This is a preprint of an article published in Journal of the American Academy of Child and Adolescent Psychiatry 2001 American Academy

y of Child and Adolescent Psychiatry. Woodward LJ, Fergusson DM. Life course outcomes of young people with anxiety disorders in adolescence. Journal of the American Academy of Child and Adolescent Psychiatry, 2001; 40(9): 1086-1093. Life Course Outcomes of Young People with Anxiety Disorders in Adolescence

Lianne Woodward Ph.D David M Fergusson Ph.D

Dr Woodward is with the University of Canterbury, Christchurch and Professor Fergusson is the Director of the Christchurch Health and Development Study, Christchurch School of Medicine, New Zealand.

Correspondence to:

Professor David Fergusson, Christchurch Health and Development Study, Christchurch School of Medicine, P O Box 4345, Christchurch, New Zealand

Telephone: Fax: Email:

64-3-372 0406 64-3-372 0405 david.fergusson@chmeds.ac.nz

Running head: Statistical experts: Word count:

Outcomes of anxious adolescents Authors 5,983

2 Acknowledgements: This research was funded by grants from the Health Research Council of New Zealand, the National Child Health Research Foundation, the Canterbury Medical Research Foundation and the New Zealand Lottery Grants Board.

ABSTRACT

Objective: This study examined associations between the extent of anxiety disorder in adolescence (14-16 years) and young peoples later risks of a range of mental health, educational and social role outcomes (16-21 years). Method: Data were gathered over the course of a 21 year longitudinal study of a birth cohort of 1,265 New Zealand children. Measures collected included: a) an assessment of DSM-III-R anxiety disorders between the ages of 14 and 16 years; b) assessments of mental health, educational achievement and social functioning between the ages of 16 and 21 years; and c) measures of potentially confounding social, family and individual factors. Results: Significant linear associations were found between the number of anxiety disorders reported in adolescence and later risks of anxiety disorder, major depression, nicotine, alcohol and illicit drug dependence, suicidal behavior, educational under-achievement and early parenthood. Associations between the extent of adolescent anxiety disorder and later risks of anxiety disorder, depression, illicit drug dependence and failure to attend university were shown to persist following statistical control for the confounding effects of socio-familial and individual factors. Conclusions: Findings suggest that adolescents with anxiety disorders are at an increased risk of subsequent anxiety, depression, illicit drug dependence and educational under-achievement as young adults. Clinical and research implications are considered.

Keywords:

Anxiety disorder, adolescence, psychiatric disorder, longitudinal study.

4 Life Course Outcomes of Young People with Anxiety Disorders in Adolescence

Anxiety disorders represent one of the least well understood mental health problems of childhood and adolescence (Costello and Angold, 1995; Zahn-Waxler et al., 2000). However, there is now growing evidence to suggest that anxiety disorders are not uncommon in children and adolescents, with the prevalence of any anxiety disorder ranging from 5.7 to 17.7% (Cohen et al., 1993; Costello and Angold, 1995; Costello et al., 1996; Fergusson et al., 1993). For example, McGee et al. (1990) found that 10.7% of 15 year olds met DSM-III criteria for an anxiety disorder, with the rate of specific disorders varying from 6% for overanxious disorder to between 1 and 2% for social phobia and separation anxiety disorder. Despite these improvements in the recognition and diagnosis of anxiety disorders in young people, the longer term outcomes associated with adolescent onset anxiety remain poorly understood. Available evidence provides a somewhat conflicting view of the longer term prognosis of children and adolescents with anxiety disorders, with some studies suggesting that they grow up to become relatively well adjusted young adults (Last et al., 1997; 1996), and others suggesting that they may be at increased risk of later mental health problems, particularly with respect to anxiety and depression (Berg et al., 1989; Feehan et al., 1993; Ferdinand and Verhulst, 1995; Flament et al., 1990; Keller et al., 1992; Pine et al., 1998). Although these findings generally suggest that adolescents with anxiety disorders may be at risk of ongoing internalizing mental health problems, it is important to note that much of this research is characterized by a number of methodological limitations. First, many existing follow-up studies have been based on samples of clinic referred children, thus limiting the generalizability of study findings (Cantwell and Baker, 1989; Flament et al., 1990; Last et al., 1996; Leonard et al., 1993). Second, there has been a tendency for studies to focus on the effects of a specific anxiety disorder such as panic disorder or obsessive compulsive disorder (Cantwell and Baker, 1989;

