is related to certain personality traits and personality disorders. • To examine differences between subjects with early onset vs. late onset of antisocial behaviour concerning personality disorders and personality traits. • To describe the construction and initial validation results of a self-report instrument for retrospective assessment of child and adolescent psychopathology in adults.
METHODOLOGY
This thesis is based on three separate patient populations. Papers I, II and V comprise investigations of a group of former child psychiatric patients who are compared to controls from the general population. In paper III, major depressed primary care
18 patients are examined, and paper V comprises an investigation of a group of forensic psychiatric patients. The patient series and the healthy subjects are described in Table 1, and the assessments and statistical methods used are presented in Table 2.
Table 1. Characteristics of participants in papers I – V Paper N M/F Age mean (SD) Sample I 137 59/78 30.7 (6.8) Former child psychiatric patients 137 59/78 30.7 (6.8) Controls II 158 63/95 30.5 (7.1) Former child psychiatric patients III 400 113/287 47.2 (12.4) Depressed primary care patients IV 61 61/0 34.0 (11.0) Forensic psychiatric patients V 164 66/98 30.6 (7.0) Former child psychiatric patients 193 91/102 30.5 (6.8) Controls Note: M/F = males/females
Subjects and Methods Papers I, II and V Subjects Former patient group Four hundred and seven former inpatients at the Department of Child and Adolescent Psychiatry, University Hospital, Uppsala, Sweden, admitted for care during 1972, 1984, 1987 or 1992, could be completely identified retrospectively. The civil population register could not identify addresses for 29 of these 407 subjects, and another 19 persons had died, leaving 359 subjects who were contacted.
Self-report instruments (see below) were mailed to each of these 359 former patients together with a stamped, self-addressed envelope and an explanatory letter in which the aim of the study was described and participant anonymity was guaranteed. Two reminders were sent to non-responders only. Seventy-nine subjects (22.0 %) returned their questionnaires but refused to participate and 116 (32.3 %) did not respond. Thus, a total of 164 subjects (45.7 %) were eligible for analysis in paper V (Table 1).
19 Another six subjects were excluded from analyses in paper II due to incomplete enquiries.
Control group Three hundred and fifty-nine age and sex-matched individuals selected from the civil population register were asked to participate as the control group. The same procedure was used for collecting data as in the former patient group. Seventy-five subjects (20.9 %) returned their questionnaires but refused to participate, and 91 (25.3 %) did not respond at all. One person answered the questions by telephone. Thus, a total of 193 subjects (53.8 %) were eligible for analysis in paper V (Table 1).
Matched pairs Out of the 164 former patients and 193 controls, 137 age and sex- matched pairs were eligible for analysis in paper I (Table 1). There were no differences regarding age, sex or presence of any personality disorder between the matched subjects and the unmatched remaining participants.
Assessments Childhood psychiatric diagnoses Based on information from the medical records, all participants in the former patient group had their childhood psychiatric diagnoses coded according to the DSM-IV. The diagnostic procedure was performed in a structured way using checklists of the DSM- IV criteria. All symptoms mentioned in the records were regarded as being of clinical importance. In a previous work, using the same procedure, good diagnostic agreement between two of the authors (M.R. and A-L.v.K.) was achieved (for detailed information see (125)). In paper II, a diagnosis of ADHD, conduct disorder and oppositional defiant disorder were referred to as a disruptive disorder. Furthermore, substance abuse and substance dependence were referred to as a substance-related disorder.
20 Table 2. Assessment methods and statistics. Paper Assessments Statistics I DSM-IV Checklist McNemar’s test DIP-Q Paired t-test GAF Student’s t-test Wilcoxon’s matched-pairs signed-rank test II DSM-IV Checklist Chi-square test DIP-Q Logistic regression analysis III DSM-III-R diagnostic criteria Chi-square test MADRS Fischer’s Exact test KSP Logistic regression analysis SCID-II Screen Mann-Whitney U-test Student’s t-test IV SCID I and II Chi-square test KSP Discriminant analysis PCL-R Student’s t-test Registered criminality V DSM-IV Checklist Cohen’s kappa coefficient CAPSI-R Cronbach’s alpha Sensitivity/Specificity
The DSM-IV and ICD-10 personality questionnaire Personality disorders were assessed by means of the DSM-IV and ICD-10 Personality Questionnaire (DIP-Q). The DIP-Q is a true/false self-report questionnaire designed to measure all DSM-IV and all ICD-10 personality disorders (115). Out of a total of 140 items in the questionnaire, 135 items reflect the diagnostic criteria of the DSM-IV and ICD-10 personality disorders. The items consist of brief statements reflecting the main aspect of the corresponding criterion, and the respondent is asked to score the statement as ‘true’ or ‘false’. The remaining five items constitute the impairment and distress scale (ID-scale) and reflect aspects of the general criteria for personality disorders (10). A self-report version of the Global Assessment of Functioning (GAF)
21 scale (20) is also included. This scale consists of the original 0-100 point GAF scale included in DSM-IV (10), but with fewer defining characteristics than in the original version. A short scale, comprising 11 ‘yes’ or ‘no’ questions concerning psychosocial and environmental problems according to axis IV in the DSM-IV, is included as a separate part of the DIP-Q.
The DIP-Q has been validated by comparing results from the questionnaire with those from a structured clinical interview (114, 116). Agreement was acceptable on both the global and cluster levels. The agreement between DIP-Q and the structured interview for any DSM-IV personality disorder as measured by Cohen’s Kappa was 0.61, and the sensitivity and specificity were 0.84 and 0.77, respectively. In papers I and II, a categorical personality disorder diagnosis was based on trait criteria for the specific personality disorder and a score of two or higher on the ID scale. A dimensional score was obtained from the proportion of fulfilled trait criteria for each personality disorder diagnosis.
Child and Adolescent Psychiatric Screening Inventory-Retrospect The Child and Adolescent Psychiatric Screening Inventory-Retrospect (CAPSI-R) questionnaire was constructed with the intention of retrospectively assessing perceived psychiatric psychopathology during childhood and adolescence. It consists of 146 items, of which 140 correspond to the following DSM-IV standardised diagnostic categories: reading disorder, mathematics disorder, disorder of written expression, developmental co-ordination disorder, expressive language disorder, stuttering, autistic disorder, Asperger´s disorder, attention deficit/hyperactivity disorder, conduct disorder, oppositional defiant disorder, Tourette´s disorder, separation anxiety disorder, selective mutism, panic disorder, obsessive-compulsive disorder, posttraumatic stress disorder, anorexia nervosa, bulimia nervosa, substance dependence, substance abuse, major depressive episode, manic episode and psychotic episode. The remaining six items reflect aspects of impairment.
22 Subjects and Methods Paper III Subjects Eligible subjects were 400 depressed primary care patients participating in a randomised, double-blind, multicenter study (48) (Table 1). Included patients were required to meet the DSM-III-R criteria for major depressive disorder and to have a minimum score of 21 on the Montgomery Åsberg Depression Rating Scale (MADRS) (described below) (109).
A clinical diagnosis of a major depressive disorder was set using the DSM-III-R criteria. The age at onset of the first major depressive episode was established by interview examination, and the cut-off for early onset major depression was set at 26 years of age.
Assessments Montgomery-Åsberg Depression Rating Scale Severity of depression was assessed by means of the MADRS, a depression rating scale designed to be particularly sensitive to treatment changes. The MADRS is a subscale constructed from the Comprehensive Psychopathological Rating Scale (11). It consists of 10 items concerning depressive symptomatology. Each item can be scored within a range from 0 to 6. The MADRS has been shown to have good psychometric properties (109).
SCID screen Personality disorders were assessed by the use of the Swedish modified version of the SCID screen questionnaire (46). The questions in this self-report are similar to the questions in the Structured Clinical Interview for DSM-III-R personality disorders (SCID II) (152). The SCID screen has been validated by comparing the results obtained from clinically performed SCID II interviews. By adjusting the cut-off level for diagnosis (one more criterion required for each of the personality disorders), the overall Kappa coefficient of agreement was 0.78 and the sensitivity and specificity were 0.87 and 0.75, respectively (46).
23 Karolinska Scales of Personality The Karolinska Scales of Personality (KSP) was used to assess personality traits (63, 142). The inventory comprises 135 items with a four-point response format grouped in 15 scales. Four scales are related to anxiety proneness: Somatic anxiety (autonomic disturbances, vague distress and panic attacks), Psychic anxiety (cognitive, social anxiety, worrying, insecurity), Muscular tension (subjective muscular tenseness and aches, difficulties in relaxing), and Psychasthenia (low degree of mental energy, easily fatigued). Three scales are related to vulnerability to non-inhibitory psychopathology: Impulsiveness (non-planning, impulsive), Monotony avoidance (need for change and action), and Socialisation (positive childhood experiences, good school and family adjustment). Six scales are related to aggression and hostility: Verbal aggression (aggressive feelings expressed in style and content of speech), Indirect aggression (undirected expressive aggression), Irritability (irritable, lacking patience), Suspicion (distrusting people’s motives), Guilt (remorseful), and Inhibition of aggression (non- assertive, sad rather than angry). Also included are: Detachment, which is related to social withdrawal, and Social desirability, which refers to conformity and control.
