American Journal of Medical Genetics Part C (Seminars in Medical Genetics) 148C:99 – 117 (2008) A R T I C L E High Risk Studies and Developmental Antecedents of Anxiety Disorders DINA R. HIRSHFELD-BECKER,* JAMIE A. MICCO, NICOLE A. SIMOES, AND AUDE HENIN The past two decades have witnessed significant growth in our understanding of the developmental antecedents of anxiety disorders. In this article, we review studies of offspring at risk for anxiety disorders, longitudinal studies of the course of anxiety disorders in clinical, epidemiologic, and at-risk samples, studies of hypothesized temperamental risk factors for anxiety, and give a brief overview of the literature on environmental risk factors. Clear developmental antecedents to anxiety disorders identified include (1) childhood anxiety disorders [in particular, separation anxiety and overanxious disorder/general anxiety disorder (GAD)], (2) behavioral inhibition which predicts later social phobia, (3) anxiety sensitivity which predicts later panic disorder, and (4) negative affectivity, which predicts a spectrum of psychopathology including anxiety disorders. Further prospective studies are needed to examine the roles of environmental factors such as parenting practices, peer influences, stressful life events, and perinatal stressors. Future studies could benefit from (1) beginning earlier in development and following individuals into adulthood, (2) assessing the overlap between multiple temperamental constructs, (3) greater use of observational measures of temperament and of parent–child and peer interactions, (4) greater attention to parental psychopathology which may confound associations noted, (5) exploration of other features of anxiety disorders (neurofunctional correlates, cognitive features, other aspects of emotional regulation) as potential precursors, and (6) intervention studies exploring whether modifying developmental antecedents can alter the course of anxiety disorders. ß 2008 Wiley-Liss, Inc. KEY WORDS: anxiety disorders; antecedents; temperament; high-risk studies; behavioral inhibition How to cite this article: Hirshfeld-Becker DR, Micco JA, Simoes NA, Henin A. 2008. High risk studies and developmental antecedents of anxiety disorders. Am J Med Genet Part C Semin Med Genet 148C:99–117. INTRODUCTION The ability to identify early in life individuals at highest risk for anxiety disorders would enhance efforts to understand the genetic and neurobiological underpinnings and developmental psychopathology of the disorders, as well as efforts to intervene preventively. Over the past two decades, information about the early risk factors for anxiety has burgeoned. In the following review, we provide a detailed overview of high-risk studies, early diagnostic predictors, and temperamental antecedents, and a broad summary of the environmental factors associated with onset of anxiety disorders; discuss the strengths and limitations of these studies; and highlight areas for further inquiry. In reviewing this literature, several methodological issues warrant consideration. First, the category ‘‘anxiety disorders’’ is extremely broad, encompassing no fewer than eight separate DSM-IV disorders, yet many studies Dina Hirshfeld-Becker, Ph.D. is Director of High Risk Studies and Anxiety Research in the CRPPP, and is Associate Professor of Psychiatry at the Harvard Medical School. Jamie Micco, Ph.D. is a researcher in the High Risk Studies Program in the CRPPP and an Instructor in Psychiatry at the Harvard Medical School. Nicole Simoes, M.A. is a graduate of Tulane University who serves as research coordinator of an NIMH-funded longitudinal study of children at risk for anxiety disorders. Aude Henin, Ph.D. is the Director of Cognitive Behavior Therapy Research in the CRPPP, an Instructor of Psychiatry at the Harvard Medical School, and a co-investigator and project director of an NIMH-funded longitudinal study of children at risk for anxiety disorder. *Correspondence to: Dina R. Hirshfeld-Becker, Ph.D., MGH Clinical and Research Program in Pediatric Psychopharmacology, 185 Alewife Brook Parkway, Suite 2000, Cambridge, MA 02138. E-mail: email@example.com DOI 10.1002/ajmg.c.30170 ß 2008 Wiley-Liss, Inc. treat these disorders as a unitary construct. Arguments for considering the anxiety disorders as a single category include considerable comorbidity between the disorders, evidence from twin and family studies suggesting shared genetic variability for several, and similarities in brain regions underlying the disorders and in effective treatment approaches (cognitive-behavioral as well as pharmacological). Arguments for regarding the anxiety disorders as discrete entities include their distinct ages of onset, courses, phenomenologies, and, in some cases, distinct patterns of association. In the present review, we include studies of risk for anxiety disorders in general, as well as more detailed consideration of the antecedents of several well-studied discrete disorders. Second, of the different methodologies used to study developmental antecedents of anxiety disorders, prospective longitudinal designs are the most informative. Capitalizing on the 100 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) moderate heritability of the anxiety disorders, offspring-at-risk studies recruit young children of parents affected with an anxiety disorder of interest and a contrast group of offspring of unaffected parents, and follow them prospectively to examine early disorders, symptom constellations, and biological or temperamental markers and relate these to onset and course of psychopathology. Such studies ensure high rates of the outcome of interest with a relatively small initial sample. In contrast, studies of unselected epidemiologic or community samples examine large cohorts of children for hypothesized early risk factors and link these prospectively to outcomes of interest. Such studies eliminate the bias introduced by clinical referral of parent probands but require very large samples to ensure sufficient representation of the disorders studied. Longitudinal studies of selected samples begin with children chosen on the basis of the presence or absence of a hypothesized risk factor, without regard to parental psychopathology, and then examine the children’s outcomes over time. These studies enable the study of factors with relatively low prevalence without introducing bias inherent in beginning with clinically-referred parents. Finally, we note that space limitations required us to prioritize which information in this rapidly growing field to include and which to exclude. Because we reasoned that high-risk and prospective longitudinal studies would be most informative about the developmental antecedents of anxiety disorders, we elected to present the most detailed information about factors studied using these methods; namely parental psychopathology, early diagnostic and temperamental markers, and anxiety sensitivity (AS). Although retrospective studies of affected adults and cross-sectional studies of affected children can be useful in hypothesis generation, they are subject to confounds of biased recall or of concurrent symptoms, which may have overlap with behavioral and physiologic precursors. Consequently, although we sought to provide a broad overview of factors associated with anxiety disorders and to underscore in particular data from prospective longitudinal studies where available, we tended to de-emphasize studies relying mainly upon crosssectional methodologies, particularly in the case of environmental risk factors (e.g., the bulk of the literature on parenting factors). This is not to say that cross-sectional studies cannot be invaluable in hypothesis generation and model building for future prospective analyses, only that we did not have room for them in our review. We have therefore made an effort to direct the reader to thorough reviews of the areas to which we have devoted less emphasis. HIGH-RISK OFFSPRING STUDIES Fourteen studies have examined rates of anxiety disorders among offspring of parents with anxiety disorders. Most included offspring of parents with panic or agoraphobia [Weissman et al., 1984; Sylvester et al., 1988; Biederman et al., 1991, 2001a, 2004, 2006b; Warner et al., 1995; Capps et al., 1996] or of combined groups of parents with anxiety disorders [Turner et al., 1987; Beidel and Turner, 1997; Merikangas et al., 1998a], although some studied offspring of parents with generalized anxiety disorder (GAD) [Johnson et al., 2006], animal phobias [Unnewehr et al., 1998], or obsessive-compulsive disorder [Black et al., 2003] in particular. Rates of ‘‘any anxiety disorder’’ or ‘‘two or more anxiety disorders’’ between offspring of parents with anxiety and offspring of non-affected controls differed significantly in all but two of the studies [Warner et al., 1995; Merikangas et al., 1998a], with rates of ‘‘any anxiety disorder’’ ranging from 21% to 68% in the at-risk offspring as contrasted with 0–26% in the control offspring. Rates of ‘‘any anxiety disorder’’ or ‘‘two or more anxiety disorders’’ between offspring of parents with anxiety and offspring of non-affected ARTICLE controls differed significantly in all but two of the studies, with rates of ‘‘any anxiety disorder’’ ranging from 21% to 68% in the at-risk offspring as contrasted with 0–26% in the control offspring. Combining all studies that reported the rate of ‘‘any anxiety disorder’’ (comprising 1,371 children), the 465 offspring of parents with anxiety disorders had a rate of ‘‘any anxiety disorder’’ of 37%, compared with 15% among the 906 offspring of non-anxious controls (Chi-square ¼ 83.96, P < 0.0001). In addition, ‘‘bottom-up’’ studies suggest that about half of parents of children presenting clinically with anxiety meet criteria for anxiety disorders themselves [e.g., Last et al., 1991]. Evidence for specificity of transmission is mixed, with some studies showing transmission of the same disorder from parent to child, for panic disorder [Unnewehr et al., 1998; Biederman et al., 2001a, 2004, 2006b], social phobia [Lieb et al., 2000; Biederman et al., 2006a] or specific phobia [Unnewehr et al., 1998], but with most also showing that parental anxiety disorders confer risk for a spectrum of disorders in the offspring. High risk studies also suggest that comorbid anxiety or mood disorders in the affected parent increase the children’s risk for anxiety disorder [Biederman et al., 2005], as does having more than one parent with an anxiety disorder [Merikangas et al., 1998b]. Additionally, an analysis of the effects of comorbid parental disorders found that social phobia and separation anxiety disorder among offspring of parents with panic were predicted by social phobia and separation anxiety disorder in the parents’ own histories, whereas agoraphobia and OCD in these offspring were predicted by parental panic disorder, even after controlling for the presence of comorbid agoraphobia and OCD in the parents [Biederman et al., 2006a]. ARTICLE In addition, several studies have also found elevated risk for anxiety disorders among offspring of parents with major depressive disorder [e.g., Weissman et al., 2006] and bipolar disorder [e.g., Grigoroiu-Serbanescu et al., 1989], although others have not. In complementary fashion, several studies found elevated rates of depression among offspring of parents with anxiety disorders [Sylvester et al., 1988; Biederman et al., 1991; Beidel and Turner, 1997]. The studies reviewed in this section should be interpreted in light of several methodological limitations. First, most focused on offspring of parents with panic and agorphobia or combined parental diagnostic groups, and most did not analyze the effects of parental comorbidity, especially within the anxiety disorders. Second, some studies did not report on all separate childhood anxiety disorders. Third, as reflected in Table I, the participants studied were mainly Caucasians, which limits the generalizability of findings. Finally, the offspring included were within the age of risk for onset of psychopathology, even for childhood psychopathology. Future studies could benefit from directly comparing offspring of parents with a range of distinct anxiety diagnoses, from examining effects of comorbid disorders within parents, and from following offspring longitudinally through adolescence or beyond. Beyond documenting the increased risk for anxiety among offspring of parents with anxiety and mood disorders, the high-risk studies provide evidence for an important early predictor