5 Flament et al., 1990), without considering the effects of other comorbid anxiety disorders. Third, research concerned with the life course outcomes of anxious youngsters has typically been concerned with mental health outcomes, with virtually no consideration given to other important aspects of social functioning such as educational attainment, occupational functioning and the timing of parenthood. An issue that has been largely ignored by existing research relates to the extent to which associations between adolescent anxiety disorders and later outcomes reflect the direct effect of anxiety on later outcomes, and/or the effects of third or confounding factors correlated with both adolescent anxiety and later outcomes. Whilst it is possible that high levels of anxiety during childhood and adolescence may place young people at risk of later mental health problems and constrained life opportunities, it is also possible that the higher rates of subsequent adversity evident amongst anxious adolescents may reflect the effects of comorbid psychiatric symptomatology or family risk factors that both encourage the development of anxiety and later life course adversity. There is some support for this latter possibility from studies showing high rates of comorbidity between anxiety disorders and other psychiatric disorders (Curry and Murphy, 1995), in addition to research linking family adversity to the development of anxiety in children and adolescents (Lieb et al., 2000; March, 1995). Against this general background, the present study reports the results of a 21-year longitudinal study of the relationship between adolescent anxiety disorder and later mental health, educational and social functioning in a birth cohort of 964 New Zealand children. The specific aims of the study were: 1. To document associations between the extent of anxiety disorder in adolescence (14-16 years) and a range of later mental health, educational and social functioning outcomes measured in late adolescence and early adulthood (16-21 years).

6 2. To examine whether associations between the extent of anxiety disorder and later outcomes persisted after statistical adjustment for a range of confounding social background, family and individual factors.

METHODS Subjects Participants were members of an unselected birth cohort that has been extensively studied as part of the Christchurch Health and Development Study (CHDS). The CHDS is a longitudinal study of a birth cohort of 1,265 children (635 males; 630 females) born in the Christchurch, New Zealand, urban region during mid 1977. These young people have been studied at birth, 4 months, 1 year, annual intervals to age 16, and again at 18 and 21 years. The analyses reported in this paper were based on a sample of 964 young people for whom complete data were available on the measures of anxiety disorder and later outcomes. This sample represented 76.2% of the initial birth cohort. Losses to follow up arose because of outmigration from New Zealand (50%), inability to trace (3%), subject refusal (37%) and mortality (10%). To examine the effects of sample loss on the representativeness of the sample, comparisons were made between the 964 young people included in the analyses and the excluded 301 cohort members on a range of social background measures collected at birth. This analysis suggested that losses to follow-up were not associated with maternal age, family size or child gender. However there were small but statistically detectable (p<.05) tendencies for this sample to underrepresent children from Maori, single parent and lower socioeconomic status families. Although these results suggest some bias in the sample, it is unlikely that this bias will materially influence the present results, since previous efforts to correct for nonrandom sample loss in this cohort have shown these effects to be negligible (Fergusson et al., 1988; Fergusson and Lloyd, 1991).

7 Measures Anxiety Disorders (14-16 years) At ages 15 and 16, sample members and their parents were interviewed separately about the extent to which, during the previous year, each young person showed symptoms of a range of anxiety disorders, including generalized anxiety disorder, separation anxiety disorder, over-anxious disorder, social phobia, agoraphobia, panic disorder and simple phobia. Fifteen years was the earliest age at which anxiety disorders were assessed in the cohort. Self reported anxiety disorders were assessed using the Diagnostic Interview Schedule for Children (Costello et al., 1982), supplemented by additional items based on DSM-III-R diagnostic criteria (American Psychiatric Association, 1987). Parent reported disorders were assessed using the parent version of the Diagnostic Interview Schedule for Children, in addition to items from the Diagnostic Interview Schedule (Robins et al., 1981). Sample members were classified as having experienced any of the above disorders if on the basis of either parent or self report at ages 15 or 16, they met DSM-III-R criteria for each disorder. A total of 11% met DSM-III-R diagnostic criteria for generalized anxiety, 4.7% for overanxious disorder, 20.6% for simple phobia, 2.9% for social phobia, 1.7% for separation anxiety disorder and 1.5% for agoraphobia-panic disorder. Since seven anxiety disorders were assessed, there were potentially 64 different patterns of anxiety disorders that children could exhibit during this period. In addition, the low base rates of several disorders made it impossible to assess the effects of these specific disorders on later outcomes. To avoid difficulties associated with the complexity of disorder patterns and the low base rate for some disorders, a count measure of each respondents exposure to anxiety disorder in adolescence was created by summing the total number of different anxiety disorders reported for each respondent between the ages of 14 and 16 years. Overall, 29.9% of the sample met criteria for at least one anxiety disorder, 9.0% met criteria for two or more disorders and 2.7% for three or more anxiety disorders.