The scales were composed for the purpose of operationalising and measuring constructs defining vulnerability for different forms of psychopathology. Most of the scales are based on hypotheses of biologically relevant temperament dimensions. In most scales, items were primarily selected based on rational-theoretical considerations rather than by using factor-analytical or empirical techniques. The scales have later been subjected to psychometric analyses (63). Personality traits, as assessed by the KSP in a non-patient sample, have been found to be stable over nine years, both with regard to absolute and relative stability (64). The mean scores for the different scales have been transformed into normative T scores (mean=50, SD=10), based on a Swedish age and gender-stratified, non-patient sample (13).
24 Subjects and Methods Paper IV Subjects Research subjects were 61 consecutive men who were referred by the courts to the Department of Forensic Psychiatry in Uppsala for major forensic examinations between 1992 and 1994 (Table 1). All subjects were detoxified from drugs and alcohol when investigated. Criminal behaviour was described in accordance with the subjects’ registered criminality. Subjects sentenced three or more times were regarded as repeatedly criminal.
Assessments The Structured Clinical Interview for DSM-III-R Assessment of psychiatric disorders was based on the use of the Structured Clinical Interview for DSM-III-R (SCID), a semi-structured interview which was designed to be administered by a clinician or researcher familiar with the diagnostic criteria used in the DSM-III-R. The SCID has two parts, one for axis I disorders (151) and another for axis II personality disorders (152). The interview starts with a set of overview questions, after which it systematically covers each criterion of all included diagnoses. Based on the answers given by the patient, the interviewer makes a clinical rating of each criterion. The scoring categories are: ”?” (inadequate information), 1 (the criterion is absent), 2 (the criterion is doubtful), and 3 (the criterion is present). The reliability of the SCID has been found to be adequate (150, 174).
Psychopathy Checklist-Revised Psychopathy traits were assessed by means of the Psychopathy Checklist-Revised (PCL-R), which is a 20-item, 40-point rating scale. The scoring is based on a semi- structured clinical interview and review of file information (66). Each PCL-R item is rated 0 (definitely absent), 1 (some or contradicting data), 2 (definitely present), or omitted (insufficient information). The score represents the extent to which a person matches the clinical construct of psychopathy.
25 KSP personality traits Personality traits were assessed using the KSP (see above).
Ethics All studies were approved by the Uppsala University Ethics Committee.
RESULTS Personality disorders in former child psychiatric patients (paper I) Former patients had a significantly higher prevalence of all DSM-IV personality disorders than controls (38.0 vs. 10.9 %, p<0.001) (Table 3). There was also a considerably higher personality disorder co-morbidity among the former patients (mean values 1.7 vs. 0.3, p<0.001). Furthermore, former patients had a greater number of fulfilled personality disorder criteria than controls (median 23 vs.11 out of 79 possible criteria, p<0.001). In the former patient group there were also significantly more fulfilled criteria for all specific personality disorders as compared to the control group.
Table 3. Prevalence of personality disorders in former child psychiatric patients as adults and in sex and age-matched controls (N=137 pairs). Former patients Controls n (M/F) % n (M/F) % χ2 p Any Cluster A PD 40 (14/26) 29.2 7 (6/1) 5.1 22.76 ∗∗∗ Any Cluster B PD 43 (13/30) 31.4 7 (5/2) 5.1 26.63 ∗∗∗ Any Cluster C PD 47 (15/32) 34.3 14 (7/7) 10.2 19.32 ∗∗∗ Any PD 52 (17/35) 38.0 15 (7/8) 10.9 22.74 ∗∗∗ Note: PD = personality disorder; M = male, F = female
There were also significantly lower self-reported GAF scores in former patients than in controls (mean values 73.4 vs. 82.3; p<0.001). Moreover, the former patients diagnosed with a personality disorder had lower GAF scores than those former patients without such a diagnosis (mean values 63.3 vs. 79.3, p<0.001). This was also true in
26 the control group (mean values 69.2 vs. 83.5, p<0.01).
Finally, former patients had more psychosocial and environmental problems than controls (mean values 4.1 vs. 1.9, p<0.001). Those former patients also diagnosed with a personality disorder had more psychosocial and environmental problems than former patients without such a diagnosis (mean values 4.7 vs. 3.7, p<0.01). This was also true in controls (mean values 3.5 vs. 1.8, p<0.001). Former patients without a personality disorder also had significantly more psychosocial problems than controls without a personality disorder (mean values 3.7 vs. 1.8, p<0.001).
Predictive value of child and adolescent psychiatric disorders for adult personality disorder (paper II) Out of the 158 former patients, 28 (17.7 %) had a childhood onset of major depression, 75 (47.5 %) a disruptive disorder, and 43 (27.2 %) a substance related disorder. Sixty- two out of the 158 former patients (39.2 %) received a personality disorder diagnosis.
After controlling for demographic variables and other axis I pathology, former patients with a childhood onset of a major depressive disorder were more than ten times as likely to have a schizoid personality disorder than those without (OR=10.4; CI = 2.23- 48.4). Furthermore, those with a major depressive disorder were almost five times as likely to have an avoidant personality disorder (OR=4.71; CI=1.83-12.1). In addition, they were also almost four times as likely to have a dependent personality disorder (OR=3.71; CI=1.22-11.3), and about three times as likely to have a borderline (OR=2.91; CI=1.17-7.25) and a schizotypal personality disorder (OR=2.94; CI=1.10- 7.88). Participants with a disruptive disorder at a young age were almost seven times as likely to have a narcissistic disorder (OR=6.78; CI=1.56-29.4) and more than five times as likely to have an antisocial disorder (OR=5.62, CI=1.96-16.2)). A substance- related disorder in childhood or adolescence was a risk factor for paranoid (OR=2.42; CI=1.04-5.65), schizotypal (OR=3.09; CI=1.25-7.67), borderline (OR=3.00; CI=1.30- 6.92) and avoidant (OR=2.82; CI=1.15-6.92) personality disorders.
27 Personality traits and personality disorders in early onset vs. late onset major depression (paper III) Out of the 400 primary care patients with major depression, 104 (24 men and 80 women) were classified as having an early onset depression with a cut-off at 26 years, and 296 (89 men and 207 women) as having a late onset depression. There were no significant differences in gender or education between the two groups. The severity of current illness was similar as determined by the MADRS instrument (mean scores 28.9 vs. 28.1). The early onset group were younger (mean age 40.1 vs. 49.7, p<0.001), had experienced more depressive episodes (mean values 6.2 vs. 2.6, p<0.001), and had more hospitalisations due to depression (15.4 vs. 4.1 %; p<0.001). Furthermore, this group had a significantly higher proportion of family members with depressive disorders (46.7 vs. 22.0 %; p<0.001).
There were co-existing personality disorders in 249 of the 400 depressed patients (62.3 %). In those with an early onset major depression, 81 (77.9 %) had a personality disorder, and in those with a late onset there were 168 (56.8 %) subjects with such a diagnosis (p<0.001). Avoidant and paranoid personality disorders were the most common axis II disorders in the entire patient population, while borderline personality disorder was also common in those with an early onset, and obsessive-compulsive personality disorder occurred frequently in those with a late onset. Those with an early onset major depression had significantly higher frequencies of all personality disorders except for schizoid, schizotypal and dependent personality disorders. However, after adjustment for age, gender and number of previous depressive disorders, the difference was statistically significant only for the paranoid, borderline and avoidant personality disorders.
There were significant differences between the early onset and the late onset groups with respect to the KSP personality traits. Thus, those with an early onset major depression had higher scores in all anxiety-related scales, i.e. Somatic anxiety, Psychic anxiety, Muscular tension and Psychasthenia. In addition, they scored higher on four out of six aggression and hostility-related scales, i.e. Indirect aggression, Guilt,
28 Suspicion and Irritability. Finally, those with an early onset rated themselves as having a lower level of Socialisation. In a logistic regression with the age and sex-corrected KSP scales dichotomised at the median value, it was demonstrated that age at onset was independently related to three anxiety-related scales, Psychic anxiety, Psychasthenia and Muscular tension, and also to Suspicion, Irritability and Socialisation.
The entire spectrum of analyses was repeated after dichotomisation of the group using 18 years as the cut-off for early onset depression. Here, 28 of the 400 patients were defined as having an early onset. Those with an early onset exhibited more paranoid, borderline, histrionic and avoidant personality disorders, i.e. observations that were very similar to those when the original cut-off level was used. After correction for age, sex and number of previous episodes, early onset was related only to avoidant personality disorder. Early onset was also related to the KSP-scales Psychic anxiety, Psychasthenia, Suspicion, Socialisation, Irritability and Social desirability. After correction for number of previous episodes, a relationship to Psychic anxiety, Psychasthenia, Suspicion, Irritability and Socialisation was found.
Personality traits and personality disorders in early onset vs. late onset of antisocial behaviour (paper IV) Out of the 61 forensic psychiatric male subjects, 27 were diagnosed as having a previous conduct disorder and 34 had not had a previous conduct disorder. Those with a conduct disorder had more cluster B personality disorders than those without (96.3 % vs. 32.4 %, p<0.001). Those with a previous conduct disorder had significantly more repeated violent criminality (78 vs. 21 %; p<0.001). This group of patients also showed significantly more mixed abuse (56 vs. 3 %; p<0.001), but not significantly more pure alcohol abuse or dependence.