of adulthood anxiety disorders: childhood disorders. Since the offspring in these studies are at moderately increased genetic risk for the adulthood disorders, which have heritabilities in the 0.3–0.4 range [Smoller and Tsuang, 1998], the disorders they manifest may represent prodromes of these disorders. If disorders in offspring parallel those reported retrospectively by affected adults, the hypothesis that childhood disorders are antecedents of the adulthood disorders gains further support. The finding that in some cases, children present specifically with their parents’ AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) disorder suggests continuity of anxiety disorders from childhood to adulthood, and is consistent with retrospective reports indicating that a substantial proportion of adults with panic disorder [Von Korff et al., 1985], social phobia [Stein, 2006], and specific phobia [Öst, 1987] report childhood onset. On the other hand, the finding that childhood separation anxiety disorder, social phobia (or DSM-III-R avoidant disorder), and GAD (or DSM-III-R overanxious disorder) are found in offspring of parents with different disorders suggests that these childhood disorders may be important developmental precursors to different anxiety disorders. These findings dovetail with retrospective accounts by adults presenting clinically with panic disorder and social phobia, of whom at least half report having had one of these three childhood disorders [Otto et al., 1994, 2001]. EARLY SYMPTOMATIC ANTECEDENTS A total of 14 studies, summarized in Table II, have prospectively examined the course of childhood anxiety disorders (in 12 samples) from early to middle childhood or from adolescence to young adulthood. The converging evidence suggests that early anxiety symptoms or disorders, whether found in clinical, community, epidemiologic, or offspring-at-risk samples, significantly increase the risk of meeting criteria for anxiety disorders later in life. For example, two epidemiologic studies found that children diagnosed with an anxiety disorder in pre- or early adolescence were nearly three times as likely as those without a prior anxiety disorder to have an anxiety disorder in young adulthood [Pine et al., 1998; KimCohen et al., 2003]. Follow-up of clinical samples have also found that childhood anxiety increases the risk for later anxiety. Moreover, studies from two large epidemiologic samples suggest that childhood major depression may also increase the risk for development of subsequent anxiety disorders [Costello et al., 2003; Kim-Cohen et al., 2003; Gregory et al., 2007]. 101 In the studies that examined individual disorders, evidence for both homotypic (i.e., predicting to the same disorder) and heterotypic (i.e., predicting to other anxiety disorders) continuity was found. For example, Pine et al.  found that social phobia, simple phobia, and childhood overanxious disorder/GAD in adolescence each predicted the corresponding disorder in early adulthood; however, overanxious disorder also predicted specific and social phobia and depressive disorders. Similarly, scales from a self-report questionnaire measuring separation anxiety disorder, GAD, social phobia and panic disorder at ages 10–12 each predicted the corresponding scale 2 years later; however each scale was also predicted in roughly equal measure by the combination of all of the others (heterotypic scales) [Ferdinand et al., 2007]. In the Dunedin Study [Gregory et al., 2007], adults age 32 with GAD, social phobia, and agoraphobia had greater rates at ages 11–15 of all three childhood disorders assessed (OAD, separation anxiety disorder and phobias), whereas adults with specific phobia only had childhood phobias (assessed using an instrument that did not distinguish social and specific phobias). In a high-risk sample [Biederman et al., 2007], separation anxiety disorder at mean age six significantly increased the risk of a range of anxiety and mood disorders 5 years later. Thus, having a childhood anxiety disorder, particularly separation anxiety disorder, overanxious disorder/GAD, or social or specific phobia appears significantly to heighten the risk for a range of later anxiety disorders. Thus, having a childhood anxiety disorder, particularly separation anxiety disorder, overanxious disorder/GAD, or social or specific phobia appears significantly to heighten the risk for a range of later anxiety disorders. (1) SADS-L (2) K-SADS-E (1) SADS-L (2) K-SADS-E N ¼ 214, ages 6–23 N ¼ 145, ages 6–24 (parent and direct child interviews) Warner et al.  PD þ MDD PD PD þ MDD PDAG þ MDD PDAG (1) DIS (2) DICA-P Mufson et al. a Biederman et al.  Sylvester et al.  AG or OCD PD (1) ADIS (2) CAS N ¼ 59, ages 7–12 (direct child interviews); 93% Caucasian, 7% African American N ¼ 125, ages 7–17 (direct ch. ints) N ¼ 121, ages 4–22 (parent interviews); 100% Caucasian Turner et al.  MDD þ PD Diagnoses of anxiety probands (1) DIS (2) DICA (1) SADS-L (2) ‘‘screening instrument’’ for diagnosis of offspring N ¼ 194, ages 6–17 (community sample; parent interviews) Weissman et al.  References Diagnostic interview Sample size, age of children, and ethnicity of children (where reported) Early-onset (<30) MDD MDD MDD MDD Dysthymic Disorder MDD without anxiety disorder Diagnoses of psychiatric control probands Neverpsychiatrically-ill controls Neverpsychiatrically-ill controls Children without known psychiatric dxs & siblings Normal controls No DSM-III diagnoses Neverpsychiatrically-ill controls Description of normal control probands Anx: SAD: SOC: Anx: SAD: PD: PDAG: Anx: SAD: OAD: Phob: Anx: Anx: SAD: SOC: OAD: Anx: SAD: PD: AG: SOC: n ¼ 60 28% 15% 12% Anx: SAD: OAD: Phob: n ¼ 14, 21% 14% 14% 7% Anx: SAD: SOC.: Anx: SAD: PD: PDAG: Anx: n ¼ 50 42% n ¼ 79 46% 20% 4% 5% n ¼ 17, 24% 12% 6% Anx: SAD: SOC: OAD: Anx: SAD: PD: AG: SOC: n ¼ 19 37% 37% 5% 5% 5% n ¼ 16 37% 25% 0% 12% n ¼ 25 48% 16% 24% 24% In offspring of anxious probands n ¼ 74 34% 18% 4% 5% n ¼ 32 34% 12% 16% n ¼ 12 8% 0% 8% 0% n ¼ 27 44% n ¼ 38 0% 0% 0% 0% 0% n ¼ 14 22% 7% 7% 7% In offspring of psychiatric control probands Rates of anxiety and depressive disorders TABLE I. Rates of Anxiety Disorders in the High-Risk Offspring of Parents With Anxiety Disorders Anx: SAD: SOC: Anx: SAD: PD: PDAG: Anx: SAD: OAD: Phob: Anx: Anx: SAD: SOC: OAD: Anx: SAD: PD: AG: SOC: n ¼ 20 20% 10% 0% 0% n ¼ 36 20% 3% 11% n ¼ 47 11% 4% 2% 2% n ¼ 48 15% n ¼ 87 2% 0% 0% 0% 1% n ¼ 29 3% 0% 0% 3% In offspring of normal controls 102 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) ARTICLE (1) SCID (2) K-SADS-E N ¼ 312, ages 5–25; mean age 6.8 (parent interviews; direct interviews for children age 12 and older); 93% Caucasian, 3% AfricanAmerican, 2% Latino, 2% Asian Biederman et al. [2001a] (1) SADS (2) K-SADS-E Anx (PD, OCD, SOC); (1) SCID (2) K-SADS PD þ MDD PD Animal Phobia PD Anx (PDAG, SOC, GAD) Anx/MDD (PD, OCD, GAD) PDAG (1) ADIS-R (2) DISC-C/ DISC-P (1) DIPS N ¼ 87, (2) Kinder-DIPS age 5–15; (parent and direct child interviews) N ¼ 32, ages 8–14; (parent and direct child interviews) 13% Caucasian, 13% AfricanAmerican, 13% AsianAmerican, 6% Latino N ¼ 129, ages 7–12; (parent and direct child interviews); 83% Caucasian, 15% AfricanAmerican, 2% East Indian N ¼ 192, ages 7–18; (parent and direct child interviews) Unnewehr et al.  Merikangas et al. [1998a] Beidel and Turner  Capps et al.  MDD N/A Parents without PDAG and MDD Neverpsychiatrically-ill controls Neverpsychiatrically-ill controls Neverpsychiatrically-ill controls MDD Substance Abuse or Dependence Neverpsychiatrically-ill controls N/A n ¼ 23 17% 48% n ¼ 141 26% 27% 4% 10% 19% 13% 4% 18% 3% 65% 32% n ¼ 26, 27% 19% 8% 0% 14% 4% 5% 15% 8% 2Anx: SAD: PD: SOC: AG: OAD: GAD: Phob: OCD: 2Anx: SAD: PD: SOC: AG: OAD: GAD: Phob: OCD: n ¼ 46 15% 20% 4% 11% 7% 11% 0% 11% 2% 2Anx: SAD: PD: SOC: AG: OAD: GAD: Phob: OCD: n ¼ 99 4% 4% 0% 2% 3% 4% 1% 10% 0% 7% 13% n ¼ 57 11% 7% 0% 0% 5% 4% n ¼ 30 n ¼ 48 8% 0% 2% 2% n ¼ 16 0% 0% 0% 0% 0% (Continued) SAD/ OAD/PD: AG/Phob: Anx: SAD: PD: SOC: OAD: Phob: SAD/ OAD/ PD: AG/Phob: Anx: SAD: PD: SOC: OAD: Phob N/A n ¼ 77 10% 3% 0% 1% 8% 1% n ¼ 58 22% 12% 2% 7% 12% 7% n ¼ 34, Anx: SAD: PD: SOC: OAD: Phob: n ¼ 24 21% 4% 13% 0% Anx: SAD: SOC.: OAD: Anx: SAD: SOC.: OAD: n ¼ 29 33% 0% 7% 3% n ¼ 28, 33% 4% 0% 21% Anx: SAD: SOC.: OAD: Anx: SAD: Avoid: OAD: GAD: N/A n ¼ 16 68% 56% 13% 25% 13% Anx: SAD: Avoid: OAD: GAD: ARTICLE AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) 103 N ¼ 78, ages 7-18 (parent and direct child interviews); 98% Caucasian 2% AfricanAmerican N ¼ 593, ages 26–39 (parent and direct offspring interviews); 91% Caucasian N ¼ 319, ages 7–18, mean age 10 (parent interviews and direct child interviews for children age 12 and older); 95% Caucasian PD þ MDD PD GAD (1) SCID-IV (2) DISC (1) SCID (2) K-SADS-E OCD Diagnoses of anxiety probands (1) SCID-IV (2) DICA Diagnostic interview MDD N/A N/A Diagnoses of psychiatric control probands Parents without PDAG and MDD No lifetime history of GAD (may have other psychiatric disorders) No lifetime history of OCD Description of normal control probands 2Anx: SAD: PD: SOC: AG: GAD: Phob: OCD: Anx: Anx: SAD: OAD: Phob: OCD: n ¼ 27, 37% 37% 15% 15% 26% 5% 15% 8% n ¼ 136 42% 36% 10% 22% 24% 4% 18% 3% 2Anx: SAD: PD: SOC: AG: GAD: Phob: OCD: n ¼ 53 28% 30% 2% 15% 7% 0% 11% 2% 2Anx: SAD: PD: SOC: AG: GAD: Phob: OCD: Anx: n ¼ 90 41% N/A Anx: SAD: OAD: Phob: OCD: N/A n ¼ 103 12% 12% 3% 6% 3% 1% 10% 0% n ¼ 503 19% n ¼ 35 26% 6% 9% 9% 3% In offspring of normal controls n ¼ 43 51% 17% 32% 21% 23% In offspring of anxious probands In offspring of psychiatric control Probands Rates of anxiety and depressive disorders Italicized percentages indicate those reported to differ significantly from rates in the non-affected control group. SADS-L, schedule for affective disorders and schizophrenia, lifetime version; CAS, children’s assessment schedule; ADIS, anxiety disorders interview schedule, R, for DSM-III-R; DISC, diagnostic interview schedule for children; DIS, diagnostic interview schedule; DICA-P, diagnostic interview for children and adolescents-parent version; KSADS, schedule for affective disorders and schizophrenia, child version, E, epidemiological version; DIPS, German version of the ADIS-R; Kinder-DIPS, German version of the ADIS-C/P; SCID, structured clinical interview for DSM-III-R, IV, for DSM-IV; DICA, diagnostic interview for children and adolescents; Anx, presence of any anxiety disorder; 2Anx, presence of two or more anxiety disorders in a child; PD, panic disorder; PDAG, panic disorder with agoraphobia; AG, agoraphobia; MDD, major depressive disorder; OCD, obsessive compulsive disorder; GAD, generalized anxiety disorder; SOC, social phobia; SAD, separation anxiety disorder; OAD, childhood overanxious disorder; Phob, specific phobia. a Follow-up of Weissman et al.  study, with overlapped sample. b Follow-ups of Biederman et al. [2001a] study. Biederman et al. b Johnson et al.  Black et al.  References Sample size, age of children, and ethnicity of children (where reported) TABLE I. (Continued ) 104 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) ARTICLE Clinical Clinical Klein  Last et al.  Epidemiologic Clinical Epidemiologic (Christchurch Health and Dev’t Study) Epidemiologic Pine et al.  Weissman et al.  Woodward and Fergusson  Costello et al.  Epidemiologic Clinical High-Risk Kim-Cohen et al.  Aschenbrand et al.  Bosquet and Egeland  b Epidemiologic (Dunedin Study) Newman et al.  b Community Sample type Ferdinand and Verlhurst References CBCL, K-SADS (at age 17) AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) (Continued) Correlations between CBCL ‘‘anxiety/depression’’ scores and anxiety symptoms at age 17.5: . Childhood CBCL score, r ¼ 0.20 (P < 0.05) . Preadolescent CBCL score, r ¼ 0.08 (NS) . CBCL score at age 16, r ¼ 0.43 (P < 0.001) Controlling for comorbidity, past anxiety disorders increased the risk of having future anxiety disorders (OR ¼ 2.0) N ¼ 155, offspring of low-income mothers, assessed numerous times from birth to age 17.5 years Ethnicity: 70% Caucasian, 9% African American, 2% Native American or Hispanic, 16% multi-racial CAPA N ¼ 1,420, North Carolina birth cohort, ages 9–13; Ethnicity: predominantly Caucasian, 25% American Indian, 8% African American, <1% Hispanic, FU at age 16 Rates of anxiety disorder at FU: . 76.9% in those with 3þ prior anxiety disorders . 45.9% in those with 2 prior anxiety disorders . 41.8% in those with 1 prior anxiety disorder . 