8 Mental Health, Educational and Social Role Outcomes (16-21 years) Mental Health (16-21 years). Sample members were questioned at ages 18 and 21 years about their mental health and substance use since the previous interview using a measure based on the Composite International Diagnostic Interview (World Health Organization, 1993), the Self Report Delinquency Inventory (Elliott and Huizinga, 1989) and custom written survey items. From this assessment, DSM-IV (American Psychiatric Association, 1994) symptom criteria were used to construct the following psychiatric and substance abuse diagnoses for each respondent. Major depression. Symptoms of major depression were assessed using CIDI items administered at ages 18 and 21. At age 18, subjects were asked to report on symptoms occurring during the periods from 16-17 and 17-18 years. At age 21, subjects were asked to report on symptoms occurring for the periods: 18-20 and 20-21 years. Using this information, assessments of major depression using DSM-IV criteria were made between the ages of 16 and 21 years. Anxiety disorders. Anxiety disorder symptoms were assessed using the CIDI at ages 18 and 21 years. Subjects were asked to report whether or not they had experienced a range of anxiety disorder symptoms since the last assessment. Anxiety disorders assessed included: generalized anxiety; panic disorders; agoraphobia; social phobia; and specific phobia. Subjects were classified as having an anxiety disorder if they met DSM-IV criteria for an anxiety disorder at either of the two assessments. Nicotine dependence. Symptoms of nicotine dependence were assessed using custom written items designed to reflect DSM-IV diagnostic criteria for nicotine dependence. These items were assessed for the intervals, 17-18 and 20-21 years. Alcohol abuse/dependence. Alcohol abuse/dependence was assessed for annual intervals between the ages of 16 and 21 years using items from the CIDI. Subjects were classified as showing alcohol dependence if they reported experiencing at least three of the following: increased

9 tolerance for alcohol; withdrawal symptoms when alcohol was ceased; heavy drinking and overuse of alcohol; unsuccessful attempts to quit or reduce drinking; spending large amounts of time in alcohol related activities; restriction of social and other activities as a result of drinking; and psychological problems due to by heavy and prolonged drinking. Illicit drug dependence. Using similar criteria to those for alcohol dependence, subjects were questioned about their dependence on other substances, including cannabis, hallucinogens, opiates, and other illicit drugs. In addition to these measures, sample members were questioned about any suicidal behavior between the ages of 16 and 21 years. Educational Outcomes Two measures of tertiary educational involvement were used to describe sample members educational achievement from 16 to 21 years. These included whether or not they had enrolled in a) a trade or skill based tertiary training course and/or b) a university or equivalent level program, by age 21. Social Role Outcomes (21 years) Two measures of social role functioning were identified. First, a measure of exposure to extended unemployment (>3 months) was created from subject reports of the frequency and duration of all periods of unemployment (excluding full time education) between the ages of 16 and 21 years. Second, a measure of respondents parenting status was developed on the basis of all reported child births between the ages of 16 and 21 years. The youngest subject to become a parent was aged 16 years. Confounding Factors To assess the extent to which associations between adolescent anxiety and later outcomes could be explained by the effects of confounding factors, the following variables were included as covariates in the analysis.

10 Social Background Four measures of family social background were considered. The first two measures were maternal age and maternal education at the time of the respondents birth. Maternal education was coded on a three point scale ranging from (1) no formal educational qualifications to (3) tertiary level qualifications. The third measure, family socioeconomic status at birth was assessed using the Elley and Irving (1976) scale of socioeconomic status for New Zealand. This scale categorizes families into six classes according to paternal occupation. Finally, a composite measure of family living standards was created by averaging interviewers annual ratings of living conditions from birth to 10 years. Family Functioning Five measures of family functioning were included. The first measure assessed the extent to which young people were exposed to parental change from birth to 14 years due to parental separation/divorce, death, remarriage and reconciliation (Fergusson et al., 1992). The second and

third measures assessed respondents exposure to regular/severe physical punishment and sexual abuse during childhood. At age 18, sample members were interviewed about the extent to which their mother and father had used physical punishment to discipline them as children (birth to age 16). These ratings were combined into a composite four-point scale based on the highest level of reported exposure to physical punishment (Fergusson and Lynskey, 1997). This scale ranged from (1) parent never used physical punishment to (4) at least one parent used physical punishment too often or too severely. The measure of sexual abuse was based on subject reports of their experience of childhood sexual abuse (CSA) prior to age 16 (Fergusson et al., 1996). Subjects were classified into four groups ranging from (1) no CSA reported to (4) subject experienced CSA involving completed or attempted oral, anal, or vaginal intercourse. The fourth measure consisted of whether or not either parent reported a prior history of alcoholism or alcohol problems at the 15 year assessment. Finally, young peoples attachment to their parents at age 15 years was assessed using