Personality traits assessed by means of the KSP revealed that subjects with a previous conduct disorder had higher scores on the scales measuring Impulsiveness, Monotony avoidance, Verbal aggression and Irritability, Suspicion and lower scores on Social
29 desirability and Socialisation. They also had significantly higher scores on the KSP Psychopathy factor and the Aggression factor. Finally, the total PCL-R score was significantly higher in the conduct disorder group as compared to those without such a previous diagnosis (mean values 29.4 vs. 10.2, p<0.001).
The Child and Adolescent Psychiatric Screening Inventory-Retrospect (Paper V) Internal reliability measured by means of Cronbach´s alpha ranged between 0.62 and 0.93 for the different subscales covering the CAPSI-R diagnoses. Autistic disorder demonstrated the lowest alpha correlation coefficient (0.62). After exclusion of one item (”As a child, did you often take part in make-believe plays?”) the alpha correlation coefficient increased to 0.76. This motivated the exclusion of this item from all subsequent analyses.
Data retrieval from the medical records showed that 140 of the 164 former patients fulfilled diagnostic criteria for at least one of the mental disorders covered in the CAPSI-R. Data from the CAPSI-R revealed that a total of 144 of the 164 patients reached the threshold for at least one mental disorder during their childhood or teenage years. One hundred and twenty-six of these subjects also reported functional impairment defined as the need for professional help.
In the control group a total of 96 (49.7 %) of the 193 subjects fulfilled criteria for a psychiatric disorder according to the CAPSI-R. Twenty of these subjects (10.4 %) also experienced functional impairment. Another five individuals reported the need for professional help but did not reach the diagnostic threshold for a CAPSI-R diagnosis.
The sensitivity and the specificity of the CAPSI-R were investigated using diagnoses from the medical records as “true” diagnoses. Determinations were made in two different populations: in the whole group of former patients and in the former patient group who had fulfilled the impairment criterion (Table 4). Sensitivity in the group of all former patients ranged from 33.3% to 83.3 %, and the specificity varied between
30 54.1 % and 95.5 %. In the 126 former patients who had fulfilled the impairment criterion, i.e. contact with health professionals, the sensitivity ranged from 47.4 % to 100 %, and the specificity ranged from 44.6 % to 94.8 %. The overall Kappa value between CAPSI-R diagnoses and those from medical records was 0.79.
Table 4. Sensitivity and specificity of the CAPSI-R. The sensitivity and the specificity were calculated in two separate populations, i.e. in all former patients and in the patients who fulfilled the impairment criterion in the CAPSI-R. Sensitivity and specificity were only calculated if five of more subjects fulfilled diagnostic criteria for a specific disorder according to the medical records. All former patients Former patients who fulfilled the impairment criterion n=164 n=126 DSM-IV Diagnoses Sensitivity Specificity Sensitivity Specificity % % % % Reading dis.1 75.0 81.8 100 80.9 Mathematics dis. 62.5 82.1 – – Dis. of written expression 72.2 76.7 83.3 73.7 Developmental coordination dis 33.3 80.3 47.4 78.5 Expressive language dis. 65.0 90.3 64.7 90.8 Stuttering 50.0 95.5 50.0 94.8 Asperger´s dis. 40.0 83.6 – – ADHD 55.6 76.8 62.5 72.9 Conduct dis. 73.5 60.0 78.6 52.4 Oppositional defiant dis. 50.9 64.0 63.4 56.5 Obsessive compulsive dis. 80.0 81.0 – – Eating dis.2 83.3 90.5 83.3 87.5 Substance dependence 66.7 84.2 70.0 79.8 Substance abuse 62.5 82.8 66.7 81.4 Major depressive episode 82.1 54.1 88.0 44.6 Notes: 1dis. = disorder, 2anorexia nervosa and bulimia nervosa are collapsed into an eating disorder group.
31 DISCUSSION
The present studies constitute an attempt to discern the relationship between child and adolescent psychopathology, above all depression and disruptive disorders, and adult personality pathology determined as deviant personality traits and personality disorders.
Methodological considerations
A common limitation in all the papers is that central assessments were performed in the form of self-reports. Self-reports are frequently used for assessment of a number of psychosocial issues, including personality and personality disorders. They are time saving and easily administered. They are also free from systematic biases and tendencies on the part of interviewers (174). It has, however, been suggested that some individuals with personality disorders may not adequately describe their personality because they do not possess sufficient insight into their own behaviour (175). With respect to the present study, however, the fact that the prevalence of any personality disorder in the controls is similar to the figures previously reported in the general population using a multitude of instruments (47, 161, 165) suggests that the above- mentioned shortcomings play a subordinate role regarding the results.
With respect to child and adolescent psychopathology, the study was retrospective. One issue of special importance here is the possibility of an effect of memory on the reported results. A number of factors influence memory. Developments in cognitive psychology have recently increased our understanding of the ways in which memories are encoded, stored and retrieved. Episodic autobiographic memory is part of declarative memory, i.e. the memory that is accessible for verbalisation. In general, it is this memory that is of prime concern in the recall of childhood experiences. Its function depends on the function, and thus on the development, of the sensory cortices, the limbic system, especially the hippocampal formation, and the functional circuitry
32 between them (110). Studies of adults have found a marked fall-off in the recall of experiences that took place before the age of five (121).
A central issue in an attempt to evaluate an instrument for assessing child and adolescent psychiatric problems is to have a “golden standard” that represents a measure of all possible lifetime mental disorders. This was not possible in the present study, since objective information on psychiatric morbidity was only obtained for those periods the former patients were admitted for psychiatric care. This fact has most likely influenced the figures for the specificity of the CAPSI-R reported in the present study. Another issue is that the “golden standard” should be interpretable in the same context as the instrument. Here, the DSM-IV diagnoses in the former patients were coded from medical records up to almost three decades after treatment, a time period during which diagnostic classification routines have changed, which might have influenced the phenomenological reporting of psychiatric symptoms. Such an effect would possibly have made the diagnostic categories less valid and thereby have influenced the specificity measurement.
A critical question in all studies is the representativity of the sample studied. In the retrospective evaluation of former child and adolescent psychiatric patients, one problem was that the response rate was somewhat low, which means that even moderate differences between the responders and the non-responders may affect the results obtained. The response rate was around 50 percent, with a dropout frequency that was higher among males and more pronounced in the former patients. The estimated effect of such non-representativity on the present results would be that the true frequency of personality disorder among the former patients could be even higher than reported. Thus, other investigators have noted that the prevalence of a disorder is over-represented, if anything, among non-responders (6, 85). It is therefore reasonable to suppose that there was a high frequency of disruptive behaviour among the male dropouts, and that this is associated with a very high frequency of personality disorders as adults (127, 131).
33 Another methodological issue is that there is no agreed-upon definition for early onset major depression. Fava et al. (52) classified early onset as occurrence of the first episode before the age of 18, while Sato et al. (140) used 23 years as a cut-off, and Parker et al. (117) used 26 years. It was recently suggested that a major genetic locus contributes to the expression of early onset major depression, where 26 years was used as a cut-off (101), and this provides certain biological support that this cut-off is appropriate. In any case, the similarity to the results in our study group, with two different cut-off levels, 18 and 26 years, suggests that the observations are possible to generalise. In this perspective, one interpretation is that the underlying processes operate in a dimensional perspective, also slightly after adolescence.
The definite strengths of this study are the length of the follow-up period and the focus on individuals with severe childhood or adolescent psychiatric problems, a group that reasonably offers the best chance for detecting psychopathology retrospectively. Furthermore, the design used was optimised to reveal differences between afflicted individuals and representative controls. The control group was sampled from the population registry in order to match the former patient population with respect to age and sex. The control group exhibited similar figures for personality disorders as those for Swedish controls in previous studies (47), as well as a similar extent of contact with professionals because of symptoms they had experienced (13 vs. 14 percent in a previous study) (170).
Adult personality disorders Personality disorders are common conditions with a major impact on health and social and occupational functioning (139, 148). The prevalence of personality disorders in the former child psychiatric patients was 38 percent (paper I), a somewhat higher figure than the 28 percent prevalence rate reported in former referrals to an adolescent psychiatric unit (128). Moreover, the occurrence of an adult personality disorder was strongly predicted by a former child psychiatric problem (paper II). The increased risk
34 for an adult personality disorder was similar in the present study and in a previous community-based group with a much shorter follow-up (81, 82). This suggests that the predictive values of Axis I disorders at a young age on the occurrence of adult personality disorders are not dependent on length of follow-up.
In the depressed primary care patients the prevalence of personality disorders was 62 percent (paper III), which is in line with the figures reported earlier in depressed samples (for review see (33)). However, the most striking observation was that the depressed individuals with an early onset of their first depressive episode showed significantly more Axis II personality pathology than those with a late onset. This is consistent with previous findings that major depression with an onset at younger ages is related to greater personality pathology in adult ages (52, 81, 82, 97, 117, 140, 149). The early onset group also showed signs of a more debilitating course of their mood disorder, seen in the form of more depressive episodes and previous hospitalisations in spite of their younger age.
Finally, 53 out of the 61 males undergoing forensic psychiatric investigations were diagnosed with a personality disorder (paper IV) (154). Furthermore, in those diagnosed with a previous conduct disorder, personality disturbances were more exaggerated. This was seen in the form of more cluster B personality disorders and a higher psychopathy score. This group also exhibited more repeated violent criminality and had more mixed drug abuse.