13.3% in those without a prior anxiety disorder . No significant differences in rate of PDA at FU between those with prior SAD, GAD, or SOC . Those with prior SAD had greater risk of developing SPEC, OCD, or PTSD/ASD (OR ¼ 4.8) DISC, CIDI N ¼ 964, New Zealand birth cohort, ages 15–16, FU period ¼ 1 to 5 years No significant difference between groups in risk of developing subsequent anxiety disorders N ¼ 85, children who received treatment for SOC, GAD, or SAD, ADIS-C/P, ADIS-IV-L ages 9–13, Ethnicity: 87% Caucasian, 5% African American, 2% Asian, 4% Other, FU period ¼ mean of 7 years K-SADS, SADS-LA N ¼ 44, children with anxiety, ages 6–12; 83 with MDD and 91 NPI controls, Ethnicity: 33% Caucasian, 32% African-American, 33% Hispanic, FU period ¼ mean of 11.9 years Anxiety at Times 1 and 2 predicted anxiety at Time 3: . Simple phobia ! Simple phobia (OR ¼ 3.79) . SOC and OAD ! SOC (OR ¼ 3.29 and 2.27) . OAD and ‘‘PD features’’ ! GAD (OR ¼ 2.48/3.39) Of those with anxiety disorders at age 21, 61.5% had a prior anxiety disorder, 18.9% had a different prior disorder and 19.5% had no prior disorder . 81.9% of initial anxiety disorders remitted by FU, with an 8% relapse rate of initial disorder . 15.5% of anxiety group developed new anxiety disorder (vs. 10% and 2.4% of ADHD and NPI groups). Compared to controls, those with prior SAD had greater rates of PD (7% vs. 0%) Controlling for sex, an anxiety disorder before age 15 predicted an anxiety disorder at age 26 (OR ¼ 2.9) DISC N ¼ 776, children ages 9–18, assessed at mean ages 13.7, 16.4, and 22.1, Ethnicity: 91% Caucasian, FU period ¼ 9 years Results Clinically elevated ‘‘Anxious/Depressed’’ Score at Time 1 significantly predicted clinically elevated ‘‘Anxious/ Depressed’’ Score at Time 2 (OR ¼ 3.0) N ¼ 1,037, New Zealand birth cohort assessed multiple times from DISC, DIS age 11 to 26, Ethnicity: 91% Caucasian, FU period ¼ 15 years DISC, DIS K-SADS N ¼ 102, children ages 5–18 with anxiety disorders, 58 children with ADHD, and 87 NPI controls, Ethnicity: 72% Caucasian, FU period ¼ 3–4 years N ¼ 961, New Zealand birth cohort assessed at ages 11, 13, 15, 18, and 21, Ethnicity: 91% Caucasian, FU period ¼ 10 years Semi-structured clinical interview N ¼ 54, children treated with imipramine for SAD, and 60 community controls, FU period ¼ 15 years Measures CBCL young adult self-report N ¼ 459, Dutch adolescents, ages 13–16, FU period ¼ 8 years Sample characteristicsa TABLE II. Prospective Studies of the Association Between Childhood Anxiety and Later Anxiety Disorders ARTICLE 105 Community Ferdinand et al.  Results RCADS (self-report questionnaire) Examined prediction of each symptom scale at Time 2 by the same scale at Time 1 (homotypic) or by all other scales at Time 1 (heterotypic). Variances reflecting homotypic vs. heterotypic continuity were: . SAD: 2.3% vs. 2.0% . GAD 3.6% vs. 4.2% . SOC: 3.7% vs. 1.3% . PD: 3.2% vs. 2.3% Controlling for parental diagnosis: . SAD at Time 1 increased risk at Time 2 for SPEC (OR 4.7), AG (OR 9.1), and PD (OR 9.2) . AG at Time 1 predicted GAD at Time 2 (OR 3.9) N ¼ 2,067, children from the Dutch general population, ages 10–12, FU period ¼ 2 years Measures . Adults with GAD, SOC, AG, PTSD, OCD or SPEC all had elevated anxiety dxs at 11–15 . Those with GAD, SOC, and AG all had elevations in all three childhood anxiety disorders, but those with SPEC only had elevated phobias . Adults with GAD, SOC, AG, PTSD, OCD all had an excess of MDD at 11–15 . Those with SOC, AG, and especially PTSD had an excess of disruptive behavior disorders at 11–15 K-SADS N ¼ 963, New Zealand birth cohort assessed at ages 11–32. DISC, DIS (Anxiety dxs assessed at 11–15 were Examined which disorders from ages 11 to 15 differed between individuals with or without anxiety disorders at age 32 SAD, OAD, and phobias) Ethnicity: 91% Caucasian, FU period ¼ 21 years N ¼ 233, offspring of parents with PD and/or MDD, ages 5–13; Ethnicity: 95% Caucasian, FU period ¼ 4.5 years Sample characteristicsa Information on ethnicity of sample included where available. CBCL, child behavior checklist; semi-structured clinical interview, schedule for the assessment of conduct, hyperactivity, anxiety, mood, and psychoactive substances; KSADS, schedule for affective disorders and schizophrenia, child version; DISC, diagnostic interview schedule for children; DIS, diagnostic interview schedule; SADS-LA, schedule for affective disorders and schizophrenia, lifetime version; CIDI, composite international diagnostic interview; CAPA, child and adolescent psychiatric assessment; ADIS-IV, anxiety disorder interview schedule for DSM-IV, C/P, child/parent version, L, lifetime; RCADS, revised child anxiety and depression scale; FU, follow-up; OR, odds ratio; PD, panic disorder; SAD, separation anxiety disorder; SOC, social phobia; OAD, childhood overanxious disorder; GAD, generalized anxiety disorder; SPEC, specific phobia; OCD, obsessive compulsive disorder; AG, agoraphobia; PTSD, post-traumatic stress disorder; ASD, acute stress disorder; ADHD, attention deficit-hyperactivity disorder; NPI, not psychiatrically ill. a Includes ethnicity where available. b Multiple follow-ups of the same sample. Epidemiologic High-Risk Sample type Gregory et al.  b Biederman et al.  References TABLE II. (Continued ) 106 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) ARTICLE ARTICLE This conclusion should be viewed in light of the methodological limitations of the studies reviewed. For example, not all assessed the full spectrum of childhood anxiety disorders, and some of the longer prospective studies were affected by changes to the classification of childhood anxiety disorders between DSM-III and DSM-IV. In addition, some of the studies had gaps between assessments during which psychopathology was not assessed, so that some anxiety disorders may have been missed. Also, with a few exceptions, the majority of studies included mainly Caucasian participants, limiting the generalizability of findings. In addition, because participants in all but one of the studies reviewed were still well within the age range for the onset of anxiety disorders, studies following these samples further into adulthood are needed to elucidate the full spectrum of risk signaled or conferred by childhood anxiety. Moreover, since anxiety disorders can onset as early as preschool-age [Egger and Angold, 2006], studies are needed which assess anxiety disorders in young children and follow them through adulthood. TEMPERAMENTAL ANTECEDENTS Beyond considering syndromatic or symptomatic antecedents, a growing number of studies have focused on temperamental characteristics that predate onset of anxiety disorders yet appear to increase their likelihood. It is beneficial to examine temperament very early in life, to avoid its being confounded by other factors (including onset of psychopathology). Behavioral Inhibition The temperamental construct most widely studied through prospective standardized observational methods as a potential antecedent to diagnosable anxiety disorders is behavioral inhibition to the unfamiliar (BI). BI refers to the persistent tendency to exhibit restraint, withdrawal, and reticence when faced AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) with novel or unfamiliar situations and people [Kagan et al., 1988]. The temperamental construct most widely studied through prospective standardized observational methods as a potential antecedent to diagnosable anxiety disorders is behavioral inhibition to the unfamiliar (BI). BI refers to the persistent tendency to exhibit restraint, withdrawal, and reticence when faced with novel or unfamiliar situations and people. It is found in 10–15% of children and is moderately stable from toddlerhood through the early elementary school years, with higher stability among children more extreme in BI. Antecedents to BI have been observed as early as age 3 months, including the tendency to respond to novel stimuli with increased distress and motor reactivity [Kagan et al., 1999]. This ‘‘high reactive’’ behavior observed in infancy has been shown in its own right to predict childhood anxiety symptoms. BI is hypothesized to be rooted in a lower threshold to limbic arousal, specifically to higher reactivity of the amygdalae and their projections to the striatum, hypothalamus, sympathetic chain, and cardiovascular system. Evidence for amygdalar reactivity derives from two recent fMRI studies of adolescents or young adults who had been classified as inhibited in toddlerhood [Schwartz et al., 2003; PerezEdgar et al., 2007], and multiple studies have documented markers of sympathetic arousal in inhibited children, including high and more stable heart rate, elevated salivary cortisol, increased startle response and increased right frontal EEG activation in inhibited youngsters (see Fox 2004; Hirshfeld- 107 Becker et al., 2004b for reviews). BI has been shown to have moderate heritability (0.41–0.64) which is even higher among children rated as extreme in BI [Smoller and Tsuang, 1998]. It has shown association with a genetic marker at the corticotropin releasing hormone (CRH) locus [Smoller et al., 2005]. Another study found a gene-environment interaction, in which children who had both the short 5-HTTallele and low social support had increased risk for BI [Fox et al., 2005]. A growing number of studies have suggested that BI is linked to risk for social phobia in childhood and adolescence, and possibly to more broad risk for anxiety or mood disorders later on. Four high-risk studies have found laboratory observed preschool-age BI to be elevated among offspring of parents with panic disorder [Rosenbaum et al., 1988, 2000; Manassis et al., 1995; Battaglia et al., 1997], although one study examining reactive temperament and BI in infancy found no association with maternal panic disorder [Warren et al., 2003]. A ‘‘bottom-up’’ study found elevated rates of anxiety disorders, particularly social phobia, in parents of children selected as BI based on laboratory observations compared with nonBI children [Rosenbaum et al., 1991]. Prospective longitudinal studies of children with laboratory-observed BI from high-risk and community samples have examined the outcome of childhood BI. Although small pilot studies that included only offspring of psychiatric, but not of non-clinical, controls suggested that BI predicted anxiety disorders in general [Biederman et al., 1990, 1993], larger, better controlled studies have clarified that BI appears prospectively associated with social anxiety in particular [Hayward et al., 1998; Schwartz et al., 1999; Biederman et al., 2001b]. For example, BI observed at preschool-age among children at risk for panic and depression predicted new onset social phobia (but no other disorders) by mean age 10 (22% vs. 8%, P < 0.05), after covarying parental dia gnosis [Hirshfeld-Becker et al., 2007]. Similarly, among 13-year olds who had been selected as inhibited or 108 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) uninhibited in toddlerhood, those with BI had higher rates of generalized social anxiety with impairment (34% vs. 9%, P < 0.05), with the effect greatest for girls (44% vs. 6%, P < 0.05) [Schwartz et al., 1999]. No differences for specific fears, separation anxiety, or performance anxiety were found. Studies suggest that BI is most predictive of anxiety disorders when it is found among offspring of parents with anxiety disorders [Rosenbaum et al., 1992; Biederman et al., 2001b] and when it remains stable across early childhood [Hirshfeld et al., 1992]. Studies using retrospective or concurrent self-report measures of BI in older children, adolescents, and adults from community and clinical samples have been mixed, with some finding links to anxiety in general [e.g., van Ameringen et al., 1998; Muris et al., 2003] and others finding stronger or more specific links to social anxiety disorder [e.g., Mick and Telch, 1998; Gladstone and Parker, 2006]. However, whereas such studies may benefit from the informant being able to report on longer intervals of behavior than a brief lab observation, they are also subject to potential recall bias and social desirability; other limitations include differences in the informants’ knowledge of normative development, the possibility that concurrent BI may be confounded by inhibition secondary to symptom onset, and the absence (in most) of data on parental psychopathology. If BI is more common among offspring of parents with anxiety, individuals endorsing BI in community or clinical samples may have a greater probability of having anxious parents, which may confound associations between BI and anxiety in