11 the parental attachment scale developed by Armsden and Greenberg (1987). The internal consistency of this scale assessed using coefficient alpha was .87. Individual Factors Seven measures of individual functioning were included. The first four measures were concerned with young peoples gender, psychiatric adjustment and substance abuse between the ages of 14 and 16 years. At ages 15 and 16 years, young people and their parents were questioned separately about the young persons mental health using items from the Diagnostic Interview Schedule for Children (Costello et al., 1982), the Self Report Early Delinquency Inventory (Moffitt and Silva, 1988); the Rutgers Alcohol Problems Index (White and Labouvie, 1989) and custom written items based on DSM-III-R criteria. Using this information, respondents were classified as being subject to for major depression, conduct disorder and alcohol abuse, if on the basis of either parent or self report, they met DSM-III-R diagnostic criteria for these disorders. The remaining three measures were as follows. At age 14, neuroticism was assessed using a short form of the Eysenck Personality Inventory (Eysenck and Eysenck, 1964). The reliability of this scale assessed using coefficient alpha was .80. At age 9, childrens intellectual ability was assessed using the revised Wechsler Intelligence Scale for Children (WISC) (Wechsler, 1974). The reliability of this scale assessed using split half methods was .93. Finally, at ages 15 and 16, young people were questioned about the extent to which their best friend and other friends were involved in a range of behaviors including the use of tobacco, alcohol or other substances, criminal offending and related behaviors. These items were then summed to provide an overall measure of the extent to which each subject affiliated with delinquent and/or substance using peers from 14 to 16 years.

12

RESULTS

Life Course Outcomes of Anxious Adolescents Table 1 shows the sample of 964 young people studied to the age of 21 years divided into four groups according to the extent of anxiety disorder in middle adolescence. These groups included: 1) those individuals who did not meet DSM-III-R diagnostic criteria for an anxiety disorder (n = 676); 2) those who met criteria for one anxiety disorder (n = 201); 3) those who met criteria for two anxiety disorders (n = 61); and 4) those who met DSM-III-R criteria for three or more anxiety disorders between the ages of 14 and 16 years (n = 26). For each group, the table shows the mental health, educational and social role outcomes by age 21. The strength of each association was tested using the Mantel Haenszel chi squared test of linearity. Significant linear associations were found between the number of anxiety disorders reported in adolescence and a range of adverse outcomes in early adulthood. Specifically, as the number of anxiety disorders increased there was a corresponding increase in young peoples subsequent risk of a range of mental health problems, including anxiety disorders (p<.0001), major depression (p<.0001), nicotine dependence (p<.05), alcohol dependence (p<.01), illicit drug dependence (p<.001) and suicidal behavior (p<.05). An increasing number of anxiety disorders was also associated with a reduced likelihood of entering university (p<.0001) or other training course (p<.0001), in addition to an increased likelihood of early parenthood (p<.0001). No association was found between the number of anxiety disorders reported and later unemployment risk. To test for gender differences in the relationships between the extent of adolescent anxiety disorder and later outcomes, the sample was stratified by gender and tests of gender by anxiety disorder interactions conducted for each of the significant outcomes shown in Table 1. No

13 significant (p<.05) interactions were found between gender and the number of anxiety disorders across all outcomes.

Social, Family, Individual, and Life Event Factors Associated with Anxiety Disorders in Adolescence (14-16 years) Although the results in Table 1 suggest the presence of clear linear associations between the number of anxiety disorders in adolescence and a range of adverse outcomes, it is possible that these associations may reflect the effects of confounding factors correlated with both adolescent anxiety and adult adversity. This issue is explored in Table 2 which shows the psychosocial profiles of adolescents with varying numbers of anxiety disorders. For ease of data display, all covariate measures have been presented in dichotomous form, with associations tested using the Mantel Haenszel chi squared test of linearity. This table shows that the extent of anxiety disorder was associated with a range of sociofamilial and individual factors. Specifically, young people with higher rates of anxiety disorders were more likely to: 1. Come from socially disadvantaged family backgrounds characterized by early motherhood (p<.01), maternal educational underachievement (p<.05), lower socioeconomic status (p<.0001) and below average living standards (p<.0001). 2. Have been raised in families characterized by parental change (p<.0001), regular/severe physical punishment (p<.0001), exposure to contact sexual abuse (p<.0001), parental alcohol problems (p<.0001) and weaker parent-child attachment (p<.0001). 3. Be female (p<.0001) and have higher rates of comorbid depression (p<.001), conduct disorder (p<.001) alcohol abuse (p<.0001), and higher rates of neuroticism (p<.0001) and deviant peer involvement in adolescence (p<.0001). There was also a tendency for those with higher rates of anxiety disorders to have obtained lower IQ scores at age 9 (p<.10).