Personality traits At present, personality disorders are organised as categories with defined cut-off points, although evidence suggests that whenever the category versus dimension distinction has been subjected to empirical validation, the results almost invariably support a dimensional representation of personality pathology (98). Alternative approaches for conceptualising and classifying personality disorders have therefore been suggested (34, 98, 156). The core of these conceptualisations is that personality disorders can be viewed as more extreme aggregations of different continuous
35 personality traits that have reached such deviance that they are associated with an impact on global functioning. Viewed in this way, assessment of such personality traits would aid our understanding of the transition from normality to disordered personality.
In the present study the KSP instrument was used to assess personality traits. The KSP was constructed for the purpose of operationalising and measuring constructs defining vulnerability to different forms of psychopathology. Most of the KSP scales, i.e. the anxiety-related and the aggressiveness-related scales and the Socialisation scale, are correlated with the separate personality disorders in the DSM (43, 63, 155).
Major depression and personality characteristics Both individuals with early and with late onset major depression (paper III) could be characterised by elevated scores in several of the KSP scales. This is analogous to previous results suggesting that individuals with a lifetime history of depression can be characterised by certain personality traits, e.g. high levels of neuroticism and introversion (74, 118, 145, 172). However, a more striking result of the present study was that individuals who experienced their first depressive episode at young ages exhibited more deviant personality traits than those with a late onset. These results can be explained in a number of ways. Early onset major depression may lead to maladaptive interpersonal functioning that eventually results in the development of personality disturbances. Alternatively, personality dysfunction (or maladaptiveness) could cause or contribute to early onset major depression and an increased risk for recurrent depressive episodes. A third, balanced explanation may be that both developmental pathways reciprocally affect one another, or else operate separately from one another.
Psychobiological differences between patients with early onset and late onset major depression have been suggested. Thus, genetic segregation analysis has recently provided evidence for a single major locus of genetic transmission of susceptibility for early-onset major depression (101). The concept of affective instability has been given a role as a predisposing factor. Affective instability can be defined as a predisposition
36 to marked, rapidly reversible shifts in affective state (146). Such shifts are extremely sensitive to meaningful environmental events such as separation, frustration regarding expectations or criticism, which in other people might induce more modest emotional responses. Such sensitivity might explain why life events in adolescence predispose to adult major depression (123). Affective instability may also interfere with the development of stable relationships and self-image (146). Furthermore, there is some evidence that the impact of experiencing a depressive episode on personality might be stronger in young individuals who are in the process of establishing their self-identity in society (132). This is supported by prospective studies suggesting that after recovery, young persons may continue to show sub-clinical symptoms of depression, negative attributions, impairment in interpersonal relationships, impairment in global functioning, early pregnancy and increased physical problems (80, 111, 124, 126, 132). Depression has also been associated with a negative cognitive style including low self-esteem, high self-criticism, significant cognitive distortions and a feeling of lack of control over negative events (12).
The observation that major depressive disorder is a risk factor for the development of schizoid and schizotypal personality disorders (paper II) can be explained in the context of two earlier observations. First, there is an association between cluster A personality disorders and schizophrenic disorders (113). Second, a certain proportion of individuals with depression in childhood and adolescence will later develop a schizophrenic disorder (36). Taken together, these observations suggest that for a small subgroup of children and adolescents there may be a developmental transition from a major depression-like picture into the adult schizophrenic spectrum. A similar finding has in fact been reported in a community sample, where participants with a lifetime history of major depression were almost four times as likely to have an elevated rate of schizoid dimensional scores (97). However, in contrast to our study, Kasen et al. (82) did not observe major depression to be a risk factor for the development of schizoid and schizotypal personality disorders. The most probable explanation for the discrepancy between these studies is that the group investigated by
37 Kasen and colleagues was a community-based sample, while the present results were based on a group of former severely affected psychiatric in-patients.
Substance use disorders and personality characteristics Psychiatric research workers have studied the reasons for addiction to alcohol over a long period of time. There is a general consensus that addicted persons exhibit an overrepresentation of criteria for both clinical and personality disorders. As general diagnostic criteria for personality disorders include a requirement that the enduring pattern is not due to the direct physiological effects of a substance, e.g. a drug abuse (10), it is difficult to make assumptions about the relationship between drug abuse and personality. However, there is clear evidence that certain personality features are present at an early stage in those prone to alcohol and drug abuse (26, 27, 31, 76, 166). Thus, in the present investigation individuals diagnosed with a substance-related disorder at an early age were found to be at risk for paranoid, schizotypal, borderline and avoidant personality disorders (papers I and II). Furthermore, evidence from cross- sectional studies is convincing, and shows a strong link between substance-related disorders and all forms of personality disorders (39, 65), particularly antisocial and borderline personality disorders (163). In the study by Lewinsohn et al. (97), substance use disorders in adolescence were significantly associated with antisocial personality disorder dimensional scores when re-assessed at age 24. In the present study, however, there was no association between substance-related disorders and antisocial personality disorder. This might be explained by a high attrition rate in males.
Disruptive disorder and personality characteristics Former patients diagnosed with a disruptive disorder were observed to have an increased risk for antisocial and narcissistic personality disorders in adulthood (papers I and II). This observation was also made by Kasen et al. (82) and Lewinsohn et al. (97), and is also supported by earlier observations of well-established associations between disruptive disorders and cluster B personality disorders (131, 176). Although there are characteristics that differentiate antisocial and narcissistic personality
38 disorders, there is also a considerable overlap, particularly for careless behaviour and interpersonal relationships that denote indifference to or exploitation of others.
Individuals afflicted with a previous conduct disorder exhibited significantly higher levels of Impulsiveness (paper IV). This is suggested to represent a specific vulnerability to disruptive behaviours and criminality, with a strong hereditary factor (3, 4). In fact, impulsivity is a major predictor for antisocial, delinquent behaviour (50, 162, 169). Impulsivity has also been related to different kinds of mental morbidity such as borderline personality disorder and type II alcoholism (146, 171). Moreover, highly impulsive individuals exhibit impaired neuropsychological functions determined in the form of disturbed sorting perceptions and an inability to show foresight and to inhibit their behaviour (3, 40).
An intriguing observation (paper IV) was that the forensic patients were characterised by a very low Socialisation score, considerably lower than the scores of those with a depressive disorder (paper III). A striking feature was that this was significantly more pronounced in those with a previous conduct disorder, who on average exhibited a socialisation score more than three standard deviations lower than that in a general population control group. The KSP Socialisation scale reflects dimensions of alienation, victimisation, resentment and dysphoria, and was originally constructed to assess psychopathic personality features (63). Later studies, however, suggest that a low score in Socialisation not only characterises psychopathic personality but is also strongly and broadly associated with the clinical severity of personality pathology and maladaptiveness (43, 155). As the scale has very little genetic origin (63), it is probable that environmental factors during development of socialised behaviour play a dominant role with respect to this personality trait. Here, socially oriented theories emphasise interpersonal interactions and cognitive processes in the development of socialised behaviour, and assume that socialisation is dependent on appropriate role- taking experiences (60). In this context, the impact of experiencing psychopathology
39 on personality might be stronger in young individuals who are in the process of establishing their sense of self-identity in society (132).
The CAPSI-R
The development of the CAPSI-R (paper V) was initiated by the fact that there are no comprehensive instruments currently available for retrospective self-report assessment of child and adolescent psychopathology in adults. Such an instrument could serve as a research tool, but also as a screening instrument in adult patients presenting for a first time psychiatric consultation.
The internal reliability, as determined by means of Cronbach´s alpha, was satisfactory for all diagnoses included in the CAPSI-R, and the sensitivity was generally acceptable to good. The lifetime prevalence rate of any particular DSM-IV psychiatric disorder in the former patients was very similar to the rate obtained from coded medical records. The corresponding figure for a CAPSI-R diagnosis was close to 50 percent in the control group, but dropped to 10 percent after incorporation of the impairment criterion defined as the need for professional help. Others have made the same observation (16, 96, 143). Despite the fact that the validation procedure was hampered by methodological shortcomings, sensitivity and specificity were satisfactory, at least in individuals who met the impairment criterion. The CAPSI-R therefore shows promise for further evaluation.
40 CONCLUSIONS
The progression of psychopathology from childhood to adulthood may be influenced by a multitude of intervening variables, both biological and psychosocial. Increased knowledge concerning intervening variables is necessary in order to get a better understanding of the connections between child and adult psychopathology. The results of the present investigations suggest that
• personality disorders are prevalent in former child psychiatric patients, • the presence of psychiatric disorders in childhood and adolescence significantly increases the risk of an adult personality disorder, • there is an association between early onset major depression and adult personality pathology, • early onset antisocial behaviour is related to a more deviant personality, more repeated violent criminality and mixed abuse, • the CAPSI-R shows promise for further evaluation, and it may be useful in recognising previous child and adolescent mental disorders in adults.
41 ACKNOWLEDGEMENTS
This thesis was completed with the help and support of supervisors, colleagues and friends. I would like to express my sincere gratitude to:
Lisa Ekselius, Associate Professor, Department of Neuroscience, Psychiatry, University Hospital, Uppsala, my main supervisor, for all the hard work and uncountable hours you have spent on this thesis and on me. You have provided me with scientific guidance, engaging collaboration and warm and generous friendship. Thank You!!