general. Questions remain as to the degree to which BI is associated with depression as well as social anxiety, since some family [Kochanska, 1991; Rosenbaum et al., 2000], prospective [Caspi et al., 1996], cross-sectional [Muris et al., 2003], and retrospective studies have observed this association, but one found that the link was accounted for by comorbid social phobia [Gladstone and Parker, 2006]. Other questions concern the degree to which BI confers risk in the absence of parental psychopathology and the degree to which inhibition to non-social (as opposed to social) stimuli, usually assessed in only a brief segment of the BI assessment, confers risk for anxiety disorders. In addition, studies of BI are limited by the lack of observational assessments which are standardized across laboratories, the low to moderate correlations between BI observed in the laboratory and reported by parents, the absence of information about the degree to which BI observed in early childhood correlated with BI as reported retrospectively on questionnaires, and the paucity of studies which have followed children assessed as preschoolers through the full period of risk for anxiety (i.e., through early adulthood). Despite these questions and limitations, it seems clear that BI measured in the preschool years is a marker of risk for social anxiety disorder in early and middle childhood and early adolescence, especially in children whose parents have anxiety disorders. Although not all inhibited preschoolers develop social anxiety by early adolescence, a substantial proportion (34–44%) do. Longitudinal follow-up of samples assessed via laboratory observation in early childhood will clarify the full spectrum of risk conferred by BI. Assessment and followup of epidemiologic or community samples in which data about parental diagnostic histories is also assessed would help answer questions about the risk conferred by BI in the absence of parental psychopathology. Other Temperamental and Cognitive Risk Factors At least three other temperamental constructs and one cognitive risk factor, AS, have received support from family and/or prospective studies as potential risk factors for anxiety disorders. Descriptions of the constructs and summaries of the pertinent evidence are presented in Table III. The construct with the most empirical support as a precursor to panic disorder and possibly other anxiety disorders is AS. Numerous ARTICLE cross-sectional studies have documented an association between AS and panic symptoms [McNally, 2002] which persists after controlling for other measures of anxiety proneness such as trait anxiety. The construct with the most empirical support as a precursor to panic disorder and possibly other anxiety disorders is AS. Numerous cross- sectional studies have documented an association between AS and panic symptoms which persists after controlling for other measures of anxiety proneness such as trait anxiety. As detailed in Table III, multiple prospective studies have shown that AS predicts onset of panic symptoms and panic disorder in adults and adolescents. Therefore it holds promise as a vulnerability marker; however, studies examining it among offspring at risk for anxiety have shown mixed results. In addition, in most studies, the variance in panic or anxiety symptoms accounted for by AS is typically small (ranging from 4% to 12%). Few studies have examined the overlap of these four similar constructs, and the literature is limited by the tendency in many studies to examine internalizing symptoms as an outcome rather than anxiety disorders in particular. In addition, the constructs tend to show poor specificity for anxiety. Neuroticism, in particular, is associated with a wide range of disorders [Mineka et al., 1998], and harm avoidance, shyness, and AS have also been associated cross-sectionally with other disorders (e.g., mood disorders). Efforts to refine the prediction of anxiety from neuroticism/negative affectivity (NA) have had some success. For example, (1) high NA in the absence of low positive affectivity differentiates anxiety from depression in a wide number of cross-sectional studies (Continued) Shyness: Initial reticence, awkwardness, or anxiety with new people or in social settings. Generally measured by self-, parent-, or teacher-report Prospective studies: . In a community sample, shyness at multiple points in childhood was linked to high trait anxiety at age 10 [Fordham et al., 1999] . In a representative sample, shyness throughout childhood was linked to higher anxiety/withdrawal at ages 13–14, with the most persistently shy children having 42% rate of anxiety problems compared to 12% in those rarely or never rated shy [Prior et al., 2000] Family studies: . Shy preschoolers had mothers with significantly higher rates of social phobia than children with other behavior problems [Cooper et al., 1999] Neuroticism/negative affectivity (N/NA): A sensitivity to negative stimuli, a predisposition to experience negative emotions or distress [Eysenck et al., 1967; Clark et al., 1994]. Measured with questionnaires in adults and children, and in a few studies with observational measures in children Prospective studies: . Three studies found that high emotionality in infancy or childhood predicted anxious (or anxious/depressed) symptoms 1.5–6 years later, but in some it predicted externalizing scales as well [Rende et al., 1993; Gjone et al., 1997; Rydell et al., 2003]. The latter of these found that fear (and fear regulation) at age 5 predicted internalizing symptoms at 1.5 years FU whereas anger (and regulation of anger) predicted externalizing symptoms . Among males in an unselected sample (the Dunedin Study), examiner-rated emotionality at age 3 predicted onset of panic and agoraphobia at ages 18–21, but not other anxiety disorders [Craske et al., 2001] . In a prospective study of 3,062 male conscripts in Zurich (followed from ages 19 to 36) questionnaire-rated neuroticism and autonomic lability were found to predict generalized anxiety disorder or panic disorder [Angst et al., 1991] . N/NA in children and adolescents predicted anxiety symptoms at 7-month follow-up, but only via the association with concurrent N/NA at FU [Lonigan et al., 2003] Family and genetic studies: . Emotionality (the tendency to get upset easily) was more common in the siblings and parents of children with anxiety plus comorbid depression [Masi et al., 2003] . Multiple large twin studies have estimated the heritability of neuroticism in the 0.30–0.60 range [e.g., Mackinnon et al., 1990] with some of the association between neuroticism and anxiety disorders accounted for by shared genetic variance [Hettema et al., 2006] Harm avoidance (HA): The tendency to respond intensely to aversive stimuli and to avoid punishment, novelty, and non-reward (conceptualized as a combination of neuroticism and introversion) [Cloninger, 1986]. Measured with self-report (or parent-report) questionnaires Prospective studies: . None known, however several studies suggest that HA changes with pharmacological or cognitive-behavioral treatment for anxiety disorders [e.g., Allgulander et al., 1998; Mortberg, 2007], and that HA pretreatment predicts higher anxiety post-treatment [e.g. Mortberg, 2007] Family and genetic studies: . HA was elevated among first degree relatives of adults with social phobia [Stein et al., 2001] . A twin study of 2,680 adult Australian twin pairs, estimated that the heritability of HA (44%) was comparable to that of Eysenck’s neuroticism (35–45%), but that the two constructs appeared to capture somewhat different dimensions of personality [Heath et al., 1994] . Two independent studies have reported that HA shows linkage to a locus on chomosome 8p21 [Cloninger et al., 1998; Zohar et al., 2003] Anxiety sensitivity (AS): The tendency to interpret sensations of physiological arousal as having harmful physical, psychological or social consequences [Reiss et al., 1985; McNally et al., 1994]. This misinterpretation of benign sensations of arousal is hypothesized to contribute to a ‘‘fear of fear’’ cycle, in which anxiety about the sensations further increases arousal, and further fuels the anxiety. AS is hypothesized to predispose to panic disorder, and other anxiety disorders. Measured via self-report questionnaires TABLE III. Prospective and Family Study Evidence for Other Precursors to Anxiety Disorder ARTICLE AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) 109 Prospective studies: . Among 151 college students, baseline AS predicted the number, frequency, and intensity of panic attacks as well as the likelihood of developing an anxiety disorder at 3-year FU [Maller et al., 1992] . Among 1,401 young adults experiencing a stressful situation (i.e., military training), high AS predicted the onset of spontaneous panic attacks after controlling for history of panic attacks and trait anxiety [Schmidt et al., 1997, 1999] . In a community sample of 404 adults, baseline AS predicted panic attacks, anxiety, and other diagnoses over 2-year FU [Schmidt et al., 2006] . Over a 1-year period, AS predicted (a) maintenance and relapse of panic disorder among affected adults and (b) the emergence of panic attacks among adults without panic disorder [Ehlers et al., 2003] . In a 4-year prospective study of more than 2,000 high school students, AS predicted the onset of panic attacks, after controlling for negative affectivity and history of major depression [Hayward et al., 2000] . In a community sample of 107 African-American adolescents, high AS predicted symptoms of panic at 6 months, though not once initial levels of panic were controlled [Ginsburg and Drake, 2002] . Multiple physiologic challenge studies in adults and children find that high AS predicts anxious response to physiologic challenges [e.g., Leen-Feldner et al., 2007] Family and genetic studies: . In two separate studies, offspring at risk for anxiety or mood disorders did not have higher levels of AS than offspring of healthy control parents ; van Beek et al., 2005] . In a sample of adolescents, an association between AS and anxiety symptoms and disorders was evident only among those with a familial risk for anxiety disorders [Pollock et al., 2002] . AS was found to mediate the relationship between parental psychopathology and child anxiety in one study [Drake et al., 2007] . Heritability of AS estimated at 45% [Stein et al., 1999] . Adults with the S/S 5-HTTLPR genotype with higher levels of reported childhood maltreatment had significantly higher levels of AS than subjects in other groups. No relationship was found for neuroticism, suggesting that the effect was specific to AS [Stein et al., 2007] TABLE III. (Continued ) 110 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) ARTICLE (see [Lonigan et al., 2004] and [PerezEdgar and Fox, 2005]); (2) high NA in the presence of low effortful control (EC)—the capacity for active voluntary regulation of attention or behavior [Posner and Rothbart, 1998]—shows links to anxiety and attentional bias for threat in some studies (see [Lonigan et al., 2004]); and (3) subcomponents of NA such as fear and anger have shown differential association with anxiety and disruptive behaviors in some studies, but not others. However, more work needs to be done to define under what circumstances NA (even with normal positive affectivity, high EC, or fearful as opposed to angry affect) leads to anxiety rather than to other outcomes. Finally, the studies of all of these constructs could benefit from the application of observational methods to enable their measurement early in childhood before the onset of other symptomatic antecedents to anxiety. Now that observational laboratory batteries have been developed to assess NA and its subcomponents (e.g., the LAB TAB) and EC [Kochanska and Knaack, 2003] in toddlerhood and early childhood, a fruitful avenue of research might be to examine these temperamental constructs early in life and to assess the degree to which they prospectively predict onset of anxiety. Similarly, research on AS has been limited by its reliance upon self-report questionnaires, which caps the age at which it can be reliably assessed in children. However, novel paradigms have been developed that use signal detection methodology to examine children’s auditory perceptual bias for heartbeat sounds [Eley et al., 2004; Pollock et al., 2006]. Such protocols may offer a means to operationalize hypervigilance to internal sensations characteristic of high AS individuals without reliance on language-ability and insight, possibly