14

Relationships Between Adolescent Anxiety and Later Outcomes Adjusted for Confounding Factors To examine the extent to which associations between the number of anxiety disorders in adolescence and later outcomes could be explained by the effects of confounding factors correlated with anxiety, the associations shown in Table 1 were adjusted for the effects of the factors listed in Table 2. This involved fitting a series of logistic regression models in which each outcome was regressed on: a) the extent of anxiety disorder between the ages of 14 and 16 years; and b) the significant factors shown in Table 2. The findings of this analysis are presented in Table 3 which shows the relationship between adolescent anxiety disturbance and later outcomes after adjustment for significant confounding factors. The results in Table 3 show that following control for confounding factors, associations between the extent of adolescent anxiety disorder and subsequent risks of nicotine dependence, alcohol dependence, suicide attempt and becoming a parent were reduced to statistical nonsignificance. This suggests that the higher rates of these outcomes found amongst highly anxious teenagers were due to the effects of other risk factors correlated with anxiety, rather than the direct effects of adolescent anxiety disorder on later functioning. However, even after control for confounding, significant associations remained between the extent of anxiety disorder in adolescence and young peoples later risks of anxiety disorder (p<.0001), major depression (p<.0001) and illicit drug dependence (p<.05). Young people with three or more anxiety disorders in middle adolescence had rates of subsequent anxiety disorders that were 3.5 times higher, and rates of depression that were double the rate of their non-anxious peers. Similarly, adolescents with three or more anxiety disorders had rates of later illicit drug dependence that were almost four times higher than their non-anxious peers. Finally, significant associations were also found between the extent of anxiety disorder and later tertiary educational participation (p<.10), with non-anxious

15 adolescents being 1.4 times more likely to enter a tertiary training course and 2.5 times more likely to go to university than adolescents with three or more anxiety disorders.

DISCUSSION

In this paper we have used prospective longitudinal data to examine the mental health, educational and social role outcomes of adolescents with varying severity of anxiety disorders. The results extend our empirical understanding of the longer term effects of adolescent anxiety disorders in several important ways. First, the use of a large representative sample helped to avoid the difficulties associated with referral bias and poor generalizability. Second, the assessment of a range of anxiety disorders over the course of adolescence allowed us to examine the impact of both single and multiple anxiety disorders on later functioning. Finally, the availability of a comprehensive database of information about the social and personal backgrounds of cohort members provided an opportunity to examine the extent to which associations between adolescent anxiety and later outcomes reflected the direct effects of anxiety and/or the effects of other factors correlated with both anxiety and later outcomes. In agreement with existing research (Berg et al., 1989; Feehan et al., 1993; Ferdinand and Verhulst, 1995; Flament et al., 1990; Keller et al., 1992; Pine et al., 1998), the present findings suggest that anxiety disordered adolescents have elevated rates of anxiety and depression as young adults. In addition there was also evidence to suggest that adolescent anxiety was associated with a number of other adverse mental health and life course outcomes. Specifically, as the number of anxiety disorders reported during adolescence increased, rates of later nicotine, alcohol and illicit drug dependence, suicidal behavior, educational under-achievement and early parenthood also increased.

16 However, it was also found that adolescents with anxiety disorders were characterized by a number of socio-familial and personal disadvantages that might also account for the difficulties they faced as young adults. After control for the effects of these confounding factors, relations between the number of anxiety disorders in adolescence and later risks of suicidal behavior, nicotine dependence, alcohol dependence and early parenthood were reduced to statistical nonsignificance. These findings tend to suggest that the elevated rates of these outcomes found amongst adolescents with anxiety disorders were a consequence of the risk factors and life processes associated with anxiety rather than the direct effects of early onset anxiety on later life course development. However, even after controlling for a wide range of social, family and personal factors, clear evidence was still found to support the presence of continuities in the later life course outcomes associated with adolescent anxiety disorder. Even after taking into account the effects of confounding factors, significant associations remained between the number of anxiety disorders reported in adolescence and young peoples later risks of anxiety disorder, major depression, illicit drug dependence and failure to attend university. Limitations While these findings clearly suggest that adolescents with anxiety disorders represent an atrisk group for both subsequent mental health problems and educational under-achievement, a number of study limitations should be acknowledged. First, the results are based on a specific cohort studied in New Zealand and it therefore remains open to examination whether similar findings will apply to other cohorts in other social contexts. Second, the assessment of anxiety and other psychiatric disorders was based on retrospective interview methods which may have resulted in some imprecision in the reporting of the severity and timing of disorder symptoms. This feature may explain the fairly high rates of disorder reported over the period from 16 to 21 years. Third, no information was available about anxiety disorder symptoms prior to age 14 years. Therefore, it was