Anne-Liis von Knorring, Professor, Department of Neuroscience, Child and Adolescent Psychiatry, University Hospital, Uppsala, my supervisor, for firmly establishing my work and scientific education in the field of Child and Adolescent Psychiatry. Thank you for sharing your professional knowledge with me and for encouraging me to start this project.
Lars von Knorring, Professor, Department of Neuroscience, Psychiatry, University Hospital, Uppsala, my supervisor, for enthusiasm and great support. You have provided me with solutions to all the practical and financial problems that have arisen. I am also grateful that as part of my scientific education, you introduced me to and involved me in teaching.
Gunilla Stålenheim, M.D., senior colleague at the Clinical Department of Psychiatry, University Hospital, Uppsala. Thank you for inviting me to collaborate with you and for sharing your data with me. You gave me a much useful advice during the time we wrote two scientific papers together.
Hans Arinell, B.Sc., Department of Neuroscience, Child and Adolescent Psychiatry, University Hospital, Uppsala, for teaching me statistics and how to work with computers. You have provided me with continual support regarding all statistical and technical problems, and you have been an excellent teacher.
Lena Bohlin, Department of Neuroscience, Psychiatry, University Hospital, Uppsala. Thanks for efficient secretarial help and for supportive friendship.
42 Ulises Penayo, M.D., Head of the Clinical Department of Psychiatry at Uppsala University Hospital, for providing me with financial support in order to complete this thesis.
Jane Wigertz, for linguistic revision.
Mimmie Willebrand and Johan Dyster-Aas, my fellow Ph.D. students, for scientific guidance and interesting discussions.
Gösta Lyttkens and Lars-Olof Janols, my clinical tutors, for introducing me to child and adolescent psychiatry and for your great contributions to my clinical training.
Goldy and Lars Ramklint, my parents, for never-ending support in all areas of life, including this scientific work. My mother Goldy, for helping me create the database, and my father Lars, for linguistic revision of several manuscripts.
Anders Holmberg, my husband and life companion, for giving me the idea to become a Ph.D. student, and for providing me with the encouragement and practical support that I needed to complete this thesis. But most of all for all the warm and generous love you give me every day.
Staffan, Håkan, Sara, Calle and Olle, our beloved children, for being part of my life, and with respect to this thesis for your positive attitude towards my Ph.D. work.
Gunilla and Nils Trowald, neighbours and close friends, for taking part in my Ph.D. project as friends, for listening, asking questions, discussing and, finally, planning the party.
And last, but not least, all the participants in these studies, who lent us their time and shared with us their thoughts and feelings. Without them no knowledge would have been gained and no studies would have been possible.
Generous financial support was given by the Swedish Medical Research Council (grants no. 9523, 12260 and 13245), the Söderström-Königska Foundation and the Märta and Nicke Nasvell Foundation.
43 REFERENCES 1. Aarkrog T. Psychotic adolescents 20-25 years later. Nord J Psychiatry 1999; 53, Suppl 42: 1-38. 2. Achenbach TM, Howell CT, McConaughy SH, Stanger C. Six-year predictors of problems in a national sample: III. Transitions to young adult syndromes. J Am Acad Child Adolesc Psychiatry 1995; 34: 658-669. 3. af Klinteberg B. The psychopathic personality in a longitudinal perspective. Eur Child Adolesc Psychiatry 1996; 5, Suppl. 1: 57-63. 4. af Klinteberg B, Magnusson D, Schalling D. Hyperactive behavior in childhood and adult impulsivity: A longitudinal study of male subjects. Person Individ Diff 1989; 10: 43-50. 5. Akiskal HS, Hirschfeld RM, Yerevanian BI. The relationship of personality to affective disorders. Arch Gen Psychiatry 1983; 40: 801-810. 6. Allgulander C. Psychoactive drug use in a general population sample, Sweden: correlates with perceived health, psychiatric diagnoses, and mortality in an automated record-linkage study. Am J Public Health 1989; 79: 1006-1010. 7. Allport GW. Personality: A Psychological Interpretation. Vol. XIV. 1937, New York: Holt. 588 pages. 8. Alnaes R, Torgerson S. DSM-III symptom disorders (Axis I) and personality disorders (Axis II) in an outpatient population. Acta Psychiatr Scand 1988; 78: 348-355. 9. Angold A, Costello EJ, Erkanli A. Comorbidity. J Child Psychol Psychiat 1999; 40: 57-87. 10. APA. Diagnostic and Statistical manual of Mental Disorders, fourth edition, DSM-IV. 1994, Washington DC: American Psychiatric Association. 11. Asberg M, Montgomery SA, Perris C, Schalling D, Sedvall G. A comprehensive psychopathological rating scale. Acta Psychiatr Scand Suppl 1978; 271: 5-27. 12. Beck AT, Brown G, Steer RA, Eidelson JI, Riskind JH. Differentiating anxiety and depression: a test of the cognitive content- specificity hypothesis. J Abnorm Psychol 1987; 96: 179-183. 13. Bergman H, Bergman I, Engelbrektsson K, Holm L, Johannesson K, Lindberg S. Psykologhandboken vid Magnus Husskliniken. Swedish ed. 1982, Stockholm, Sweden: Karolinska Hospital. 14. Bernstein DP, Cohen P, Skodol A, Bezirganian S, Brook JS. Childhood antecedents of adolescent personality disorder. Am J Psychiatry 1996; 153: 907-913. 15. Bernstein DP, Fink L, Handelsman L, Foote J, Lovejoy M, Wenzel MA, Sapareto E, Ruggiero J. Initial reliability and validity of a new retrospecttive measure of child abuse and neglect. Am J Psychiatry 1994; 151: 1132-1136. 16. Bird HR, Canino G, Rubio-Stipec M, Gould MS, Ribera J, Sesman M, Woodbury M, Huertas-Goldman S, Pagan A, Sanchez-Lacay A, Moscoso M. Estimates of the prevalence of childhood maladjustment in a community survey in Puerto Rico. Arch Gen Psychiatry 1988; 45: 1120-1126.
44 17. Bird HR, Gould MS, Staghezza BM. Patterns of diagnostic comorbidity in a community sample of children aged 9 through 16 years. J Am Acad Child Adolesc Psychiatr 1993; 32: 361-368. 18. Birmaher B, Ryan ND, Williamson DE, Brent DA, Kaufman J, Dahl RE, Perel J, Nelson B. Childhood and adolescent depression: a review of the past 10 years. Part I. J Am Acad Child Adolesc Psychiatr 1996; 35: 1427-1439. 19. Bodlund O, Ekselius L, Lindström E. Personality traits and disorders among psychiatric outpatients and normal subjects on the basis of the SCID screen questionnaire. Nord J Psychiatry 1993; 47: 425-433. 20. Bodlund O, Kullgren G, Ekselius L, Lindström E, von Knorring L. Axis V- Global Assessment of Functioning scale: evaluation of a self-report version. Acta Psychiatr Scand 1994; 90: 342-347. 21. Boyle MH, Offord DR, Racine YA, Szatmari P, Sanford M, Fleming JE. Adequacy of interviews vs checklists for classifying childhood psychiatric disorder based on parents reports. Arch Gen Psychiatry 1997; 54: 793-799. 22. Brewin CR, Andrews B, Gotlib IH. Psychopathology and early experience: a reappraisal of retrospective reports. Psychol Bull 1993; 113: 82-98. 23. Buss AH, Plomin R. A Temperament Theory of Personality development. Vol. 255. 1975, New York: John Wiley. 24. Buss AH, Plomin R. The EAS approach to temperament. The study of temperament; Changes, Continuities and Challenges., Plomin R, Dunn J. Editors. 1986, Lawrence Erlbaum: Hillsdale, NJ. 25. Buss DM, Craik KH. Why not measure that trait? Alternative criteria for identifying important dispositions. J Pers Soc Psychol 1985; 48: 934-946. 26. Cadoret RJ, Yates WR, Troughton E, Woodworth G, Stewart MA. Adoption study demonstrating two genetic pathways to drug abuse. Arch gen psychiatry 1995; 52: 42-52. 27. Cadoret RJ, Yates WR, Troughton E, Woodworth G, Stewart MA. An adoption study of drug abuse/dependency in females. Compr Psychiatry 1996; 37: 88-94. 28. Caspi A, Moffitt TE, Newman DL, Silva PA. Behavioral observations at age 3 years predict adult psychiatric disorders. Longitudinal evidence from a birth cohort. Arch Gen Psychiatry 1996; 53: 1033-1039. 29. Christian RE, Frick PJ, Hill NL, Tyler L, Frazer DR. Psychopathy and conduct problems in children: II. Implications for subtyping children with conduct problems. J Am Acad Child Adolesc Psychiatry 1997; 36: 233-241. 30. Cleckley H. The mask of sanity. 5 ed. 1976, St Louis, MO: Mosby. 31. Cloninger CR, Sigvardsson S, Bohman M. Childhood personality predicts alcohol abuse in young adults. Alcohol Clin Exp Res 1988; 12: 494-505. 32. Cohen RJ, Swerdlik ME. Psychological testing and assessment: an introduction to tests and measurement. 4 th ed. 1998, Mountain View, California: Mayfield Publishing Company. 146-214 . 33. Corruble E, Ginestet D, Guelfi JD. Comorbidity of personality disorders and unipolar major depression: a review. J Affect Disord 1996; 37: 157-170. 34. Costa PT, McCrae RR. Personality disorders and the five factor models of personality. J Pers Disord 1990; 4: 362-371.