making them suitable for use with young children. PHYSIOLOGIC MARKERS OF RISK FOR ANXIETY DISORDER Although the literature on physiologic markers is scant, studies have suggested ARTICLE that infants and young offspring at risk for anxiety disorders show markers of higher physiologic arousal, including elevated heart rates under stress [Battaglia et al., 1997], salivary cortisol [Warren et al., 2003], increased startle responsivity [Grillon et al., 1997], and disturbed sleep [Merikangas et al., 1999; Warren et al., 2003]. Other physiologic factors point to risk for panic disorder. For example, in healthy adults with no history of panic attacks, family history of panic disorder was associated with increased likelihood of panic symptoms after exposure to 35% CO2, and ventilatory abnormalities during inhalation of 5% CO2 [Coryell, 1997; Coryell et al., 2006b], suggesting that these may be markers of risk for panic disorder. In the same sample, adults who had displayed abnormal ventilatory response were more likely to develop panic attacks over 4-year follow-up [Coryell et al., 2006a]. Similarly, in a large study of unselected youths, a history of respiratory problems at age 15 predicted panic onset by 18–21 in females, as did having a parent with a respiratory disorder [Craske et al., 2001]. In males, having had asthma by age 18 or having had frequent colds and infections at age 3 was also associated with panic disorder. Despite these promising studies, far more work on physiologic precursors to anxiety among children at risk and in other samples is needed. HYPOTHESIZED ENVIRONMENTAL RISK FACTORS Twin studies in children and adults estimate much larger contributions to anxiety from unique or unshared environmental factors than from shared environmental factors (e.g., [Hettema et al., 2005; Spatola et al., 2007]). True to these estimates, few sociodemographic or family environment factors have been found to predict anxiety disorders in longitudinal studies, although a few studies found low SES to be a predictor (e.g., [Moffitt et al., 2007]). Several different shared (e.g., parenting behaviors) and unshared environmental factors, including life events, peer rela- AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) tionships, and perinatal factors, have been examined for their influence on anxiety. We discuss them briefly below, and direct the reader to more thorough reviews (see for example our prior reviews of environmental factors in Fredman et al.  and Ollendick and Hirshfeld-Becker ). Parenting Behaviors Experimental studies in animals have suggested that parenting behaviors can influence anxious behaviors in offspring. For example, in rhesus monkeys and rats, infants bred to be ‘‘high-reactive’’ showed decreased anxiety behaviors after being fostered by highly nurturant rhesus or rat mothers [Suomi, 1997; Caldji et al., 1998]. Such findings Experimental studies in animals have suggested that parenting behaviors can influence anxious behaviors in offspring. For example, in rhesus monkeys and rats, infants bred to be ‘‘high-reactive’’ showed decreased anxiety behaviors after being fostered by highly nurturant rhesus or rat mothers. raise the hypothesis that parenting behaviors in humans might similarly influence the development of anxiety. At least 47 cross-sectional studies of parenting and child anxiety, using mainly self-report and observational methodology, have examined two general parenting dimensions—warmth versus criticism/rejection, and overcontrol versus autonomy granting [see, Rapee, 1997; Wood et al., 2003; Ginsburg et al., 2004; McLeod et al., 2007 for reviews]. In general, these studies suggest that parents of anxious children tend to be more overprotective and less granting of autonomy [e.g., Siqueland et al., 1996; Gruner et al., 111 1999; Whaley et al., 1999; Bogels et al., 2001; Hudson and Rapee, 2001] or less warm and accepting [Whaley et al., 1999; Hudson and Rapee, 2001] than parents of non-anxious children. Lower warmth and responsiveness are theorized to foster poorer emotional regulation, whereas overprotection and high parental control are hypothesized to lead to lower self-efficacy [Chorpita and Barlow, 1998]. Other studies have suggested that parents of anxious children tend to model anxiety, particularly in the case of specific phobias [Gruner et al., 1999; Whaley et al., 1999; Muris et al., 2000], and that they tend to encourage or reinforce anxious avoidance [Barrett et al., 1996; Chorpita et al., 1996; Dadds et al., 1996]. Some of the anxiety-related behaviors observed in parents of anxious children have also been found to be more common in anxious parents [Hirshfeld et al., 1997; Whaley et al., 1999; Lieb et al., 2000; Woodruff-Borden et al., 2002; Turner et al., 2003]. In particular, anxious parents of school-age children appear to display lower warmth and greater disengagement than non-anxious parents; but do not tend to restrict autonomy [Woodruff-Borden et al., 2002; Turner et al., 2003; Dibartolo and Helt, 2007]. A recent meta-analysis subsuming 47 cross-sectional studies with 12,879 participants examined the degree to which parental rejection and control were associated with childhood anxiety disorders or trait anxiety [McLeod et al., 2007]. Most relied on self- or parentreport, with 16 using direct observation of parent–child interactions. The authors found an overall effect size of 0.21, with parenting behaviors accounting for only 4% of the total variance in measures of child anxiety symptoms or disorders. The effect size was 0.35 for diagnosed anxiety disorders. Parental rejection and control accounted for 4% and 6% of the variance in child anxiety, respectively. Parental autonomy granting (examined in only seven of the studies) accounted for the greatest proportion of variance in childhood anxiety (ES ¼ 0.42, 18%), whereas parental warmth accounted for the least (ES ¼ 0.06, <1%). 112 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) Because this literature is crosssectional, it remains unknown whether parental control gives rise to anxiety in children, arises in response to the child’s anxiety, or interacts reciprocally with child anxiety. Only a few empirical studies have explored these effects. One found that mothers reported greater overprotection of their anxious than non-anxious children, suggesting that parenting behavior is shaped by child anxiety [Hudson and Rapee, 2005]. However, these investigators found no differences in mothers’ observed overprotection toward anxious versus nonanxious children during interactive tasks; mothers of anxious children were more intrusive and involved toward all their children [Hudson and Rapee, 2001, 2002]. On the other hand, Moore et al.  found that mothers of anxious children demonstrated lower autonomy granting and lower warmth during a conversation task, regardless of maternal anxiety status, consistent with the hypothesis that children with anxiety may shape their mothers’ behavior. Clearly, prospective studies are needed to help eludicate the directionality of effect between child anxiety and parental behaviors. One prospective study, of mainly African American low income families, found that greater criticism and lower autonomy-granting by parents when children were 6 years old were significantly associated with more symptoms of anxiety at 6-year follow-up, but only in children of anxious mothers [Ginsburg et al., 2004]. This suggests that family-genetic high-risk status and parental behaviors interact to contribute to child anxiety. Other prospective studies have shown support for observed maternal control and criticism (derisiveness) predicting greater stability in inhibition from ages 2 to 4 [Rubin et al., 2002] and for observed anxious-resistant attachment style at 12 months predicting a twofold increase in anxiety disorders at age 17.5 [Warren et al., 1997]. A third prospective study supported a learning hypothesis: it found that increased time in day-care in infancy predicted reduced BI in toddlerhood [Fox, 2004], suggesting that greater naturalistic exposure to new people and settings may reduce inhibition. Beyond its highly cross-sectional nature and the clear need for further prospective studies, there are several other limitations to the literature on the relationship between parenting behaviors and the development and maintenance of child anxiety disorders. First, other than the Ginsburg et al.  study, most studies of parenting behavior and child anxiety have included predominantly Caucasian families, limiting the generalizability of the findings. Second, studies have relied mainly on self-report and observational measures. Self-report measures are subject to biases in participants’ perceptions of parenting behaviors, whereas for observational measures, different types of autonomy-granting situations observed and observational coding schemes are used across research labs. Studies suggest that variations in the observational situations used (e.g., structured versus unstructured tasks) can influence the parenting behaviors observed [Ginsburg et al., 2006]. Third, not all studies took into account parental psychopathology; this variable is important to consider and control for, since it is common among parents of clinically anxious children and since behaviors such as low warmth or high criticism (found more commonly among anxious parents) may serve as proxies for parental anxiety. Despite these limitations, experimental modifications of parental behaviors (i.e., interventions teaching parents to model and reinforce non-anxious coping instead of avoidance) have shown efficacy in reducing children’s anxiety symptoms [Rapee et al., 2005], particularly when parents themselves are anxious [Cobham et al., 1998]. Therefore, parents can learn to respond to a child’s anxious behaviors in ways that can make a positive difference in the child’s outcome. [For a more detailed treatment of these issues, see Hirshfeld-Becker and Biederman, 2002; Rapee, 2002]. Life Events Retrospective studies of adults and children have implicated past (e.g., ARTICLE parental separations) and recent life events (e.g., stressors, threats) in the development of anxiety disorders. Recent prospective studies have confirmed these results, finding links between health problems, divorce, abuse, and loss at ages 6–7 and increased odds of anxiety disorder 1.5 years later [Kroes et al., 2002]; between parental separation, conduct disorder in a father or stepfather, and poverty before age 5 and anxiety disorders at age 15 (controlling for maternal depression and anxiety) [Phillips et al., 2005]; and between poor academic performance in first grade and anxiety in 7th grade [Grover et al., 2005]. Another study found no relationship between parental divorce and anxiety disorders 10 years later [Nomura et al., 2002]. Peer Factors Cross-sectional studies have found associations between peer rejection and social anxiety. A recent prospective study followed 144 ninth-graders over a year and found that being left out or rejected by peers accounted for significant variance in later social phobia symptoms, whereas being teased or bullied did not [Storch et al., 2005]. Cross-sectional studies have found associations between peer rejection and social anxiety. A recent prospective study followed 144 ninth-graders over a year and found that being left out or rejected by peers accounted for significant variance in later social phobia symptoms, whereas being teased or bullied did not. The association between social anxiety and peer relationships is likely reciprocal [Morris, 2001]. [For further discussion of the impact of peer factors on social anxiety and psychosocial adjustment, see Storch and Ledley, 2005]. ARTICLE Perinatal Factors Animal studies have suggested that experimentally-induced stress during pregnancy may influence the offspring’s stress responses and fearful behavior [e.g., Clarke and Schneider, 1993]. Indeed, in a prospective study of a large community cohort (N ¼ 6,996), rates of mother-rated emotional problems in boys at age 6.75 years was predicted by the mothers’ rating of their own anxiety at 32 weeks gestation (odds ratio 1.64 [1.06–2.52]), even when maternal anxiety at two post-natal assessments was covaried [O’Connor et al., 2003]. Although the study did not assess maternal anxiety disorders (which could have a genetic association with childhood emotional problems), the control for post-natal anxiety can be considered to partially control for maternal anxiety disorder. Retrospective studies of children found associations between