17 not possible to examine the effects of very early onset anxiety disorders on later outcomes, including adolescent anxiety. Clinical Implications These limitations not withstanding, current findings raise concerns about the likely impact of adolescent anxiety disorders on later life course development. There is clearly a need to develop more effective methods of identifying and treating anxious adolescents. One strategy that may be helpful could be the development of school based screening and counselling programs which address the needs of anxious children and adolescents by means of referral to specialist services, in addition to the provision of ongoing educational support and monitoring. This would seem especially relevant with respect to adolescents who have persistent and/or comorbid anxiety disorders. Finally, an important challenge for future research in this area will be to develop a better understanding of the underlying process or processes by which anxiety disorders increase an individuals risk of developing other mental health problems such as depression and illicit drug dependence, and which also deter them from seeking higher educational qualifications.

18 REFERENCES

American Psychiatric Association (1987), Diagnostic and Statistical Manual of Mental Disorders (3rd ed. - rev). Washington, DC: American Psychiatric Association American Psychiatric Association (1994), Diagnostic and Statistical Manual of Mental Disorders (4th ed). Washington: American Psychiatric Association Armsden GC, Greenberg MT (1987), The inventory of parent and peer attachment: Individual differences and their relationship to psychological well-being in adolescence. J Youth Adolesc 16:427-454 Berg CZ, Rapoport JL, Whitaker A, Davies M, Leonard H, Swedo SE, Braiman S, Lenane M (1989), Childhood obsessive compulsive disorder: A two-year prospective study of a community sample. J Am Acad Child Adolesc Psychiatry 28:528-533 Cantwell DP, Baker L (1989), Stability and natural history of DSM-III childhood diagnoses. J Am Acad Child Adolesc Psychiatry 28:691-700 Cohen P, Cohen J, Kasen S, Velez C, Hartmark C, Johnson J, Rojas M, Brook J, Streuning EL (1993), An epidemiological study of disorders in late childhood and adolescence I. Age-and gender-specific prevalence. J Child Psychol Psychiatry 34:851-867 Costello A, Edelbrock C, Kalas R, Kessler M, Klaric SA (1982), Diagnostic Interview Schedule for Children (DISC). Bethesda, MD: National Institute of Mental Health Costello EJ, Angold A (1995), Epidemiology. In J. S. March (Ed.), Anxiety Disorders in Children and Adolescents . New York: Guilford Press Costello EJ, Angold A, Burns BJ, Stangl DK, Tweed DL, Erkanli A, Worthman CM (1996), The Great Smoky Mountains study of youth: Goals, design, methods, and the prevalence of DSM-III-R disorders. Arch Gen Psychiatry 53:1129-1136

19 Curry JF, Murphy LB (1995), Comorbidity of anxiety disorders. New York: Guilford

Elley WB, Irving JC (1976), Revised socio-economic index for New Zealand. NZ J of Ed Studies 11:25-36 Elliott DS, Huizinga D (1989), Improving self-reported measures of delinquency. In M. W. Klein (Ed.), Cross-National Research in Self-Reported Crime and Delinquency (pp. 155-186). Dordrecht: Kluwer Academic Publisher Eysenck HM, Eysenck SBG (1964), Manual of the Eysenck Personality Inventory. London: London University Press Feehan M, McGee R, Williams SM (1993), Mental health disorders from age 15 to age 18 years. J Am Acad Child Adolesc Psychiatry 32:1118-1126 Ferdinand RF, Verhulst FC (1995), Psychopathology from adolescence into young adulthood: An 8 year follow-up study. Am J Psychiatry 152:1586-1594 Fergusson DM, Fergusson JE, Horwood LJ, Kinzett NG (1988), A longitudinal study of dentine lead levels, intelligence, school performance and behaviour. Part III. Dentine lead levels and attention/activity. J Child Psychol Psychiatry 29:811-824 Fergusson DM, Horwood LJ, Lynskey MT (1992), Family change, parental discord and early offending. J Child Psychol Psychiatry 33:1059-1075 Fergusson DM, Horwood LJ, Lynskey MT (1993), Prevalence and comorbidity of DSM-III-R diagnoses in a birth cohort of 15 year olds. J Am Acad Child Psychiatry 32:1127-1134 Fergusson DM, Lloyd M (1991), Smoking during pregnancy and its effects on child cognitive ability from the ages of 8 to 12 years. Paediatr & Perinatal Epidemiol 5:189-200 Fergusson DM, Lynskey MT (1997), Physical punishment/maltreatment during childhood and adjustment in young adulthood. Child Abuse Negl 21:617-630