45 35. Costello EJ, Angold A, Keeler GP. Adolescent outcomes of childhood disorders: the consequences of severity and impairment. J Am Acad Child Adolesc Psychiatry 1999; 38: 121-128. 36. Crow TJ, Done DJ, Sacker A. Childhood precursors of psychosis as clues to its evolutionary origins. Eur Arch Psychiatr Clin Neurosci 1995; 245: 61-69. 37. Cui XJ, Vaillant GE. Antecedents and consequences of negative life events in adulthood: a longitudinal study. Am J Psychiatry 1996; 153: 21-26. 38. de Chateau P. Mortality and aggressiveness in a 30-year follow-up study in child guidance clinics in Stockholm. Acta Psychiatr Scand 1990; 81: 472-476. 39. DeJong CA, van den Brink W, Harteveld FM, van der Wielen EG. Personality disorders in alcoholics and drug addicts. Compr Psychiatry 1993; 34: 87-94. 40. Dolan M, Deakin WJ, Roberts N, Anderson I. Serotonergic and cognitive impairment in impulsive aggressive personality disordered offenders: are there implications for treatment? Psychol Med 2002; 32: 105-117. 41. Dolan-Sewell RT, Krueger RF, Shea MT. Co-occurrence with syndrome disorders. In Handbook of personality disorders, Livesley WJ. Editor. 2001, The Guilford Press: New York. pp. 84-104. 42. Dunn J. Individual differences in temperament. In Scientific Foundations of Developmental Psychiatry, Rutter M. Editor. 1980, Heineman Medical: London. 43. Ekselius L, Hetta J, von Knorring L. Relationship between personality traits as determined by means of the Karolinska Scales of Personality (KSP) and Personality Disorders according to DSM-III-R. Person Individ Diff 1994; 16: 589-595. 44. Ekselius L, Lindström E, von Knorring L, Bodlund O, Kullgren G. Personality disorders in DSM-III-R as categorical or dimensional. Acta Psychiatr Scand 1993; 88: 183-187. 45. Ekselius L, Lindström E, von Knorring L, Bodlund O, Kullgren G. Comorbidity among the personality disorders in the DSM-III-R. Person Individ Diff 1994; 17: 155-160. 46. Ekselius L, Lindström E, von Knorring L, Bodlund O, Kullgren G. SCID II interviews and the SCID Screen questionnaire as diagnostic tools for personality disorders in DSM-III-R. Acta Psychiatr Scand 1994; 90: 120-123. 47. Ekselius L, Tillfors M, Furmark T, Fredriksson M. Personality disorders in the general population: DSM-IV and ICD-10 defined prevalence as related to sociodemographic profile. Person Individ Diff 2001; 30: 311-320. 48. Ekselius L, von Knorring L, Eberhard G. A double-blind multicenter trial comparing sertraline and citalopram in patients with major depression treated in general practice. Int Clin Psychopharmacol 1997; 12: 323-331. 49. Eysenck HJ. A model for personality. 1981, Berlin: Springer-Verlag. 50. Farrington DP. The Twelfth Jack Tizard Memorial Lecture. The development of offending and antisocial behaviour from childhood: key findings from the Cambridge Study in Delinquent Development. J Child Psychol Psychiatry 1995; 36: 929-964.
46 51. Farrington DP, Hawkins JD. Predicting participation, early onset and later persistence in officially recorded offending. Crim Behav Ment Health 1991; 1: 1-33. 52. Fava M, Alpert JE, Borus JS, Nierenberg AA, Pava JA, Rosenbaum JF. Patterns of personality disorder comorbidity in early-onset versus late-onset major depression. Am J Psychiatry 1996; 153: 1308-1312. 53. Feehan M, McGee R, Williams SM, Nada-Raja S. Models of adolescent psychopathology: childhood risk and the transition to adulthood. J Am Acad Child Adolesc Psychiatry 1995; 34: 670-679. 54. Fendrich M, Weissman MM, Warner V. Longitudinal assessment of major depression and anxiety disorders in children. J Am Acad Child Adolesc Psychiatry 1991; 30: 38-42. 55. Ferdinand RF, Verhulst FC. Psychopathology from adolescence into young adulthood: an 8-year follow- up study. Am J Psychiatry 1995; 152: 1586-1594. 56. Fergusson DM, Horwood LJ, Lynskey M. The childhoods of multiple problem adolescents: a 15-year longitudinal study. J Child Psychol Psychiatry 1994; 35: 1123-1140. 57. Frick PJ, O'Brien BS, Wootton JM, McBurnett K. Psychopathy and conduct problems in children. J Abnorm Psychol 1994; 103: 700-707. 58. Goldacre M, Hawton K. Repetition of self-poisoning and subsequent death in adolescents who take overdoses. Br J Psychiatry 1985; 146: 395-398. 59. Goldsmith HH, Buss AH, Plomin R, Rothbart MK, Thomas A, Chess S, Hinde RA, McCall RB. Roundtable: what is temperament? Four approaches. Child Dev 1987; 58: 505-529. 60. Gough HG. Theory and measurement of socialization. J Consult Psychol 1960; 24: 23-30. 61. Gray JA. The neuropsychology of anxiety. 1982, London: Oxford University Press. 62. Gunderson JG. Diagnostic controversies. In Review of psychiatry, Tasman A, Riba MB. Editors. 1992, APA: Washington DC. pp. 9-24. 63. Gustavsson JP. Stability and validity of self reported personality traits. Contributions to the evaluation of the Karolinskas Scales of Personality (PhD thesis). 1997, Karolinska Institutet: Stockholm. 64. Gustavsson JP, Weinryb RM, Göransson S, Pedersen NL, Åsberg M. Stability and predictive ability of personality traits across 9 years. Person Individ Diff 1997; 22: 783-791. 65. Hallman J, von Knorring L, Oreland L. Personality disorders according to DSM-III-R and thrombocyte monoamine oxidase activity in type 1 and type 2 alcoholics. J Stud Alcohol 1996; 57: 155-161. 66. Hare RD. The Hare Psychopathy Checklist-Revised. 1991, Toronto, Ontario, Canada: Multi-Health Systems. 67. Hare RD, Hart SD, Harpur TJ. Psychopathy and the DSM-IV criteria for antisocial personality disorder. J Abnorm Psychol 1991; 100: 391-398. 68. Harpur TJ, Hare RD, Hakstian AR. Two-factor conceptualization of psychopathy: Construct validity and assessment implications. Psychol Assess 1989; 2: 6-17.
47 69. Harrington R, Fudge H, Rutter M, Pickles A, Hill J. Adult outcomes of childhood and adolescent depression. Arch General Psychiatr 1990; 47: 465- 473. 70. Harris EC, Barraclough B. Excess mortality of mental disorder. Br J Psychiatry 1998; 173: 11-83. 71. Hart SD, Kropp PR, Hare RD. Performance of male psychopaths following conditional release from prison. J Consult Clin Psychol 1988; 56: 227-232. 72. Henry B, Moffitt TE, Caspi A, Langley J, Silva PA. On the "remembrance of things past": A longitudinal evaluation of the retrospective method. Psychol Assess 1994; 34: 49-68. 73. Hinshaw SP. Conduct disorder in childhood: conceptualization, diagnosis, comorbidity and risk status for antisocial functioning in adulthood. In Experimental personality and psychopathology research, Fowles DC, Sutker P,Goodman SH. Editors. 1994, Springer: New York. pp. 3-44. 74. Hirschfeld RM, Klerman GL, Clayton PJ, Keller MB. Personality and depression. Empirical findings. Arch Gen Psychiatry 1983; 40: 993-998. 75. Hodges K. Structured interviews for assessing children. J Child Psychol Psychiatry 1993; 34: 49-68. 76. Iacono WG, Carlson SR, Taylor J, Elkins IJ, McGue M. Behavioral disinhibition and the development of substance-use disorders: findings from the Minnesota Twin Family Study. Dev Psychopathol 1999; 11: 869-900. 77. Jablensky A. The nature of psychiatric classification: issues beyond ICD-10 and DSM- IV. Aust N Z J Psychiatry 1999; 33: 137-144. 78. Jackson HJ, Whiteside HL, Bates GW, Bell R, Rudd RP, Edwards J. Diagnosing personality disorders in psychiatric inpatients. Acta Psychiatr Scand 1991; 83: 206-213. 79. Jensen PS, Brooks-Gunn J, Graber JA. Dimensional scales and diagnostic categories: constructing crosswalks for child psychopathology assessments. J Am Acad Child Adolesc Psychiatry 1999; 38: 118-120. 80. Kandel DB, Davies M. Adult sequelae of adolescent depressive symptoms. Arch Gen Psychiatry 1986; 43: 255-262. 81. Kasen S, Cohen P, Skodol AE, Johnson JG, Brook JS. Influence of child and adolescent psychiatric disorders on young adult personality disorder. Am J Psychiatry 1999; 156: 1529-1535. 82. Kasen S, Cohen P, Skodol AE, Johnson JG, Smailes E, Brook JS. Childhood depression and adult personality disorder: alternative pathways of continuity. Arch Gen Psychiatry 2001; 58: 231-236. 83. Kendler KS. Diagnostic approaches to schizotypal personality disorder: a historical perspective. Schizophrenia Bull 1985; 11: 538-553. 84. Kenrick DT, Stringfield DO. Personality traits and the eye of the beholder: Crossing some traditional philosophical boundaries in the search for consistency in all of the people. Psychol Rev 1980; 87: 88-104. 85. Kessler RC, McGonagle KA, Zhao S, Nelson CB, Hughes M, Eshleman S, Wittchen HU, Kendler KS. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Arch Gen Psychiatry 1994; 51: 8-19.