pregnancy or birth complications and childhood anxiety disorders [Velez et al., 1989; Rapee and Szollos, 2002] even after controlling for parental anxiety disorders [Hirshfeld-Becker et al., 2004a]. However, a retrospective study of adults found no associations for most anxiety disorders, and a lower rate of pregnancy complications for social phobia [Foley et al., 2001]. CONCLUSION A review of the literature on developmental antecedents to anxiety disorders points to several clear risk factors and many more that remain to be explored. The evidence that offspring of parents with anxiety disorder are at elevated risk for a spectrum of anxiety disorders (with a risk ratio compared with the offspring of unaffected parents on the order of 2.5) is clear, as is the finding that anxiety disorders in childhood and adolescence (particularly separation anxiety disorder, overanxious disorder/GAD, and possibly social or specific phobias) predict those in early adulthood (with odds ratios in the 2.0–3.0 range; see Table II). In addition, the evidence suggesting that preschool-age behavioral inhibition (BI) predicts social anxiety AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) disorder in childhood and early adolescence is strong (with odds ratios corrected for parental psychopathology on the order of 3.0 [Hirshfeld-Becker et al., 2007]; and risk ratios not corrected for parental psychopathology in the order of 3.77 for adolescents in general and 7.33 for girls only [Schwartz et al., 1999]), as is the evidence that AS increases vulnerability for the onset of panic disorder and likely other anxiety diagnoses (accounting for about 4–12% of the variance in panic or anxiety symptoms). Although not specific to anxiety, NA also appears to predict later anxiety symptoms. Promising evidence is accruing in support of hypotheses about parenting factors in anxiety disorders (with effect sizes in the 0.2–0.4 range), peer factors in social phobia, and stressful life events, including perinatal stressors (with odds ratios in the 1.0– 2.0 range), for anxiety disorders in general. As can be seen from the risk and odds ratios and effect sizes reported in the literature, parental psychopathology, early anxiety disorders, and behavioral inhibition appear to have moderate associations with subsequent anxiety disorders, whereas parenting factors show small to moderate associations, and perinatal factors may show relatively small associations. However, because most studies of one factor do not control for others, these associations cannot be assumed to be additive. More studies are needed that examine multiple risk factors in the same individuals and examine their combined contributions over time to the prospective risk for anxiety disorders. We can identify several additional fruitful areas for future study: (1) Prospective longitudinal studies: First, although several important long-term longitudinal studies have been conducted, many more are needed. Ideally, such studies would begin in early childhood or during gestation. Staggered longitudinal designs (simultaneously following several different cohorts beginning perinatally, in toddlerhood, in early and middle childhood, and in adolescence) could speed up the time needed to understand the course of hypothesized risk factors. In particular, 113 prospective studies examining the role of harm avoidance, EC, parenting factors, peer factors, and life events are needed. In addition, given the potential overlap in constructs, it would be useful to measure multiple constructs in the same samples and study their overlap systematically. Twin studies could be useful in determining the degree to which different constructs share common genetic and environmental variance. Prospective studies of offspring at risk, in whom the expected base rate of onset of anxiety is higher, can complement studies of representative community samples to examine these issues. (2) Greater use of observational measures: Although informative, questionnaires are subject to biases which can be mitigated by complementary assessment with observational measures. As discussed above, because anxiety disorders can onset very early, the study of temperamental precursors could benefit from the development of creative ways to observe temperamental characteristics through standardized laboratory tasks early in life. In addition, more ecologically valid observational methods are needed for assessing parent and peer interactions. For BI, better standardization of laboratory measures across studies, and prospective comparisons of retrospective questionnaires with early laboratory measures would be beneficial. (3) Greater attention to parental psychopathology: Because of the clear risk conferred by parental anxiety and possibly mood disorders, and because associations with parental psychopathology could potentially confound other associations with anxiety (e.g., parenting behaviors, perinatal anxiety, stressful events, temperament) it is important for investigators to assess and covary lifetime history of anxiety disorders in parents in order to understand whether other factors mediate, moderate, or are independent of the effects of parental anxiety disorder. (4) Examination of other potential precursors: Our review did not exhaust all of the features found to be crosssectionally associated with anxiety disorders. In particular, neurophysiologic 114 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) (e.g., right frontal EEG asymmetry) and neurofunctional differences (e.g., amygdalar activation) associated with anxiety could be fruitfully explored as potential vulnerability factors for anxiety among at-risk populations. Similarly, cognitive (e.g., bias for threat, catastrophic interpretations, underappraisal of coping resources) and emotional features of anxiety (e.g., discomfort intolerance [Schmidt et al., 2007]) should be explored to determine whether they predate onset of anxiety disorders. Other factors may also be explored; for example, a recent study found that language impairment assessed at age 5 predicted social phobia at age 19 [Voci et al., 2006]. More studies examining interactions between known genotypes and particular vulnerability factors in predicting anxiety outcomes would be particularly exciting. (5) Greater inclusion of diverse populations among the individuals studied: As reflected in our review, the majority of prospective studies of developmental antecedents to anxiety disorders have included mainly Caucasian participants. Greater efforts to focus on more diverse populations are necessary in order to improve our understanding of the role of these risk factors among other groups. (6) Intervention studies: Finally, since a major impetus for the study of developmental antecedents is the informing of preventive interventions, another important approach is to conduct intervention studies which modify hypothesized risk factors and examine whether onset of anxiety disorders can be averted. Studies intervening to reduce BI [Rapee et al., 2005] and AS (e.g., [McNally, 2002]) have been initiated, and others are underway. The next 20 years promise to be as exciting as the previous in the evolution of our understanding of the etiology of the anxiety disorders. REFERENCES Allgulander C, Cloninger CR, Przybeck TR, Brandt L. 1998. Changes on the temperament and character inventory after paroxetine treatment in volunteers with generalized anxiety disorder. Psychopharmacol Bull 34:165–166. Angst J, Vollrath M. 1991. The natural history of anxiety disorders. Acta Psychiatr Scand 84:446–452. Aschenbrand SG, Kendall PC, Webb A, Safford SM, Flannery-Schroeder E. 2003. Is childhood separation anxiety disorder a predictor of adult panic disorder and agoraphobia? A seven-year longitudinal study. J Am Acad Child Adolesc Psychiatry 42:1478–1485. Barrett PM, Rapee RM, Dadds MM, Ryan SM. 1996. Family enhancement of cognitive style in anxious and aggressive children. J Abnorm Child Psychol 24:187–203. Battaglia M, Bajo S, Strambi LF, Brambilla F, Castronovo C, Vanni G, Bellodi L. 1997. Physiological and behavioral responses to minor stressors in offspring of patients with panic disorder. J Psychiat Res 31:365– 376. Beidel D, Turner S. 1997. At risk for anxiety: I. Psychopathology in the offspring of anxious parents. J Am Acad Child Adolesc Psychiatry 36:918–924. Biederman J, Rosenbaum JF, Hirshfeld DR, Faraone SV, Bolduc EA, Gersten M, Meminger SR, Kagan J, Snidman N, Reznick JS. 1990. Psychiatric correlates of behavioral inhibition in young children of parents with and without psychiatric disorders. Arch Gen Psychiatry 47:21–26. Biederman J, Rosenbaum JF, Bolduc EA, Faraone SV, Hirshfeld DR. 1991. A high-risk study of young children of parents with panic disorder and agoraphobia with and without major depression. Psychiatry Res 37:333– 348. Biederman J, Rosenbaum JF, Bolduc-Murphy EA, Faraone SV, Chaloff J, Hirshfeld DR, Kagan J. 1993. A 3-year follow-up of children with and without behavioral inhibition. J Am Acad Child Adolesc Psychiatry 32:814–821. Biederman J, Faraone S, Hirshfeld-Becker D, Friedman D, Robin J, Rosenbaum J. 2001a. Patterns of psychopathology and dysfunction in a large sample of high-risk children of parents with panic disorder and major depression: A controlled study. Am J Psychiatry 158:49–57. Biederman J, Hirshfeld-Becker DR, Rosenbaum JF, Herot C, Friedman D, Snidman N, Kagan J, Faraone SV. 2001b. Further evidence of association between behavioral inhibition and social anxiety in children. Am J Psychiatry 158:1673–1679. Biederman J, Monuteaux MC, Faraone SV, Hirshfeld-Becker DR, Henin A, Gilbert J, Rosenbaum JF. 2004. Does referral bias impact findings in high risk offspring for anxiety disorders? A controlled study of high risk children of non-referred parents with panic disorder/agoraphobia and major depression. J Aff Disorders 82:209–216. Biederman J, Petty C, Faraone SV, HirshfeldBecker DR, Henin A, Pollack MH, Rosenbaum JF. 2005. Patterns of comorbidity in panic disorder and major depression: Findings from a non-referred sample. Depress Anxiety 21:55–60. Biederman J, Petty C, Faraone SV, Henin A, Hirshfeld-Becker DR, Pollack MH, de Figuerido S, Feeley R, Rosenbaum JF. 2006a. Effects of parental anxiety disorders in children at high risk for panic disorder: A ARTICLE controlled study. J Affect Disord 94:191– 197. Biederman J, Petty C, Hirshfeld-Becker DR, Henin A, Faraone SV, Dang D, Jakubowski A, Rosenbaum JF. 2006b. A controlled longitudinal five year follow-up study of children at high and low risk for panic disorder and major depression. Psychol Med 36:1141–1152. Biederman J, Petty C, Hirshfeld-Becker DR, Henin A, Faraone SV, Fraire M, Henry B, McQuade JD, Rosenbaum JF. 2007. Developmental trajectories of anxiety disorders in offspring at high risk for panic disorder and major depression. Psychiatr Res 153:245– 252. Black DW, Gaffney GR, Schlosser S, Gabel J. 2003. Children of parents with obsessivecompulsive disorder—A 2-year follow-up study. Acta Psychiatr Scand 107:305–313. Bogels S, van Oosten A, Muris P, Smulders D. 2001. Familial correlates of social anxiety in children and adolescents. Behav Res Ther 39:273–287. Bosquet M, Egeland B. 2006. The development and maintenance of anxiety symptoms from infancy through adolescence in a longitudinal sample. Dev and Psychopathol 18:517– 550. Caldji C, Tannenbaum B, Sharma S, Francis D, Plotsky PM, Meaney MJ. 1998. Maternal care during infancy regulates the development of neural systems mediating the expression of fearfulness in the rat. Proc Natl Acad Sci USA 95:5335–5340. Capps l, Sigman M, Sena R, Henker B, Whalen C. 1996. Fear, anxiety and perceived control in children of agoraphobic parents. J Child Psychol Psychiat 37:445–452. Caspi A, Moffitt TE, Newman DL, Silva PA. 1996. Behavioral observations at age 3 years predict adult psychiatric disorders. Arch Gen Psychiatry 53:1033–1039. Chorpita BF, Barlow DH. 1998. The development of anxiety: The role of control in the early environment. Psychol Bull 124:3–21. Chorpita B, Albano A, Barlow D. 1996. Cognitive processing in children: Relation to anxiety and family influences. J Clin Child Psychol 25:170–176. Clark L, Watson D, Mineka S. 1994. Temperament, personality, and the mood and anxiety disorders. J Abnorm Psychol 103:103–116. Clarke A, Schneider M. 1993. Prenatal stress has long-term effects on behavioral responses to stress in juvenile rhesus monkeys. Dev Psychobiol 26:293–304. Cloninger C. 1986. A unified biosocial theory of personality and its role in the development of anxiety states. Psychiatric