20 Fergusson DM, Lynskey MT, Horwood LJ (1996), Childhood sexual abuse and psychiatric disorder in young adulthood: I. Prevalence of sexual abuse and factors associated with sexual abuse. J Am Acad Child Psychiatry 35:1355-1364

Flament MF, Koby E, Rapoport JL, Berg CJ, Zahn T, Cox C, Denckla M, Lenane M (1990), Childhood obsessive-compulsive disorder: A prospective follow-up study. J Child Psychol Psychiatry 31:363-380 Keller MB, Lavori PW, Wunder J, Beardslee WR, Schwartz CE, Roth J (1992), Chronic course of anxiety disorders in children and adolescents. J Am Acad Child Adolesc Psychiatry 1992: Last CG, Hansen C, Franco N (1997), Anxious children in adulthood: A prospective study of adjustment. J Am Acad Child Adolesc Psychiatry 36:645-652 Last CG, Perrin S, Hersen M, Kazdin AE (1996), A prospective study of childhood anxiety disorders. J Am Acad Child Adolesc Psychiatry 35:1502-1510 Leonard HL, Swedo SE, Lenane MC, Rettew DC, Hamburger SD, Bartko JJ, Rapoport JL (1993), A two to seven year follow-up study of 54 obsessive compulsive children and adolescents. Arch Gen Psychiatry 50:429-439 Lieb R, Wittchen H-U, Hofler M, Fuetsch M, Stein MB, Merikangas KR (2000), Parental psychopathology, parenting styles, and the risk of social phobia in offspring: A prospectivelongitudinal community study. Arch Gen Psychiatry 57:859-866 March JS (1995), Anxiety Disorders in Children and Adolescents. New York: Guilford McGee R, Feehan M, Williams S, Partridge F, Silva PA, Kelly J (1990), DSM-III disorders in a large sample of adolescents. J Am Acad Child Adolesc Psychiatry 29:611-619 Moffitt TR, Silva PA (1988), Self-reported delinquency: Results from an instrument for New Zealand. Aust NZ J Crim 21:227-240

21 Pine DS, Cohen P, Gurley D, Brook J, Ma Y (1998), The risk for early-adulthood anxiety and depressive disorders in adolescents with anxiety and depressive disorders. Arch Gen Psychiatry 55:56-64 Robins L, Helzer JE, Croughan J, Ratcliff RS (1981), NIMH Diagnostic Interview Schedule: Its history, characteristics and validity. Arch Gen Psychiatry 38:381-389

Wechsler D (1974), Manual for the Wechsler Intelligence Scale for Children - Revised. New York: Psychological Corporation White HL, Labouvie EW (1989), Towards the assessment of adolescent problem drinking. J Stud Alcohol 50:30-37 World Health Organization (1993), Composite International Diagnostic Interview (CIDI). Geneva: World Health Organization Zahn-Waxler C, Klimes-Dougan B, Slattery MJ (2000), Internalizing problems of childhood and adolescence: Prospects, pitfalls, and progress in understanding the development of anxiety and depression. Dev Psychopathology 12:443-466

Table 1. Rates (%) of mental health, educational and social role outcomes by number of anxiety disorders (14-16 years). Number of anxiety disorders (14-16 years) Measure Mental Health Outcomes (16-21 years) Anxiety disorder (%) Major depression (%) Nicotine dependence (%) Alcohol dependence (%) Illicit drug dependence(%) Suicide attempt (%) Educational Achievement (16-21 years) Entered university (%) Entered tertiary education/training (%) Social Role (21 years) Unemployed (3 months) (%) Became a parent (%) 46.0 6.0 53.2 8.9 47.5 14.5 57.7 30.8 2.5 21.7 >.20 <.0001 37.5 54.5 23.7 37.9 24.1 43.1 26.9 34.6 10.9 19.9 <.001 <.0001 13.3 26.2 25.3 8.1 3.4 5.2 41.8 42.8 32.3 10.9 5.5 8.5 45.9 62.3 26.2 13.1 6.6 11.5 76.9 84.6 46.2 23.1 19.2 11.5 127.5 76.8 5.3 7.5 11.9 6.6 <.0001 <.0001 <.05 <.01 <.001 <.05 None (N = 676) 1 (N = 201) 2 (N = 61) 3 or more (N = 26) 2 (1) p