48 86. Kjelsberg E. A long-term follow-up study of adolescent psychiatric in-patients. Part III. Predictors of disability. Acta Psychiatr Scand 1999; 99: 243-246. 87. Kjelsberg E, Dahl AA. High delinquency, disability and mortality-a register study of former adolescent psychiatric in-patients. Acta Psychiatr Scand 1998; 98: 34-40. 88. Kjelsberg E, Dahl AA. A long-term follow-up study of adolescent psychiatric in-patients. Part II. Predictors of delinquency. Acta Psychiatr Scand 1999; 99: 237-242. 89. Kjelsberg E, Sandvik L, Dahl AA. A long term follow-up study of adolescent psychiatric in-patients. Part I. Predictors of early death. Acta Psychiatr Scand 1999; 99: 231-236. 90. Kovacs M. Presentation and course of major depressive disorder during childhood and later years of the life span. J Am Acad Child Adolesc Psychiatry 1996; 35: 705-715. 91. Kraepelin E. Psychiatrie. 5th ed. 1896, Leipzig: Barth. 92. Kullgren G. Personality disorders among psychiatric in-patients. Nordic J Psychiatry 1992; 46: 27-32. 93. Kuperman S, Black DW, Burns TL. Excess suicide among formerly hospitalized child psychiatry patients. J Clin Psychiatry 1988; 49: 88-93. 94. Lahey BB, Loeber R, Quay HC, Frick PJ, Grimm J. Oppositional defiant and conduct disorders: issues to be resolved for DSM-IV. J Am Acad Child Adolesc Psychiatry 1992; 31: 539-546. 95. Larsen FW, Dahl V, Hallum E. A 30-year follow-up study of a child psychiatric clientele. I. Demographic description. Acta Psychiatr Scand 1990; 81: 39-45. 96. Lewinsohn PM, Hops H, Roberts RE, Seeley JR, Andrews JA. Adolescent psychopathology:I. Prevalence and incidence of depression and other DSM-III- R disorders in high school students. J Abnorm Psychol 1993; 102: 133-144. 97. Lewinsohn PM, Rohde P, Seeley JR, Klein DN. Axis II psychopathology as a function of Axis I disorders in childhood and adolescence. J Am Acad Child Adolesc Psychiatry 1997; 36: 1752-1759. 98. Livesley WJ. Handbook of personality disorders. 2001, New York: The Guilford Press. 84-104 99. Lynham DR. Early identification of chronic offenders: who is the fledging psychopath? Psychol Bull 1996; 120: 209-234. 100. Magnusson D. Individual development from an interactional perspective: a longitudinal study. 1988, Hillsdale: Erlbaum, N.J. 101. Maher BS, Marazita ML, Zubenko WN, Spiker DG, Giles DE, Kaplan BB, Zubenko GS. Genetic segregation analysis of recurrent, early-onset major depression: Evidence for single major locus transmission. Am J Med Genet 2002; 114: 214-221. 102. Maier W, Lichtermann D, Klingler T, Heun R. Prevalences of personality disorders (DSM-III-R) in the community. J Pers Disord 1992; 6: 187-196. 103. Mattison RE, Hooper SR. The history of modern classification of child and adolescent psychiatric disorders: an overview. In Assessment and Diagnosis of Child and Adolescent Psychiatric Disorders. Volume I: Psychiatric Disorders,
49 Hooper SR, Hynd GW,Mattison RE. Editors. 1992, Lawrence Erlbaum Publishers: Hillsdale, N.J. 104. Maughan B, Rutter M. Retrospective reporting of childhood adversity: issues in assessing long-term recall. J Pers Disord 1997; 11: 19-33. 105. McCann BS, Scheele L, Ward N, Roy-Byrne P. Discriminant validity of the Wender Utah Rating Scale for attention- deficit/hyperactivity disorder in adults. J Neuropsychiatr Clin Neurosci 2000; 12: 240-245. 106. McGee R, Feehan M, Williams S, Anderson J. DSM-III disorders from age 11 to age 15 years. J Am Acad Child Adolesc Psychiatry 1992; 29: 611-629. 107. McGee R, Feehan M, Williams S, Patridge F, Silva PA, Kelly J. DSM-III disorders in a large sample of adolescents. J Am Acad Child Adolesc Psychiatry 1990; 29: 611-619. 108. Millon T. Disorders of personality: DSM-IV and beyond. 2 nd ed. 1996, New York: John Wiley & Sons, Inc. 17-19. 109. Montgomery SA, Asberg M. A new depression scale designed to be sensitive to change. Br J Psychiatry 1979; 134: 382-389. 110. Nelson CA. The nature of early memory. Prev Med 1998; 27: 172-179. 111. Nolen-Hoeksema S, Girgus JS, Seligman ME. Predictors and consequences of childhood depressive symptoms: a 5-year longitudinal study. J Abnorm Psychol 1992; 101: 405-422. 112. Offord DR, Boyle MH, Racine YA, Fleming JE, Cadman DT, Blum HM, Byrne C, Links PS, Lipman EL, MacMillan HL, et al. Outcome, prognosis, and risk in a longitudinal follow-up study. J Am Acad Child Adolesc Psychiatry 1992; 31: 916-923. 113. Oldham JM, Skodol AE, Kellman HD, Hyler SE, Doidge N, Rosnick L, Gallaher PE. Comorbidity of axis I and axis II disorders. Am J Psychiatry 1995; 152: 571-578. 114. Ottosson H, Bodlund O, Ekselius L, Grann M, von Knorring L, Kullgren G, Lindström E, Söderberg S. DSM-IV and ICD-10 personality disorders: a comparison of a self report questionnaire (DIP-Q) with a structured interview. Eur Psychiatry 1998; 13: 246-253. 115. Ottosson H, Bodlund O, Ekselius L, von Knorring L, Kullgren G, Lindström E, Söderberg S. The DSM-IV and ICD-10 personality questionnaire (DIP-Q): Constructions and preliminary validation. Nord J Psychiatry 1995; 49: 285- 291. 116. Ottosson H, Ekselius L, Grann M, Kullgren G. Cross-system concordance of personality disorder diagnoses of DSM-IV and diagnostic criteria for research of ICD-10. J Pers Disord 2002; in press. 117. Parker G, Wilhelm K, Mitchell P, Austin MP, Roussos J, Gladstone G. The influence of anxiety as a risk to early onset major depression. J Affect Disord 1999; 52: 11-17. 118. Pendse B, Westrin A, Engstrom G. Temperament traits in seasonal affective disorder, suicide attempters with non-seasonal major depression and healthy controls. J Affect Disord 1999; 54: 55-65.
50 119. Pepper CM, Klein DN, Anderson RL, Riso LP, Ouimette PC, Lizardi H. DSM- III-R axis II comorbidity in dysthymia and major depression. Am J Psychiatry 1995; 152: 239-247. 120. Pilkonis PA, Frank E. Personality pathology in recurrent depression; nature, prevalence, and relationship to treatment respons. Am J Psychiatry 1988; 145: 435-441. 121. Pillemer DB, White SH. Childhood events recalled by children and adults. Adv Child Dev Behav 1989; 21: 297-340. 122. Pine DS, Cohen P, Gurley D, Brook J, Ma Y. The risk for early-adulthood anxiety and depressive disorders in adolescents with anxiety and depressive disorders. Arch Gen Psychiatry 1998; 55: 56-64. 123. Pine DS, Cohen P, Johnson JG, Brook JS. Adolescent life events as predictors of adult depression. J Affect Disord 2002; 68: 49-57. 124. Puig-Antich J, Lukens E, Davies M, Goetz D, Brennan-Quattrock J, Todak G. Psychosocial functioning in prepubertal major depressive disorders. II. Interpersonal relationships after sustained recovery from affective episode. Arch Gen Psychiatry 1985; 42: 511-517. 125. Ramklint M, Stålenheim EG, von Knorring A-L, von Knorring L. Triiodothyronine (T3) related to conduct disorder in a forensic psychiatric population. Eur J Psychiat 2000; 14: 33-41. 126. Rao U, Ryan ND, Birmaher B, Dahl RE, Williamson DE, Kaufman J, Rao R, Nelson B. Unipolar depression in adolescents: clinical outcome in adulthood. J Am Acad Child Adolesc Psychiatry 1995; 34: 566-578. 127. Rasmussen P, Gillberg C. Natural outcome of ADHD with developmental coordination disorder at age 22 years: a controlled, longitudinal, community- based study. J Am Acad Child Adolesc Psychiatry 2000; 39: 1424-1431. 128. Rey JM, Morris-Yates A, Singh M, Andrews G, Stewart GW. Continuities between psychiatric disorders in adolescents and personality disorder in young adults. Am J Psychiatry 1995; 152: 895-900. 129. Roberts RE, Attkisson CC, Rosenblatt A. Prevalence of psychopathology among children and adolescents. Am J Psychiatry 1998; 155: 717-725. 130. Robins LN. Deviant children grown up. 1966, Baltimore: William and Wilkins. 131. Robins LN, Price RK. Adult disorders predicted by childhood conduct problems: results from the NIMH epidemiologic catchment area project. Psychiatry 1991; 54: 116-132. 132. Rodhe P, Lewinsohn PM, Seeley JR. Are adolescents changed by an episode of major depression? J Am Acad Child Adolesc Psychiatry 1994; 33,9: 1289- 1298. 133. Rutherford MJ, Cacciola JS, Alterman AI, McKay JR. Use of the Psychopathy Checklist-Revised with women methadone patients. Assessment 1996; 3: 43-54. 134. Rutter M. Psychopathology and development: II. Childhood experiences and personality development. Aust N Z J Psychiatry 1984; 18: 314-327. 135. Rutter M. Resilience in the face of adversity. Protective factors and resistance to psychiatric disorder. Br J Psychiatry 1985; 147: 598-611. 136. Rutter M. Pathways from childhood to adult life. J Child Psychol Psychiat 1989; 30: 23-51.