Development 3:167–226. Cloninger CR, Van Eerdewegh P, Goate A, Edenberg HJ, Blangero J, Hesselbrock V, Reich T, Nurnberger J Jr, Schuckit M, Porjesz B, Crowe R, Rice JP, Foroud T, Przybeck TR, Almasy L, Bucholz K, Wu W, Shears S, Carr K, Crose C, Willig C, Zhao J, Tischfield JA, Li TK, Conneally PM, et al. 1998. Anxiety proneness linked to epistatic loci in genome scan of human personality traits. Am J Med Genet 81:313– 317. Cobham V, Dadds M, Spence S. 1998. The role of parental anxiety in the treatment of child- ARTICLE hood anxiety. J Cons Clin Psychol 66:893– 905. Cooper PJ, Eke M. 1999. Childhood shyness and maternal social phobia: A community study. Br J Psychiatry 174:439–443. Coryell W. 1997. Hypersenstivity to carbon dioxide as a disease-specific trait marker. Biol Psychiatry 41:259–263. Coryell W, Dindo L, Fyer A, Pine DS. 2006a. Onset of spontaneous panic attacks: A prospective study of risk factors. Psychosom Med 68:754–757. Coryell W, Pine D, Fyer A, Klein D. 2006b. Anxiety responses to CO2 inhalation in subjects at high-risk for panic disorder. J Affect Disord 92:63–70. Costello EJ, Mustillo S, Erkanli A, Keeler G, Angold A. 2003. Prevalence and development of psychiatric disorders in childhood and adolescence. Arch Gen Psychiatry 60: 837–844. Craske MG, Poulton R, Tsao JC, Plotkin D. 2001. Paths to panic disorder/agoraphobia: An exploratory analysis from age 3 to 21 in an unselected birth cohort. J Am Acad Child Adolesc Psychiatry 40:556–563. Dadds MR, Barrett PM, Rapee RM, Ryan S. 1996. Family process and child anxiety and aggression: An observational analysis. J Abnorm Child Psychol 24:715–734. Dibartolo PM, Helt M. 2007. Theoretical models of affectionate versus affectionless control in anxious families: A critical examination based on observations of parent-child interactions. Clin Child Fam Psychol Rev 10: 253–274. Drake KL, Kearney CA. 2007. Child anxiety sensitivity and family environment as mediators of the relationship between parent psychopathology, parent anxiety sensitivity, and child anxiety. J Psychopathol Behav Assess (in press). Egger HL, Angold A. 2006. Common emotional and behavioral disorders in preschool children: Presentation, nosology, and epidemiology. J Child Psychol Psychiatry 47:313– 337. Ehlers A. 1995. A 1-year prospective study of panic attacks: Clinical course and factors associated with maintenance. J Abnorm Psychol 104:164–172. Eley TC, Stirling L, Ehlers A, Gregory AM, Clark DM. 2004. Heart-beat perception, panic/ somatic symptoms and anxiety sensitivity in children. Behav Res Ther 42:439–448. Eysenck H. 1967. Biological bases of personality. Springfield, IL: Charles C. Thomas. Ferdinand RF, Verhulst FC. 1995. Psychopathology from adolescence into young adulthood: An 8-year follow-up study. Am J Psychiatry 152:1586–1594. Ferdinand RF, Dieleman G, Ormel J, Verhulst FC. 2007. Homotypic versus heterotypic continuity of anxiety symptoms in young adolescents: Evidence for distinctions between DSM-IV subtypes. J Abnorm Child Psychol 35:325–333. Foley DL, Thacker LRn, Aggen SH, Neale MC, Kendler KS. 2001. Pregnancy and perinatal complications associated with risks for common psychiatric disorders in a population-based sample of female twins. Am J Med Genet 105:426–431. AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) Fordham K, Stevenson-Hinde J. 1999. Shyness, friendship quality, and adjustment during middle childhood. J Child Psychol Psychiat 40:757–768. Fox NA. 2004. Temperament and early experience form social behavior. Ann N Y Acad Sci 1038:171–178. Fox NA, Nichols KE, Henderson HA, Rubin K, Schmidt L, Hamer D, Ernst M, Pine DS. 2005. Evidence for a gene-environment interaction in predicting behavioral inhibition in middle childhood. Psychol Sci 16:921–926. Fredman SJ, Hirshfeld-Becker DR, Smoller J, Rosenbaum JF. 2003. Antecendents of adult anxiety disorders. In: Nutt DJ, Ballenger J, editors. Anxiety disorders. Oxford: Blackwell Science, Ltd. Ginsburg GS, Drake KL. 2002. Anxiety sensitivity and panic attack symptomatology among low-income African-American adolescents. J Anxiety Disord 16:83–96. Ginsburg G, Grover R, Ialongo N. 2004. Parenting behaviors among anxious and non-anxious mothers: Relation with concurrent and long-term child outcomes. Child Fam Behav Ther 26:23–41. Ginsburg GS, Grover RL, Cord JJ, lalongo N. 2006. Observational measures of parenting in anxious and nonanxious mothers: Does type of task matter? J Clin Child Adolesc Psychol 35:323–328. Gjone H, Stevenson J. 1997. A longitudinal twin study of temperament and behavior problems: Common genetic or environmental influences? J Am Acad Child Adolesc Psychiatry 36:1448–1456. Gladstone GL, Parker GB. 2006. Is behavioral inhibition a risk factor for depression? J Affect Disord 95:85–94. Gregory AM, Caspi A, Moffit TE, Koenen K, Eley TC, Poulton R. 2007. Juvenile mental health histories of adults with anxiety disorders. Am J Psychiatry 164:301–308. Grigoroiu-Serbanescu M, Christodorescu D, Jipescu I, Totoescu A, Marinescu E, Ardelean V. 1989. Psychopathology in children aged 10–17of bipolar parents: Psychopathology rate and correlates of the severity of the psychopathology. J Affect Disord 16: 167–179. Grillon C, Dierker L, Merikangas K. 1997. Startle modulation in children at risk for anxiety disorders and/or alcoholism. J Am Acad Child Adolesc Psychiatry 36:925–932. Grover RL, Ginsburg GS, Ialongo N. 2005. Childhood predictors of anxiety symptoms: A longitudinal study. Child Psychiatry Hum Dev 36:133–153. Gruner K, Muris P, Merckelbach H. 1999. The relationship between anxious rearing behaviours and anxiety disorders symptomatology in normal children. J Behav Ther Exp Psychiatry 30:27–35. Hayward C, Killen J, Kraemer K, Taylor C. 1998. Linking self-reported childhood behavioral inhibition to adolescent social phobia. J Am Acad Child Adolesc Psychiatry 37:1308– 1316. Hayward C, Killen JD, Kraemer HC, Taylor CB. 2000. Predictors of panic attacks in adolescents. J Am Acad Child Adolesc Psychiatry 39:207–214. 115 Heath AC, Cloninger CR, Martin NG. 1994. Testing a model for the genetic structure of personality: A comparison of the personality systems of Cloninger and Eysenck. J Pers Soc Psychol 66:762–775. Hettema JM, Prescott CA, Myers JM, Neale MC, Kendler KS. 2005. The structure of genetic and environmental risk factors for anxiety disorders in men and women. Arch Gen Psychiatry 62:182–189. Hettema JM, Neale MC, Myers JM, Prescott CA, Kendler KS. 2006. A population-based twin study of the relationship between neuroticism and internalizing disorders. Am J Psychiatry 163:857–864. Hirshfeld DR, Rosenbaum JF, Biederman J, Bolduc EA, Faraone SV, Snidman N, Reznick JS, Kagan J. 1992. Stable behavioral inhibition and its association with anxiety disorder. J Am Acad Child Adolesc Psychiatry 31:103–111. Hirshfeld D, Biederman J, Brody L, Faraone S, Rosenbaum J. 1997. Expressed emotion toward children with behavioral inhibition: Associations with maternal anxiety disorders. J Am Acad Child Adolesc Psychiatry 36:910–917. Hirshfeld-Becker DR, Biederman J. 2002. Rationale and principles for early intervention with young children at risk for anxiety disorders. Clin Child Fam Psychol Rev 5:161–172. Hirshfeld-Becker DR, Biederman J, Faraone SV, Robin JA, Friedman D, Rosenthal JM, Rosenbaum JF. 2004a. Pregnancy complications associated with childhood anxiety disorders. Depress Anxiety 19:152–162. Hirshfeld-Becker DR, Biederman J, Rosenbaum JF. 2004b. Behavioral inhibition. In: Morris TL, March JS, editors. Anxiety disorders in children and adolescents. 2nd edition. NY: The Guilford Press. p 27–58. Hirshfeld-Becker DR, Biederman J, Henin A, Faraone SV, Davis S, Harrington K, Rosenbaum JF. 2007. Behavioral inhibition in preschool children at risk is a specific predictor of middle childhood social anxiety: A five-year follow-up. J Dev Behav Pediatr 28:225–233. Hudson JL, Rapee RM. 2001. Parent-child interactions and anxiety disorders: An observational study. Behav Res Ther 39:1411–1427. Hudson JL, Rapee RM. 2002. Parent-child interactions in clinically anxious children and their siblings. J Clin Child Adolesc Psychol 31:548–555. Hudson JL, Rapee RM. 2005. Parental perceptions of overprotection: Specific to anxious children or shared between siblings? Behav Change 22:185–194. Johnson JG, Cohen P, Kasen S, Brook JS. 2006. A multiwave multi-informant study of the specificity of the association between parental and offspring psychiatric disorders. Compr Psychiatry 47:169–177. Kagan J, Reznick JS, Snidman N. 1988. Biological bases of childhood shyness. Science 240: 167–171. Kagan J, Snidman N, Zetner M, Peterson E. 1999. Infant temperament and anxious symptoms in school age children. Dev Psychopathol 11:209–224. 116 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) Kim-Cohen J, Caspi A, Moffit TE, Harrington H, Milne BJ, Poulton R. 2003. Prior juvenile diagnoses in adults with mental disorders. Arch Gen Psychiatry 60:709– 717. Klein R. 1995. Is panic disorder associated with childhood separation anxiety disorder? Clinical Neuropharmacology 18:S7–S14. Kochanska G. 1991. Patterns of inhibition to the unfamiliar in children of normal and affectively ill mothers. Child Dev 62:250– 263. Kochanska G, Knaack A. 2003. Effortful control as a personality characteristic of young children: Antecedents, correlates, and consequences. J Pers 71:1087–1112. Kroes M, Kalff AC, Steyaert J, Kessels AG, Feron FJ, Hendriksen JG, van Zeben TM, Troost J, Jolles J, Vles JS. 2002. A longitudinal community study: Do psychosocial risk factors and child behavior checklist scores at 5 years of age predict psychiatric diagnoses at a later age? J Am Acad Child Adolesc Psychiatry 41:955–963. Last CG, Hersen M, Kazdin AE, Orvaschel H, Perrin S. 1991. Anxiety disorders in children and their families. Arch Gen Psychiatry 48:928–934. 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. Leen-Feldner EW, Reardon LE, Zvolensky MJ. 2007. Pubertal status and emotional reactivity to a voluntary hyperventilation challenge predicting panic symptoms and somatic complaints: A laboratory-based multi-informant test. Behav Modif 31:8–31. Lieb R, Wittchen HU, Hofler M, Fuetsch M, Stein MB, Merikangas KR. 2000. Parental psychopathology, parenting styles, and the risk of social phobia in offspring: A prospective-longitudinal community study. Arch Gen Psychiatry 57:859–866. Lonigan CJ, Phillips BM, Hooe ES. 2003. Relations of positive and negative affectivity to anxiety and depression in children: Evidence from a latent variable longitudinal study. J Counseling and Clinical Psychology 71:465–481. Lonigan CJ, Vasey MW, Phillips BM, Hazen RA. 2004. Temperament, anxiety, and the processing of threat-relevant stimuli. J Clin Child Adolesc Psychol 33:8–20. Mackinnon A, Henderson A, Andrews G. 1990. Genetic and environmental determinants of the lability of trait neuroticism and the symptoms of anxiety and depression. Psychol Med 20:581–590. Maller RG, Reiss S. 1992. Anxiety sensitivity in 1984 and panic attacks in 1987. J Anx Disord 6:241–247. Manassis K, Bradley S, Goldberg S, Hood J, Swinson R. 1995. Behavioural inhibition, attachment and anxiety in children of mothers with anxiety disorders. Can J Psychiatry 40:87–92. Mannuzza S, Klein RG, Moulton JLr, Scarfone N, Malloy P, Vosburg SK, Klein DF. 2002. Anxiety sensitivity among children of parents with anxiety disorders: A controlled high-risk study. J Anxiety Disord 16:135– 148. Masi G, Mucci M, Favilla L, Brovedani P, Millepiedi S, Perugi G. 2003. Temperament in adolescents with anxiety and depressive disorders and in their families. Child Psychiatry and Human Dev 33:245–259. McLeod BD, Wood JJ, Weisz JR. 2007. Examining the association between parenting and childhood anxiety: A meta-analysis. Clin Psychol Rev 27:155–172. McNally RJ. 1994. Panic disorder: A critical analysis. New York: Gilford Press. McNally RJ. 2002. Anxiety sensitivity and panic disorder. Biol Psychiatry 52:938–946. Merikangas KR, Dierker LC, Szatmari P. 1998a. Psychopathology among offspring of parents with substance abuse and/or anxiety disorders: A high-risk study. J Child Psychol Psychiat 39:711–720. Merikangas KR, Swendsen JD, Preisig MA, Chazan RZ. 1998b. Psychopathology and temperament in parents and offspring: Results of a family study. J Affect Disord 51:63–74. Merikangas KR, Avenevoli S, Dierker L, Grillon C. 1999. Vulnerability factors among children at risk for anxiety disorders. Biol Psychiatry 46:1523–1535. Mick M, Telch M. 1998. Social anxiety and history of behavioral inhibition in young adults. J Anx Disord 