23 Table 2. Social background, family functioning and individual characteristics of anxious adolescents (14-16 years). Number of anxiety disorders (14-16 years) Measure Social Background Factors Mother aged <21 years (birth) (%) Mother lacked formal educational qualifications (birth) (%) Family unskilled/semi-skilled socio-economic status (birth) (%) Below average family living standards (0-10 years) (%) Family Functioning Experienced parental change (0-14 years) (%) Highest decile family conflict score (0-10 years) (%) Regular/severe physical punishment (0-16 years) (%) Contact sexual abuse (0-16 years) (%) Parental history alcohol problems (15 years) (%) 23.8 19.9 8.6 4.4 9.6 41.8 31.9 18.2 16.7 15.8 44.6 35.1 16.4 10.9 21.3 65.0 36.4 28.0 28.0 26.9 24.0 15.8 18.9 35.4 16.8 <.0001 <.0001 <.0001 <.0001 <.0001 12.2 46.4 22.0 20.1 17.7 57.1 30.0 28.1 9.7 54.8 32.3 38.7 42.3 57.7 46.2 42.3 9.6 6.3 13.4 19.6 <.01 <.05 <.0001 <.0001 None 1 2 3 2 (1) p

24

Number of anxiety disorders (14-16 years) Measure Lowest quartile parental attachment score (15 years) (%) Individual Factors Female (%) Depression (14-16 years) (%) Conduct disorder (14-16 years) (%) Alcohol abuse (14-16 years) (%) Highest quartile neuroticism score (14 years) (%) Lowest quartile of IQ distribution (9 years) (%) Highest decile deviant peer affiliations (14-16 years) (%) 42.2 7.2 7.3 7.9 5.2 21.4 9.2 65.5 20.7 10.8 14.8 35.1 30.4 12.8 72.6 29.0 17.7 17.7 49.2 46.2 17.7 88.5 42.3 19.2 30.8 78.3 20.0 38.5 60.8 66.8 11.8 22.5 101.6 6.0 20.1 <.0001 <.0001 <.001 <.0001 <.0001 <.10 <.0001 None 1 2 3 2 (1) 16.1 p

20.5

27.2

34.4

46.2

<.0001

25 Table 3. Rate (%) of mental health, educational and social role outcomes by number of anxiety disorders (14-16 years) after adjustment for confounding factors. Number of anxiety disorders (14-16 years) Measure Mental Health Outcomes (16-21 years) Anxiety disorder (%) Major depression (%) Nicotine dependence (%) Alcohol dependence (%) Illicit drug dependence (%) Suicide attempt (%) Educational Achievement (16-21 years) Entered university (%) Entered tertiary education/training (%) Social Role (21 years) Became a parent (%)
1

None (N = 676)

1 (N = 201)

2 (N = 61)

3 or more (N = 26)

Significant covariates 1

17.4 28.5 27.4 9.2 3.0 5.9

28.8 38.2 26.4 9.6 4.8 6.5

43.7 49.2 25.4 10.0 7.5 7.2

60.0 60.5 30.4 10.4 11.5 7.9

<.0001 <.0001 >.60 >.70 <.05 >.40

1-4, 7 1-3, 9, 10 3, 5, 10, 12 3, 5, 12 3, 4, 6, 9 3, 6, 9, 13

34.2 50.2

26.0 45.0

19.0 39.9

13.4 35.0

<.05 <.10

1, 6, 8, 11, 14, 15 1, 6, 8, 11, 12, 14

7.5

8.1

8.6

9.2

>.60

1, 7, 10, 11

Covariates: 1 = gender; 2 = depression (14-16 years), 3 = childhood sexual abuse (0-16 years); 4 = neuroticism (14 years); 5 = alcohol abuse (14-16 years); 6 = conduct problems (14-16 years); 7 = maternal age (birth); 8 = family living standards (0-10 years); 9 = attachment to parents (15 years); 10 = parental changes (0-14 years); 11 = IQ score (9 years); 12 = deviant peer affiliations (14-16 years); 13 = parental history of alcohol problems; 14 = family socioeconomic status (birth); 15 = maternal educational achievement (birth).