51 137. Rutter M. Child psychiatric disorder. Measures, causal mechanisms, and interventions. Arch Gen Psychiatry 1997; 54: 785-789. 138. Rydelius PA. The development of antisocial behaviour and sudden violent death. Acta Psychiatr Scand 1988; 77: 398-403. 139. Samuels JF, Nestadt G, Romanoski AJ, Folstein MF, McHugh PR. DSM-III personality disorders in the community. Am J Psychiatry 1994; 151: 1055- 1062. 140. Sato T, Sakado K, Uehara T, Narita T, Hirano S. Personality disorder comorbidity in early-onset versus late-onset major depression in Japan. J Nerv Ment Dis 1999; 187: 237-242. 141. Schalling D. Contributions to the validation of some personality concepts (Ph D thesis). 1970, Department of Psychology, Stockholm University: Stockholm, Sweden. 142. Schalling D, Asberg M, Edman G, Oreland L. Markers for vulnerability to psychopathology: temperament traits associated with platelet MAO activity. Acta Psychiatr Scand 1987; 76: 172-182. 143. Shaffer D, Fisher P, Dulcan MK, Davies M, Piacentini J, Schwab-Stone ME, Lahey BB, Bourdon K, Jensen PS, Bird HR, Canino G, Regier DA. 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 1996; 35: 865-877. 144. Shaffer D, Schwab-Stone M, Fisher P, Cohen P, Piacentini J, Davies M, Conners CK, Regier D. The Diagnostic Interview Schedule for Children- Revised version (DISC-R): preparation, field testing, interrater reliability and acceptability. J Am Acad Child Adolesc Psychiatry 1993; 32: 643-650. 145. Shea MT, Leon AC, Mueller TI, Solomon DA, Warshaw MG, Keller MB. Does major depression result in lasting personality change? Am J Psychiatry 1996; 153: 1404-1410. 146. Siever LJ, Davis KL. A psychobiological perspective on the personality disorders. Am J Psychiatry 1991; 148: 1647-1658. 147. Sjöbring H. Personality structure and development. Acta Psychiatr Scand 1973; Suppl. 244. 148. Skodol AE, Gunderson JG, McGlashan TH, Dyck IR, Stout RL, Bender DS, Grilo CM, Shea MT, Zanarini MC, Morey LC, Sanislow CA, Oldham JM. Functional impairment in patients with schizotypal, borderline, avoidant, or obsessive-compulsive personality disorder. Am J Psychiatry 2002; 159: 276- 283. 149. Skodol AE, Stout RL, McGlashan TH, Grilo CM, Gunderson JG, Shea MT, Morey LC, Zanarini MC, Dyck IR, Oldham JM. Co-occurrence of mood and personality disorders: a report from the Collaborative Longitudinal Personality Disorders Study (CLPS). Depress Anxiety 1999; 10: 175-182. 150. Skre I, Onstad S, Torgersen S, Kringlen E. High interrater reliability for the Structured Clinical Interview for DSM-III-R Axis I (SCID-I). Acta Psychiatr Scand 1991; 84: 167-173.
52 151. Spitzer RL, Williams JBW, Gibbon M, First MB. Structured Clinical Interview for DSM-III-R, Patient Edition/Non-patient Edition,(SCID-P/SCID-NP). 1990, Washington, D.C.: American Psychiatric Press, Inc. 152. Spitzer RL, Williams JBW, Gibbon M, First MB. Structured Clinican Interview for DSM-III-R Axis II Disorders,(SCID-II),. 1990, Washington, D.C.: American Psychiatric Press, Inc. 153. Storm-Mathiesen A, Vaglum P. Conduct disorder patients 20 years later: a personal follow-up study. Acta Psychiatr Scand 1994; 89: 416-420. 154. Stålenheim EG, von Knorring L. Psychopathy and Axis I and Axis II psychiatric disorders in a forensic psychiatric population in Sweden. Acta Psychiatr Scand 1996; 94: 217-223. 155. Svanborg P, Gustavsson PJ, Mattila-Evenden M, Asberg M. Assessment of maladaptiveness: a core issue in the diagnosing of personality disorders. J Pers Disord 1999; 13: 241-256. 156. Svrakic DM, Whitehead C, Przybeck TR, Cloninger CR. Differential diagnosis of personality disorders by the seven-factor model of temperament and character. Arch Gen Psychiatry 1993; 50: 991-999. 157. Szatmari IP, Boyle M, Offord DR. ADHD and conduct disorder: degree of diagnostic overlap and differences among correlates. J Am Acad Child Adolesc Psychiatry 1989; 28: 865-872. 158. Thomas A, Chess S. Temperament and Development. 1977, New York: Brunner/Mazel. 159. Thomas A, Chess S. Temperament and follow-up to adulthood. Ciba Found Symp 1982; 89: 168-175. 160. Thomsen PH. A 22- to 25-year follow-up study of former child psychiatric patients: a register-based investigation of the course of psychiatric disorder and mortality in 546 Danish child psychiatric patients. Acta Psychiatr Scand 1996; 94: 397-403. 161. Torgersen S, Kringlen E, Cramer V. The prevalence of personality disorders in a community sample. Arch Gen Psychiatry 2001; 58: 590-596. 162. Tremblay RE, Pihl RO, Vitaro F, Dobkin PL. Predicting early onset of male antisocial behavior from preschool behavior. Arch Gen Psychiatry 1994; 51: 732-739. 163. Tyrer P, Gunderson J, Lyons M, Tohen M. Extent of comorbidity between mental state and personality disorders. J Pers Disord 1997; 11: 242-259. 164. Ward MF, Wender PH, Reimherr FW. The Wender Utah Rating Scale: an aid in the retrospective diagnosis of childhood attention deficit hyperactivity disorder. Am J Psychiatry 1993; 150: 885-890. 165. Weissman MM. The epidemiology of personality disorders: A 1990 Update. J Pers Disord 1993; 7: 44-62. 166. Wennberg P, Bohman M. Childhood temperament and adult alcohol habits: a prospective longitudinal study from age 4 to age 36. Addict Behav 2002; 27: 63-74. 167. WHO. ICD-10: The ICD-10 Classification of Mental and Behavioural Disorders: Clinical Descriptions and Diagnostic Guidelines. 1992, Geneva: World Health Organization.
53 168. Widiger TA, Trull TJ. Personality and psychopathology: an application of the five-factor model. J Pers 1992; 60: 363-393. 169. Vitacco MJ, Rogers R. Predictors of adolescent psychopathy: the role of impulsivity, hyperactivity, and sensation seeking. J Am Acad Psychiatry Law 2001; 29: 374-382. 170. von Knorring A-L, Andersson O, Magusson D. Psychiatric care and course of psychiatric disorders from childhood to early adulthood in a representative sample. J Child Psychol Psychiat 1987; 28: 329-341. 171. von Knorring AL, Bohman M, von Knorring L, Oreland L. Platelet MAO activity as a biological marker in subgroups of alcoholism. Acta Psychiatr Scand 1985; 72: 51-58. 172. von Knorring L, Perris C, Eisemann M, Perris H. Discrimination of former depressed patients from healthy volunteers on the basis of stable personality traits assessed by means of KSP. Eur Arch Psychiatry Neurol Sci 1984; 234: 202-205. 173. Zelkowitz P, Paris J, Guzder J, Feldman R. Diatheses and stressors in borderline pathology of childhood: the role of neuropsychological risk and trauma. J Am Acad Child Adolesc Psychiatry 2001; 40: 100-105. 174. Zimmerman M. Diagnosing personality disorders. A review of issues and research methods. Arch Gen Psychiatry 1994; 51: 225-245. 175. Zimmerman M, Pfohl B, Stangl D, Corenthal C. Assessment of DSM-III personality disorders: the importance of interviewing an informant. J Clin Psychiatry 1986; 47: 261-263. 176. Zoccolillo M, Pickles A, Quinton D, Rutter M. The outcome of childhood conduct disorder; implications for defining adult personality disorder and conduct disorder. Psychol Med 1992; 22: 971-986. 177. Östman O. Child and Adolescent psychiatric patients in adulthood. Acta Psychiatr Scand 1991; 84: 40-45.
54