12:1–20. Mineka S, Watson D, Clark LA. 1998. Comorbidity of anxiety and unipolar mood disorders. Annu Rev Psychol 49:377–412. Moffitt TE, Caspi A, Harrington H, Milne BJ, Melchior M, Goldberg D, Poulton R. 2007. Generalized anxiety disorder and depression: Childhood risk factors in a birth cohort followed to age 32. Psychol Med 37:441–452. Moore PS, Whaley SE, Sigman M. 2004. Interactions between mothers and children: Impacts of maternal and child anxiety. J Abnorm Psychol 113:471–476. Morris TL. 2001. Childhood anxiety disorders: Etiology, assessment, and treatment in the new millennium. Curr Psychiatry Rep 3: 267–272. Mortberg E, Bejerot S, Aberg-Wistedt A. 2007. Temperament and character dimensions in patients with social phobia: Patterns of change following treatments? Psychiatry Res 152:81–90. Mufson L, Wessman MM. 1992. Depression and anxiety in parents and children: A direct interview study. J Anx Disord 6:1–13. Muris P, Meesters C, Merckelbach H, Hulsenbeck P. 2000. Worry in children is related to perceived parental rearing and attachment. Behav Res Ther 38:487–497. Muris P, Meesters C, Spinder M. 2003. Relationships between child- and parent-reported behavioral inhibitionand symptoms of anxiety and depression in normal adolescents. Pers Individ Differ 34:759–771. Newman DL, Moffitt TE, Caspi A, Magdol L, Silva PA, Stanton WR. 1996. Psychiatric disorder in a birth cohort of young adults: Prevalence, comorbidity, clinical significance, and new case incidence from ages 11 to 21. J Consult Clin Psychol 64:552– 562. Nomura Y, Wickramaratne PJ, Warner V, Mufson L, Weissman MM. 2002. Family discord, parental depression, and psychopathology in ARTICLE offspring: Ten-year follow-up. J Am Acad Child Adolesc Psychiatry 41:402–409. O’Connor TG, Heron J, Golding J, Glover V, Team AS. 2003. Maternal antenatal anxiety and behavioural/emotional problems in children: A test of a programmin hypothesis. J Child Psychol Psychiatry 44:1025–1036. Ollendick TH, Hirshfeld-Becker DR. 2002. The developmental psychopathology of social anxiety disorder. Biol Psychiatry 51:44–58. Öst L-G. 1987. Age of onset of different phobias. J Abnorm Psychol 96:223–229. Otto M, Pollock M, Rosenbaum JF, Sachs GS, Asher RH. 1994. Childhood history of anxiety in adults with panic disorder: Association with anxiety sensitivity and comorbidity. Harvard Rev Psychiatry 1:288–293. Otto MW, Pollack MH, Maki KM, Gould RA, Worthington JJI, Smoller JW, Rosenbaum JF. 2001. Childhood history of anxiety disorders among adults with social phobia: Rates, correlates, and comparisons with patients with panic disorder. Depress Anxiety 14:209–213. Perez-Edgar K, Fox NA. 2005. Temperament and anxiety disorders. Child Adolesc Psychiatr Clin N Am 14:681–706, viii. Perez-Edgar K, Roberson-Nay R, Hardin MG, Poeth K, Guyer AE, Nelson EE, McClure EB, Henderson HA, Fox NA, Pine DS, et al. 2007. Attention alters neural responses to evocative faces in behaviorally inhibited adolescents. Neuroimage 35:1538–1546. Phillips NK, Hammen CL, Brennan PA, Najman JM, Bor W. 2005. Early adversity and the prospective prediction of depressive and anxiety disorders in adolescents. J Abnorm Child Psychol 33:13–24. 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. Pollock RA, Carter AS, Avenevoli S, Dierker LC, Chazan-Cohen R, Merikangas KR. 2002. Anxiety sensitivity in adolescents at risk for psychopathology. J Clin Child Adolesc Psychol 31:343–353. Pollock RA, Carter AS, Amir N, Marks LE. 2006. Anxiety sensitivity and auditory perception of heartbeat. Behav Res Ther 44:1739– 1756. Posner MI, Rothbart MK. 1998. Attention, selfregulation and consciousness. Phil Trans R Soc Lond B 353:1915–1927. Prior M, Smart D, Sanson A, Oberklaid F. 2000. Does shy-inhibited temperament in childhood lead to anxiety problems in adolescence? J Am Acad Child Adolesc Psychiatry 39:461–468. Rapee R. 1997. Potential role of childrearing practices in the development of anxiety and depression. Clin Psychol Rev 17:47–67. Rapee RM. 2002. The development and modification of temperamental risk for anxiety disorders: Prevention of a lifetime of anxiety? Biol Psychiatry 52:947–957. Rapee RM, Szollos AA. 2002. Developmental antecedents of clinical anxiety in childhood. Behav Change 19:146–157. Rapee RM, Kennedy S, Ingram M, Edwards S, Sweeney L. 2005. Prevention and early intervention of anxiety disorders in inhib- ARTICLE ited preschool children. J Consult Clin Psychol 73:488–497. Reiss S, McNally R. 1985. Expectancy model of fear. In: Reiss S, Bootzin R, editors. Theoretical Issues in Behavior Therapy. San Diego, CA: Academic Press. p 107– 121. Rende RD. 1993. Longitudinal relations between temperament traits and behavioral syndromes in middle childhood. J Am Acad Child Adolesc Psychiatry 32:287–290. Rosenbaum JF, Biederman J, Gersten M, Hirshfeld DR, Meminger SR, Herman JB, Kagan J, Reznick JS, Snidman N. 1988. Behavioral inhibition in children of parents with panic disorder and agoraphobia: A controlled study. Arch Gen Psychiatry 45: 463–470. Rosenbaum JF, Biederman J, Hirshfeld DR, Faraone SV, Bolduc EA, Kagan J, Snidman N, Reznick JS. 1991. Further evidence of an association between behavioral inhibition and anxiety disorders: Results from a family study of children from a non-clinical sample. J Psychiat Res 25:49–65. Rosenbaum JF, Biederman J, Bolduc EA, Hirshfeld DR, Faraone SV, Kagan J. 1992. Comorbidity of parental anxiety disorders as risk for child-onset anxiety in inhibited children. Am J Psychiatry 149:475–478. Rosenbaum JF, Biederman J, Hirshfeld-Becker DR, Kagan J, Snidman N, Friedman D, Nineberg A, Gallery DJ, Faraone SV. 2000. A controlled study of behavioral inhibition in children of parents with panic disorder and depression. Am J Psychiatry 157:2002– 2010. Rubin KH, Burgess KB, Hastings PD. 2002. Stability and social-behavioral consequences of toddlers’ inhibited temperament and parenting behaviors. Child Dev 73:483– 495. Rydell AM, Berlin L, Bohlin G. 2003. Emotionality, emotion regulation, and adaptation among 5- to 8-year-old children. Emotion 3:30–47. Schmidt NB, Lerew DR, Jackson RJ. 1997. The role of anxiety sensitivity in the pathogenesis of panic: Prospective evaluation of spontaneous panic attacks during acute stress. J Abnorm Psychol 106:355–364. Schmidt NB, Lerew DR, Jackson RJ. 1999. Prospective evaluation of anxiety sensitivity in the pathogenesis of panic: Replication and extension. J Abnorm Psychol 108:532– 537. Schmidt NB, Zvolensky MJ, Maner JK. 2006. Anxiety sensitivity: Prospective prediction of panic attacks and Axis I pathology. J Psychiatr Res 40:691–699. Schmidt NB, Richey JA, Cromer KR, Buckner JD. 2007. Discomfort intolerance: Evaluation of a potential risk factor for anxiety psychopathology. Behav Ther 38:247– 255. Schwartz C, Snidman N, Kagan J. 1999. Adolescent social anxiety as an outcome of inhibited temperament in childhood. J Am Acad Child Adolesc Psychiatry 38:1008– 1015. Schwartz CE, Wright CI, Shin LM, Kagan J, Rauch SL. 2003. Inhibited and uninhibited infants ‘‘grown up’’: Adult amygdalar AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) response to novelty. Science 300:1952– 1953. Siqueland L, Kendall P, Steinberg L. 1996. Anxiety in children: Perceived family environments and observed family interaction. J Clin Child Psychol 25:225–237. Smoller JW, Tsuang MT. 1998. Panic and phobic anxiety: Defining phenotypes for genetic studies. Am J Psychiatry 155:1152–1162. Smoller J, Yamaki L, Fagerness J, Biederman J, Racette S, Laird N, Kagan J, Snidman N, Faraone S, Hirshfeld-Becker D, Tsuang MT, Slaugenhaupt SA, Rosenbaum JF, Sklar PB. 2005. The corticotropin releasing hormone gene and behavioral inhibition in children at risk for panic disorder. Biol Psychiatry 57:1485–1492. Spatola CA, Fagnani C, Pesenti-Gritti P, Ogliari A, Stazi MA, Battaglia M. 2007. A general population twin study of the CBCL/618DSM-oriented scales. J Am Acad Child Adolesc Psychiatry 46:619–627. Stein MB. 2006. An epidemiologic perspective on social anxiety disorder. J Clin Psychiatry 67:3–8. Stein MB, Jang KL, Livesley WJ. 1999. Heritability of anxiety sensitivity: A twin study. Am J Psychiatry 156:246–251. Stein MB, Chartier MJ, Lizak MV, Jang KL. 2001. /PUBYR> Familial aggregation of anxiety-related quantitative traits in generalized social phobia: Clues to understanding ‘‘disorder’’ heritability? Am J Med Genet 105:79–83. Stein MB, Schork NJ, Gelernter J. 2007. Geneby-environment (serotonin transporter and childhood maltreatment) interaction for anxiety sensitivity, an intermediate phenotype for anxiety disorders. Neuropsychopharmacology (in press). Storch EA, Ledley DR. 2005. Peer victimization and psychosocial adjustment in children: Current knowledge and future directions. Clin Pediatr (Phila) 44:29–38. Storch EA, Masia-Warner C, Crisp H, Klein RG. 2005. Peer victimization and social anxiety in adolescence: A prospective study. Aggressive Behav 31:437–452. Suomi SJ. 1997. Early determinants of behaviour: Evidence from primate studies. Br Med Bull 53:170–184. Sylvester C, Hyde T, Reichler R. 1988. Clinical psychopathology among children of adults with panic disorder. In: Dunner D, Gershon E, Barrett J, editors. Relatives at risk for mental disorder. New York: Raven Press, Ltd. p 87–102. Turner SM, Beidel DC, Costello A. 1987. Psychopathology in the offspring of anxiety disorders patients. J Consult Clin Psychol 55:229–235. Turner SM, Beidel DC, Robertson-Nay R, Tervo K. 2003. Parenting behaviors in parents with anxiety disorders. Behav Res Ther 41:541– 554. Unnewehr S, Schneider S, Florin I, Margraf J. 1998. Psychopathology in children of patients with panic disorder or animal phobia. Psychopathology 31:69–84. van Ameringen M, Mancini C, Oakman J. 1998. The relationship of behavioral inhibition and shyness to anxiety disorder. J Nerv Mental Dis 186:425–431. 117 van Beek N, Perna G, Schruers K, Muris P, Griez E. 2005. Anxiety sensitivity in children of panic disorder patients. Child Psychiatry Hum Dev 35:315–324. Velez CN, Johnson J, Cohen P. 1989. A longitudinal analysis of selected risk factors for childhood psychopathology. J Am Acad Child Adolesc Psychiatry 28:861–864. Voci SC, Beitchman JH, Brownlie EB, Wilson B. 2006. Social anxiety in late adolescence: The importance of early childhood language impairment. Anxiety Disorders 20:915– 930. Von Korff M, Eaton W, Keyl P. 1985. The epidemiology of panic attacks and panic disorder: Results of three community surverys. Am J Epidemiol 122:970–981. Warner V, Mufson L, Weissman MM. 1995. Offspring at high and low risk for depression and anxiety: Mechanisms of psychiatric disorder. J Am Acad Child Adolesc Psychiatry 34:786–797. Warren S, Huston L, Egeland B, Sroufe L. 1997. Child and adolescent anxiety disorders and early attachment. J Am Acad Child Adolesc Psychiatry 36:637–644. Warren SL, Gunnar MR, Kagan J, Anders TF, Simmens SJ, Rones M, Wease S, Aron E, Dahl RE, Sroufe LA. 2003. Maternal panic disorder: Infant temperament, neurophysiology, and parenting behaviors. J Am Acad Child Adolesc Psychiatry 42:814–825. Weissman M, Leckman J, Merikangas K, Gammon G, Prusoff B. 1984. Depression and anxiety disorders in parents and children: Results from the Yale Family Study. Arch Gen Psychiatry 41:845–852. Weissman MM, Wolk S, Wickramaratne P, Goldstein RB, Adams P, Greenwald S, Ryan ND, Dahl RE, Steinberg D. 1999. Children with prepubertal-onset major depressive disorder and anxiety grown up. Arch Gen Psychiatry 56:794–801. Weissman MM, Wickramaratne P, Nomura Y, Warner V, Pilowsky D, Verdeli H. 2006. Offspring of depressed parents: 20 years later. Am J Psychiatry 163:1001–1008. Whaley S, Pinto A, Sigman M. 1999. Characterizing interactions between anxious mothers and their children. J Consult Clin Psychol 67:826–836. Wood JJ, McLeod BD, Sigman M, Hwang W, Chu BC. 2003. Parenting and childhood anxiety: Teory, empirical findings, and future directions. J Child Psychol Psychiatry 44:134–151. Woodruff-Borden J, Morrow C, Bourland S, Cambron S. 2002. The behavior of anxious parents: Examining mechanisms of transmission of anxiety from parent to child. J Clin Child Adolesc Psychol 31:364–374. Woodward LJ, Fergusson DM. 2001. Life course outcomes of young people with anxiety disorders in adolescence. J Am Acad Child Adolesc Psychiatry 40:1086–1093. Zohar AH, Dina C, Rosolio N, Osher Y, Gritsenko I, Bachner-Melman R, Benjamin J, Belmaker RH, Ebstein RP. 2003. Tridimensional personality questionnaire trait of harm avoidance (anxiety proneness) is linked to a locus on chromosome 8p21. Am J Med Genet Part B Neuropsychiatr Genet 117B: 66–69.